The State of GIS in Developing Countries: A Diffusion and GIS & Society Analysis of Uganda, and the Potential for Mobile Location-Based Services A Dissertation SUBMITTED TO THE FACULTY OF UNIVERSITY OF MINNESOTA BY Sami Eria IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY Advisor: Robert B. McMaster September 2012 © Sami Eria 2012 i Acknowledgements I would like to thank my Advisor, Robert B. McMaster for guiding me through this doctoral dissertation. I would also like to thank my PhD Committee members, Francis Harvey (Geography), Steven Manson (Geography), and Mohamed Mokbel (Computer Science) for providing thoughtful comments and suggestions for improvement throughout the research process. I would also like to acknowledge the financial support of the Interdisciplinary Center for the Study of Global Change (ICGC) at the University of Minnesota, and the Compton Foundation. In addition, I would like to thank the Department of Geography, University of Minnesota, for providing me with a Graduate Teaching and Research Assistantship through the course of my doctoral studies. ii Dedication This doctoral dissertation is dedicated to my father, Dr. M.A.F. Eria, and mother, Rehana (R.I.P). iii Abstract Geographical Information Systems (GIS) were first introduced in the developing world by the United Nations Environmental Programme (UNEP) to promote environmental conservation activities in in the mid-80s. There have been far fewer studies on the diffusion of GIS in developing countries as compared to industrialized countries. All diffusion studies use a common framework, the Diffusion of Innovations theory. Most of the studies carried out in developing countries have focused on the public sector, and ignored the nongovernmental organizations (NGO), International Organizations (IO), academic, and private sectors. What is the current state of GIS in all these sectors in developing countries? A holistic understanding of the current state of GIS is important to governments in developing countries for the planning of geospatial infrastructure. It is also important to the World Bank, United Nations, and international development agencies, such as the US Agency for International Development (USAID) for planning effective policies on technology transfer to developing countries. The research in this dissertation is based on a case study of GIS diffusion in Uganda. I argue that there are a number of political, social, economic, and technological factors that have led to varying degrees of GIS diffusion in Uganda’s public, academic, NGO, IO and private sectors. I also argue that future trends in GIS in developing countries point to a shift from desktop to mobile platforms because of the ubiquity and pervasiveness of mobile phones. What is the potential for mobile location-based services (LBS) as alternative tools for promoting spatial thinking and spatial awareness, and for supporting spatial decision making in civil society? A mixed methods methodology is employed in this research, and facets of iv diffusion of innovations theory are tested. Further, the research analysis utilizes the five perspectives of GIS and Society discourse to investigate the evolution and current state of GIS in Uganda, and the potential for LBS. v Table of Contents Acknowledgements........................................................................................................................... i Dedication ........................................................................................................................................ ii Abstract ........................................................................................................................................... iii Table of Contents ............................................................................................................................. v List of Tables ..................................................................................................................................xiii List of Figures ................................................................................................................................. xiv Chapter 1 : Introduction .................................................................................................................. 1 1.1 Motivation and Problem Definition ....................................................................................... 1 1.2 Aim, Foci, Research Objectives and Research Questions ...................................................... 3 1.3 Research Contributions ......................................................................................................... 5 1.4 Importance/Significance of Research .................................................................................... 7 1.4.1 “Hardness” of Research ................................................................................................. 9 1.5 Justification of Study Site..................................................................................................... 11 1.5.1 Geography .................................................................................................................... 12 1.5.2 History .......................................................................................................................... 15 1.5.3 Economy ....................................................................................................................... 17 1.5.4 Political Structure ......................................................................................................... 18 1.5.5 Role of GIS and Spatial Data in Uganda........................................................................ 19 1.6 Approach to Research ......................................................................................................... 20 1.6.1 Description of Methods................................................................................................ 20 vi 1.6.2 Novelty of Approach to Research ................................................................................ 23 1.6.3 Advantages of the Mixed Methods Approach ............................................................. 24 1.7 Potential Biases and Limitations of Research ...................................................................... 25 1.7.1 Case Study Approach.................................................................................................... 25 1.7.2 Mixed Methods ............................................................................................................ 26 1.7.3 Limitations of Diffusion Theory .................................................................................... 27 1.8 Overview of the Literature .................................................................................................. 30 1.9 Definition of Terminologies ................................................................................................. 33 1.10 Structure of Dissertation ................................................................................................... 35 Chapter 2 : Literature Review ........................................................................................................ 36 2.1 Introduction ......................................................................................................................... 36 2.2 GIS - Geographical Information Systems or Geographic Information Science? .................. 37 2.3 Situating GIS within Information Systems Literature .......................................................... 38 2.4 Theories for Understanding the Acceptance and Use of Information Systems .................. 38 2.4.1 The Technology Acceptance Model ............................................................................. 40 2.4.2 Social Cognitive Theory ................................................................................................ 41 2.4.3 Unified Theory of Acceptance and Use of Technology ................................................ 42 2.5 Theories for Understanding the Diffusion of Technology ................................................... 43 2.5.1 Diffusion of Innovations ............................................................................................... 44 2.6 Understanding the Evolution and State of GIS as the Diffusion of Innovations .................. 57 2.6.1 Genesis of GIS ............................................................................................................... 58 2.6.2 GIS Diffusion in North America, Europe and Australia ................................................. 60 2.6.3 GIS Diffusion in Africa, Asia and South America .......................................................... 62 2.6.4 Evolution of GIS in the Developing World with a Focus on Africa ............................... 70 vii 2.7 GIS and Society .................................................................................................................... 85 2.7.1 History of GIS and Society ............................................................................................ 86 2.7.2 The Five Perspectives of GIS and Society ..................................................................... 89 2.7.3 Relevance of GIS and Society to Developing Countries with a focus on Africa ........... 95 2.8 ICT, Mobile Phones and Location-based Services ............................................................... 96 2.8.1 LBS Technology ............................................................................................................. 97 2.8.2 Applications of LBS in Developed Countries .............................................................. 101 2.8.3 Applications of LBS in Developing Countries ............................................................. 101 2.9 Conclusion ......................................................................................................................... 103 Chapter 3 : Research Methodology ............................................................................................. 104 3.1 Introduction ....................................................................................................................... 104 3.2 Philosophical Perspectives on Research ............................................................................ 104 3.2.1 Qualitative Research Paradigm .................................................................................. 106 3.2.2 Quantitative Research Paradigm ............................................................................... 110 3.2.3 Mixed Methods Research Paradigm .......................................................................... 111 3.3 Choosing Mixed Methods Research .................................................................................. 112 3.4 Methods Used ................................................................................................................... 113 3.4.1 Semi-structured In-depth Personal Interviews .......................................................... 113 3.4.2 Survey Questionnaires ............................................................................................... 120 3.4.3 Focus Group and Case Study ...................................................................................... 123 3.4.4 Workshop ................................................................................................................... 124 3.5 Data Analysis...................................................................................................................... 124 3.6 Strengths, Weaknesses of and Potential Biases of Methods Used ................................... 127 3.7 Conclusion ......................................................................................................................... 129 viii Chapter 4 : Evolution of GIS in Developing Countries – A Case Study of Uganda ....................... 130 4.1 Introduction ....................................................................................................................... 130 4.2 The Historical Development of Mapping Science in Uganda ............................................ 131 4.3 The Evolution of GIS in Uganda ......................................................................................... 139 4.3.1 Introduction of GIS at Public, Academic, Non-governmental and Private Sector Institutions .......................................................................................................................... 139 4.3.2 Training and Capacity Building in GIS in Uganda........................................................ 152 4.4 Analysis .............................................................................................................................. 156 4.4.1 Diffusion of Innovations perspective ......................................................................... 156 4.4.2 GIS and Society in Uganda: An Intellectual History Perspective ................................ 166 4.5 Conclusion ......................................................................................................................... 169 Chapter 5 : Diffusion of GIS in Developing Countries – A Case study of Uganda’s Public and Academic Sectors ......................................................................................................................... 171 5.1 Introduction ....................................................................................................................... 171 5.2 The Public Sector and Academic Sectors ........................................................................... 172 5.2.1 Public Sector ............................................................................................................... 172 5.2.2 Academic Sector ......................................................................................................... 178 5.3 Results ............................................................................................................................... 182 5.3.1 The Public Sector ........................................................................................................ 183 5.3.2 The Academic Sector .................................................................................................. 229 5.4 The State of GIS in Uganda: A Diffusion of Innovations Analysis ...................................... 249 5.4.1 The Innovation: Perceived Attributes ........................................................................ 250 5.4.2 Communication Channels .......................................................................................... 257 5.4.3 Time ............................................................................................................................ 258 5.4.4 Social System .............................................................................................................. 267 ix 5.4.5 Criticism of Diffusion Research .................................................................................. 278 5.5 The State of GIS in Uganda: A GIS and Society Perspective ............................................. 283 5.5.1 Critical Social Theory Perspective .............................................................................. 283 5.5.2 Institutional Perspective ............................................................................................ 288 5.5.3 Legal and Ethical Perspective ..................................................................................... 291 5.5.4 Intellectual History Perspective ................................................................................. 292 5.5.5 Public Participation Perspective ................................................................................. 296 5.6 Conclusions ........................................................................................................................ 299 Chapter 6 : Diffusion of GIS in Developing Countries - A Case study of Uganda’s Non- governmental Organization, International Organization and Private Sectors ............................ 301 6.1 Introduction ....................................................................................................................... 301 6.2 The NGO, IO and Private Sectors ....................................................................................... 302 6.2.1 NGO Sector ................................................................................................................. 302 6.2.2 IO Sector ..................................................................................................................... 307 6.2.3 Private Sector ............................................................................................................. 309 6.3 Results ............................................................................................................................... 315 6.3.1 NGO Sector ................................................................................................................. 315 6.3.2 IO Sector ..................................................................................................................... 325 6.3.3 Private Sector ............................................................................................................. 338 6.4 The State of GIS in Uganda: A Diffusion of Innovations Analysis ...................................... 355 6.4.1 The Innovation: GIS .................................................................................................... 355 6.4.2 Communication Channels .......................................................................................... 360 6.4.3 Time ............................................................................................................................ 364 6.4.4 Social System .............................................................................................................. 366 6.5 The State of GIS in Uganda: A GIS and Society Perspective ............................................. 370 x 6.5.1 Critical Social Theory Perspective .............................................................................. 370 6.5.2 Legal and Ethical Perspective ..................................................................................... 373 6.5.3 Institutional Perspective ............................................................................................ 376 6.5.4 Intellectual History Perspective ................................................................................. 379 6.6 Conclusions ........................................................................................................................ 380 Chapter 7 : The Future of GIS in Developing Countries: The Potential for Mobile Location-based Services in Uganda ....................................................................................................................... 382 7.1 Introduction ....................................................................................................................... 382 7.2 Results and Diffusion Analysis ........................................................................................... 387 7.2.1 Mobile Phone Penetration ......................................................................................... 388 7.2.2 Access to ICT ............................................................................................................... 401 7.2.3 Innovative Uses of Mobile Phones for Spatial Decision Making ................................ 414 7.2.4 Potential for Location-Based Services in Supporting Spatial Thinking ....................... 429 7.2.5 Location in Mobile Applications and the Link to GIS .................................................. 438 7.2.6 The Role of Mobile Telecom Service Providers .......................................................... 444 7.2.7 The Role of Content Providers ................................................................................... 457 7.2.8 Challenges to Location-based Services in Uganda ..................................................... 470 7.2.9 Weaknesses of the Analysis ....................................................................................... 475 7.3 Diffusion Analysis of the Google SMS Application in Wainah Village ............................... 476 7.3.1 The Innovation ........................................................................................................... 477 7.3.2 Communication Channels .......................................................................................... 480 7.3.3 Time ............................................................................................................................ 482 7.3.4 Social System .............................................................................................................. 483 7.4 Conclusion ......................................................................................................................... 485 Chapter 8 : Summary and Conclusions ........................................................................................ 487 xi 8.1 Aim and Research Questions ............................................................................................. 487 8.2 Summary of Research Findings.......................................................................................... 488 Research Question 1: How did GIS evolve as a technology in developing countries? ........ 488 Research Question 2: What is the current state of GIS technology in the various sectors of the economy in developing countries? ............................................................................... 493 Research Question 3: What is the future potential for emerging mobile-device based geospatial technologies, known as location-based services (LBS), as alternative tools to GIS for spatial decision making in developing countries? ......................................................... 496 8.3 Contributions and Implications ......................................................................................... 500 8.4 Suggestions for Further Research ...................................................................................... 501 References ................................................................................................................................... 503 Appendix ...................................................................................................................................... 550 Appendix A: Institutions Interviewed ...................................................................................... 551 Appendix B: Interview Questions - Academic Sector .............................................................. 554 Appendix C: Interview Questions – Public, Private, NGO, IO Sectors ..................................... 557 Appendix D: Interview Questions – Telecom Industry ............................................................ 560 Appendix E: Survey Questionnaire – Mobile Phone Usage ..................................................... 562 Appendix F: Social Consent Form ............................................................................................ 566 Appendix G: Situating GIS within Information Systems Literature ......................................... 568 Appendix H: Theories for Understanding the Acceptance and Use of Information Systems . 575 1. Technology Acceptance Model (TAM) ............................................................................ 575 2. Social Cognitive Theory (SCT) .......................................................................................... 577 3. Unified Theory of Acceptance and Use of Technology (UTAUT) ..................................... 579 Appendix I: Theories for Understanding the Diffusion of Technology .................................... 582 1. Technological Determinism ............................................................................................. 582 2. Economic Determinism ................................................................................................... 583 xii 3. Social Interactionism ....................................................................................................... 584 Appendix J: Philosophical Perspectives on Research .............................................................. 586 1. Ontology, Epistemology, Methodology and Method ...................................................... 586 2. Constructivist, Critical Theory, and Postmodernist Perspectives ................................... 588 3. Interpretive Perspectives, Symbolic Interactionism ....................................................... 589 4. Positivism ........................................................................................................................ 591 5. Postpositivism ................................................................................................................. 592 xiii List of Tables TABLE 2-1: REGIONAL CENTERS FOR REMOTE SENSING, PHOTOGRAMMETRY, SURVEYING AND GEODESY IN AFRICA, 1972-1998 .......................................................................................................................... 79 TABLE 4-1: INTRODUCTION OF GIS IN UGANDA’S PUBLIC, PRIVATE, NGO AND ACADEMIC SECTORS, 1989- 2002 ................................................................................................................................................... 143 TABLE 5-1: PUBLIC SECTOR INSTITUTIONS AT WHICH INTERVIEWS WERE PERFORMED ........................... 183 TABLE 5-2: PROJECT-DRIVEN GIS AT PUBLIC SECTOR INSTITUTIONS .......................................................... 188 TABLE 5-3: GIS USAGE AT PUBLIC SECTOR INSTITUTIONS INTERVIEWED ................................................... 193 TABLE 5-4: GIS DATA COLLECTION METHODS USED AT PUBLIC SECTOR INSTITUTIONS ............................ 201 TABLE 5-5: GIS EMPLOYEE TRAINING AT PUBLIC SECTOR INSTITUTIONS ................................................... 208 TABLE 5-6: GIS EDUCATION AT UNIVERSITY LEVEL ..................................................................................... 211 TABLE 5-7: ACADEMIC INSTITUTIONS AT WHICH INTERVIEWS WERE PERFORMED ................................... 230 TABLE 5-8: GIS AT PUBLIC UNIVERSITIES IN UGANDA ................................................................................. 233 TABLE 5-9: GIS SOFTWARE AT ACADEMIC SECTOR INSTITUTIONS ............................................................. 237 TABLE 6-1: NGO SECTOR INSTITUTIONS AT WHICH INTERVIEWS WERE PERFORMED ............................... 317 TABLE 6-2: GIS USAGE AT NGO SECTOR INSTITUTIONS .............................................................................. 317 TABLE 6-3 IO SECTOR INSTITUTIONS AT WHICH INTERVIEWS WERE PERFORMED .................................... 326 TABLE 6-4: GIS USAGE AT IO SECTOR INSTITUTIONS .................................................................................. 330 TABLE 6-5: PRIVATE SECTOR INSTITUTIONS AT WHICH INTERVIEWS WERE PERFORMED ......................... 339 TABLE 6-6: GIS USAGE AT PRIVATE SECTOR INSTITUTIONS ........................................................................ 345 TABLE 7-1: A COMPARISON OF TECHNOLOGICAL CAPABILITIES OF SMART PHONES OWNED BY RESPONDENTS BASED ON YEAR OF RELEASE .................................................................................... 401 TABLE 7-2: HARDWARE AND SOFTWARE OF COMPUTERS OWNED BY RESPONDENTS.............................. 404 TABLE 7-3: ACCESS OF SURVEY RESPONDENTS TO THE INTERNET/WEB .................................................... 405 TABLE 7-4: INTERNET SERVICE PROVIDERS USED BY THE SURVEY RESPONDENTS ..................................... 410 TABLE 7-5: REASONS FOR ACCESSING THE INTERNET/WEB ....................................................................... 411 TABLE 7-6: COST OF ACCESSING THE INTERNET PER MONTH FOR SURVEY RESPONDENTS ....................... 411 TABLE 7-7: OPINION OF SURVEY RESPONDENTS: IS THE COST OF ACCESSING THE INTERNET AFFORDABLE? ........................................................................................................................................................... 412 TABLE 7-8: MOBILE INTERNET USAGE TRENDS OF SURVEY RESPONDENTS ............................................... 413 TABLE 7-9: SPATIAL QUERIES PERFORMED BY SURVEY RESPONDENTS USING MOBILE PHONES .............. 415 TABLE 7-10: POPULAR MOBILE PHONE BASED APPLICATIONS ................................................................... 416 TABLE 7-11: SURVEY RESPONDENTS’ INTEREST IN FUTURE LBS APPLICATIONS FOR VARIOUS SPATIAL QUERIES ............................................................................................................................................. 431 TABLE 7-12: SURVEY RESPONDENTS THAT USE GOOGLE MAPS AND GOOGLE EARTH ............................... 432 TABLE 7-13: SURVEY RESPONDENTS’ SUBSCRIPTION TO VARIOUS MOBILE TELECOMMUNICATION SERVICE PROVIDERS ......................................................................................................................................... 447 TABLE 0-1: A CLASSIFICATION OF INFORMATION SYSTEMS; ...................................................................... 570 xiv List of Figures FIGURE 1-1: UGANDA, POLITICAL BOUNDARIES AT INDEPENDENCE, 1962 .................................................. 13 FIGURE 1-2: KAMPALA, A TERRESTRIAL VIEW ............................................................................................... 14 FIGURE 2-1: THE DIFFUSION PROCESS – THE S-SHAPED DIFFUSION CURVE ................................................. 46 FIGURE 2-2: THE FIRST COUNTRIES IN THE DEVELOPING WORLD TO RECEIVE GIS TECHNOLOGY, THROUGH UNEP-GRID ........................................................................................................................................... 76 FIGURE 2-3: REGIONAL CENTERS FOR SPACE SCIENCE AND TECHNOLOGY EDUCATION IN DEVELOPING COUNTRIES .......................................................................................................................................... 80 FIGURE 2-4: SIMPLE LBS ARCHITECTURE ..................................................................................................... 100 FIGURE 3-1: THE RELATIVE NUMBER OF INSTITUTIONS CONTACTED FOR INTERVIEWS, CATEGORIZED BY SECTOR ............................................................................................................................................... 116 FIGURE 3-2: INSTITUTIONS THAT PARTICIPATED IN PERSONAL INTERVIEWS, CATEGORIZED BY SECTOR .. 117 FIGURE 3-3: HIERARCHICAL INSTITUTIONAL STRUCTURE OF THE MINISTRY OF WATER AND THE ENVIRONMENT, UGANDA, SHOWING DIRECTORATES, DEPARTMENTS AND PARASTATAL AGENCIES GOVERNED BY THE MINISTRY ............................................................................................................ 118 FIGURE 3-4: MOBILE PHONE QUESTIONNAIRE SURVEY RESPONDENTS BY PROFESSION ........................... 127 FIGURE 5-1: S-SHAPED CURVE: DIFFUSION OF GIS IN UGANDA’S PUBLIC AND ACADEMIC SECTORS, 1986- 2010 ................................................................................................................................................... 267 FIGURE 6-1: INTERVENTIONS OF THE USAID FUNDED SPRING PROJECT IN NORTHERN UGANDA ............. 327 FIGURE 6-2: A MAP SHOWING CONFLICT AFFECTED AREAS IN NORTHERN UGANDA ................................ 331 FIGURE 6-3: TRAVEL MAP OF SOUTH EASTERN KAMPALA .......................................................................... 341 FIGURE 6-4: HIGHWAYS IN UGANDA MANAGED BY THE UGANDA NATIONAL ROADS AUTHORITY ........... 342 FIGURE 6-5: DOWNTOWN KAMPALA TOURIST MAP .................................................................................. 343 FIGURE 6-6: S-SHAPED CURVE: RATE OF ADOPTION OF GIS IN THE NGO, IO AND PRIVATE SECTORS IN UGANDA, 1986-2010 ......................................................................................................................... 365 FIGURE 7-1: LBS CONCEPT ........................................................................................................................... 385 FIGURE 7-2: SURVEY RESPONDENTS BY PROFESSION ................................................................................. 389 FIGURE 7-3: AGE RANGE OF SURVEY RESPONDENTS .................................................................................. 390 FIGURE 7-4: EDUCATION LEVEL OF SURVEY RESPONDENTS ....................................................................... 390 FIGURE 7-5: DISTRICTS/TOWNS OF RESIDENCE OF SURVEY RESPONDENTS ............................................... 391 FIGURE 7-6: NUMBER OF MOBILE PHONES OWNED BY EACH SURVEY RESPONDENT ............................... 393 FIGURE 7-7: MOBILE PHONE BRANDS USED BY SURVEY RESPONDENTS .................................................... 394 FIGURE 7-8: YEAR OF RELEASE OF MOBILE PHONE MODELS OWNED BY SURVEY RESPONDENTS ............. 395 FIGURE 7-9: DIFFUSION CURVE OF MOBILE PHONE ADOPTION IN UGANDA ............................................. 397 FIGURE 7-10: TECHNOLOGICAL CAPABILITIES OF MOBILE PHONES OWNED BY SURVEY RESPONDENTS .. 399 FIGURE 7-11: PORTABLE USB WIRELESS 3G MODEM (“INTERNET EVERYWHERE”) FROM ORANGE TELECOM, UGANDA ........................................................................................................................... 406 FIGURE 7-12: AIRTEL TELECOMMUNICATION: 2G AND 3G COVERAGE IN UGANDA .................................. 408 FIGURE 7-13: MTN GEOGRAPHICAL NETWORK COVERAGE OF UGANDA AS OF JULY 2009 ....................... 409 FIGURE 7-14: MTN MOBILE MONEY KIOSK ................................................................................................. 417 FIGURE 7-15: MTN VILLAGE PHONE ............................................................................................................ 418 FIGURE 7-16: GOOGLE SMS SEARCH ........................................................................................................... 420 FIGURE 7-17: GOOGLE SMS POSTER ........................................................................................................... 423 FIGURE 7-18: GOOGLE SMS TIPS ................................................................................................................. 424 FIGURE 7-19: GOOGLE SMS – FARMER’S FRIEND ........................................................................................ 424 FIGURE 7-20: GOOGLE SMS – FARMER’S FRIEND APPLICATION – ARCHITECTURE ..................................... 425 xv FIGURE 7-21: SMALL RESTAURANT BUSINESS: OWNER RECEIVES BULK ORDERS FOR FOOD FROM CORPORATE CUSTOMERS ON HER MOBILE PHONE VIA SMS AND DELIVERS FOOD PACKAGES TO THEIR OFFICES BY BIKE TAXI, “BODA-BODA” ..................................................................................... 427 FIGURE 7-22: MOBILE TECHNOLOGIES: USER FRIENDLINESS AND DEVICE DEPENDENCY .......................... 435 FIGURE 7-23: FCIT’S OPENXDATA SCHOOL ATTENDANCE PROJECT ARCHITECTURE .................................. 443 FIGURE 7-24: INTERNET INFRASTRUCTURE OF ORANGE TELECOM ............................................................ 448 FIGURE 7-25: MOBILE COMMUNICATION IN THE LAKE VICTORIA RESCUE PROJECT’S MOBILE POSITIONING SYSTEM .............................................................................................................................................. 456 FIGURE 7-26: BROSDI STAFF IN MAYUGE DISTRICT THAT TOOK PART IN THE FOCUS GROUP DISCUSSION ........................................................................................................................................................... 462 FIGURE 7-27: BROSDI CENTER IN RURAL MAYUGE DISTRICT, WAINAH VILLAGE: SATELLITE BROADBAND INTERNET CONNECTION THROUGH AFSAT ....................................................................................... 463 FIGURE 7-28: INTERNET ROOM AT BROSDI CENTER IN MAYUGE DISTRICT, WAINAH VILLAGE, PROVIDES INTERNET ACCESS TO LOCAL FARMERS AND YOUTH ........................................................................ 463 FIGURE 7-29: T-MOBILE G1 GOOGLE PHONE USED BY BROSDI STAFF IN MAYUGE DISTRICT TO ACCESS THE WEB, AND THE GOOGLE SMS MOBILE APPLICATIONS ...................................................................... 467 FIGURE 0-1: RELATIONSHIP BETWEEN IT AND IS ........................................................................................ 568 FIGURE 0-2: PROCESS OF CONVERSION OF DATA INTO INFORMATION IN AN INFORMATION SYSTEM .... 569 FIGURE 0-3: THE PYRAMID MODEL CATEGORIZING INFORMATION SYSTEMS IN THE 1980S ..................... 571 FIGURE 0-4: TECHNOLOGY ACCEPTANCE MODEL ....................................................................................... 575 FIGURE 0-5: SOCIAL COGNITIVE THEORY – TRIADIC RECIPROCAL CAUSATION IN THE CAUSAL MODEL OF SCT ..................................................................................................................................................... 578 FIGURE 0-6: UNIVERSAL THEORY ON ACCEPTANCE AND USE OF TECHNOLOGY ........................................ 580 FIGURE 0-7: BASIC CONCEPTUAL FRAMEWORK UNDERLYING USER ACCEPTANCE MODELS ..................... 580 1 Chapter 1 : Introduction 1.1 Motivation and Problem Definition Geographical Information Systems (GIS) were first introduced in the developing world by the United Nations Environmental Programme (UNEP) to promote environmental conservation activities in in the mid-80s (Mooneyhan, 1998). There have been far fewer studies on the diffusion of GIS in developing countries as compared to industrialized countries. All diffusion studies use a common framework, the Diffusion of Innovations theory (Rogers, 1993, 1995, 2003). Examples of GIS diffusion research in developing countries include studies carried out in Botswana (Cavric, Nedović-Budić, & Ikgopoleng, 2003), Ghana (Karikari, Stillwell, & Carver, 2005), India (Barrett, Sahay, & Walsham, 2001; Puri & Sahay, 2004; Sahay, 1998), and Brazil (Borges & Sahay, 2000; Câmara, Fonseca, Monteiro, & Onsrud, 2006; Câmara, Fonseca, Onsrud, & Monteiro, 2004). Some studies have employed other theoretical frameworks, such as Actor Network Theory; for example, see the work of Noongo (2007) on GIS implementation in Namibia. However, each of these studies has focused on the public sector, and fallen short of addressing GIS diffusion in other sectors, namely the nongovernmental organizations (NGO), International Organizations (IO), academic, and private sectors. What is the current state of GIS in the public sector and other sectors in developing countries? An understanding of the current state of GIS in developing countries is important for governments in developing countries in planning of geospatial infrastructure. It is also of interest to the World Bank, United Nations, and international development agencies, such 2 as the US Agency for International Development (USAID) for planning effective policies on technology transfer to developing countries. In this dissertation research, I examine the evolution and current state of GIS in developing countries based on a case study of one such developing country, Uganda. One of the objectives in the 2010-2015 National Development Plan (NDP) of Uganda is to improve physical planning infrastructure in the country by establishing a national GIS center and a national spatial database (GoU, 2010). I argue that there are a number of political, social, economic, and technological factors that have led to varying degrees of GIS diffusion in Uganda’s public, academic, NGO, IO and private sectors. Any future plans to develop national GIS infrastructure need to carefully consider these factors. I also argue that future trends of GIS in developing countries point to a shift from desktop to mobile platforms because of the pervasiveness and ubiquity of mobile phones. What is the potential for mobile location-based services (LBS) as alternative tools for promoting spatial thinking and spatial awareness, and for supporting spatial decision making in civil society? Mixed methods analysis is employed in this dissertation research. The research analysis tests facets of Diffusion of Innovations theory, and employs the five perspectives of GIS and Society discourse to investigate the evolution and current state of GIS in Uganda, and the potential for mobile location-based services (LBS). 3 1.2 Aim, Foci, Research Objectives and Research Questions The aim of this research is to critically examine the diffusion of GIS in developing countries, and to assess the impact of GIS technology on Society. Uganda is used as a case study to draw broader implications for developing countries in general. The research focuses on investigating the evolution, current state, and future of GIS in Uganda. My research has three main objectives. First, I investigate the evolution of GIS in Uganda. GIS is inherently linked to the constituent disciplines of mapping science, that is, land surveying, photogrammetry, cartography and remote sensing. In Uganda, land surveying, cartography, and photogrammetry were introduced by the colonial British government between 1900 and 1962, thus, I investigate the history of these disciplines as precursors to GIS. This is followed by an investigation of the evolution of the mapping science disciplines between 1962 to the present, relative to the entry point and evolution of GIS as a discipline. This understanding of the chronological development of the mapping science disciplines as precursors to the introduction of GIS is important for a better understanding of the current level of GIS diffusion in Uganda, which then leads to the second objective of this research. Second, I investigate the current state of GIS in Uganda. In the context of this research, the “state” of GIS refers to the penetration level of the technology into the public, private, non-governmental organizations (NGO), International Organizations (IOs), and academic sectors. I also examine the impacts of GIS on society in Uganda based on the five perspectives of GIS and Society discourse. An understanding of the current state of GIS is important to the Government of Uganda for future planning of a national GIS center 4 and national spatial database. It is also important to the World Bank, United Nations, and international development agencies, such as the US Agency for International Development (USAID) for planning effective policies on technology transfer to developing countries. Third, I investigate the potential for mobile LBS technology as alternative tools to GIS for promoting spatial thinking and spatial awareness, and for supporting spatial decision making in civil society. I assess the Information and Communication Technology (ICT) and mobile telecommunication infrastructure in Uganda, and then examine the political, social, economic, institutional and technological challenges to the development and use of LBS technology. LBS is dependent upon ICT and mobile telecommunication infrastructure. An understanding of the current state of the infrastructure, and the challenges to the development of LBS technology would allow for an assessment of the potential for LBS as alternative tools to GIS for promoting spatial thinking, spatial awareness, and for supporting spatial decision making in Uganda. There are three core research questions investigated in this research project. One, how did GIS evolve as a technology in developing countries? Two, what is the current state of GIS technology in the various sectors of the economy in developing countries? Three, what is the future potential for emerging mobile-device based geospatial technologies, known as location-based services, as alternative tools to GIS for spatial decision making in developing countries? Next, I discuss the contribution of this dissertation research to academia and, broadly, to society. 5 1.3 Research Contributions On the academic level, this research contributes to GIS literature based on Diffusion of Innovations theory (also known as GIS Diffusion literature), and also to the body of literature known as GIS and Society, specifically in developing countries. A contribution is also made to the literature on Information and Communication Technology for Development (ICT4D) and Mobile for Development (M4D). Facets of Diffusion of Innovations theory (Rogers, 1995, 2003) were tested, and the five perspectives of GIS and society (McMaster & Harvey, 2010; Nyerges, McMaster, & Couclelis, 2011) were used as a framework for examining the impact of GIS on society in developing countries, based on a case study of Uganda. A lot of the literature on GIS diffusion has a Western focus, with only a limited number of publications on GIS diffusion in the developing world. A lot of the literature on GIS diffusion in industrialized countries started emerging in the mid-80s to early-90s, for example, see Arnaud 1993; Masser 1993; H. J. Onsrud 1991; Wiggins 1993. This literature focused on GIS diffusion in organizations in Europe, North America and Australia, especially in local government and public sector agencies. In the past two decades, a studies have emerged on GIS diffusion in developing countries, such as Botswana, Namibia, Egypt, China, Indonesia, and Brazil, for example, see Cavric, Nedović-Budić, and Ikgopoleng 2003; Noongo 2007; Salem 1994; Barrett, Sahay, and Walsham 2001; Sahay and Walsham 1996; Puri and Sahay 2004; Yue et al. 1991; Shupeng 1987; Sipe and Dale 2003; Borges and Sahay 2000. However, most of these have focused on GIS in the public and academic sectors, ignoring the private, NGO, and 6 IO sectors. This dissertation research complements this past research by contributing the NGO, IO and private sector perspectives on GIS diffusion in developing countries, in addition to the public and academic sectors. Literature on GIS and Society also tends to have a Western focus, although some parallels can be drawn with the developing world on similar issues in developed countries, such as issues regarding indigenous peoples (for example, see Laituri 2011) and marginalized groups (for example, see Harjo 2006; Leitner et al. 1998). Nonetheless, there is a gap in the literature on critical issues related specifically to developing countries. Much as some work has been done in Public Participatory GIS (see Kyem 2004a; Kyem and Saku 2009; Rambaldi, Mccall, and Weiner 2006; Harris et al. 1995; Weiner and Harris 1999), and more generally in GIS and Society (see Koti 2004), a lot more research is needed to fully understand the impact of GIS on society in the developing world. This research offers a developing countries’ perspective to each of the five major foci (McMaster & Harvey, 2010; Nyerges et al., 2011) of GIS and Society. M4D and ICT4D is a growing body of literature that has garnered a lot of interest in scholarly work especially over the past ten years with the exponential diffusion of ICT and mobile telecommunication technology in the developing world starting in the late 1990s. The impacts of mobile phones, and Information and Communication Technology (ICT) on the livelihoods of local citizens in the developing world have been widely discussed in the literature, for example, see Donner 2010; Aker and Mbiti 2010; James and Versteeg 2007; Furuholt and Matotay 2011; Akpan-obong et al. 2009; Masuki et al. 2010. However, few studies have focused on location-specific and context-aware 7 applications, such as location-based services (LBS) as tools for enhancing spatial thinking and facilitating spatial decision making. Mobile phone technology and LBS can be viewed as an alternative to GIS in developing countries, which tend to be infrastructure poor environments (Cartwright, 1993), for facilitating greater citizen agency (Unwin, 2010; Zanello & Maassen, 2011)and participatory inclusiveness in society. This dissertation research complements and extends existing literature on the potential for LBS in developing countries. Next, I discuss the significance of this research. 1.4 Importance/Significance of Research My research directly addresses a need for GIS and spatial database management systems in Uganda, as specified in a recent government document. In April 2010, the Government of Uganda published the National Development Plan 2010-2015 (GoU, 2010). The section on Physical Planning states the following: Uganda has 112 gazetted towns of which 93 have up-to-date structure plans. However, a big number of these towns have not translated the bulk of their Structure Plans into implementable detailed physical development plans. In addition, Uganda lacks a coherent rural land use plan. The country, therefore, continues to experience haphazard developments in both urban and rural areas. (GoU 2010, 167) The lack of effective physical planning implementation is due in part to the “lack of up- to-date planning information, including topographic maps, cadastre information and land 8 tenure maps, among others” (GoU 2010, 168). One of the strategies proposed in the National Development Plan to address the problem of poor physical planning in the country is for the central and local governments to “establish a land use (physical planning) database and to computerize physical planning operations” (GoU 2010, 168). The proposed intervention specifically prescribes the introduction of GIS technology into the institutional structure of central and local government, and for the training of staff in geospatial technology. The exact statement on this issue is as follows: i) Establish and operationalize an appropriate institutional structure at the Local Government level. ii) Train all Central Government physical planning staff in advanced Geographical Information System (GIS) skills. iii) Roll out GIS training to all district and urban Local Governments to impart adequate GIS skills for all relevant Local Government technical staff. iv) Establish a GIS centre and a national spatial database to adequately back up land use data repository and work stations for all trained Planners. v) Support relevant security and law and order organs to build spatial databases linked (real time) to national spatial records of street layouts, utility maps, addresses, architectural plans, land use and ownership among others. (GoU 2010, 169) 9 The document also expresses the government’s desire to increase participation of marginalized groups in the physical planning process to “ensure increased implementation and public acceptance” (GoU 2010, 169) of physical plans. The results of my research are important and significant in that they directly address the issues and problems that implementers of GIS in the Central and Local governments will likely face in the next five years. An understanding of the evolution and current state of GIS will be useful in informing decisions about how to move forward in establishing a GIS center and a national spatial database, institutionalizing GIS into the public sector, and training of physical planning and other technical staff in GIS. An understanding of the potential for leveraging mobile phone technology for encouraging societal participation in the spatial decision making process would also be very useful as the government looks to the future and the role of ubiquitous mobile technology in civil society with regard to spatial awareness creation in the citizenry. Next, I discuss the “hardness” of this research. 1.4.1 “Hardness” of Research This research is “hard” on many levels. First, to trace the evolution, and establish the current state of GIS in developing countries in general, and Uganda in particular, is a nontrivial engagement mainly due to the complex societal nature of the problem. Such a research project in the social sciences would produce the most detailed and accurate results through carefully structured personal interviews with research subjects. The specific media used to carry out interviews depends on the infrastructure of the country. On the subject of GIS diffusion, telephone and mail surveys have been used by 10 researchers in Europe and in the United States (for instance, Masser and Campbell 1995; Cambell 1993). However, in a developing country like Uganda, the mail and telephone infrastructure do not support such techniques. As a result, I visited all research subjects in person and carried out personal interviews, recording the sessions using a digital voice recorder. To avoid bias in interpretation, all interviews were manually transcribed word- for-word. Second, in as far as the interview questions were concerned; respondents were not comfortable answering some questions on issues of institutional politics of power, corruption and dissent. The personal responses, facial expressions, sighs, and change in tone of voice made it possible for me to get a sense of the friction involved in discussing these sensitive issues. A telephone or mail in survey would definitely not have allowed for the detection of these nuances in the responses. Third, to be able to effectively analyze the transcribed data, diffusion of innovations theory was employed cautiously. Notwithstanding the broad applicability of this theory, which could apply to innovations ranging in diversity from seatbelts in cars to Nano technology, testing facets of the theory with respect to GIS technology requires a critical examination of social issues, not from the perspective of the individual as was originally intended, but from that of the organization. Moreover, the theory has to be used with caution in the developing countries context because of biases associated with the Western model of development based on capital-intensive, labor-saving technology, which is the complete opposite of the situation in the developing world (Rogers, 2003). See section 1.7 for a detailed discussion on potential biases of diffusion theory. 11 Next, I explain why I chose Uganda as my study site. 1.5 Justification of Study Site Much as my research focuses on GIS in developing countries, I used Uganda, a country in East Africa equivalent in surface area to the state of Oregon, and almost equivalent in terms of total population to California. The reasons for choosing Uganda were three – first, I am a native of Uganda; I know the culture, local languages and dialects; I am familiar with the institutional set up in the public sector and have a preexisting personal network of contacts in the realm of geospatial science in academia and the public sector which I built during my undergraduate education at Makerere University Kampala. Second, Uganda is a developing country with a lot of similarities to other developing countries around the world. The socio-economic, political, and technological challenges facing the country are very similar to those you might find in Nigeria, Zimbabwe, Honduras, Peru, Haiti, and Vietnam. In terms of the politics, Uganda, like many other countries in the region, obtained independence in the early sixties, followed by a period of political instability and civil war, in turn followed by a period of peace, reconciliation and development (GoU and UNFPA 2007). In terms of the economy, like many other developing countries, the majority of Uganda’s population earns its income from subsistence agriculture; 42% to be exact (UBOS, 2010), and the national average monthly household income in 2010 was approximately $120 (UBOS, 2010). The public infrastructure in the country is not as developed as it is in the developed countries, but meets the basic requirements for trade and industry to develop (Ayogu, 1999). In terms of the social context, the demographics are composed of a multi ethnic heterogeneous 12 society, with increasing migration from rural to urban areas (Nyakana, Sengendo, & Lwasa, 2007) in search of service oriented employment. Third, Uganda has at least experienced some level of GIS penetration based on my previous knowledge about the country, which makes it a suitable candidate for a deeper investigation. I first encountered GIS in the country during my undergraduate studies in Kampala between 1997and 2001. After graduation, I worked with land surveying consultancies and civil engineering contractors for two years, and noticed the pervasiveness of Computer Aided Design (CAD) software in the construction industry. I further noticed that only a handful of institutions, such as the National Environmental Management Authority (NEMA) and the Department of Surveys and Mapping (the equivalent of the US Geological Survey), used GIS at that time. Of course, the latter are public sector institutions, which begs the question, has GIS penetrated the private and NGO sector at all. Uganda, one of three East African nations, was chosen as the study site for this dissertation research because the socio-economic, political and historical context of this country is representative of typical developing countries in the world. An overview of the geography, history, economy and politics of Uganda follows. 1.5.1 Geography Uganda is a located in the east central part of Africa, although geo-politically categorized as part of East Africa. It is a land locked country bordered by Kenya to the east, South Sudan to the north, DR Congo to the West, and Tanzania and Rwanda to the south. It is located between 1° 29’ South and 4° 12’ North latitude, 29° 34 East and 35° 0’ East 13 longitude (UBOS 2007). One third of the country lies south, and two thirds north of the Equator. The country’s territory covers an area of 241,038 square kilometers (91,344 square miles), of which the land area covers 197,323 square kilometers (UBOS 2007), an area slightly larger than the state of Minnesota, and almost equivalent to the State of Oregon. Figure 1-1: Uganda, Political Boundaries at Independence, 1962 14 Its territory includes almost half of Lake Victoria, also indigenously known as Lake Nalubaale, the second largest fresh water lake in the world in terms of total surface area (EB, 2012), and source of the River Nile which flows north to the Mediterranean Sea. The territory includes four other smaller lakes, Albert, Kyoga, Edward, and George, and the area is categorized as part of the African Great Lakes region. Figure 1-2: Kampala, a terrestrial view Uganda’s terrain is largely hilly in the south, gradually undulating to the north, and for the most part a highland plateau at an average height of 1200 meters (4000 ft.) above sea level, sitting in between the two arms of the Great African Rift valley formed through geological faulting (UgandaPeople, 2011). The terrain is characterized by a number of volcanic and block mountains, including Mount Elgon (peak: 4324 meters) and the Ruwenzori mountain ranges (peak: 5113 meters) (UgandaPeople, 2011). The vegetation is predominantly grassland Savannah, with about 20% of the surface area covered by thick tropical forests. The climate is tropical in nature, with two rainy and two dry seasons per year (UgandaPeople, 2011). 15 Uganda’s population is approximately 32 million people (GoU & UNFPA, 2010). The Capital City, Kampala, has a population of about 1.7 million people (GoU & UNFPA, 2010). The country’s demographics are characterized by an ethnically diverse population consisting of three main African ethnic groups based on similarity of language spoken: Bantu, Nilotic and Central Sudanic. Of these, the Bantu are predominant. Each ethnic group is composed of several tribes. The Baganda tribe is the largest in the country, and falls under the Bantu ethnic group. It constitutes about 18% of the total population of Uganda (UBOS, 2006). In fact, Uganda gets its name from the pre-colonial kingdom of Buganda that dominated the sub region economically and politically in the late 1800s. The next largest Bantu tribes include the Banyankole (members of the current ruling party are mostly from this tribe), Basoga, Bakiga, Bafumbira, Banyoro, Batooro, and Bagisu, respectively (UBOS, 2006). Among the Nilotic ethnic group are tribes such as the Teso, Langi, and Acholi, while the Central Sudanic tribes include the Lugbara and Madi. There are a number of other smaller tribes in each of the three ethnic groups that all speak distinct languages, sharing some similarities only with those in one of the three ethnic groups (UBOS, 2006). The official language of Uganda is English, the lingua franca Swahili, although, for all practical purposes, it is arguably Luganda (the language of the Baganda tribe) (UBOS, 2006). 1.5.2 History Present day Uganda was historically organized by Kingdom. The most prominent of these kingdoms in the mid-19 th Century was the kingdom of Buganda, although there were three other Bantu speaking kingdoms neighboring to the west, Bunyoro, Tooro, and 16 Ankole (Lwanga-Lunyiigo, 1987). To the north, the Nilotic peoples were organized in smaller units, chiefdoms, and as a result did not feature as prominently as the Bantu kingdoms. Trade in those days was mainly among neighboring tribes. The first outsiders to arrive were the Arabs in the 1840s, and they traded with the African tribes in ivory and slaves. The first European to arrive was John Hannington Speke in 1862, a British explorer, who met “Kabaka” (King) Mutesa I of Buganda Kingdom. Speke was followed by James Augustus Grant and Henry Morton Stanley in 1865, also British explorers. These explorers were followed by Christian missionaries of the Protestant (Anglican) and Catholic denominations, though Islam was already being practiced by many people in the Kingdom due to their prior contact with the Arabs. In 1890, during the scramble for Africa, Britain and Germany signed an agreement, in which Uganda came under the British sphere of influence. In 1894, present day Uganda was officially declared as a British Protectorate. This was the official colonization of present day Uganda till it gained its independence in 1962. (Batungi, 2008; Churchill, 1908; Gray & Peters, 1960; Lwanga-Lunyiigo, 1987; McGrath, 1976; Nkurunziza, 2006) After independence, a streak of regime changes and coups followed. The first president Sir Edward Mutesa II of the Buganda tribe was toppled by the Prime Minister, Milton Obote of the Langi tribe in 1965, who was in turn toppled by Iddi Amin in 1971 in a bloodless coup. Amin was infamous for his brutal and tyrannical rule that lasted almost a decade, and included the expulsion of all Ugandan Asians (of south Asian descent) and non-Africans from the country. He was overthrown in 1979 with the help of the Tanzanian army and Ugandan dissidents, including the current president, Yoweri 17 Museveni. Milton Obote took power for a second time in 1980, and was toppled by Tito Okello of the Acholi tribe in 1985; who was in turn overthrown by Yoweri Kaguta Museveni of the Banyankole tribe in 1986. (Lwanga-Lunyiigo, 1987) Between independence in 1962 and Museveni’s takeover in 1986, Uganda was very unstable as a state, and experienced several civil wars and coups; however, from 1986 to date, it has enjoyed relative peace and prosperity under the current president. Between 1986 and 2006, a long insurgency gripped the northern part of the country, perpetrated by the Lord’s Resistance Army (LRA) of Joseph Kony, a bandit with no plausible political agenda other than he wanted to rule the country based on the ten commandments in the Bible. The LRA got a lot a media attention in the West because of the capture and engagement of child soldiers in the rebellion, and the brutality of the rebels toward their own tribal people. After several failed operations, this insurgency was finally crushed by President Museveni’s National Resistance Army (NRA) and the remnants of the LRA were driven into the neighboring DR Congo and Central African Republic forests in 2006. At the time of this dissertation, the US government has deployed a small contingent of US special troops to assist Ugandan troops pursue the leader of the Lord’s Resistance Army, Joseph Kony, who is alleged to be hiding in the forests of the Central African Republic. (GoU & UNFPA, 2010) 1.5.3 Economy Agriculture is the major backbone of Uganda’s economy with the export of coffee, tobacco and fish accounting for over 45% of the country’s export earnings (UBOS, 2006). Other cash crops include hides and skins, vanilla, vegetables, flowers, tobacco, 18 cotton, tea and maize (corn). Over 70% of Uganda’s population is employed in the agricultural sector (UBOS, 2006). Industry is underdeveloped due to a lack of appropriate supporting infrastructure (Hirabe, 2009), for instance, a lack of sufficient electricity on the national grid. A few of the existing industries include those involved in the production of soft drinks, alcoholic beverages, steel, plastics, soap, sugar, and construction materials. There also exist processing plants for diary, poultry, and beef products. Local industrial products serve local markets, for the most part. The extractive industry has picked up in the past decade, especially with the recent discovery of crude oil (NAPE, 2009; Tuhumwire, 2009). Traditionally, Uganda’s major extraction industries have included copper, limestone, beryllium, salt, and gold. The fishing industry is also very important, however, it is mostly engaged in by local fishermen at subsistence scale, and petty trade. Uganda’s GDP for 2010 was about $17 Billion (WorldBank, 2010). 1.5.4 Political Structure Uganda transitioned to a quasi-democracy after the arrival of the current president, Yoweri Kaguta Museveni in 1986, although it took 20 years for the first multi-party elections to be held, in 2006 (RTI, 2010). The country’s system of governance is based on the British system, and in fact, many of the laws are still based on the laws written by the colonial British government. Government consists of three branches, the Executive, Judiciary, and legislature (parliament). There have been several reports of poor governance, corruption, and poor service delivery (RTI, 2010) and as a result, Uganda’s political structure cannot be regarded as a 19 properly functioning democracy, and it appears that the current president has no plans of relinquishing power in the near future. This is common place in many developing countries, not only in Sub-Saharan Africa, but also in Asia, South America, and the Middle East and North Africa, (for example, see the Arab Spring articles by Hayes 2011; M. Taylor 2011; Warf 2011; Snider & Faris 2011). 1.5.5 Role of GIS and Spatial Data in Uganda Uganda has enormous potential in terms of its natural resources. With the discovery of oil (NAPE, 2009), a boom could be expected to occur in all sectors of the economy. There is a dire need for planning for organized development to take place. GIS and spatial data infrastructure will need to be put into place for development to occur in an orderly fashion in the next decade (Sentongo, 2003; Wegener & Junius, 1993). It is in this vein that the current state of GIS needs to be established. And this is exactly the justification for the second research question in this dissertation, what is the current state of GIS in Uganda? Of course, related to the second is the first question, how did GIS evolve as a technology in Uganda over time? The government of Uganda realized the importance of GIS by making it one of the priority areas in the national development plan 2010-2015 (GoU, 2010). Though not the focus of this research, the National Development Plan also recognizes the need for a national spatial database, which can also be referred to as national spatial data infrastructure (NSDI). A lot of research has been done on this subject within the global context (Budhathoki & Nedović-Budić, 2007a; Chan, Feeney, Rajabifard, & Williamson, 2001; Yola Georgiadou, Harvey, & Miscione, 2007; Harvey, 2011; Harvey & Tulloch, 20 2006; H. J. Onsrud, 2004), regional context (Ezigbalike, Selebalo, Faïz, & Zhou, 2000; Janssen & Dumortier, 2007; Makanga & Smit, 2010; Masser, 2010; F. D. R. Taylor, 2005) and the national context (Lwasa et al., 2006; Lwasa, Nasirumbi, Amadra, Muhwezi, & Dielmann, 2005; Muhwezi, 2006; Musinguzi, Bax, & Tickodri-Togboa, 2004; Nasirumbi, 2006). Next, I discuss the approach used to carry out this dissertation research. 1.6 Approach to Research In this dissertation research project, I employ mixed methods to investigate my research questions. This involves a combination of both classical qualitative and quantitative research methods in the same research project (Creswell, 2012). In this section, I present a description of the methods used in this dissertation research, make a case for novelty, and explain the advantages of these methods over others. 1.6.1 Description of Methods Mixed methods research is formally defined as, “A class of research where the researcher mixes or combines quantitative and qualitative techniques, methods, approaches, concepts, or language into a single study” (Johnson & Onwuegbuzie, 2004, 17). I chose this research methodology mainly because I sought to take advantage of both qualitative and quantitative approaches to research, and in my particular case, I exploit the strengths of both methodologies to obtain the most appropriate form and quantity of data to address both the overarching and nuanced questions about the evolution and current “state” of GIS in developing countries. 21 By definition, qualitative analysis is “a process of examining and interpreting data in order to elicit meaning, gain understanding, and develop empirical knowledge” (Corbin and Strauss 2008, 1). In contrast, “The quantitative research paradigm is based on positivism. Science is characterized by empirical research, and all phenomena can be reduced to empirical indicators which represent the truth” (Sale, Lohfeld, and Brazil 2002, 44). The ontological position of the quantitative research paradigm is that there is only one truth, an objective reality that exists independent of human perception (Sale et al., 2002). Next, I discuss data collection methods used in this research. Data Collection Methods For purposes of data collection, I divided my research project into two methodological categories corresponding to the two major research objectives of the project: (1) the evolution and current state of GIS, and (2) the potential for mobile location-based services in Uganda. In line with these, I used the following data collection methods: (1) in-depth personal structured interviews with research subjects, 2) survey questionnaires personally administered to research subjects, (3) a focus group discussion as part of a case study of an nongovernmental organization (NGO). My unit of measurement for each category, that is to say, the two research objectives, respectively, is as follows: (1) organizations, which I also refer to as institutions, and 2) individuals. I divided the organizations or institutions covered under this research into five categories: 1) public, 2) private, 3) NGOs, (4) International Organizations (IO), and 5) academic sector institutions. 22 Next, I discuss the qualitative analysis methods used in this research. Qualitative Analysis: Grounded Theory, Diffusion of Innovations Theory, and GIS and Society Analysis of the transcribed interview data was performed using qualitative methods, specifically, grounded theory (Corbin & Strauss, 2008). I used this technique to extract concepts from my qualitative data, which consisted of word-for-word transcriptions of in- depth personal interviews. The interviews were recorded using a digital voice recorder. As the name suggests, “grounded theory” allows social scientists to build their own theory that is grounded in observation data. Grounded theory is “a specific methodology developed by Glaser and Strauss (1967) for the purpose of building theory from data” (Corbin & Strauss, 2008). The epistemology of grounded theory as a methodology “came in a two-step evolution, involving both the tradition of Chicago Interactionism (embedded in symbolic interaction (Blumer, 1969)) and the philosophy of pragmatism inherited largely from John Dewey and Georg Mead” (Corbin & Strauss, 2008). Further, I tested facets of Diffusion of Innovations theory (Rogers 1993, 1995, 2003) to critically analyze the concepts that had been extracted from the data based on grounded theory. This was done so as to provide answers to all three research questions (see section on Research Questions). In addition, these same extracted concepts were qualitatively analyzed based on the five perspectives of GIS and Society discourse so to examine the impact of GIS on society in Uganda (H. J. Campbell, 1996; Masser, 1993; McMaster & Harvey, 2010; Nyerges et al., 23 2011; H. J. Onsrud & Pinto, 1991). This was done so as to provide answers to the first two research questions (see section on Research Questions). Next, I discuss the quantitative analysis methods used in this research. Quantitative Analysis As for the survey data collected using personally administered paper questionnaires, I used a simple quantitative methods approach to analysis. I computed percentages based on the total number of respondents in the sample data to analyze various attributes related to mobile phone usage and the potential for location based services. This was done so as to provide answers to the third research question (see section on Research Questions). Next, I discuss the novelty of the mixed methods approach to research. 1.6.2 Novelty of Approach to Research I used Diffusion of Innovations theory to provide a broad theoretical framework for analysis of my qualitative data. This research further used the five perspectives of GIS and Society discourse (McMaster & Harvey, 2010) to assess the impacts of GIS on Society. This research tests facets of diffusion theory and, further, analyzes the data through the GIS and Society lens so as to offer a complete understanding about the evolution and present state of GIS in Uganda, and by extension in developing countries. It employs a mixed methods approach to data analysis, which to the best of my knowledge has not been employed in previous studies of GIS diffusion in developing countries. 24 This dissertation research complements both GIS Diffusion and GIS and Society research by contributing perspectives and concepts that are relevant specifically to developing countries. The novelty of the research is attributed to employing two conceptual frameworks in the same research project, and further, employing both qualitative and quantitative (mixed) methods. Next, I discuss the advantages of the mixed methods approach to research. 1.6.3 Advantages of the Mixed Methods Approach My approach to this research is mixed methods, an approach that combines the collection and analysis of both qualitative and quantitative data. Personal interviews were carried out with research subjects representing institutions in Uganda, and short survey questionnaires were administered to individuals about their use of mobile phones. The advantage with mixed methods research is that shortfalls of one approach are complemented by the other. Mostly, qualitative methods of research have been employed in GIS and Society (Barrett et al., 2001; J. Corbett et al., 2006; Kyem, 2004a; Leitner et al., 1998; Sahay & Walsham, 1996; Weiner & Harris, 1999), while GIS diffusion research has mainly been carried out using quantitative methods (for example, Masser and Cambell 1996; Masser 1993; Pinto and Harlan J. Onsrud 1993), based on mail-in and telephone surveys. The shortfall of purely quantitative surveys is that due to the a priori determined questions asked in a survey questionnaire, there is little room for “unknown” information to emerge out of the survey. The disadvantage of purely qualitative methods, such as personal interviews, is that respondents might go off on tangents and detract from the relevant issues. The advantage of quantitative methods is that they are usually more 25 objective, concise, and good for summarizing the big picture, while with qualitative methods; the advantage is that they are extremely detailed and can be used to infer social processes much better than quantitative methods. I carried out detailed personal interviews with respondents in public, private, NGO, IO and academic sector institutions, about their use of GIS, and I also administered survey questionnaires to individuals about their use of mobile phones for spatial decision making. At the analysis stage, I transcribed my interview data, and analyzed it using principles of grounded theory (Babbie, 2007; Corbin & Strauss, 2008) to extract concepts within the textual data. I further tallied these concepts in a database and performed statistical analysis of the concepts obtained earlier, in essence using qualitative data as input into a quantitative analysis. Further, I carried out descriptive statistical analysis on my survey data on mobile phone and LBS usage. Next, I discuss potential biases in this dissertation research. 1.7 Potential Biases and Limitations of Research Potential limitations of this dissertation research are of two types: (1) limitations associated with the methods used, and (2) limitations associated with the theoretical framework used. Under the first category, I discuss limitations associated with the case study approach and mixed methods. Under the second category, I discuss the four criticisms of Diffusion of Innovations theory. 1.7.1 Case Study Approach One of the major limitations of this research is that it is based on a case study of only one developing country, Uganda. The findings are extrapolated to draw conclusions about 26 developing countries in general. It is possible that social, political, economic and institutional issues in Uganda do not necessarily apply to other countries in the developing world, and thus this could cause a bias in the results. To reduce the effects of such a bias, the results from this case study were compared with those from other countries. 1.7.2 Mixed Methods Another limitation regards the methodology used in this research, mixed methods, which is a combination of qualitative and quantitative research methods. A qualitative research method used was in-depth personal interviews. These are subject to personal biases introduced into the qualitative research analysis by the views of the respondents interviewed. Results of analysis could differ depending on the respondents’ socio- economic status, level of education, and degree of familiarity with GIS leading to incoherency in the final conclusions drawn from the analysis. The nature of in-depth personal interviews is that they take much longer to organize and administer as compared to other research methods such as mail, telephone, and paper surveys. Moreover, it is only possible to carry out personal in-depth interviews with a much smaller number of respondents as compared to questionnaire surveys within a given time period. To reduce the effects of this bias, whenever possible, at least two or three employees working for the same institution were interviewed so as to check the accuracy of the information provided. A quantitative research method used in this research was the paper-based survey questionnaire method. This method is prone to biases if the sample does not represent the 27 general population under study. One of the ways to reduce potential bias introduced by a non-representative sample is to use large sample sizes. In this research, a sample of 182 individuals was used to investigate the penetration of mobile phones in the Ugandan market, and a subset (of the 182 individuals) sample of 101 individuals was used to assess the potential for location-based services. A larger sample would probably have reduced the effect of biases introduced into the analysis because of sample size. 1.7.3 Limitations of Diffusion Theory The diffusion of innovations theoretical framework was used to perform mixed methods analysis in this research. There are four main criticisms of diffusion research, (1) the pro- innovation bias, (2) the individual-blame bias, (3) the recall problem, and (4) the issue of equality (Rogers, 2003). Pro-innovation Bias The pro-innovation bias is the implication in diffusion research that an innovation should indeed be diffused and adopted by all members of a social system, that it should be diffused more rapidly, and that the innovation should be neither re-invented nor rejected (Rogers, 2003). The main reason for this bias is because much diffusion research is funded by change agencies. Further, successful diffusion leaves traces that can be investigated by diffusion researchers, while an unsuccessful diffusion effort leaves little or no visible traces (Rogers, 2003). In this paper, I have tried to reduce the effects of these biases by highlighting both the positive and the negative consequences of GIS diffusion. For example, much as GIS is a useful planning tool used in the public sector, its (negative) effect on job security, maintenance costs in terms of licenses, and the 28 dependency on donor funding has also been highlighted. I also discuss the fact that re- invention is important for diffusion of GIS technology; however, one struggles to find signs of GIS software customization (re-invention) at Ugandan institutions. Individual-blame bias Diffusion research tends to “side with change agencies that promote innovations rather than with individuals who are potential adopters” (Rogers 2003, 118), the so-called individual-blame bias. This is often because the sponsors of diffusion research tend to be change agencies. The implication of this is that blame for unsuccessful diffusion of innovations tends to be on the individual rather than the social system leading to an individual-blame bias in the research findings; for example, in the case of GIS diffusion, the blame falls on potential adopters rather than on GIS technology vendors, International donors, and technocrats in government. In this research paper, I have made considerable efforts to address this bias by analyzing the roles of the various members of the social system including international donors, International Organizations like UNEP-GRID, and individuals in the institutions adopting GIS technology. Recall Problem The recall problem in diffusion research results from the methodology used as regards measuring the time of adoption of an innovation, and in determining causality (Rogers, 2003). In-depth personal interviews tend to ask respondents to draw on their memory of events leading up to the adoption of GIS in their organizations. Oftentimes, their memory is inaccurate. One way to reduce the effects of this bias is to do some archival research to establish exact dates of introduction of the technology and other information relevant to 29 the early diffusion of an innovation. In this research project, whenever possible, archival documents were sought and used to check the accuracy of transcribed interview data. Equality and Equity With regard to the issue of equality in diffusion of innovations research, the cultural importance and appropriateness of the diffusion paradigm in developing nations is questioned (Rogers, 2003). Diffusion research seems to promote economic growth through industrialization and urbanization, capital-intensive laborsaving technology transferred from industrialized nations, centralized planning mainly by government economists, and ignores the causes of underdevelopment which have to do with unfair trade relationships with industrialized countries (Rogers, 2003). Indeed some technological innovations have negative consequences on society in developing nations such as increasing the socio-economic gap between the rich and the poor, inequitable distribution of individual incomes, unemployment, and rural to urban migration. To reduce the effects of this bias in this research, a section was devoted to the positive and negative consequences of GIS diffusion in Uganda. A final limitation is in answering the question, has Uganda attained a critical mass of adopters signaling the take-off stage of the S-shaped curve? There was insufficient data collected in this particular research to adequately answer this question. A larger sample of institutions would be needed that encompasses all sectors of the institutional framework so as to gauge the percentage level of adoption of GIS technology over time. Future research should address this important question. 30 Next I present a general overview of the literature on diffusion of innovations, and GIS and Society. 1.8 Overview of the Literature The literature on Diffusion of Innovations theory is very broad (for example, see Rogers 1993; Rogers 1995; Valente and Rogers 1995; Rogers 2003), however, there is a narrower body of literature, “Diffusion of GIS,” specific to social issues regarding GIS technology. Most of this literature surfaced between 1990 and 2000 (for example, see Wiggins 1993; Miellet 1996; Zwart 1993; Masser and Campbell 1995; Rogers 1993; H. J. Onsrud and Pinto 1993; Pinto and Onsrud 1995; Nedović-Budić and Godschalk 1996; Harvey 2001; Chan and Williamson 1999a; Chan and Williamson 1999b; Chan and Williamson 2000). The period 2000-2011 has seen a shift in the literature focus from GIS Diffusion to Spatial Data Infrastructure (SDI) (for example, see Rajabifard et al. 2003; Masser 2005; Masser 2011; Masser 2007; Masser, Borrero, and Holland 2003; De Man 2007; H. Onsrud et al. 2005; H. J. Onsrud and Craglia 2003; McDougall, Rajabifard, and Williamson 2007; Budhathoki and Nedović-Budić 2007b; Grus, Crompvoets, and Bregt 2008; Harvey 2011; Harvey and Tulloch 2006; Yola Georgiadou, Harvey, and Miscione 2007). This is because most of the literature on GIS diffusion and SDI, in both these time periods, mainly addresses issues in developed countries in Europe (Brunn, Dahlman, & Taylor, 1998; Cipriano, 2006; Junius, Tabeling, & Wegener, 1996; Masser & Campbell, 1996; Miellet, 1996; Van Loenen & Kok, 2004), North America (Guptill & Eldridge, 1998; Harvey & Tulloch, 2004; Nedović-budić, Pinto, & Warnecke, 2004; H. J. Onsrud, 31 2004; H. J. Onsrud & Pinto, 1991), and Australia (Chan & Williamson, 1999a, 1999b; Clarke, Hedberg, & Watkins, 2003; Jacoby, Smith, Ting, & Williamson, 2002; Najar, Rajabifard, Williamson, & Giger, 2007; Williamson, Rajabifard, & Binns, 2007). The issues that GIS diffusion addresses, mainly the implementation of GIS and the institutional issues involved in that implementation, were important and interesting in the 1990s when GIS was picking up in the developed countries. By the beginning of the new century, most of these countries had already implemented institutional GIS within their countries, and thus, their research focus turned to spatial data quality and the infrastructure necessary to allow efficient sharing of that data. However, this is not the trend of events in the developing world. Literature addressing the issues in the developing world is plentiful mostly for issues on SDI (Câmara, Fonseca, Monteiro, & Onsrud, 2005; Y. Georgiadou, Puri, & Sahay, 2005; Karatunga, 2002; Lwasa et al., 2005; Miscione & Vandenbroucke, 2011; Musinguzi, 2003a; Nasirumbi, 2006; Onah, 2009; Puri, Sahay, & Georgiadou, 2007; Smit, Makanga, Lance, & Vries, 2009; F. D. R. Taylor, 2005), however, one would struggle to find literature specifically on GIS diffusion (examples include: Cavric, Nedović-Budić, and Ikgopoleng 2003; Gibson 1998). I argue that the diffusion of GIS needs more research to address the gap in the literature, and this is mainly because to understand SDI issues in developing countries, one needs to understand how GIS technology diffused within the context of developing countries. My research, thus, makes a contribution toward filling in the gap in GIS diffusion literature with respect to developing countries. 32 The link between GIS Diffusion and GIS and Society literature is that the former body of research was a precursor to the latter. Whereas GIS Diffusion literature answers questions about the adoption of GIS technology among members of a social system via certain communication channels over time, GIS and Society asks a set of related questions focused on the impact of GIS on society, and the impact of society on GIS. Thus, GIS and Society can be arguably considered a subset of GIS Diffusion. In fact, some of the same questions asked by GIS and Society scholars in 1993 were also posed by Diffusion of GIS scholars around the same time. This is evident in the book, Diffusion and Use of Geographic Information Technologies, edited by Masser and Onsrud (1993), which is a collection of papers on the diffusion of GIS, mainly in European countries, that resulted out of a NATO Workshop on “Modeling the Diffusion and Use of Geographic Information Technologies” in Sounion, Greece, in April 1992. The research agenda of GIS and Society has its roots in discussions held between geographic information scientists and critical human geographers in the early 1990s (Nyerges et al., 2011). In the fall of 1993, a heated debate between scholars in the two disciplines in Friday Harbor, Washington, resulted in a new research agenda called GIS and Society (Nyerges et al., 2011; Obermeyer, 1993; Sheppard, 1993), which progressed in several directions. Two key publications followed on the subject of GIS and Society, one edited by John Pickles, Ground Truth (1995) and the other a special issue by the Cartography and Geographic Information Science Journal, GIS and Society, edited by Eric Sheppard and Tom Poiker (Poiker, 1995). An iconic meeting sponsored by the National Center for Geographic Information and Analysis (NCGIA) in New Haven, 33 Minnesota (Harris & Weiner, 1996) resulted in a new model of GIS that was more inclusive. This was called GIS II, or Public Participation GIS (PPGIS) or community- based GIS (Nyerges et al., 2011). Next, I provide a definition of some of the common terminologies used in this dissertation. 1.9 Definition of Terminologies There are various terms and terminologies used throughout this dissertation. In this section, I clarify on the exact meanings of each. GIS. This acronym can mean Geographical Information Systems (outside North America (Lo & Yeung, 2007)), or Geographic Information Systems, or Geographic Information Systems and Science, or Geographic Information Science. Geographic Information Systems can be defined as “a computer based systems for storing and processing geographic information” (Longley et al. 2010, 13). Geographic Information Science can be defined as “a set of basic research issues arising from the use of Geographic Information Technologies”, which also calls for a description of Geographic Information technologies and geographic information. “Geographic information technologies apply digital methods to geographic information”, while geographic information is defined as “an abstraction of primitive tuples linking geographical information to primitive descriptors” (Goodchild, Mark, and Sheppard 1999, 1). Evolution of GIS. I define the evolution of GIS as the chronological development of GIS as a technology in various sectors of the economy of a country, right from the introduction of the technology into institutions, to the present day. 34 State of GIS. I define the “state” of GIS as the current penetration level of GIS technology into the various sectors of a country’s economy, and the socio-economic and political context that has affected the diffusion of GIS into that country’s public, private, NGO, IO and academic sector institutions. Implementation of GIS. In this dissertation, the term implementation of GIS can be understood to mean the diffusion of GIS into an institution, an understanding that I have adopted from the work of Noongo (2007). GIS and Society. GIS and Society is a research area in GIScience that focuses on “how geographic information technologies affect people, and organizations and, conversely, how people and organizations use the technology, and by using it, cause it to change” (Nyerges et al., 2011). Diffusion of Innovations. Diffusion is defined as “the process by which (1) an innovation (2) is communicated through certain channels (3) over time (4) among the members of a social system. The four main elements are the innovation, communication channels, time and the social system” (Everett M. Rogers 2003, 11). A plot of the adoption percentage versus time produces as an S-shaped curve, which is a typical temporal pattern for all innovations. A critical mass of adopters is required before an innovation can take off (Rogers, 2003). GIS Diffusion. I define GIS diffusion as the process by which (1) GIS technology (2) is communicated through certain channels (3) over time into the institutional framework of a given society, which includes the public, private, NGO, IO and academic sectors. This 35 definition is based on the formal definition of diffusion of innovations offered by Rogers (2003). Spatial Thinking and Spatial Awareness. I define spatial thinking as one’s ability to incorporate the notion of space, place and time into one’s daily life (Manson, Kne, Dyke, Shannon, & Eria, 2012; NRC, 2006). Spatial awareness can be defined as a person’s conscious knowledge about attributes in space around his/her location (J. Campbell, Hardy, & Barnard, 2010; Nasirumbi, 2006; Toppen, 1991). Next, I outline the general structure of this dissertation. 1.10 Structure of Dissertation This dissertation consists of eight chapters. Chapter one and two are the introduction and literature review, respectively. Chapter three discusses the methodology. Chapters four to seven present the findings of the dissertation research. Chapter four presents a descriptive analysis of the evolution of GIS in Uganda, and more generally in developing countries. Chapter five examines the current state of GIS in Uganda in the public and academic sectors. Chapter six investigates the current state of GIS in Uganda in the NGO, IO and private sectors. Chapter seven projects the future of GIS and investigates the potential for mobile phones and location-based services in Uganda. Finally, the summary and conclusions follow in chapter eight. 36 Chapter 2 : Literature Review 2.1 Introduction This chapter is a review of the literature that relates to the evolution and current state of GIS in both the developed and developing world. Broadly, this chapter presents a survey of literature on theoretical frameworks for understanding the adoption, use and diffusion of technology. Further, a review of the literature on GIS diffusion, GIS and Society, Information and Communication Technology (ICT), Mobile phones, and Location-based Services (LBS) is presented. I first provide a definition of Geographical Information Systems and Science in section 2.2 within the context of current technological advances in the field of GIS. In section 2.3, I situate GIS within the broader Information Systems literature. Theories and frameworks for studying the evolution and state of GIS are presented in section 2.4. This is followed by section 2.5, where I briefly discuss various theories for understanding the diffusion of technology, and then focus on the theory chosen for this dissertation research, Diffusion of Innovations. In section 2.6, I discuss the evolution and state of GIS in, both the developed, and developing world, as seen through the lens of Diffusion of Innovations theory. I discuss the GIS and Society conceptual framework for studying the impact of GIS on Society in section 2.7. A review of the literature on ICT, mobile phones, and LBS in developing countries is presented in section 2.8. 37 2.2 GIS - Geographical Information Systems or Geographic Information Science? The acronym GIS can mean Geographical Information Systems (outside North America (Lo & Yeung, 2007)), or Geographic Information Systems and Science, or Geographic Information Science (also GIScience, for short). There is a rich history of how the field widely known today as GIS obtained its current name (and acronym). First, let me provide relatively recent definitions of GIS as we know it today. Geographic Information Systems can be defined as “a computer based systems for storing and processing geographic information” (Longley et al. 2010, 13). Geographic Information Science, on the other hand, can be defined as “a set of basic research issues arising from the use of Geographic Information Technologies” (Goodchild, Mark, & Sheppard, 1999, 1). This also calls for a definition of Geographic Information technologies and geographic information. “Geographic information technologies apply digital methods to geographic information”, while geographic information is defined as “an abstraction of primitive tuples linking geographical information to primitive descriptors” (Goodchild, Mark, and Sheppard 1999, 1). The main difference between GIS and GIScience is the emphasis in the latter on the inclusion of people (society) in the use and practice of GIS (Longley et al., 2010). In addition, the “Science” in GIS further emphasizes the multi-disciplinarity of GIS and its use in Science-based disciplines like environmental sciences, engineering and the social sciences (Longley et al., 2010). 38 Next, I situate GIS within the broader Information Systems literature. 2.3 Situating GIS within Information Systems Literature GIS consist of multiple technologies, and draw upon theories and paradigms from various disciplines, including geography, urban planning and design, and computer science. Within the discipline of computer science is the field of Information Systems (IS). Some would situate GIS within the sub-discipline of IS, although I argue that the hierarchical relationship between the two disciplines is weak, mainly because IS as a sub-discipline of computer science developed along a totally parallel path for the purpose of supporting activities in business administration and management. There is little emphasis on space and place in the field of IS, which are strong attributes of Geography, and by extension, GIS. A more detailed discussion of the relationship between GIS and the broader Information Systems literature can be found in Appendix G: Situating GIS within Information Systems Literature. 2.4 Theories for Understanding the Acceptance and Use of Information Systems So as to understand the evolution and state of GIS in developing countries, it would be logical to use a theory that explains how a technology is invented by a society, deployed within that society, gets widely used, is accepted, and thrives thereafter, possibly reaching a maxima, and then declines to become obsolete. Such a theory would be very helpful for understanding GIS in the developed world where most of the technologies were originally invented. 39 On the contrary, such a theory would not immediately be appropriate for understanding the evolution and current state of GIS within the context of developing countries where invention of new technologies is uncommon. Thus, a theory that explains the concept of diffusion of technology from the western world to developing countries would make more sense in this case. Related to the diffusion of technology is the notion that diffusion does not necessarily indicate success in as far as acceptance and use of a given technology is concerned. Thus, theories that explain society’s acceptance and eventual widespread use of a technology would be helpful in understanding the state of a technology (in this case GIS) in a given developing country. I briefly discuss three theories that explain the acceptance and use of Information Systems, and compare the strengths and weaknesses of these relative to the theory chosen for this dissertation, Diffusion of Innovations theory. The three theories include: (1) the Technology Adoption Model (TAM), (2) Social Cognitive Theory (SCT), and (3) the Universal Theory of Adoption and Use of Technology (UTAUT). By 2003, there were two dominant types of theories that were increasingly used to study the evolution of Information Systems (Liang & Chen, 2003): (1) theories for human related research, and (2) theories for systems related research. The former were on the increase, and there was a corresponding decline in the latter. Liang and Chen (2003) stipulated the causes for these trends as being driven by two factors, (1) the development of new technology, and (2) organizational needs. Two of the most commonly used theories include the Technology Acceptance Model (TAM), and Social Cognitive Theory 40 (SCT). A third theory, the Unified Theory of Acceptance and Use of Technology (UTAUT) (Venkatesh, Morris, Davis, & Davis, 2003), is a unified theory that marries together several theories on user acceptance and the use of technology in general. 2.4.1 The Technology Acceptance Model TAM theorizes that the major factor leading to the adoption of a technology is the attitude of the user towards using that technology. This attitude is determined by the user’s perceived ease of use, and perceived usefulness of the technology. Design features of the technology do not have a direct effect on the attitude of the user. (Davis Jr., 1985) This theory’s strength lies in understanding technology adoption based the user’s perceptions of a technology, rather than the effectiveness of design of the technology. TAM exhibits some overlap with diffusion of innovations theory in that the latter addresses the same issues as “perceived attributes of an innovation,” for example, in terms of relative advantage and compatibility. A weakness of TAM that has been highlighted by some is the deterministic nature of the theory. A user’s intention to use a technology is supposed to translate into actual use of the technology, according to TAM. However, this theoretical assumption has been criticized by some because it does not account for re-evaluation and reflection, which could lead to a change in actual adoption and use of the technology (Bagozzi 2007 in Chuttur 2009). Diffusion of innovations Theory, however, takes this weakness into account by tracing the innovation-decision process that could lead to adoption or rejection of an innovation. For a detailed description of this theory, see the work of Davis 41 Jr. 1985, a summary of which can be found in Appendix H: Theories for Understanding the Acceptance and Use of Information Systems. 2.4.2 Social Cognitive Theory Social Cognitive Theory rests on the premise that human motivation and action is determined by a triad of factors, that is, (1) behavioral, (2) cognitive, and (3) environmental factors (Bandura, 1976, 1986, 2001a). The factors combine in a relationship called reciprocal determinism, which essentially means that people are both products and producers of their environment (Bandura, 1986). The theory argues that people learn by observing behavior, and developing expectations about the outcomes of such behavior. Further, people go through a process of self-regulation and self-constraint which leads to a modification in their behavior. This latter process is called self-efficacy. (Bandura, 1986) The strengths of SCT for understanding technology adoption lies in self-efficacy, which essentially involves constant self-reflection in light of stimulants in the external environment. SCT has elements of Diffusion of Innovations theory with respect to influences of a social system on the adoption or rejection of an innovation or technology, for example, social norms, social structure, change agents and opinion leaders. Further, both theories explore the effect of user observation on eventual behavior of potential adopters of an innovation or technology. In Diffusion of Innovations theory, this observability aspect is one of the perceived attributes of an innovation. A weakness of SCT that has been cited (for example, see Compeau and Higgins (1995)) is the effect of expectations (developed on the basis of observations of technology use) on 42 actual outcomes of adoption. Observability has not necessarily led to adoption of a technology. For a detailed description of SCT, see the work of Albert Bandura (1976; 1986), a summary of which can be found in Appendix H: Theories for Understanding the Acceptance and Use of Information Systems. 2.4.3 Unified Theory of Acceptance and Use of Technology The Unified Theory of Acceptance and Use of Technology (UTAUT) is a theoretical framework that combines eight theories, including TAM, SCT, and Diffusion of Innovations theory. The basic conceptual framework of this theory posits that actual use of Information Technology (IT) by an individual is affected directly by the intentions to use IT, which in turn depends on individual reactions to using IT. There are four key moderating variables that influence technology adoption: (1) experience, (2) voluntariness of use, (3) gender and (4) age. Further, there are four constructs that are direct determinants of user acceptance and usage behavior: (1) performance expectancy, (2) effort expectancy, (3) social influence, and (4) facilitating conditions. (Venkatesh, 2000; Venkatesh & Bala, 2008; Venkatesh & Davis, 2011; Venkatesh et al., 2003) The main strength of UTAUT is that it combines elements from several well-known theories on technology use and adoption, and identifies the most significant elements. In comparison to Diffusion of Innovations theory, the influence of the social system is clearly considered as one of the important determinants of user acceptance of a technology. Other elements in Diffusion of Innovations theory are inherently part of some of the other constructs alluded to in UTAUT. For a detailed description of this theory, see the work of Venkatesh and others (2003), a summary of which can be found 43 in Appendix H: Theories for Understanding the Acceptance and Use of Information Systems. 2.5 Theories for Understanding the Diffusion of Technology To understand the evolution of GIS in developing countries, and the current “state” of GIS, I employ the concept of “diffusion of technology.” Diffusion here is similar in meaning to “acceptance, adoption and use” of technology discussed above. The concept, diffusion of technology, models the process by which a given technology enters, permeates, and becomes common place in a given society. The concept of diffusion captures the processes involved in the evolution of a technology in a society all the way up to the present “state” of the technology. Theories for understanding the diffusion of technology include: 1) Technological Determinism, 2) Economic Determinism, 3) Social Interactionism, and 4) Diffusion of Innovations. Each of these theories has its merits and demerits. In this dissertation research, I chose Diffusion of Innovations theory. The main reason for choosing this theory is that it is the broadest of all the theories, and provides a fair balance between the constructs or concepts that constitute the other theories mentioned above. It provides a balance between the role of technological factors (technological determinism), economic factors (economic determinism), and social factors (social Interactionism), and in addition, provides a few other constructs not considered in the other theories, for example, the notion of time (temporal factors) and communication. A summary of the theoretical constructs of technological determinism, economic determinism, and social 44 interactionism can be found in Appendix I: Theories for Understanding the Diffusion of Technology. Next, I discuss the Diffusion of Innovations theoretical framework based on the work of Rogers (2003). 2.5.1 Diffusion of Innovations “Diffusion is the process in which an innovation is communicated through certain channels over time among the members of a social system” (Rogers, 2003, 5). Diffusion is a special type of communication in which messages exchanged are about new ideas (Rogers, 2003). “Communication is a process in which participants create and share information with one another in order to reach a mutual understanding” (Rogers, 2003, 5). Communication is a two-way process of convergence (or divergence) in which there is interaction between a change agent and a client, and this continues through several cycles. The change agent seeks to persuade a client to adopt an innovation, but the reverse is also possible where the client goes to a change agent with a specific problem, and the innovation may be recommended as a possible solution (Rogers, 2003). The newness of the idea is a special character of the type of communication in diffusion (Rogers, 2003). This newness means that there is a degree of uncertainty that’s involved in diffusion (Rogers, 2003). “Uncertainty is the degree to which a number of alternatives are perceived with respect to the occurrence of an event and the relative probability of these alternatives. Uncertainty implies a lack of predictability, of structure, of information. Information is a means of reducing uncertainty. A technological innovation 45 embodies information and thus reduces uncertainty about cause-effect relationships in problem solving” (Rogers, 2003, 6). Rogers (2003) asserts that Diffusion is a type of social change, “defined as the process by which alteration occurs in the structure and function of a social system” (Rogers, 2003, 6). Social change occurs as ideas are invented, diffused, and adopted or rejected, leading to certain consequences (Rogers, 2003). There are four main elements in the diffusion of innovations: 1) the innovation, 2) communication channels, 3) time, and 4) the social system (Rogers, 2003). This follows directly from Rogers’ (2003) definition of diffusion, “the process by which an innovation is communicated through certain channels over time among the members of a social system” (Rogers, 2003, 11). These elements are identifiable in each and every diffusion study (Rogers, 2003), and in the case of this dissertation, in the study of GIS diffusion in developing countries. The process of diffusion is best illustrated by an S-shaped curve shown in Figure 2-1. 46 Figure 2-1: The Diffusion Process – the S-shaped Diffusion Curve Source: (Rogers, 2003) 2.5.4.1 The Innovation “An innovation is an idea, practice, or object that is perceived as new by an individual or other unit of adoption” (Rogers, 2003, 12). Newness does not mean the actual time lapse from the time of discovery of an innovation. If an innovation seems new to an individual, then it is an innovation (Rogers, 2003). Thus, the perceived newness of an idea for an individual is what defines newness (Rogers, 2003). Technological Innovations, Information and Uncertainty Often the terms “innovation” and “technology” are used as synonyms because a lot of diffusion research has focused on technological innovations (Rogers, 2003). A 47 technology consists of two components, the hardware and software. The hardware embodies the material or physical object, while the software the information base for the technological tool. “The diffusion of software innovations has been investigated, although a methodological problem in such studies is that their adoption cannot be so easily traced or observed. Such idea-only innovations have a relatively lower degree of observability and thus a lower rate of adoption” (Rogers, 2003, 13). Thus, “knowing of a technological innovation creates uncertainty about its consequences in the minds of potential adopters. Will the innovation solve the adopter’s perceived problem? The potential advantage of a new idea impels an individual to exert effort to learn more about the innovation. Once such information seeking activities reduce uncertainty about the innovation’s expected consequences to a tolerable level, a decision concerning adoption or rejection can be made” (Rogers, 2003, 14). Perceived Attributes of innovations To think that all innovations are equivalent units of analysis would be to oversimplify. For example, compare the short time it took for personal computers (PCs) to achieve widespread adoption, as compared to using seat belts in cars, which took decades in the US (Rogers, 2003). Individuals perceive characteristics of innovations differently, which also explains their different rates of adoption. There are five characteristics (Rogers, 2003). (1) Relative advantage is “the degree to which an innovation is perceived as better than the idea it supersedes” (Rogers, 2003, 15). This could be measured in economic terms, social prestige factors, convenience, and satisfaction. Just how objective an innovation is 48 not as important as whether an individual perceives the innovation as advantageous – the relative advantage of the innovation. (2) Compatibility is “the degree to which an innovation is perceived as being consistent with the existing values, past experiences, and needs of potential adopters” (Rogers, 2003, 15). “An idea that is incompatible with the values and norms of a social system will not be adopted as rapidly as an innovation that is compatible” (Rogers, 2003, 15). (3) Complexity is “the degree to which an innovation is perceived as difficult to understand and use” (Rogers, 2003, 16). Ideas that are simpler to understand and relatively less complex are adopted more rapidly, than those that require the adopter to develop new skills and understandings. (4) Trialability is “the degree to which an innovation may be experimented with on a limited basis” (Rogers, 2003, 16). For example, in the classical diffusion study of hybrid corn in Iowa, (Ryan & Gross, 1950 in Rogers, 2003), farmers more readily adopted the genetically modified seed after trying it on a partial basis. This was because the level of uncertainty reduces with each try, and increases the chances for adoption. (5) Observability is “the degree to which the results of an innovation are visible to others” (Rogers, 2003, 16). If individuals are easily able to see results of an innovation, then they are more likely to adopt it. In fact, this even stimulates peer discussion of the new idea as colleagues, friends, neighbors of the adopter evaluate the innovation. 49 Re-invention Re-invention refers to the degree to which an individual can customize an invention to suit his or her needs during the process of adoption. An innovation diffuses more rapidly when it can be re-invented, and its adoption is more likely to be sustained (Rogers, 2003). 2.5.4.2 Communication Channels “Communication is the process by which participants create and share information with one another in order to reach a mutual understanding” (Rogers 2003, 18). Diffusion is a unique kind of communication whereby the message content or information exchanged is directly concerned with a new idea (Rogers, 2003). This involves (1) an innovation or idea, (2) an individual who has knowledge of the innovation, (3) another individual who does not have the knowledge, and (4) the communication channel. “The communication channel is the means by which messages get from one individual to another” (Rogers 2003, 18). These channels include, (1) mass media, for instance radio, television, newspapers, and the Web, and (2) interpersonal channels, which includes face to face information exchanges between two or more people (Rogers, 2003). Rogers (2003) points out a very interesting fact that most people do not evaluate an innovation on the basis of scientific studies of its consequences, but instead depend mainly upon subjective evaluation conveyed to them by peers who have adopted the innovation already. Rogers (2003) asserts that diffusion is a very social process that involves interpersonal communication relationships. 50 Homophily and Heterophily One of the important principles of human communication is that the transfer of ideas occurs most frequently between two individuals who are similar, or homophilous (Rogers, 2003). “Homophily is the degree to which two or more individuals who interact are similar in certain attributes, such as beliefs, education, socioeconomic status, and the like. Heterophily, the opposite of homophily, is the degree to which two or more individuals who interact are different in certain attributes” (Rogers 2003, 19). One of the problems in the diffusion of innovations, however, is that the participants are often times quite heterophilous, for example, a change agent is more technically competent than his or her clients (Rogers, 2003). This difference leads to ineffective communication. When the individuals are all technically homophilous, there would be no exchange of new information (Rogers, 2003). For effective diffusion there must be some degree of heterophily present between two participants in terms of technical knowledge, and homophily in all other variables (education, socio-economic status and so on). (Rogers, 2003) 2.5.4.3 Time Time is one of the four elements in the diffusion process, and its inclusion is one of the strengths of diffusion theory. The time dimension is involved in diffusion in three ways (Rogers, 2003): “(1) the innovation-decision process by which an individual passes from first knowledge of an innovation through its adoption or rejection, (2) the innovativeness of an individual or other unit of adoption (the relative earliness/lateness with which an innovation is adopted) compared with other members of a system, and (3) an innovation’s 51 rate of adoption in a system, usually measured as the number of members of the system who adopt the innovation in a given time period” (Rogers 2003, 20). The Innovation-Decision Process “The innovation-decision process is an information-seeking and information-processing activity in which an individual obtains information in order to gradually decrease uncertainty about the innovation” (Rogers 2003, 20). There are five main steps in the innovation-decision process: (1) knowledge, (2) persuasion, (3) decision, (4) implementation, and (5) confirmation (Rogers, 2003). “Knowledge is gained when an individual (or some other decision making unit) learns of the innovation’s existence and gains some understanding of how it functions. Persuasion takes place when an individual forms a favorable or unfavorable attitude toward the innovation. Decision occurs when an individual engages in activities that lead to a choice to adopt or reject the innovation. Implementation takes place when an individual puts an innovation into use. Re-invention is especially likely to occur at the implementation stage. Confirmation occurs when an individual seeks reinforcement of an innovation- decision that has already been made, but he or she may reverse this previous decision if exposed to conflicting messages about the innovation” (Rogers 2003, 20). Innovativeness and Adopter Categories “Innovativeness is the degree to which an individual or other unit of adoption is relatively earlier in adopting new ideas than the other members of a system” (Rogers 2003, 22). “Adopter categories are classifications of members of a social system on the basis of 52 innovativeness, and these categories include: (1) innovators, (2) early adopters, (3) early majority, (4) late majority, and (5) laggards” (Rogers 2003, 22). Innovators are active information seekers about new ideas; they are highly exposed to mass media, and have extensive interpersonal networks beyond their local system. Further, innovators are able to cope with higher levels of uncertainty about an innovation than are other adopter categories (Rogers, 2003). Rate of Adoption The rate of adoption in diffusion research is defined as “the relative speed with which an innovation is adopted by members of a social system” (Rogers 2003, 23). A cumulative frequency plot of individuals adopting an innovation over time produces the signature S- shaped diffusion curve shown in Figure 2-1. In the beginning, only a few individuals adopt an innovation; these are the innovators. After a while, more and more people start adopting the innovation over time, thus the curve starts to climb upwards. After a given time period, the curve starts to level off, as fewer individuals remain who have not yet adopted the innovation. Thus, the S-shaped curve reaches its asymptote, which marks the end of the diffusion process. (Rogers, 2003) There are differences in the rate of adoption, which changes the steepness (gradient) of the S-shaped curve; the steeper the curve, the more rapid the rate of adoption. One of the obvious questions looked at in diffusion research is why some innovations produce steeper curve, meaning there is a higher rate of adoption, than other innovations. This rate of adoption is measured for an innovation in a system, rather than for an individual as a unit of analysis (this variable is innovativeness). The system might be a community, an 53 organization, or some other structure. Innovations that are perceived by individuals as possessing greater relative advantage, compatibility, complexity, trialability and observability are usually adopted at a faster rate than others. (Rogers, 2003) 2.5.4.4 A Social System A social system is defined as “a set of interrelated units that are engaged in joint problem solving to accomplish a common goal (Rogers 2003, 23). The members or units of a social system could be individuals, informal groups, organizations, and/or subsystems. In a system, all members cooperate to solve a common problem, and seek a common objective or goal, and this sharing of a common goal binds the system together (Rogers, 2003). Since diffusion occurs in a social system, it constitutes a boundary. The social system affects diffusion through factors such as norms and values, the influence of opinion leaders and change agents, the types of innovation-decisions in the system, and the consequences of innovation (Rogers, 2003). Social Structure and Diffusion Structure is defined by Rogers (2003, 23) as “the patterned arrangements of the units in a system.” This structure gives regularity and stability to human behavior in a system, and thus allows one to predict behavior in that system with some degree of accuracy. Structure is a type of information, and helps reduce uncertainty (Rogers, 2003). An example of structure is a hierarchical system of government in public sector agencies like ministries in Uganda. At the top is the Minister, followed by his/her deputy, then the permanent secretary and so on. Orders come from the top, and are expected to be 54 followed by lower ranking officials at the bottom. This is a bureaucratic social structure, and it is a formal structure. Informal structures also exist, and these are composed of the interpersonal networks linking members in a system, “tracing who interacts with whom and under which circumstances” (Rogers 2003, 24). This informal structure is called a communication structure, and is usually created in a system in which homophilous sets of individuals are grouped together in cliques. When people in a social system come together for the first time, there is no communication structure, as every individual has an equal probability of speaking with another individual, however, after a given time period, different networks of individuals in cliques begin to emerge (Rogers, 2003). These aspects of communication structure predict the behavior of individual members of a system, and also if and when they would adopt an innovation. (Rogers, 2003) “The structure of a social system can facilitate or impede the diffusion of innovations” (Rogers 2003, 25). The impact of social structure is studied widely by sociologists and psychologists. According to Rogers (2003), this is an under-researched area in diffusion research compared to other aspects of diffusion. System Norms and Diffusion Norms are “the established behavior patterns for the members of a social system” (Rogers, 2003). Acceptable and expected behavior of members within a social system is determined by norms (Rogers, 2003). A social system’s norms can be a barrier to change, for example, pork is not consumed by Muslims and Jews, and therefore a pork factory stands no chance in Palestine and Israel. 55 Opinion Leaders and Change Agents “Opinion leadership is the degree to which an individual is able to influence other individuals’ attitudes or overt behavior informally in a desired way with relative frequency” (Rogers 2003, 27). Opinion leaders provide information and advice about innovations to other individuals in a social system (Rogers, 2003). Opinion leadership does not depend on the individual’s formal position or status in the system; instead it is earned by the individual’s technical competence, social accessibility, and conformity to the norms of the social system (Rogers, 2003). If the social system is oriented towards change, opinion leaders are innovators, but when the system is opposed to change, then opinion leaders reflect this norm (Rogers, 2003). Because opinion leaders conform to the norms of a social system, they serve as a model for the innovation behavior of their followers (Rogers, 2003). Opinion leaders take advantage of their communication networks consisting of linked individuals to exert their influence on a social system (Rogers, 2003). Professionals that represent change agencies are called change agents (Rogers, 2003), for example, marketing executives who work for Environmental Systems Research Institute. A change agent is “an individual who influences clients’ innovation-decisions in a direction deemed desirable by a change agency” (Rogers 2003, 27). Change agents seek the adoption of new ideas, but could also attempt to slow down diffusion by preventing the adoption of undesirable innovations. These change agents usually use opinion leaders in a social system to promote diffusion (Rogers, 2003). Because change agents are highly qualified professionals, usually in a technical field, they are heterophilous from their typical clients (Rogers, 2003). 56 Types of Innovation-decisions The social system further influences the decision of members in the system to adopt or reject an innovation in two ways, (1) an innovation can be adopted/rejected by an individual member of the system or (2) an innovation can be adopted/rejected by the entire social system, which can decide to adopt an innovation by a collective or an authority decision (Rogers, 2003). Thus, a distinction can be made between (1) Optional innovation-decisions – a decision made by an individual independent of the decisions of other members of the social system, (2) Collective innovation-decisions – these are based on consensus among members of the system, (3) Authority innovation-decisions – choices made by relatively few individuals in system who possess power, status, or technical expertise, and here individual members have no influence (Rogers, 2003). The fastest rate of adoption of innovations stems from authority decisions (Rogers, 2003). Optional decisions can be made usually faster than collective decisions. An example of an optional decision turned authority decision is the seat belt requirement in cars in the US in the 1960s, which started out as optional, but when a law was passed, became compulsory (Rogers, 2003). Consequences of Innovations Consequences are “the changes that occur to an individual or to a social system as a result of the adoption or rejection of an innovation” (Rogers 2003, 30). There are three types of consequences: (1) Desirable versus undesirable consequences, depending on if the effects of an innovation are functional or dysfunctional in a social system, (2) Direct versus indirect consequences, depending on whether changes occur in immediate response to an 57 innovation, or as a secondary effect of the direct consequences of an innovation, (3) Anticipated versus unanticipated consequences, depending on whether or not the changes are intended by the members of a social system (Rogers, 2003). “Change agents usually introduce innovations into a client system that they expect will have consequences that will be desirable, direct, and anticipated. But often such innovations result in at least some unanticipated consequences that are indirect and undesirable for the system’s members” (Rogers 2003, 31). Having presented the Diffusion of Innovations theoretical framework, based on the work of Rogers (2003), next I discuss the application of this theory to understanding the evolution and current state of GIS in developing countries. 2.6 Understanding the Evolution and State of GIS as the Diffusion of Innovations Of all the theories discussed in this chapter, I chose to employ the diffusions of innovations theory for understanding the evolution and state of GIS in developing countries, and this was for two reasons: (1) it is a broad theory covering several aspects of social processes, and (2) it is a relatively straight forward and simple theory to understand, and has cross disciplinary appeal. I will now proceed to first present a brief history of GIS, followed by a discussion of literature on the diffusion of GIS in North America and Europe, and end the subsection by tracing the developments of GIS in the developing world. 58 2.6.1 Genesis of GIS The backbone for the evolution of GIS is computer technology (Foresman, 1998). There were other influences, certainly, including the shift in Geography toward quantitative methods in the United States in the 1950s, the environmental regulation age of the 1960s and 1970s, the ARAPANET Web network age of the 1970s, and the Global Change age of the 1980s. Eras in terms of GIS evolution started with the Pioneering Age in the 1950s to early 70s, followed by a Research and Development Age from the early 70s to the mid-80s, followed by the Implementation and Vendor Age up to the mid-90s, in turn followed by the Client Applications and Network Age in the 90s and 2000s. (Foresman, 1998) Many agree that the first GIS was the Canada Geographic Information System (Tomlinson, 1987, 1998) developed in 1963 under the direction and supervision of Dr. Roger F. Tomlinson under a project of the Canadian government to create an inventory of Canada’s natural resources. An important role in the founding of GIS as a discipline in academia was played by the Urban Regional Information Systems Association (URISA), founded by Dr. Edgar Horwood, a professor at University of Washington’s Department of Civil Engineering and Urban Planning. Before 1960, Horwood was offering a course on “geocoding” and computer mapping techniques. He had developed databases and linked records to geographical features, an operation he called geocoding. The software package developed by Horwood was well received at a workshop he gave in Chicago, well attended by regional scientists, quantitative geographers, and cartographers. (Tomlinson, 1988) 59 One of the earliest institutions to use GIS was the Minnesota Land Management and Information System (MLMIS). This institution resulted out of a joint research project between the State Planning Agency of Minnesota, and the Center for Urban and Regional Affairs (CURA) in 1967 (McMaster & Manson, 2010). The goal of the project was to map the lake shores of Minnesota, the “State of 10,000 lakes.” The Statewide Lakeshore Development Study led to the production of a statewide land use map based on spatial data from the United States Public Land Survey System (PLSS). The combination of the software, hardware, and the Environment Planning Programming Language (EPPL) used to produce this map was arguably one of the first true GIS systems in the United States. (McMaster & Manson, 2010; Tomlinson, 1988) The Harvard Laboratory for Computer Graphics, founded by Howard Fisher in 1965 was a milestone in the early history of GIS, because it led to the development of one of the first GIS software packages in the United States, the Synagraphic Mapping Package (SYMAP) (Chrisman, 1998, 2006). After having worked at the Harvard Laboratory for Computer Graphics and Spatial Analysis in 1969, Jack Dangermond left for California and founded Environmental Systems Research Institute (ESRI) in 1973 (Chrisman, 2006). This is the current industry leader in GIS software on a global scale and was one of the first successful commercial GIS companies in the world. Next, I discuss the diffusion of GIS in the developed world. 60 2.6.2 GIS Diffusion in North America, Europe and Australia GIS diffusion literature largely spans the period 1990 to 2000. This was the period when discussions about the diffusion of GIS technology strongly emerged in North America, Europe and Australia. A significant body of literature encompassing GIS diffusion can be traced back to a 1993 publication edited by Ian Masser and Harlan J. Onsrud entitled Diffusion and Use of Geographic Information Technologies. The book was a publication that resulted from papers prepared by participants in the NATO Advanced Research Workshop on “Modeling the Diffusion and Use of Geographic Information Technologies” held in 1992 in Sounion, Greece. Researchers at the National Center for Geographic Information and Analysis (United States), Economic and Social Research Council (United Kingdom), and others from various European countries including Germany, France, the Netherlands, and Portugal, among others, brought their diffusion research to the table for a comprehensive discussion about the need for GIS to address the needs of individuals, organizations, and institutions so as to ensure optimal use of geographic information innovations (Masser & Onsrud, 1993b). “While many of the concepts supporting GIS emerged in the 60s and 70s, the technology to support GIS has only become available during the last five to ten years. This has led to a multi-billion dollar industry …” (Masser and Onsrud 1993, 1). The diffusion of GIS was definitely assisted by developments in the computer industry. Hardware costs were significantly reduced with the invention of the personal computer in the early 70s by Hewlett Packard at the Palo Alto Research Center. This allowed GIS to be accessible to a 61 larger population, and to many more organizations than ever before. Sure there were still constraints in the technology (Goodchild, 1988) going from analog to digital computing (Tomlinson, 1988), but one thing was clear – the new cartography looked very promising. The main questions explored in the 1992 Sounion GIS Diffusion workshop were as follows (Masser & Onsrud, 1993b): 1. Research methodology and interdisciplinary perspective – What can GIS researchers learn from research elsewhere on organizational behavior and technological innovation? 2. Assessing and modeling the diffusion of geographic information innovations – What are the main features of the GIS diffusion process? What criteria should be used to measure the organizational and institutional impacts of GIS? 3. Assessing cultural and institutional issues – To what extent is the adoption and utilization of GIS facilitated or impeded by the institutional and organizational context within which it takes place? How do natural cultural factors affect the diffusion process? 4. Mechanisms for facilitating the diffusion of GIS technology – What mechanism might be used at the regional, national, and international levels to facilitate the diffusion of GIS? Masser and Onsrud (1993) note that Rogers’ (1993) article in the same publication (Rogers, 1993) stipulates that the historical development of GIS over the past quarter century indeed has all the hall marks of a classic diffusion process. “He (Rogers) argues 62 that most of the key elements of this model, the S-shaped curve form, the notion of the critical mass which is required before an innovation can take off in terms of its widespread adoption, and the importance that must be attached to the activities of a limited number of champions in the early phases, can be used without any difficulty in the analysis of GIS diffusion” (Masser and Onsrud 1993, 3). At this workshop, it was evident that GIS had indeed taken off in Europe and North America, for example, in the British local government (Masser, 1993), the German Department of Surveying (Wegener & Junius, 1993), Australia (Zwart, 1993), Portugal (Arnaud, 1993), Italy (Rumor, 1993), Canada (Cartwright, 1993), and the United States (Huxhold, 1993). Next, I discuss the literature on GIS diffusion in developing countries. 2.6.3 GIS Diffusion in Africa, Asia and South America In the developing world, however, the diffusion of GIS was not going as smoothly as it was in the developed countries for several reasons. Whether GIS was appropriate technology (AT) for developing countries was an issue raised by some (F. D. R. Taylor, 1986, 1998; Yapa, 1991). Yapa (1991) argues that GIS is not appropriate technology for the developing world because of its high cost of purchase, maintenance, and expertise, however, he also recognizes the importance of GIS as a complementary technology to AT because it is useful in uncovering local natural resources. In this sense, GIS can indeed be considered an integral part of AT. He further suggests that to make GIS more accessible in developing countries, four things could be done, (1) relax proprietary rights over GIS software, (2) develop “public domain” GIS, (3) develop GIS software that is built on top 63 of existing software infrastructure in the third world, and (4) develop local information systems that are more inclusive of indigenous knowledge and encourage public participation. Taylor (1986) argues that the introduction of computer assisted cartography in Nigeria in the mid-1980s was characterized by a false start because of the antithetical nature of computer assisted cartography relative to the principles of AT. Computer assisted cartography required expensive main frame computers, was capital intensive and minimized labor. Further, the government of Nigeria was unaware of the importance of maps and frankly uninterested in spending hard earned foreign exchange on acquiring expensive computing machinery. He asserts that AT should be considered as technology that is appropriate for performing the tasks at hand. In order for computer assisted cartography to be considered appropriate, there is a need for a thorough understanding of the tasks and the socio-economic setting in which they are to be performed. Nonetheless, some evidence of GIS diffusion in developing countries emerged in the 80s and 90s, for example in India, Brazil, Sudan, Egypt and South Africa (Borges & Sahay, 2000; Cartwright, 1993; Harris et al., 1995; Sahay, 1998; Sahay & Walsham, 1996; Salem, 1994; Weiner & Harris, 1999). Salem (1994) examines the current status of GIS in Egypt and argues that the growth of GIS is linked to a growing interest in the sound management of Egypt’s environment and natural resources. The availability of inexpensive computers is also partly the reason for the boom. He asserts that despite this interest, the slow penetration of GIS in Egypt is partly due to a fundamental problem resulting from the way GIS training is carried out at organizations. This training is 64 usually organized by the donor agencies that originally set up the GIS system at the receiving agencies in Egypt, such as USAID, and the German and Dutch international development agencies. Such training consists of short term courses, from one to six weeks, administered by visiting expatriates from the parent countries. Much as these courses are useful, they do not allow for a deep understanding of the principles of GIS given the ad hoc manner in which the training is administered. He suggests that a deeper level of training at the university level is required so as to impart GIS skills in future well-groomed GIS specialists in Egypt, and this will require incorporating GIS into the curriculum at the many Egyptian universities. Sahay and Walsham (1996) developed a theoretical framework that emphasized the importance of context and process in GIS implementation within an organization based on their longitudinal study of Information Systems implementation in the Ministry of Forestry and Environment in India. For successful GIS implementation, they stress the importance of understanding the influence exerted by the various groups of actors involved in the actual implementation process of GIS. Sahay (1998) goes beyond simple narratives of the failure in implementation of IT systems by managers in the developing countries to argue that notions of time and space are at the heart of the problem. With reference to technology transfer, GIS technology is developed in the West, to be implemented in the global south. The temporal and spatial context within which the technologies are developed, versus that within which the technology is implemented are in conflict, which is an obvious misnomer with regard to the implementation of IT systems in developing countries. Notions of time and space, in India for example, are 65 very different from those in the West. Whereas GIS emphasizes a positivistic epistemology, based on a rationalist view of the world, the receiving society in India has embedded in its culture a subjective, place-based, cosmological and traditional understanding of the world. Space and place are not well represented by a GIS because GIS development is closely linked to the quantitative revolution of the 80s in geography departments in Western universities. This is a problem in the transfer of GIS technology to developing countries. Further, there is a need for a gradual approach to the implementation of information systems in an organization, and a critical assessment of anticipated costs is needed. The importance of public participatory GIS, indigenous knowledge and the link with nature and environment has been stressed by some authors (Harris et al., 1995; Weiner & Harris, 1999). In South Africa, Harris and others (1995) used a public participatory GIS framework to implement a multi-media based community integrated GIS to involve local communities in unpacking the political ecology of the central Lowveld in a post- apartheid era. The authors demonstrated the importance of making GIS relevant to local communities in developing countries by including indigenous knowledge into the GIS to reduce the power differential created by GIS; that is, between the technocrats, or experts, and marginalized peoples. According to Cartwright (1993), urban information systems have seen slow penetration levels into the institutional framework of urban planning agencies in the third world. This is mainly because of a lack of computer literacy among the staff in the concerned organizations. He suggests that there is a need for better access to hardware, user-friendly 66 software, adequate training and advice. He also argues that excessive enthusiasm for information systems can have negative impacts, for example, managers acquiring sophisticated information systems that no one knows how to use. There is a need for a careful needs assessment before such decisions to acquire sophisticated software and hardware are made. Next, I discuss GIS literature specifically with respect to African countries. Literature on GIS in Africa emerged in the 2000s, in particular, literature on GIS in Kenya, Uganda, Botswana, Namibia, Nigeria, and Lesotho (Cavric et al., 2003; Fadahunsi, 2010; Kalande & Ondulo, 2006; Karikari et al., 2005; Koti, 2004; Letsie, 2008; Musinguzi et al., 2004; Musinguzi, Tickodri-togboa, & Bax, 2010; Noongo, 2007; Uwayezu, 2010). The major road blocks to GIS diffusion in developing countries are articulated in a study of the land sector in Ghana (Karikari & Stillwell, 2005; Karikari, Stillwell, & Carver, 2003) and in the study of the diffusion of GIS in Botswana (Cavric et al., 2003). According to the case study of GIS in Ghana’s land sector (Karikari & Stillwell, 2005; Karikari et al., 2003), some of these hindrances to GIS diffusion include a lack of adequate human and financial resources in developing countries, challenges in service delivery due to inappropriate organizational structures, poor maintenance of hardware and equipment, and inadequate and inconsistent spatial data at various public sector agencies in terms of coverage, accuracy and format. Moreover, there is a lack of policy regarding land registration regulation, which translates into challenges in using GIS for developing a national digital cadastre. This last finding resonates with that of Kalande and Ondulo (2006) who assert that there is a lack of geoinformation policy in 67 Uganda, Kenya and Tanzania which is currently preventing cooperation in terms of sharing of geospatial data among the three East African nations. Uganda and Kenya, for instance, have a number of policies that are essential for supporting a geoinformation (GI) policy; however, most of the legislation is outdated and does not adequately cover GI. For the development of national spatial data infrastructure (NSDI), there is an urgent need to go beyond mere copyright laws, which currently exist to protect intellectual property rights, to the formulation of contractual and extraction laws that prevent the risk and fear of unfair advantage. Other challenges to NSDI development in developing countries are attributed to the under development and foreign aid (Musinguzi et al., 2004, 2010). There is a lack of adequate ICT infrastructure, limited trained personnel, and limited funding from the government. Foreign aid agencies involved with GIS lack coordination in their activities related to the use of GIS (Lance & Bassole, 2006; Musinguzi et al., 2004). In addition, there is a lack of standards with regard to metadata, accuracy, spatial data quality, which in turn affects GIS data interoperability (Alford, 2009; R. M. Moyo & Chuba, 2001; Musinguzi et al., 2004, 2010; Woldai, 2002). Moreover, the appropriate uses of GIS in developing countries are those that will first consider the socio-economic, political and institutional contexts within which the GIS is to be implemented (Cartwright, 1993; Karikari et al., 2003; F. D. R. Taylor, 1998). The introduction of a new technology within the institutional framework of a developing country may initially represent cultural imperialism (Cartwright, 1993). The diffusion of an innovation in any foreign society will depend on the socio-technical relationship 68 between the technology, its users and its uses. “Is the technology appropriate to the circumstances in which it is being applied, given the resources available and the results desired?” (Cartwright 1993, 264) Indeed, institutional constraints are a major factor to consider in the diffusion of GIS in developing countries (Cavric et al., 2003; F. D. R. Taylor, 1986, 1998, 2004). In Botswana, the lack of a central GIS institution at a national level to help with the implementation of GIS is identified as one of the major reasons for the failure of GIS diffusion in that country (Cavric et al., 2003), which resonates with findings of other authors (Felicia O Akinyemi, 2011; Felicia Olufunmilayo Akinyemi, Uwayezu, & Simbizi, 2011; Budhathoki & Nedović-Budić, 2007a; Cartwright, 1993; Yola Georgiadou, Budhathoki, & Nedović-Budić, 2011; Lwasa et al., 2006, 2005; Muhwezi, 2006; Nyemera, 2008). In addition, the lack of trained personnel and data standards have also been identified as challenges to GIS diffusion (Cavric et al., 2003) which echoes the findings of other authors (R. M. Moyo & Chuba, 2001; Musinguzi et al., 2004, 2010; Nebert, Reed, & Wagner, 2007; Nyemera, 2008; Salem, 1994). It is very interesting to note that the authors that actively published on GIS diffusion in the 1980s and 1990s eventually started publishing within a new body of literature: Spatial Data Infrastructure (SDI). This makes sense in the developed countries that were transitioning from the diffusion of GIS into a more mature form of GIS with a new set of problems, especially problems surrounding data access, copyright, organization, ontology and so on. There are various publications on SDI from the late 1990s through the 2000s to the present day, for example, see De Vries and Lance 2011; Miscione and 69 Vandenbroucke 2011; Georgiadou, Budhathoki, and Nedovic-Budic 2011; Vancauwenberghe, Crompvoets, and Vandenbroucke 2011; Nedovic-Budic, Pinto, and Warnecke 2011; Dessers et al. 2011; Koerten and Veenswijk 2011; Richter, Miscione, and De 2011; C. A. Davis and Fonseca 2011; Nedovic, Crompvoets, and Georgiadou 2011; Silva 2011; Janssen, Crompvoets, and Dumortier 2011. This literature is quite comprehensive and covers both the developing world and the developed countries. In developing countries context, the SDI literature mainly focuses on the challenges to NSDI in these countries. Some have linked the poor state of NSDI in developing countries to the lack of financial support and commitment from national government, a lack of coordination in GIS activities among donor agencies (Lance & Bassole, 2006), a lack of metadata and spatial data quality standards (Musinguzi et al., 2004) which leads to uncertainty and a lack of trust among agencies using GIS data, which in turn leads to redundancy in geospatial data due to duplication of data collection efforts (De Vries & Lance, 2011; Nyemera, 2008). Others have argued that the problem is due to a lack of geoinformation policy and unwillingness to share spatial data at zero cost (Felicia O Akinyemi, 2011; Felicia O. Akinyemi, 2007; Uwayezu, 2010). Despite the roadblocks, a successful NSDI framework has been implemented in South Africa with a decentralized top-down approach to SDI policy implementation integrating SDI activities at the national, provincial, and local levels of government, although a bottom-up approach would be desirable so as to be more inclusive of user-focused and user-driven policy issues at the bottom of the power pyramid (Smit et al., 2009). A detailed analysis and review of the state of NSDI on the 70 African continent is provided by Makanga and Smit (2010), and it reveals that as of 2008, there were only three countries on the continent with NSDI clearinghouses, Kenya, Gabon, and Chad, despite the fact that there are nineteen African countries with NSDI coordination bodies. Next, I discuss the evolution of GIS in developing countries, with a focus on Africa. 2.6.4 Evolution of GIS in the Developing World with a Focus on Africa In this subsection, I draw upon the work of Emma N. Noongo (2007). Chapter three of her PhD dissertation, “GIS implementation in Namibia,” is entitled GIS in Developing Countries, and it synthesizes her comprehensive literature search and qualitative research on the history, implementation and development of GIS in the developing world since the 1970s, with a special focus on the challenges faced by African countries. I also draw upon D. Wayne Mooneyhan’s book chapter, International Applications of GIS (Mooneyhan, 1998), which provides insights into the introduction of GIS in Africa, South America, and Asia. I will also draw upon additional literature that relates to the subject (for example, Borges & Sahay, 2000; Cartwright, 1993; Drichi, 2002; Fadahunsi, 2010; Gowa, 2009; Sahay & Walsham, 1996; Taylor, 1986, 1998; Tchindjang et al., 2005; Turyareeba & Drichi, 2001; UNEP, 1972; Yapa, 1991). According to the Free Dictionary on the Web (thefreedictionary.com), the definition of a developing country is as follows: “(Economics) a non-industrialized poor country that is seeking to develop its resources by industrialization.” Generalizing from this definition, the countries located in Central America, South America, Africa, and Asia have (stereotypically) been labeled (by societies in Europe and North America) as developing 71 countries. My reference to developing countries is in light of the understanding stipulated hitherto. The entry of GIS into developing countries is solely attributed to the efforts of one organization: the United Nations Environmental Programme (UNEP), and more generally, the parent organization, the United Nations (UN) (Mooneyhan, 1998). In 1972, a United Nations conference was held on the Human Environment in Stockholm, Sweden (UNEP, 1972). The theme of this conference was to protect the earth’s environment through environmental monitoring and assessment. Excerpts of the summary of the conference proceedings were as follows: “34. The Conference was launching a new liberation movement to free men from the threat of their thralldom to environmental perils of their own making....” “36. Developing countries could ill-afford to put uncertain future needs ahead of their immediate needs for food, shelter, work, education and health care. The problem was how to reconcile those legitimate immediate requirements with the interests of generations yet unborn. Environmental factors must be an integral part of development strategy; one of the most encouraging outcomes of the preparatory process had been the emergence of a new synthesis between development and environment.” “38. The vast benefits which the new technological order had produced were undeniable, but man's activities had created serious imbalances. Not only each society but the world as a whole must achieve a better balance among the major elements that determined the level and quality of life it could provide for its 72 members-population and its distribution, available resources and their exploitation, and pressures placed on the life systems that sustained it.” “39. Conference action was required on the proposed Declaration on the Human Environment; on the proposed Action Plan for the Human Environment; on the proposed Environment Fund – which would be additional to moneys, which Governments made available for development programmes; and on the vitally needed organizational and financial machinery by which it would be possible to continue the work begun with the calling of the Conference.” “40. Certain priorities that required urgent and large-scale action had emerged from the preparatory process; these included water supplies, ocean and sea pollution, and the urban crisis. In addition, there were other areas for priority action: the need for understanding and controlling the changes man produced in the major ecological systems; the need for accelerating the dissemination of environmentally sound technologies and for developing alternatives to existing harmful technologies; the need to avoid commitment to new technologies before adequately assessing their environmental consequences; the need to encourage broader international distribution of industrial capacity; and the need to assist developing countries to minimize environmental risks in their development strategies.” (UNEP, 1972) 73 This conference led to the establishment of a new arm of the UN called UNEP, headquartered in Nairobi, Kenya. The organization’s mission was “to provide leadership and encourage partnership in caring for the environment by inspiring, informing, and enabling nations and peoples to improve their quality of life without compromising that of future generations” (Mooneyhan 1998, 350). The primary responsibility of UNEP was to assess the state, trends, and problems of the environment on a global basis, and to provide early warning of impending environmental dangers (Mooneyhan, 1998). At the conference, a coordinating mechanism was also created called “Earthwatch” whose role was to coordinate the activities of the UN bodies with governments and scientists in the quest to gather data for the comprehensive assessment of environmental issues. Within UNEP, a specific organization was established to coordinate the monitoring and assessment activities, Global Environment Monitoring System Programme Activity Center (GEMS/PAC). It was also in 1972 that the National Aeronautics and Space Administration (NASA) launched the Earth Resources Technology Satellite (ERTS1, later known as Landsat) in support of a civil space science program. High quality satellite data for the entire earth surface became available for the first time in history to anyone who requested for it under the so called “open skies” data release policy of NASA. (Mooneyhan, 1998) The availability of satellite data had two profound impacts on the history of GIS development. First, georeferenced data was available for the entire earth’s surface for the first time, and this data supported spatio-temporal applications in GIS such as environmental monitoring and change detection. Second, Landsat satellite data was huge 74 in terms of size, at the time, relative to the available computing capacity in terms of computer technology – processing times were impractical. This led to research and development in both software and hardware technologies to cope with the available data. The NASA Earth Resources Laboratory in Mississippi was involved in this development in the 1970s, and by 1983 had released software with high image processing capability, and a raster-based GIS capability known as ELAS (Earth Resources Laboratory Applications Software). (Mooneyhan, 1998) In 1984, UNEP signed a Memorandum of Understanding with NASA in which NASA was to provide software, hardware, and technical expertise to transfer image processing and GIS technology to UNEP and its member countries. NASA also agreed to provide all of its regional and global datasets to UNEP for distribution to its member countries at zero cost. UNEP named this project the Global Resource Information Database (GRID), and the first center was opened in Geneva in 1985 and Nairobi in 1986. UNEP received more donations at this time, including hardware from a couple of computer companies (and important to this dissertation), including vector GIS software from Environmental Systems Research Institute (ESRI). This fact is of major significance to this dissertation because it establishes the roots of ESRI software in Uganda, and other developing countries in general (see research findings in Chapters five and six). Further, the Swiss Directorate for Development Cooperation also funded a training program for developing country participants in that same year (Mooneyhan, 1998). This is also significant to my research because a lot of early training in GIS in Uganda was done in Switzerland according to my interviews with research subjects (see Chapter five). 75 During the early days of UNEP (1985-1989), GRID conducted a number of case studies and demonstration projects with strategic national and international partners as “proof-of- concept” for GIS applications (Mooneyhan, 1998). “The purpose of these projects was to confirm to UNEP’s governing bodies, national governments and the International community that spatial data models and GISs were necessary tools for complex environmental assessments and resource management” (Mooneyhan 1998, 352). Country participants confirmed if they had the necessary data for projects in their countries, while GRID provided all the satellite data. “The countries in which GRID helped carry out national or sub-national case studies include: Argentina, China, Costa Rica, Indonesia, Kenya, Nepal, Panama, Peru, Saudi Arabia, Thailand, and Uganda” (Mooneyhan 1998, 353). It can be argued that GIS was introduced into developing countries by UNEP- GRID in the period 1985-1989. A map showing this fact is illustrated in Figure 2-2. 76 Figure 2-2: The First Countries in the Developing World to Receive GIS Technology, through UNEP-GRID 2.6.4.1 The Introduction of GIS in Uganda A national case study in Uganda was carried out by GRID in collaboration with the Ministry of Environment Protection (Mooneyhan, 1998), which had recently been established in 1987 (Gowa, 2009) immediately after a civil war of succession that ended in 1986. The case study was specifically oriented towards an analysis of the current distribution of forest and wetlands, and changes in these patterns over a ten-year period. It included a retrospective analysis, based on satellite data, by a team of Ugandan 77 resource managers and ultimately concluded with the creation of a nationwide thematic atlas in digital format (Mooneyhan, 1998). The inventorying of forests in Uganda was carried out under the National Biomass Study, a project established in 1989 by the Ministry of Environment Protection in collaboration with the Norwegian Forestry Society, and funded by the Norwegian Agency for International Co-operation (NORAD). This project was executed by the Forestry Department, under the Ministry of Environment Protection (Drichi, 2002). 1989 was an important year in the GIS history of Uganda because it was in the same year that Uganda’s first (GIS) environment information center, the National Environment Information Center (NEIC) was established, although this had been in the pipeline already under a government (Ministry of Environment Protection) project initiated in 1987 (Gowa, 2009). NEIC was established with the support of UNEP-GRID and the World Bank Technical Division, Africa Region (AFTEN) (Gowa, 2009). It can be inferred that GIS was introduced in Uganda through the efforts of UNEP-GRID in the period 1987-89. 2.6.4.2 Early Initiatives that Supported the Diffusion of GIS in Africa The need for regional cartographic centers in Africa was first voiced at the first United Nations Cartographic Conference held in Nairobi, Kenya, in 1963 (Tchindjang et al., 2005). In 1964, at the 8 th session of the United Nations Economic Commission for Africa (UNECA) in Addis Ababa, Ethiopia, a call was made for the creation of regional centers for training in photo interpretation, photogrammetry, and airborne geographical surveys (Tchindjang et al., 2005). Between 1972 and 1987, five regional centers for remote 78 sensing had been built and established on the African continent by the United Nations Economic Commission for Africa (Ribot, 1989, in Tchindjang et al. 2005). These included centers in Burkina Faso, Egypt, Kenya, Nigeria and the Democratic Republic of the Congo (Zaire). The Zaire center was largely non-operational. In 1998, two additional centers were opened in Nigeria, and Morocco (Balogun, 2004; Touzani, 2006). A table showing the centers and their roles in capacity building in Africa is shown below (Table 2-1). 79 Table 2-1: Regional Centers for Remote Sensing, Photogrammetry, Surveying and Geodesy in Africa, 1972-1998 Source: (Balogun, 2004; Tchindjang et al., 2005; Touzani, 2006) Country Center Name Creation Date Training Period Training Program & Schedule Training Duration No. of Students Burkina Faso Centre Régional de Télédétection de Ouagadougou (CRTO) Remote sensing Regional Centre of Ouagadougou 1978 1978- 1988 Professional and specialized training based on remote sensing, topography, hydrogeology and cartography as well as air photographs 6-9 months 268 Egypt Remote Sensing Centre, Cairo 1978 Training and involvement in remote sensing projects at national level No Data No Data Kenya Regional Centre for Services in Surveying, Mapping and Remote Sensing (RCSSMRS) Nairobi 1975 1979- 1989 Remote sensing, cartography, geodesy and photogrammetry. 1-3 weeks, 6- 12 months 1000 Nigeria Regional Centre for Training in Aerial Surveys (RECTAS) Ile Ife 1972 Sensitization and Training in remote sensing and photogrammetry. 6-26 months 500 African regional Centre for Space Science and Technology Education (ARCSSTE-E) 1998 Develop skills and Knowledge in four prime areas – Remote Sensing and Geographic Information Systems – Satellite Meteorology – Satellite Communications – Basic Space Science and Atmospheric 80 Sciences. DR Congo (Zaire) Unknown 1978 Has never functioned No Data No Data Morocco African Centre for Space Science and Technology (CRASTE-LF) 1998 Figure 2-3: Regional Centers for Space Science and Technology Education in Developing Countries Source: (Touzani, 2006) In the developing world in general, a map by Touzani (2006) shows three additional regional centers affiliated with the United Nations Committee on the Peaceful Uses of 81 Outer Space (COPUS) by June 2003 – see Figure 2-3. In addition to the regional centers in Morocco and Nigeria, the additional three centers are in Mexico, India and Brazil. The focus of the pioneering regional centers established in Africa, Latin America and Asia in the 1970s was clearly on remote sensing, aerial photogrammetry, land surveying, cartography and earth science. Of course, GIS was still in its infancy in the 1970s, and in fact, widely known as automated cartography in the 1980s (Sui & Morril, 2004; F. D. R. Taylor, 1986, 1998). These regional centers were the backbone for the diffusion of GIS into the developing world. The organization solely credited with this diffusion is the United Nations, through its agencies such as UNEP/GRID, UN-ECA, and UN-COPUS (Balogun, 2004; Mooneyhan, 1998; Noongo, 2007; Tchindjang et al., 2005; Touzani, 2006) with financial support of the World Bank agencies like AFTEN. UN-ECA established the first regional remote sensing centers between 1972-1978, while UNEP/GRID leveraged and harnessed the little capacity that had been built by these centers to actually introduce GIS technology between 1985 and 1989 to developing countries by establishing GIS centers in host countries in collaboration with the governments in those developing countries, for example, in the case of Uganda (Gowa, 2009), through the Ministry of Environmental protection. There is one more agency that played a critical role in the diffusion of GIS in the developing world, the United Nations Institute for Training and Research (UNITAR). This agency of the UN was responsible for capacity building in developing countries (Mooneyhan, 1998). Between 1985 and 1989, UNITAR trained several people in developing countries in GIS. One such project of UNEP/UNITAR provided hardware, 82 software, data and training to over 100 persons in eleven African countries, and resulted in the establishment of national GIS centers in each of the countries. These included (Mooneyhan, 1998): 1. Botswana – center established in the Department of Town and Regional Planning, Ministry of Local Government, Lands and Housing 2. Burkina Faso – center established in the Department of Forestry, Ministry of Environment, and Tourism 3. Cote d’ Ivoire – center established in the Service Autonome de Teledetection, Direction Generale et Controlle de Grandes Traveaux 4. Ghana – center established in the Environment Protection Council 5. Kenya – center established in the National Environment Secretariat 6. Lesotho – center established in the Environment Division, Lesotho Highlands Development Authority 7. Mali – center established in the Project d’Inventaire des Resource Terrestres 8. Mozambique – center established in the Environment Division, National Institute for Physical Planning 9. Niger – center established in the Department of Hydraulics, Ministry of Environment and Hydraulics 10. Tanzania – center established in the National Environment Management Council 11. Uganda – center established in the Ministry of Environment Protection, and later evolved in to the National Environment Information Center (NEIC), and then into the National Environment Management Authority (NEMA) (Gowa, 2009) Similar programs in the Asia/Pacific region involved the following countries: Bangladesh, Bhutan, Cambodia, China, Fiji, India, Indonesia, Lao P.D.R., Maldives, Myanmar, Nepal, Pakistan, Sri Lanka, Thailand, Vietnam, and Western Samoa (Mooneyhan, 1998). 83 GRID has four centers spread across the globe that have regional responsibility, are funded by UNEP and staffed by UNEP, as well. These centers include (Mooneyhan, 1998): 1. Geneva, Switzerland – opened in 1985, responsible for Europe plus global dataset maintenance and distribution 2. Nairobi, Kenya – opened in 1986, responsible for Africa and UNEP Internal 3. Bangkok, Thailand – opened in 1989, responsible for Asia and the Pacific 4. Sioux Falls, South Dakota – opened in 1991, responsible for North America, and for the generation of new global and regional datasets There are seven national or regional centers that receive some funding from UNEP, but are staffed and managed entirely by the national governments or regional organizations; these centers include (Mooneyhan, 1998): 1. Arendal, Norway – opened in 1989, responsible for Norway and polar region datasets 2. Tsukuba, Japan (National Institute for Environmental Studies, NIES) – opened in 1990, responsible for Japan, and supercomputing processing requirements 3. Warsaw, Poland – opened in 1991, responsible for Poland and selected regional environmental studies 4. Sao Jose dos Campos, Brazil (Instituto Nacional des Pesquisas Espacias, INPE) – opened in 1992, responsible for Brazil and datasets for the Amazon region 5. Kathmandu, Nepal (Intenational Center for Integrated Mountain Development, ICI-MOD) – opened in 1992, responsible for datasets of the Himalayan region and GIS training for eight member countries 6. Apia, Western Samoa (South Pacific Regional Environmental Program, SPREP) – opened in 1993, responsible for 22 island countries in the region; and 7. Copenhagen, Denmark, opened in 1994, responsible for Denmark and selected regional environmental studies 84 2.6.4.3 The Lackluster Diffusion of GIS in Developing Countries Despite the regional centers opened by UNEP and the training provided by UNITAR, the diffusion of GIS skills and technology has been relatively slow in the developing world, and this could be because of several reasons. Mooneyhan (1998) argues that the training efforts of UNEP/GRID/UNITAR have been a failure, not in terms of the quality of training, but the less than impressive long terms effects of the training. One reason for this is the lack of hardware and software in developing countries. There is no use providing training to persons, yet they do not have the equipment to continue practicing what they have learned after they return to their home countries. Mooneyhan (1998) advises that trainees should be allowed to return home to their countries with hardware and software so as to ensure success in the long term training objectives. Second, when trainees returned home to their institutions, they were often times assigned duties that had nothing to do with GIS at all. This was because these institutions had their own priorities, methods of work and operating rules. In some cases trainees returned home to find a changed political situation, or no support for follow-up activities. Mooneyhan (1998) advises that whenever serious training efforts are provided to trainees, the providers should be certain that the selected institution has a clear mandate from the appropriate governing body and that there is commitment to institutionalize the new capability. Third, the lack of follow-up by the trainer led to a demoralizing effect on the trainees. After initial training, the trainees had real need for support in terms of technical and moral support, yet this was not available. Much as UNITAR tried to create a newsletter 85 as a means of maintaining a regional professional community and forum, this effort was less than effective in terms of follow-up activities, because the trainees problems were more complicated than that. Mooneyhan (1998) suggests that help desk support should have been provided for particular systems that had been selected for operational use, and at that time (1998) had envisioned that the Internet would be a possible solution were it to become affordable in developing countries. Fourth, the problem of “brain drain” eventually occurred, and this is common problem in any technology transfer program. The trainees, as a result of their training, achieved the status of better trained professionals with rare skills, which made them more market and thus, more mobile, both internally within their own countries and institutions, and abroad. Thus, the skills acquired did not necessarily transfer to the home institutions that they were meant for. And one cannot blame the new graduates, as the opportunities at home were not as attractive as abroad. Mooneyhan (1998) advises that this problem could be tackled by sensitizing the bureaucracy in the home country to the newer technology so as to influence a transition from old analog methods to new digital methods, for example, from paper/pen cartography to GIS, which would lead to job creation in the home institutions. Next, I discuss the constructs of GIS and Society as a framework for understanding the impacts of GIS on society. 2.7 GIS and Society GIS and Society is a research agenda that emerged out of an intellectual debate in the early 90s in the United States about the impact of GIS on society, and vice versa, how 86 society shaped GIS. In this dissertation, I use GIS and Society to frame my discussion on the impacts of GIS on the society in Uganda, and developing countries in general. First, I provide some historical background to GIS and Society as a research agenda in the Geography discipline. Second, I highlight the five conceptual perspectives of GIS and Society, under which most of the literature falls. Third, I discuss the relevance of GIS and Society to developing countries. 2.7.1 History of GIS and Society In the early 90s, there was an intellectual rift between two groups of scholars in the discipline of Geography. These two groups included: (1) the GIS specialists and (2) the social theorists. The first group included cartographers, remote sensing specialists and others in geography whose background was in the mapping sciences. These scholars, who were also geographers, were more aligned with the technical aspects of GIS, and had a tendency to employ quantitative methods of inquiry and research in their work, ignoring a host of concepts embedded in the traditions of the social sciences. The second group consisted mainly of human geographers whose work and research was more aligned with qualitative methods of inquiry and heavily engaged social theories, not only in the tradition of geography, but also other social sciences, such as political science, sociology, and anthropology. This rift had been growing over the years, especially with the emergence of GIS as a promising and strong sub-discipline within Geography in the late 80s, with the establishment of several supporting institutions over the years, such as the National Center for Geographic Information and Analysis (NCGIA) in 1998 (NCGIA, 1990), 87 funded by the National Science Foundation (NSF); and the exclusion of human geographers from the debates within the GIS community. In a way, one can situate the critiques voiced by the human geographers as critiques of technological determinism (for example, Latour 1987). Human geographers proposed that there should be an interdependence between technology and society, because at the end of the day, technology is only a means to address problems of society (Sheppard, 1993), the impact on society being more important than the technology used to achieve the societal objective. Roger P. Miller (1995) wrote in a paper presented in 1992 to the GIS/LIS conference (and later published in 1995 in The Journal of Cartography and Geographic Information Systems): “Increasingly complex social issues make it imperative that GIS practitioners be aware of theoretical debates in the human sciences, where arguments over ontological status and epistemological approaches have subsumed technical and methodological issues. Understanding the debates around deconstruction, postmodernism, and structuration is crucial to the development of useful geographic information systems that deal with social constructs. GIS needs to be informed about social theory to remain relevant in contemporary research, but GIS also can contribute to the further articulation of social theory.” (Miller 1995, 1) In 1992, GIS had been re-positioned by Goodchild (1992), one of the most distinguished scholars in GIS, as a Science instead of a System, and this proposal gained positive reviews in the GIS community. His main argument was that GIS was not a mere set of 88 technologies, but instead, a combination of technologies and people (society). In light of these arguments and rhetoric, to diffuse the heated debate between the two camps, a workshop was called in 1993 by Tom Poiker (Poiker, 1995) at Friday Harbor, University of Washington, sponsored by the NCGIA, so as to find some common intellectual ground between the feuding camps (Couclelis, Nyerges, & McMaster, 2011; Sheppard, 1993). This was followed by another meeting in South Heaven, Minnesota in 1996, also sponsored by the NCGIA. Two key publications resulted out of these workshops and interactions that followed, one by Pickles, Ground Truth (Pickles, 1995), and another, a special edition of the journal of Cartography and Geographic Information Systems edited by Eric Sheppard of the University of Minnesota (Sheppard, 1995). A new research agenda emerged, and it was called GIS and Society. Projects on GIS and Society were initiated so as to investigate the issues raised in previous workshops, such as the Varenius Project (Elmes et al., 2005; Goodchild et al., 1999; Harris & Weiner, 1996). Many publication on the subject have resulted (for example, Couclelis, Nyerges, and McMaster 2011; Nyerges, McMaster, and Couclelis 2011; Sheppard et al. 1999; Obermeyer 1993; Harvey and Chrisman 2004). Delegates of the University Consortium for Geographic Information Science (UCGIS) suggested a set of formal research questions that would form the backbone a GIS and Society research agenda, and these questions were translated into the five perspectives on GIS and Society as they are known today (Elmes et al., 2005; Getis, Anselin, Lea, Ferguson, & Miller, 2005). 89 2.7.2 The Five Perspectives of GIS and Society There are five perspectives of GIS and Society: (1) The Critical Social Theory Perspective, (2) The Institutional Perspective, (3) The legal and ethical perspective, (4) The intellectual history perspective, and (5) The public participation perspective (McMaster & Harvey, 2010; Nyerges et al., 2011). I discuss each of these next. 2.7.2.1 The Critical Social Theory Perspective This perspective is concerned with issues of power and representation in GIS (McMaster & Harvey, 2010; Nyerges et al., 2011). The GIS experts enjoy privilege and power as a result of their “rare” knowledge of GIS, and make decisions which can marginalize members of society who do not have the skills to use GIS. Further, representations of geographical features is at the discretion of the GIS experts, and this privilege could be misused for all sorts of reasons, for example, reshaping of district boundaries so as to allow a political party to win elections. In study of environmental equity and risk assessment, McMaster, Leitner and Sheppard (1997) showed how GIS could be used in a responsible way to look out for the interests of underrepresented neighborhoods. This study showed that in the Phillips neighborhood of Minneapolis, toxic release sites were located too close to unsuspecting marginalized populations; the African Americans and Latino population. The basic question under the social theory perspective is, “Does the use of GIS benefit society?” Another issue of concern is protecting the confidentiality of people inherent in spatial data provided to researchers (McMaster & Harvey, 2010), for example, in public health, demography, and in the social sciences. If this data was resolved to the individual 90 level, and landed in the wrong hands, it might affect the individuals’ quality of life, for example, insurance companies could hike up their rates because of the discovery in the data of a “preexisting condition”. 2.7.2.2 The Institutional Perspective This perspective is concerned with the implementation of GIS in institutions and the social issues that surround the process of implementation (McMaster & Harvey, 2010; Nyerges et al., 2011), such as costs and benefits, maintenance and coordination issues. It also addresses the development of theories and techniques for determining the impact of GIS on policy decisions and on expectations about the agencies carrying out the implementations. It further addresses the impact of GIS implementation on interactions between agencies, on citizens’ relations with government agencies and also people’s actions with respect to the use and management of land (Elmes et al. 2005, in R. B. McMaster and Harvey 2010). An organization that decides to implement a GIS will face myriad challenges. There will be impacts on the institutional structure of the organization, for example, a Ministry of Environment Protection in an African country. Will a new department have to be set up? How will decision making in the Ministry change as a result of this new GIS department? What about the costs of implementation – out of which budget will these come out? There will be a need to hire new staff with appropriate GIS skills, and a plan for regular training. Where will spatial data be obtained from? These are all issues that the institutional perspective on GIS and Society addresses. 91 2.7.2.3 The Legal and Ethical Perspective The legal and ethical perspective is concerned with the institutional processes and pricing mechanisms that govern access to spatial data (McMaster & Harvey, 2010; Nyerges et al., 2011). With the availability on the market of proprietary spatial databases, there is a need for the examination of governmental and legal regulations of such data, and the related ethical and legal implications of the changing institutional processes surrounding access to data (Elmes et al. 2005, in R. B. McMaster and Harvey 2010). With the increasing use of GIS data, ethical questions have emerged, such as, who is responsible for errors in spatial databases? This is important in the event of a disaster resulting from the use of spatial data, for example, accidental contamination of rivers by factories located in a watershed. Who is responsible for the luxury cruise ship, the Costa Concordia, that run aground off the coast of Italy in January 2012 due to a navigational error made by the captain? (BBC, 2012) What about intellectual property rights surrounding spatial data and related computer algorithms for modeling the data? Are spatial databases legal documents (McMaster & Harvey, 2010)? There are other related issues within this perspective. The issue of privacy and surveillance of society have been bellowed by some (G. C. H. Cho, 2008; Dobson & Fisher, 2003; McMaster & Harvey, 2010). High resolution satellite imagery has become pervasive on Web mapping applications like Google Earth. What are the implications for the privacy of citizens if this powerful data is used by government agencies for surveillance of citizens? Is it morally correct to track former inmates using GPS devices 92 (Dobson & Fisher, 2003)? McMaster and Harvey (2010) point out that some steps have been taken by the Urban and Regional Information Systems Association (URISA) to address these legal and ethical concerns about GIS data; “a GIS code of ethics” published in 2003. URISA acts as a governing body in the GIS community. This code of ethics articulates proper conduct of GIS professionals within the GIS community. The section on privacy has been extracted from the URISA (2003) website below: IV. Obligations to Individuals in Society The GIS professional recognizes the impact of his or her work on individual people and will strive to avoid harm to them. Therefore, the GIS professional will: 1. Respect Privacy •Protect individual privacy, especially about sensitive information. •Be especially careful with new information discovered about an individual through GIS-based manipulations (such as geocoding) or the combination of two or more databases. 2. Respect Individuals •Encourage individual autonomy. For example, allow individuals to withhold consent from being added to a database, correct 93 information about themselves in a database, and remove themselves from a database. •Avoid undue intrusions into the lives of individuals. •Be truthful when disclosing information about an individual. •Treat all individuals equally, without regard to race, gender, or other personal characteristic not related to the task at hand. (URISA, 2003) 2.7.2.4 The Intellectual History Perspective Another perspective on GIS and Society is the intellectual history perspective. This perspective concerns itself with tracing the evolution of GIS technologies. What are the dynamics that led to certain GIS technologies being selected over others? Why is it that the raster data structure used by Clark Labs succeeded, while the quad-tree data structure employed by the Tydac Corporation in their SPANS software failed to get a foothold in the market (McMaster & Harvey, 2010)? Why is ESRI the most successful GIS software vendor, globally, today and not Intergraph? What were the societal and institutional influences that shaped GIS as we know it today? Were there any productive alternative technologies that were overlooked and why (Elmes et al. 2005 in R. B. McMaster and Harvey 2010)? For instance, the concept of topology was borne out of the US Census bureau’s work on their proprietary DIME file data structure used for storing census data (McMaster & Harvey, 2010). This, together with the pioneering work of the Harvard 94 laboratory for Computer Graphics and Analysis, eventually institutionalized the topology-based georelational model in various GIS packages. 2.7.2.5 The Public Participation Perspective The fifth and final perspective on GIS and Society is the public participation perspective. This perspective seeks to find out how GIS can be more inclusive as a practice and as a technology, in essence, how can GIS be democratized (Kyem, 2004b; Pickles, 1995; Ramasubramanian, 2011). This perspective on GIS and Society led to a new research group within GIS and Society called Public Participatory GIS (PPGIS), or sometimes called simply Participatory GIS (PGIS). The objective of PPGIS is to empower local communities, community organizations, and grass roots groups with GIS technology, data and techniques. One of the most cited works in PPGIS is of Giacomo Rambaldi who has made a huge effort to bring GIS to some of the most remote parts of the world to some of the most marginalized communities, especially those in developing countries (J. Corbett et al., 2006; Rambaldi et al., 2006). Other works have been done within the public participation perspective on GIS and Society under banners such as community- based GIS (Assimakopoulos, 1996; S. Elwood, 2006a; Ghose, 2011; Leitner et al., 1998; Weiner & Harris, 1999), and indigenous people’s GIS (Kyem, 2000; Kyem & Saku, 2009; Laituri, 2011; Rundstrom, 1995). Next, I discuss the relevance of GIS and Society perspectives to developing countries. 95 2.7.3 Relevance of GIS and Society to Developing Countries with a focus on Africa A scan of the literature on GIS and Society eventually reveals a gap in as far as GIS in developing countries is concerned. The limited literature on GIS and Society in the developing world, has mainly centered on PPGIS in Ghana and South Africa (Kyem, 2000, 2004a; Kyem & Saku, 2009; Weiner & Harris, 1999), indigenous people’s GIS (Laituri, 2011; Rundstrom, 1995), issues surrounding the implementation of GIS for land management in Ghana and South Africa (Karikari & Stillwell, 2005; Karikari et al., 2003, 2005; Weiner & Harris, 1999), peri-urbanization in Kenya (Koti, 2004), SDI in Uganda, and other African countries (De Vries & Lance, 2011; GSDI, 2011; Kalande & Ondulo, 2006; Musinguzi et al., 2004; Onah, 2009; Smit et al., 2009), GIS implementation in Namibia, Botswana and Ethiopia (Beshah, 2003; Cavric et al., 2003; Noongo, 2007) and geo-information policy in Africa (Kalande & Ondulo, 2006; Lwasa, 2006; UNEP- DEWA, 2007). This research is very important and significant for development practices in developing countries. Projects in these countries oftentimes lack geographical organization and planning, evidence of which I provide in chapter four. The impacts of GIS on society in developing countries have been subtle, as the findings will reveal in chapters four through six. Questions asked in GIS and Society literature, originally meant to examine the impacts of GIS in developed countries, I argue, are also relevant to developing countries, in particular in examining the causes for the current levels of penetration of GIS technology in these countries. My research complements work done by aforementioned researchers 96 and provides GIS and Society perspectives based on the social systems, cultural norms and the political context within Uganda, and developing countries in general. Next, I review the literature on ICT, mobile phones and LBS. 2.8 ICT, Mobile Phones and Location-based Services Information and Communication Technology (ICT) has been cited by some as a promising avenue for promoting information diffusion (Zanello & Maassen, 2011), and in fact, Vinton Cerf, chief Technology Evangelist at Google and inventor of the Internet, has cautiously predicted that “underdeveloped countries could do some technology leap- frogging using things like mobile communications” (ACM, 2005). Cerf suggested that many people in the near future, especially in developing countries, will have their first interaction with the Internet via a mobile device, not the laptop. He noted how mobile devices have become more aware of their environment so as to request and receive location-aware and context-aware information from the Internet (Cerf, 2009) to offer geo- location based services, also known as location-based services (LBS). Moreover, mobile phone penetration in developing countries is growing. In Uganda, for example, mobile-cellular telephone subscribership per 100 inhabitants of the total population of 32 million people (GoU & UNFPA, 2010) in 2011 was at 48.38% (ITU, 2012), while in 2006 it was only 6.84% (ITU, 2012). This growth is significant because it reiterates Cerf’s cautious prediction about technology leapfrogging in underdeveloped countries (ACM, 2005). Developing countries tend to be characterized by poor infrastructure, thus, “ICT can provide a revolutionary tool to organize and get in touch with people at very low scale costs” (Zanello and Maassen 2011, 370). 97 I argue that mobile phones and location-based services have great potential as alternatives to full scale GIS technology in the future in infrastructure-poor environments to allow citizens to participate in spatial decision making by increasing spatial thinking and spatial awareness in society. Next, I describe the architecture behind LBS. 2.8.1 LBS Technology Location-based services (LBS), also known as Mobile Location Services (Jagoe, 2003), are one of the important applications of distributed GIS (Peng & Tsou, 2003). LBS refer to a system that offers real-time information about a location and its surrounding area (Jagoe, 2003; Peng & Tsou, 2003). This information allows users to determine the location of a destination and the best route to get there (Peng & Tsou, 2003). The information provided to the user of the mobile device is location-specific and customized based on the user’s personal profile (Raper, Gartner, Karimi, & Rizos, 2007a, 2007b). Location-based services can be categorized in two possible categories based on the perspective of the user. First is the End-user application perspective, which includes applications that provide Traffic and weather information, directions, entertainment, wireless advertising, store location, and so on (Jagoe, 2003). Second is the Developer and Vendor application perspective, which includes applications for mapping, routing (directions), GPS Navigation (real-time turn-by-turn navigation), proximity searches, destination guides, tracking (E-911, vehicles, assets, friend or buddy finders), telematics, location-based billing, advertising, and so on (Jagoe, 2003). LBS Technology is supported by cellular and wireless networks architecture. Cellular and wireless networks use radio waves as a means of communication. Radio signals used for 98 wireless communications are referred to as microwaves because of their very small wavelengths. The microwave spectrum ranges from 0.4GHz to 100 GHz, and their high bandwidth makes them ideal for communication. (Jagoe, 2003) Cellular Networks Mobile phone systems are comprised of a network of cells each with a powerful radio transmitter at its center. This is because, one, the radio signals most effective for carrying digital voice and data are short ranged, and two, the cellular design is modular and can provide redundancy and failover capabilities. The base stations are typically connected to each other via high-speed fiber, and then to the public phone system and the Internet. As a mobile device moves through the network it is passed from one base station to another accessing services through the base station of the cell it is in. (Jagoe, 2003) Wireless Data Sending wireless data over a standard second generation (2G) GSM (Global Systems for Mobile Communications) mobile network requires full-time use of the voice channel, and allows rates of about 14.4 kbps. Third-generation (3G) mobile systems provide an always-on data connection at vastly faster rates than is possible in 2G mobile communications. An intermediate step that is a less expensive and a relatively straightforward upgrade from GSM is GPRS (General Packet radio Service), sometimes also called 2.5G. GPRS is a packet-switched network, which uses bandwidth only when sending and receiving data. This allows it to be shared by numerous mobile devices at the same time, just as dial-up Internet users share one fast Internet connection from an ISP. The specification for GPRS allows it to provide up to 115 kbps. The mobile client device 99 communicates to the LBS application server using the mobile/cellular network. (Jagoe, 2003) LBS Architecture LBS architecture is based on client-server architecture. A mobile LBS client application generates a query, such as “where is my nearest gas station?” and sends it over the cellular wireless network to a server-side LBS application server. The application server consists of several layers, the most important being the presentation, business logic and the data access layers. (Jagoe, 2003) LBS server-side applications reside in the presentation layer, which work with the business logic layer to provide a response to the incoming query. The business logic layer obtains the current location of the client from a location server, and performs the necessary computations to provide an appropriate response. Depending on the query, necessary content is retrieved through the data access layer from various content servers, for example, a personalization server (for personal user-profile and preference information) and the commerce server (for billing and payment information). The response is passed back to the server-side LBS application in the presentation layer, which packages the information appropriately and in turn sends it to the client LBS application on the mobile device, through the communication layer over the Internet. This communication takes place via the Hyper Text Transfer Protocol (HTTP) on top of the wireless cellular network. This final response received by the mobile client could be in the form of driving directions from the current location of the user to the nearest gas 100 station (in textual form), and a map of the route. The architecture just described can be viewed in Figure 2-4. (Jagoe, 2003) Figure 2-4: Simple LBS Architecture Source: (Jagoe, 2003) Geopositioning in LBS Geopositioning is the determination of the location of a mobile device. This is because it provides the locational context for LBS. There are various position determination techniques (PDTs) in LBS. These include: (1) the Global Navigation Satellite Systems (GNSS), for example, the US’s Global Positioning Systems (GPS), Russia’s Global Navigation Satellite System (GLONASS), and the European Union’s GALILEO; (2) 101 Terrestrial wave-based positioning systems, such as Time of Arrival (TOA), Time Difference on Arrival (TDOA), Angle of Arrival (AOA), Cell-ID/Beacon/Proximity; and (3) Non-signal based PDTs, such as Micro-Electro-Mechanical Systems (MEMS), for example, accelerometers, gyroscopes, and digital compasses (Raper et al., 2007b). Raper and others (2007a), however, have challenged researchers to build a universal positioning device that can meet the demands of LBS in light of the diversity of constraints in geopositioning for LBS, for example, GPS positioning only works for outdoor environments, while Cell-ID works for indoor environments, but is characterized by lower accuracy. 2.8.2 Applications of LBS in Developed Countries Applications of LBS in developed countries are vast, and include mobile guides, intelligent transportation systems, location-gaming, assistive technology, and location- based health applications (Raper et al., 2007a). LBS have played significant roles in enhancing the users’ experience of urban areas and cities (Rizos, 2007) and influencing perceptions of space and place (Evans, 2011). The largest concerns about LBS technology today in the Western world revolve around the issue of privacy of user data (Abbas, 2011; Evans, 2011; Raper et al., 2007b) and accuracy of positioning (Curran et al., 2011; Curran & Hubrich, 2009; Dahunsi & Dwolatzky, 2012; Raper et al., 2007b; Zandbergen, 2012). 2.8.3 Applications of LBS in Developing Countries Scholarly literature specifically on LBS applications in developing countries is limited. This dissertation research borrows from literature describing mobile phone applications 102 in the developing world, especially literature on applications that leverage location of the mobile device. Bodies of literature that encompass such topics include ICT for Development (ICT4D) and Mobile for Development (M4D), for example, see (Aker & Mbiti, 2010; Banks & Burge, 2004; Dholakia & Kshetri, 2003; Donner, 2010; Furuholt & Matotay, 2011; Martin & Abbott, 2008; Masuki et al., 2010; Sey, 2008; Wicander, 2010; WorldBank & InfoDev, 2012). In these studies, general trends in mobile phone applications with relevance to LBS in the developing world can generally be categorized as (1) mobile-health, (2) agricultural, (3) social development applications. South Africa and India are examples of developing countries witnessing significant LBS penetration today. The most advanced LBS deployment and cellular network on the African continent is found in South Africa (Dahunsi & Dwolatzky, 2012). Two LBS applications used in South Africa are an emergency location service, and a person location service, and these employ the low accuracy cell-ID geopositioning technique (Dahunsi & Dwolatzky, 2012). In India, the future LBS market has been estimated at $20 Billion in light of the fact that India’s 3G and Mobile broadband infrastructure is ready to support an LBS market (Sharma, 2012). The most popular LBS application used currently in India is navigation, while the future holds promise for applications such as person location and tracking, for example, parents tracking their children; fleet management; and Point of Interest (POI) location (Sharma, 2012). Issues that could affect the future success of LBS in India include privacy, a feasible business model, marketing of the services, and infrastructure constraints (Sharma, 2012). 103 2.9 Conclusion This chapter has reviewed the literature with respect to the theoretical underpinnings of GIS diffusion, the various perspectives on GIS and Society, the history of GIS, and ICT, mobile and LBS. The chapter started by situating GIS within the broader Information Systems literature. GIS can be seen as a Decision Support System; however, as the literature review has shown, the Information Systems literature was formed from business applications of computers, and does not necessarily have an explicit link to GIS literature, which is embedded in the fields of geography, and urban and regional planning; and not computer science and business management. The chapter then reviewed theoretical frameworks for understanding the adoption, use and diffusion of technology. There are various theories on the acceptance and adoption of GIS and Information Systems technologies, however, the one most suited to this dissertation research is the Diffusion of Innovations theory mainly because it a broad framework that ties together most of the constructs posited across the other theoretical frameworks on technology use and adoption. This was followed by a review of the literature on the genesis of GIS as a technology and academic discipline, its evolution in the industrialized world, and the developing world. So as to ascertain the impact of GIS on society in developing countries, the five perspectives on GIS and Society discourse were reviewed as a framework for analysis. Finally, a review of the literature on ICTs, mobile phones, and LBS technology in the developing world was presented. The literature review in this chapter provides the framework for the empirical chapters that follow; that is, chapters four to seven. 104 Chapter 3 : Research Methodology 3.1 Introduction In this dissertation research project, I employ mixed methods to investigate my research questions. Mixed methods research is formally defined as, “A class of research where the researcher mixes or combines quantitative and qualitative techniques, methods, approaches, concepts, or language into a single study (Johnson & Onwuegbuzie, 2004, 17). I chose this research methodology mainly because I sought to take advantage of both qualitative and quantitative approaches to research, and in my particular case, I exploit the strengths of both methodologies to obtain the most appropriate form and quantity of data to address both the overarching and nuanced questions about the evolution and current “state” of GIS in developing countries, and the potential for LBS. In this chapter, I discuss the philosophical perspectives on the well-known research paradigms in qualitative, quantitative, and mixed methods research in section 3.2. This is followed by a justification for the use of mixed methods for this doctoral research in section 3.3. In section 3.4, I describe the actual methods used during field work, and in section 3.5, I discuss methods that I used for carrying out my analysis, including grounded theory and descriptive statistical analysis. Some weaknesses and the potential for biases in the methods used are presented in section 3.6. 3.2 Philosophical Perspectives on Research Research in the social sciences can be categorized under three major “research paradigms,” that is qualitative, quantitative and mixed methods research (R Burke 105 Johnson & Onwuegbuzie, 2004; R. B. Johnson, Onwuegbuzie, & Turner, 2007). By definition, a research paradigm is “a set of beliefs, values and assumptions that a community of researchers has in common regarding the nature and conduct of research (Kuhn, 1977 in R. B. Johnson et al., 2007, 130). A similar definition is provided by Denzin and Lincoln (1994, 99): “A paradigm is a basic set of beliefs that guide action. Paradigms deal with first principles or ultimates. They are human constructions. They define the worldview of the researcher-as-bricoleur. These beliefs can never be established in terms of their ultimate truthfulness.” These beliefs include ontological beliefs, epistemological beliefs, axiological beliefs, aesthetic beliefs and methodological beliefs (R. B. Johnson et al., 2007). A research paradigm, in other words, is a research culture, and it consists of the trilogy, that is qualitative, quantitative, and mixed methods research (R. B. Johnson et al., 2007). The term paradigms was originally coined by Thomas Kuhn (1962, viii) and defined as “universally recognized scientific achievements that for a time provide model problems and solutions to a community of practitioners.” Denzin and Lincoln (1994), interestingly enough, draw a distinction between the terms “paradigm” and “perspective.” They assert that perspectives are not as solidified, or as well unified, as paradigms, although a perspective may share many elements with a paradigm, such as a common set of methodological commitments. To understand the meaning of each of these research paradigms, it is necessary to define some key terms in philosophy: ontology and epistemology, and how they relate to methodology and method. These are important terms because they are part and parcel of 106 the definition of a given paradigm, which encompasses the three elements: epistemology, ontology, and methodology (Denzin & Lincoln, 1994). A discussion on ontology and epistemology can be found in Appendix J: Philosophical Perspectives on Research. In the next three sub-sections, I discuss the three commonly used research paradigms: qualitative, quantitative and mixed methods. 3.2.1 Qualitative Research Paradigm The qualitative paradigm is based on interpretivism, and constructivism (Sale et al., 2002). Ontologically, in qualitative research, there are multiple realities or truths based on one’s construction of reality (Sale et al., 2002). “Reality is socially constructed and is constantly changing” (Sale et al., 2002, 45). At an epistemological level, there is no access to reality independent of our minds, and no external referent with which to compare claims of truth (Sale et al., 2002). “The investigator and the object of study are interactively linked so that findings are mutually created within the context of the situation which shapes the inquiry” (Guba & Lincoln, 1994 in Sale et al., 2002). This means that reality does not exist prior to the activity of investigation, and further, reality ceases to exist when we no longer focus on it (Smith, 1983 in Sale et al., 2002). “The emphasis on qualitative research is on process and meaning” (Sale et al., 2002, 45), as opposed to causality. The methods commonly used in qualitative research include personal and in-depth interviews, focus group interviews or discussions, ethnographic studies, participant observations, and oral histories, among others (Sale et al., 2002). Unlike in quantitative methodology, the number of research subjects (“samples”) is usually small. The emphasis 107 is not on numbers, but on in-depth information and good articulation of descriptive information on the topic under investigation. A small group of respondents, or the sample, is not meant to be necessarily a mathematically verifiable representation of the larger population, as it is in quantitative statistical methods. So as to understand the qualitative research paradigm, I will briefly describe the foundations of this research paradigm, which are built on constructivism, critical theory, and interpretivism. A discussion on constructivism, critical theory, post-modernism, interpretivism and symbolic interactionism can be found in Appendix J: Philosophical Perspectives on Research. 3.2.1.1 Grounded Theory Grounded theory method (GTM) is a qualitative methodology whose objective is to build theory grounded in data (Corbin & Strauss 2008). Grounded theory’s epistemology is rooted in a two-step evolution, involving the traditions of Chicago Interactionism (Blumer, 1969) and the Philosophy of Pragmatism, often related to the work of John Dewey and Georg Mead (Dewey, 1938; Strauss, 1995). The philosophy of pragmatism (Dewey, 1938) states that knowledge is created through action and interaction of self-reflective beings (Corbin & Strauss, 2008). The activity of thinking has temporal aspects as the actor reassess the effectiveness of his/her actions, changes these actions with the passing of time, based on an envisioned end of the action. Past memories and recollections also affect an actor’s action, thus, pragmatism emphasizes the importance of process (Corbin & Strauss, 2008). Pragmatism also supports the notion that knowledge is socially constructed, and the duality of person and 108 group (collectivity) is disqualified because a single person’s understanding of reality is inherited by that person as a result of being socialized within the framework of the collective perspectives of a broader group of people (Corbin & Strauss 2008). “Truth” to pragmatists is defined based on a key assumption in that truth is equivalent to “this is what we know for the time being, but as time passes, it may be judged partly or even wholly wrong” (Corbin & Strauss, 2008). This position is consistent with postpositivism. The test of validity of an idea (“Truth”) is determined by the consequences of the ideas, in terms of operations to be performed, that establish connectivity within concrete experience (Dewey, 1929 in Corbin & Strauss, 2008). Pragmatists believe that some social knowledge is cumulative and provides the basis for the evolution of thought and society (Corbin & Strauss 2008). The methodological implications of the above philosophies for grounded theory are that there are no simple explanations for things, rather, “events are the result of multiple factors coming together and interacting in complex and often unanticipated ways” (Corbin & Strauss, 2008, 8). Any methodology that tries to understand experience and explain situations will have to be complex. It is important to obtain multiple perspectives on events and build variation into analytic schemes (Corbin & Strauss 2008). Experience can only be understood within larger events in social, political, cultural, racial, gender- related, informational, and technological framework, and so these should be essential aspects of analysis in grounded theory methodology (Corbin & Strauss 2008). Process is very important to the grounded theory method of inquiry and analysis because experience, and therefore any action/interaction that follows, is likely to be formed and 109 transformed as a response to consequence and contingency (Corbin & Strauss 2008). The GTM does not necessarily attempt to reduce understanding of action/interaction/emotion to one explanation or theoretical scheme; however, the approach of using concepts of various levels of abstraction forms the basis of analysis (Corbin & Strauss, 2008). “Concepts provide ways of talking about and arriving at shared understandings among professionals” (Corbin & Strauss, 2008, 8). “Grounded theory is an attempt to derive theories from the analysis of the patterns, themes, and common categories discovered in observational data” (Babbie, 2007, 296). “It is an inductive approach to the study of social life that attempts to generate a theory from the constant comparing of unfolding observations. This is very different from hypothesis testing, in which theory is used to generate hypotheses to be tested through observations” (Babbie, 2007, 296). GTM employs the constant comparative method (Babbie, 2007) and this involves four stages (Babbie 2007; Corbin & Strauss 2008). One, comparing incidents applicable to each category – here, the researcher compares findings from one case with other cases to determine if findings are similar across multiple cases. This is the process of creating concepts, in other words, specifying the nature and dimensions of the many concepts arising from the data. Two, integrating categories and their properties – here, the researcher notes relationships among concepts. Three, delimiting the theory – as the patterns of relationships among concepts become clearer, the researcher can ignore some of the concepts initially recorded, but now evidently irrelevant to the inquiry. Thus, there is reduction in the number of categories, and the theory becomes simpler. Four, writing 110 theory – this is the final step in which the researcher puts his/her findings into words to be shared with an others. Because putting one’s own ideas into words actually improves one’s own grasp of a topic, this final part of the GTM is considered part of the research process. Next, I discuss the quantitative research paradigm. 3.2.2 Quantitative Research Paradigm “The quantitative research paradigm is based on positivism. Science is characterized by empirical research, and all phenomena can be reduced to empirical indicators which represent the truth” (Sale, Lohfeld, and Brazil 2002, 44). The ontological position of the quantitative research paradigm is that there is only one truth, an objective reality that exists independent of human perception (Sale et al., 2002). Epistemologically, the investigator and investigated are independent entities, allowing the former to study a phenomenon without influencing it, or being influenced by it (Sale et al., 2002). “Inquiry takes place as though a one-way-mirror” (Guba and Lincoln 1994, in Sale, Lohfeld, and Brazil 2002, 44). “The goal of quantitative research is to measure and analyze causal relationships between variables within a value free framework” (Denzin and Lincoln 1994, in Sale, Lohfeld, and Brazil 2002, 44). Techniques used to make sure that there is no relationship between the inquirer and research subjects includes randomization, blinding, well-structured protocols and written or orally administered questionnaires with predetermined responses (Sale et al., 2002). The sample sizes here are much larger than those employed in qualitative research. This is to ensure that the data are statistically representative of the larger population, and to 111 ensure that statistical methods can be safely applied to carry out data analysis on the sample population to make conclusions about the larger population. (Sale et al., 2002) Quantitative methods use positivist and post-positivist philosophy. A discussion on positivism and post-positivism can be found in Appendix J: Philosophical Perspectives on Research. Next, I discuss the mixed methods research paradigm. 3.2.3 Mixed Methods Research Paradigm Mixed methods research is defined as a research paradigm in which both qualitative and quantitative methodologies have been used in the same research project (R Burke Johnson & Onwuegbuzie, 2004; Rocco, Bliss, Gallagher, & Pérez-Prado, 2003; Sale et al., 2002; Tashakkori & Teddlie, 2003). Any researcher that has collected data that includes close-ended items with definitive responses, numerical or textual, and then also open-ended items, like “opinion questions” all in the same survey has carried out mixed method research (Tashakkori and Teddlie 2003 in Rocco et al. 2003). Mixed methods according to Giddings (2006) have been applied since the 1950s, especially in evaluation research, and to explore issues and problems. Thus, it was mostly used at the first two stages of social science research, namely, the exploratory and descriptive stages. It was in the early 1990s that mixed methods was elevated to the status of an emerging new paradigm, and that it was the solution to the qualitative-quantitative paradigm wars (Giddings, 2006). The ground was fertile for mixed methods to be recognized as a paradigm because it was reconciliatory. On the one hand, logical positivism had come under serious criticism from philosophers of science since the 112 1950s, such as Thomas Kuhn (1962; 1977), while on the other, scholars such as Karl Popper (1959) heavily criticized qualitative epistemologies such as Marxism and historicism for being superstitious (Giddings, 2006). This ushered in a middle ground position on positivism – postpositivism – in the 1980s (see Appendix J: Philosophical Perspectives on Research). Mixed methods research methodology was employed in this dissertation research. Next, I explain why this methodology was the most appropriate for this dissertation research. 3.3 Choosing Mixed Methods Research Mixed methods approaches are relatively new – only about twenty years old (Creswell, 2012). One of the major authorities in mixed methods research, John W. Creswell, answers the question, when should mixed methods be used? “Mixed methods is (sic) useful when either quantitative or qualitative data alone does not give a full understanding of the problem.” In this research, the research questions would best be answered by leveraging the strengths of both qualitative and quantitative methods because of the complex nature of the problem. To be able to answer the first research question, how has GIS evolved as a technology in Uganda, a qualitative approach to research would be ideal as it provides a detailed account of the historical events that shaped GIS as a technology in the country. In order to answer the second question, what is the current state of GIS in Uganda, a combination of a qualitative and quantitative approach would be most appropriate; qualitative data provides an opportunity for an in-depth analysis of the current state of the technology with respect to type of software, hardware, human expertise available, the 113 impact of GIS on society, and vice versa, while quantitative analysis of the qualitative data (stored in a database) provides a sense of the level of penetration of the technology in the various sectors of the country’s economy. In answering the third question, what is the potential for mobile LBS technology as a tool for society’s access to geographical information in Uganda, a quantitative approach to data collection and analysis seems a logical choice so as to collect a large amount of data related to mobile phone usage, and location-based services software innovation in the country. It was decided, thus, that a mixed methods approach to research would be the most appropriate methodology to employ in this research. I discuss the actual methods used, next. 3.4 Methods Used This dissertation research was based on a mixed methods research paradigm. The methods employed include semi-structured in-depth personal interviews, paper survey questionnaires, a focus group and case study, and a workshop. I discuss each of these methods next. 3.4.1 Semi-structured In-depth Personal Interviews Semi-structured in-depth personal interviews with respondents, lasting fifteen minutes to two hours, were carried out and recorded with a digital voice recorder to answer the first two research questions: (1) how has GIS evolved as a technology in Uganda? And (2) what is the current state of GIS in Uganda? These interviews were carried out with individuals representing institutions at which they were employed, in most cases. 114 Institutions were chosen as the unit of analysis. I adopt the following definition of an institution: “an established organization or foundation, especially one dedicated to education, public service, or culture” (Farlex, 2012). Institutions can be either for-profit businesses, or non-profit organizations; large multinational corporations or small family owned shops; and government or privately owned entities. I adopt a very broad definition of the term “institution” so as to adequately capture the diversity of institutions and their application of GIS technology. Interview responses given by individuals regarding GIS at their institutions were interpreted as being representative of the state of GIS at their respective institutions. In a few cases, the respondents made general comments about the state of GIS in other institutions in the country, as well, based on their personal experiences with those other institutions, for example, in terms of their capacities as data providers, educators, and so on. The selection of institutions in this dissertation research relied on a two-step strategy. First, four broad categories of institutions were identified based on conventional government structure in many countries around the world: (1) public sector institutions, (2) private sector institutions, (3) nongovernmental organizations (NGOs) and international organizations (IOs), and (4) academic sector institutions. A total of 91 institutions were contacted for interviews. The relative number of institutions by sector is shown in Figure 3-1. Out these, respondents from 53 institutions accepted to carry out personal-interviews. The number of the institutions at which respondents accepted to in- person interviews is shown in Figure 3-2. 115 The criteria I used for selection of the four types of institutions included: (1) my personal knowledge about probable presence and application of GIS in various institutions in Uganda, (2) recommendations from my personal network of friends, family, and GIS professionals about institutions that probably use GIS, (3) recommendations from my early interview respondents about institutions that most probably use GIS, (4) an Internet search for institutions that focus on problems and issues that are characteristic themes for GIS analysis, for example, natural resources, utilities, transportation, crime and so on, and (5) a “foot patrol” search for institutions along certain streets in the City of Kampala well-known to be lined by office premises of public sector institutions, and non- governmental organizations, for example, parliament avenue and Bukoto street, respectively. Public sector institutions include government institutions at two basic levels, (1) central government, and (2) local government. Central government institutions include government ministries, under which a hierarchy of smaller institutions falls. According to the constitution of Uganda (GoU, 1995; Kisembo, 2006), the government of Uganda is structured under a hierarchy of institutions, starting with a Ministry at the top, followed by Directorates, in turn followed by Departments. For example, the Ministry of Lands, Housing and urban Development, the Directorate of Lands, and the Department of Surveys and Mapping is one such hierarchy. Some Ministries have within their hierarchical structure institutions that do not fall under any directorate, for example, under the Ministry of Water and the Environment is the agency National Environment Management Agency (NEMA) that enjoy a semi-autonomous relationship with the 116 governing Ministry. An example of this general hierarchy is provided in Figure 3-3 below. A Ministry in Uganda can be equated to a Department in the United States, for instance, the Ministry of Defense in Uganda equates to the Department of Defense in the US. Figure 3-1: The relative number of institutions contacted for interviews, categorized by sector Various public sector institutions were selected in this dissertation research. Many of the institutions chosen for interviews were at the departmental level, for example, the Department of Surveys and Mapping, under the Ministry of Lands, Housing and Urban Development; the Wetlands Department, under the Ministry of Water and the Environment. Semi-autonomous parastatal agencies were chosen as well, for example, 117 the National Environmental Management Authority, the National Wildlife Authority, and the National Forestry Authority. Figure 3-2: Institutions that participated in personal interviews, categorized by sector Private sector institutions chosen for this research were only but a few, and this was due to the general lack of GIS in the private sector. Examples of the institutions chosen include: WE Consult, a Dutch-based consultancy that focuses on water management issues, Tullow Oil, Irish Oil and Natural Gas Company, GeoInformation Communication (GIC), ESRI’s GIS business partner in Uganda, Umeme, a South African-based electricity distribution utility company, and Mobile Telecommunication Network (MTN), a mobile phone service provider. 118 Examples of Local NGOs chosen for this research included organizations such as Nature Uganda, Advocates for Coalition for Development and Environment (ACODE), Northern Uganda Malaria, AIDS and Tuberculosis Program (NUMAT), and Environmental Alert. Examples of IOs chosen for interviews include: Center for Disease Control (CDC), Infectious Disease Institute (IDI), UN Office for the Coordination of Humanitarian Assistance (UN-OCHA), United States Agency for International Development (USAID), and the World Wildlife Fund (WWF). Figure 3-3: Hierarchical institutional structure of the Ministry of Water and the Environment, Uganda, showing directorates, departments and parastatal agencies governed by the Ministry For the purposes of this research, academic sector institutions are defined as Departments within University faculties, or schools, or colleges. In the Commonwealth tradition of education adopted by countries in Africa colonized by Great Britain, a faculty equates to 119 a college at a US university. Schools have similar meanings both in the US and in English speaking African countries. The structure of many universities in Uganda is hierarchical in nature, with a governing council at the top, the “University Senate”. Makerere University is the largest public university in Uganda. The University Senate directly governs various academic departments at the University grouped into faculties, schools, and institutes. The largest administrative unit is a faculty, followed by a school, and then an institute; the larger the unit, the greater the size, and/or number of Departments that constitute it – every department belongs to one of the three administrative units. Uganda has 5 public universities, and 26 private universities (Owoeye & Oyebade, 2011) serving its Higher Education needs. Of these, I interviewed respondents at 4 public universities – Makerere University, Mbarara University, Gulu University and Kyambogo University. Makerere University had the largest number of respondents due the fact that it is the largest university in the country. I carried out personal interviews with professors, technical staff and teaching assistants at 8 academic departments, in total, at the 4 public universities mentioned above; 5 of these departments were at Makerere University. Private universities were not selected for interviews because of a resource constraint relating to the relatively larger number of private universities as compared to the public universities; 26 private universities versus 5 public universities. It is suggested that future research consider private universities to complement this research about GIS diffusion in Uganda’s academic sector. 120 Personal interviews were also carried out with technical staff of institutions involved in the telecommunication industry so as to collect data to answer my third research question, what are the major technological, socio-economic and political challenges to the development and use of LBS technology in Uganda? Five in-depth personal interviews were carried out with employees of the major telecom service providers and telecom infrastructure providers in Uganda: MTN, Orange Telecom, Zain Telecom, Nokia- Siemens and Ericsson, respectively. A complete list of all institutions chosen for this dissertation research, and a list of the interview questions can be found in the Appendices A – D, while the social consent form can be viewed in Appendix F. Next I discuss the survey questionnaire method. 3.4.2 Survey Questionnaires In order to answer the third research question, what are the major technological, socio- economic and political challenges to the development and use of LBS technology, I employed a paper survey questionnaire to collect data. My respondents were selected based on two criteria, (1) geographical location, and (2) socio-economic status, the two criteria being inter-related. The first criterion was based on the geographical location of the localities where the respondent lived and worked, that is, the urban areas, versus the suburbs, versus rural areas. The second criterion, socio-economic status, tried to ensure that I captured a diverse set of respondents in terms of their professions, education level, and income brackets. These two criteria helped me capture a sample of respondents that would represent the societal structure in Uganda. And in many ways, the two are linked because socio- 121 economic status, and the locations where respondents live and work are correlated – the more educated a respondent is, the more likely he/she lives in a city, town, urban or suburban area, these directly impact the respondent’s likelihood of owning a cell phone and cell phone usage habits. Since my research question directly investigates the socio- economic and political challenges to the development of location-based cell phone technology, the two criteria used in selecting respondents were deemed appropriate. In total, the questionnaire was personally administered to 182 respondents who answered questions about their mobile phones, for example, the type of mobile phones they carried, and their network providers. Out of these, 101 respondents answered ICT- and LBS- related questions. They were asked whether they owned computers, what type of access they had to the Internet, who were their Internet Service Providers, whether they used their phones for making spatial decisions, and whether they would consider using their phones for certain types location-based services, for example, for reporting to authorities the locations of extremely large potholes on roads in their neighborhoods. The 182-sample was to establish current trends in the type of mobile phones owned by Ugandans. The 101-sample, which was a subset of the larger 182-sample, was used to assess society’s access to ICT, and the potential for LBS in. The reason for the difference in sample sizes was that the larger sample allowed for quick and broad sampling with fewer questions so as to increase response rates for the “mobile phone trends” part of the survey. The second sample was smaller because the survey was much longer, and thus, would probably have a lower response rate. This was the “ICT and LBS” part of the overall 122 survey. One combined questionnaire was used in both cases: 182 respondents answered the “mobile phone trends” part of the survey, and 101 of the 182 respondents continued on to the “ICT and LBS” part of the survey to provide even more information. Two research assistants were used to administer the surveys in two rural areas, Mayuge and Katakwi districts. I personally administered the surveys in the capital city, Kampala, and in the surrounding districts of Wakiso and Jinja. These latter districts can generally be considered urban in nature. Much as the surveys were carried out in these five districts, some of the respondents actually commuted to work in these districts from other surrounding districts, such as Mukono, Iganga, Soroti and the municipality of Entebbe. Examples of spatial decisions respondents might make using cell phones include, using mobile phones to transfer money from urban areas to relatives in rural areas; paying utility bills using mobile phones without physically going to the offices of a utility company; small business operators using text messages to take orders for goods and services, for example, fish mongers, restaurant operators; customers paying for goods and services remotely through mobile money transfer to traders; and rural farmers in villages using text message-based services to do a Google search for information on how to concoct a pesticide using natural organic ingredients. The complete questionnaire used for data collection can be viewed in Appendix E. A chart showing the relative percentage of respondents based on their professions is shown in Figure 3-4. 123 3.4.3 Focus Group and Case Study In this dissertation research, one focus group interview was carried out with employees working for a nongovernmental organization (NGO), the Busoga Rural Open Source and Development Initiative (BROSDI), in a rural town in Eastern Uganda. This was part of a case study to investigate the use of mobile location-based services (LBS) for improving access to location-specific agricultural information in Uganda. A focus group, also known as a focused group interview, is qualitative method of data collection that involves the interviewing of a carefully selected group of people on a specific topic (focus) facilitated by a moderator (Sim, 1998). Focus groups often times comprise of 4-6 or 8-12 participants (Sim, 1998). The difference between a focus group and a one-on-one personal interview is that with the former, information from members of the group is cross-checked by the other members, and diverse opinions emerge out the discussion. This process allows for the obtaining of a rich set of information and a deeper understanding of the topic under discussion. A case study approach to research tries to understand a general issue by studying the dynamics present in a single setting (for example, an organization), and can combine both qualitative and quantitative data collection methods, such as archives, observations, interviews, and questionnaires (Eisenhardt, 1989). Analysis of data resulting out of case study research may be done based on qualitative (for example, grounded theory) or quantitative (for example, descriptive statistics) techniques. A case study of BROSDI was done by carrying out one in-depth personal interview with two staff members of the NGO, and a focus group discussion with three grass roots 124 facilitators in a small village in Mayuge district, Eastern Uganda. The focus group discussion took place at the same time that a site visit was made to the NGO’s office in small rural village in Mayuge district. During this site visit, personal observations were also made of the NGO’s activities in the village. The topic under investigation was the use of a mobile text-message-based location-based services application for improving livelihoods of rural farmers in Mayuge district. 3.4.4 Workshop Finally, I attended a one-day preconference workshop on mobile applications for development (6th Annual International Conference on Computing and ICT Research) in Uganda at Makerere University’s faculty of Information and Communication Technology (today School of Information and Communication Technology) in Kampala (see Kizza et al. 2010 for conference proceedings). This was with relevance to the third research question in this dissertation. Presentations by Orange Telecom Uganda, Mobile Monday Kampala (MoMoKla – a grass roots mobile applications development group) and SMS-Media (a mobile applications content provider) were recorded using a digital voice recorder and transcribed. Qualitative analysis of the transcriptions followed. I also witnessed live demonstrations of various LBS applications being developed by innovators at Makerere University. 3.5 Data Analysis Having employed a mixed methods approach to research in this dissertation, I applied both qualitative and quantitative analysis methods. To analyze the qualitative in-depth 125 personal interviews, I first transcribed my recorded interview data verbatim to pick up any vocal cues and intonations that might help in my analysis. I used analysis techniques embedded in the grounded theory method (GTM) (Corbin & Strauss 2008; A. Strauss 1995) to dig out concepts or codes in the transcribed data. Examples of concepts include phrases such as project-driven GIS, expensive GIS licenses, donor funding, the World Bank, and so on. Concepts or codes were later generalized to provide an understanding of broader issues, and finally, irrelevant concepts were eliminated leaving only the important concepts. In the example of concepts given above, these were generalized into a broader concept about the heavy dependence of GIS in developing countries on international donor funding and aid. The broader concepts were combined together to form a theoretical understanding about the evolution and current state of GIS in developing countries. Examples of broad concepts in the data include, dependency on donor funding, a lack of skilled GIS personnel, data sharing problems, and a lack of government support for and interest in GIS technology. The theoretical understandings developed from the grounded theory method was then compared with elements of the well-known diffusion of innovations theory (Rogers, 1993, 2003), leading to confirmation of, or departure from, the well- established theory. Diffusion of innovations theory provides a framework for understanding the adoption or rejection of innovations as a function of social behavior in a particular society. 126 I used commercial database software, Microsoft SQL Server, to store and analyze my qualitative data. Concepts or codes formed the columns of my tables, similar to dependent variables in the quantitative tradition, and row represented institutions, for example, public sector institutions, such as the Department of Surveys and Mapping; and academic sector institutions, such as Makerere University’s Department of Forestry. Graphical plots were created in a quantitative analysis of the dependent variables in the database, for example, a plot of the total number of institutions that adopted GIS technology over time. To achieve this, the data was exported from the database into a spreadsheet format. I analyzed the data in the survey questionnaires based on descriptive statistical analysis methods. These data were initially transferred from paper to a digital commercial database, Microsoft SQL Server. A column in the database table represented a particular attribute of data (dependent variable), for example, number of mobile phones carried by respondent, or GPS capability of the phone; while the rows represented each of the 182 respondents. I carried out analysis on this data using descriptive statistical methods. To plot the analytical results in graphical layouts, I exported my final database into a spreadsheet format. 127 Figure 3-4: Mobile phone questionnaire survey respondents by profession Next, I discuss the strengths, weaknesses, and potential biases of the methods used in this dissertation research. 3.6 Strengths, Weaknesses of and Potential Biases of Methods Used This study employs a mixed methods approach to research. Mixed methods research does not aim to replace quantitative and qualitative methods. Instead, it is meant to draw from the strengths and minimize the weaknesses of both in single research studies and across studies (R Burke Johnson & Onwuegbuzie, 2004). Despite its strengths, critiques have voiced concern over some weaknesses in the methodology, which is sometimes questionably equated to epistemology (for example, see R Burke Johnson & Onwuegbuzie, 2004, 15). 128 Quantitative purists from the positivist tradition insist on an approach to research that is based on objectivity, observation, evidence, confirmation and falsification, the so called concept behind “science.” However, the positivist approach is one that is, arguably, well suited for inanimate objects rather than human beings (J. K. Smith, 1983). This is mainly because humans are complex beings who exercise inherent subjectivity, have emotions and values – all factors not strongly addressed by the hard sciences. The need for qualitative approaches seems necessary to address this. Mixed methods approaches attempt to harness the strengths in both qualitative and quantitative approaches. In this dissertation research, there are some weaknesses and biases that need mention, nonetheless. Much as the qualitative in-depth semi-structured interviews provided rich information, there is no guarantee that the respondent revealed all the necessary information required, for example, information on certain politically oriented questions regarding governance and corruption in institutions. This could be because of the nature of the interview venue, and the fact that a digital voice recorder was making an actual record of the interview. It is possible that the interviewee did not fully trust the words in the consent form about data privacy. As regards the quantitative data resulting from the interviews, for example, consideration of the actual number of institutions interviewed, a larger sample of institutions could have produced more concrete results from the quantitative analysis. For instance, the number of academic institutions interviewed could have been larger, not to mention a more inclusive in terms of private universities and academic institutions. In addition, the survey sample for mobile phone and LBS potential analysis could have been larger so as to 129 convey even stronger conclusive analysis results. Surveys also, by definition, suffer from the fact that questions are determined a priori, leaving no room for open discussion on the subject under investigation. 3.7 Conclusion This dissertation research employed a mixed methods approach to research which is a combination of both qualitative and quantitative research methods, benefitting from the strengths of both methods. Semi-structured in-depth personal interviews were carried out with respondents at institutions that use GIS, and paper surveys were administered to mobile phone users in Uganda. In addition, a case study of a local NGO was carried, and this involved an in-depth personal interview, a focus group discussion and personal observation. Analysis of the data was carried out using both qualitative and quantitative methods, specifically, grounded theory and descriptive statistics, respectively. Despite the weaknesses inherent in qualitative and quantitative methods, such as subjectivity, and sample size, mixed methods research benefits from the strengths of one method complementing the weaknesses of the other. 130 Chapter 4 : Evolution of GIS in Developing Countries – A Case Study of Uganda 4.1 Introduction The entry of GIS into developing countries is solely attributed to the efforts of one organization: the United Nations Environmental Programme (UNEP), and more generally, the parent organization, the United Nations (UN). As a result of UNEP’s efforts, GIS was introduced into a few developing countries in Africa, Asia and South America between 1985 and 1989. Related to the activities of UNEP were the efforts of the UN Economic Commission on Africa (UNECA) to establish cartographic centers in developing countries. As a result, five regional remote sensing and cartographic centers were established in Africa between 1972 and 1987 (Ribot, 1989, in Tchindjang et al. 2005). Early capacity building of local citizens in GIS was largely the work o the United Nations Institute for Training and Research (UNITAR), and the International Institute for Aerospace Survey and Earth Science (ITC) in the Netherlands (Mooneyhan, 1998; Toppen, 1991). Much as the early intellectual history of GIS has been somewhat documented in the literature, not much is known about the process of evolution of GIS in developing countries. What were the social, political, economic and technological influences that facilitated or hampered early GIS diffusion in the developing world? An understanding of this complete intellectual history of GIS technology and its diffusion in developing 131 countries is important for a holistic understanding of the current state of GIS in these countries. I argue that GIS diffusion in developing countries was heavily influenced by donor agencies in the West, and was closely related to the evolution of mapping science. However, it was the field of environmental conservation that led to the initial introduction of GIS in the developing world by various donor agencies, including UNEP. This evolutionary trend is responsible for the current state of GIS in the third world today with respect to the public, private, academic, nongovernmental and International Organization sectors. The structure of this chapter is as follows: I present my qualitative findings about the historical development of mapping science in Uganda in section 4.2. In section 4.3, I present my findings on the evolution of GIS in Uganda. Finally, in section 4.4, I perform a diffusion of innovations, and a GIS and society (intellectual history) analysis. 4.2 The Historical Development of Mapping Science in Uganda The evolution of GIS in Uganda is closely linked to the development of mapping science in the country. Mapping science historically includes the fields of photogrammetry, and cartography, although the disciplines of remote sensing and GIS have been added to this list (NRC, 2002). In many developing countries, the discipline of land surveying was a precursor to photogrammetry and remote sensing and still forms the backbone of mapping activities at national mapping agencies. In this section, the history of land surveying and cartography is traced to genesis of mapping science in Uganda. This is 132 followed by an account of the relatively recent history of photogrammetry and remote sensing in the country. The origin of land surveying as a discipline is closely linked to colonization of present- day Uganda by Britain. On 1 st July 1890, Britain and Germany signed a treaty, the Anglo- German Agreement of 1890, also known as the Heligoland-Zanzibar Treaty, which set out the spheres of influence of the two European superpowers in present-day Uganda, Kenya and Tanzania (Batungi, 2008; Gray & Peters, 1960; Sanderson, 1963). After this treaty, Uganda and Kenya were declared as British spheres of influence, while Tanzania was declared German. In 1894, Britain declared Uganda its protectorate when the Imperial British East Africa Company (IBEAC) transferred its interests in the region to the British government. The boundaries around the kingdom of Buganda, the center of British influence, were extended outward to include territory approximately equal to the area within the political boundaries of present-day Uganda (Lwanga-Lunyiigo, 1987). The actual surveying and mapping of Uganda was closely preceded by the mapping of the boundary between the British and German spheres of influence in the areas around Mount Kilimanjaro, following the Anglo-German agreement of July, 1890, also known as the Heligoland-Zanzibar treaty. The field work from the Indian Ocean to Lake Jipe, in the north-western part of modern day Kenya, was compiled by plane table surveys and included in three map sheets. This plane table survey was carried out by two assistants loaned by the Survey of India. In the last decade of the 19 th Century, the surveying and mapping of British East Africa was being performed in various forms, (1) boundary mapping by Anglo-German 133 Boundary Commissions, (2) topographical and exploratory mapping, and (3) the mapping for the Uganda railway that was constructed mainly by Indian laborers from British India, from Mombasa in present day Kenya to Kisumu on eastern the shores of Lake Victoria. Much as the construction of the “Lunatic Express” lasted from1896-1901, the initial survey of the proposed path was carried out between 1890 and 1894 by a British surveyor who was working in 1890 for the Imperial British East Africa Company in British India. He was hired by the British government to carry out the East African railway survey.(McGrath, 1976) The delimitation of the boundary between the British and German spheres of influence in East Africa west of Lake Victoria was performed between 1902 and 1904, while the delimitation of the boundary east of the lake was began in 1904, both by a combined Anglo-German Commission for the delimitation of the boundary. The office closely related with the surveying and mapping of boundaries of the protectorate of Uganda was the British War office, under instruction of the British Foreign Office, while technical support was provided by Royal Geographical Society in London and the British military, for example, in terms of hiring and acquiring surveying equipment. In 1904, a new entity was created called the Directorate of Military Operations, under which the Topographic Section of the General Staff was directly responsible for the delimitation of boundaries in British East Africa. By 1910, the boundaries of the spheres of influence of Britain, Germany and Belgium in East Africa were largely completed, a process that lasted about twenty years from planning to finish. An extension of the mapping from the boundary towards inland regions was done in 1907-08 in western Uganda, and partially attempted 134 in 1910-11, by which time a policy had been put in place by the colonial government for systematic mapping of the British territory. (McGrath, 1976) The result of this mapping exercise between 1890 and 1910 was the consolidation of the War office as the authority on boundary surveys and maps in the British African territories. The Foreign Office, whose responsibilities for the territories were eventually passed on to the Colonial Office, both consulted the War Office on surveying and mapping issues in the territories in Africa. The Topographic Section of the General Staff (after April 1907, the Geography Section) was the single most important agency in the eventual development of local surveying and mapping departments in the colonies and protectorates. (McGrath, 1976) In October, 1905, the construction of a triangulation network that was to run from Cape Town in South Africa to Cairo in Egypt was started in the Orange River Colony of South Africa. The method used for this geodetic survey was plane table surveying. Later, systematic topographic mapping in East Africa began with the creation of the first survey department, first in the Uganda Protectorate in 1901 when R. C. Allen was appointed Chief Surveyor, followed by a second one in the East Africa Protectorate (present day Kenya) in 1903. Both Survey Departments were responsible for cadastral surveying and mapping in accordance with the Crown Ordinance that resulted from the Uganda Agreement of 1900. On 10 th March 1900, the Uganda Agreement (also known as the Buganda Agreement) was signed between the British government and the Kingdom of Buganda (Batungi, 2008; Nkurunziza, 2006). This agreement was significant because it led to the formal 135 recognition of two types of land tenure in the Kingdom of Buganda, (1) Mailo (from the English word “mile”), a freehold type of land tenure to be held by the beneficiaries in absolute ownership to perpetuity, and (2) Crown land tenure, also a freehold tenure type. Mailo lands were to be owned by the “Kabaka”, or King of Buganda (958 sq. miles), and one thousand of his chief (8000 sq. miles), including a few private land owners. Crown land was owned by the Colonial government vested in the Queen of England, and represented in the Protectorate by the Governor (Batungi, 2008). It is estimated that the land covered under the 1900 Uganda agreement was about 20% of the surface area of Uganda. Britain’s main interest in the agreement was to find some way of taking control of unoccupied land to establish lucrative plantation farms (Batungi, 2008). As a result of the 1900 Uganda agreement, there was an obvious need for cadastral land surveying services in the Kingdom of Buganda, and the rest of Uganda as well. All administrative departments of the colonial administration were strategically located on a hill directly opposite the location of the native Buganda government, whose powers after the 1900 Uganda Agreement had been considerably reduced to the areas surrounding the hill on which the Kabaka’s (King’s) palace was located – Mengo hill. The British colonial government strategically established its administrative headquarters on Old Kampala hill, overlooking the native Buganda government headquarters on Mengo hill. This was to keep a check on the diminished powers of the native government and to allow the British colonialists to administer colonial rule on the rest of the Kingdom. However, due to tensions between the native and colonial governments on the two hills, the latter decided to move their administrative headquarters 40km south to the shores of 136 lake Victoria, an area called Entebbe, as early as 1908 (Churchill, 1908, in Nkurunziza, 2006). Personal interviews with respondents at the Department of Surveys and Mapping revealed that the first colonial administration departments in the early 1900s, including the Department of Surveys and Mapping, were located on Old Kampala hill, a suburb of the city of Kampala known today as Old Kampala. As early as 1908, administrative departments of the colonial government started moving south to Entebbe, which eventually developed into a small satellite town. In 1937 the Department of Surveys and Mapping moved from Old Kampala hill to Entebbe where a new building had been constructed. This building is the current location of the Department of Surveys and Mapping even today. Surveying operations in the early 1900s consisted of plane table and chain surveying. The surveying data were plotted manually using cartographic techniques employing pen, paper and ink. Because the British surveyors and cartographers needed help in performing their duties, they set up a training facility for transferring surveying and drafting skills to the local population. The facility, called the Survey Training School was established in Entebbe, at a locality known as Katabi. This facility had already been constructed and was operational by the time the Department of Surveys and Mapping moved to its permanent address in Entebbe in 1937. Students were trained in the discipline of surveying, but at a very limited skill level – plane tabling, chain surveying, map compilation and map production (cartography). Students graduated with certificates in surveying and cartography. (Y. Okia, personal communication, July 16, 2010) 137 There is a close relationship between the development of mapping science and issues of land management and cadastre in the Buganda Kingdom. The main reason for the need for surveying was for the documentation of land titles. The land owners under the two types of tenure that resulted after the 1900 Uganda agreement, Mailo and Crown land, would require official land titles. Land titles for Mailo land were to be based on boundaries of blocks of land owned by powerful people in the Kingdom of Buganda, which included the royal family and chiefs, while crown land, which consisted of unoccupied lands in the Kingdom of Buganda, was to be officially titled for the British colonial government. The British first surveyed the old Buganda counties, partitioned first into blocks, and then surveyed at the parcel level. Within these blocks, chiefs were apportioned large pieces of land based on instructions from the Kabaka. “These lands were all surveyed using plane table surveying techniques and land titles produced. These titles were stored at the office of the Registrar of titles in Mengo: In the 1960s, the titling was decentralized to the district level, for example, land titles of land in Masaka town would be kept by the registrar of titles in Masaka, not Kampala. Each district had its own Registry for Mailo Land. However, these Registries were still under the Ministry of Lands. (Y. Okia, personal communication, July 16, 2010) In 1937, a new building was constructed in Entebbe which housed the colonial headquarters for the Department of Surveys and Mapping. This building is still the location of the Department of Surveys and mapping in the current administration of Uganda today. By 1937, the Survey Training School was still located in Katabi, near Entebbe, however, in 1969, it moved from Katabi to its current location at Gowers road 138 in Entebbe. The STS was an institution created the colonial administration under a policy that every administrative department had to have a related training facility to supply locally trained skilled workers to that department. Surveying activities between the 1940s and 1960s were organized by the colonial government in a decentralized system of administration. There was a hierarchy of surveying institutions with a Ministry at the top, followed by a department. Under each department were branch offices. Each branch office was responsible for the mapping of a certain region of the country, for example, the eastern region had a branch office in Mbale, which was responsible for mapping districts like Bukede and Tororo, all the way up Karamoja. Each surveyor at the branch office was given a base topographic map of the area and was supposed to update the map with time. Under each branch office, smaller units existed, called camps. Each district had one or more camps with surveyors. After these surveyors had updated their topographic maps with new detail, they were responsible for sending their maps from their respective camps to the regional branch office, which would compile all the maps from the various camps and send the compilation of maps to the central Department of Surveying and Mapping in Entebbe, which would then update the national map. Updates for rural areas were required every 5-10 years. Between the 1950s and 1960s, photogrammetry was used actively by the colonial government for topographic mapping in Uganda. As narrated by one of the interviewees, “This was revolutionary in that the mapping exercise was much faster than traditional ground surveying. Traditional ground surveying techniques were abandoned in this 139 period. At that time, the Department had an analogue aerial photogrammetric camera, a photogrammetric lab with an analogue stereoplotter, and a photography section for developing negatives, positives and plates. The department cooperated with the Police Air Wing and flew its own photography. It also had a powerful process camera that was used in the process of map reproduction – for photographing existing maps, and creating new plates for the printing press. They also had a printing press, which later broke down and due to the lack of spares, was left to decay. Up to today, the Department is hoping to repair it, but funding is a problem” (Y. Okia, personal communication, July 16, 2010). Next, I present my findings about the evolution of GIS in Uganda in the public, academic, nongovernmental, International Organization, and private sectors. 4.3 The Evolution of GIS in Uganda The evolution of GIS in Uganda started with the introduction of GIS in the public sector by UNEP-GRID, followed by the academic, NGO and private sectors. Simultaneously as this was happening, training and capacity building efforts were under way to transfer skills to the staff at the local institutions. In this section, I discuss the introduction of GIS in the various sectors of Uganda’s economy in section 4.3.1, and then discuss the history behind training and capacity building in GIS in the country in section 4.3.2. 4.3.1 Introduction of GIS at Public, Academic, Non-governmental and Private Sector Institutions The introduction of GIS in Uganda coincides with the creation, in 1989, of an agency called the National Environment Information Center (NEIC) under the Ministry of Environment Protection (Gowa, 2009). 140 According to Gowa (2009), shortly after a new regime and ruling party had come into power in January 1986, the newly government in Uganda established a ministry to cater to issues regarding environmental management in the country. In 1987, the Ministry of Environment Protection held talks with the United Nations Environmental Programme (UNEP) about launching a donor supported project to establish a center for the management of environmental information in Uganda (Gowa, 2009). In 1989 a user needs assessment was completed to specify information and capacity building needs for such a center to be established (Gowa, 2009). With financial and technical support from UNEP and the World Bank Technical Division, Africa Region (AFTEN), a center was established within the Ministry of Environment Protection: the National Environment Information Center (NEIC). The mandate of this center was to provide environmental information to support development through collaboration with sector institutions in the country. NEIC was to realize its objectives through the establishment of an Environmental Information System (EIS). (Gowa, 2009) An EIS was implemented in form of a GIS. The United Nations Environmental Programme – Global Resource Information Database (UNEP-GRID) simultaneously introduced Idrisi and PC Arc/INFO GIS software at NEIC in 1989. According to my personal interviews with employees of the now defunct NEIC, UNEP-GRID started the process of capacity building in GIS by training local personnel at the center in vector GIS using PC Arc/INFO, and raster GIS using Idrisi. One of the first people to receive GIS training at NEIC were Frank Turyatunga in Idrisi, and Filipo Mpabulungi in PC Arc/INFO, both employees of NEIC in the early days. 141 “And when I started working, I started on Arc/INFO, basically. I think I was the only person who was doing PC Arc/Info at that time. That was in 1991. And it became the general program (PC Arc/INFO), because they were being given out by UNEP – it used to distribute to different countries which wanted to take it on… Personally, I started on Arc/Info. My other colleagues did not know Arc/Info. They had done something in Idrisi, but I think they were more keen on management, than in the real practical work” (F. Mpabulungi, personal communication, July 19, 2010). NEIC started collecting GIS data from secondary sources by trying to collect secondary data from various institutions, such as the Department of Statistics, and the Department of Surveys and Mapping, that owned spatial and non-spatial environmental data (Gowa, 2009). However, this effort was soon abandoned for two reasons: one, the lack of storage capacity due to a limitation in computing resources, and two, due to an intense inter- institutional political problem regarding the contestation of ownership rights of data in Uganda (Gowa, 2009). This was the sign of the failure of future National Spatial Data Infrastructure (NSDI) efforts in Uganda, which have not succeeded up to today, more than twenty years later. Institutions such as the Department of Statistics and the Department of Surveys and Mapping contested the legitimacy of the mandate of NEIC as a repository of spatial and non-spatial statistical data for Uganda, each institution claiming the mandate for the specific data that they produced. For example, the Department of Statistics claimed to have the mandate to produce and own all statistical data, for example, census data, while the Department of Surveys and Mapping claimed its institutional right and mandate to produce and own all topographic spatial data. This 142 political tension was partially diffused through the creation of a network of cooperation, the Environmental Information Network (EIN), to which institutions that produced spatial and non-spatial environmental data were admitted (Gowa, 2009). However, this was an early sign that getting institutions to cooperate on matters of data sharing was going to be a challenge in the future. NEIC Later became NEMA in 1995. Between 1991 and 1995, NEIC played a significant role in providing GIS generated environmental information, which was used to provide content for the first two significant publications on the state of the environment of Uganda, that is, the 1994 State of the environment report (SOE), and the National Environment Management Policy. Both these reports directly informed the debate that led to the legislation to protect the environment, the National Environment Act Cap 153 (Gowa, 2009). In 1995, NEIC morphed into the information unit of the National Environmental Management Authority (NEMA), a semi-autonomous agency “charged with the responsibility of coordinating, monitoring, regulating and supervising environmental management in the country” (NEMA, 2012). In its six years of existence, NEIC had faced several challenges as articulated by Gowa (2008). NEIC remained a small and technically constrained institution because of several reasons: one, inadequate institutional mechanisms to support the dissemination of information to potential users; two, limitations with regard to data availability, accessibility, quality, and standardization; and three, inadequate legislation with regard to data sharing among institutions and data propriety. A review by NEMA in 1995 based on passed lessons learnt led to recommendations being made so as to integrate EIS into the development 143 process in Uganda. The main recommendations made included, one, the development of an EIN at both national and district levels, two, the formulation of a strategy for the integration of environment information into the development planning process, and three, the development of a training program in support of the EIN at both the national and district levels. Table 4-1: Introduction of GIS in Uganda’s public, private, NGO and academic sectors, 1989-2002 YEAR INSTITUTION NAME SECTOR SOFTWARE 1989 National Environment Information Center (NEIC) Public Idrisi; ARC/INFO 1989 National Forestry Authority (NFA) Public Idrisi 1990 Wetlands Management Department Public Idrisi 1991 Makerere University – Department of Surveying Academic Idrisi 1991 Makerere University - MUIENR Academic Idrisi 1995 Department of Surveys and Mapping, CAMPUS Project Public ArcView 1995 National Environmental Management Authority (NEMA) Public Idrisi; ARC/INFO 1996 Famine Early Warning System Network (FEWSNET) NGO ArcView 1996 Uganda Wildlife Authority (UWA) Public ArcView 1996 Department of Geological Survey and Mines Public MapInfo 1997 Makerere University Department of Geography Academic ArcView 1998 National Water and Sewerage Corporation Public ArcView 1999 National Agricultural Research Laboratories (NARL) Public ArcView 2000 Makerere University – Department of Forestry Academic ArcView 2000 MTN Uganda - Mobile Telecom Network Private MapInfo 2000 Northern Uganda Data Center - NUDC Public ArcView 2001 UBOS - Uganda Bureau of Statistics Public ArcView 2002 GeoInformation Communication (GIC) – ESRI partner Private ArcView 144 Around the same time that GIS was introduced by UNEP-GRID at NEIC, in 1989, another institution, the Department of Forestry under the Ministry of Environment Protection (Drichi, 2002), had GIS introduced into its workflow through a donor funded project supported by the Norwegian Agency for Development Cooperation (NORAD) (J. Mutyaba, personal communication, July 19, 2010) in collaboration with the Norwegian Forestry Society (Drichi, 2002). This project was called the National Biomass Study (NBS), and its main objective was to map out and inventory the volume of biomass resources in Uganda, that is, forests and woodlands (Drichi, 2002; Turyareeba & Drichi, 2001). GIS methods used for mapping out the forests using SPOT satellite imagery from 1989 to 1993, covering the entire country (Drichi, 2002). Specifically, Idrisi software from Clarke Labs was used to perform unsupervised classification, followed by supervised classification. Inventory plots were used for estimating the amount of biomass contained in the areas classified as forest. Various tree parameters were measured in the inventory plots, for example, tree height and canopy size, resulting in a computation of biomass for the inventory plot. These inventory plots were overlayed with the land cover classification maps resulting from the Idrisi analysis of satellite imagery, and based on the relative locations of the inventory plots and the areas classified as forest or woodland, the biomass was extrapolated for the underlying areas. During the course of the NBS, the Forestry Department realized the need for other operational GIS layers so as to carry out meaningful analysis relative to the results of the biomass study. The task of digitizing of 1:50,000 scale topographic maps from the 145 Department of Surveys and Mapping was undertaken by the Department of Forestry so as to produce GIS layers such as roads, water bodies and population centers . However, due to the tension resulting from questions about ownership of data, and mandates relating to specific datasets, eventually, the Department of Forestry’s mandate was reduced to the creation, collection, and storage of only forestry data. (J. Mutyaba, personal communication, July 19, 2010) In 1990, GIS was introduced at yet another institution in Uganda, the Wetlands Inspection Division, under the Ministry of Environmental Protection. This was under a program funded by the Dutch government, the Wetlands Management Program/Project, which was a project meant to protect wetlands in Uganda. It should be noted that it was this program/project that led to the creation of the Wetlands Inspection Division, and not vice versa. The project undertook the mapping of wetlands in Uganda using remote sensing methods employing Idrisi software. Satellite imagery, Landsat and SPOT was used to carry out unsupervised land cover classification. A lot of ground truthing was required so as to determine whether the wetlands classified by Idrisi were permanent or seasonal. This extensive and laborious ground truthing was carried out by staff of the Wetlands Inspection Division with the help of topographic paper maps obtained from the Department of Surveys and Mapping. Field crews identified nearby villages next to wetlands, and then proceeded to personally visit the sites to determine whether the wetlands were permanent or seasonal. In addition, the exercise involved capturing of related attribute data, for example, the current use of the wetlands, for instance, for farming, fishing, brick making, or reed harvesting for making floor mats; pH of the 146 water; water quality; and level of degradation of the wetland, for instance, how much of it had been drained. The attribute data was stored in 89 Microsoft Access Databases, and constituted a total of 4.4 million records. The storing of attribute data in a separate database was mainly because Idrisi, a raster GIS, had no tools for storing attribute data. This is a limitation of raster GIS which is best suited for representing single attribute values at the pixel level, and has no capability of storing geographical entities as objects. This project ended in 1994. When ArcView GIS was introduced in the period following the completion of the Wetlands Management Program/Project, a new challenge was discovered – how to integrate the attribute data in stored in the Access database with ArcView GIS? Specifically, there were two challenges. One, ArcView was not able to automatically integrate attribute data from an external database into its own software environment. Two, ArcView was not able to handle data sizes exceeding 2GB, and for all practicality, 500MB. As a result, the Access database lay dormant till 2008 when ArcGIS software was introduced at the institution, and a new staff member was hired with the necessary skills required to perform the task of GIS database integration. (P. Omute, personal communication, July 16, 2010) In the early 1990s, GIS was introduced for the first time in the academic sector. Makerere University Institute of Environment and Natural Resources (MUIENR) (D. Mfitumukiza, personal communication, June 28, 2010) and the Department of Surveying (M. Musinguzi, personal communication, June 23, 2010), both at Makerere University 147 Kampala, a government owned university, saw the introduction of GIS as a course in their respective curricula. A precursor to the project that introduce GIS at the Department of Surveys and Mapping in Uganda was the Kampala Mapping Project which had started in 1993, funded by a loan from the World Bank to the government of Uganda (Y. Okia, personal communication, July 16, 2010). The main objective of the project was updating the map of the greater Kampala area using photogrammetric methods. The latest topographic maps at the Department of Surveys and Mapping at that time were outdated and had last been compiled in the 1960s. Institutions responsible for managing the city’s infrastructure, for example, Kampala City Council (KCC), had a dire need for up-to-date topographic maps of Kampala city for effective urban planning and management. In fact, the original contract had been awarded to KCC; however, due to the lack of skilled capacity at KCC, it was decided that the project be delegated to the Department of Surveys and Mapping. The German company that won the Kampala Mapping Project contract introduced the first computers at the Department of Surveys and Mapping in 1993. These were the early 90s versions of desktop computers and carried 286 processors. One of the computers had a hard disk memory size 1GB, while the other computers only 256 MB. These computers were pre-installed with Computer Aided Design (CAD) software, AutoCAD Release 12, and photogrammetric software for semi-analytical photogrammetric functionality. Data resulting from the use of the computers was stored externally on square-shaped magnetic tape cartridges and large-format diskettes. There was not much knowledge about computers or computer aided mapping at the Department. Only John Kitaka, a senior 148 cartographer at the department who had attended graduate school in computer aided cartography and mapping at the International Institute for Geo-Information Science and Earth Observation (ITC) in the Netherlands, had had some exposure to computers at that time, and was, thus, made head of the project at the Department. (J. Kitaka, personal communication, July 16, 2010) The Germans reasoned that it would be cheaper for them to train on-site staff in Uganda in semi-analytical photogrammetric methods to produce the desired maps, rather than flying photography in Uganda and sending the photographs to Germany for map feature extraction and map production (J. Kitaka, personal communication, July 16, 2010). Analogue stereoplotters that were already present at the Department from the time of independence in 1962 were linked up to the new computers resulting in a setup known as analytical photogrammetric stereoplotters. The advantage with the analytical photogrammetry is that aerial triangulation of control points is automated, and feature extraction is done using CAD software instead of manual analogue photogrammetric methods. The Kampala Mapping project resulted in the production of 340 topographic maps of the greater Kampala area at a medium scale of 1:25,000 covering an area of 600 square kilometers. The extent of the mapping was as follows: “And this extended from Kajjansi in the south, halfway between Kampala and Entebbe, and then in the north, it went up to Kawanda, in the East it reached Mukono, and the West it reached Busega – a swamp there, around Kyengera on Masaka-Mbarara road” (J. Kitaka, personal communication, July 16, 2010). 149 Following the Kampala Mapping project, GIS was introduced at the Department of Surveys and mapping by a French company under a project that started in December 1995 (M. Murindwa, personal communication, July 28, 2010). This was under a project called Computer Aided Mapping Project Uganda Surveys (CAMPUS), funded by the French Government, and implemented by a French company. The GIS software introduced was ArcView 3.1. The major objective of the CAMPUS project was to develop spatial databases for all the districts in Uganda (J. Kitaka, personal communication, July 16, 2010) so as to support the new decentralization policy of the government of Uganda, a bill that had been passed by Uganda’s law makers in 1993, the local government Act of Uganda (GOU, 2005), to decentralize planning activities in support of development activities (GoU & UNFPA, 2010). A secondary objective of the project was to computerize the Department of Surveys and Mapping, and further to train the staff in automated mapping methods. In the first phase of the project, 1995-1996, the staff members on the project at the Department were trained in the scanning and digitizing of paper maps using digitizing tables and Demeter CAD software. Topographic map sheets at a scale of 1:50,000 were scanned and digitized. A total of 316 map sheets covered the entire country. The second phase of the project was funded by the European Union (EU) and lasted from 1997-1999. It was during this phase of the project that ArcView 3.1 GIS software was introduced to convert the CAD and DXF (Data Exchange Format) files produced from the digitizing exercise in phase 1 to GIS format shapefiles. Since the digitized maps were out of date, spatial data collection of new features had to be done to update the spatial 150 database. This exercise begun in 1998 and handheld GPS receivers (Garmin 12XL) were used for the job. Geographic features collected included roads, health centers, schools, administrative buildings, vegetation, hydrography and so on. Each of these categorical data was used to create thematic layers in ArcView GIS, at the parish and county levels for each district in Uganda. As of July 2010, ninety per cent of Uganda has been mapped and updated exhaustively. The only the Capital City, Kampala remains to be updated. After phase II of the CAMPUS project ended, the government of Uganda’s Ministry of Finance took over the funding of the Project under a special budget called the “developmental budget”. This type of budget is different from the usual “current budget” that most other public institutions, including the parent Department of Surveys and Mapping, are under. From 1996 to 2002, various institutions had GIS introduced into their workflow. By 1996, ArcView GIS was already being used by a USAID supported non-governmental organization called the Famine Early Warning System Network (FEWSNET). This NGO was using GIS for mapping food security conditions in Uganda as a step towards preparedness in times of drought conditions, especially in the north-eastern part of the country. FEWSNET partners include the Food and Agricultural Organization (FAO) and the Ministry of Agriculture, Animal Industry and Fisheries (MAAIF). Also in 1996, ArcView GIS was introduced at Uganda Wildlife Authority (UWA) to assist in patrol coverage and the formulation of general management plans. In the same year, MapInfo GIS was introduced at the Department of Geological Survey and Mines, Ministry of Energy and Mineral Development. An employee who went to Belgium for GIS training 151 returned with a student version of MapInfo and introduced it at the department. This department is a repository for geological data, which it had historically been storing up to 1996 in paper map format. In 1997, the Department of Geography at Makerere University had GIS introduced into its curriculum. Both undergraduate and graduate level courses are taught. At some point, the Department owned up to 30 computers, but at the moment, only 10 computers remain, the rest having been vandalized. This means that GIS is currently taught at a theoretical level. The National Water and Sewerage Corporation (NWSC), a government owned utility parastatal currently under the Ministry of Water and the Environment, had ArcView GIS introduced in its workflow in 1997 with assistance from the German government. NWSC’s mandate is to provide water and sewerage services to the major districts in Uganda. In 1999, through the activities of the Environment Information Network (EIN), the National Agricultural Research Laboratories (NARL) secured funding from the National Environmental Management Authority to establish a GIS lab at NARL’s research station in Kawanda. ArcView GIS software was used initially. In 2000, ArcView GIS was introduced at Makerere University’s Department of Forestry, including courses at the undergraduate level in GIS. In the same year, Mobile Telecom Network (MTN), a mobile telecommunications company introduced MapInfo GIS into the company’s radio planning and network optimization department. Also in 2000, Karamoja Data Center, introduced ArcView GIS into their work flow. In 2001, Uganda Bureau of Statistics (UBOS) introduced ArcView to take advantage of the spatial analysis capabilities of GIS. 152 2002 saw the first ESRI authorized reseller in Uganda open shop in Kampala, GeoInformation Communication (GIC). Most of the licenses for the ESRI family of GIS products is currently handled by GIC and ESRI Eastern Africa in Nairobi, Kenya. I discuss the history behind training and capacity building in GIS at Ugandan institutions next. 4.3.2 Training and Capacity Building in GIS in Uganda One of the major factors that supported the diffusion of GIS in Uganda was the academic institutions that trained staff of public sector institutions in developing countries in GIS, remote sensing and photogrammetry to equip them with the necessary skills needed to support GIS workflows in their home countries. The International Institute for Aerospace Survey and Earth Science (ITC) in the Netherlands (Toppen, 1991), today known as the faculty of Geo-Information Science and Earth Observation, University of Twente, was one of the first academic institutions in the world to offer formal training to students from developing countries, mainly staff of public sector institutions. The incoming students from various African countries were sponsored by their home governments to pursue certificates and master’s degrees in GIS, cartography, photogrammetry and remote sensing. ITC, specifically, was offering a wide range of GIS and Land Information Systems (LIS) courses, “especially for students from abroad (third world countries)” (Toppen 1991, p.7) One of the earliest people to undergo master’s level training in photogrammetry at ITC from Uganda was Filipo Mpabulungi, as he notes in a personal interview: “… We had already started talking about GIS in ITC. I was there in 1986/87… So basically, with 153 what they were calling computer aided cartography, basically that was the birth of GIS, although it was called Computer Aided Cartography” (F. Mpabulungi, personal communication, July 19, 2010). There were other institutions that played key roles in providing early GIS education and training to public sector employees from Uganda. The Royal Museum for Central Africa in Tervuren, Belgium, administered six-month long certificate level training in GIS to employees from the Wetlands Inspection Division and the Department of Geological Survey and Mines. The Regional Centre for Services in Surveying, Mapping and Remote Sensing (RCSSMRS) in Nairobi that had been set up by the United Nations Economic Commission for Africa (UNECA) (Tchindjang et al., 2005) offered short term training and workshops in GIS to employees of public sector institutions in Uganda, for example, staff from the National Agricultural Research Laboratories, and Kampala City Council. GIS training sessions were organized by UNEP/GRID/UNITAR between 1985-1989 in developing countries, including Uganda, but as Mooneyhan (1998) observes, this program was mainly a failure for myriad reasons, which I discuss in the next section. The surveying and mapping departments of the UK, Netherlands and Germany also trained staff from the Department of Surveys and Mapping, Uganda, in cartography and GIS (M. Murindwa, personal communication, July 28, 2010). In Uganda, the first academic institution to provide GIS training and certificates was Makerere University Institute of Environment and Natural Resources (MUIENR) in the early 1990s. It was also arguably the first academic institution to obtain GIS software (D. Mfitumukiza, personal communication, Jun 28, 2010). The Departments of Surveying, 154 Geography, Forestry, Veterinary Medicine and Computer Science soon followed suite at Makerere University, some offering undergraduate courses, while others, graduate level courses and certificates in GIS. Kyambogo University and Gulu University are the other public universities that offer GIS courses in their curricula today. A majority of the institutions in Uganda that use GIS in their workflows today employ the ESRI family of software. Looking at the historical trends in software usage in Table 4-1, Idrisi and PC Arc/INFO and ArcView were the first software introduced into Uganda by UNEP-GRID and other international development agencies. This was one reason, however, not the only reason for the successful diffusion of ESRI software in Uganda. The second reason for this trend is the academic institutions that provided early GIS education to international students from Uganda; these universities, and colleges used ESRI software in the training of their students. The most influential institution in this respect was the International Institute for Aerospace Survey and Earth Science (ITC) in the Netherlands. At some point in its curriculum in the 1990s, ITC started using ESRI software, in addition to its own proprietary ILWIS GIS software, in its courses. A comprehensive list of software used at Dutch universities by 1991 is provided by Toppen (1991). Students were able to return to Uganda with student copies of GIS software, which they introduced into the workflow of their parent institutions back home. The fact that ArcView was the most user-friendly GIS at that time meant that students had an obvious preference for it over other GIS software. The user-friendliness of GIS software in the 1990s was an increasingly important issue in the early 1990s, and ESRI’s ArcView GIS software, based on the Windows graphical user interface was a major breakthrough 155 in GIS user interface and design allowing it to edge the competition from other GIS software vendors, such as MapInfo Corporation. Moreover, these two reasons for the successful diffusion of ESRI software in Uganda were mutually reinforcing and allowed for the growth in the usage of this family of GIS software products. In 2002, GeoInformation Communication (GIC) was incorporated as the first ESRI value added reseller in Uganda. Its two main functions included, one, providing training services, and two, licenses to the various public, private, academic and NGO sector institutions in the country. GIC is under the ESRI Eastern Africa regional office based in Nairobi, Kenya. All licenses for ESRI software at Ugandan institutions are obtained directly from ESRI Eastern Africa, or indirectly through GIC. The historical timeline of commercial GIS software licensing and training in East Africa started with a company called Thunder Associates in Nairobi Kenya in the early 1990s. Thunder Associates closed shop in 1995, but for four years prior to that, the company had been distributing licenses for ESRI software in East Africa. In 1995, Oakar Services emerged as the new distributor for ESRI software licenses in the region located in Nairobi, Kenya. It also carried out GIS software training. In 2007, ESRI Eastern Africa began operations as the primary distributor of ESRI software in East Africa based in Nairobi, Kenya. Oakar Services’ focus shifted to the Erdas Imagine suite of software, although it still provides training in ESRI based GIS software. (A. Ori-Okido, personal communication, June 29, 2010) In the next section, I present a Diffusion of Innovations, and a GIS and Society Analysis (intellectual history perspective) of the evolution of GIS in Uganda. 156 4.4 Analysis This paper employs two lenses for analyzing the evolution of GIS in Uganda, the diffusion of innovations theoretical perspective and a GIS and Society perspective. The analysis is based on mixed methods applied to data acquired from personal interviews carried out with 91 public, private, NGO/IO and academic sector institutions in Uganda. Diffusion analysis hinges on four elements of diffusion, (1) the innovation, (2) communication channel, (3) time, and (4) society (Chan & Williamson, 1999a; H. J. Onsrud & Pinto, 1991; Pinto & Onsrud, 1995; Rogers, 1993, 1995, 2003; Valente & Rogers, 1995). GIS and Society’s intellectual history perspective analyzes the evolution of GIS technologies, how certain technologies succeeded, while others not, and the societal, personal, and institutional influences on the selection of GIS technologies in institutions (Couclelis et al., 2011; Elmes et al., 2005; McMaster & Harvey, 2010; Nyerges et al., 2011). I provide a Diffusion of Innovations perspective on the evolution of GIS in Uganda next. 4.4.1 Diffusion of Innovations perspective There are four main elements in the diffusion of innovations: 1) the innovation, 2) communication channels, 3) time, and 4) the social system. This follows directly from Rogers’ (2003) definition of diffusion, “the process by which an innovation is communicated through certain channels over time among the members of a social system” (Rogers, 2003, 11). I proceed to discuss each of these elements as they apply to the case of GIS diffusion in Uganda. 157 4.4.1.1 The Innovation The new idea or innovation in the case of this dissertation is GIS technology. I perform my analysis on the basis of the following questions related to the diffusion of GIS in Uganda: (1) what were the major factors that influenced the early adopters of GIS? (2) What challenges were faced by the early adopters of GIS? (3) How do the perceived attributes of GIS technology, for instance its relative advantage, affect its rate of adoption? (4) Has Uganda attained a critical mass of adopters signaling the take-off stage of the S-shaped curve, meaning that a communication network is in place allowing a self- sustainable rate of adoption of GIS technology? (5) What role do communication channels play in the adoption of GIS? (6) What about the role of a society’s social system? Factors that Influenced Early Adopters In answering the first question, what were the major factors that influenced the early adopters of GIS, qualitative analysis of the interview data revealed that the major factors that allowed for the early adoption of GIS at institutions in Uganda include, (1) the type of tasks carried out at that institution based on its mandate, (2) government support for public sector institution activities, (3) and donor agency support for public sector institutions. Since early adopters of GIS were mainly public and academic sector institutions, my analysis revolves around these. The institutions that adopted GIS first were those that dealt with environmental issues. In Uganda, the first three institutions to adopt GIS, according to Table 4-1, were all under the Ministry of Environment Protection. These institutions enjoyed strong government 158 support, in terms of budgetary support and staffing, for their activities because of the perceived importance of a sustainable environment relative to achieving the goal of development in Uganda. The most significant factor, however, was the support provided by donor agencies, and International Organizations. In Uganda, GIS was actually introduced by UNEP-GRID, and NORAD, both agencies providing funding for software licenses, hardware, and training opportunities for local staff. Challenges Faced by Early Adopters In answering the question, what challenges were faced by the early adopters of GIS technology, my analysis of the interview data revealed the following challenges: (1) the cost of licensing the software, and related costs of maintaining the hardware and software, (2) insufficient training facilities within the country, (3) a lack of awareness about the importance of GIS technology, (4) the lack of a support network of GIS specialists in the country, (4) a culture of corruption and unethical institutional practice, (5) bureaucratic roadblocks in the acquisition of equipment and in the hiring of new GIS staff, (6) brain drain, meaning the loss of skilled personnel to developed countries, (7) a project-driven instead of problem-driven approach to the use of GIS, and a (8) dependency on donor funding to support GIS activities. The cost of licensing GIS software is one of the most significant challenges to the early diffusion of GIS software. For the ESRI family of software, the most pervasive GIS software, license prices are based on the institutional sector. For academic institutions like Universities, a special license package called Lab Kit can be purchased for $5000. This provides 25 single-use licenses. For unlimited floating licenses at academic 159 institutions, an institution has to pay $40,000. This package would make the most economic sense if, instead of every single department operating their own GIS courses, a single GIS center was created from pooled common resources of the various stakeholders, that is the various departments each independently offering GIS courses at the same university. However, the reality is that the various departments have not been able to come together to address this institutional problem, and instead operate in a decentralized capacity independent of the other departments with little or no cooperation among departments. Public and Private sector institutions are considered commercial entities by ESRI, thus, the costs of licenses are much more expensive than those for academic institutions. A respondent that I interviewed at one of the public sector institutions explained the cost of licensing GIS software at the institution: “ArcGIS, ArcInfo – 5 floating licenses for $9200 one-time fee for version 9.2. You would need to pay an extra $9200 for an upgrade to the next version, usually every year. If you do not subscribe, then you do not get support. In the 5 floating licenses, we have 3 Spatial Analyst licenses, and 2 3D Analyst licenses. That is within the $9200 package, though there is a way they compute how much each costs, but I have given you the total sum. Because each of the these extensions, e.g. Spatial Analyst and 3-D Analyst, each is an extra $500, so probably, if you do the math, the licenses without Extensions cost about $6700 for the 5 floating ArcGIS licenses.” Moreover, there were not enough training facilities for capacity building in GIS skills in Uganda in the early days. Institutions like NEIC relied heavily on the UNEP-GRID 160 UNITAR program for training opportunities in GIS, and as Mooneyhan (1998) explains, this attempt by UNEP was not particularly a success for myriad reasons. One reason for this was the lack of hardware and software in developing countries. Second, when trainees returned home to their institutions, they were often times assigned duties that had nothing to do with GIS at all. This was because these institutions had their own priorities, methods of work and operating rules. Third, the lack of follow-up by the trainer led to a demoralizing effect on the trainees. After initial training, the trainees had real need for support in terms of technical and moral support, yet this was not available. Fourth, the problem of “brain drain” eventually occurred, and this is common problem in any technology transfer program. The trainees, as a result of their training, achieved the status of better trained professionals with rare skills, which made them more market and thus, more mobile, both internally within their own countries and institutions, and abroad. Thus, the skills acquired did not necessarily transfer to the home institutions that they were meant for. There is a flourishing culture of corruption and unethical institutional practice at Ugandan public sector institutions which was a hindrance to early GIS adoption. An illustration of how this is significant is in the national cadastre system. It takes anywhere from a few weeks to a few years to obtain a land title in Uganda, depending on how much of a bribe one is able to provide the concerned authorities. This unethical practice flourishes because the process of obtaining a land title, right from the initial surveying of the land to the actual issuance of the title involves a number of “mechanical” steps, and human intervention. The actual production of a title deed map is done manually using paper/pen 161 cartography, the records are stored in paper files and folders, and the files and folders have to pass through thick bureaucratic tape before they are eventually approved for the issuance of a title. Introducing GIS to automate the cartographic and data storage aspects of the cadastre would mean the reduction of middle-man intervention, and the potential loss of earnings from bribes. This makes GIS undesirable to bureaucrats and technocrats benefiting unethically from the current inefficient cadastre management system. Moreover, the level of bureaucracy at public sector institutions is a significant hindrance to the adoption of GIS technology. Procurement and acquisition of new equipment, say GIS hardware, at government ministries is a non-streamlined process that can take months to bear fruit; the hiring of new skilled staff, for example, recent graduates from the local universities well conversant with GIS, is very difficult, and this is partly because of job insecurity within the institutions themselves; and government budgets for institutions that do mapping and environmental activities are insufficient, for example, until recently, the total annual budget allotted to the Department of Surveys and Mapping was $25000 – not even enough for mapping a single city, leave alone an entire country. Perhaps one of the most significant deterrents to the early adoption of GIS in Uganda was the way that GIS was introduced into the country. It was donor supported and project driven. With the end of the project came the end of GIS at a given institution, in many cases. Because the government did not appreciate the importance of the technology, it did not provide budgetary considerations in its annual allotment of fiscal funding to institutions that used GIS. Thus, licenses expired, hardware decayed from a lack of maintenance, and GIS staff members were reassigned to non-GIS roles over time. 162 Perceived Attributes In answering the question, how perceived attributes of GIS technology affected its rate of adoption, the relative advantage of GIS technology over old technologies such as paper, pen and ink cartography was a significant factor leading to early adoption. Going from analog to digital methods helped expedite the map production process at the Department of Surveys and Mapping, for example. The excitement that computers and automated mapping techniques brought is easily inferred from a statement one of my interviews made: “Yeah, and for the first like 3 years there was no computer, then all of a sudden, the Germans brought them and we said, eh! At that time they were 286 processors – they were running AutoCAD. The biggest size we had was like 1GB. It was very ‘big’. Most of these others were 256MB (he laughs).” However, with regard to compatibility of GIS technology with existing methods of mapping, GIS posed a threat to the existing status quo. Paper, pen and ink cartography was being actively used by professional cartographers in institutions such as the Department of Surveys and Mapping, National Water and Sewerage Corporation, and the Kampala City Council. The adoption of GIS would mean that the old profession of cartography would be rendered obsolete, and lead to possible job losses, or require fresh training for an aging staff no longer interested in learning new methods. GIS, thus, was not compatible with existing mapping methods in Uganda, and this slowed the adoption of the technology in some institutions. Moreover, the level of complexity of GIS technology was high. Early versions of GIS ran on platforms requiring command line based interaction with the software, for example, 163 PC Arc/INFO. This was a hindrance to adoption of GIS. As one of my interviewees noted, “Yeah, but we didn’t use it much. It was command line based. So you have to put in (a lot of effort)…. Making a small mistake, like you have a dot (he laughs)… it was not user-friendly for the staff.” Later, the advent of ArcView which had a Graphical User Interface (GUI) made the GIS software more user friendly, and hence, more adoptable. This user friendliness of ArcView, and the fact the version 3.x was distributed free of charge by ESRI in the late 90s, explains the current popularity of ArcView software, which is still being used to this day in spite of newer software releases by ESRI, for example ArcGIS, later. 4.4.1.2 Time: Rate of Adoption In answering the question, has Uganda attained a critical mass of adopters signaling the take-off stage of the S-shaped curve, there was insufficient data collected in this particular research to answer this question. A larger sample of institutions would be needed that encompasses all sectors of the institutional framework so as to gauge the percentage level of adoption of GIS technology over time. Future research should address this question. 4.4.1.3 Communication Channels: Heterophily and Homophily In answering the question, what roles do communication channels play in the adoption of GIS, in the case of Uganda, I argue that heterophilous communication played a larger role than homophilous communication. According to Rogers (2003), the transfer of ideas occurs most frequently between two individuals who are similar, or homophilous. “Homophily is the degree to which two or more individuals who interact are similar in 164 certain attributes, such as beliefs, education, socioeconomic status, and the like. Heterophily, the opposite of homophily, is the degree to which two or more individuals who interact are different in certain attributes” (Rogers 2003, 19). Because of the lack of knowledge about GIS in the early 90s, homophilous communication among peers was not common place. There were few opportunities to interact with other GIS enthusiasts at conferences or workshops, although a few significant efforts were made to encourage homophilous communication, for example, the Environment Information Network (Gowa, 2009), and workshops organized by the Regional Mapping Center in Nairobi. In most cases, GIS experts from UNEP-GRID and other expatriates were the major agents as communication channels in a heterophilous relationship with Ugandan staff at public sector institutions. 4.4.1.4 Social System In answering the question, how significant is the role played by a society’s social system, my analysis shows that social structure and norms are significant to the diffusion of GIS technology. Uganda’s public, private, academic and NGO sector institutions are part of a larger society where culture, norms and values play a major role in an individual’s decision making process. There is a culture of communal living, communal rights and ownership, and a respect for cultural values, norms, practices, and above all, a cautious attitude toward those in authoritative and powerful positions of governance, for example, local politicians. Political leaders use patronage as a tool to seek political favors in return for loyalty (Adjibolosoo & Ofori-Amoah, 1998; Adu-Febiri, 1998; Lawal, 2007; OAG, 2004; Obuah, 2010; Samatar, 1999). This situation is common in developing countries, 165 for example, the countries in North Africa that experienced the Arab Spring (Hallward, 2011; Hayes, 2011; Schwartz, 2011; Snider & Faris, 2011). This is in direct opposition to the ideology of capitalism in the West, which encourages individualistic tendencies and ownership, for example, with respect to land, efficiency and time management, profit oriented commerce, a free market, a credit-based system, freedom of expression, and open political space. In developing societies, GIS as a technology is viewed by some with suspicion, and associated with this western way of life, which is not necessarily in agreement with the social norms in Uganda. GIS tends to empower the policy makers and not necessarily the local citizens. There is a fear among local NGOs, for example, that if they adopt GIS, the results of their GIS analysis could directly challenge the official analytical results of the government agencies, for example, the Uganda Bureau of Statistics (UBOS). For example, the NGO might find that its GIS analysis shows that 50% of the forests in the country are degraded, while the official government statistics state that only 20% are degraded. The NGO risks being witch- hunted and taunted as an anti-government organization, possibly working for the opposition party – the backlash could be serious. Due to this fear for authority, such an NGO will not adopt GIS, and will instead opt to use official government data for its work so as to work in harmony with the ruling party. One of my interviewees, a staff member of an environmental NGO explained, “And every time we have wanted GIS related material, we have had to outsource it to NFA. And then the way our work is structured as NGOs, if we produced our own work, we would have a lot of questions – authenticity of this work. People would question – but who has done that (work)? So that’s why 166 sometimes we prefer to work with government institutions, so that if we have a report, they own the results as well, because as an advocacy NGO, we wouldn’t want to do work on our own. We would want to work in partnership with them (the government institutions) so that when we are criticizing and we are making our analysis, they can own up to the analysis and say yes, it is true, we did the analysis together, and this is what came out. So when we make our proposals (recommendations) to them, it is easy for them to take it up than when we do our work independently alone as NGOs, because that will be like giving hearsay and all that and it is easy for government to rubbish our work. And you know our situation, right? If you are too critical of the government, you are branded the ‘opposition’. So we want to remain relevant, but work in partnership with the government” (anonymous, personal communication, June 24, 2010). Next, I analyze the evolution of GIS in Uganda from the intellectual history perspective of GIS and Society. 4.4.2 GIS and Society in Uganda: An Intellectual History Perspective The intellectual history perspective is concerned with tracing and understanding the dynamics through which dominant GIS technologies are selected out of a set of possible options, and how the processes leading to these selections are linked to institutional, governmental and personal influences (McMaster & Harvey, 2010; Nyerges et al., 2011). It further examines the adopted technologies in light of other promising technologies that were overlooked (McMaster & Harvey, 2010). This section discusses the influences of external change agents on GIS, followed by societal, institutional and personal influences. 167 4.4.2.1 Influences of External Change Agents GIS was introduced into Uganda by UNEP-GRID in the environmental management public sector in Uganda in 1989. This was as a result of initial efforts that can be traced back to the 1964 session of the UN Economic Commission for Africa that called for the creation of regional centers for training in photo interpretation, photogrammetry, and airborne geographical surveys (Tchindjang et al., 2005), which were followed by the launch of the Landsat satellites in the 1970s, the parallel development of microcomputers, and GIS software for these platforms. UNEP’s relationship with ESRI (Mooneyhan, 1998) played a key role in the dissemination of GIS software in developing countries. It is because of the relationship fostered between UNEP-GRID and ESRI that the dominant GIS software technology in Uganda in the early days was indeed ESRI based ArcView software as is evident in Table 4-1. Another reason for this trend at that time was the fact that ArcView was simply the most user friendly vector GIS software at the time, and one of the 3.X versions of this software was available free of charge. This allowed ESRI’s brand of software to gain a foothold in terms of a user base in Uganda. The cost of licensing subsequent versions of ESRI software was definitely a hindrance to Ugandan institutions, as is evident from the responses from my interview subjects: “The type of GIS software we use depends on the type of resources we have, but we are basically using ILWIS, ArcView 3.2, 3.1, 3.3. We have not yet got into ArcMap because you need resources to access these recent versions of ESRI” (M. Isabirye, PhD, National Agricultural Research Laboratories, Personal Communication, July 28, 2010). 168 4.4.2.2 Societal, Personal and Institutional Influences There are also societal, personal, and institutional influences that governed the selection process of GIS software in Uganda in the early days. First, the fact that UNEP-GRID introduced both Idrisi and ArcView into Uganda, and further organized training for these particular GIS software through the UNITAR program (Mooneyhan, 1998) already introduced a bias that tended to influence strategic software selection decisions in other related public institutions toward the already “known” software. Second, the fact the fact that the majority of Ugandans at public institutions using GIS did their certificate and Master’s level training at the International Institute for Aerospace Survey and Earth Science (ITC), Netherlands, which primarily exposed their students to ILWIS and ArcView software. Upon the completion of their studies, students returned to Uganda and had a natural preference for GIS software that they were already familiar with, for example, ILWIS, ArcView and MapInfo. Unfortunately, ILWIS was not as user friendly as ArcView, hence the preference for the latter. As one of my interview respondents’ asserts, with reference to the pervasive use of ArcView GIS due to the waiving of licensing fees by ESRI, “Yeah, it was a marketing strategy; people got used to it and …. So those are the programs. We have Idrisi to a certain extent, but we don’t use it so much; even MapInfo. (Interviewer: doesn’t MapInfo require a license also?) Yeah, it does. You see, some of these programs, what happens is that, when you go for trainings, short courses and all that, during the training, they become part of the deductive material they give you, and they say, you will be able to go with the program of ArcView, MapInfo… whatever they are using in their training, they 169 make sure that you can also get … just to ensure that you can continue using it after” (M. Isabirye, PhD, National Agricultural Research Laboratories, Personal Communication, July 28, 2010). Another interviewee also stressed the preference for the ESRI family of software, the project-driven nature of GIS projects, and the related preference for the software based on the funders of a project: “We use ArcGIS in particular because of the type of analysis or work that we do. Originally we were using MapInfo, but then we found that when we wanted to do some complex analysis, we couldn’t handle it with MapInfo. And it was just introduced last year (ArcGIS) in 2009, because currently we are running a project with Spatial Dimension (PTY) Ltd. of South Africa” (A. Alaba, Department of Geological Survey and Mines, Personal Communication, August 04, 2010). 4.5 Conclusion GIS diffusion in developing countries was heavily influenced by donor agencies in the West, and was closely related to the evolution of mapping science. This had negative implications for sustenance of the diffusion of GIS as an innovation in the public, private, NGO and academic sector institutions as GIS was bound to the chains of donor funded projects. However, the pervasiveness of ESRI-based ArcView software was certainly supported by this trend, and also by influence of academic training institutions in Europe that trained students from Uganda in GIS, like the International Institute for Aerospace Survey and Earth Science (ITC) in the Netherlands. From an intellectual history perspective of GIS and Society, ArcView software became pervasively used at Ugandan institutions because of the links to UNEP-GRID and ITC, the free version of ArcView 170 3.X, and the user-friendliness of the software. Early adopters of GIS in Uganda faced various challenges including inadequate training, unaffordable GIS licenses, and a lack of awareness. The evolution of GIS in Uganda can be used to draw broader conclusions about the diffusion of GIS in other developing countries so as to better understand the development and use of the software in the developing world today. 171 Chapter 5 : Diffusion of GIS in Developing Countries – A Case study of Uganda’s Public and Academic Sectors 5.1 Introduction Ever since its advent, GIS has played an increasingly important role in the public and academic institutions in developed nations. Myriad publications on the diffusion of GIS in Europe, Australia and North America are in circulation (Chan & Williamson, 1999a; Masser & Campbell, 1996; Miellet, 1996; H. J. Onsrud, 1991), and they mostly indicate a classical diffusion pattern based on the four elements of diffusion as defined by Rogers (2003). Diffusion of GIS in the developed world has been characterized by institutional or organizational challenges related to social structure of the organizations concerned, amongst others. Literature on the diffusion of GIS in developing countries is limited. Some work has been done on GIS diffusion in India, Brazil, China and Egypt (Barrett et al., 2001; Borges & Sahay, 2000; Cartwright, 1993; Sahay & Walsham, 1997; Salem, 1994; Yue et al., 1991), but a lot more research remains to be carried out about the diffusion of GIS in developing countries within the unique socio-economic and political contexts of those countries. The objective of this dissertation chapter is to answer the research questions, what is the current state of GIS in developing countries? I critically examine the diffusion of GIS technology in the public and academic sectors of developing countries, using Uganda as a 172 case study, and further, use a GIS and Society framework to analyze the impact of GIS on society and vice versa. The findings in this research will help stakeholders in the GIS industry, namely, government officials, the international donor community, GIS consultants, academics, and society at large better understand the dynamics behind the diffusion of software technology from developed countries to developing countries so as to make better decisions about technology transfer in the future. In sections 5.2, I provide working definitions for the public and academic sectors then proceed to review the literature on GIS diffusion and GIS and Society in each of these sectors. Section 5.3 presents the findings of this research with respect to GIS diffusion, and GIS and Society in the public and academic sectors of Uganda. A Diffusion of Innovations, and GIS and Society analysis follows in sections 5.4 and 5.5, respectively. 5.2 The Public Sector and Academic Sectors In this section, a literature review of GIS in the public and academic sectors is provided. For each sector, a working definition of the sector is first provided, followed by a review of the literature on GIS diffusion in industrialized countries, and then developing countries. 5.2.1 Public Sector Broadly defined, the public sector refers to, “that part of a nation’s economy concerned with providing basic government services” (InvestorWords, 2012). The public sector includes services and infrastructure for the common good a nation as whole provided by the ruling government, for example, the police, transportation networks, public transit, 173 agricultural and environmental services, security, healthcare, education, urban and regional planning services, and so on. 5.2.1.1 Developed Countries The nature of many of the services in the public sector tend to inherently entail a geographical element, which makes the application of GIS for planning an attractive option for technocrats and field experts. The use of GIS for planning of public sector services is widely reported in GIS diffusion literature about developed countries such as the United Kingdom, the Netherlands, the European Union Countries and the United States, for example, see (Chan & Williamson, 1999a, 1999b, 2000; Ciancarella, Craglia, Ravaglia, Secondini, & Valpreda, 1996; Crompvoets et al., 2008; Gilfoyle & Thorpe, 2004a, 2004b; Masser & Campbell, 1996; Masser & Craglia, 1996; Volman, 2004; Wiggins, 1993). Most of the literature on GIS diffusion in the industrialized nations tends to be from the 90s and early 2000s because this was the time that GIS diffusion was actively taking place in the industrialized nations. By the mid-90s, there were high rates of adoption of GIS in public sector agencies in the US, and western Europe, for example, over 78% in the US (Wiggins, 1993), about 29% in the UK (Gilfoyle & Thorpe, 2004a; Masser & Campbell, 1996), 24% in municipalities and 43% in County local governments in Denmark (Kiib, 1996), 65% in Italy (Ciancarella et al., 1996), 68% in France’s large municipalities (Miellet, 1996), and 75-82% in the local governments of large cities in Germany (Junius et al., 1996). There were a few industrialized nations with low GIS 174 penetration rates, for example, only 4% adoption in Portugal (Arnaud, Vasconcelos, & Geirinhas, 1996), and a low rate in Greek municipalities (Assimakopoulos, 1996). The causes of low adoption rates in some European countries resonate with the findings in this paper about developing countries. In Greece, for example, GIS diffusion is linked to external funding from the EU as International aid, and funding from UNEP. Moreover, organizational difficulties also pose a problem to the adoption of GIS, for example, a lack of economic resources for maintenance, low priority given to GIS in decision making, dependency on donor funding to support GIS, and a lack of social networking within the Greek GIS community to allow for exchange of ideas and sharing of scarce resources. (Assimakopoulos, 1996) Across the board in the US and Europe, the most pervasive GIS software used by public sector agencies in the period 1993-96 was PC Arc/Info and/or ArcView from the GIS software vendor, ESRI: 90-95% in US municipalities and counties (Wiggins, 1993), 59% in Italian provincial local government agencies (Ciancarella et al., 1996), and 22% in the UK (Masser & Campbell, 1996). The other prominent GIS software products commonly in use at that time included MapInfo from MapInfo Corporation (today owned by Pitney Bowes), and GeoMedia from Intergragh Corporation. The early applications of GIS technology in the public sector of developed nations were vast; however, there were commonalities in the types of applications, which fueled the diffusion of GIS. In the US, there was a predominance of land parcel, and demographic mapping applications; the most common geographic features being mapped based on GIS 175 methods included streets, political boundaries and land parcel boundaries (Wiggins, 1993). In the UK, the first GIS applications included studies carried out in the 1960s; sub regional planning studies in Nottingham and Derbyshire, and transportation studies such as Merseyside. In the early 70s, the Local Authority Management Information System (LAMIS) was developed by International Computers Limited at Leeds. In the early 70s, the British government funded the development of a national gazetteer, a digital system of postal addresses. It was also around the same time, the Ordnance Survey of the UK started digitizing base scale maps. (Gilfoyle & Thorpe, 2004a) In Germany, the diffusion of GIS in the public sector was mainly linked to activities of the surveying and mapping departments and the development of an effective land information management system (Junius et al., 1996). In Portugal, early GIS applications included land use planning for protected areas by the Ministry of Environment, and municipal planning by the local government, including water and sewerage planning (Arnaud et al., 1996). In Italy, early GIS applications included, map production, environmental monitoring, network management, and military applications (Ciancarella et al., 1996; Cremona & Ciancarella, 2006). 5.2.1.2 Developing Countries The diffusion of GIS in the public sector of developing countries has been research by some, for example, see Cavric, Nedović-Budić, and Ikgopoleng 2003; Gibson 1998; Borges and Sahay 2000; Câmara et al. 2004; Câmara et al. 2005. GIS diffusion in Botswana started between 1985 and 1990, and is linked to donor funding from UNEP- 176 GRID and NORAD (Cavric et al., 2003), which resonates with the findings in this paper on diffusion in Uganda. Training of government employees in GIS was mainly provided by the International Institute for Geo-Information Science and Earth Observation (ITC), Netherlands. This influenced the trends in early GIS software in Botswana, which included ITC’s proprietary ILWIS and ESRI’s PC Arc/Info (Cavric et al., 2003), which also echoes the findings in this paper about GIS diffusion in Uganda. Between 1991 and 1995, a number of government agencies had started exploring and adopting GIS technology including, the Town and Regional Planning Department, the Mapping and Surveying Department, Lands Department, and the Geological Survey. This differs with the findings in this paper about Uganda, where GIS diffusion started in the environment sector, and a little later in the Department of Surveys and Mapping, and the Department of Geological Surveys and Mines. Unlike Botswana, the City Planning Department experienced very late and unsuccessful GIS diffusion in the early 2000s, while the National Planning Authority in the Ministry of Finance has never even tried to implement GIS technology as of August 2010. Similar to Uganda, Botswana’s initial slow GIS diffusion rate was attributed to a lack of coordination among agencies actively involved in the diffusion process. These included the University of Botswana, Town and Regional Planning Department, the Power Corporation, the Mapping and Surveying Department, Crop and Forestry Production Department, and the Roads Department. Each of these agencies was in competition to take the national coordination of GIS leading role (Cavric et al., 2003). This directly resonates with the current situation in Uganda with various government agencies 177 squabbling over the leadership role with regards to being a central authority on GIS and geospatial data in the country (Amadra, 2003a, 2003b; Gowa, 2009; Karatunga, 2002; Lwasa et al., 2005; Muhwezi, 2005, 2006; Musinguzi et al., 2004). This is one of the major obstacles to GIS diffusion in developing countries. Cavric et al. (2003) identify two other factors for the slow diffusion of GIS in Botswana; a shortage of trained staff and a lack of standardized and reliable data. For effective diffusion of information systems, Gibson (1998) recommends corrective action from the government to address the lack of skilled personnel to support informatics technology diffusion in a country. There is also a need for an on-going process of evaluation and feedback to assess the success of the implementation of an information system so as to make necessary adjustments (Gibson, 1998). Sahay and Walsham (1997) identified four factors that slow down GIS implementation in organizations in developing countries: institutional arrangements, sustainability, data management and power-relations (Sahay & Walsham 1996 in: Borges & Sahay 2000). These factors resonate with the findings in this paper strongly, as will be discussed in a later section. Despite these setbacks in the developing world, successful GIS diffusion has been reported in Brazil (Borges & Sahay, 2000; Câmara et al., 2006) and India (Barrett et al., 2001; Puri, 2006; Sahay & Walsham, 1996). Next, I present a literature review on GIS diffusion in the academic sector in developed, and developing countries, respectively. 178 5.2.2 Academic Sector The academic sector is that part of the economy that focuses on education as a service, or as a commodity. The academic sector includes institutions such as universities, K-12 schools, and research institutions, and these are either public- or private-owned. 5.2.2.1 Developed Countries Literature on GIS diffusion in the academic sector in developed countries (Chrisman, 1998; Dale, 1991; Heywood & Petch, 1991; Jackson, Schell, & Taylor, 2009; Kemp & Goodchild, 1991; Kerski, 2008; Mark, Usery, & Mcmaster, 2005; Thompson, 1991; Toppen, 1991; UCGIS & AAG, 2006; Usery & Mcmaster, 2005; Wise & Burnhill, 1991) addresses issues concerning GIS courses taught at institutions of higher learning, software used in the classroom, the relationship between the institutions of higher learning and the other sectors and the role of the academic sector in GIS capacity building in a nation’s socio-economic development. The academic origins of GIS in the West are well articulated by Chrisman (1998). The academic sector in the developed nations played a key role in developing the tools that constitute GIS today. From the 1920s through the 1950s, John K. Wright at the American Geographical Society had made contributions to cartography and quantitative geography. These influenced the work of A. H. Robinson’s work on cartography that simplified the dimensions of geographical measurements as points, lines and polygons. The Department of Geography at the University of Washington under Donald Hudson was one of the earliest locations of early GIS work in the 1950s. The teaching of quantitative Geography at this university inspired a new crop of graduate students in 1955 that later became 179 influential in defining the very principles of the field to be known as GIS, for example, Brian Berry and Waldo Tobler. Before 1960, Edgar Horwood, a faculty member with the Department of Civil Engineering and Urban Planning developed an automated method to perform geocoding. This inspired Howard Fisher to found the Harvard Lab for Computer Graphics at Harvard University in 1965, which produced the created a sophisticated computer-aided cartographic software product, the first GIS in the United States, called SYMAP. Jack Dangermond worked as a graduate research assistant (master’s in landscape architecture) at the Harvard Lab before founding ESRI in 1969. “Between 1985 and 1992, the advent of Idrisi, MapInfo, PC Arc/Info and ArcView desktop software diffused GIS within organizations” (Kerski 2008, 1). This diffusion created a demand for GIS education in the 1990s. Kerski (2008) conceptualizes GIS education in four distinct ways: (1) professional development, (2) research about GIS, (3) teaching about GIS, and (4) teaching and learning with GIS. In Uganda’s academic sector, the all four conceptualizations of GIS education apply; however, the last two characterize GIS education most strongly at institutions of higher learning. The principles of GIS and the applications of GIS in various fields are taught in over five academic departments at Makerere University, the largest public university in Uganda. The most significant academic institution in Europe in terms of capacity building in GIS in developing nations is arguably the International Institute for Geo-Information Science and Earth Observation (ITC) in the Netherlands. Many Ugandan students attended this institution of higher learning in 80s and 90s to pursue their master’s degrees in GIS, Photogrammetry and Automated Cartography. Toppen (1991) teases out the various 180 needs of employers with regard to GIS education in the Netherlands in the early 90s by classifying GIS jobs into three categories in line with conventional organizational structure: (1) GIS managers, (2) GIS users, (3) GIS developers. He notes that based on a particular category, there is a difference in the balance between Geography, GIS, and Computer Science. He goes on to suggest a delicate mix of courses that should constitute the curriculum to teach GIS at institutions of higher learning, with particular regard to each of the three categories. Courses include geography, computer science, information management, spatial analysis, planning, GIS theory, GIS applications, business management, and hands-instruction. GIS software at Dutch universities by the early 90s was quite diverse; however, Arc/Info was the most pervasive throughout Dutch Universities. Other GIS and CAD software included Idrisi, ILWIS, GIMMS, MapInfo, Atlas Graphics, Erdas and AutoCAD. In the United States, the curriculum for GIS was strongly influenced by the activities of the National Center for Geographic Information and Analysis (NCGIA) (Kemp & Goodchild, 1991). The NCGIA proposed a core curriculum for the development of three types of GIS courses at Universities, (1) Introduction to GIS, (2) Technical Issues in GIS, and (3) Application Issues in GIS. Laboratory exercises in each of these courses were strongly emphasized so as to give students hands-on practice with using GIS software and hardware. A recent publication of the Association of American Geographers (AAG) and the University Consortium for Geographic Information Science (UCGIS) called the Geographic Information Science and Technology Body of Knowledge (UCGIS & AAG, 181 2006), edited by David DiBiase (of ESRI Educational Services) and others, provides a detailed and well-structured model curriculum for GIS in the United States. UCGIS is a consortium of GIS educators and researchers in the United States formed in 1994 (Usery & Mcmaster, 2005) that provides a forum for consensus on matters of GIS education and research in the US. The curriculum is a hierarchical outline composed of three tiers, (1) Knowledge Areas, (2) Units, and (3) topics. The first tier consists of ten knowledge areas in Geographic Information Science and Technology (GIS&T); the second tier consists of several constituent units that are meant to be coherent sets of topics that embody concepts, methodologies, and applications in GIS&T; and the third tier consists of topics, which are the constituent topics for each unit. 5.2.2.2 Developing Countries The Literature on GIS in the academic sector in developing countries (Aina, 2009; Dunn, Atkins, Blakemore, & Townsend, 1999; Dunn, Atkins, & Townsend, 1997; Hall, 1999; Jackson et al., 2009; Lauriault & Taylor, 2007a; Owoeye & Oyebade, 2011; F. D. R. Taylor, 1986, 2004, 2005; Teferra & Altbach, 2004; Yapa, 1998) examines challenges and problems of GIS education in developing countries. The literature covers related subjects on GIS education in developing nations. Challenges facing GIS education in emerging countries are financial and technical in nature and must be addressed for appropriate use of GIS technology in emerging nations (Hall, 1999). For more effective teaching of GIS courses to students from developing countries in Europe, courses need to be tailored towards real world problem solving so as to be more appropriate to the students’ needs (Dunn et al., 1999). The major challenges facing higher education 182 research in many developing nations include a lack of sufficient funding for research, inadequate staff development, and a lack of organized data warehouses available for researchers in higher education (Owoeye & Oyebade, 2011). Other challenges to higher education in Africa include political and economic instability, brain drain, a lack of academic freedom, inadequate support for research from parent governments, gender imbalance in higher education and lack of private sector involvement in higher education. Aina (2009) asserts that there is a lack of skilled geomatics personnel in Saudi Arabia, thus, calling for initiatives to promote geomatics education in that country, which echoes the need for GIS education in the curricula of developing nations in general by some authors (Lauriault & Taylor, 2007b; F. D. R. Taylor, 1986, 2006). There is also call for developing nations to consider the constantly evolving field of GIS relative to institutional challenges in academia (Jackson et al., 2009). Some have questioned the appropriateness of GIS technology in developing nations and have examined issues related to the transfer of capital-intensive technology from the west to labor-intensive economies in developing countries (Yapa, 1991). Next, the results and findings of this mixed methods research are presented for the public and academic sectors in Uganda, respectively. 5.3 Results In the first part of this section, the findings about the current state of GIS in the public sector is presented under the following subtopics: (1) the introduction of GIS into the institutions, (2) GIS software and hardware usage, (3) applications of GIS, (4) GIS data collection, sources, and sharing, (5) training and education in GIS, (6) government 183 funding for GIS, (6) Government funding for GIS activities, (8) the impact of GIS on society, and (9) the challenges to the diffusion of GIS. In the second part of this section, the findings about the current state of GIS in the academic sector are presented under the following subtopics: (1) introduction of GIS into the curriculum, (2) GIS software and hardware usage, (3) impact of GIS on society, and (4) Challenges to GIS education. 5.3.1 The Public Sector The public sector has seen the highest level of GIS penetration in Uganda as compared to the other sectors of the economy. Table 5-1 shows the public sector institutions that were selected for interviews in this research, and Table 5-3 shows that 18 of the 21 public sector institutions interviewed use GIS to facilitate their activities. These institutions are government owned, and fall under various government ministries, such as the Water and Environment; Lands, Housing and Urban Development; Tourism, Trade and Industry; Finance; Energy and Mineral Development; and Local Government. Of all these Ministries, the largest number of institutions in the sample selected that use GIS are under the Ministry of Water and Environment. Table 5-1: Public Sector institutions at which interviews were performed ID Institution Name Ministry 1 NEMA - National Environmental Management Authority Water and Environment 2 NFA - National Forestry Authority Water and Environment 3 Wetlands Management Department Water and Environment 4 Department of Surveys and Mapping - Cartography Section Lands, Housing and Urban Development 5 Department of Surveys and Mapping - CAMPUS Project Lands, Housing and Urban Development 6 UWA - Uganda Wildlife Authority Tourism, Trade and Industry 7 Department of Geological Survey and Mines Energy and Mineral Development 184 8 NARL - National Agricultural Research Laboratories Agriculture, Animal industry and Fisheries 9 Northern Uganda Data Center - NUDC Office of the Prime Minister 10 UBOS - Uganda Bureau of Statistics Finance 11 UNRA - Uganda National Roads Authority Works and Transport 12 Ministry of Agriculture, Cartographic Office Agriculture, Animal Industries and fisheries 13 KCC - Kampala City Council Local Government 14 Mbarara Police Internal Affairs 15 UNHRO - Uganda National Health Research Organization Health 16 Department of Petroleum Exploration and Production Energy and Mineral Development 17 NPA - National Planning Authority Finance 18 Uganda Police, Central Police Station, Kampala Internal Affairs 19 NW&SC - National Water and Sewerage Corporation, and Kampala Water Ltd Water and Environment 20 Department of Meteorology Water and Environment 21 DWD - Directorate of Water Development Water and Environment 5.3.1.1 Early GIS Introduction, Project-driven GIS and Donor Funding GIS was first introduced into the public sector by UNEP-GRID at NEIC (today NEMA), and by Norwegian government’s NORAD at the Department of forestry (today NFA) in 1989; both institutions under the then Ministry of Environment Protection (today Ministry of Water and Environment). See Table 5-2. This was shortly followed by the introduction of GIS at the Wetlands Division (today Wetlands Department), Ministry of Environment Protection in 1990. The Ministry of Environment Protection was, thus, the pioneer umbrella institution under which GIS was introduced into Uganda’s public sector. It is evident that the initial purpose of GIS in Uganda was to help global efforts towards sustainable environmental practices and to control environmental degradation. 185 In the mid-to-late 90s, GIS was introduced at the Department of Surveys and Mapping, Uganda Wildlife Authority (UWA), the Department of Geological Survey and Mines, and the National Agricultural Research Laboratories/Organization (NARL/NARO). In this time period, GIS was being used to promote national development, as updated maps were being seen as a necessary ingredient for development. The tourism and mineral public sectors introduced GIS into their workflows during this time period. The 2000s saw the introduction of GIS into various other public sector institutions in the finance, agricultural, transport, and utility sectors of the economy. An interesting trend in the early introduction of GIS at public sector institutions in Uganda is that in almost all cases GIS was introduced through a project funded by an international donor agency or government. International donor agencies that funded projects that led to the eventual introduction of GIS included UNEP-GRID, International Atomic Energy Agency (IAEA), the World Bank, European Union and African Development Bank (ADB). Development arms of developed countries such as Norway (NORAD – Norwegian Agency for International Development), France, Germany, Belgium, and the Netherlands also supported projects that led to the introduction of GIS in Uganda’s public sector. A “project”, or more strictly, an international development project, within the context of international development is a program with a fixed time duration specifically designed for economic and social needs of developing countries (Ahsan & Gunawan, 2010). The main objectives of International development projects include: poverty alleviation, improvement of living standards, environment and basic human rights protection, 186 assistance to victims of natural and people-caused disasters, capacity building and the development of basic physical and social infrastructures (Ahsan & Gunawan, 2010; Clements, 1993). Funders of such projects are usually highly developed countries such as the United States, Britain, Germany, Japan, Norway, Sweden, the Netherlands, and a few others. Much of the funding for these projects comes in form of loans to developed countries for the implementation of the projects (Ahsan & Gunawan, 2010), which have to be paid back within an extended time period. In some cases, grants are issued which do not have to be paid back. The five major multilateral development banks involved in issuing loans to developing countries include the World Bank Group, the Inter-American Development Bank, the Asian Development Bank, the African Development Bank and the European Bank for Reconstruction and Development. Any international development project consists of three sets of stakeholders: the funding agency, the host country (the benefactors), and the project implementers/contractors. The implementers are often foreign companies from the developed world. It is often the case that international development projects fail to meet their objectives because of several reasons, for example, a lack of consideration for local cultural contexts during project implementation (Clements, 1993; Diallo & Thuillier, 2004; Muriithi & Crawford, 2003), and a host of other factors (see Gow 1988). It is quite clear from Table 5-2 that the introduction of GIS in Uganda was project-driven and supported by donor funding. It was the expectation of many of these projects that GIS technology would somehow be self-sustaining and diffuse to other public sector institutions over time. However, a qualitative analysis of the interview data reveals that 187 GIS technology in Uganda’s public sector is still very dependent on foreign aid, and is still project-driven and project-dependent. One of the interview respondents stated the following on this subject: “The type of GIS software we use depends on the type of resources we have, but we are basically using ILWIS, ArcView 3.2, 3.1, and 3.3. We have not yet got into ArcMap because you need resources to access these recent versions of ESRI. In terms of resources, you need a lot of money to buy the license. And then after you have bought the license, you need maintenance – every year you have to pay some money. Am not aware of the exact figures, but am told by those who have it that it is a ‘biggy’. Actually some of these colleagues of mine are running away from it because once the project funding ends, you cannot sustain it anymore” (M. Isabirye, NARO, personal communication, July 28th, 2010). One of the most surprising findings is that the National Planning Authority of Uganda does not employ GIS in its workflow, although there is strong intention to establish a national scale enterprise GIS in the near future. One of the interview respondents at the National Planning Authority, Ministry of Finance, explained: “Well, right now, we are in the process of conceptualizing the use of GIS in the planning process of this country. I trust that you have a lot of literature on this that GIS is one of the best tools that can assist in planning – all sorts of planning – they really use a lot of GIS. So we have not been having any GIS as the National Planning Authority. But we realize that if we want really to correlate several things, the simplest thing that every person can understand is where we actually map out something using GIS. So we have realized that right now there are 188 several islanded GIS systems in the country. There are people who have a GIS, for example, National Roads Authority has a GIS, Uganda Wildlife Authority has a GIS, Electric Transmission has a GIS, so we have realized that they are islanded, but now for us who are in charge of National Planning, we want something which is holistic. We want a bird’s eye view of all these things, how one impacts the other” (P. Wakabi, NPA, personal communication, August 9 th , 2010). Table 5-2: Project-driven GIS at public sector institutions ID Year GIS Introduc ed Project- driven GIS Donor Agency Project Name Project Start/End Foreign Contractor/ Facilitator 1 1989 1 UNEP- GRID, World Bank (AFTEN) National Environment Information Center (NEIC) 1987-1989 Norwegian Forestry Society 2 1989 1 Norwegian Government (NORAD) National Biomass Study Phase I 1989 - 1992 3 1990 1 Dutch Government National Wetlands project 1986 4 1995 1 World Bank Greater Kampala Mapping Project 1993 German Company 5 1995 1 French Government CAMPUS Project Phase I 1995-1997 French Company 6 1996 7 1996 1 World Bank, NORAD, African Developmen t Bank Sustainable Management of Mineral Resources Project 2003-2011 8 1999 1 European Union 1995 9 2000 1 GTZ, UNDP, World Bank, 189 Italian Government 10 2001 11 2002 1 World Bank, European Union, African Dev Bank 12 2003 1 IAEA 2003 13 2003 1 World Bank Nakivubo Channel Project 2002 14 2009 15 16 17 1 UNDP and other donors approached in 2010 Future NSDI (not in existence yet) 18 1 Belgian government Hiring of 2000 ICT graduates by CID 2011 19 1998 1 German Government 1998 5.3.1.2 GIS Software and Hardware Usage Of the 21 public sector institutions interviewed, 18 use GIS in their workflows. The most pervasive GIS software used is ArcGIS 9.x (57%) and ArcView 3.x (52%), both from Environmental Systems Research Institute (ESRI). Other software used includes Idrisi from Clark Labs (33%), Erdas Imagine (19%), and Google Maps and Google Earth (38%). The reason for the widespread use of ESRI’s ArcView and ArcGIS software is two-fold. One, it has to do with the project-driven nature of GIS in Uganda. In the implementation of projects, foreign contractors choose the software and hardware to be used, in essence acting as change agents. In-house training in software usage is usually 190 provided to the local staff members in the host country by the contractors, and a preference for the initially introduced software application develops among the staff. Two, the user-friendliness of ESRI GIS software is probably the reason that the contractors, or change agents, selected this software in the first place. This is manifested in many of the interviews where respondents attribute their preference for ESRI software over others, for example, MapInfo. One respondent explained, “We use ArcGIS in particular because of the type of analysis or work that we do. Originally we were using MapInfo, but then we found that when we wanted to do some complex analysis, we couldn’t handle it with MapInfo” (A. Alaba, Department of Geological Survey and Mines, personal communication, August 4 th , 2010). It is surprising to note that many public sector institutions intensively use ArcView 3.x in their daily GIS workflows, despite the fact that they have access to ArcGIS. In fact, many interview respondents expressed their preference for ArcView 3.x over ArcGIS 9.x. One respondent stated, “ArcGIS is good, but it needs licenses, and I hear you have to keep paying annually, which is terrible. That has pushed off many people, and it’s not that we are not using it. We are using it. Yes, we have it. The only thing is that we don’t use it much like we do ArcView. I go in to execute some modules, I go in when I want to do something, and then I jump out and come back to ArcView 3.2a. For example, this one (ArcView) cannot do some things, if you are handling images – ArcView is not good, whereas ArcMap ArcGIS can handle that” (J. Kitatka, Department of Surveys and Mapping, personal communication, July 16 th , 2010). 191 It is no surprise that the cost of licenses for GIS software is a major obstacle to the diffusion of GIS in Uganda’s public sector. To understand this issue, one first needs to understand the licensing criteria for GIS software employed by ESRI. Public sector institutions are charged the full possible licensing fee as private sector institutions. The first type of license is the unlimited “floating” license for ArcGIS with full functionality, the so-called ArcInfo license level. For this type of license, an institution is charged a one-time fee of $40,000. This fee includes free support services for one year from the local ESRI partner in the region, in the case of Uganda, GeoInformation Communication and ESRI Eastern Africa. This price does not, however, include updating of the software in the event of a new release, and also does not include training of staff members. However, the most important fact about the unlimited floating licenses is that an unlimited number of computers at the institution can use ArcGIS software at any one given time. It would seem at first glance that for large public sector institutions, this type of license would be the most ideal since there is no limit on the number of computers licensed to use the software. However, not a single public sector institution interviewed utilizes this type of license. Most of these institutions prefer the alternative single floating license option, at one of three licensing levels, ArcView (least functionality), ArcEditor (medium level functionality), or ArcInfo (full functionality). At one of the institutions interviewed, each license of ArcGIS cost $1340, and an additional $500 for each extension, for example, the Spatial Analyst. In the case of this particular institution, a total of 5 floating licenses and 5 extensions were purchased at a total cost of $9,200. 192 It is because of the licensing costs that many institutions rely on project funding to obtain GIS licenses, and once the project ends, are unable to renew their annual subscription for support or an upgrade. It is often the case that institutions use outdated versions of ArcGIS, for example, of the 12 institutions using ArcGIS software, 7 institutions use ArcGIS 9.2 or earlier versions with no plan to upgrade to ArcGIS 9.3 or higher. The high licensing costs also explains the fact that many institutions prefer to employ ArcView 3.x GIS software. This is because ArcView 3.x requires no license. For marketing reasons, ESRI strategically made ArcView 3.x freely available to the world in the late 90s with no licensing restrictions which allowed for the quick diffusion and adoption of ESRI software on a global scale. In Uganda, this historical fact has led to the pervasiveness of ArcView 3.x GIS software today in the country’s public sector. As one of the interview respondents explained, ArcView can handle most GIS workflows, save for the more complex ones where ArcGIS is needed, for example, when handling imagery. Since many institutions’ GIS workflow requirements are basic, ArcView 3.x does the job and, thus, there is no need to incur licensing costs in acquiring ArcGIS 9.x. 193 Table 5-3: GIS usage at public sector institutions interviewed ID GIS Used ArcV iew 3.X ArcGIS 9.X Map Info Idrisi Erdas ILW IS Ge oV IS Google Maps/ Earth Aut oC AD 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 3 1 1 1 1 1 4 1 1 1 1 5 1 1 6 1 1 1 7 1 1 1 1 1 1 8 1 1 1 1 1 1 9 1 1 1 1 1 10 1 1 1 1 1 11 1 1 12 1 1 1 1 1 13 1 1 1 1 1 14 1 15 16 1 1 17 18 19 1 1 1 20 1 ? ? ? ? ? ? ? ? ? 21 1 ? ? ? ? ? ? ? ? ? Total No. 18 11 12 2 7 4 3 2 8 3 Total %age 86 52 57 10 33 19 14 10 38 14 As far as GIS hardware is concerned, almost all of the institutions run GIS software on desktop PCs and in some instances on personal laptop computers. At least one plotter is available for printing large maps at each institution. The Department of Surveys and Mapping, however, is unable to print maps on a large scale because of a broken-down printing press that would have facilitated the mass production of maps to be supplied to 194 local planners and administrators in various towns in the country. “So we had again down there our printing press, though it has collapsed – the machine which was there – because of spare parts. It became expensive to maintain it. But we have been planning to replace it, but the funding is a problem” (Y. Okia, Department of Surveys and Mapping, personal communication, July 17 th , 2010). In all institutions interviewed, there is no dedicated Information Technology (IT) department in charge of maintaining the computer hardware and software – in fact, there is simply no system in place that ensures the regular maintenance of GIS software, and hardware. Such tasks are left to the GIS analyst, and are not included as part of the institution’s fiscal budget. 5.3.1.3 Applications of GIS Many of the applications of GIS are in environmental degradation monitoring, agriculture, topographic mapping, and basic planning activities. The complexity of GIS analyses performed at the various institutions varies, however. In most cases, very basic GIS analysis is performed, the most common workflow involving the visualizing of spatial data and creation of thematic maps showing features of interest with appropriate symbology. At some institutions, however, complex GIS workflows are employed involving sophisticated GIS processing and operations. A few examples of GIS workflows at various institutions are provided next. The National Forestry Authority (NFA) uses GIS to carry out land cover mapping of Uganda. It uses SPOT and Landsat imagery to perform an unsupervised classification, from which land cover classes are obtained and verified by ground truthing. Because the mandate of this institution concerns the inventorying of forests in the country, the first 195 project at NFA was to determine the total amount of woody biomass stored in the forests in Uganda (Drichi, 2002; Turyareeba & Drichi, 2001). To be able to do this, field inventory plots were used to obtain estimates of woody biomass per square areal unit, from which estimates were extrapolated for the rest of the country based on the land cover classification. The National Environmental Management Authority (NEMA) is the environment watchdog for Uganda. NEMA morphed out of the National Environmental Information Center (NEIC), the result of a project in 1989 to create an Environmental Information System in Uganda, under the Ministry of Environmental Protection, with donor funding from the World Bank (AFTEN) and UNEP-GRID. Today, NEMA’s main role is to manage environmental information in Uganda by collecting data from various primary sources, such as UBOS, Department of Surveys and Mapping, NFA, Makerere University, and so on. NEMA repackages environmental data, performs sophisticated GIS analysis that incorporates remote sensing techniques to produce national state of the environment reports, and national environment atlases for Uganda. Issues addressed in the reports and atlases produced include encroachment on wetlands, pollution, and deforestation. It has also assumed a leading role in organizing all producers of environmental data in the country under a network called the Environmental Information Network (EIN) (Gowa, 2009). NEMA also works closely with UNEP-GRID on issues concerning environmental degradation (Kitutu, 2003; NEMA, 2009; UNEP, 2008; UNEP-DEWA, 2007). 196 The mandate of the Department of Surveys and Mapping is for the establishment of survey and geodetic controls, quality checks of cadastral surveying jobs, survey of government land and international boundaries, production and printing of topographical maps, and for the production of a National Atlas (MoLHUD, 2012). This department’s role in Uganda is similar to the role played by the USGS in the United States. The cartography section within this department uses GIS mainly for creating cartographic products resulting from topographical and cadastral survey data. Before the introduction of computer aided mapping and GIS at the Department in the mid-90s, topographic and cadastral maps were produced using traditional paper and ink cartography. In 1993, a project implemented by a German company, funded by the World Bank, the Mapping of Greater Kampala project introduced computers and digital mapping to the Department of Surveys and Mapping. The objective of this project was to produce large scale maps for the Kampala City Council. GIS was employed for the first time to digitize existing topographic paper maps at large data scales – 1:2500. The cartography section in this department concentrates on producing topographic and cadastral survey maps. Another quasi-section, a spin-off of a project in 1995 called CAMPUS (Computer Aided Mapping Project Uganda Surveys) project, housed within the same Department, concentrates on producing thematic maps of socio-economic variables related to health, education, and urban planning based on features collected using handheld GPS receivers. Maps and digital data are available for purchase in both digital and paper format. The Department of Wetlands Management under the Ministry of Water and Environment is charged with the management of seasonal and semi-permanent wetlands in Uganda 197 (MWE, 2012). This department was formed as a result of a project called the National Wetlands Program, funded by the Dutch government in1986 under the Ministry of Environment Protection. Under this program, a wetlands inspection division was created, which eventually morphed into a larger unit in 2009, a department. In the early days of the National Wetlands Program project, this institution used Idrisi to perform land cover classification with an objective to identify seasonal and permanent wetlands in the country – Idrisi was the first GIS software used as early as 1990. Ground truthing exercises are also carried out to ascertain the permanency of features identified as wetlands through the classification exercise. Attribute data about the wetlands was stored separately in Microsoft Access databases because raster GIS was unable to link attribute data to geographical features at that time. This workflow resulted in very large Access databases of environmental data that was disconnected from spatial data. In the late 90s, ArcView 3.x was introduced, however, there was still no way of linking Access to ArcView. It was not until 2008 when the institution acquired ArcGIS and a skilled GIS analyst that it was able to convert its Access data to ArcGIS ArcSDE format and link it to spatial features in vector format (P. Omute, Department of Wetlands Management, personal Communication, July 16 th , 2010). National Water and Sewerage Corporation (NWSC) is a government owned utility services parastatal organization under the Ministry of Water and Environment. Its mandate is “to operate and provide water and sewerage services in areas entrusted to it, on a sound, commercial and viable basis” (NWSC, 2012). The main GIS task performed at this institution is mapping of water meters, water lines, and sewer lines. In addition, 198 footprints of buildings and other structures are also mapped relative to the water infrastructure. All this data allows for a decision support system to be realized, and facilitates planning and generation of water usage, consumer consumption, and billing reports. Prior to 1998, this was done using traditional surveying and pen and paper cartography. In 1998, a German company introduced ArcView GIS to NWSC, and for the first time, digitizing of old paper maps was performed to convert all data from analog to digital form. In 2007, ArcGIS was introduced and most recently, ArcGIS Server, and ArcPad as well. AutoCAD, however, is the main work horse of the organization for cartographic editing mainly because of licensing issues with ArcGIS Server. NWSC currently has an Arc Editor license, which puts limits on the size of data that can be handled by the ArcSDE geodatabase. This license cost $12,000 and included the 3D Analyst and Spatial Analyst extension, but soon turned out to be insufficient for the large data sizes handled by the organization. NWSC plans to upgrade their license to ArcGIS Server Enterprise, which has no limits on the geodatabase size, however, this will cost a total of $40,000 (G. Akon, NWSC, personal communication, July 12 th , 2010). This organization has benefited from the use of GIS mostly in cutting loses due to the effective management of billing of customers which has greatly improved revenue collection, and thus, purchasing an enterprise level license is not necessarily an obstacle, unlike for other public sector organizations. Apart from GIS, another reason for the financial success of NWSC is the creation of, and partnership with, a private company, Kampala Water Ltd, owned and operated by staff of NWSC. This public-private partnership has boosted 199 performance, and promises to be a successful future model for driving performance in the public sector in Uganda. Uganda Bureau of Statistics is the principal agency for the collection, storage and dissemination of statistical data in Uganda (UBOS, 2012). It is a semi-autonomous agency under the Ministry of Finance. The activities at UBOS are geared towards serving the needs of the parent Ministry – the Ministry of Finance, which is their biggest client for statistical data services. UBOS is responsible for collecting socio-economic, macro- economic, labor, price, and house-hold survey, and Census statistics for Uganda which are collected at varying time intervals (weekly, monthly, annual, decadal) and also vary by unit of measurement (household, organization, district). All government endorsed research must be based on official statistics from UBOS to achieve favorable recognition, otherwise, any research based on personal data would be considered as “your own opinion or hearsay” by the government of Uganda (B. Muhwezi, UBOS, personal communication, July 19 th , 2010). GIS was introduced at UBOS in 2001, and an integration with GPS for data collection meant that for the first time, a geospatial component of archived statistical data was realized. Despite the availability of this data, UBOS use of GIS is mainly for the provision of spatial data to other public sector, and private sector agencies in raw form so as to allow them to add value to it. UBOS does not necessarily perform highly sophisticated GIS analysis on its data. And even then, the capacity to handle the large archival data sets is limited in terms of skilled GIS personnel and funding from the government. It is because of this limitation that UBOS has chosen to digitize only a subset of their datasets. Access to these datasets is provided through a 200 geoportal called Uganda Clusters, an initiative created a working group called the GeoInformation Working Group for the purpose of providing Web access to geospatial data in Uganda to local consumers of data, especially non-governmental organizations working on rebuilding northern Uganda after an extended period of civil strife caused by the 20-year long Joseph Kony insurgence (UgandaClusters 2012; B. Muhwezi, UBOS, personal communication, July 19 th , 2010). 5.3.1.4 GIS Data Collection, Data Sources and Data Sharing Data collection for GIS is either collected directly by each institution (primary data sources), or indirectly obtained from other institutions (secondary data sources). Only a handful of institutions collect primary geospatial data. The main methods employed in primary data collection (see Table 5-4) involve the use of handheld GPS receivers (83%), topographic paper maps (89%), satellite imagery (56%), land surveying techniques (22%), and aerial photographs (11%). These percentages are calculated based on the total 18 institutions that actually employ GIS in their workflows. LiDAR techniques have not been applied by any of the institutions interviewed in this research exercise, though there is a strong intention to move in this direction for certain types of mapping, for example, the NFA is interested in using LiDAR in the future for forest mapping to obtain a more accurate estimate of woody biomass (J. Mutyaba, personal communication, July 19 th , 2010). 201 Table 5-4: GIS Data Collection Methods Used at Public Sector Institutions ID Handheld GPS Land Surveying Topographic Paper Maps Satellite Imagery Aerial Photographs 1 1 1 1 2 1 1 1 1 3 1 1 1 4 1 1 1 1 1 5 1 1 1 6 1 1 1 7 1 1 8 1 1 1 1 9 1 1 1 10 1 1 11 1 1 12 1 1 1 13 1 1 1 1 14 15 16 1 1 17 1 18 19 1 1 1 20 21 Total 15 4 16 10 2 Total %age 83 22 89 56 11 In Uganda, the institutional structure in the public sector is such that each Department within a Ministry has a specific mandate to collect a given class of spatial data, for example, forests and woodlands data is NFA’s mandate, topographic and political boundary data is the Department of Surveys and Mapping’s mandate, wetlands data is the Department of Wetlands Management’s mandate, statistical data is UBOS’s mandate, and geological data is the Department of Geological Survey and Mine’s mandate. Due to a 202 lack of a clearly defined geoinformation policy (Kalande & Ondulo, 2006) in Uganda, these mandates have resulted in a culture of assumed “ownership” of spatial data by each of the data producing institutions, which has in turn led to each institution’s lack of willingness to share “their” data in a transparent manner. This situation is problematic at many levels, including the fact that the data is not private, but public data. Nonetheless, data producing institutions insist that their data must be purchased at arbitrary prices. There are no clearly formulated price structures governing the purchase of this public data in many of the public sector institutions. One of my interview respondents explained, “… the coming of this digital era is a problem, but traditionally, with the analog maps, the cost was Uganda Shillings 10,000 ($4) per map for a Scale of 1:50,000. In fact, we would charge the same if you wanted a 1:1,000,000 scale map, or so. Even the larger scale, the 1:2500, it cost the same. Now, when GIS came, the computer technology came, now you can also access data in the soft form. That is where we have failed to come up with a realistic cost” (Y. Okia, Department of Surveys and Mapping, personal communication, July 17 th , 2010). The effects of this unclear public data pricing problem are twofold. One, institutions that are unwilling to purchase public spatial data from data producing institutions have decided to collect their own data, in effect, duplicating efforts in terms of collecting already existing digital data. One of my interview respondents asserted, “And don’t forget that in Uganda here, it is not easy to get information from any other offices. People keep their data, they don’t want to share. We are still actually behind. We may be having very many organizations, we may be having the data, but they cannot release that data to 203 other organizations to use it. Lands and Survey (Department of Surveys and Mapping) – much as it is an organization that has the mandate and authority for producing maps in Uganda, geographic maps, and for producing everything; then other organizations get from them, but right now … ok, I understand their budget does not allow them. So there is no need for, maybe Northern Uganda Data Center to sit back, when they want information to use, and they just wait for the other guys who will not be able to collect for them what they want, so if they have some money, it is better that they just go and collect it and they come and use it. The problem we have in Uganda is sharing: lack of sharing, lack of access, and lack of coordination” (B. Nyemera, Northern Uganda Data Center, personal communication, July 20 th , 2010). This finding is consistent with previous studies, for example, see Nyemera 2008; De Vries & Lance 2011; Makumbi 2010. Two, a culture of not wanting to share public spatial data has become inculcated into the public sector institutions. This situation is further exacerbated by a lack of geoinformation policy enforced by law. It is no surprise, thus, that Uganda’s decade-long efforts to create a National Spatial Data Infrastructure have so far failed miserably because of the institutional roadblocks to data sharing woven into the original mandates awarded to different institutions regarding data ownership. This finding is consistent with previous work in this area, for example, Goa (2009) explained that the early efforts of National Environmental Information Network in the mid-90s to act as a quasi-repository for environmental geospatial data in Uganda met a lot of resistance from the Department of Statistics, the Department of Surveys and Mapping, and other data generating 204 institutions because each claimed a sole mandate to generate and “own” their respective datasets. Admission of these major institutions into the Environment Information Network temporarily diffused the tension, but future efforts to create a central data repository still failed due to these institutional challenges of territorial ownership. An initiative that had a lot of promise to move Uganda towards an NSDI was the National Integrated Monitoring and Evaluation (NIMES) program (GoU, 2006; NIMES, 2004; Sentongo, 2003), a strategy framework for improving performance monitoring of the Poverty Eradication Action Plan (PEAP) initiative created by the government of Uganda to meet goals of poverty eradication in accordance with the United Nations Millennium development goals (Lwasa, 2006; Matheson, Shall, & Ogeda, 2008; OPM, 2008). However, there is skepticism about the success of NIMES, especially in as far its role in promoting GIS and the creation of NSDI in Uganda. With reference to NIMES and geoinformation policy in Uganda, an interview respondent expressed his frustration and disappointment, “It’s not there. We have been described as a society that is very good at coming up with very well thought-through policies, but also as a society that is really bad, really way back in terms of implementing policy. In respect to geospatial information, we don’t have a policy. We have a loose network which we started in 2006 of a national spatial data infrastructure – some GIS professionals, and a community of practitioners of GIS. It is still loose. We try to drive the agenda on policy and mainstreaming in the government and then we met a stumbling block. The stumbling block is that one of the organizations in the Ministry of Finance was establishing a similar initiative. And then we said, well if we have this, then you guys you are doing 205 nothing. You should not tell us what to do, we already have this. If you go to the Ministry of Finance, no not Ministry of Finance, but the Office of the Prime Minister and asked for the secretariat for NIMES, which is the National Integrated Monitoring and Evaluation Strategy, it even has a website. Check out how functional the GIS component of NIMES is. If it is functional, then I’m wrong to give my comment, and I will apologize and withdraw my comment. If it is not that functional to the expectation, then there was a clear misunderstanding of Spatial Data Infrastructure (SDI). But am convinced it is not functional, why? This is because a parallel initiative has started and USAID, the Embassy of the US, PEPFAR, UNICEF, WFP have started an initiative and discussion that we started in 2006. But that initiative is only serving humanitarian organizations and NGOs (Uganda Clusters). It is not serving the wider community. We understand the importance of humanitarian aid, NGOs, and development assistance, but they are not the only users of geospatial information products. Researchers are users of geoinformation products. The administrators at districts are users of geoinformation products.” (S. Lwasa, Department of Geography, Makerere University, personal communication, July 2 nd , 2010) Similar problems involving institutional challenges are cited in the limited success of the Local Government Information Communication System (LoGICS) program (Lwasa, 2006; Lwasa et al., 2005; Ofori-amoah, 2004), a GIS capacity building program started in 2002, whose major objective was to build GIS capacity among staff in local government institutions in line with the decentralization policy of the Government of Uganda (Asiimwe & Musisi, 2007; GoU, 1995; Kisembo, 2006). One respondent had the 206 following comment on the minimal success of LoGICS: “Ministry of Local Government went ahead or even started much earlier to introduce what they call LoGICS which is integrative of statistics, management information and GIS. Again, go to Ministry of Local Government, I challenge you; find out how functional LoGICS is. And don’t only talk to local government officials, go to some of the districts they’ll tell you how functional LoGICS is, and how it is helping them – so find out whether LoGICS is functional” (S. Lwasa, Department of Geography, Makerere University, personal communication, July 2 nd , 2010). 5.3.1.5 Training and Education in GIS In the sample of 19 public sector institutions interviewed, apart from the Department of Surveys and Mapping, Cartography section (see Table 5-5), employees that work with GIS technology at most organizations range from 2-7 employees per organization. Relative to employees in other professions, this number is quite small and is reflective of the relatively low importance given to the contribution GIS makes to an institution’s productivity. There is a lack of appreciation for the potential of GIS in Uganda’s public sector institutions. Several interview correspondents attributed this problem to a lack of awareness about the importance of GIS in government, and thus a lack of government funding to support GIS activities, including funding for new GIS personnel hires. “Yes, in every section – the structure does not allow us to hire new employees. You cannot be in a whole project and you are only 3 people! The problem is the structure. You know with the government, some time back they made a structure, where they said in this section, there are so many people, in the other section, so many and so on… Unless, you 207 write to the Ministry of Public Service, convince them why you need those people, then they come back to you and say, ok, you need those people, you first contact Ministry of Finance, and they allow you to recruit? Is there money for them? No. So there is a long bureaucratic process. It is not (as easy as) just employing (new personnel). Is there money for that extra person or for the three or four (hires), is it there?” (M. Murindwa, Department of Surveys and Mapping, personal communication, July 28 th , 2010) From Table 5-5, it is clearly evident that almost all institutions that use GIS technology have some sort of training for their employees in form of on-site training, off-site training, and hands-on “on-the-job” training. The main institutions that provide training services to public sector institutions include GeoInformation Communication (GIC), the official ESRI business partner in Uganda (about 50% of all cases), ESRI Eastern Africa in Nairobi, Regional Center for mapping in Nairobi, Makerere University Institute for Environment and Natural Resources (MUIENR), and International Development project implementers, for example, French, German, and Japanese companies. These all provide in-house (on-site) or off-site training. Another form of training is through local pre-conference GIS workshops organized by GIC in Uganda, and regional workshops organized by ESRI Eastern Africa and the Regional Center for Mapping in Nairobi, and also in Addis Ababa. Before the decline and deterioration of the Department of Geography, Makerere University, training was also offered there in form of short courses to some public sector institutions in the mid-2000s. Yet another form of training is through short courses in GIS offered by universities in Europe, for example, ITC in the Netherlands, and the Royal Museum for Central Africa 208 in Belgium. Further, bilateral partnerships between certain Ugandan public sector institutions and surveying and mapping departments in Europe allowed for the training of employees in European countries, for example, partnerships with surveying departments in the UK, Germany, the Netherlands, and Switzerland. Employees would go to Europe to do short courses in GIS, remote sensing, photogrammetry, and cartography lasting 3-6 months and then return to their parent institutions, not only with newly obtained GIS skills, but also with GIS software. Table 5-5: GIS Employee Training at Public Sector Institutions Id No. of GIS Employ ees Training Type On- site Off- site Hands -on Duration Training in Europe Training Institution s 1 2 2 7 ESRI 1 GIC 3 1 6 months 1 Royal Museum for Central Africa (Belgium) 4 15 ESRI 1 1 1 1 week 1 GIC, Project Implemente rs, e.g. French and German Companies 5 3 ESRI 1 1 1 week 1 GIC; Project Implemente rs, e.g. French company; Regional Center for Mapping, Nairobi 209 6 3 1 1 GIC, MUIENR, Local GIS Workshops 7 5 ESRI 1 1 Regional GIS Workshops , e.g. in Ethiopia; ITC Alumni training 8 4 ESRI 1 9 10 ESRI 1 1 1 11 1 1 12 ESRI 1 week; 3 - 4 months GIC (ESRI) workshops, ITC, Ethiopian University 13 2 ESRI 1 1 GIC 14 15 16 5 ESRI 1 1 1 week GIC, ESRI Eastern Africa 17 18 19 ESRI 1 1 1 week GIC preconferen ce workshops; Department of Geography, Makerere University; ESRI User Conference , USA 210 An investigation of the university level GIS education of interview respondents revealed that most respondents studied for their undergraduate degrees at Makerere University in Uganda and were introduced to GIS while pursuing a diverse variety of majors including land surveying, Geography, Forestry, Geology, Agriculture, Information Technology and Economics. There were several respondents who had also been exposed to GIS at tertiary institutions such as the Survey Training School in Entebbe (MSE, 2009) that offered certificate- and Diploma-level education in Cartography, land surveying and, recently, GIS. After 1996, this institution was restructured and renamed the Institute of Survey and Land Management. It currently offers 1-year post-graduate diplomas in GIS. 211 Table 5-6: GIS Education at University level ID BSc. Degree BSc. Course MSc. Degree MSc. Course 1 Ardhi University, Arusha, Tanzania Land Surveying ITC Netherlands. Photogrammetry 2 Makerere University Forestry Makerere University - MUIENR Environmental Information Management 3 Makerere University Land Surveying Curtin University of Technology, Australia GIScience 4 Makerere University Mathematics and Physics ITC Netherlands Computer Assisted Mapping 5 Nkumba University Information Technology 6 Baraton University, Kenya Zoology 7 Makerere University Geology ITC Netherlands Geoinformation 8 Makerere University Agriculture ITC Netherlands Soil Science and Soil Survey 9 Makerere University Urban Planning (Geography) ITC Netherlands GeoInformation Management 10 11 Makerere University Economics Makerere University Information Systems 12 Makerere University Land Surveying Addis Ababa University of Science; ITC Netherlands GIScience 13 Makerere University Geography Jerusalem University Municipal Service Delivery Systems 14 15 16 17 18 19 Makerere University Economics Lund University, Sweden Online GIS program 212 A number of respondents had obtained their masters level education, as well. A majority of them had obtained their Masters degrees from the International Institute for Geo- Information Science and Earth Observation (ITC), Netherlands. Others had obtained Masters Degrees in related fields, such as Photogrammetry, Environmental Information Management, Information Systems, and Municipal Service Delivery Systems from foreign universities in Australia, Sweden, Ethiopia, and Israel, and also from Makerere University, MUIENR, in Uganda. 5.3.1.6 Government Funding for GIS activities The level of support for GIS activities at public sector institutions by the government of Uganda is minimal. Because of the project-driven nature of GIS at the various institutions, government budgets for public sector institutions have not traditionally included a component to support Information Technology (IT), or GIS in particular. Much as developing countries generally suffer from a lack of sufficient funding, and hence infrastructure, in most sectors of the economy, the GIS and mapping sector definitely stands out as one of the least funded. An examination of funding provided to the Department of Surveying and Mapping, the national mapping agency, by the Government of Uganda sheds some light on this dilemma. According to an interview respondent, since 1986, this Department has received a total of about $20,000 annually for its activities, the most important of which is to keep the national maps up-to-date. Because of this meager budget, this Department has not been able to update the national map, let alone perform any other activities in any meaningful way. In fact, the most recent maps of Uganda by 1995 had been compiled in 213 the 1960s – they were out of date by more than two decades. Updates and recompilations of the national map have so far been performed in an ad-hoc fashion based on available donor funded projects, for example, the mapping and recompilation of the central region of Uganda between the Equator and 1°N. This project lasted from 1993 to 1995, funded by the World Bank, and implemented by a Japanese company. Maps at a scale of 1:50,000 were produced, as a result, and these are arguably the most recent topographic maps of Uganda available today. Map updates for specific districts and regions have also been performed under various government and donor funded projects in piecemeal fashion, for instance, the large scale mapping of Greater Kampala between 1993-1995, funded by the World Bank; the mapping of major urban centers of Uganda in the 90s, funded by the government of Uganda under the National Action Plan (NAP) and recently under the Northern Uganda Social Action Fund (NUSAF); and the mapping of the entire country based on medium and high resolution satellite imagery in 2002 under an East Africa regional project called Africover funded by the Italian government – this project covered Egypt, Sudan, Somalia, Eritrea, Djibouti, Tanzania, Kenya, Rwanda, Burundi, and the DR Congo. In the latter two cases, funds were mismanaged by top government officials in the Department and the objectives related to topographic mapping set out at the start of the projects were never achieved. In the case of the Africover project, about $1 Million was mismanaged within the Department of Surveys and Mapping. One respondent lamented, “Not only is funding for GIS and mapping limited, but even worse, the little that is made available is 214 mismanaged due to corruption and greed. This is an internal problem” (Anonymous, personal communication). Moreover, updating of old base maps has its limits. The most recent base map of the country dates back to the mid-1960s. Today, after forty years, a national map recompilation is necessary due to errors and distortions that have accumulated due to simplistic map updates over the years. Complete base map recompilation would require a descent amount of much needed funding. There is some light at the end of the tunnel. Recently, government funding to the Department of Surveys and Mapping increased to about $120,000 per quarter, which equates to about $0.5 Million per year. With this increased financial backing of the government of Uganda, it is hoped that an up-to-date topographic national base map will soon be realized, at least at a medium data scale of 1:50,000. It is also hoped that a broken-down printing press at the Department will be replaced to allow for the mass printing of paper maps needed by local government planners at the district level. It is only in recent years that the government of Uganda has recognized the importance of surveying, mapping and GIS in the development of the country. The National Development Plan 2010/11-2014/15 (GoU, 2010) specifically expresses the government’s recognition of GIS and mapping activities for the future development of Uganda. The section on Physical Planning states the following: Uganda has 112 gazetted towns of which 93 have up-to-date structure plans. However, a big number of these towns have not translated the bulk of their Structure Plans into implementable detailed physical development plans. In 215 addition, Uganda lacks a coherent rural land use plan. The country, therefore, continues to experience haphazard developments in both urban and rural areas. (GoU 2010, 167) The lack of effective physical planning implementation is due in part to the “lack of up- to-date planning information, including topographic maps, cadaster information and land tenure maps, among others” (GoU 2010, 168). One of the strategies proposed in the National Development Plan to address the problem of poor physical planning in the country is for the central and local governments to “establish a land use (physical planning) database and to computerize physical planning operations” (GoU 2010, 168). The proposed intervention specifically prescribes the introduction of GIS technology into the institutional structure of central and local government, and for the training of staff in geospatial technology. The exact statement on this issue is as follows: i) Establish and operationalize an appropriate institutional structure at the Local Government level. ii) Train all Central Government physical planning staff in advanced Geographical Information System (GIS) skills. iii) Roll out GIS training to all district and urban Local Governments to impart adequate GIS skills for all relevant Local Government technical staff. iv) Establish a GIS center and a national spatial database to adequately back up land use data repository and work stations for all trained Planners. 216 v) Support relevant security and law and order organs to build spatial databases linked (real time) to national spatial records of street layouts, utility maps, addresses, architectural plans, land use and ownership among others. (GoU 2010, 169) The document also expresses the government’s desire to increase participation of marginalized groups in the physical planning process to “ensure increased implementation and public acceptance” (GoU 2010, 169) of physical plans. 5.3.1.7 The Impact of GIS in the public sector on Society The most significant impact of GIS in the public sector on society is the contribution made by GIS to service delivery. Publicly owned utility parastatals such as National Water and Sewerage Corporation use GIS to manage various aspects of water delivery to communities, including efficient reading of water meters at client premises using Mobile GIS methods, and automated billing. Uganda National Roads Authority (UNRA) uses GIS to keep track of road construction projects around the country, allowing for better management of road maintenance and construction projects. The use of GIS in the daily workflows of public institutions has allowed for better service delivery to society because of increased efficiency in data storage, and better analytical capabilities based on geographical information and GIS technology. GIS in the public sector has allowed for relatively easier access to spatial data. NGOs and private sector companies today can obtain forestry data from the National Forestry Authority, statistical and political boundary data from UBOS, Wetlands data from the Department of Wetlands, and topographic data from the Department of Surveys and 217 Mapping. Prior to the introduction of GIS, such data was hardly accessible as it existed mostly in paper map format at the various institutions mandated with the collection of the data. The use of GIS in natural resource management is one of the most important impacts of GIS on society. The National Biomass Study was the first long term study of the total woody biomass in the country from which an estimate could be made as the level of deforestation the country had experienced over the decades allowing for appropriate interventions to be made to protect forests in the future. The national wetlands study established the total acreage of wetlands in the country, and estimated the rate of wetlands conversion for other land uses. Both these studies used GIS and GPS technologies to quickly capture and analyze geospatial data. The Uganda Wildlife Authority uses GIS today to monitor the movements of wild animals, as well as illegal activities, such as the poaching of wild game in the various national game parks in the country. Tourism being one of the major earners of foreign exchange benefits immensely from the use of GIS. Further, NEMA uses GIS actively to monitor society’s involvement in environmental degradation resulting from human impacts. During the long insurgency in northern Uganda due to the army of the infamous anti- government rebel leader, Joseph Kony, GIS played a significant role in helping villagers fleeing capture by use of maps to avoid potentially dangerous areas where the rebel forces most probably operated. Villagers were handed maps showing such areas by local leaders with assistance from International Organizations such as the United Nations Office for the Coordination of Humanitarian Affairs (UNOCHA) and local NGOs 218 operating in the region. Following the end of the insurgency in 2006, efforts to rebuild northern Uganda were guided by GIS data capture, storage and analysis. The Northern Uganda Data Center under the office of the Prime Minister uses GIS to map the locations of various service delivery entities such as schools and health centers in northern Uganda to help NGOs better plan their development initiatives in the troubled region. 5.3.1.8 Challenges to the Diffusion of GIS in the Public Sector There are myriad challenges to the diffusion of GIS in Uganda’s public sector. Being a developing country, many of these challenges are based on economic factors linked to insufficient funding to support GIS activities. However, there are also political, social, cultural and technological factors that have impacted the diffusion of GIS in Uganda. I will discuss these in the following order: economic, political, social and technological factors. Economic factors The project-driven nature of GIS at many public sector institutions has resulted in many institutions relying to donor funding to support GIS activities. The end of a donor funded project deals a hard blow to GIS activities at many institutions – GIS licenses expire, or can’t be renewed to allow for software updates, maintenance of computer hardware and peripheral devices, for example, plotters cannot be done, to mention but a few problems. Project-driven GIS has also meant that cheaper GIS options like open source GIS have not had a chance to diffuse into Uganda because of preferences of the foreign implementers. This is a fundamental problem within the approach to technology transfer, and indeed such an approach begs the question is commercial GIS appropriate 219 technology? Commercial GIS licenses are too expensive for public sector institutions to purchase without donor support, which makes a case for an alternative, namely, freely available open source software. The main issue with the latter is the lack of user- friendliness and technical support, and this could explain the lack of adoption and diffusion of open source technology in the developing world, similar to the slow adoption of the command line based PC ARC/INFO commercial software. For many years, the government of Uganda has not provided sufficient funding to support GIS and mapping activities at their public sector institutions. It is only in 2010 that GIS has officially been recognized as a necessary part of physical planning workflows at various levels of local government (GoU, 2010). As a result of a lack of sufficient funding, many institutions have been unable to renew GIS licenses, and purchase necessary hardware. The Department of Surveys and Mapping, cartography section, for example, has a limited number of computers, five to be exact, to carry out the mapping of the entire country. In fact, most of the cartographic and cadastral mapping is still done using old-fashioned paper, pen and ink cartographic techniques. Save for a few land parcels belonging to large private companies and corporations, the majority of the country’s cadaster is stored in box files in form of hand-drawn paper prints of land parcels. This is also the case at Kampala City Council (KCC), where the cartography section still employs analog cartographic techniques to carry out its mapping. The CAMPUS project at the Department of Surveys and Mapping has only 8 computers and a plotter for thematic mapping of the entire country. In terms of software, most institutions rely on the free 220 version of ArcView 3.x to perform their day-to-day GIS tasks. Not that there is anything wrong with using ArcView, but the full analytical capabilities of ArcGIS would probably be a better option for more powerful GIS analysis than ArcView, not to mention the inconvenience associated with data size limitations inherent in ArcView GIS. The Department of Geological Survey and Mines is also severely handicapped by a lack of equipment, for example, equipment for geophysical and seismic surveys, mining cadastre surveys. A respondent at this Department complained, “We have some people who are employed here who are surveyors by profession, but the instruments were not there. So, for monitoring purposes, we are now acquiring the surveying instruments, but for now, we depend on the topographic maps” (A. Alaba, Department of Geological Survey and Mines, personal communication, August 4 th , 2010). GIS, just like other forms of information technology, is a dynamic field with software versions changing every two years on average. There is a need for scheduled training in new analytical functionality and GIS operations. However, many public sector institutions do not have a clearly defined program or budget for training of local staff. This causes inefficiency in GIS workflows, and prevents newer staff members from picking up GIS skills to inform their work. Currently, most of the training is carried out by GIC; however, there is a need for more private and academic sector institutions to provide sufficient and timely training to public sector institutions. Most of the training provided by GIC is introductory level GIS training, and hardly any advanced level training for more experienced GIS staff is available. To acquire such training, staff members would have to incur a higher cost because they would seek it from foreign 221 academic institutions and universities, for example, through ESRI Eastern Africa in Nairobi, ITC in the Netherlands, the Regional Center for Mapping in Nairobi, and through regional GIS pre-conference and workshop-based training. In addition to these complexities, there a shortage of adequately trained and skilled GIS personnel for hire in many public sector institutions. For a technology to diffuse successfully there must be an economic model that includes support from the private sector. In the case of Uganda, there is a lack of such support for the GIS industry. Public sector institutions have failed to package their services in such a way that private sector companies can take advantage of geospatial data and services available through public sector agencies. The only source of income from GIS activities is through the sale of geospatial data, and even that is not well-defined in terms of pricing and access to the private sector. And sales of GIS data have gone down recently, as one respondent at NFA noted, due to the fact that there are no copyright policies in place that prevent copying of the same data by several clients, eliminating the need to buy data from the data producer. In some limited cases, UBOS and the Department of Surveys and Mapping, the NFA provide special consulting services to clients with specific needs, but these services are provided in ad-hoc manner with no real business model in place. There is a need to define a working economic model that markets and advertises GIS services and geospatial data so as to make GIS a profitable industry. Due to a lack of sufficient government funding for GIS and mapping activities, the topographic and thematic maps available from the Department of Surveys and Mapping, and the Department of Geological Survey and Mines is outdated. In both cases, the most 222 current map updates date back to the 60s. These outdated maps are a source of frustration to other public sector agencies that rely on these maps to be able to add other thematic detail to the maps based on their mandates, for example, NFA, NUDC, and the Department of Wetlands Management. This leads to data duplication as these institutions decide to take matters into their own hands and use any available funds to recreate spatial data sets. Political Factors With regard to the sharing of geospatial data, there is no specific data sharing policy in place. This is a source of frustration to both data producers and consumers. The former feel that their efforts to collect and store geospatial data are not being rewarded, and thus resort to imposing stringent and ad-hoc policies for the sale of public data, while the latter feel that they have a right to obtain public data free of charge. An interesting case of this scenario involves two public sector institutions, Kampala City Council (KCC) and the Department of Surveys and Mapping. An interview respondent at KCC complained, “That is the biggest problem we have in this country. We wouldn’t have started this GIS unit if we had opportunity to access that information. And I would tell you that it is not only Kampala City Council, but if you go to UBOS, they have also a fully-fledged GIS Unit. They are doing their own stuff. If you go to NFA, they have a fully-fledged GIS unit; they are doing their own stuff, NW&SC, UEDCL, and so on. Why, because we fail to get service from the government institution, which is this Surveys and Mapping Department. First of all, accessing information is quite difficult, and then secondly, it requires some money, and getting current information was not easily available, so you 223 would want to get the latest topographic information for a given area, and they cannot produce it within time. So I think many institutions have decided to go their own ways and produce their own maps and products, so that they can do their work faster” (J. Ssemaambo, KCC, personal communication, August 19 th , 2010). Excessive bureaucracy or red tape in the way government carries out its business was cited by some interview respondents as a major problem affecting the diffusion of GIS in the public sector. There are two effects of this. One, the official procurement process for GIS hardware and software is too lengthy, complicated, and sometimes flawed. After a lengthy bidding process, top management in government departments sometimes decide to issue supply contracts to quack companies. In severe cases, pirated GIS software was supplied to a public institution at a cost of $40,000 yet legitimate software could have been obtained from the official ESRI partner in Uganda at the same cost. Because possible kickbacks and bribes, the contract was awarded to the wrong supplier. Two, there is excessive bureaucracy in hiring new employees at public sector institutions, causing many sections within Departments to be understaffed. An interview respondent complained, “The major problem facing this Department is a typical problem in government institutions in Uganda - bureaucracy in everything. There are long bureaucratic processes in procuring equipment, and in hiring new employees. The CAMPUS project and the Department of Surveys and Mapping, in general, are acutely understaffed. For the whole CAMPUS project, there are only 3 employees that do GIS. This is not enough. The few others are simply interns who are unpaid. And there is no way of paying them, as this money is not accounted for in the Departmental budget. If 224 this Department wanted to hire a new employee, they would need to apply to the Ministry of Public service explaining why they needed the new employee. Then the Ministry of Public Service would communicate back to our Department asking if there are funds available for the new hire. Our Department would in turn then contact the Ministry of Finance, and in most cases, the Ministry will simply reply that there is no money for a new hire. The current staffing structure is outdated, since it was done several decades ago, when there was not much (work) at the Department, yet now, work has grown in volume, but staffing has remained constant. And each of the sections under this Department is handicapped when it comes to hiring new employees. For example, there are an overwhelming number of files that the Surveying section has to handle in terms of private cadastral surveys that need approval for land title deeds acquisition.” There is an institutional problem in the collection of spatial data at public sector institutions with regard to the centralization of data collection at the top level of government, namely the departments under the Ministries. This put enormous pressure on a few employees to perform data collection tasks for the entire country. Not only is this cumbersome, but also slow and inefficient. There is a need to decentralize the collection of spatial data to lower levels of local government so as to streamline data collection. A respondent at the Department of Wetlands Management advised, “And we were looking to even try to decentralize the data updates, so that it comes from each district. In other words, train district environmental officers to collect this data and be able to update them locally, the other side, while centrally updating the main central database. Previously, during the project people would go from the Departmental headquarters in Kampala, to 225 the field, and collect the data, but with this idea of decentralization, the government is saying, why don’t you involve the district officers? They give you the data, and then your work is just to manage it, instead of you saying, ok, this is a project, now we are going to recruit more people to go and collect that data, yet there are people lying in the district (who can collect the data on your behalf). And you know, these (district) officers know their localities very well” (P. Omute, Department of Wetlands Management, July 16th, 2010). Social Factors A general problem affecting many public sector institutions in Uganda is the low salaries and wages of civil servants. An average civil servant working in a Ministry in Uganda earns about $200 per month. The motivation and morale of employees in the public sector is very low. Under such circumstances, the diffusion of GIS technology is hampered because employees might not necessarily feel the need to learn new skills, or go the extra mile to be innovative and perform additional GIS tasks, teach others in the department how to use GIS software, or even explain the results of their GIS analysis to colleagues. In fact, there is a problem of job security. GIS Analysts, like other professionals, are afraid that if their skills are mastered by other individuals in the Department, they might become dispensable and be replaced. An economist at one of the public sector institutions lamented, “The only problem with GIS here is that these people (the GIS Analysts) have refused to decentralize the knowledge of GIS. It is like a monopoly of theirs, they are very proud people, and so, it is kept with them only. We (economists) have not benefitted 226 from it (GIS technology). We need to get that information and use it for policy analysis and whatever. They can generate the information, but it requires interpretation from us. They don’t involve us. Actually, if we could partner with them, they come up with the information, and then we interpret it, give it an economic touch, put it in the economic perspective, and then see the social economic importance of that. We would do so much.” Related to the problem of job security is the threat of GIS to older professions like paper, pen and ink cartography. This threat is plausible. Using KCC and the Department of Surveys and Mapping as examples, these institutions have historically employed cartographers who specialized in the old fashioned analog methods of drawing maps using paper, pen and ink cartography. Land surveyors provide the angle and distance measurements to the cartographers, and the latter plot the features on paper. With the coming of GIS, the skills of these draughtsmen have become obsolete. The logical thing would be to retrain the cartographers in the new digital cartography, but old habits die hard. And the older generation are not necessarily interested in learning new skills at their age. Thus, there is resistance to the adoption of GIS in these two departments as it would require an institutional overhaul of the departments to phase out one profession, and introduce another. Jobs of several employees could be at stake. A common problem in developing countries is the misappropriation of public funds. Many interview respondents complained about rampant corruption within their institutions at various scales, mainly at top management levels within Ministries and Departments. One respondent explained that the amount of money appropriated in the annual fiscal budget by the Government is never delivered to the receiving Departments 227 as promised, “Our annual government funding is approximately 200 million shillings (about $100,000). It is not consistent– it keeps changing from year to year, but it has never exceeded 250 million shillings. Even if they give you (assign) 200 million, at the end you will find they have given you only 100 million shillings (about $50,000). They don’t give you what they have said (promised). They give you money according to what is in the treasury. So they don’t give you all the money promised under the budget. Every quarter, they give you a fraction. By the end of the year, the total cannot come to that original figure promised in the budget.” Due to low salaries, uncontrolled corruption, and impunity for perpetrators of corrupt practices, many GIS professionals in Uganda have sought greener pastures in other countries. A number of students who travel to Europe and the United States for further studies prefer to seek jobs in developed countries to avoid disappointing work ethics and meager salaries when they return home. This phenomenon of brain drain is a major challenge to the diffusion of GIS because much needed skills in GIS in developing countries are lost to employers in developed countries. Technological Factors For any technology to flourish, infrastructure needs to be in place to support it. Institutional GIS requires and IT support department, at the very least, for the proper functioning and utilization of the technology. Many public sector institutions do not have a functioning IT department. Maintenance of computer hardware and software is left to GIS analysts, who might not necessarily have the skills required to perform such tasks. 228 Moreover, many Departments in Ministries do not have a reliable Internet connection. For those that do, the connection is too slow. GIS sections within departments are unable to upload their GIS analysis results to a server to be shared over a network, whether via intranet or the Internet. This is counter to the notion that GIS data at public institutions is indeed public data, and needs to be shared with the public, for example, information regarding wetlands, forests, topography, geology and so on. In fact, many departments do not own a website domain. All ministries own website domains publishing information about their ministries, but these websites do not have any links to geospatial data and services provided by the respective departmental units. An interview respondent at the Department of Surveys and Mapping lamented: “We have Internet, we have been having Internet, but it goes off. No payment and we are out of Internet. Recently we got it, but it is still not stable. But again, the Ministry again wants … I don’t even know how we connect Internet, either the Ministry does not know … because about a week ago, I was hearing the Ministry wants to give our Department Internet. But we have been connecting ourselves to the Internet on a personal basis using personal wireless USB modems from Orange (Telecom).” The supply of electricity in Uganda is very unstable. Power-cuts for extended periods of time are common – outages from 6 to 24 hours in a day. Not only does this hinder day to day activities at public institutions, computers are at risk of crashing from unexpected shutdowns from power outages. A GIS analyst working on a complex GIS analytical task could easily lose his or her work in the event of a power outage. Automated ArcGIS Python geoprocessing scripts that normally run for hours at a time would fail to execute 229 to completion in the event of such an outage. All this amounts to frustration in a GIS workflow. 5.3.2 The Academic Sector Academic sector institutions in this research are broadly defined as institutions of higher learning: public and private universities, and technical colleges. Academic departments at four public universities were chosen for interviews in this research. Faculty members that use and teach GIScience in their curriculum were selected at the following universities: Makerere University, Kyambogo University, Gulu University, and Mbarara University. These public universities are fairly spread out geographically around Uganda. Makerere University, the largest university, and Kyambogo university are located in the central region; Gulu University in the northern region; Mbarara University in the western region. In terms of enrollment numbers, the chosen universities have the highest enrollment numbers of students on a regional and national scale. Table 5-7 shows the Departments, and the faculties and institutes under which they fall, that participated in interviews at the respective universities. These include Surveying, Environment and Natural resources, Geography, Forest Management, Information Systems, and Computer Science. Makerere University, the largest public university in the country, has multiple departments teaching of GIS in their curriculum, including the Departments of Surveying, Geography, Forest Management, Information Systems and the Institute of Environment and Natural Resources (MUIENR). The Computer Science Department at Mbarara University and Kyambogo University also offers GIS in their curriculum. Gulu University’s Computer Science Department offers short GIS courses to non-degree 230 seeking students affiliated with local non-governmental organizations with specific training needs. Gulu University offers GIS in its curriculum through the Department of Biosystems Engineering, Faculty of Agriculture and the Environment. However, this Department relies on the computer hardware and GIS software available in the Department of Computer Science as it has none of its own. Table 5-7: Academic Institutions at which interviews were performed ID Name Faculty/School/Institute Department Year GIS Introduced 1 Makerere University Technology Surveying Early 90s 2 Makerere University MUIENR – Makerere University Institute of Environment and Natural Resources Environmental and Information Management Early 90s 3 Makerere University Arts Geography 1997 4 Makerere University Forestry Forest Management 2000 5 Makerere University Computing and Informatics Technology Information Systems 2005 6 Mbarara University Computer Science Computer Science 7 Gulu University Science Computer Science 2009 8 Kyambogo University Science Computer Science 5.3.2.1 GIS Introduction into the Curriculum The first academic institution to use GIS for research, and introduce it into its curriculum, was MUIENR in the early 90s. This institution captured and stored geospatial data with regard to biodiversity in Uganda. Because MUIENR was considered a primary producer of environmental information on biodiversity, it was one of the eight institutions, and the only academic institution, to be inducted into the Environment Information Network (EIN) following a 1995 structural review of the National Environment Information 231 Center (NEIC) that subsequently led to the creation of the National Environmental Management Authority (NEMA), into which NEIC was inducted (Gowa, 2009). It follows that project-oriented and donor-driven nature of GIS within public sector institutions was spinning off into the academic sector within a process of GIS diffusion. The objectives of eight institutions incorporated into the EIN, including MUIENR, were to build capacity in environmental planning, data presentation, and the development of data standards. MUIENR’s role was mainly to build capacity in GIS and maintain a database of biodiversity data, which it produced and stored. The Department of Surveying, rebranded as the Department of Geomatics and Land Management in 2012, was the next academic institution to introduce GIS into its curriculum, also in the early 90s, mainly through its working relationship with MUIENR. The Department of Geography introduced GIS into its curriculum in 1997, followed by the Department of Forestry in 2000, both departments being at Makerere University. In 2005, GIS was introduced at Makerere University’s Faculty of Computing and Informatics Technology, Department of Information Systems. Gulu University’s Department of Computer Science had GIS technology introduced at the institution in 2009 by the Uganda Wildlife Conservation Society (WCS) in collaboration with USAID. The primary purpose of GIS at this institution is for carrying research relevant to the activities of the Uganda Wildlife Conservation Society, and to build capacity within the environmental NGO sector in Gulu district. GIS is not currently taught as a course within the Department’s curriculum. Interestingly enough, the Department of Biosystems Engineering, Faculty of Agriculture and the Environment, at 232 Gulu University actually teaches GIS as a course within its curriculum, however, it does not own its own computer hardware and software, and instead relies on the Department of Computer Science for its technological needs. A similar situation exists at Mbarara University that had GIS technology introduced at the Department of Computer Science mainly for the purposes of carrying of district level geospatial research, for example, to assist the local police in crime mapping activities. GIS does not exist as a course taught within the curriculum of the Department, but instead serves an external objective to assist local government institutions in south western Uganda to carry out focused geospatial research based on existing GIS capacity and skills of faculty members at Mbarara University’s Department of Computer Science. Departments that teach GIS, the number of students they enroll, and the number of faculty members that teach GIS courses, and the level at which they teach the courses are shown in Table 5-8. Six out of the eight Departments teach GIS within their curriculum. Six Departments teach GIS at the undergraduate level, while three teach it at the Master’s level. Only the Department of Information Systems, Faculty of Computing and Informatics Technology has a student doing GIS at the PhD level. The Departments of Geography, and Surveying (Geomatics and Land Management), have the largest number of faculty members that teach GIS, followed by MUIENR, and the Department of Forestry, all at Makerere University; the other public universities have only one or two faculty members that teach GIS. 233 Table 5-8: GIS at Public Universities in Uganda ID GIS Taught Level Enrollment per Semester No. of GIS Faculty No. of Computers No. of ArcGIS Licenses 1 1 Bachelors 60 4 130 50 2 1 Bachelors, Masters 100 3 40 ? 3 1 Bachelors, Masters 100 - 180 5 10 ? 4 1 Bachelors, Masters 30 2 60 ? 5 1 Masters, PhD 35 1 > 1000 Open Source 6 60 ? 7 60 ? 8 1 Bachelors 90 2 60 ? Enrollment in GIS courses occurs in two ways, (1) as a required course in a given program, and (2) as an elective course in a program. For example, as part of the undergraduate degree in Land Surveying in the Department of Surveying (Geomatics and Land Management), GIS is a required or core course that students have to take, whereas, in the Masters of Information Systems at the Faculty of Computing and Informatics Technology, it is an elective course. The highest number of students that enroll for a GIS course in a semester is at the Department of Geography (100 - 180) at Makerere University, followed by MUIENR (60 undergraduates and 40 graduate students), Kyambogo University’s Computer Science Department (90), the Department of Surveying (Geomatics and Land Management) (60), Information Systems (35), and Forestry (30) at Makerere University. GIS courses taught at Ugandan public universities have various names. The Department of Surveying (Geomatics and Land Management) offers Geographical Information 234 Systems as a core course in the fourth year of the Bachelor’s in Land Surveying degree program. This is an introductory level GIS course on the principles of GIS. Remote Sensing is offered as a required course in the third year, just prior to the introductory level GIS course. Courses that are relevant to understanding data collection techniques for GIS are also offered, for example, photogrammetry, Satellite Geodesy, Physical Geodesy, Geometrical Geodesy, Geometrical Astronomy, Principles of Land Surveying and Cartography. At the Department of Forestry, Makerere University has a more application centered to teaching GIS. In the first of the Bachelor’s in Forestry degree program, students take a required course called Forest Resource Assessment and Management, which introduces principles and applications of GIS in Forestry, and includes practical exercises involving data collection using handheld GPS units. This is followed by a more detailed GIS course in the fourth year of the program called Resource Assessment and GIS, which introduces students to GIS analytical methods. At the Master’s level, the Forestry Department offers a GIS course called Resource Assessment and GIS with a focus on specific applications of GIS in forestry. The Department of Geography offers a course called Introduction to GIS as a core Geography course in the 2 nd year of all programs within the Department. In the 3 rd year, GIS Applications is offered as an elective course tailored to the various programs in the Department. A third course called Advanced GIS is offered in the advanced stages of the degree program as an elective course depending on students’ interests. 235 MUIENR offers a course entitled Principles and Applications of GIS at the Bachelor’s level, while at the Master’s level, the Principles of GIS is offered as a required course, and the Applications of GIS is an elective course. 5.3.2.2 GIS Software and Hardware Usage All of the eight academic sector institutions interviewed use GIS technology in their daily work flows, either for teaching purposes or for research (see Table 5-9). The most pervasive GIS software used is Environmental Systems Research Institute’s (ESRI) ArcGIS 9.x (75%) and ArcView 3.x (62.5%) software. Other types of GIS software used include ILWIS (50%), QGIS (37.5%), Idrisi (25%), Google Maps/Earth (25%), and gvSIG (12.5%). Based on this distribution, it is quite clear that commercial GIS software from ESRI is more popular than open source GIS such as ILWIS, gvSIG and QGIS. The reason for this trend is the impact of donor funding on the penetration of certain types of software in the Ugandan academic sector. The link between commercial GIS software and project (donor) funding exists in the academic sector just as it does in the public sector. Take for example, the Department of Information Systems, Faculty of Computing and Informatics Technology, Makerere University uses open source Quantum GIS to teach GIS courses to university students. However, it also uses ESRI’s ArcGIS to teach short courses on a contract basis to employees of local non-governmental organizations and public sector institutions. This Department has a preference for Open Source software because it is available free-of- charge, and licenses for commercial GIS software are too expensive. ArcGIS was introduced in this Department under a NUFFIC (Netherlands Organization for 236 International Cooperation in Higher Education) project. Renewing of licenses was dependent on continued project funding, which made the department vulnerable to donor support. As a result, a decision was made by the Department to switch to open source Quantum GIS. Because computer science students are more tech savvy than students in other departments, the use of open source GIS technologies to teach GIS courses was feasible in this particular department. A respondent at this institution explained, “We use Quantum GIS. The reason (why we do not use commercial GIS software) is this: we have had problems with licensing, especially with payments and all these things. Even initially, I know that departments or even universities cannot afford the licensing fees. It’s quite expensive, for example, we needed 10 licenses, and we had to pay … officially, it was coming to around $15,000 – initial – then annual was around $5000 to be able to access all the …” (P. J. Ogao, Faculty of Computing and Informatics Technology, Makerere University, personal communication, June 29 th , 2010). The major disadvantage with open source software is that they are generally less user- friendly, have less intuitive user interfaces (UIs), and poor tech support. The major reason for the popularity of commercial GIS software, not only in Uganda, but on a global scale, is the user-friendliness of such software, for example, ESRI’s ArcGIS. However, the successful diffusion of ArcGIS software in Uganda has been due to two major reasons, (1) the influence of change agents, for example, UNEP-GRID, and donor agencies of developed countries, for example, NUFFIC that that funded foreign project implementing agencies that introduced ArcGIS at Ugandan academic sector institutions through project funding; and (2) the graduate level training of the faculty members that teach GIS at 237 Ugandan institutions, at ITC (the International Institute for Geo-Information Science and Earth Observation) in the Netherlands, an institution that popularized ArcGIS amongst its foreign students, many of whom were from developing countries like Uganda. Four out of the eight respondents in the interviews did their post-graduate training in GIS at ITC, Utrecht University and the University of Groningen in the Netherlands. Two respondents did their post-graduate training in England, Sweden and Norway at Universities that primarily employed ArcGIS software to teach GIS to their students. Three of the eight respondents were holders of PhD degrees and senior lecturers in their Departments, while the remaining five were holders of Master’s level degrees, employed as lecturers, assistant lecturers, and teaching assistants. Table 5-9: GIS Software at Academic Sector Institutions ID GIS Used ArcV iew 3.X Arc GIS 9.X Map Info Idrisi Erd as ILW IS gvSIG Google Maps/Ear th QGIS 1 1 1 1 1 2 1 1 1 1 1 3 1 1 1 1 4 1 1 1 1 5 1 1 6 1 7 1 1 1 1 1 1 1 8 1 1 1 1 1 Total No. 8 5 6 1 2 0 4 1 2 3 Total %age 100 62.5 75 12.5 25 0 50 12.5 25 37.5 GIS hardware used at academic institutions in Uganda is mainly Personal Computer- (PC) based, mainly desktop computers (See Table 5-8). The most well-equipped 238 Departments at Makerere University include the Department of Information Systems, Faculty of CIT (greater than 1000), Department of Surveying, Faculty of Technology (130), Department of Forestry (60), and MUIENR (40) (MUK-FCIT, 2009). The most ill-equipped Department in terms of GIS hardware is the Department of Geography with about 10 computers. A respondent lamented, “There are less than 10 computers in our GIS lab. In the last two years we lost, through vandalism, up to 20 computers. It’s terrible. We are trying to recapitalize the GIS unit. GIS at the undergraduate level is currently taught only at the theoretical level” (S. Lwasa, personal communication, July 2 nd , 2010). Much as there are about 60 computers all installed with ArcGIS software at the Department of Forestry, Makerere University, the annual renewal of GIS licenses is unsustainable. It is often the case that the licenses expire and students have to seek GIS technology resources from the better equipped Faculty of Computing and Informatics Technology. In fact, a respondent at the Department of Forestry explained that the two major constraints facing the GIS teaching in the Department are, (1) a lack of skilled GIS personnel, and (2) inadequate technological resources in the Department: “I think the challenge we have, or what could be easily addressed is partly on our side, with our curricula, to incorporate it a little more than what we’ve been doing. But of course, we’ve been putting it a little at a less stage because of the resource persons and the facilities in terms of the software. So even to do exams, we struggle to go to the Faculty of Technology (Department of Surveying) and Faculty of Computing and Informatics Technology – students have to go there to use their computers and software. If the faculty 239 of Forestry was well-equipped, in terms of software, and teachers and everything, GIS is something we would teach almost every academic year at different levels” (J. J. Namaalwa, personal communication, June 29 th , 2010). The quality of Information and communication technology (ICT) infrastructure at academic institutions affects the teaching of GIS at these institutions. Through collaborations between public sector institutions of higher learning, the government of Uganda, and International donor agencies, the ICT infrastructure of the four main public universities has been progressively improving. According to the Computing and Informatics Technology Report (MUK-FCIT, 2009), various improvements have been made to the ICT infrastructure at Makerere University, Kyambogo University, Mbarara University and Gulu university. For example, training of University staff in ICT skills was performed through a €5.7 million NUFFIC grant: The NPT Project on ‘Strengthening ICT Training and Research Capacity in the Four Public Universities in Uganda’, 2007 – 2011 (MUK-FCIT 2009, 11). The following improvements were made at the four public universities between 2007 and 2009: about 1,500 academic staff had been trained in IT skills; 2000 computers were installed to create engineering labs, a GIS lab, other types of ICT labs – about 1000 of these computers were networked. Makerere University also has free Wi-Fi availability on campus for students. All these developments are supporting the diffusion of GIS at the four public universities in Uganda, mainly in the faculties and departments associated with the field of Computer Science and Information Technology, and not necessarily other faculties and 240 Departments, for example, the faculty of Forestry, and the Department of Geography at Makerere University. 5.3.2.3 Impact of GIS in the Academic Sector on Society The teaching of GIS at institutions of higher learning in Uganda has had a positive impact on society in two ways. One, students who have taken GIS courses are able to broaden their analytical skills to apply GIS knowledge within their fields to solve problems from a geographical perspective. Two, society has benefitted from the GIS skills of new graduates as they are employed in the health and environmental sectors by non- governmental organizations (NGO) and International Organizations (IO) to provide support in the implementation of projects in rural and urban areas to improve livelihoods of local citizens. I provide two examples of university students’ projects where GIS skills attained through a university education have directly contributed to solving of real-world problems in Uganda within the context of developing nations and the Millennium Development goals. The Faculty of Computing and Informatics Technology organized two workshops on the use of Google Earth, and Google Maps in 2009. As a result of these workshops, interest amongst students in GIS increased leading to a decision by some students to enroll in a GIS course, and upon completion, to include a GIS element in their final projects. One such project by a student, Prosy Nassaka, was on the creation of a prototype mobile- based monitoring tool to locate faulty Automated Teller Machines (ATM) in the City of Kampala (MUK-FCIT, 2010). 241 Another research project at FCIT, OpenXdata, led to the development of an open source client-server software tool for capturing patient data using a mobile phone in rural areas and sending that information in real time to a central server hypothetically located at a hospital so that doctors can prescribe treatment for patients in real time. A similar research project, cu@School, based on the OpenXData open source platform led to the creation of a client-server system for monitoring the weekly attendance of primary and secondary school students and their teachers (Ndiwalana, 2011) so as to detect and geovisualize unusual absence (from school) as an indicator of an anomalous event, for example, a possible disease outbreak at a school. This project was implemented by FCIT at Makerere University in conjunction with SNV, an International not-for-profit organization based in The Hague, The Netherlands. One of my respondents, a former student at FCIT, Makerere University, explained that GIS skills that he picked up during his undergraduate studies in Information Technology helped him find a job at Data Care Uganda, a small company specializing in the creation Monitoring and Evaluation (M & E) database Systems. The company won a project tender to supply the Department of International Development (DFID), Uganda, with a custom M & E system with mapping capabilities. Technologies to develop the system include open source technologies, including PHP, MySQL, Apache Web Server, and Web GIS technologies, such as UMN Map Server. One of the major problems in project implementation by the respondent was access to private web hosting facilities. Another project being implemented by the same company is the Water Atlas project for the Ministry of Water and Environment, Department of Water Development. The objective 242 of this project was to capture geospatial data related to water management using handheld GPS units and to visualize the data in form of a map necessitating the need for a GIS component in the workflow. 5.3.2.4 Challenges to GIS Education in the Academic Sector There are several challenges facing GIS in the academic sector in Uganda. These can be classified as follows: institutional, technological, economic and social challenges. Institutional challenges At Kyambogo University, GIS courses are offered in multiple departments: computer Science, Lands and Architectural Studies, and the Department of Geography. The same situation exists at Makerere University where GIS courses are offered in at least five departments. Much as this can be viewed as a strength, some interview respondents argued that this as a limitation in terms of a lack of central mobilizing body to mobilize resources to support GIS at the University level. Some respondents suggested that a university-level GIS Center is needed, and would allow for the mobilization of resources to address technological and financial roadblocks to the use GIS and teaching of GIS courses at the various departments at Makerere University. For example, it would be cheaper to purchase a university-wide license for the proprietary ArcGIS software at a cost of $40,000 and obtain an unlimited number of floating licenses, and then distribute these licenses to all the Departments that offer GIS courses. Currently, each Department that teaches GIS purchases and manages its own GIS licenses, costs which when summed up could exceed those of purchasing a university-wide unlimited license package. The poorly funded departments have no choice but to seek donor funding for GIS licenses, or 243 forge partnerships with public sector agencies such as the National Forestry Authority so as to benefit from their GIS licensing. Class sizes at the various universities have progressively become larger over the years and this poses a challenge to the teaching of GIS, a practical course that requires lab sections in addition to lecture sections. The Department of Geography and the Department of Forestry Management, for example, have to send their students over to FCIT to do their practical exams because of a shortage of computers and GIS licenses in their departments. An interview respondent explained, “Note that sometimes we have problems with our licenses, so our students go to the GIS lab at MUIENR, or Faculty of CIT (Computer and Informatics Technology) to use their resources. That is actually where the students do their exams from because the software is more reliable and so we usually go there; even sometimes for teaching, we take the students there” (J. Namaalwa, Faculty of Forestry, Makerere University, personal communication, June 29 th , 2010). The effect of large class sizes goes beyond the quality of teaching of GIS – it affects lecturers who are swamped with teaching, and have no time to focus on research, a finding that is echoed by some scholars (see Owoeye & Oyebade 2011, 16). The syllabi for teaching GIS vary by department because of the different foci of each of the departments at the various universities. However, one critical problem that educators in the respective departments need to identify is whether the material taught is actually relevant to addressing real problems in society. Do the skills acquired by students actually address the needs of the private, public, and NGO sectors? Some departments teach GIS at only a theoretical level, with no lab sections. Students graduating out of such 244 programs will not be adequately equipped to apply their GIS skills in the real world. Partnerships between the various departments and the private, public and NGO sectors could help identify relevant skills required in the respective sectors, and in turn inform the updating of syllabi. Some respondents explained that there is a shortage of lecturers to teach GIS courses in their departments. This problem is related to large class sizes that put enormous pressure on the few qualified faculty members teaching GIS classes, often times with no support from teaching assistants. Even if the Departments wanted to hire new faculty members to teach GIS, there are very few on the job market. One cause of this problem could be the effect of brain drain, and poor remunerations at public universities in Uganda. Students who end up pursuing their graduate studies in the United States, Canada, Europe and Australia tend to remain in developed countries after their studies to seek greener pastures. A lecturer at FCIT lamented, “I know in ITC I trained with so many Ugandans, but when I came here (to Uganda), I was wondering, where are these guys?” One of the causes of brain drain is poor salaries at public universities in Uganda. The other problem is that young faculty members have few opportunities to join the various departments as lecturers because older and aging professors in those departments see new younger faculty members are threats to their own job security, and to the existing status quo in the departments, especially if corrupt practices have been benefiting them. The development of any field, including GIS, is promoted by cutting edge research. Without geospatial data, GIS research cannot move forward. Because of a problem in the public sector regarding the free sharing of public geospatial data, universities have found 245 it extremely hard to obtain, store and create an organized repository for geospatial data. Uganda has no NSDI and no geospatial policy in place, which has resulted in a “every man for himself” attitude. There is no collaboration among the various departments at Makerere University in terms of sharing of public geospatial data. Duplication of effort is common in data collection, which is a waste of already scarce financial resources. The creation of a university-level GIS Center would definitely help in the sharing of public geospatial data at the various universities; however, such an initiative is easier said than done due to a struggle for power and control. Power dynamics among the various departments are a source of disunity. Each department would seek sole control of such a powerful GIS Center, associating it with the inflow of funding. This problem is similar to the struggle to establish an NSDI in Uganda which resulted in a power struggle among the various stakeholders, ministries, and academic institutions in the country for control of that central repository of data. Technological Challenges The ICT infrastructure to support the teaching of GIS in the academic sector in Uganda varies in its quality by faculty and department. Makerere University’s FCIT has the best ICT infrastructure in the country with about 2000 computers, and other peripheral hardware devices. The Faculty of Technology and MUIENR also have a fairly good ICT infrastructure in place with a number of computers and plotters. The same cannot be said about the Faculty of Arts, and the Faculty of Forestry at Makerere University which struggle to teach GIS labs to their students due to a shortage of functional computers and GIS software licenses. 246 The choice between commercial and open source GIS poses a challenge to the faculty members that teach GIS courses. Because FCIT is a more technically oriented faculty, faculty members are conversant with open source GIS technologies, and indeed, use GIS software such as Quantum GIS to teach their students. However, in the less technically inclined departments, the use of ArcGIS and ArcView is preferred because these software are more user-friendly and most faculty are accustomed to them. Economic Challenges Just like the public sector, the academic sector relies heavily on donor funding. As a result, GIS at public universities is project-driven. Funding for GIS software and hardware relies on funding provided by donor agencies, for example, NUFFIC at Makerere University, and USAID and the Uganda Wildlife Conservation Society at Gulu University. Much as this funding helped kick-start GIS at these universities, it does not breed a culture of self-reliance and sustainability, as one of my respondents explained: “I can tell you the biggest stumbling block (in Africa). I published an article in the Chronicle on the 23rd of June, I don’t know if you’ve read it, the nine problems that hinder partnerships in Africa. And it is focusing on universities. And it lists 9 points, and to me it blames the global north universities. What I see is not the assumptions of global north universities; what I see is our own failures (the global south), and we are unable to reflect on those, find our weaknesses and take action on our weaknesses. And then we blame somebody else and we expect somebody else to come and solve our problems. It’s in Africa where a donation of a $1000 computer, somebody will still expect that there will be an additional annual donation of $50 to maintain it. It is absolutely ridiculous!” 247 (S. Lwasa, Department of Geography, Makerere University, personal communication, July 2 nd , 2010) At various departments, such as the Department of Geography, Makerere University, there are inadequate ICT facilities to support the teaching of GIS at the institution. The Department of Geography is a particularly bad case of this inadequacy. Having less than computers and 100-180 students per semester in GIS courses, it is no wonder that some lecturers take to teaching GIS at only the theoretical level. Those determined to offer students some hands-on practical training have to negotiate with the better equipped FCIT to allow students from the Department of Geography to use the latter’s computers and GIS software to carry out labs. There is also a lack of GPS receivers in the Department of Forestry. Acquiring of data collection skills in disciplines such as forestry is important to successfully completing the program yet students have to struggle to share the few GPS receivers to get hands-on experience using them. The high cost of GIS software licenses is a problem that was reiterated by all respondents in this research. ArcGIS has become the standard and preferred software to use for classroom instruction; however, licensing it can set the institution back thousands of dollars. ArcView 3.x, the free version of ESRI software that was released in the late nineties is still in active use at many academic institutions mainly because it can be installed on a computer without the need for a license. Due to the evolving nature of technology, it is not recommended for students to be left behind. It would be ideal for academic institutions to employ the most recent versions of ArcGIS and other GIS software for instruction in the classroom. This would expose students to the cutting edge 248 technologies available on the market, and would make students’ GIS skills more marketable. In recognition of this limitation, ESRI pledged to supply free GIS software to 100 African Universities starting in 2012 (ESRI, 2011). Social Challenges One of the respondents argued that the mindset within the institutional framework in Uganda is responsible for the lack of progress in Uganda, not only in the field of GIS, but in general: “I can summarize the stumbling blocks as our own untapped capacities, lack of leadership, the addiction to aid, and lack of vision to mobilize our own small resources to move forward, institutional challenges. Even when innovation occurs, the institution sits on it, ‘it is not important, you are moving so fast maybe as an individual’, things like that. It has very little to do with resources. It has very little to do with a lack of resources. It’s a mindset, and it’s an inter-generational gap” (S. Lwasa, Department of Geography, Makerere University, personal communication, July 2 nd , 2010). He also argued that the system of education in Uganda is modeled on the colonial cookbook style education, and needs to change to address the actual needs of the society. Due to a poor work ethic, mismanagement of funds at academic sector institutions is rampant. Funds allocated for ICT facilities and services are often times diverted by individuals in powerful positions in the government. This has left some academic Departments at these institutions vulnerable to the acquisition of substandard software and hardware technology to keep the departments afloat. For instance, a certain department at Makerere University was supplied pirated copies of ArcGIS licenses for $40,000 which eventually was of no use for obvious reasons leading to a colossal loss to 249 an already cash-strapped institution. Many departments have also suffered financial losses due to vandalism and theft of computers and equipment from their premises. The low salaries for the majority of employees in academic sector institutions have led to a shortage of qualified faculty to teach GIS courses. Often times, highly qualified students who come to developed countries for post-graduate studies have little incentive to return to their home countries and instead seek opportunities in the West leading to the so-called brain drain effect. A challenge that the academic sector has to deal with in order for GIS to be better utilized by the economy of Uganda is to make the skills students are handed down more relevant to the private sector. There is a general lack of awareness about GIS in the private, public and NGO sectors. For example, one of my respondents, the managing director of a solid waste management company had no idea about what GIS entailed, and how it could relevant to his business in terms of solving one of his biggest problems – the travelling salesman problem. Next, I present an analysis of the state of GIS in Uganda from a diffusion of Innovations perspective. 5.4 The State of GIS in Uganda: A Diffusion of Innovations Analysis In this paper, I investigate the current state of GIS in Uganda’s public and academic sectors, where “state of GIS” refers to the penetration level of the technology into the named sectors. However, to understand the current penetration level of GIS, one needs to understand how GIS penetrated into the institutional framework of the two sectors. Diffusion of innovations theory provides a theoretical framework for studying and 250 analyzing the process by which an innovation is adopted by society over time, and thus, is used to analyze the diffusion of GIS into developing countries, using Uganda as a case study. Diffusion is “the process by which (1) an innovation (2) is communicated through certain channels (3) over time (4) among the members of a social system” (Rogers 2003, 11). In this section, I analyze facets of the four elements of diffusion with respect to the diffusion of GIS in Uganda. This analysis allows for a better understanding of the process by which GIS technology has been adopted in developing countries, allowing for a confirmation of diffusion theory, or extension of the theory, with respect to developing countries. In this research, the unit of analysis is the organization (or institution), and not individuals. Most diffusion research is concerned with the diffusion of innovations among individuals; however, in the case of GIS technology, diffusion occurs not on individual basis, but on the basis of collective and authority innovation decisions. 5.4.1 The Innovation: Perceived Attributes Diffusion theory states that the perceived attributes of an innovation, as perceived by individuals, lead to the different rates of adoption of that innovation (Rogers, 2003). There are five characteristics of innovations that explain different rates of adoption of an innovation, according to diffusion theory: (1) relative advantage, (2) compatibility, (3) complexity, (4) trialability, (5) observability. Generally, those innovations with greater relative advantage, compatibility, trialability, and observability, and less complexity will be adopted more rapidly. 251 5.4.1.1 Relative Advantage The relative advantage of GIS technology over the technology it superseded can be measured in terms of economic terms, Rogers (2003) explains, but social prestige factors, convenience and satisfaction could be additional measures. In the case of Uganda, the technology that GIS replaced was paper, pen and ink analogue cartography, a very laborious and slow method used to produce maps still actively used today in many public sector institutions even after the introduction of GIS in those institutions, for example, the Kampala City Council, and the Department of Surveys and Mapping. Much as GIS has been readily adopted by institutions such as the National Forestry Authority, this is not the case in the two institutions mentioned previously – at least not at a significantly large scale, despite the fact that GIS is used to some extent in both KCC and the Department of Surveys and Mapping. Analog cartographic methods still continue to dominate the map making exercise at these two institutions. This is antithetical to Roger’s (2003) Diffusion theory. Why this anomaly? This is mainly because GIS poses a relative disadvantage as opposed to a relative advantage. The new technology threatens the already existing profession of draughting and analog cartography. If the institution were to adopt GIS, many draughtsmen and cartographers would be out of a job, or would require re-training in digital cartographic methods, an exercise that itself would be costly. One might argue that retraining old employees would make economic sense in that the costs incurred in the training program would produce returns in terms of time saving to produce maps. This argument would make sense if the map producers are interested in time as a dimension of 252 productivity; however, in cultural setting where time is not perceived with the same value as it is in the western world, this argument soon crumbles. Employees would rather spend longer periods performing their mapping tasks for the sake of job security, as long as monthly salaries are paid by the government. Unless a government or donor funded project is in phase, there is no time-delivery requirement on updating the topographical maps of the country or the City of Kampala. The greater the perceived relative advantage of an innovation, the more rapid its rate of adoption (Rogers, 2003). Much as the relative advantage, in terms of job security, has been a hindrance to the diffusion of GIS technology in departments such as KCC and the Department of Surveys and Mapping, this relative advantage in terms of economic returns has led to rapid adoption in the more “corporate” public sector agencies such as the National Water and Sewerage Corporation, a utility parastatal agency mandated with supplying water and sewerage services to various towns and districts in Uganda. The impact of GIS in this institution was a more efficient billing system managed with a GIS. This led to a substantial increase in revenues, and paved the way for a public/private partnership between NW&SC and a private company, Kampala Water. In this case, the perceived relative advantage of GIS technology, in economic terms, led to a rapid adoption of GIS within the organization. The two different outcomes of GIS adoption in the public sector, on the basis of relative advantage, imply that those public sector institutions that are of a more corporate nature, like the NW&SC, are more likely to perceive the relative advantage of a technological innovation like GIS in economic terms than the more traditional bureaucratic public 253 sector institutions that have no motivation for profit gain as a result of their day to day operations. 5.4.1.2 Compatibility Compatibility of an innovation is defined as “the degree to which an innovation is perceived as being consistent with the existing values, past experiences and needs of potential adopters” (Rogers, 2003). An innovation that is compatible will more likely be adopted by potential adopters. In Uganda’s public and academic sectors, GIS technology is perceived, in some cases, as compatible with the needs of the institutions adopting the technology, while in others not. In the public sector, at KCC, for instance, GIS is not compatible with the needs of the users because it is seen as a threat to jobs of analog cartographers. Also, the introduction of GIS technology would imply that a new department would need to be created, which implies expensive restructuring. This restructuring is unattractive to institutions with a strict bureaucratic organizational culture, such as KCC and the Department of Surveys and Mapping. At NW&SC, on the other hand, GIS is compatible with the needs of the organization as it greatly helps employees manage water and sewerage utility services in the country. At same time, the jobs of employees are not threatened as old fashioned cartographers are afforded training in the new technology so as to allow for a smooth transition to digital technology. In the academic sector, for example, at Makerere University, GIS technology allowed various academic departments to create and introduce new courses in the emerging field of GIS. 254 5.4.1.3 Complexity Complexity of an innovation can lead to either rapid adoption or rejection of the innovation. Complexity is defined as the degree to which an innovation is difficult to understand. GIS technology is a relatively complex innovation that requires advanced university level training, or persistent on-the-job training. Without access to such training, the diffusion of the technology is hampered. The user friendliness of ArcView and ArcGIS, the GIS software from ESRI, is one of the reasons for the rapid diffusion and adoption of the technology in the majority of Uganda’s public sector institutions. Idrisi, and MapInfo, the other GIS software applications that competed with ArcView and ArcGIS lost out because the software was simply too complex and non-intuitive to use. For similar reasons, many institutions abandoned ESRI’s PC Arc/Info command line- based software in preference for the visual-based ArcView when it was first released in the mid-90s. 5.4.1.4 Trialability “Trialability is the degree to which an innovation may be experimented with on a limited basis” (Rogers 2003, 16). A technology that is trialable by potential users is more likely to be adopted because it represents less uncertainty to the users considering it for adoption. In the case of Uganda’s public and academic sectors, GIS technology was not necessarily tried by public sector and academic sector employees before the decision to adopt was made. International organizations, such as UNEP-GRID, the change agents, introduced the GIS technology into these institutions based on their expert opinions about the technology. Also, donor funding often dictates which technologies are to be 255 introduced to developing countries, matters in which the recipients of donor aid have no say. Thus, trialability did not play any major role in the introduction of GIS in Uganda. However, trialability did certainly play a role in users adopting ESRI GIS software, ArcView and ArcGIS, as these software were more user friendly than the competing brands, Idrisi, MapInfo and Erdas. Also, it turned out that there was a preference for vector GIS over raster GIS at many of the public and academic sector institutions. 5.4.1.5 Observability Observability is defined as “the degree to which the results of an innovation are visible to others” (Rogers 2003, 16). The observability of computing technology is relatively less than that of consumer goods. In fact, observability is one of the main hindrances of the diffusion of GIS in Uganda. The National Planning Authority (NPA), an institution under the Ministry of Finance mandated with planning of the country’s infrastructure, does not possess a GIS, much as it does have a fully operational IT department. An interview with a respondent at the NPA revealed that one of reasons for this scenario is that technocrats at the institution have not had significant exposure to GIS technology, and are not sure what the benefits of such technology are. Much as they are aware of the existence of such technology in the various sectors of the economy, it is the opinion of the NPA that the technology is “islanded” with no central authority to collectively mobilize GIS operations for the national good. There is a general lack of awareness about GIS technology because it is not easily observable. In recent times, the observability of GIS has gradually increased. Regional GIS conferences held in Uganda, Kenya, and Ethiopia have gradually increased awareness about GIS technology within the public and academic sectors, in 256 essence, increasing observability, however, the impact on general awareness about the technology within the two sectors, especially amongst technocrats and decision makers, has been minimal. 5.4.1.6 Re-Invention The concept of re-invention in the diffusion of an innovation is very important. Re- invention is the degree to which an innovation is changed or modified by its adopters in the process of adoption and implementation so as to better suit the their needs (Rogers, 2003). With reference to GIS technology, re-invention would involve the customization of GIS software. Proprietary software from ESRI, for example, ArcGIS desktop can be customized using proprietary software development kits (SDKs) based on software industry standards. Open source GIS software, for example, Quantum GIS and UMN Map Server, on the other hand, can be customized using freely available industry standard software tools. In both cases, a very narrow and high level of technical skill is required. Such technical knowledge is in short supply in Uganda, currently and thus re-invention has not played any major role in the diffusion of GIS technology in the public and academic sectors. The only institution with the ability to re-invent GIS technology is the Faculty of Informatics and Computing Technology at Makerere University and this is mainly because of the strength of the faculty in computer science and informatics skills. Indeed, FCIT is one of the only institutions that employs open source Quantum GIS to teach GIS courses to their students. The technical ability of the students allows them to re-invent open source GIS technology while doing their academic projects, and later when they are employed. For 257 instance, one of the respondents, a former student at FCIT used UMN Map Server to implement a Web GIS application for the UK’s Department of International Development (DFID) project while working for a local NGO in Kampala. 5.4.2 Communication Channels GIS as an innovation was first introduced in Uganda by UNEP-GRID and NORAD in the public sector. The first institutions to receive GIS technology were NEIC and the Department of Forestry in the Ministry of Environment Protection in 1989. Thus, GIS was introduced in Uganda via certain communication channels. In diffusion theory, “communication channels are the means by which messages get from one individual to another” (Rogers 2003, 18). In the case of Uganda, interpersonal communication channels were responsible for the introduction of GIS in Uganda. These involved face-to- face exchanges between representatives of International Organizations, such as UNEP- GRID and Ugandan government officials in the Ministry of Environment protection. The nature of the interpersonal channels was heterophilous, in that the degree to which the individuals who participated in the exchange of information were significantly different with respect to attributes such as their cultural background, education level, socioeconomic and political status. Heterophily, as opposed to homophily, is defined as the degree to which two or more individuals who interact are different in certain attributes (Rogers 2003, 19). The change agents, representatives of UNEP-GRID in Uganda, introduced GIS technology to employees working at NEIC, Ministry of Environment protection. These change agents were expatriates, and not at of a higher socio-economic status than the Ugandans they interacted with. According to Rogers, such 258 heterophilous power-relations are a distinctive problem in the diffusion of innovations. In the case of Uganda, I argue that the converse is true, in that heterophily encouraged the diffusion of GIS technology into the public sector of Uganda mainly because the imbalance in power relations in favor of the change agents justified the importance of the technology to the receivers of the technology. Further, the fact that UNEP-GRID provided the technology to Ugandan institutions at “zero cost” (through a World Bank and AFTEN funded project) allowed for easier introduction of the technology into the public sector. 5.4.3 Time Time is an important element in diffusion studies. This time element is involved, first, in the (1) innovation-decision process that involves an individual passing from initial awareness of an innovation all the way through to adoption or rejection of the innovation. Second, time is also involved in the (2) relative earliness/lateness of adoption of an innovation. Third, time is involved in the (3) rate of adoption of an innovation, measured by the number of members of a social system that adopt an innovation in a given time period (Rogers, 2003). 5.4.3.1 The Innovation-Decision Process The innovation-decision process involves five main steps: (1) knowledge, (2) persuasion, (3) decision, (4) Implementation, and (5) confirmation (Rogers, 2003). Knowledge Knowledge of GIS technology was gained by many public sector institutions through change agents, mainly International donor agencies, such as UNEP-GRID. These change 259 agents provided in-house and off-site training to potential Ugandan adopters of GIS so as to persuade them to adopt the technology. Persuasion and Decision The persuasion stage was an information gathering stage during which Ugandans at public sector institutions obtained relevant information regarding GIS technology, and this reduced uncertainty about the Implementation of the technology which led to a decision to implement the Technology. However, this decision to implement the technology was not purely based on knowledge and persuasion. Institutions in developing countries are prone to adopting technologies that are given to the adopters as a result of donor assistance, and the decision to adopt is not necessarily theirs to make, but the donors’ decision instead. Due to the nature of the power-relations in the donor- developing countries’ context, the innovation decision process is rather one-sided from donor to receiver. Implementation Implementation of GIS technology in terms of acquiring and installing hardware and software was achieved in Uganda’s public sector mainly through donor funding from institutions such as the World Bank, and NORAD in the public sector. In the academic sector, implementation of the technology at MUIENR, for example, was through assistance from NEMA, one of first institutions to implement GIS in Uganda, albeit through donor funding. The Faculty of Forestry, Makerere University, similarly implemented GIS through assistance from the National Forestry Authority (NFA), a public sector institution that had itself initially benefited from donor funding of its 260 Biomass project. With the exception of the Faculty of Computing and Informatics Technology, Makerere University that implemented GIS through direct donor funding from the Dutch government, there seems to be a general trend in the academic sector of a strong relationship with an affiliate institution in the public sector into which GIS has already diffused previously through donor funding. A lack of re-invention of GIS technology in Uganda’s public sector is one of the factors that has led to slow diffusion of the technology. There has been a general lack of expertise to be able to customize GIS software appropriate for specific tasks in the public sector. For example, the Department of Surveying and Mapping has a need for customization of GIS software to be able to efficiently create a national cadaster and land information system, but has no skilled capacity to customize its ArcGIS software to realize such a cadaster. In fact, it still heavily relies on the “free” older version of ESRI software, ArcView, which has very limited customization (“scripting”) capabilities. It follows that most of the country’s cadastral maps are still in analog paper form. The only hope for a national cadaster is through an on-going World bank project at the Ministry of Lands, Housing and Urban Development, the Land Information System (LIS) project (MLHUD, 2011a). Another example of the lack of re-invention with regards to GIS technology at Uganda’s public sector institutions is at the Department of Wetlands Management. For over ten years, attribute data about seasonal and permanent wetlands at this department was stored in Microsoft Access databases, while geospatial data (the geometry of wetlands) was stored separately in ArcView GIS project files between 1990 and 1994. Due to a lack of skilled capacity, it was not possible to combine the two 261 datasets until 2008 when the Department obtained ArcGIS, and employed a skilled GIS analyst, trained at a foreign University, with the necessary customization knowledge required to do the merging of the data. Confirmation Confirmation of GIS technology occurred at some public sector institutions upon reinforcement of the initial innovation-decision made by donor agencies on behalf of Ugandan public sector institution employees. However this was not the case at all the institutions. The positive results of GIS technology were readily observable at the National Forestry Authority, for example, where estimations of the entire country’s biomass were made based on satellite imagery and Idrisi software. At Kampala City Council, however, the uncertainty about the technology with regard to its effect on job security and organizational culture led to resistance to change from analog cartography techniques to new digital methods. Another reason for resistance to GIS technology is due to organization culture with regards to work ethic and corruption. Uganda has been listed as one of the most corrupt countries in the world, ranking 143 rd out of a total of 182 countries considered by Transparency International (2011). A corrupt organization benefits from inefficiency in the daily work flows of public sector organizations. This translates into long wait times for basic services to the public, and in some cases, the need for an outright bribe to obtain value added services. Due to frustration among the population being served, local citizens would rather pay a bribe than wait years to obtain a basic service. Examples of such services include the obtaining of a land title for newly purchased land, the reconnection 262 of a piped water or power line after the disconnection of an illegal connection. GIS technology would streamline these processes leading to faster return on investment times, which would only benefit the institution and not the poorly paid employees who rely on inefficiency of analog mapping and document management methods to obtain kick-backs to supplement their salaries. In this case, organizational culture is indeed a hindrance to the confirmation of GIS technology at public sector institutions in Uganda. 5.4.3.2 Adopter Categories Adopter categories in diffusion theory include: (1) innovators, (2) early adopters, (3) early majority, (4) late majority, and (5) laggards. Innovators I argue there were no innovators as far as GIS technology is concerned in Uganda. Innovators are those who independently seek information about new ideas, are exposed to mass media, and their interpersonal networks extend over a wide area, reaching outside their local system (Rogers, 2003). An innovator is venturesome and ready to cope with high levels of uncertainty and risk in trying out a new innovation. GIS technology was not necessarily sought by the Ministry of Environment Protection; instead, it was simply “forced” into the organization by a larger global institution, UNEP-GRID. At an even larger level, the initial conference on the human environment in the early 70s (UNEP, 1972) was at the heart of the creation of UNEP as a UN agency, and the introduction of GIS in developing countries twenty years later. In fact, the innovators of GIS Technology, under the lens of diffusion theory, were actually UNEP-GRID who 263 experimented with various GIS technologies before forming a working relationship with ESRI to supply GIS software to developing nations. Early Adopters Early adopters of GIS technology were the public sector institutions affiliated with the Ministry of Environment Protection. It is not because of the were innovators in their own right that they were early adopters, but instead, the fact that they were chosen by a higher authority, UNEP-GRID and the donor agencies funding and pushing the environment protection agenda on a global scale that landed the ball in their court. Thus, early adopters included NEIC, the Department of Forestry (today known as the National Forestry Authority), and the Wetlands Management Division (today known as the Department of Wetlands Management) – the benefactors of International donors interested in protecting the global environment. Nonetheless, these early adopters served as role models for those public sector institutions that followed, such as the Department of Surveying and Mapping, and the Department of Geological Surveys and Mines in the public sector, and MUIENR, the Department of Surveying, and the Department of Geography at Makerere University (in the academic sector) all in the mid- to late-90s. These institutions helped trigger the critical mass required to spark off the diffusion of GIS technology in Uganda. Early Majority An early majority of adopters was realized in the early to mid-2000s with new adopters such as UBOS, Karamoja Data Center (today known as Northern Uganda Data Center), the Ministry of Agriculture, the National Agricultural Research Organization (NARO), 264 and the Uganda National Roads Authority in the public sector; the Faculty of Forestry at Makerere University in the academic sector. Late Majority and Laggards Because the diffusion of GIS is still on-going at Ugandan public and academic sector institutions, it would be premature to talk of a late majority and laggards. Institutions that adopted GIS in the mid-2000s to 2010 include the Faculty of Computing and Informatics at Makerere University, and Mbarara University. Based on these findings, a general trend appears that the late majority, and laggards, if at all these latter adopters can be referred to as such, seem to be academic sector institutions, while the early adopters seem to have been institutions in the public sector. The reason for this trend is definitely related to power dynamics within the institutional infrastructure of Uganda, as public sector institutions tend to be much more powerful and better exposed to donor funding, which is at the heart of diffusion of GIS in Uganda. The late majority and laggards, contrary to diffusion theory, is not characterized by suspicion of an innovation by the late adopters in the case of Uganda. Instead, it seems to be characterized by the power relations between the public and academic sectors, the former being exhibiting much more power and influence over the latter. In fact, all the universities in the academic sector considered in this study are government-owned universities. This means that in essence they are part of the public sector, although they can be considered within a sub-category, the academic sector. 265 5.4.3.3 Rate of Adoption The rate of adoption of GIS can be visualized by plotting the percentage of adoption of GIS versus time. This makes sense if the total sample of targeted adopters is known, and in diffusion theory such analysis is done with reference to individuals as adopters, for example, farmers adopting a new hybrid corn variety (Rogers, 2003; Ryan & Gross, 1950). In this research, organizations instead of individuals represent the unit of analysis. This is not a problem as long as the total number of potential adopting organizations is known. However, such an analysis is problematic because the definition of an organization or institution in the public and academic sectors is rather arbitrary. For example, I have used Departments at Ministries as the unit of analysis in the public sector, and academic departments at Universities as the unit of analysis in the academic sector. These units of analysis are arbitrarily defined as institutions or organizations. The justification and assumption used in this research project is that these units represent the largest unit within a larger institution, for example a government Ministry, or a University faculty, that have a unique function within the encompassing institution. Thus, justifying a quantitative measure of the number of institutions or organizations that have adopted GIS is problematic with regard to the definition of an acceptable unit of analysis. This definition and assumption in turn determines the empirical diffusion analysis based on the resulting total number of institutions or organizations in Uganda, especially with regard to plotting of the S-shaped diffusion curve. Moreover, not all institutions actually need GIS in their daily operations, for example, Departments under the Ministry of Justice, Constitutional Affairs, and Attorney General. 266 Including such institutions within the sample would imply a Pro-innovation bias in the results. Indeed, the sample of institutions covered in this research only represents 6 out of the total 21 Ministries (Commonwealth-Network, 2012) in Uganda. Thus, the analysis of the data is to be viewed with caution as a larger sample would have probably been more representative of the complete institutional framework in the country. In light of this limitation regarding sample size and semantic definition of an organization or institution, I make an assumption that my definition of an institution or organization is acceptable within the limits of my research questions and I proceed to plot a diffusion curve based on my sample of public and academic sector institutions (See Figure 5-1). I argue that the rate of adoption of GIS in Uganda started slowly between 1986 and 1990, picked up some momentum between 1991 and 1995, reaching a maximum between 1996 and 2000. Between 2001 and 2005, the diffusion process started losing momentum, and between 2006 and 2010, this momentum dropped even further. This trend is typical in diffusion of innovations, and is manifested in an S-shaped curve (H. J. Campbell, 1993; Chan & Williamson, 1999a; Rogers, 1993, 2003). 267 Figure 5-1: S-Shaped Curve: Diffusion of GIS in Uganda’s Public and Academic Sectors, 1986-2010 5.4.4 Social System The social system is the fourth and final element in the process of diffusion of innovations. By definition, a social system is “a set of interrelated units that are engaged in joint problem solving to accomplish a common goal” (Rogers 2003, 23). The members or units of a social system may be individuals, groups, or organizations. Diffusion occurs within a social system because the social system constitutes a boundary within which the innovation diffuses. Rogers (2003) identifies five characteristics of a social system that affect diffusion within a social system: (1) social structure, (2) norms, (3) roles of opinion leaders and change 268 agents, (4) types of innovation-decisions, and (5) the consequences of innovation (Rogers, 1993, 2003). I analyze the diffusion of GIS in Uganda’s public and academic sectors through the social system lens next. 5.4.4.1 Social Structure The social structure of public and academic sector institutions is characterized by heavy bureaucracy and vertical hierarchical relationships, while the communication structure within these institutions is characterized by homophilous relationships among employees within the organizations. Bureaucracy is an impediment to the diffusion of an innovation in general, and GIS in particular. The public procurement process has been identified as one of the impediments to development, for example, see Agaba & Shipman 2006; Basheka 2008. Acquisition of GIS software and hardware must go through a public procurement process that can take several months, and in some cases, years to complete. Much as this process ensures a fair bidding process for the supply of goods and services to public sector institutions, the downside is that it frustrates efforts of employees to quickly obtain the necessary GIS software and equipment. Some institutions have had a less stringent public procurement process than others, for instance, the National Forestry Authority (formerly the Department of Forestry), and thus GIS has blossomed within this organization. On the other hand, there are cases of extremely sluggish bureaucratic procurement procedures hampering the diffusion of GIS at institutions such as the National Environmental Management Authority (NEMA) and the Department of Surveys and Mapping. 269 One respondent explained why the Department of Forestry, in 1995, moved faster than NEMA in terms of GIS diffusion: “They (Department of Forestry) took a much faster step. Because when we (NEMA) wanted to buy equipment, for us we started having these World Bank programs. Actually, these people started after 1995; because at first they had someone, when they were starting the Biomass Study, they were doing it manually, going to pick things and so on. So the introduction of GIS was when they got another person in charge of coordinating that program, because it was a Norwegian forestry program – the Norwegian Forest Society – which was financing them. But for them, their conditions for purchase, and so, on were very simple, and easy, so they moved fast, and they bought equipment very quickly and software, and they did direct purchase. Whereas for us, we passed through the normal laborious procurement process – red-tape – and we got our equipment after 5 years, yet we were supposed to get it in about 2 months” (F. Mpabulungi, NEMA, personal communication, July 19 th , 2010). Another effect of excessive bureaucracy is the complexity it causes in the hiring process of new GIS personnel at public and academic sector institutions. The Department proposing a new hire for a new GIS position has to justify this hire to its parent Ministry, which then contacts the Ministry of Public Service, which in turn contacts the Ministry of Finance, which often times responds, “There is not enough funding for a new hire,” according to one of the interviewees at the Department of Surveys and Mapping, CAMPUS project. The result of this is that many GIS departments in Uganda’s public and academic sectors are understaffed, yet there is an urgent need for skilled personnel. 270 As far as communication structure is concerned, public and academic sector institutions generally exhibit homophily with regard to communication among the employees in the organization. However, these homophilous relationships occur in cliques based on different specializations within the organization. GIS specialists tend to communicate more with fellow GIS Analysts, cartographers, and physical planners, for example, at Kampala City Council. This allows for easier GIS diffusion, however, there are certain professionals who feel they are left out of the conversation. For instance, one of the respondents, an economist at one of the public sector institutions in Kampala, lamented the lack of communication between his department and the GIS department at his institution. He felt that better communication between the two would lead to an exchange of information and ideas, which in turn would lead better analysis of their socio- economic environmental data, and more effective decision making. He suggested that the reason for the seeming lack of information exchange between the two parties was possibly due to job insecurity on the part of GIS Analysts within the organization to avoid competition for their jobs from other professionals within the same organization. 5.4.4.2 Norms System norms are the established behavior patterns for members of a social system (Rogers, 1993, 2003). With relevance to the diffusion of GIS technology, there are several norms that were identified by respondents that affect the diffusion of GIS technology in the public and academic sectors: (1) a culture of corruption and impunity, (2) old-fashioned attitudes towards power-relations within the organizational hierarchy, (3) poor attitudes towards work and service delivery, (4) heavy reliance on donor 271 funding, and (5) poor attitudes towards sharing of public data. I expand on each of these behavioral patterns next. Corruption Corrupt practices by powerful officials in the various public and academic sector institutions have hampered the diffusion of GIS at these institutions. Some interview respondents confided that top officials within their organization blatantly diverted public funds for personal use, and mismanaged whatever little funds were allocated to the organization under the annual fiscal budget. The source of the problem in some cases is at the top level of the organizational pyramid; at the level of the Ministry, for example, in any given year, a public sector department of a Ministry would never receive 100% of the funds allocated to it under the national budget drafted by the Ministry of Finance because a percentage of the allocated funds disappeared from the national treasury before disbursement to the respective department. It should be noted that the funds allocated to some of these departments are inadequate, to begin with, and to make matters worse, even the little is lost to misappropriation. Of greater concern is the culture of impunity whereby corrupt officials expect never to be brought to book, and even if they are, expect to be acquitted. Corruption affects the diffusion of GIS because it leads to insufficient funds for securing and maintaining GIS software licenses, hardware, and peripheral equipment, such as GPS receivers. Old Fashioned Attitudes towards Power-relations in Organizations In Uganda’s public sector, there are old fashioned attitudes towards power-relations within the organizational hierarchy. Those higher up in the hierarchy do not expect 272 criticism from their subordinates lower in the hierarchy. These powerful officials sometimes get into such positions through personal friendships and relations with powerful politicians, including family based relationships. Because many positions are oftentimes filled based on personal relations, and not merit, this often leads to incompetence in the top leadership of the organization. Patronage to such leaders becomes the norm within the organization, leading to a decline of performance amongst staff whose only concern becomes patronage to the boss so as to keep their job, with little to no criticism of the operations of the organization. Poor Attitudes towards Work There is generally a very poor attitude towards work – a poor work ethic – in the public sector and academic sectors in Uganda. Because of poor remunerations, public servants (employees in the public sector) tend to have little motivation to do their work, choosing often times to slack off and slow down the organization’s progress. 2011saw several strikes in various sectors of the economy, for example, demonstrations and strikes by secondary school teachers, university faculty and staff, and medical doctors, over low government salaries. As a result, there is little motivation to do public work, which cascades into poor service delivery in the public sector. A poor attitude towards work slows down the diffusion of GIS because it manifests in a lack of interest in learning new GIS skills on-the-job, and a general lack of interest in using the technology. Reliance on Donor Funding “Addiction to aid” is a problem that many developing countries face (Alesina & Dollar, 2012; D. Moyo, 2009). There is a heavy reliance on donor funding to support service 273 delivery in the public and academic sectors in Uganda (Mwenda, 2007). It is in this vein that GIS diffusion in Uganda is directly linked to this foreign aid, in form of grants, and loans. Much as this aid allowed the introduction of GIS Technology into Uganda; it certainly did not nurture and support a sustainable development model. Often times, at the end of the donor supported project that introduced GIS technology into an institution, GIS software licenses expired and public sector agencies were unable to secure funding to renew them, and for hardware maintenance. This was mainly because GIS technology had not yet received recognition as an institutional asset worthy of specific budgetary appropriation by the Ministry of Finance. As such, GIS in Uganda has not yet been institutionalized, and simply remains “project” GIS. Poor Attitudes towards Sharing of Public Data One of biggest problems facing the diffusion of GIS in Uganda is the lack of free access to public geospatial data (Amadra, 2003a; Muhwezi, 2006; Musinguzi, 2003a; Nyemera, 2008). Public sector institutions that are data producers have a tendency to charge large exorbitant prices for the purchase of their data. The sale of data to the private sector makes economic sense if entrepreneurs are seeking to add value to the data and resell it; however, charging other public sector agencies trying to acquire this data is politically and ethically contentious, especially if the use of the data is in the interests of the common good. As a result of this problem, some public sector organizations have opted to go out and perform their own data collection exercises. In many instances, this turns out to be largely an expensive and wasteful pursuit due to duplication of efforts in data collection (Ernest, 274 2010; Karatunga, 2002; Lwasa et al., 2006; Nyemera, 2008). The main reason for this trend is the lack of a data sharing policy in Uganda (Ernest, 2010; Kalande & Ondulo, 2006; Nyemera, 2008; Uwayezu, 2010). 5.4.4.3 Roles of Opinion Leaders and Change Agents Opinion leaders and change agents play important roles in the diffusion of innovations within a social system. Opinion leadership is “the degree to which an individual is able to influence other individuals’ attitudes informally with relative frequency” (Rogers 2003, 27). Opinion leaders may support innovation, or oppose it. They are usually more exposed to all forms of external communication, and are more cosmopolite, have a higher socio-economic status, and are more innovative (Rogers, 2003). They are in a unique and influential position within the social system’s communication structure, and are at the center of interpersonal communication networks (Rogers, 2003). In Uganda’s public sector, GIS diffusion was promoted by heads of the various departments within the Ministries that GIS was initially introduced, for example, the Department of Forestry, the Wetlands Management Division, and NEIC. In the academic sector, heads of department and specific faculty members who had had exposure to GIS technologies during their post-graduate studies in Europe, for example, at ITC in the Netherlands, acted as opinion leaders, for example, faculty members at MUEINR, Department of Surveying, and the Department of Geography, Makerere University. These opinion leaders echoed the promotional considerations about GIS technology issued by the change agents. A change agent is “an individual who influences clients’ 275 innovation-decisions in a direction deemed desirable by a change agency” (Rogers 2003, 27). The major change agents in Uganda with regard to GIS diffusion in the public and academic sectors were professional expatriates who worked for international donor agencies and organizations, such as the World Bank, UNEP-GRID, and NORAD. These agencies funded projects that introduced GIS technology to Ugandan institutions. They worked hand in hand with implementers of the projects, usually GIS consultants from foreign-owned GIS companies, who drew upon expertise of other change agents in terms of technology support, including GIS professionals from the US Geological Survey, Oakar Services, ESRI Eastern Africa, GeoInformation Communication (GIC – ESRI’s value-added reseller in Uganda), and the Regional Center for Mapping, Nairobi. In terms of GIS education and training, ITC-Netherlands played a leading role in educating Ugandans in the science of GIS. Public sector institutions sponsored a number of local employees to seek their master’s level degrees in GIS at this University in Holland. Upon the return of these employees back to Uganda, the recent GIS graduates become the new opinion leaders which aided the farther diffusion of GIS within their respective organizations. 5.4.4.4 Types of Innovation-decisions In a social system, there are three types of innovation-decisions that lead to the eventual adoption or rejection of an innovation: (1) optional (2) collective, and (3) Authority Innovation decisions (Rogers, 2003). 276 Optional Optional innovation decisions are those made by an individual independent of the decisions of the other members of the system. There are several cases of such decisions leading to adoption of GIS technology, for example, the Department of Geological Surveys and Mines’ decision to adopt GIS resulted from the one of the employees at this institution going to Belgium for a Master’s degree in GIS and then returning to the Institution with a student copy of MapInfo GIS software. Another example is the adoption of ILWIS and ArcView 3.x GIS software are the National Agricultural Research Laboratories (NARL), also commonly known as the Kawanda Research Station, where the current head of the GIS Department introduced it at the institution upon returning from his studies at ITC, Netherlands. Because ILWIS was the main GIS software used for training at ITC in the early to mid- 90s, this influenced the innovation-decision at NARL. In the academic sector, the department of Geography, Makerere University, exhibits optional-decision making with regards to which software faculty members use to teach GIS to their courses. This choice often is influenced by the educational background, skills and preference of individual lecturers. The Department of Surveying (known today as the Department of Geomatics and Land Management) exhibits similar trends with individual faculty members deciding the type of software to use for their individual courses, for example, ArcView 3.x and ArcGIS for teaching the Principles of GIS, while TNTMips for remote sensing courses. 277 Collective Collective innovation-decisions are choices to adopt or reject an innovation based on Consensus amongst members of the social system (Rogers, 2003), and all units must conform to the system’s decision once it is made. Owing to the high levels of bureaucracy in public sector institutions in Uganda, such an innovation-decision is a rarity. Cases of collective innovation-decisions can be found in the academic sector, for instance, the decision to adopt open source Quantum GIS at the Faculty of Computing and Informatics Technology was as a result of consensus among faculty members after a deal to acquire ArcGIS software went sour. Authority Authority innovation-decisions are “choices to adopt or reject an innovation that are made by relatively few individuals in a system who possess power, status, or technical expertise” (Rogers 2003, 28). An individual member of the system has little influence in the innovation-decision. The majority of innovation-decisions at public and academic sector institutions in Uganda are authority innovation-decisions. For example, the decision to adopt GIS at NEIC (today known as NEMA) was originally made by top government officials in the Ministry of Environment Protection, in conjunction with change agents such as UNEP-GRID, and the World Bank’s AFTEN. Diffusion was not in any way dependent upon consensus amongst the employees in the Ministry. Another example is the decision to adopt Idrisi GIS at the Department of Forestry which was again a decision made by top government officials in the Ministry of Environment Protection in conjunction with the change agents, NORAD, and the Norwegian Forestry Society. 278 5.4.4.5 Consequences of Innovations The last element of diffusion with regard to a social system is the consequences of innovations. Rogers (2003) distinguishes between three types of consequences: (1) Desirable versus undesirable, (2) direct versus indirect, (3) anticipated versus unanticipated. Rogers (2003) states that change agents often mean to introduce innovations into a client social system that they expect will have desirable, direct and anticipated consequences, although often times, the undesirable, indirect, unanticipated consequences can result. In the case of Uganda, the introduction of GIS has often had desirable, direct and anticipated consequences. At the Department of Surveys and Mapping, CAMPUS project, the diffusion of GIS led to the rapid mapping of service delivery facilities, as point features, all over the country using handheld GPS units that ultimately allowed for better planning of district level local government resources. This was a direct, desirable and anticipated consequence of the innovation. However, the undesirable, indirect and unanticipated consequences of GIS diffusion include, (1) the creation of a dependency on donor funding to maintain GIS software licenses, (2) the duplication of efforts in geospatial data creation by various data producers, which is a wastage of project funds, and (3) a threat to job security of current professionals in the fields of analog cartography, and photogrammetry. 5.4.5 Criticism of Diffusion Research Despite all the contributions that diffusion of innovations theory and research has made to our understandings of human behavioral change over time, there are several criticisms 279 of this theoretical framework. There are four main criticisms of diffusion research, (1) the pro-innovation bias, (2) the individual-blame bias, (3) the recall problem, and (4) the issue of equality (Rogers, 2003). 5.4.5.1 Pro-innovation Bias The pro-innovation bias is the implication in diffusion research that an innovation should indeed be diffused and adopted by all members of a social system, that it should be diffused more rapidly, and that the innovation should be neither re-invented nor rejected (Rogers, 2003). The main reason for this bias is because much diffusion research is funded by change agencies. Further, successful diffusion leaves traces that can be investigated by diffusion researchers, while an unsuccessful diffusion effort leaves little or no visible traces (Rogers, 2003). In this paper, I have tried to highlight both the positive and the negative consequences of GIS diffusion, for instance, much as GIS is a useful planning tool used in the public sector, its effect on job security, maintenance costs in terms of licenses, and the dependency on donor funding has also been highlighted. I have also discussed the fact that re-invention is important for diffusion of GIS technology; however, one struggles to find signs of GIS software customization at Ugandan public and academic sector institutions. 5.4.5.2 Individual-blame Bias Diffusion research tends to side with change agencies that promote innovations rather than with individuals who are potential adopters (Rogers, 2003). This is often because the sponsors of diffusion research tend to be change agencies. The implication of this is that blame for unsuccessful diffusion of innovations tends to be on the individual rather than 280 the social system leading to an individual-blame bias in the research findings; for example, in the case of GIS diffusion, the GIS technology vendors, International donors, and technocrats in government. In this paper, I have made considerable efforts to address this bias by analyzing the roles of the various members of the social system including international donors, International Organizations like UNEP-GRID, and individuals in the institutions adopting GIS technology. 5.4.5.3 Recall Problem Time is one of the four elements of diffusion; however, the time variable tends to be problematic in terms of the, so called, recall problem. Diffusion research depends upon self-reported recall data from respondents with respect to the date of adoption of a new innovation (Rogers, 2003). The accuracy of the recall data depends on how involved the individual was with the innovation, how far back the innovation was adopted, individual differences in memory, education level, and so on. To avoid these problems, certain measures can be put in place, such as carrying out field experiments, longitudinal studies, use of archival records, and case studies of the innovation process based on data from multiple respondents, providing a validity check on each other’s data (Rogers, 2003). It is this last approach that was used in this research on GIS diffusion in Uganda. Much as respondents were chosen from various institutions in the public and academic sectors, in some cases more than one respondent was chosen from the same institution, and in multiple cases, some institutions were all under a common larger institution, and thus information provided by one respondent was checked by another. For example, NEIC, 281 NFA, the Department of Wetlands Management are all under the Ministry of Water and Environment; and the departments of Forestry, Surveying, and Computing and Informatics Technology are all academic departments at Makerere University. Interview information from each of these departments allowed for cross checking of the responses from the interlinked institutions relative to the diffusion of GIS in the larger institution, that is to say, the Ministry of Water and Environment, and Makerere University, respectively. In addition to interviewing multiple respondents from the same institution, some archival research was also done to establish exact dates of introduction of the technology and other information relevant to the early diffusion of an innovation. In this research project, whenever possible, authentic documents were sought and used to check the accuracy of transcribed interview data. 5.4.5.4 Issue of Equality The issue of equality in the Diffusion of innovations has often been ignored by diffusion researchers (Rogers, 2003). Diffusion tends to cause inequality in socio-economic benefits of the innovation to the individuals in a social system. Diffusion tends to widen the socio-economic gap between the higher and lower-socioeconomic status segments in society, especially in developing nations (Rogers, 2003). Diffusion in developing nations tends to be similar to developed nations in Europe and North America, for example, rate of adoption occurs following the familiar S-shaped curve over time, innovators are characterized by higher social status, greater cosmopoliteness, and greater tolerance for uncertainty than are other adopter categories (Rogers, 2003). 282 With regard to development in developing nations, an intellectual issue that arose in the 70s was cultural appropriateness of diffusion research in parts of world with very different sociocultural conditions (Rogers, 2003). The basic conception of development in the west consisted of four main elements: (1) economic growth through industrialization, (2) Capital-intensive, labor-saving technology transferred from the industrialized nations, (3) centralized planning by government economists to speed up the process of development, and (4) the causes of underdevelopment , which lay mainly within the developing nation, rather than in their trade or external relationships with industrialized countries (an example of individual-nation blame, rather than world-system blame). Following this time, a major shift occurred in the conceptualization of development; development today is defined, rather, as a participatory process of social change in a society to bring about both material and social advancement, including greater equality, freedom and other valued qualities (Rogers, 2003). There is a greater concern with equality of the benefits of development towards villagers, the urban poor and women as the main target audiences for development programs. Related to the discussion about equality is the appropriateness of the diffusion paradigm to developing nations (Rogers, 2003; Schumacher, 1973; Yapa, 1991). Because of the capital-intensive, labor-saving, bureaucracy promoting, individual-blaming, and pro- innovation bias of western technology, the transfer of such technologies to developing nations questions is bound to have effects on socio-economic equity. GIS technology has 283 the potential to put old fashioned cartographers out of work, reduce the time to completion of projects, reduce corruption, and improve the urban planning process. This is a mixture of negative and positive consequences which are bound to create socio- economic gaps in society. GIS technology tends to concentrate on large corporate-like organizations like government ministries and large powerful universities, ignoring the smaller institutions. In Uganda, government and donor funding for various Ministries varies, for example, the Ministry of Water and Environment is better funded than the Ministry of Education and Sports (GOU, 2012). This produces a differential impact on society with regard to the diffusion of innovations, and technology transfer. Next, I present an analysis of the state of GIS in Uganda from the five perspectives of GIS and Society. 5.5 The State of GIS in Uganda: A GIS and Society Perspective There are five perspectives of GIS and Society, (1) critical social theory, (2) institutional, (3) legal and ethical, (4) intellectual history, and (5) the public participation perspective (McMaster & Harvey, 2010; Nyerges et al., 2011). In this section, I examine the state of GIS in the public and academic sectors in Uganda, and the impact of GIS on society, through each one of these perspectives. 5.5.1 Critical Social Theory Perspective The critical social theory perspective is concerned with issues of marginalization and empowerment, access, surveillance, and other issues of social equity. It is also concerned with limitations of current GIS representations of populations, resource distribution, culture, and other social aspects. Further, this perspective is concerned with how the 284 evolution of GIS technologies reflects upon the societal structures of those who develop and utilize them, and asks the important question, does GIS benefit society. Another issue of concern within this perspective is GIS and confidentiality in light of the fact that spatial and non-spatial information can easily be linked to identify research subjects. This issue is of prime concern as spatial data becomes more ubiquitous and accessible. (Nyerges et al., 2011) Does GIS Benefit Society? The impact of GIS in Uganda’s public and academic sectors has impacted society at various levels. From the point of service delivery by the public sector, GIS has boosted efficiency in the delivery services by government departments and agencies under various Ministries, for example, the National Water and Sewerage Corporation uses GIS to capture, store and update a digital spatial database of water meters and water lines, and links this data to customer data to allow for efficient billing and infrastructure maintenance. This allows for efficiency in service delivery to customers. In the years prior to GIS introduction at NW&SC, the management of spatial data was done using analog paper cartographic techniques that were slow and laborious. In the environmental sector, the applications of GIS in the management of Uganda’s forest reserves, wetlands, lakes and rivers has led to more effective conservation policies, for example, the first national State of the Environment report by NEIC in 1994 (Gowa, 2009) was a major component in the National Environment Action Plan policy in Uganda in the mid-90s, and was a direct result of the GIS analysis of environmental data by 285 NEIC. The conservation of natural resources is beneficial to farmers, fishermen, and society at large. GIS has allowed some institutions to create new units within public and academic sector institutions creating employment for new GIS professionals. At the same time, GIS technology is seen as a threat to existing jobs in the older traditional fields of paper/pen/ink cartography, which has prevented the diffusion of GIS in some public sector institutions, for example, Kampala City Council. In this case, GIS is seen as an agent that can cause a widening of the socio-economic gap in society. Because of the capital-intensive, labor saving nature of western technology, older professions such as land surveying, cartography, and photogrammetry are under threat. Professionals in these fields must undergo training in GIS so as to secure their jobs in the public sector. The academic sector, on the contrary, views GIS technology as a strength, in that it has led to the creation of new courses in the various academic departments at public universities, for example, GIS courses are taught in over five academic departments at Makerere University, and also at Kyambogo, and Gulu Universities. At Mbarara University, and Gulu universities, it is mainly used in facilitating research activities in the surrounding communities, for example, in the environmental sector in Gulu, and crime analysis in Mabarara district. Privacy and Surveillance The threat of GIS to privacy is not an issue at the moment in Uganda. This is mainly because there are no well-established systems in place to identify citizens at a national level, for example, there is no system similar to the US system of unique social security 286 numbers that can lead to the identification of a given citizen. About 77% of the 33 million people in Uganda (GoU & UNFPA, 2010) live in rural areas (Nyakana et al., 2007). Specific identities of Ugandans are not stored in a sophisticated national database because the infrastructure for such a system does not exist. A pension scheme, similar to the social security system in the United States, is the National Social Security Fund (NSSF). However, NSSF identity numbers are only used to identify working class Ugandans in formal employment ignoring 77% of the population living in rural areas. And even then, NSSF numbers are not available to the public, and are not tied to any geographical data. It is not until recently that the government of Uganda has attempted to introduce a national identity card system, a project which is still in its infancy, and plagued with accusations of mismanagement of tax payers’ money (AllAfrica, 2012). The use of biometric photo and finger print technology was first tried in the mid-2000s (R. G. Smith, 2007) and most recently in the 2011 presidential elections (NTV, 2011), however, no central system of citizen identification has yet been established. There exists no risk to a breach of privacy in the health insurance industry, either. Health insurance in Uganda does not work in the same way that it does in industrialized nations. Most of the 78% of Uganda’s citizens that lie in rural areas are not insured, despite efforts to introduce donor support community health insurance schemes among rural populations (Basaza, Criel, & Van der Stuyft, 2008; Bennet, Creese, & Monasch, 1998; Ekman, 2004; Setel et al., 2007). 287 Due to a combination of factors, privacy and surveillance issues, with regard to geospatial data and society, are not particularly a major concern in Uganda, because the infrastructure to support ICT, and by extension, GIS, is still very poor. These factors include the (1) lack of a street address system, (2) the lack of a national digital cadaster, and (3) poor postal and utility infrastructure. There is no practical street address system in existence in Uganda, although a quasi- functioning “plot number” system tied to the surveyed cadaster is used loosely to reference well known urban locations. According to some authors, only 15% of the total land in Uganda has actually been surveyed (Batungi, 2008) and recorded in the national cadaster. No national cadaster exists in digital form – it all is still in analog form – although there are efforts through a World Bank Funded project, the Land Sector Strategic Plan, to develop a national the Land Information System (MLHUD, 2011a). Postal mail services are mainly through post office boxes located at centralized post offices in the various towns with no possibility for house-to-house delivery. Basic infrastructure such as power lines are only found in urban areas of towns (GoU & UNFPA, 2010; Matthews, 2010). Dialog between various institutional stakeholders about the creation of a National Spatial Data Infrastructure is ongoing, but unfortunately, a consensus on which institution should have the mandate for this has not been reached (Amadra, 2003b; Gowa, 2009; Karatunga, 2002; Lwasa et al., 2006, 2005; Muhwezi, 2006) , as various institutions squabble over anticipated financial benefit upon being selected as the mandated institution. 288 5.5.2 Institutional Perspective The second perspective on GIS and Society is the institutional perspective, which is concerned with the implementation of GIS within institutions (Couclelis et al., 2011; McMaster & Harvey, 2010; Nyerges et al., 2011). Issues addressed within this perspective include, costs of GIS implementation at institutions, equity and distribution of the costs, and benefits of GIS to society, and maintenance of geospatial data by institutions, the impact of GIS on policy decisions and on the interaction between society and government institutions (Nyerges et al., 2011). Costs of GIS Implementation The implementation of GIS at Ugandan institutions, as noted earlier, has been mainly donor-driven project GIS, and the costs of implementation have largely been financed by funding from International Organizations such as UNEP-GRID and development agencies in the United States and Europe. Benefits of GIS to Society The benefits of GIS to Ugandan society have been numerous, both in the academic and public sector. GIS has promoted better service delivery and management of utility services like piped water by the NW≻ improved environmental management by NEMA, NFA, and the Wetlands Management Department (Fuller et al., 1998; Musinguzi, 2003b, 2003c; UgandaWetlands, UBOS, ILRI, & WRI, 2009); and made digital data available to public and academic sector institutions for research. GIS has played a major role in planning peace building and development efforts in northern Uganda after twenty years of the Joseph Kony insurgency, for example, 289 initiatives such the USAID-supported SPRING project in northern Uganda (EMG, AVSI, Agrisystems, & Foundation, 2008; USAID, 2010) that uses MapInfo to store locations of food storage facilities. Another example of an institution using GIS for peace, recovery and development in northern Uganda is the Northern Uganda Data Center (NUDC), under the Office of the Prime Minister. NUDC, in liaison with local district planning offices has equipped local planners, surveyors and decision makers in the various districts in northern Uganda with GIS software, equipment, and provided training in GIS to build capacity in GIS proficiency. Maintenance of GIS Data by Institutions The CAMPUS project under the Department of Surveys and Mapping has also created a digital database mapping of major public infrastructures in Uganda, for example, schools, health centers, district offices, water points, and so on. This data is available to the public for purchase. However, there is a duplication of efforts in data collection between NUDC and CAMPUS which is an indicator a lack of cooperation among certain institutions. This finding is consistent with other authors’(Amadra, 2003a; Ernest, 2010; Muhwezi, 2005; Nyemera, 2008). The lack of a clear mandate for an NSDI, and absence of a legal framework regarding geoinformation policy is at the root of this data sharing problem (Kalande & Ondulo, 2006; Musinguzi et al., 2004; Ofori-amoah, 2010; Uwayezu, 2010). Society’s Access to GIS Society’s access to GIS technology and geoinformation products, such as maps is minimal. During this research, there was no evidence of community-based mapping centers, similar to the ones described by Leitner et al. (1998) that empower local citizens 290 with spatial analytical skills and decision making tools to better their livelihoods. As a result, a significant power differential exists between the GIS expert and the average citizen in the various institutions in Uganda with regard to GIS technology. An important role is played by so-called Internet cafes, public access points for Internet services in rural and urban areas. Access to the World Wide Web has, at least, had a noticeable impact of the interest of society in maps, and spatial thinking. The use of Web maps such as Google Maps and Google Earth is reasonably widespread, mainly for informational purposes. These Web applications, unlike in the industrialized nations, are not used for routing as no street address system exists in Uganda. However, directions based on land marks identifiable on Web maps serve similar purposes. Research is needed on the reasons for the lack of a general street address system in Uganda. Is GIS Appropriate Technology? One of the major challenges that GIS faces in Uganda is the project-driven nature of GIS technology. GIS technology diffused into Uganda as a result of international donor funded projects at the Ministry of Environment Protection, and the Ministry of Lands and Housing in the 90s. Unfortunately, after the project funding ended, many of the GIS departments that had been set up within the various departments within these ministries struggled to maintain software licenses, and hardware equipment. Thus, questions were raised about the appropriateness of GIS technology to developing countries like Uganda that could not necessarily afford to maintain the technology in the absence of donor support. 291 However, the real problem here is the lack of adequate support from the central government for GIS and mapping in Uganda. Annual budgetary support for public sector institutions hosting GIS technology is less than adequate. This means a constraint on the little available resources, both financial and human, to support GIS technology in the public and academic sectors. For example, the Ministry of Agriculture, Animal industries and Fisheries has only one employee that uses GIS; the Department of Surveys and Mapping’s CAMPUS project is severely understaffed with skilled personnel for GIS work, and yet are unable to hire new employees due to excessive bureaucracy and lack of interest in the department’s activities in the Ministry of public service, and the Ministry of Finance. 5.5.3 Legal and Ethical Perspective The third perspective on GIS and Society is the legal and ethical perspective, which is concerned with pricing mechanisms governing access to spatial data, the institutional processes that lead to these, proprietary spatial databases, ethics in GIS and legal aspects of GIS (Nyerges et al., 2011). Of particular concern is the ownership and accuracy of geospatial data, ethical debates surrounding creation, digital representation, and use of the data. Pricing Mechanisms and Institutional Processes The right of institutions to capture, store, and distribute certain types of geospatial data in Uganda are driven by specific mandates awarded by the central government of Uganda to specific institutions. For example, ownership rights to topographic and geopolitical spatial data of Uganda is under the official government mandate of the Department of 292 Surveys and Mapping, Ministry Lands, Housing and Urban Development, while that for Forest data is under the mandate of the NFA. Ownership and Accuracy of Data It follows that there is a loose understanding of data ownership at the institutional level, but no official government data sharing policy among institutions (Chaminama, 2009; Kalande & Ondulo, 2006; Schwabe, 2010; Tukugize, 2005), although there is indication of some efforts towards this objective (GSDI, 2011). This problem is further complicated by a lack of policy on geodata capture standards and metadata, which has implications for data interoperability (Musinguzi et al., 2010; Woldai, 2002). 5.5.4 Intellectual History Perspective The fourth perspective on GIS and Society is the intellectual history perspective. This perspective is concerned with tracing and understanding the dynamics through which dominant GIS technologies are selected out of a set of possible options, and how the processes leading to these selections are linked to institutional, governmental and personal influences (McMaster & Harvey, 2010; Nyerges et al., 2011). It further examines the adopted technologies in light of other promising technologies that were overlooked. Public Sector In the case of Uganda, the introduction of GIS into the public sector was through international-donor-supported projects. UNEP-GRID, NORAD and the World Bank were the first change agents to introduce GIS into the country, in the form of Idrisi and PC ARC/INFO. 293 Today, the most pervasive GIS software technology is ArcView GIS and ArcGIS from ESRI. The intellectual history behind this trend has to do with the fact that ESRI and UNEP had formed a strategic partnership in the early-to-mid-80s (Mooneyhan, 1998). UNEP’s primary concern in the mid-80s was to foster better environmental management practices in developing countries in light of the objectives of the 1972 conference on the Human Environment in Stockholm (UNEP, 1972). Much as Idrisi software from Clark Labs, Clark University, was also introduced into Uganda around the same time as PC ARC/INFO, for example in the Department of Forestry, and later, MapInfo in the Department of Geological Survey and Mines, users in public sector institutions in Uganda developed a preference for ArcView, and later ArcGIS, over time. This was mainly because of the user friendliness and a user-interface design of ESRI software. Another reason for the success of ESRI software is tied to the special relationship between ESRI and the International Institute for Geo-information Science and Earth Observation (ITC), Netherlands. ITC was the main university at which Ugandans in GIS, Cartography and Photogrammetry in the 80s and 90s, the decades in which GIS started diffusing into developing countries around the world. This university used a range of GIS software in its lecture rooms in the early 90s, including ESRI’s PC ARC/INFO (Toppen, 1991), and the university’s own proprietary software, ILWIS; however, in the mid-90s, there was a shift toward the ESRI software, especially after the release of ArcView, a revolutionary GIS software in terms of the Windows-based user interface design in the 294 90s. This directly led to preference for ESRI software at various public sector institutions in Uganda today. Academic sector Uganda’s educational sector also experienced a diffusion of GIS technology into its institutional framework and curriculum in the early to mid-90s, however, unlike in the public sector, this diffusion was not directly supported by International donor funding; instead, it was indirectly supported by international donor funding through initial funding provided to public sector institutions. Public sector agencies used their alliances with related academic sector institutions in the same field to introduce GIS technology at the centers for higher learning and education in the country. First of these was MUIENR in the early 90s, at Makerere University, which was benefactor of support from NEIC, which in turn benefitted from donor funding from the World Bank, and UNEP-GRID. Other academic institutions followed suit such as the Department of Surveying, Department of Forestry, and the Department of Geography, at Makerere University. At all of these academic departments, ESRI software has become the GIS software of preference, just like in the public sector. The curriculum at several university departments includes courses on the principles, and applications of GIS, and their foci vary according to the departmental foci. For example, the Department of Forestry focuses on the forestry applications of GIS, while the Department of Information Systems focuses on Web applications and open source GIS. These courses are taught mainly in undergraduate programs, although a few departments offer GIS courses at the master’s level as well. Because of the historical link between the 295 GIS diffusion and donor support in the public sector, GIS in the academic sector suffers from similar symptoms of donor dependency in terms of software license renewal and hardware maintenance. In fact, most departments at Makerere University, Kyambogo university, Mbarara University, and Gulu University heavily rely on donor funding to support the continuity of GIS usage in their departments, for example, the Faculty of Computing and Informatics Technology, Makerere University benefitted from a €5 million grant from NORAD to upgrade its ICT infrastructure, part of which was used for the purchase of about 2000 new computers and software (MUK-FCIT, 2009). Because of the preference for ESRI software, training in GIS technology is mainly linked to ESRI software. Outside of Academia, the main training institution within the boundaries of Uganda is GeoInformation Communication (GIC), the official ESRI value- added services partner in Uganda. Public sector employees also train at regional training centers, for example, the Regional Center for Mapping, Nairobi; ESRI Eastern Africa; and Oakar Services, Nairobi. European academic and research institutions that have trained the bulk of GIS analysts from Ugandan public academic sector institutions include ITC, Netherlands; and, the Royal Museum for Central Africa, Belgium. The latter of these institutions was the entry-point for MapInfo software into Uganda, much as this GIS software failed to succeed in the Ugandan market. One of biggest challenges facing GIS in Uganda today is the loss of skilled GIS personnel to industrialized countries (Mooneyhan, 1998). Often times, Ugandan GIS students in Europe and North America seek greener pastures after completing their studies abroad, and find better paying jobs by several magnitudes compared to their own 296 country. This is problem facing various sectors in the third world, and needs to be addressed urgently to avoid the impact of a paralyzing “brain drain” on fragile economies leading to a shortage of skilled GIS personnel. Low salaries in the public sector in Uganda are common place, and bureaucracy levels too high to support the institutionalization of GIS in the public sector, which directly affects the diffusion of GIS technologies according to some authors (see (Rumor, 1993; Wegener & Junius, 1993). Many GIS departments are understaffed and underfunded by the government; hiring of new GIS staff is an uphill task due to excessive bureaucracy in the public service hiring process, and a lack of funding; not to mention, high levels of corruption (Transparency International, 2011) and mismanagement of public funds that could be better used to improve infrastructure in the various public sector institutions, including ICT and GIS infrastructure. 5.5.5 Public Participation Perspective The fifth perspective on GIS and society is the public participation perspective. This is concerned with the involvement of community and grass roots groups in the use and application of GIS so as to empower them. It also addresses the cultivation of local and indigenous knowledge into GIS representations of space and place. Further, it examines the relationships between community groups and institutions of power at various scales, for example, at the municipal, state, and national levels. There was no evidence of significant community participation in GIS and mapping activities in the majority of the public and academic sector institutions that were considered in this research. Some involvement of local citizens in mapping activities goes 297 as far as helping reconnaissance teams from the Department of Geological Surveys and Mines, the Department of Surveys and Mapping, and the CAMPUS mapping project during in the course of their mapping exercises to identify real world geographic features of interest to be included in an updated map. This can be considered as inclusion of indigenous and local geographical knowledge into the final GIS output. Next, I provide five examples of involvement of local citizens in mapping at Ugandan institutions, that is, at (1) the NFA, (2) NARO, (3) NUDC, (4) Department of Petroleum Exploration and Production, and (5) the Local Government in Hoima town. NFA The National Forestry Authority (NFA) has two units within the agency that regularly interacts with local communities in the management of forests in Uganda, especially private-owned forests. These are the Advisory Services and Community Partnerships units. The former is a public relations unit, while the latter educates the public on issues of forest management. Maps are used to aid interactions with the community members. Owners of private forests are provided with maps of their forested lands by forestry officers from NFA so as to help them manage their forests. NARO The National Agricultural Research Laboratories/Organization (NARL/NARO) interacts with members of local communities to identify new geographic features of interest on aerial photography. These newly identified features are then incorporated into a GIS database. In this case, indigenous and local knowledge is incorporated into the GIS. In some instances, projects undertaken by NARL/NARO specifically call for public 298 participation, and such cases, local community members walk together with scientists from NARL/NARO along transects to identify features of interest, in line with the objectives of the research project, to include in a map update. NUDC The Northern Uganda Data Center (NUDC) has made efforts to include views from local citizens in northern Uganda to inform the planning process of peace, recovery and development in that part of the country. NUDC GIS and mapping experts organize meetings with local community members in northern Uganda to identify localities on maps, based on boundaries of counties, and smaller political units called parishes, as priority areas for development projects. For example, local villagers identify counties and parishes on a map where schools and health centers should be built first by the government of Uganda. These preferences are noted in the field and stored in the GIS database by NUDC. Often times, projects are implemented by the government of Uganda at the locations specified by the local people. Petroleum Exploration The Department of Petroleum Exploration and Production uses GIS to manage land parcels affected by oil exploration activities in western Uganda. Sensitization of the local population about oil exploration and mining activities around their lands is done by the local government officials, Local Council Levels I-V, in the districts where such prospecting for oil is planned. In the event that oil is discovered below the land, the land owner’s compensation is processed accordingly with the help of a GIS to store the location of the affected land parcels, and compute the acreage of land affected. 299 Civic Engagement by Local Government There is some commitment by the government of Uganda to solicit views from local citizens so as to actively involve them in decision making, and development project planning. For example, in a recent development project to build markets in Hoima town in western Uganda, sensitization of local citizens by local government authorities was done so as to inform them about the positive and negative effects of the project (M. Musinguzi, Department of Surveying, Makerere University, personal communication, June 3 rd , 2010). Such sensitization would be even more effective if a GIS component was included in the sensitization exercise. GIS can play an active role if it is used in the public participatory model to empower citizens; however, such interventions involving public participatory GIS methods have not been applied in any significant fashion in Uganda. 5.6 Conclusions GIS technology was introduced into Uganda’s public and academic sector institutions two decades ago. The diffusion of the technology has been relatively slow as compared to its diffusion in industrialized nations like the United States, and the UK. The most outstanding difference between the two cases of diffusion has been the fact that GIS diffusion in the developing countries was driven by international donor funded projects, while in the developed world, by market forces. There has been some impact of GIS on society in Uganda, however, only indirectly through a few public and academic sector institutions. The current state of GIS in Uganda can be characterized by a dependency on donor funding to sustain GIS licenses, a lack of awareness about the technology and its benefits among high ranking government officials, inadequate support for GIS activities 300 in the public sector, a lack of geo-information policy, and the absence of any form of spatial data infrastructure. However, despite all the problems, there a resilient group of GIS specialist, public servants and academicians, who optimistically advance an agenda for the recognition of GIS as a useful technology for development, and as an academic field that holds immense promise for young Ugandans at institutions of higher learning. 301 Chapter 6 : Diffusion of GIS in Developing Countries - A Case study of Uganda’s Non-governmental Organization, International Organization and Private Sectors 6.1 Introduction There is a growing interest in the diffusion of GIS technology in developing countries because of the important role played by GIS in planning in various sectors of the economy. Non-governmental organizations (NGOs), International Organizations (IOs), and the private sector have adopted GIS at various rates in developing countries yet little has been published on the current state of GIS in these sectors, particularly in developing nations. What is the current state of GIS in the NGO, IO and private sectors in developing nations? This research is based on a case study of Uganda. Diffusion of innovations theory (Rogers, 1993, 1995, 2003) is used as a theoretical framework to investigate the diffusion and state of GIS in the three sectors in Uganda. The state of GIS is inferred from indicators such as the level of penetration of GIS technology (the innovation), the communication channels through which diffusion is occurring, the time to adoption or rejection, and the influences of the social system on GIS adoption. Further, the GIS and Society framework is employed to investigate the impacts of GIS on Society. There are five perspectives of GIS and Society: (1) critical social theory, (2) 302 legal and ethical, (3) institutional, (4) legal and ethical, and (5) the intellectual history perspective (McMaster & Harvey, 2010; Nyerges et al., 2011). Each one of these is examined with respect to the impact of GIS in the NGO, IO and private sectors in Uganda on Society. The findings of this research are significant for future implementation of GIS projects by the NGO, IO and private sectors in developing countries, and also, for international donor agencies funding development projects, such as the United States Agency for International Development (USAID). In the next section, each of the three sectors, NGO, IO and private is defined formally, and literature relevant to GIS diffusion is reviewed. 6.2 The NGO, IO and Private Sectors GIS diffusion in the NGO, IO and private sectors has taken place at different rates, and undergone various social influences, both in the developed and developing nations. This section reviews the literature on GIS diffusion in each of the three sectors with relevance to developed and developing countries, respectively. 6.2.1 NGO Sector A non-governmental organization (NGO) is defined as “any non-profit, voluntary citizens' group which is organized on a local, national or international level” (NGN, 2012). An NGO is task-oriented and driven by people with a common interest, and performs a variety of service and humanitarian functions including bringing citizen concerns to Governments, advocating and monitoring policies, and encouraging political participation through provision of information. Some NGOs are organized around 303 specific issues, such as human rights, environmental or health issues. NGOs also provide analysis and expertise, serve as early warning mechanisms and help monitor and implement international agreements. (NGN, 2012) The definition of NGOs is adopted to refer to non-profit organizations, community-based organizations (CBOs), and grass roots organizations (GROs). 6.2.1.1 Developed Countries In industrialized nations, GIS has been found to be a very useful tool for NGOs, or non- profit organizations, for example, in France where the use of GIS by community based organizations has encouraged local communities to think about and further explore issues related to the environment (Roche & Humeau, 1999). In a study of GIS implementation by grassroots organizations in northern California, Sieber (1993) found that the process of implementation differs considerably from public sector agencies because GROs are more fragile in nature due to the inability to attract and retain resources, and they lack the capacity to hold together a loosely knit network of individuals with diverse goals and strategies, many of whom work on voluntary basis with little commitment to their work. Beyond commitment, quality and continuation of work are the most important factors for effective GIS work (Sieber, 1993). Further, GROs do not work with official mandates, and are not necessarily required to provide public accountability. Their organizational structure loose, that is, participatory, decentralized and informal, and their organizational culture, receptive to innovations. This is different from the centralized and bureaucratic organizational structure in public sector organizations which are obstacles to adoption. However, some GROs can exhibit 304 organizational culture that is antithetical to adoption of new technologies; for example, environmental GROs may associate technology with confining the environment to an exploitable resource, and thus, reject technological innovations. (Sieber, 1993) All these factors notwithstanding, Sieber (1993) concluded that GROs organizationally, GROs do not follow traditional factors of public sector organizations with regard to GIS implementation; instead, they substitute resources and employ different implementation models that suit each organizational culture, while minimizing the impacts. It follows that GIS technology is, in fact, within reach of these organizations. The use of GIS in community-based organizations has been investigated by some (S. Elwood, 2006a; S. A. Elwood, 2002; Harris & Weiner, 1996; Leitner et al., 1998; McMaster et al., 1997). This research, grounded in GIS and Society literature, is relevant to understanding the diffusion of GIS in non-governmental organizations. GIS can either empower or marginalize communities (Leitner et al., 1998; McMaster et al., 1997). GIS empowers decision makers in ways that can lead to environmental injustice to communities (McMaster et al., 1997), marginalizing them (S. A. Elwood, 2002). To make GIS more accessible to community-based organizations, alternative models have been proposed by Leitner et al. (1998) including community-based (in-house) GIS facilities, university/community partnerships, “map rooms”, Internet map servers, and neighborhood GIS centers. GIS in the hands of community based organizations has the potential to empower communities in creating spatial narratives against undesired urban planning projects that threaten livelihoods, and give local communities a stronger voice in the local politics of their neighborhoods, as shown in the research by Elwood (2006a) 305 and Elwood (2006b) in urban planning and community-based activism in Chicago. GIS diffusion in developing countries is reviewed next. 6.2.1.2 Developing Countries Literature on GIS diffusion in developing countries reveals that there are similar challenges in various parts of the world, for example, in Eastern Europe and the former USSR (Brunn et al., 1998), China (Shupeng, 1987; Yue et al., 1991), Indonesia (Sipe & Dale, 2003), Colombia, Brazil and Peru (Gibson, 1998). These challenges include technological challenges with respect to software and hardware maintenance, and internationalization (translation into local languages); data accuracy, accessibility and standards; and institutional challenges. In developing countries, GIS implementation, and by extension, GIS diffusion, has often been explained within the framework of a factors approach or a process approach, but Ramasubramanian (1999) argues that implementation should be assessed within the context of the policies that necessitated the introduction of GIS. “Any assessment of GIS implementation in developing countries must acknowledge the uniqueness of country- specific problems” (Ramasubramanian 1999, 378). Four characteristics important to the implementation of GIS in developing countries identified by Ramasubramanian (1999) include: (1) clarity in problem definition, (2) forging strategic alliances, (3) incremental planning, and (4) building local knowledge. The importance of indigenous knowledge in the success of technology diffusion and transfer to developing countries has long been ignored by western scholars (Puri & Sahay, 2004). It was not until the Rio Earth Summit of 1992 that the role of indigenous knowledge began to be recognized in sustainable 306 development (NRC, 2002; Puri & Sahay, 2004). Local knowledge is useful in the reinvention of GIS technology (Rogers, 1993, 2003) to meet the needs of indigenous communities in developing countries (Kyem, 2000; Kyem & Saku, 2009; Laituri, 2011; Rundstrom, 1995). NGOs provide a link between indigenous communities and GIS technology. In an Actor-Network Theory- (ANT) informed case study of GIS-based natural resource management by a local NGO in northern Namibia, Noongo (2007) found that the NGO’s intervention led to a more participatory, transparent and efficient decision making process in natural resource management in the region. The collaborative use GIS technology addressed user needs, facilitated the building of trust in the community, improved data collection, quality, and sustainability, and fostered creativity. Challenges included limited GIS software and hardware resources, ArcView GIS software and the same computer hardware still being used ten years after the initial donation of the technology in the mid- 90s by Environmental Systems Research Institute (ESRI), a lack of communication and coordination among the various non-governmental organizations in Namibia, a lack of sufficient funding, and a lack of staff training. These factors led to the slow implementation of GIS within the organization. The role played by the executive manager, as a champion of the diffusion process, cannot be ignored. Further, a significant role was played by a visiting scholar from the United States, a PhD student in the adoption of GIS technology within the NGO as a heterophilous change agent. Next, I review the GIS diffusion literature in the IO sector in developing countries. For obvious reasons, no literature review is provided with respect to GIS diffusion in 307 developed countries because, by definition, IO activities are limited to developing countries. 6.2.2 IO Sector By definition, International organizations (IOs) are “entities established by formal political agreements between their members that have the status of international treaties; their existence is recognized by law in their member countries; they are not treated as resident institutional units of the countries in which they are located” (OECD, 2012). Examples of IOs include the agencies of the United Nations, such as the Office for the Coordination of Humanitarian Affairs (OCHA), the United Nations Environmental Programme (UNEP), and the United Nations Economic Commission for Africa; financial IOs such as the International Monetary Fund (IMF), and the World Bank Group; regional IOs such as the African Union, and the Organization for Economic Co-operation and Development (OECD). Government agencies responsible for international development, such as United States Agency for International Development (USAID), and the Norwegian Agency for Development Cooperation (NORAD), are considered IOs in this paper, although they are agencies created by their parent governments, not through international treaties as is the case with IOs. For example, “USAID is an independent federal government agency responsible for providing economic and humanitarian assistance around the globe” (USAID, 2012); “The Norwegian Agency for Development Cooperation (NORAD) is a directorate under the Norwegian Ministry of Foreign Affairs (MFA)” (NORAD, 2012). 308 6.2.2.1 Developing Countries IOs have played a primary role as change agents in GIS diffusion in developing countries. The introduction of GIS in developing countries is attributed to the activities of UNEP (Cartwright, 1993; Mooneyhan, 1998), through the Global Resource Information Database (GRID) and Global Environmental Monitoring System (GEMS) projects (Mooneyhan, 1998). IOs are involved directly or indirectly in the implementation of donor funded projects in developing countries, for example, USAID- and UN agency- supported projects in Uganda (HPDMH et al., 2009; UNEP-DEWA, 2007; USAID, 2010); and European Space Agency- (ESA) supported projects in Zambia (Mwape 2010, 19). Often times, GIS is a component of the project so as to perform geospatial tasks effectively to meet the goals and objectives set out in the project. Thus, IOs act as change agents and provide heterophilous communication channels for the diffusion of GIS through training and capacity building of local staff in host countries, for example, the training in GIS provided through the United Nations Institute for Training and Research (UNITAR) in the mid-80s in various developing countries (Mooneyhan, 1998). After the earthquake in Haiti, several IOs, including UN agencies and US government agencies were involved in disaster relief operations in Haiti. The main challenge to GIS analysis was the lack of geospatial data for Haiti, however, through crowd sourcing, a novel solution to the data problem was found (Zook, Graham, Shelton, & Gorman, 2010). Lack of data standards is also a challenge facing IOs in the implementation of information systems projects in developing countries, for example, health information 309 systems in Botswana, Ethiopia, Thailand and South Africa (Braa, Mohamed, Hanseth, Shaw, & Heywood, 2007; Braa, Monteiro, & Sahay, 2004; Braa & Muquinge, 2004). International organizations have been at the root of diffusion of GIS technology in most developing countries, however, there some authors call for less dependence on this IO- support and a move towards support through local and regional collaboration among countries (Cavric et al., 2003). Next, I discuss GIS diffusion in the private sector of developed and developing countries. 6.2.3 Private Sector By definition, the private is “the part of the economy that is not state controlled, and is run by individuals and companies for profit. The private sector encompasses all for-profit businesses that are not owned or operated by the government” (Investopedia, 2012). Because of the multi-disciplinary nature of GIS, applications of the technology can be found in utility companies, such as gas and electric companies, water and sewerage companies; the telecommunications industry, such as cell phone and landline service providers; in retail and wholesale companies, such as large supermarkets; and in insurance companies. GIS consultancy work is often a source of employment for GIS analysts, providing contract-based services to government agencies, for example, in the United States; the Dewberry Corporation provides GIS consultancy services to the US government. 6.2.3.1 Developed Countries In the United States, GIS created a new niche in the private sector starting in the early 1970s. Private sector GIS consisted of GIS software companies, GIS consultancy 310 companies, and customers, mainly academic departments, government agencies, and international organizations. One of the earliest commercial GIS software companies was ESRI, which was initially registered as a non-profit organization in 1969 (Dangermond & Smith, 1988). ESRI started producing GIS software for sale to customers in the early 1970s. Early GIS software was written by ESRI in FORTRAN, and this included AutoMap II and GRID. However, it was ARC/IFNO that created the greatest customer demand for GIS software in the mid-1980s. “Yet, by the time the mid-1980s arrived, potential users came pounding at ESRI's door. It rapidly became difficult just to keep up with requests for information and advice. As the decade passed, instead of having to press GIS solutions on unaware and often unwilling potential users, user organizations and the Requests for Proposals (RFPs) they wrote began to specify GIS technology as the required means of meeting certain objectives. By the mid-1980s this had gone so far that there was a steady inflow of RFPs all calling for fairly standard turnkey GISs, which were, by then, becoming ‘off the shelf items’” (Dangermond and Smith 1988, 305). A major driver of GIS applications in the private sector in the 1980s was the increase in awareness about environmental degradation and sustainability in the 70s (Dangermond & Smith, 1988). As a result of this awareness, there was a passage of a series of laws requiring public and private projects to perform careful analysis of environmental impacts. A need for consultancy services emerged for environmental impact assessments with regard to urban housing and construction, waste water management, biodiversity monitoring, flood plain monitoring, forest conservation and so on. ESRI was one of the companies that provided these consultancy services using GIS. The main clients for such 311 services were government agencies such as the Department of Housing and Urban Development, the California state waste water management agency, the Corps of Engineers, the Fish and Wildlife service, and so on. All the projects were flowing directly from the new regulatory requirements, and cities, counties, and states were being considered as units of project implementation. Because of the sheer size of data involved, agencies started considering automation as opposed to older manual methods (Dangermond & Smith, 1988). Another major driver of GIS in the private sector was the invention of the PC. The reason for the success of PC-based GIS lies in the fact that smaller user organizations were now able to afford GIS technology, which was not the case with the older main frame computers of the previous decade. As a result of PC-based systems, GIS found business applications in the late 80s in businesses as diverse as real estate development, recreation and vacation planning, and car navigation. (Dangermond & Smith, 1988) A case for GIS consultancy in the private sector was also created by the need for customized GIS software for various public sector agencies. For example, ESRI developed customized software for the Corps of Engineers, a flood plain information system for estimating flood damage. Utility, energy companies, and other private sector organizations all had specific needs with regard to meeting the new regulatory requirements, and also to automate their own workflows. This created a market for GIS software customization as a professional service in the private sector (Dangermond & Smith, 1988). 312 Interest in GIS and awareness in the private sector was promoted greatly by the increase in importance of GIS in the academic sector, with an increasing number of journals in the late 80s devoted to GIS as a an academic field of study (Dangermond & Smith, 1988). Other factors that promoted the diffusion of GIS was the increasing user-friendliness of GIS software, the decreasing costs of acquiring the software, and the increasing reliability and integrity of the software. Another factor that led to the promotion of GIS technology in the private sector was support from the military as nations tried to maintain a strategic advantage in technology over their adversaries. One of the challenges to GIS in the US in the 80s was the lack of spatial analytical and modeling tools in GIS (Dangermond & Smith, 1988; Goodchild, 1987; Tomlinson, 1987, 1988). There was a need for these tools to be developed to address important modeling problems in fields such as climatology, environmental ecology, and public health. Another challenge to the development of GIS in the private sector in the US was poor sharing of geographic data by government agencies and the private sector (Dangermond, 1995). Institutional inertia was singled out as one of the causes of this tendency of public agencies not to be interested in sharing their data: “Public organizations, in particular, have mandates that they are required to fulfill, and they have little interest in bending to accommodate the needs to others” (Craig 1995, 115). However, the public does have access to certain public databases that contain matters of public record, some census data, property data, building permit data, birth and death records, property transfers, some tax records, and so on (Dangermond, 1995). Most GIS data used by the public come from 313 government sources, although some private sector organizations are increasingly repackaging GIS data on its way to GIS users (Dangermond, 1995). Other issues regarding data access in the developed world include government legislation and policy, the impact of the internet, intellectual property rights, liability issues, privacy issues, the pricing of geospatial public sector data, and the tragedy of the information commons (Kennedy, 1998; H. J. Onsrud, 1998). The issue of greatest contention with regard to public sector data access by the private sector is pricing (Kennedy, 1998). The private sector in the developed world is a strong critic of the cost-recovery models employed by public sector agencies in North America, Europe and Australia. There are various arguments advanced by the private in support of this, for example, the public sector should not charge for geospatial data because these data have already been paid for, and should not be charged for a second time; high data prices dissuade the private sector from investing in new and innovating developments; and treating data as a commodity puts public sector agencies into unfair completion with the private sector (Kennedy, 1998). According to some scholars, the role of private sector in the new information age is to create value-added products. This should not be government’s role because this insinuates unfair completion and “the private sector is better equipped to fully develop the commercial of value added products” (Kennedy 1998, 136). A win-win strategy is the public-private sector partnership model; an example of this is in Ontario, Canada, where the provincial government has gone into a long-term partnership with a private sector company to automate the province’s land registry system and operate an information 314 utility to disseminate the land records data. The resulting company, Teranet Land Information Services Inc. is jointly owned by the government of Canada and a group of private sector investors (Kennedy, 1998). Issues surrounding data access has led to a whole new discourse on spatial data infrastructure (SDI), which is important for the private sector involvement in GIS (Chan & Whitworth, 2003; Masser, 2011; Rajabifard, Binns, Masser, & Williamson, 2006). The creation of people-relevant data will encourage private sector involvement in SDI development (Rajabifard et al., 2006). 6.2.3.2 Developing Countries In the early 90s, the diffusion of information systems for urban management and planning in developing countries was only just starting, however, it was very slow (Cartwright, 1991), and this was due to myriad reasons, beyond just technological hindrances: lack of sufficient funds to acquire information systems, lack of inter-agency co-operation, data collection challenges, and high costs of training and support. There was little to no private sector involvement in GIS at that time because it was too early in the diffusion stage. Private sector support for GIS in developing countries is critical to the diffusion of GIS in the developing world, especially with regard to the development of geoinformation for continents like Africa (Schwabe, 2005). Public participation in GIS, and public-private partnerships are crucial to the successful development of a geoinformation industry in developing countries (Gyamfi-Aidoo, Schwabe, & Govender, 2005; Schwabe, 2005, 2010). 315 Challenges to the GIS in the private sector in developing countries echo the findings of several researchers, and these include, a lack of funding, data sharing, awareness, capacity building, and geoinformation policy (Kalande & Ondulo, 2006; Lwasa et al., 2005; Nasirumbi, 2006; Nyemera, 2008; Onah, 2009; Tukugize, 2004, 2005; U-Consult, 2004). There is a need for public private partnerships for successful implementation of GIS and SDI in Uganda (Lwasa et al., 2005; Nasirumbi, 2006). Next, the results and findings of this mixed methods research are presented for the NGO, IO and private sectors in Uganda, respectively. 6.3 Results Results of mixed methods analysis of institutions in the (1) NGO, (2) IO and (3) private sectors are presented in this section. The findings about the current state of GIS in each of the three sectors are presented under the following categorical issues: (1) early GIS introduction into the institutions, (2) GIS software and hardware usage, (3) applications of GIS, (4) GIS data collection, sources and sharing (5) training and education in GIS, and (6) the challenges to the diffusion of GIS. 6.3.1 NGO Sector The NGOs that accepted the invitation to participate in interviews for this research project are shown in Table 6-1. These NGOs’ focus areas include environmental conservation, food security monitoring, biodiversity conservation, improving livelihoods of the poor, and public health. Six of these NGOs are located in Kampala city, and one in Gulu district. As can be seen in Table 6-2, only 3 out 7 of the NGOs use GIS in their daily workflows. 316 6.3.1.1 Early GIS Introduction, Project-driven GIS and Donor Funding The earliest case of GIS introduction in Uganda’s NGO sector was at FEWSNET in 1996. At the other NGOs, GIS was introduced in the mid-to-late 2000s (see Table 6-2). GIS came into Uganda’s NGO sector (1) through International Organizations, and (2) through personal initiatives of employees. It was through USAID project funding that ArcView 3.x and ArcGIS 9.x were introduced at FEWSNET and NUMAT. At Nature Uganda, a personal initiative by one of the employees led to the introduction of ArcView 3.x, inspired by a Master’s level course in GIS at the University of Cape Town, South Africa. The activities of many of the NGOs in Uganda are sponsored by international donors. A respondent at Uganda Land Alliance stated, “We have a number of donors. But in the Karamoja program, we have two particular ones: Ford Foundation and DanChurchAid (a Danish religious organization)” (J. Mwebe, Uganda Land Alliance, personal communication, 01 July 2010). At NUMAT, a USAID-funded project in northern Uganda, GIS was introduced through the support and recommendation of USAID personnel based at the USAID headquarters in Kampala. 317 Table 6-1: NGO sector institutions at which interviews were performed ID Institution Name Area of Expertise 1 Famine Early Warning System Network (FEWSNET), Kampala Food security monitoring 2 Nature Uganda, Kampala Wildlife Conservation, especially birds 3 Advocates Coalition for Development and Environment (ACODE ), Kampala Environmental conservation 4 FXB Uganda Program, Kampala Livelihoods of the poor 5 Uganda Land Alliance, Kampala Customary land ownership issues and community mapping 6 National Association of Professional Environmentalists (NAPE), Kampala Environmental conservation 7 Northern Uganda Malaria, AIDS, and Tuberculosis (NUMAT) Program, Gulu district Public health 6.3.1.2 GIS Software and Hardware Usage ArcView 3.x and ArcGIS 9.x are the only types of GIS software being used by the 3 NGOs in the sample interviewed. The hardware used consists of personal computers. The use of plotters for printing maps was only evident at FEWSNET. None of the NGOs interviewed use GPS devices for data collection. Table 6-2: GIS usage at NGO sector institutions ID GIS Used Year GIS Introduced ArcView 3.X ArcGIS 9.X 1 1 1996 1 1 2 1 2007 1 3 4 5 6 7 1 2009 1 Total No. 3 2 2 Total %age 43 29 29 6.3.1.3 Applications of GIS Nature Uganda uses GIS for biodiversity monitoring; specifically focusing on the conservation of birds in Uganda. FEWSNET uses GIS for monitoring food security 318 conditions in drought prone areas of Uganda, for example, northern and north-eastern Uganda. NUMAT employs GIS in monitoring the spread and transmission of Malaria, HIV/AIDS, and Tuberculosis from a public health standpoint. FEWSNET combines a number of datasets to do predictive modeling of food security conditions for Uganda, with a special focus on drought-prone areas. Datasets used include, climate data, food price data, nutrition data (from the Ministry of Health, UNICEF, World Food Program), data on political stability, and so on. The combination of these datasets leads to a 6-month prediction on food security in the country. However, the modeling methods are based on subjective interpretations and somewhat arbitrary – based on discussions and subjective classifications. However, the food security classification is in line with the integrated food security phase classification (IPC) scheme of the Food and Agricultural Organization, according to a respondent at FEWSNET. Nature Uganda is part of the East African Natural History Society formed in 1909 by a group of scientists from Great Britain during the colonial era, and its focus is biodiversity conservation and sustainable natural resource management (NatureUganda, 2012). Nature Uganda works mainly on the conservation of birds in general, and water birds, for example the fish eagle, in particular. ArcView GIS for mapping of bird conservation and environmentally degraded sites, around Uganda with the aim of supporting proposal writing for new conservation projects. The activities at this NGO are guided by academics and highly qualified environmentalists. The NGO works hand-in-hand with public sector agencies involved in wildlife conservation, for example, the Uganda Wildlife Authority (UWA). 319 Applications of GIS are mainly at the level of creating maps for presentation and inventorying. There was no evidence of sophisticated GIS analysis or geoprocessing at any of the NGOs, apart from FEWSNET. This could be because of a lack of sufficient knowledge about advanced GIS functionality owing to a lack of adequate training and exposure to GIS functionality. 6.3.1.4 GIS Data Collection, Data Sources and Data Sharing The three NGOs rely on public sector agencies for their data needs, such as NFA, the Ministry of health, and Department of Surveys and Mapping. FEWSNET obtains secondary data from IOs such as the Food and Agricultural Organization (FAO), United Nations Children’s Fund (UNICEF), and the World Food Program. This data is either downloaded from the respective websites of these IOs, or obtained in person from local country offices in Uganda. NUMAT obtains its operational attribute data from the Ministry of Health’s national Health Management Information System, a paper based system. “All that data we get from a facility, through a system called Health Management Information System or HMIS. It is a paper based data management system. We have reports at every level, from each clinic within a rural facility collects data on patients; they summarize it and send it to the in charge of the clinic, the in charge sends it to the district, the district sends it to the Ministry of Health” (D. Sera, NUMAT, personal communication, 17th August, 2010). Further, it obtains its geospatial data in form of shapefiles from IOs operating within Uganda, such as UNOCHA, USAID, United Nations High Commission for Refugees (UNHCR), and from government agencies such as UBOS. 320 6.3.1.5 Training and Education in GIS Training at NUMAT was provided by IOs operating locally within Gulu district on a voluntary basis, for example, UNOCHA and the Danish Mine Control Group. Professional GIS training offered by GIC was considered too expensive at $100 per day per person. At Nature Uganda, there was no indication of a training program with regards to GIS – the employee using GIS software was already trained in GIS during her post- graduate studies abroad. FEWSNET offers training workshops in GIS to other NGOs that use GIS, for example, Associación Cultural para el Desarrollo Integral (ACDI), and the Association of Volunteers in International Service (AVSI), and also public sector agencies, such as the Ministry of Agriculture. The respondent explained that training is preferably carried out using ArcView 3.x instead of ArcGIS because ArcView GIS software is freely available, unlike ArcGIS. Upon finishing the training, the software can be installed free of charge on the client’s machines without the worries of license fees. 6.3.1.6 Challenges to GIS Diffusion Awareness According to one of the respondents, GIS adoption by the NGO sector has been minimal because of a lack of awareness about the technology. “GIS has not spread to so many institutions. It has remained a tool of research institutions, and big research institutions; actually, government-based, and also academic institutions. It has not really been adopted by the private (and NGO) sector so much. But I want to tell you that GIS is a tool that is not being given the attention that it deserves. So GIS in this country is still young. It has 321 not yet reached a level where people see it as a very good tool for management” (H. Muloopa, ACODE, personal communication, 24 th June, 2010). Appreciation for GIS In fact, respondents at some of the NGOs argued that GIS is not considered an important part of their operations in terms of project delivery requirements. “We appreciate the importance of GIS, but we can do without GIS. We have done without it, we can plan without GIS. It is like an additional thing. We have never been penalized for submitting a report without a map. Most projects under USAID in Uganda are only just starting out with GIS. I think NUMAT is the most advanced (project). I don’t know how many projects you have gone to, but I think we are a step ahead of them. It is not one of the major requirements to have to report geographically. At the mission itself (USAID headquarters at the US Embassy in Kampala), it is only of recent that they have recruited a person ; before that, there was no GIS person in the mission, so GIS was not … so I think GIS is one of those new things that has just started” (S. Sera, NUMAT, personal communication, 17 th August, 2010). As a result, most NGOs outsource their GIS needs to public sector institutions like the National Forestry Authority (NFA). “When the few people who are there in the department (National Forestry Authority) produce the maps, we really use them as source documents, but there are few people who can produce such work” (anonymous respondent, personal communication, May - August, 2010). 322 Skilled Personnel A lack of skilled GIS personnel and Insufficient training of employees in GIS is also a challenge to GIS in the NGO sector. At NUMAT, at the time of initial introduction of the technology, there was only one employee who was equipped with GIS skills, who departed from the NGO shortly after. Upon his departure, the software remained unused for three months because there was no one qualified to use it. The cost of training provided by the local ESRI partner, GIC, is considered expensive by NGOs. A respondent at NUMAT explained, “We asked GIC if they could train us, but the training was so expensive. They were saying $100 per day or $300 for 3 days training per person – it was too expensive. The program could not afford, so what we decided to do was just look for a local consultant here, and set out the terms of reference, and got somebody competent in GIS, because here actually in the northern region, we have some organizations which are very active in GIS, one of them is a Danish Mine control group, so they map land mines all over the region, so they use a lot of GIS technology. Then we have OCHA as well. So we sent out this Ad and we had people that responded and out of those, we were able to pick one of the local resources here, who came and we had 4-5 days training” (D. Sera, NUMAT, Personal Communication, 17th August, 2010). Data Quality NUMAT expressed frustration in the quality of the geospatial data as numerous errors in the data were detected during their work. “The challenge we have is that we have so many shapefiles, we don’t know which one is the most accurate. Sometimes when we plot our data, some of it ends up in Sudan. So, I think in Uganda, one of the challenges is 323 the accuracy of the shapefiles” (S. Sera, NUMAT, personal communication, 17th August, 2010). Data Availability Moreover, some NGOs expressed a general lack of awareness about where to obtain geospatial data in Uganda; for example, NUMAT was unaware of digital geospatial data availability through the Department of Surveys and Mapping, the national mapping agency. Political Climate The political climate has been cited as one of the reasons preventing NGOs from using GIS in their work. One respondent explained that research findings of NGOs must be in line with those of government agencies. If an NGO’s findings contradict the ruling government’s findings, then that NGO could be branded as siding with the opposition. For example, according to the government of Uganda, the annual deforestation rate on privately owned forests is 1.9% (NFA, 2009); however, if an NGO was to challenge this number based on its own GIS research, then this would result in undesired political consequences. A respondent at one of the NGOs explained, “We recognize GIS as a very useful tool for environmental management, but we haven’t done a lot of advocacy for it, even within our organization. Ever since I came here, I have never even talked to my bosses about GIS, how useful it is, and how useful it is to the organization. And every time we have wanted GIS related material, we have had to outsource it to NFA. And then the way our work is structured as NGOs, if we produced our own work, we would have a lot of questions 324 (with respect to) the authenticity of this work. People would question, ‘who has done that (work)?’ So that’s why sometimes we prefer to work with government institutions, so that if we have a report, they own the results as well. As an NGO, we wouldn’t want to do work on our own. We would want to work in partnership with them (the government institutions) so that when we are criticizing, and making our analysis, they can own up to the analysis and say yes, it is true, we did the analysis together, and this is what came out. So when we make our proposals (recommendations) to them, it is easy for them to take it up than when we do our work independently as NGOs, because that will be like hearsay; and it will be easy for government to rubbish our work. And you know our situation, right? If you are too critical of the government, you are branded the ‘opposition’. So, we want to remain relevant, but work in partnership with the government. Not just to come up and criticize it, collect our own data and even not work with them. That is our own approach basically, so I think that’s why we have not found it necessary to have our own unit of GIS, capture our own data, or parallel data to that of government, because sometimes government could lie, you know? Government could give information that is actually erroneous, for reasons of popular politics – you know, government has done this several times. But still, we work with them sometimes when they are doing this research, so that we manage quality, we do quality control, and also be participatory” (anonymous respondent, personal communication, May-August, 2010). Donor-dependency and Project-driven GIS The link between GIS diffusion and donor funding cannot be ignored. Of the three NGOs, two receive funding for GIS directly through USAID, while the third through 325 international grants. That is not to say that the other NGO, such as, NAPE, FXB Uganda, Uganda Land Alliance and ACODE, do not receive donor funding. The lack of GIS diffusion at these NGOs is due to a combination of factors: lack of awareness about, and appreciation for GIS technology, political implications of independent research findings contradicting official government publications, a lack of skilled GIS personnel, and the absence of a GIS champion in the organization. Next, I present findings on GIS diffusion in the IO sector. 6.3.2 IO Sector There were eight IO sector institutions that participated in personal interviews, and these are shown in Table 6-3. The focus areas of these organizations include peace building, social services, humanitarian services, public health, children’s welfare services and environmental conservation. Five of these IOs are located in Kampala city, two in Gulu district and two in Entebbe town. Of the eight IO sector institutions interviewed, all eight organizations had and were using GIS technology in their daily organizational workflows. 6.3.2.1 Early GIS Introduction GIS was introduced at most of the IOs in Uganda in the 2000s. In Gulu district, northern Uganda, the introduction of GIS at UNOCHA coincided with the United Nation’s establishment of institutional structures called clusters in 2005, at the end of a 20-year insurgency caused by Joseph Kony’s Lord’s Resistance Army (LRA). 326 Table 6-3 IO sector institutions at which interviews were performed ID Institution Name Area of Expertise 1 Stability, Peace, and Reconciliation in Northern Uganda (Spring) - USAID Project, Gulu Peace building/social services 2 United Nations Office for the Coordination of Humanitarian Affairs (UNOCHA), Gulu Humanitarian services/social services 3 Center for Disease Control (CDC); Uganda Virus Research Institute (UVRI), Entebbe Public Health 4 World Wide Fund for Nature (WWF), Uganda Environmental conservation 5 United Nations Office for the Coordination of Humanitarian Affairs (UNOCHA), Kampala Peace building/Social services 6 United States Agency for International Development (USAID), Kampala International development 7 United Nations International Children's Emergency Fund (UNICEF), Kampala Children’s welfare/Social services 8 Infectious Disease Institute (IDI), Kampala Public health The aim of the USAID-funded SPRING project was to mitigate the causes and consequences of the twenty year insurgency caused by Joseph Kony’s Lord’s Resistance Army (LRA) in northern Uganda through the issuance of $50,000-$300,000 grants to local communities in various sub-counties in the Acholi and Langi regions of northern Uganda. This has led to the construction of about thirty warehouses for storage of agricultural produce for the local population. GIS was introduced into the workflow of the project in 2008 mainly to keep track of the developments achieved during the course of the project, for example the food warehouses. MapInfo GIS software was introduced into the project by an employee at the start of the project. This person can be considered a heterophilous change agent. Using Garmin handheld GPS receivers for data collection, appropriate maps were generated, an example of which can be seen in Figure 6-1. These 327 maps are fed into USAID’s central database mapping system for keeping track project activity around the country. Figure 6-1: Interventions of the USAID funded Spring Project in northern Uganda Source: (EMG et al., 2008) The introduction of ArcView GIS at the Kampala-based UNOCHA took place in 2002. Prior to this time, Harvard Graphics, graphical CAD software, was already being used at this institution, but had major functional limitations, hence the move to ESRI software. The introduction of ArcGIS at UNOCHA coincided with the arrival of a new employee at this institution who had previously had 1 month’s hands-on training in ArcView 3.x at his previous employer’s, the Karamoja Data Center’s Acholi Program, today Northern Uganda Data Center; this is a public sector agency under the Office of the Prime Minister 328 (Karatunga, 2003). On joining the Acholi Program, this employee had a bachelor’s degree in information systems from a foreign university, but no formal GIS education. Thus, at UNOCHA he was sent for further training in GIS at Oakar Services in Nairobi, and at GeoInformation Communication in Kampala. This same employee later joined USAID and is the head GIS analyst in-charge of GIS support to all USAID projects in Uganda. Support provided by this employee was reported by respondents at organizations in northern Uganda, such as UNOCHA and NUMAT in Gulu district. The year of introduction of GIS at USAID and UNICEF could not be established by the respondents interviewed. GIS was introduced at IDI in 2006 through an initiative spearheaded by the Director of the institute at that time. It was due to his personal interest in GIS that the technology diffused on this organization. The director already has a preference for the ESRI suite of software, and thus ArcGIS was chosen as the GIS software to be used in the organization. 6.3.2.2 GIS Software and Hardware Usage Table 6-4 shows that 5 IOs use ArcGIS 9.x, 4 use Google Maps and Google, 2 use ArcView GIS, 1 uses MapInfo, and 1 uses open source GIS software: Quantum GIS, OpenLayers and OpenStreetMap. The dominance of ESRI software amongst IOs is consistent with the public sector in Uganda (see chapter five). Licensing for ArcGIS is not an issue in the IO sector because these organizations are well funded, unlike local NGOs. UNICEF decided to use open source GIS source mainly because the organizational culture within the Technology for Development department at 329 UNICEF “subscribes to open source software development” (S. Blaschke, UNICEF, personal communication, 10 th July, 2012). UNOCHA, Kampala, has two ArcView (the lowest) level licenses for ArcGIS, which implies that the sophisticated tools found in the more expensive ArcEditor and ArcInfo license levels are not necessary for the agency to carry out its tasks. The head of the Information Management Unit (IMU) at this agency also expressed a personal preference for ESRI software having worked with the suite for over twenty years. Prior to 2007, UNOCHA in the Gulu district office was using Harvard Graphics software. In 2007, the organization tried to switch to MapInfo which they had learnt of, and obtained from, UNHCR in northern Uganda. However, due to the user unfriendliness of MapInfo, the organization reverted back to Harvard Graphics, and also realized that there was a need for a highly specialized GIS Analyst Lead in the organization. This led to the hiring for a new GIS Analyst (from Nepal) to take over as head of the GIS unit at UNOCHA in Uganda. It was also at this time that the Gulu-based organization switched to the ESRI ArcGIS suite of software. This was followed by the training of local Information Management Assistants (IMAs) in the use of ArcGIS software. (M. F. Akello, UNOCHA, Gulu, personal communication, 17 th August, 2010). Personal computers and hardware peripherals, such as plotters and GPS devices are in common use within the IOs. This is an indication that GIS is adequately funded and supported by the parent organizations in the United States and Europe. 330 Table 6-4: GIS usage at IO sector institutions ID GIS Used Year GIS introd uced No. of GIS Employ ees ArcV iew 3.X Arc GIS 9.X Map Info Google Maps/Earth QGIS/OpenL ayers/Open StreetMap 1 1 2008 2 1 2 1 2005 1 1 3 1 2007 4 1 1 4 1 2007 1 1 5 1 2002 2 1 1 6 1 5 1 1 7 1 1 1 8 1 2006 1 1 1 Total No. 8 2 5 1 4 1 Total %age 100 25 62.5 12.5 50 12.5 6.3.2.3 Applications of GIS At the USAID Spring project in Gulu, GIS is being used to keep track of the developments achieved during the course of the project, for example the construction of food warehouses. It also used to make socio-political maps, for example, a map showing conflict affected areas in northern Uganda (see Figure 6-2). However, GIS is not considered a key component of this USAID project. In response to the question, how important is GIS to the activities carried out at your institution and is there any training planned for the Information Technology (IT) and Monitoring and Evaluation (M & E) employees in GIS, one of the respondents answered: “No, we don’t really. Any kind of GIS initiative has really just been generated from USAID’s offices in Kampala, and the Embassy. So we are not … and when we started our project there was no mandated GIS requirements, so we … It is not something that is built in to our project in a big way” (H. 331 Aaronson, USAID Spring project, personal communication, 17 th August 2010). In fact, another respondent disclosed that their MapInfo GIS software was being underutilized, and that there is a need for greater awareness about the usefulness of the technology. In the SPRING project, GIS is used for mapping applications only, and not for any sophisticated geographical analysis. Figure 6-2: A map showing conflict affected areas in northern Uganda Source: (EMG et al., 2008) At the UNOCHA headquarters in Kampala, GIS is used to map points of interest collected using handheld GPS devices, for example, health centers, schools, and internally displaced people’s camps (IDPs). The Gulu office’s role is more of a data collection, and distribution agency – not much GIS is used in this office, most of the 332 heavy duty plotting and analysis being done by the head office of the agency in the capital city. The GIS software is mainly used for educational purposes, and for printing of maps using on a plotter for distribution to local authorities. These maps help the local authorities in making meaning interventions for development of northern Uganda in a post-conflict era. CDC uses GIS to map the locations of CD4 machines around Uganda so as to monitor interventions in HIV/AIDS prevention campaigns in the country. They also mapped the locations of healthcare facilities in one of the districts in Uganda to assess society’s access to healthcare facilities in that district, especially in light of an Anthrax outbreak that had just occurred in the nearby Queen Elizabeth national park. IDI’s mandate is to slow down the spread of HIV/AIDS in Uganda. The organization partnered with Kampala City Council (KCC) on its Kampala Outreach project. KCC has six health centers evenly spread out in Kampala. These clinics provide health care services, especially those related to HIV/AIDS prevention, to residents around each of the respective catchment areas. IDI used 2 km buffers to identify local health service providers around the existing KCC clinics so as to support and leverage them to improve outreach of health services to the residents closest to them in each catchment. This would be cheaper than constructing new clinics. In another project, IDI used GIS to geocode patients residences based on patient records so as to analyze accessibility of the patients to the six KCC clinics. Through this analysis, the organization wanted to understand patient decisions with regard to clinic choice. One of the major findings of this study was that patients prefer to go to clinics that are close to 333 their work places in the city, and not necessarily to those that are close to their residences in the suburbs. This allowed IDI to target the appropriate clinics for support. USAID’s GIS unit mainly plays the role of coordinating GIS activities among its partner IOs and NGOs, especially those in northern Uganda. At the time of the humanitarian crisis due to the Kony war in northern Uganda, between 1990 and 2005, there were over three hundred NGOs (and IOs) operating in the region (D. Mutazindwa, USAID, personal communication, 5 th July, 2010). With so many organizations, USAID thought it important to formulate a policy for coordination of the activities of these organizations: the Who is doing What Where (the 3 Ws) paradigm. All USAID funded and partner agencies in northern collect some form of geospatial data using handheld GPS devices. USAID’s GIS unit is responsible for collecting this data into a central database repository in support of the 3 Ws paradigm. Of particular interest is the location of internally displaced peoples (IDP) camps where social services are needed the most. 6.3.2.4 GIS Data Collection, Data Sources and Data Sharing Not many IOs collect their own geospatial data, UNOCHA in Gulu, USAID’s SPRING project being the only ones identified in this research as performing some form of primary data collection using handheld GPS receivers. Most of the IOs rely on secondary data from public sector organizations, such as Uganda Bureau of Statistics (UBOS). The level of cooperation among the various IOs is very high, so there is a lot of data sharing among themselves. ESRI’s official GIS partner in the country, GeoInformation Communication (GIC) is also a source of data for many of the IOs. 334 There is a general outcry from some IOs regarding the frustration in obtaining data from certain public sector agencies, for example, the Department of Surveys and Mapping (DSM) under the Ministry of Lands, Housing and Urban Planning. When asked if his organization obtained data from the DSM, one respondent lamented, “Not really because those guys need money, and are very expensive. So, most of our data is from UBOS, when they are doing pre-census mapping, in preparation for the Census of 2012 for Uganda. So what they do, they map everything they come across, e.g. water points, the road network … all those things. Yes they have a database, at the moment they have finished almost 53 districts out of 112 districts” (anonymous respondent, personal communication, May-August, 2010). An attempt to create a national spatial data infrastructure in Uganda, the Uganda SDI initiative, failed in 2004 mainly due to a lack of official government policy and support (B. J. Muhwezi, UBOS, personal communication, 19 th July 2010). As a result, a group of GIS professionals from various public, academic, NGO, and IO sectors came together to form a geoinformation working group to discuss access to geospatial data in Uganda (Lwasa et al., 2006, 2005; Muhwezi, 2006). A direct outcome of the activities of this working group is an SDI initiative called Uganda Clusters (UgandaClusters, 2012). This is the closest Uganda has come to an NSDI. Uganda Clusters was set up primarily to allow humanitarian agencies and NGOs access to geospatial data in Uganda. Uganda Clusters allows Web access to various geospatial data to NGOs, and IOs working on various development and humanitarian projects in the country. 335 6.3.2.5 Training and Education in GIS The GIS departments at the various IOs in Uganda have very small GIS departments. Apart from the USAID headquarters that has five GIS employees and the CDC that has four, the other IOs employ one or two people as GIS analysts. In fact, GIS at some IOs is sub-function performed under the IT Services and M & E departments. Training of these employees is considered important only if GIS is a major part of the workflow of the organization. The educational background of most the GIS Analysts at the IOs is in statistics, Information Technology and Information Systems, for example, the employees at IDI, CDC, UNOCHA, and USAID. Most of these personnel are either self-taught in GIS, learnt on the job, or underwent some basic training in GIS after they joined their organizations. The CDC has an active training program for its GIS employees. Employees are sent for training to ESRI Eastern Africa in Nairobi, and ESRI South Africa. Further training is provided through academic institutions such as Makerere University’s Faculty of Computing and Informatics Technology in form of tailor-made courses for organizations. At the USAID SPRING project, there is no systematic training planned for employees in GIS mainly because GIS does not constitute a major part of the organization’s activities. The two employees that work with GIS technology, employed in the IT and M & E department, are certified Microsoft IT professionals who received some on-the-job training in GIS when they joined the IO from the person who initially introduced GIS into the project. They rely on the central USAID headquarters for support in GIS activities – there is only one point person that coordinates GIS activities and provides 336 technical support to all USAID projects in Uganda, and this is the same person that the SPRING project relies on. However, there appears to be inter-agency cooperation, in terms of technical support in GIS, among the various IOs and NGOs in northern Uganda, for example, consultative collaboration among SPRING, UNOCHA, UNHCR and NUMAT. Training for all Information Management Assistants (IMAs) of UNOCHA in each of the districts in northern Uganda was initially carried out in 2008. This training was needs- based training to respond to the post-war peace building initiatives war-ravaged northern Uganda. Prior to 2008, UNOCHA, with the help of IMAs, authored and published so- called Red Zone/Green Zone maps and handed these to the people affected by the Kony war in northern Uganda. Red zones indicated high risk zones, which the public was advised to avoid, while the green zones indicated safe zones. GIC, the official ESRI business partner in Uganda, was contracted to carry out training in ArcGIS for these IMAs. The training was carried out in a central location, Soroti district, and lasted three weeks. After the training, all personnel were given a copy of ArcGIS software on their laptops to take back to their parent districts. However, after this initial training, no follow-up training was offered. As one of the respondents revealed, the effect of this lack of follow up training has been that she has forgotten most of the skills she was initially taught. Moreover, because all the heavy-duty GIS work is done by the UNOCHA headquarters in Kampala, the IMAs have been relegated to data collectors and disseminators, with little use for their GIS skills and knowledge. This has further led to the decay in their GIS knowledge and skill level. “So 337 from there I started using it, but because of the workload, I was now to oversee the whole Acholi region, but I do not practice it so much. I have left most of that to the people in Kampala. We have Kitgum, Gulu, Amuru, Nwoya, Pader, Agago, Lamwo – 7 districts. It’s very big. So even if I sit sometimes, I want to do my GIS practice, I get my coordinates, I begin plotting so that I make a map out of it, but people are always coming bringing their requests (for maps), so I have almost totally given up and left mapping using GIS. When I get the information (the GPS data), I just send it to Kampala” (M. F. Akello, UNOCHA, Gulu, personal communication, 17 th August, 2010). The GIS specialist at IDI did his initial training in GIS at GeoInformation Communication, Kampala. This training was in introductory ArcGIS. After this initial training, he trained himself through personal initiative using online tutorials. Given his strong technical background in statistics, mathematics and databases, it was not too difficult for him to utilize free Web tutorials in GIS. 6.3.2.6 Challenges to GIS Diffusion Understaffed Departments GIS departments at IOs are acutely understaffed. For instance, the GIS Department at the USAID headquarters in Kampala employs only five GIS analysts. These employees are in charge of providing support to all USAID-funded projects in the country. Moreover, the majority of those employed as GIS analysts do not seem to undergo regular training. This could be due to, either the lack of importance of GIS to the activities of the IO (for example, the SRPING project), or because a general lack of appreciation of the potential for GIS in the workflow of the IOs. 338 NSDI The lack of an NSDI in Uganda is problem with regard to data access for IOs. Despite initiatives throughout the 2000s (Sentongo 2003; NIMES 2004; GoU 2006; S. Lwasa et al. 2006; Muhwezi 2006; Lwasa et al. 2005), there is still no NSDI in Uganda. A practical alternative to the NSDI is Uganda Clusters, a recent initiative that allowed for the creation of a onestop Web based repository for geospatial data in Uganda (UgandaClusters, 2012). A promising initiative is the development of Uganda’s national Land Information System under a world bank funded project the Land Sector Strategic Plan 2001-2011(MLHUD, 2011a, 2011b). Next, I present findings on GIS diffusion in the private sector. 6.3.3 Private Sector There were 11 private sector organizations and companies that agreed to participate in this research. All of these organizations are located in the capital city, Kampala. These included various industry foci, for example, mobile telecommunication, surveying and engineering, oil and natural gas, printing, waste disposal, utilities, and GIS consultancy. 6 out of the 11 companies use GIS technology in their workflows. Each company has either one or two GIS analysts. 6.3.3.1 Early GIS Introduction The diffusion of GIS in the private sector is mainly through large foreign-owned multinational corporations and companies operating in Uganda, such as the South African-based telecommunications company Mobile Telecommunications Network Uganda (MTN) that introduced GIS into their workflows in 2000; WE-Consult, a Dutch- 339 based water, environment and geoservices consultancy that adopted GIS in 2005; Umeme, a South African-owned power utility company that adopted GIS in 2005; Nokia Siemens, a Finish multinational data networking and telecommunications company; and Tullow Oil Uganda, a UK-based oil and gas exploration company that adopted GIS in 2005. Local Ugandan-owned private sector businesses that use GIS are virtually non- existent. Apart from the local official ESRI partner in Uganda, GeoInformation Communication (GIC) that distributes licenses for the ArcGIS suite of software and provides value added training services, no other locally-owned private sector business exhibited the use and adoption of GIS software during this research. Table 6-5: Private sector institutions at which interviews were performed ID Name Industry 1 MAP (U) Ltd Surveys and Mapping 2 MTN Uganda - Mobile Telecom Network Mobile Telecommunication 3 GeoInformation Communication - GIC GIS support and Training 4 Umeme Utility Company - electricity distribution 5 WE-Consult, Water Environment and GeoServices Water and Environment consultancy 6 Tullow Oil plc - Uganda Oil and Gas 7 BIN-IT Services Ltd Domestic Waste Disposal 8 TourGuide Publications, an imprint of Fountain Publishers Printing and Publishing of tourist maps 9 Nokia Siemens Mobile Telecommunication 10 Mitland (U) Ltd Surveys and Mapping 11 GeoMaps Africa Ltd Surveys and Mapping GeoMaps Africa, Map Uganda, and Mitland, all surveying and mapping consultancies use AutoCAD software from the company AutoDesk to meet their objectives of creating maps and computing earthworks volumes for civil engineering projects; BIN-IT services, 340 a domestic waste disposal company, uses a combination of paper-based topographic maps from Kampala City Council (KCC) and Google Maps/Earth to manually solve their travelling salesperson problem while planning routes for garbage collecting around Kampala city; and TourGuide publications, a printing and publishing company, uses design software from Adobe Systems, Adobe Illustrator, to author and publish paper- based tourist maps (TourGuide, 2010a, 2010b). See Figure 6-3, Figure 6-4, and Figure 6-5 for travel and tourist maps published by TourGuide (2010) publications. GIC was incorporated as a company in 2002 as the official ESRI partner in Uganda. It delivers value added services on behalf of ESRI, such as training and support. GIC operates under ESRI Eastern Africa in Nairobi. Prior to 2007, the distributor of ESRI’s software in Uganda, Kenya, and Tanzania was Oakar Services in Nairobi, a Kenyan company. Prior to that, between 1991 and 1994, the distributor was another Kenyan GIS company, Thunder Associates. The most influential personality in the diffusion of GIS in East Africa is probably Willy Simons, a GIS professional who received his professional GIS training at Wageningen University in Holland (MSc. Agricultural Engineering) and has lived in Kenya since the early 90s. He worked for Thunder Associates in the early 90s, was the Managing Director at Oakar Services from 1999 to 2007, and has been the Managing Director of ESRI Eastern Africa from 2007. 341 Figure 6-3: Travel Map of South Eastern Kampala Source: (TourGuide, 2010a) GIC is the first point of contact for GIS software acquisition by organizations in the various sectors in Uganda, the Managing Director, Amadra Ori-Okido, explained: “We deal with NGOs. We have a number of NGOs operating in the country. We deal with UN agencies, UNOCHA, UNDP, World Food Program, and so on. We deal with government, Ministry of Lands, Water, Umeme, UEDCL (Uganda Electricity Distribution Company Ltd), UETCL (Uganda Electricity Transmission Company Limited), MTN (Mobile Telecom Network), Universities, Makerere University, Faculty of ICT (Informatics and Communication Technology), Department of Geography, MUIENR (Makerere University Institute of the Environment and Natural Resources), Surveying, Veterinary Medicine, Makerere University Medical School’s IDI (Infectious Disease Institute)” (A. Ori-Okido, GeoInformation Communication, personal communication, 29 th June, 2010). 342 Figure 6-4: Highways in Uganda managed by the Uganda National Roads Authority Source: (TourGuide, 2010a) The beginnings of GIC were very humble indeed, having registered as a limited liability company in 2002, sighting in niche in the Ugandan market. “Even OAKAR services didn’t manage to penetrate the Ugandan market by the early 2000s. We (the National Forestry Authority) were buying the software (PC Arc/Info and ArcView) from South Africa from a company called GIMS, who were the distributors for South Africa. Now also they have been bought off by ESRI South Africa, recently, I think effective January 343 this year, 2010” (A. Ori-Okido, GeoInformation Communication, personal communication, 29 th June, 2010). Figure 6-5: Downtown Kampala Tourist Map Source: (TourGuide, 2010b) Between 2002 and 2005, “there was no business,” according to the Managing Director of GIC, Amadra Ori-Okido. There was simply inadequate knowledge about GIS technology 344 in Uganda’s public, private, and NGO sectors at that time. “So GIS was an alien technology, and it was very difficult actually to penetrate (2002-2005). Now it is moving very fast. Actually, in the last two years, it was like the wind has been blowing this way. You talk of GIS (back in 2002), ‘what is this?’ The only thing you could use was present a map. Of course, nowadays you talk of cartography, like in Entebbe (Department of Surveys and Mapping)… we say, no, no, no, we now have a technology that can make maps; you don’t have to train as a cartographer. Even, to get people to employ; it is very difficult” (A. Ori-Okido, GeoInformation Communication, personal communication, 29th June, 2010). So, the trend started changing in about 2008, and all of a sudden, the number of institutions adopting GIS technology started increasing allowing GIC to run a sustainable GIS value added services business. One of the major challenges, however, is the lack of skilled GIS personnel. 6.3.3.2 GIS Software and Hardware Usage Five companies employ ESRI’s ArcGIS suite of GIS software; two companies use MapInfo, one uses ArcView GIS, and seven companies use Google Earth and Google Maps. The surveying and mapping companies all use the CAD software, AutoCAD. The utility company, Umeme, was found to have a complete package of desktop, server and mobile-based ArcGIS software, while the other companies used only the desktop software. All companies use desktop personal computers to host their GIS software. They also own plotters and hand held GPS receivers for data collection. Because most of the companies in Uganda’s private sector are powerful multinational corporations, obtaining and maintaining licenses for GIS software and acquisition of hardware is not a challenge, 345 as opposed to the NGO and public sectors. Licensing of ArcGIS software is done through GIC in Kampala, or ESRI Eastern Africa in Nairobi. Table 6-6: GIS usage at private sector institutions ID GIS Used Year GIS Introd uced No. of GIS Employees ArcV iew 3.X Arc GIS 9.X Map Info Google Maps/Eart h Auto CAD 1 0 1 1 2 1 2000 1 1 1 3 1 2002 1 4 1 2005 2 1 1 5 1 2005 2 1 1 6 1 2005 1 1 1 7 0 1 8 0 1 9 1 2006- 2010 1 10 0 1 1 11 0 1 Total Number 6 1 5 2 7 3 Total % 55 9 45 18 64 27 Only a few companies have adopted server GIS, for example, MTN, and Tullow Oil that use ArcGIS Server. The same companies also employ Mobile GIS, ArcPad. Desktop GIS, however, is the main form of GIS employed in the private sector. Tullow Oil and Umeme license their ArcGIS installations at the ArcEditor license levels. The choice of GIS software at two companies was based on a decision made by the managers in the company. In one rare case, the choice of software was made based on the recommendation of the GIS analyst. 346 6.3.3.3 Applications of GIS There are several applications of GIS in the private sector. The power utility company, Umeme uses ArcPad to locate transformers, customers, meters, power lines, voltage poles and other electricity related infrastructure. Twenty-three employees were trained to use ArcPad for data collection, and these are supervised by a GIS Manager. This information is downloaded to ArcGIS desktop and used to manage utility services. Maps are produced, printed, and handed over to external contractors to guide them in infrastructure maintenance exercises, for example, the replacing of old and rotten wooden electricity poles. Umeme also uses ArcGIS server applications developed using the Flex API to allow access to their central geodatabase. This is instrumental in facilitating the sharing geospatial information across the organization because the company has several office locations. MTN Uganda uses MapInfo for line of sight or visibility analysis to ensure effective geographical coverage around their cell phone towers or masts. Further, GIS is used for location analysis to find the best possible locations for cell phone towers in new areas. Related to this, GIS is used for preliminary planning by teams heading out into the filed to build new masts and install telecom equipment. For simulation and network planning, telecom industry specific software, such as ASSET and NirSoft is used. WE-Consult uses GIS for hydrological mapping. “We plot boreholes, differentiating water sources, and mapping all the water sources at the moment. We are working on a new national water atlas of Uganda, so it is all thematic maps, and there are nearly no topographic maps. It is really easy work – nothing complicated with the software” (Y. 347 Tylle, WE-Consult, personal communication, 2 nd August 2010). GIS is used for creating reconnaissance maps that are issued to field operatives, for example, if field operatives go out to make river flow measurements, topographic maps showing locations of rivers, roads and nearby landmarks are created. At Nokia Siemens, GIS is used for transmission and network planning. Line of sight analysis is performed when planning for new sites for antennas so as to reduce the likelihood of signal degradation in the surround areas of coverage. MapInfo and Planet, a network planning software, is used for this purpose. TourGuide publications creates very aesthetically pleasing tourist maps and publishes these as pamphlets and gazetteer books, for example, the Kampala A-Z (TourGuide, 2012). These cartographic products are created using Adobe Illustrator and other cartographic design packages. Geospatial data used for creating the cartographic layouts is obtained from the Department of Surveys and Mapping in digital exchange format (DXF), and then manipulated in Illustrator. Scale is computed manually and applied to achieve correct geographic proportions and choose appropriate symbology for the maps. The company also performs map updates since the maps obtained from the national mapping agency are often out of date by several decades. First, to perform updates on geographical features, Google Maps is employed as a data source. Since the satellite imagery on Google Maps is often times not more than two years old, this Web mapping application meets the objectives of the company. Screenshots of Google Maps are captured, co-registered with existing vector maps in Illustrator, based on manual 348 computations of scale, and digitized accordingly to capture new geographic features in vector format. Second, printouts of the updated maps are made and handed over to field crews that go out into the city to record attributes for the new geographic features, and update old ones. Extensive field notes are taken describing the geographic features, which are then brought back to the office to update the digital maps in Illustrator. Names of businesses, health centers, schools, government buildings and other details are recorded in the field, and then used to label the geographic features. The company’s business model is based around revenues from companies that wish to be included in the map gazetteer; they pay for their company name to be reflected in the gazetteer. Another source of revenue is organizations that make orders for specific maps to be authored, specifically customized to reflect their missions and mandates. For example, the Uganda National Roads Authority (UNRA) made an order to author and print pamphlets showing the highway and local street network in Uganda at various scales, including the organization’s logo, possibly to improve public relations with citizens, and also to earn some revenue from the sale of tourist maps. Tullow Oil uses GIS for plotting the location of oil wells, both those currently being drilled, and prospective wells; mapping of the shoreline of lake Albert; and geovisualization. No sophisticated GIS analysis is performed. Geospatial data is collected using Trimble handheld GPS receivers. The topographic map base is provided by the Department of Surveys and Mapping, Entebbe, and land use maps are obtained from the National Forestry Authority. 349 6.3.3.4 GIS Data Collection, Data Sources and Data Sharing The source of geospatial data for most companies in the private sector is UBOS, and the Department of Surveys and Mapping. Other data providers include the National Forestry Authority, MUIENR, KCC and NFA. Data collection for map updates is primarily performed using handheld GPS receivers. In one unique case, Google Maps is used as a primary source of satellite imagery to allow for map updates, however, this is done in a non-GIS environment using graphical software packages, such as Adobe Illustrator. 6.3.3.5 Training and Education in GIS At the power utility company, Umeme, twenty-three employees were trained in the use of ArcPad by a South African GIS consultant. This training lasted two weeks. Since the GIS manager has a strong academic and professional background in GIS, there is no need for further training on his part. The network supervisor did some training in GIS at GIC in Kampala. This training was based on standard ESRI instructor-led courses which typically last three full days. Tullow Oils GIS analyst has a Master’s degree in GIS from Europe, and undergoes training at GIC in Kampala, and ESRI Eastern Africa in Nairobi. The GIS analyst at WE- Consult received GIS training during his bachelor’s degree in Geography in Germany. At MTN, the employee handing GIS analysis has a bachelor’s degree in electrical engineering and a masters in Computer Science (data communications and Networks). Since GIS constitutes about 20% of his work, there is no need for regular training in GIS 350 for this employee. However, recently, a group of other employees at MTN underwent training in the use of ArcGIS Server at GIC. 6.3.3.6 Challenges to GIS Diffusion Data Quality It is often the case that data obtained from the various data producing organizations, such as UBOS, NFA and Uganda Clusters does not tie in together perfectly. This could be because of varying data scales or simply to the lack of enforced standards for data interoperability. A correspondent at a private GIS consultancy explained, “There are several institutions here in Uganda that provide us with data. There is UBOS that gives us statistical data, which is one thing, but the georeferencing is somehow wrong when I look at these maps. Then there is Uganda Clusters. You can check on Google. It is a group of GIS experts working here, and they are mainly using the same shapefiles, for example, some projections are better, I don’t know, some town shapefiles are better from Uganda Clusters, but some river shapefiles are better from UBOS, so we are always trying to get everything (data)” (Y. Tylle, WE-Consult, personal communication, 2nd August, 2010). Pirated Software In one particularly bad case, a fake company supplied illegal GIS licenses to a university and caused a loss of over $40,000 to the University’s department. This was a loss of revenue to the official distributors of GIS licenses in the private sector. This reflects a bigger problem regarding ethics and the business climate in Uganda. Mismanagement of funds and the culture of unethical practice have negatively impacted the diffusion of GIS in the country to some extent. 351 Decentralization of Government The decentralization of government has had a negative impact on the diffusion of GIS in the private sector. A correspondent explained the divisions at the Department of Surveys and Mapping, Uganda’s national mapping agency: “Again, you look at Lands (The Ministry of Lands, Housing and Urban Development) and Surveys (the Department of Surveys and Mapping), but there are different units with different budgets, and it is also again impossible for them to come together. Each one like Kingdoms, this one wants to have this much control and identity …” (anonymous correspondent, personal communication, May-August 2010). If various departments within a Ministry were able to unite under one umbrella, and pool their funds, they would be able to purchase an unlimited number of GIS licenses for about $40,000. However, each department prefers to buy its own single licenses for $5,000 to $10,000 each. High Cost of GIS Licenses The cost of GIS licensing has proved to be a hindrance to GIS adoption in the NGO and private sectors, and not at all in the IO sector. This is because local NGOs and private companies usually do not possess the finances required. Asked if this was a challenge, the GIC Managing Director explained, “Well, it is not really a hindrance, because I mean ESRI invests almost half of their income/profits back into research and development … so in terms of pricing, some people say this and that, but it’s an asset. I always tell people when they complain that the software is expensive, consider buying a fleet of cars, which you must all the time maintain and run spending additional money, but this is a software you buy once, and you have a computer, you can use it for years … Now which one is much cheaper? It is an investment; you have to make decisions and make choices. I don’t 352 think it is a big hindrance. The problem could be just the appreciation of the value of the GIS itself. You can get lower and middle level officers who know what to do with it, but then you have top management people who might not really appreciate the value, and they don’t approve the money for it” (A. Ori-Okido, GIC, personal communication, 29th June, 2010). Appreciation for GIS Some private universities would rather teach GIS at a theoretical level than pay for GIS licenses. Often GIS faculty at private universities demand for the free student evaluation editions of ArcGIS software, not wanting to pay licensing fees (A. Ori-Okido, GIC, personal communication, 29 th June, 2010). This could be due to a lack of appreciation for GIS technology, or simply a lack of funding for licenses. However, such an attitude does not allow GIS to flourish in the private sector as distributors of GIS licenses, such as GIC, lose business opportunities. Further, learning GIS at a theoretical level is a raw deal for students in private universities as they would not receive the necessary skills required to take up GIS-based employment in the job market. Skilled Personnel There is generally a lack of skilled GIS personnel to employ in the NGO, IO and private sectors. There is a need for more capacity building initiatives throughout all sectors. The problem of brain drain still remains; highly educated GIS professionals seek greener pastures in industrialized countries, leaving a vacuum in their home countries. 353 Branding of GIS Software With regard to the branding of ESRI’s ArcGIS software, there has been some confusion amongst managers in various sectors about which package to acquire. GIC’s Managing Director narrates a scenario where the manager of one of Uganda’s prominent public sector agencies was totally confused about ESRI’s software, “Unfortunately, the boss has some knowledge of GIS, which is good, but he started with ArcView 3.x. So he got stuck and he has been confusing ArcMap, ArcCatalog and ArcView 3.x and ArcView (the license level), because when you talk about ArcView (the license level), the man goes back to ArcView 3.x, the old software package. Yes, then there is this confusion of ArcGIS. For them ArcGIS is ArcMap, and ArcMap is … total confusion! So I think doing a lot of this Arc, Arc, Arc is good as a branding (strategy), but it also creates a lot of confusion in the minds of the users” (A. Ori-Okido, GIC, personal communication, 29 th June, 2010). Awareness There is a need for greater awareness creation about GIS in the various sectors of Uganda’s economy. GIC helps add value to GIS by providing training facilities, organizing regional GIS conferences, distributing GIS literature in form of newsletters and magazines. GIC currently has the capacity to train 20 people at a time in its on-site GIS computer lab. Efforts made in advertising GIS services in the newspapers have so far had zero impact on increasing adoption. In the future, informative programs on TV and radio might be a useful communication channel for increasing awareness about GIS. One of the most effective ways of increasing awareness has been through the annual regional user conferences in Uganda, Kenya and Tanzania. Students, teachers, professionals, 354 managers, and the broad GIS community come together to participate in an educational experience, and this also allows for the creation of personal and professional networks. NSDI One of the biggest challenges to the diffusion of GIS has been the lack of an NSDI. Initiatives to develop an NSDI started in the early 2000s. Workshops were held, reports were written, yet nothing materialized in terms of an NSDI. A notable initiative that failed was the National Integrated Monitoring and Evaluation Strategy (NIMES). At the root cause of the problem was leadership and the lack of a mandate. No one organization had the mandate to own the NSDI, and all stakeholders involved were struggling to claim ownership of such a project. Some of the participants in the workshops included UBOS, Karamoja Data Center (today, Northern Uganda Data Center), Department of Surveys and Mapping, and Makerere University’s Department of Geography. Each of these institutions sought ownership of the NSDI, and as a result, it failed. Today, Uganda Clusters is an initiative that attempts to play the role of some sort of an NSDI, however, it serves mainly NGOs, and IO; not the other sectors. The deepest hope for an NSDI is in a possible spin-off from the Land Information System project currently being implemented by the Ministry of Lands, Housing and Urban Development, as part of a World Bank funded project, the Land Sector Strategic Plan (MLHUD, 2011a). Funding for the project is also provided by the central govern, local government, and the private sector. The most important anticipated outcome of this project is the emergence of policy on geospatial data and institutional changes supporting NSDI. 355 Next I present an analysis of the current state of GIS at the NGO, IO and private sector institutions in Uganda based on Diffusion of Innovations Theory (Rogers, 1993, 1995, 2003). 6.4 The State of GIS in Uganda: A Diffusion of Innovations Analysis Diffusion is “the process by which an innovation is communicated through certain channels over time among the members of a social system” (Rogers 2003, 11). This section uses the diffusion of innovations theoretical framework to analyze the state of GIS in Uganda’s NGO, IO and private sectors. Each of the four elements of diffusion is analyzed in turn: (1) the innovation, (2) communication channels, (3) time, and (4) the social system. 6.4.1 The Innovation: GIS In this section, I discuss GIS as an innovation based on (1) the perceived attributes of the innovation, (2) technology clusters, and (3) re-invention. 6.4.1.1 Perceived Attributes The perceived attributes of innovations are those characteristics of an innovation that are favorably or unfavorably perceived by a society leading to adoption or rejection of the innovation (Rogers, 2003). These include: (1) relative advantage, (2) compatibility, (3) complexity, (4) trialability, (5) and observability. Of greatest importance to GIS diffusion in Uganda’s NGO, IO and private sectors is observability, “the degree to which the results of an innovation are visible to others” (Rogers 2003, 16) – the greater the observability, the higher the expected rate of adoption of an innovation. There is a general lack of awareness about GIS technology in the NGO 356 and private sector, and this is mainly because there is not enough publicity about the applications of GIS within those sectors. This is not the case with the IO sector that is influenced by well-funded international support in terms of expertise and heterophilous communication networks consisting of experts in GIS from the parent countries of the IOs. Observability of GIS as an innovation could possibly be increased by a GIS software vendor campaign targeted specifically at NGOs and private sector companies to demonstrate the potential for GIS as a tool for analysis and decision making. This would allow potential adopters to try out the technology (trialability), assess its relative advantage, compatibility and complexity. This would in essence increase awareness about the technology. An example of the lack of awareness about GIS technology in the private sector is a waste disposal company in Kampala that uses outdated paper maps to do routing for its dump trucks, yet this is a classical travelling salesperson problem that could be easily solved by GIS technology. 6.4.1.2 Technology Clusters One of the biggest challenges to the diffusion of GIS technology in Uganda is the lack of up-to-date geospatial data. The most recent topographic map of Uganda dates back to the 1960s, after which, only small regions, especially around major urban centers have been updated (Okia & Kitaka, 2010). Resources that could be employed in the training of new GIS staff, acquiring of additional licenses, and widening of professional networks are instead spent on updating maps. Due to a poor attitude towards data sharing (Felicia Olufunmilayo Akinyemi et al., 2011; Nyemera, 2008; Uwayezu, 2010), and the lack of a 357 geospatial data structure (De Vries & Lance, 2011; Musinguzi et al., 2010), this has led to duplication efforts in data collection (Nyemera, 2008). The geospatial data challenge in Uganda and its relationship with GIS is best explained through the concept of technology clusters in diffusion theory. A technology cluster consists of “one or more distinguishable elements of technology that are perceived as being closely interrelated” (Rogers 2003, 14). For the successful diffusion of GIS technology, a technology cluster package that includes geospatial data together with the technology would immensely increase chances of adoption. Rogers (2003) asserts that to treat closely related innovations as if they were independent from one another would involve a “dubious assumption” yet diffusion of the two innovations is closely interlinked. The diffusion of GIS is dependent on the availability of geospatial data, and if GIS vendors were to package their software together with user specific geospatial data as a technology cluster, this would greatly increase the relative advantage and adoption of GIS technology among organizations. Such a technology cluster would allow for easier re-invention of GIS technology in terms of adding value to existing geospatial data by limiting data collection efforts to only updating of attribute data for geographic features, eliminate costs of map updates, and reduce wastage of scarce resources due to redundancy in data. The ability of an organization to change or modify an innovation to suit its needs, the concept known as re-invention (Rogers, 2003) in diffusion theory, tends to increase adoption of an innovation. Clusters of GIS technology and geospatial data would promote re-invention in both these innovations because savings made in costs of data collection 358 would free funds for advanced training in GIS software customization, conferences, workshops and wider networks of collaboration. 6.4.1.3 Re-invention All the NGOs, IOs, and private sector institutions identified in this research to be using GIS had a very low degree of re-invention with respect to the customization of GIS software to suit their needs. This could have been due to the lack of skilled personnel to carry out such a task, or there was simply no immediate need for such an exercise. I, however, believe that the lack of highly skilled personnel in these sectors is probably the most likely cause. Many of the employees in these institutions had only basic training in GIS, often times on-the-job training lasting only a week or two. And even then, there is no sustained GIS training program within the organizations. This could be because of the relatively low importance given to GIS as a technology supporting development and activities of organizations. Thus, managers do not recommend budgetary allocations toward further development of skills of GIS staff. Further, it is often the case that these institutions are acutely understaffed in terms of GIS personnel. USAID, for example, only employs five GIS personnel to monitor activities of its projects in the entire country. A distinction must be made between NGOs, IOs, and the private sector in terms of their training capacity of GIS staff within their organizations. Local NGOs in Uganda are constrained by low budgets and a voluntary staff base, and thus an almost non-existent training record of employees in GIS. On the other hand, IOs such as USAID and UNOCAHA provide much better support to their GIS staff in terms of training opportunities mainly because the more professional organizational culture and financial 359 base backed up by funding from their parent countries in the West. When it comes to private companies, this trend is almost exactly identical. Local private companies have no capacity in terms of GIS training, while the large multi-national corporations and foreign- owned companies often send their staff for regular training. 6.4.1.4 Critique The cost of acquiring and maintaining new technology is scarcely addressed in diffusion of innovations theory. This could be because of the capital-intensive and credit-driven nature of commerce in developed nations as there is a positive relationship between credit and a country’s gross domestic product (GDP) (Hofmann, 2001). Industrialization and development in emerging economies in Eastern Europe, and the Balkans has been supported by loans from banks to the private sector at relatively reasonable interest rates allowing for economic growth (Cottarelli, Dell’Ariccia, & Vladkova-Hollar, 2005); the rapid industrialization of South Korea after the 1960s is related to credit infrastructure (Y. J. Cho, 1997); and that of Japan, as well (Kruse, 1997). Developed countries are able to use a system of private credit to support private sector development because they have strong legal systems in place (Djankov, McLiesh, & Shleifer, 2005), unlike developing countries. In Uganda, credit in the private sector is characterized by short term bank loans carrying very high interest rates (Ayogu, 1999) which does not encourage private sector growth. This has a negative impact on the diffusion of GIS in developing countries. In my opinion, diffusion of an innovations theory does not offer theoretical understandings of the relationship between economics and diffusion, especially in the developing countries’ context. Theories on appropriate technology and technology 360 transfer are more appropriate in this case (for example, see Schumacher 1973; Willoughby 1990; Cartwright 1993; Hazeltine and Bull 1999; Yapa 1991; Loh, Marshall, and Meadows 1998). 6.4.2 Communication Channels Communication is “the process by which participants create and share information with one another in order to reach a mutual understanding” (Rogers 2003, 18). Communication channels play a very important role in the diffusion of any innovation. This section discusses the (1) types of communication channels, and (2) the effect of heterophily and homophily. 6.4.2.1 Types There are generally three kinds of communication channels: (1) mass media, (2) interpersonal, and (3) interactive (Rogers, 2003). Mass Media The impact of mass media, radio, television, newspapers and magazines, on the diffusion of GIS in Uganda has been almost insignificant. As a respondent put it, “We tried in the past to go to the newspapers, but you know, the reading culture here in Uganda is not very good. Somebody sees something – very strange. And even to advertise our services, initially, there was no impact. So we just stopped that kind of … may be in the future we will go back and maybe start TV programs to talk about GIS” (A. Ori-Okido, GeoInformation Communication, personal communication, 29 th June, 2010). 361 Interpersonal Interpersonal channels, involving face-to-face exchange between individuals, have shown the greatest potential in terms of GIS diffusion in Uganda. “What we are doing or what we are supposed to do as a business is to come up with a business strategy. We have now 3 people who are going out to contact clients – face-to-face marketing, door-to-door marketing. Then we have been participating in the trade show for the past two years. So many people go to the trade show. That’s the biggest gathering you can get in one full year – for one full week, you get students, you get managers, CEOs of big corporations who don’t have chance to be visited. They get a chance to go and discover that there is GIS. Then you kind of create awareness” (A. Ori-Okido, GeoInformation Communication, personal communication, 29 th June, 2010). The regional ESRI user conference has also been an effective interpersonal communication channel that has been leveraged by GIS companies like GIC in Uganda. Another annual event, the GIS Day, marked in every country every year at institutions of higher learning is also one such interpersonal channel. However, both of these have a limitation in that only experts and novice GIS users attend such functions and conferences, which does not allow for diffusion amongst those who have no idea about GIS. Interactive Interactive communication over the Internet has not had any major impact in terms of creating awareness about GIS technology in Uganda. None of the respondents interviewed during this research first learnt of GIS through surfing the Web or getting an email about GIS technology. However, the World Wide Web has definitely allowed for better access to geographical information, for example, through Internet map server 362 technology such as Google Maps and Bing Maps; however, this has not necessarily translated into greater awareness about GIS as a technological innovation. Some companies use Google Maps imagery to update paper topographic maps obtained from the Department of Surveys and Mapping, and Kampala City Council, for example, BIN- IT services, and TourGuide Publications. Respondents at these companies, nonetheless, had no formal understanding of the term GIS, or any knowledge about the existence and purpose of the technology. Moreover, there is no functional street address system in the country that would encourage spatial thinking in local citizens, for example, to think in terms of networks, and utilize Google Maps or Bing Maps to access the routing functionality. 6.4.2.2 Heterophily and Homophily For effective diffusion of GIS technology, some level of heterophily, with regard to knowledge about the innovation is necessary, in a two-way communication process (Rogers, 2003). In addition, homophily in terms of other variables, such as education and socio-economic status further supports diffusion (Rogers, 2003). In Uganda’s NGO sector, GIS technology was initially introduced through heterophilous communication links of NGOs to the IO sector and is characterized by support from donor agencies such as USAID, and poor training in the use of the technology. This could explain the slow diffusion of GIS in this sector. In the IO sector, the same heterophilous traits in communication are found, however, different from the NGO sector is the fact that these organizations are well funded and 363 adequate training in GIS is provided. Often times, however, these organizations are understaffed with respect to expertise in GIS relative to the amount of work to be done. GIS diffusion in private sector companies is also subject to heterophilous exchange of information. GIS has been introduced by employees in top management positions in the companies, who generally tend to be highly educated professionals from industrialized countries. The only private sector players in terms of GIS diffusion tend to be multinational corporations that are economically very powerful. Local Ugandan private sector businesses that could be potential adopters of GIS are unaware of the technology, simply can’t afford to purchase GIS licenses, or have no specific need for GIS, for example, land surveying consultancies that use AutoCAD. In this case, the relative advantage of GIS is not significant enough to lead to adoption. The choice of GIS software is often guided by preferences of the parent multinational companies based in in Europe or North America. ArcGIS and MapInfo are the two commonly used GIS software today, for example, in the telecommunications industry. GIS employees in the multinational companies undergo more regular training than in the other two sectors, the frequency depending on the degree of importance of GIS to the companies’ business model. In all the three sectors, GIS personnel tend not to be holders of GIS masters degrees, instead, they are Information Technology and Information Systems professionals trained on the job in GIS through heterophilous communication channels between them and foreign GIS experts working as expatriates in Uganda. Training in all three sectors is mainly provided by GeoInformation Communication, the local ESRI partner in Uganda, and this is mainly because ESRI software is the dominant 364 GIS software in each of the sectors. MapInfo training in provided on the job by foreign experts who are already familiar with the software. 6.4.3 Time The time element in diffusion of innovations is an important dimension. This section discusses the time element with respect to (1) the rate of adoption, and (2) innovativeness and adopter categories. 6.4.3.1 Rate of Adoption The rate of adoption of GIS in Uganda’s NGO, IO and private sectors is shown in Figure 6-6. The shape of the cumulative frequency-based curve portrays the early stages of an S- shaped curve which shows that the adoption of GIS in Uganda’s NGO, IO and private sectors is on an upward trend. This result, however, has a limitation in that an S-shaped curve is typically based on the entire population of individuals or units under study, not a sample population, which is the case in this research. Thus, the critical mass needed for GIS to take off as an innovation, which occurs typically at the 10% adoption level, cannot be inferred based on this data. A more comprehensive study including all the NGO, IO, and private sector institutions would be needed to achieve that goal. Nonetheless, a conclusion that can certainly be drawn from the graphical plot is that GIS usage in the three sectors is on the rise. Also, GIS diffusion in these three sectors definitely lagged behind that in the public and academic sectors by about 5 years. The reason for this trend is because change agents such as UNEP-GRID and ESRI concentrated on public and academic sector institutions in developing countries before turning their attention to other sectors. 365 Figure 6-6: S-Shaped Curve: Rate of adoption of GIS in the NGO, IO and Private sectors in Uganda, 1986-2010 6.4.3.2 Innovativeness and Adopter Categories Innovativeness is “the degree to which an individual or other unit of adoption is relatively earlier in adopting new ideas than the other members of the system”(Rogers 2003, 22). Adopter categories, “the classification of members of a social system on the basis of innovativeness, include (1) innovators, (2) early adopters, (3) early majority, (4) late majority, and (5) laggards” (Rogers 2003, 22). Innovators, Early Adopters and Early Majority The innovators and early adopters of GIS were in the IO and private sectors; not so much in the NGO sector. This trend could be explained by the fact that the IO sector in Uganda 366 had the strongest relationship with other International Organizations that acted as change agents in the diffusion of GIS in developing countries, for example, UNEP and ESRI. Private sector companies using GIS are typically multinational corporations, and they too had a higher likelihood of contact with change agents promoting GIS adoption, such as ESRI and the MapInfo Corporation. Both the IOs and private sector also have a strong economic base for investment in innovative technologies. Late Majority and Laggards The NGO sector, on the other hand, can be considered as late majority and laggards in terms of their innovativeness, and this could be due to two reasons. One, they did not have adequate exposure to change agents, and two; they do not possess the financial resources needed for purchasing new innovations such as GIS. 6.4.4 Social System The Social System in Uganda’s NGO, IO and private sectors had an impact on GIS diffusion within these sectors. This section discusses the effects of the four elements of a social system on GIS diffusion in Uganda’s NGO, IO and private sectors. These include (1) social structure, (2) opinion leaders and change agents, (3) types of innovation decisions, and (4) consequences of innovations. 6.4.4.1 Social Structure The social structure of NGOs is not as bureaucratic and hierarchical as it is in the public sector. There is room for exploration and maneuverability in the NGO sectors, which is conducive to GIS diffusion. However, the organizational culture in NGOs is 367 characterized by voluntary service and a lack of permanence which is antithetical to diffusion of GIS. In the IO sector, the level of bureaucracy is higher than in the NGO sector. The effect of this is that the diffusion of GIS has been slowed to some extent in the sector mainly because decisions regarding the choice of software and the number of employees to hire for GIS tasks rest with the top management of the IO. However, because IOs are much larger in size compared to NGOs favors diffusion of GIS because of the positive correlation between size of an organization and diffusion (Rogers, 2003). For example, the decision to move from Harvard Graphics and adopt ESRI software at USAID was made by the organization’s leaders, which led to the diffusion of ESRI software in other affiliated NGOs and IOs, such as UNOCHA and NUMAT in Gulu district. Private sector companies that have adopted GIS are mainly multinational corporations, and thus, most of their social structure traits are similar to IOs. However, they exhibit less bureaucracy, are more willing to try new innovations, hire new employees to support the innovation, and provide adequate training. For example, Umeme, a private sector utility company possesses a full blown GIS, including server and mobile GIS, and has a large number of employees that use the technology; while, USAID has only the desktop version of ArcGIS, and has only five GIS employees to support affiliated NGOs in the entire country. The difference in the social structure of the two organizations is clearly different. 368 6.4.4.2 Opinion Leaders and Change Agents The role played by opinion leaders and change agents is important in GIS diffusion. Certain opinion leaders within NGOs, IOs, and private companies promoted the use of GIS software in their organizations, for example, high ranking officials at USAID and UNOCHA promoted the use of, first Harvard Graphics, then, ESRI software; at UNICEF, the use of open source Quantum GIS software was promoted by an organizational culture of norm within the organization that promotes the use of open source software; at NUMAT, MapInfo GIS software was introduced by an expatriate at the start of the project; and at IDI, ArcView GIS software was introduced and encouraged by the first director of the institute. Opinion leaders themselves are the first point of contact for change agents working for companies marketing their GIS software, for example, GeoInformation Communication, the ESRI local business partner in Uganda. GIC is an influential change agent that has been contacting various opinion leaders in NGOs, IOs, and private companies. The opinion leaders tend to be managers and generally at the top levels of organizational hierarchy. A social system, like an organization, will be more responsive to an innovation if it is supported by an opinion leader (Rogers, 2003). 6.4.4.3 Types of Innovation Decisions The types of innovation-decisions in Uganda’s NGO, IO and private sectors have mostly been authority innovation-decisions, and not so much collective or optional innovation- decisions. Authority innovation-decisions are made by those in a social system that 369 possess power, status and technical expertise, and usually, it is a small group of powerful individuals in the system (Rogers, 2003). In the NGO sector, the few that have adopted GIS had the technology introduced into their organizations by an employee who had knowledge about GIS when he/she joined the organization, for example, this was the case at Nature Uganda. In the IO sector, the decision to adopt GIS was made at the top of the management hierarchy, for example, at USAID and UNOCHA. Similarly, in the private sector, the decision to adopt GIS in many companies was made by top management, for example, at WE-Consult, MTN Uganda, and Umeme, the electricity distribution company. 6.4.4.4 Consequences of Innovations The consequences of GIS adoption in Uganda’s NGO, IO and private sectors are mainly positive. Desirable consequences of GIS adoption include an increase in efficiency in the three sectors. Umeme is able to manage its billing for power distribution more efficiently due to GIS. Nature Uganda is able to use GIS in carrying out research on the spatial distribution of birds in Uganda, promoting biodiversity conservation. USAID is able to meet its goal of keeping track of its projects in line with its principle, “Who does What Where” (the 3 Ws) using GIS. Next, I perform an analysis of the results through the lens of the various perspectives of the GIS and Society framework (McMaster & Harvey, 2010; Nyerges et al., 2011). 370 6.5 The State of GIS in Uganda: A GIS and Society Perspective There are five perspectives on GIS and Society; these include: (1) critical social theory, (2) institutional, (3) legal and ethical, (4) intellectual history, and (5) public participation perspective (McMaster & Harvey, 2010; Nyerges et al., 2011). In this section, I examine the impact of GIS on society in Uganda as a result of GIS diffusion in the NGO, IO, and private sectors based on four of the five perspectives listed. The public participatory perspective is not applicable to the three sectors due to a lack of data showing evidence of Public Participatory GIS (PPGIS) in the IO, NGO and Private sectors. 6.5.1 Critical Social Theory Perspective The critical social theory perspective addresses various issues in the relationship between GIS and society; some of these include: (1) the marginalization and disempowerment of society related to GIS use by powerful institutions, (2) access to GIS and geographic information by society, (3) limitations of current GIS representations of populations, locational conflicts, and resource distribution, (4) data confidentiality, and (5) the benefits of GIS to society (McMaster & Harvey, 2010; Nyerges et al., 2011). This section discusses (1) marginalization, empowerment, and society’s access to GIS technology, (2) citizen participation in GIS data collection, and (3) benefits of GIS to society. 6.5.1.1 Marginalization, Empowerment and Access to GIS NGO, IO and private sector institutions’ use of GIS in Uganda has not been inclusive of society in terms of giving society access to GIS technology and geospatial data. As a result GIS has not empowered society in this respect, however, it has not necessarily led to the marginalization of society, either. This is because there is some degree of indirect 371 interaction between GIS in the NGO, IO and private sectors and society. For example, NGOs, IOs and private companies often carry out field work collecting geospatial and attribute data using handheld GPS devices. During reconnaissance and needs assessment surveys, field operatives interact with local citizens to establish the locations of geographic features of interest for topographic and thematic mapping. 6.5.1.2 Citizen Participation in Data Collection Various IOs, NGOs and private sector organizations involve the local population in data collection for GIS. In this section, I provide five examples. UNOCHA UNOCHA in Gulu performs data collection of point features in northern Uganda, such as, water wells, boreholes, health centers, schools, and so on. During the survey, local community leaders and members act as guides because they know their lands best, and also provide other related useful information regarding the current condition of built infrastructure, for example, whether certain water wells are operational or have dried up. SPRING SPRING, a USAID funded project, in Gulu consults with local communities in northern Uganda before deciding where to construct food storage warehouses. Factors that might affect the exact location of these structures include current family structures in the communities, priority being given to locations close to families consisting of vulnerable women and children, and child-headed families. Certain social issues have to be taken into account regarding land wrangles as a result of previously displaced groups of people 372 returning from internally displaced people’s (IDP) camps to reclaim their customary lands. UNICEF UNICEF in Kampala actively consulted with district health officials during their RapidSMS project. This project developed mobile phone-based GIS-integrated technologies to help government and NGO partners monitor health services in local communities. Technologies were developed to perform integrated community case management, support tracking of supplies, improve rates of antenatal visits, and facilitate transfer of medical results and behavioral change messaging. Further, these mobile phone technologies support social monitoring on issues affecting local communities, for example, monitoring secondary school teacher and student absenteeism, and water point functionality. MTN Uganda MTN Uganda interacts with local communities when siting new base stations on hilltops, and also when analyzing network coverage. Local residents are more familiar with the communities they live in than the company’s engineers, thus, they act as local guides. Moreover, the company has to engage in local land acquisition deals with owners of the land on which cell phone towers are to be sited. The local people also provide information regarding the strength of the mobile service provider’s signals in their residential areas, which serves as a check on visibility analysis done in the office using GIS. 373 Tullow Oil Uganda Tullow Oil Uganda’s operations in western Uganda involve oil prospecting and drilling. During the course of their operations, field operatives consult with the local population while carrying out environmental impact assessment of oil prospecting and drilling sites. Often times, located on the sites are graves of relatives, and sites of cultural importance to the local communities. So as to acquire sites for prospecting and drilling, the company needs to work with the local land owners to work out appropriate compensation for their land. 6.5.1.3 Benefits of GIS to Society With reference to the social critical theory perspective, based on the examples given above, there is some indirect benefit that society gets from GIS in the form of pre-GIS- analysis interactions between the local population and the NGO, IO and private sector. Challenges remain regarding the extent to which GIS represents unique cultural phenomena adequately in final presentations of thematic maps to authorities in power and decision makers. Further, society’s access to GIS technology and geospatial data is limited as this research did not find any evidence of actual community-based GIS usage. Thus, GIS does not necessarily empower communities directly, however, there is no evidence to show that GIS marginalizes society in Uganda, either. 6.5.2 Legal and Ethical Perspective The legal and ethical perspective is concerned with the ethical use of GIS with respect to its impact on society (McMaster & Harvey, 2010; Nyerges et al., 2011). Issues addressed under this perspective include pricing mechanisms for geospatial data, intellectual 374 property rights of data producers, accuracy of data, privacy of subjects, and surveillance (McMaster & Harvey, 2010; Nyerges et al., 2011). Issues of interest to the Ugandan case, with respect to the NGO, IO and private sectors include (1) intellectual property rights, geoinformation policy and SDI, (2) data accuracy, and (3) privacy of citizens. 6.5.2.1 Intellectual Property Rights, GeoInformation Policy, and SDI Hardly any local NGOs collect their own primary geospatial data in Uganda due to a lack of adequate financial resources – in most cases they acquire data from government agencies such as NFA and from academic institutions such as MUIENR, for instance, Nature Uganda and ACODE acquire data from MUIENR; and Uganda Land Alliance outsources its GIS and mapping activities to the Department of Surveys and Mapping. The IO and private sectors, on the other hand, are better equipped, financially and personnel-wise, to collect their own geospatial data, which they do using GPS handheld devices and using mobile GIS, for example, Umeme collects locations of power utility infrastructure using ArcPad. However, base data is acquired from the usual government agencies that are data producers, such as UBOS, Department of Surveys and Mapping, NFA, the Wetlands Management Department and MUIENR. A critical source of data for NGOs and IOs is the SDI initiative, Uganda Clusters (UgandaClusters, 2012). Intellectual property rights regarding data, and the related pricing mechanisms have been a source of controversy sowing “seeds of discord” (Gowa 2009, 6) since the early 90s when GIS was first introduced in Uganda at NEIC. NGOs and IOs have lamented the high prices to acquire data from the Department of Surveys and Mapping, and some have 375 even challenged the legal right of a government agency to charge a fee for geospatial data that would normally be considered public data because it was acquired using public funds. To date, however, there exists no geoinformation policy and SDI remains a challenge in Uganda (Ernest, 2010; Kalande & Ondulo, 2006; Musinguzi et al., 2004, 2010; Uwayezu, 2010). It is due to the lack of a geospatial one-stop that Uganda Clusters, an SDI initiative by local NGOs and IOs was launched to provide these organizations with critical data for them to be able to carry on with humanitarian and environmental conservation work. IOs and NGOs are more willing to share their data among each other than are public sector agencies. Private sector agencies do not tend to share their data for the sake of privacy and intellectual property rights issues. 6.5.2.2 Data Accuracy Accuracy of data, due to a lack of enforced geospatial data standards poses a problem to interoperability (Musinguzi et al., 2010) among agencies. WE-Consult, a private environmental and water engineering company complained about the tendency of data from UBOS, Uganda Clusters, and the Department of Surveys and Mapping not “tying up” correctly (Y. Tylle, WE-Consult, personal communication, 2nd August 2010). This could possibly be due to differences in data scale and data acquisition methods at the various data producing agencies. 6.5.2.3 Privacy of Citizens Privacy of personally identifiable geospatial data is of some concern in the NGO, IO and private sectors, however, the threat of this occurring is minimal. This is mainly due to the 376 fact that the identity of individuals based on database design concepts has not yet been implemented in Uganda. 85% of the population lives in rural areas (UBOS 2010, xi), civil registration of citizens is inadequate (Setel et al. 2007, 4), and the concept of a single number identifier of an individual, similar to the social security number in the United States, only applies to middle class citizens in the formal employment sector, which is a minority of the population (less than 15%). 6.5.3 Institutional Perspective The institutional perspective of GIS and Society addresses issues such as (1) the costs and benefits of GIS implementation within institutions, (2) equity in distribution of costs among social groups and (3) maintenance of geospatial data by organizations (McMaster & Harvey, 2010; Nyerges et al., 2011). Further, it examines the interactions between government agencies, and society with regard to the use of GIS, for example, for land management (McMaster & Harvey, 2010). This section discusses the following issues with respect to the institutional perspective in Uganda’s IO, NGO and private sectors: (1) costs and benefits of GIS implementation, (2) degree of institutionalization of GIS, (3) political implications of GIS, and (4) the role of GIS in land management. 6.5.3.1 Costs and Benefits of GIS Implementation The implementation of GIS in the NGO sector has been hampered by the high cost of GIS licenses, and software. This could explain the small number of local NGOs using GIS in their workflows. For those that are indeed using GIS, they prefer to implement the free ArcView 3.x version, for example, Nature Uganda. NUMAT, on the other hand, can afford to implement the more expensive ArcGIS software because it is supported by 377 funding from USAID. NGOs such as NAPE, an environmental advocacy NGO, are not even aware of the existence of GIS technology. The IO and private sectors, on the other hand are not constrained by financial support for GIS, and most of these institutions have ArcEditor license level ArcGIS installations in their organizations. However, again, the financial aspects regarding GIS implementation cannot be ignored. All IO budgets are financially supported by their parent countries in the West, while all private sector companies that can afford to implement GIS tend to be multinationals and not local companies. 6.5.3.2 Degree of Institutionalization of GIS The level of institutionalization of GIS in the IO and private sector depends on the degree to which GIS is considered critical for an institution to carry out its tasks. For example, the USAID SPRING project uses MapInfo mainly for the purpose of mapping locations of its food warehouses, and not for any substantial spatial analysis, while USAID uses GIS to keep track of all its projects in Uganda. In the former case, GIS is seen as an add- on to the IT department’s activities, and thus, there is no need to hire GIS specialists and organize regular GIS training. In the latter case, however, GIS has been institutionalized as a department and five employees hired as GIS analysts, and these undergo regular training. 6.5.3.3 Political Implications of GIS Some environmental advocacy NGOs, such as ACODE, prefer not to use GIS in their organizations because this could create tensions between the NGOs and the ruling government. The results of GIS analysis could contradict official government findings, 378 for example, only analysis based on data from UBOS, the government statistics agency under the Ministry of Finance, is considered official and analysis based on primary data collected by NGOs is regarded as “hearsay” possibly leading to an NGO being labeled as siding with the opposition parties. This could result in backlash for NGOs that dare contradict government research. 6.5.3.4 Role of GIS in Land Management Currently, Uganda’s land cadastre is only partially digitized. It is estimated that only 15- 20% of the total land in Uganda has been formally surveyed and recorded in the land registry (N. Batungi, Makerere University Kampala, personal communication, 26 th June 2010). However, the digitization of land records in Uganda will eventually be done under an ongoing World Bank funded project in the Ministry of Lands, Housing and Urban Development in which the creation of a Land Information System is one of the major objectives (MLHUD, 2011a). This LIS will be based on customized GIS technology from ESRI (R. Oput, Ministry of Lands, Housing and Urban Development, personal communication, 16 th July, 2010). Currently, GIS is used for land management in the Department of Surveys and Mapping only if the client is a big land owner, for example, the Sugar Corporation, Kakira Sugar. In this case, GIS aids in quick plotting and computation of land acreage. In all other cases, surveyed land is simply hand drawn by draughtsmen in the drawing office and stored in paper format – and only in those cases where the land owner can afford these 379 services which implies that not all land that is surveyed is necessarily registered in the land registry. Moreover, land tenure in Uganda is characterized by four different kinds of tenure: (1) Freehold, (2) Leasehold, (3) Mailo, and (4) Customary land tenure (GoU, 2008; Okuku, 2006). The Mailo and Customary land tenure systems are not based on any formal GIS coordinate system, which creates challenges in terms of digital representation in GIS. NGOs such as the Uganda Land Alliance are working with local communities in the rural north eastern parts of Uganda (the Karamoja region) to register communal tribal lands under customary land tenure system so as to promote productivity of the land, and to empower the local marginalized pastoral nomadic tribal communities. Because this NGO does not possess the necessary skills in surveying and mapping, this task has been outsourced to the Department of Surveys and Mapping (J. Mwebe, Uganda Land Alliance, personal communication, 1 st July, 2010). 6.5.4 Intellectual History Perspective The final perspective examined in this section is the intellectual history perspective. The evolution of GIS in Uganda’s NGO, private, and private sectors is linked to external influences of the GIS industry in industrialized nations. As mentioned earlier, penetration of GIS in the NGO sector is very minimal; however, all IOs in this research employ GIS in their work. The fact that IOs are based in developed countries supports the thesis that the introduction of GIS at the Ugandan-based branches of these organizations was influenced by change agents in their parent countries. 380 ArcGIS software is by far the most widely used software amongst IOs in Uganda, and this is because of the working relationship developed by ESRI with government and non- governmental agencies in the United States and Europe since the 1980s (for example, see Dangermond and Smith 1988). This same influence is characteristic of ESRI’s software dominance in the private sector because institutions in the private sector that use GIS are mainly multinational corporations and not locally owned companies. 6.6 Conclusions The diffusion of GIS in the NGO, IO and private sectors in developing countries is strongly related to support and funding of these institutions from industrialized countries. The NGO sector has had the least diffusion of GIS mainly because local NGOs are unable to afford software licenses, and because of lower levels of interaction with change agents. The IO and sector has the highest levels of penetration among the three sectors because of the inherent financial and technical support from their parent countries in the West, and also due to higher levels of interaction with change agents promoting GIS adoption. The private sector has relatively high levels of adoption, however, this is only the case with multinational corporations operating in Uganda, and the reasons behind the adoption rates are similar to the IOs. GIS has not diffused to locally owned companies mainly because of a lack of awareness about GIS, the absence of a business model to support GIS, a lack of qualified GIS personnel to employ, and high costs of acquiring GIS licenses. Overall, there is a strong link between GIS diffusion in the NGO, IO and private sectors and links of these institutions with industrialized countries; the stronger 381 the link, the higher the adoption rate. The findings of this research are significant for future implementation of GIS projects by the NGO, IO and private sectors in developing countries, and also, for international donor agencies funding development projects, such as the United States Agency for International Development (USAID). Further research is needed to identify interventions for promoting partnerships to create inter-linkages among the various sectors in developing, for example, public-private partnerships, so as to promote the diffusion of GIS across sectors. 382 Chapter 7 : The Future of GIS in Developing Countries: The Potential for Mobile Location-based Services in Uganda 7.1 Introduction There is a growing body of literature on the use of Information and Communication Technology for development (ICT4D) and mobile communication technology for development (M4D), (for example see Kizza et al. 2010, Tian, Shi, and Yang 2009; Hellstrom 2010; Donovan and Donner 2010; Wicander 2010; WorldBank and InfoDev 2012; WorldBank 2010; McDonald 2011, Chu, Liu, and Wu 2010; Tian, Shi, and Yang 2009; Arminen 2007; Isomursu et al. 2011). Information and Communication Technology (ICT) and mobile communication technology have played significant roles in improving livelihoods in developing countries where a large number of people live on less than a dollar a day (WorldBank & InfoDev, 2012). Mobile software applications have been developed to address various problems in developing countries, especially in the health sector (Hellstrom, 2010; ICITD, MUBS, & PC-TECH, 2010), the agricultural sector (Hellstrom, 2010; Martin & Abbott, 2008) and social development sector (UNICEF, 2012a). Other innovative uses of mobile phones in developing countries include mobile money transfer (Kiyaga, 2012; Morawczynski, 2008), and mobile banking (Donner & Tellez, 2008; Hannes, 2011). Mobile technologies have been linked economic growth, and increase in productivity, and thus improving 383 livelihoods of the poor (Banks, 2009; Banks & Burge, 2004; Bayes, von Braun, & Akhter, 1999; Donovan & Donner, 2010; Hellstrom, 2010; WorldBank, 2010). In Africa, mobile phone usage covers 60% of the population (Aker & Mbiti, 2010). Mobile phones have the potential to improve consumer and producer welfare in the economies of African countries (Aker & Mbiti, 2010). Mobile phones provide incremental benefits, such as savings in time and costs of travel; transformational benefits from services such as mobile-money transfer; and production benefits, such as job creation, for example, selling prepaid airtime scratch cards (Wicander, 2010). Some authors have criticized ICT4D and M4D research for overestimating the impacts of mobile phone technology and have proposed the use of better measures than simply mobile phone service coverage area, and the number of subscribers (James & Versteeg, 2007). Others have criticized mobile phone technology itself as leading to the shrinking of social networks, rather than enhancing them (Shrum et al., 2011). On the contrary, some have lauded mobile phones as “citizen sensors” platforms for empowerment (Yola Georgiadou, Bana, et al., 2011). In fact, the importance of mobile phones has been pointed out as a relatively new trend in GIS – Mobile GIS (Akingbade, Navarra, & Georgiadou, 2009). There are quite a number of mobile phone applications specifically in Uganda, the case study in this paper, that have been discussed in the literature. There has been a lot of innovation in the mobile health (m-health), social development, and agricultural sector, not to mention the private sector. Mobile applications have been developed to help people in rural areas find their nearest health centers via SMS, get a diagnosis from a doctor 384 located tens of miles away through the intervention of a community health worker with a smart phone, get tips on organic farming methods to improve crop yields, and get information about sexual and reproductive health via SMS (Blaschke & Weikel, 2010; GrameenFoundation, 2010; Hellstrom, 2010; Kiyaga, 2012; MUK-FCIT, 2010; MobileHealthUganda, 2010; Muturi, 2008; NewAgriculturist, 2009; Pfaff, 2010; UNICEF, 2008, 2012b). Mobile software application development is relevant to GIS diffusion in developing countries because of its application in Spatial Data Infrastructure (SDI), Volunteered Geographic Information, and neogeography (Foth, Bajracharya, Brown, & Hearn, 2009; Yola Georgiadou, Budhathoki, et al., 2011; Goodchild, 2009; Goodchild, Johnson, Maguire, & Noronha, 2005; Hudson-Smith, Crooks, Gibin, Milton, & Batty, 2009; Rana & Joliveau, 2009; Raubal, 2011; J. Smith & Kealy, 2003; Sui, 2011; Zook et al., 2010). Mobile phones provide a platform for collecting data in remote areas of the world, especially in system-poor environments typical of developing countries. Mobile applications that specifically leverage location are classified as location-based services (LBS). 385 Figure 7-1: LBS Concept The concept of LBS is shown in Figure 7-1. A user can use his or her mobile phone to make a location-specific query to a server over a telecom provider’s wireless network, such as “where is my nearest gas station,” and receive an instant response over the same network. The location of the mobile device can be determined based on GPS capability on the mobile phone, or through s network-based positioning service provided by the telecom provider. Content to that is drawn upon for a location-specific response is provided by various third-party content providers. 386 LBS have found application in serving as mobile tourist guides, navigation, fleet management, vehicle and people tracking, location games, indoor guidance systems, and many others (Blankenbach & Norrdine, 2011; Curran et al., 2011; Espeter & Raubal, 2009; Hudson-Smith et al., 2009; Ludford, Frankowski, Reily, Wilms, & Terveen, 2006; Paay & Kjeldskov, 2008; Raper et al., 2007a; Schmid, 2008; Timko, 2006; Wac & Ragia, 2008). LBS applications have had an intriguing impact on the lives of people in the developed world (Raubal, 2011), however certain risks need to be considered such as geoslavery and privacy (Dobson & Fisher, 2003; Raubal, 2011), and in fact Raubal (2011) calls for further research on the effects of LBS on society. In this chapter, I investigate the potential for developing and utilizing LBS software applications in developing countries, using Uganda as a case study. I seek answers to the third research question in this dissertation research: What is the current state of ICT and Mobile telecom infrastructure in Uganda, and what are the major technological, socio- economic and political challenges to the development and use of LBS technology? In addition, what are some innovative ways in which local citizens use their mobile phones for spatial decision making today? What are some of the current technology trends with respect to LBS technology in developing countries? Answers to these questions allow for the assessment of the potential for LBS in Uganda as alternative tools to GIS for promoting spatial thinking, spatial awareness, and for supporting spatial decision making. The findings are also significant for guiding future research, development and implementation of LBS systems and applications in 387 developing countries. I use a mixed methods approach to research, and Diffusion of Innovations theory to analyze the data. I argue that LBS can be used as a tool for instilling spatial thinking in society, especially in developing countries which tend to lack adequate ICT infrastructure, but where mobile phones are ubiquitous. I further argue that LBS can make an impact in the field of GIS, and Geography broadly, in enhancing spatial thinking (Businge, 2012; Goodchild, 1995, 2008, 2009; Goodchild et al., 2005; Janelle & Goodchild, 2011; NRC, 2002, 2006; Schuurman, 2009). Diffusion of innovations theory can be used to understand the adoption and impacts of LBS on society as has been done by some researchers (Arminen, 2007; Chu et al., 2010; Isomursu et al., 2011; Tian et al., 2009). Finally, I argue that the future of GIS with respect to the collection and use of geospatial information by civil society lies with leveraging mobile devices. In the next section, I present the results and findings. 7.2 Results and Diffusion Analysis Results of the survey and a diffusion analysis of the results will be presented in this section. Diffusion is “the process by which an innovation is communicated through certain channels over time among the members of a social system” (Rogers 2003, 11). First, the level of mobile phone penetration in the Ugandan market is analyzed. This is followed by an analysis of the respondents’ access to Information and Communication Technology (ICT). Thereafter, I examine the innovative ways in which local citizens use their mobile phones for spatial decision making. I then analyze the potential for LBS in Uganda for spatial decision support, delving deeper into the issues based on a case study 388 of a local NGO, the Busoga Rural Open Source Development Initiative (BROSDI). The role of Mobile Telecommunication Companies and content providers in LBS infrastructure is examined next. Finally, I conclude the section by presenting the challenges to LBS diffusion in Uganda and identifying weaknesses of the analysis. 7.2.1 Mobile Phone Penetration To understand the level of penetration of mobile phones in Uganda, information was collected on (1) the background of survey respondents, and (2) indicators of mobile phone trends. 7.2.1.1 Background of Survey Respondents Background information was collected on the survey respondents. This included: (1) professional background, (2) age-range, (3) education level and (4) their districts of residence. Professional background. 28% of the survey respondents were university students, mainly from Makerere University, the largest public university in Uganda; 20% were high school students (see Figure 7-2). The rest of the respondents were from a mixture of professional backgrounds including land surveyors, teachers, farmers, bankers, administrators, Information Technology professionals, and so on. Figure 7-2 shows the distribution of the various professions. Overall, the survey respondents represented a good mix of professional backgrounds, although there is a potential bias that will be introduced in the results because 48% of the sample were either university or high school students. 389 Figure 7-2: Survey Respondents by Profession Age Range. 65% of the survey respondents were in their late teens or twenties; 21% in their 30s and 40s, and only 2% in their 60s. Since over half the respondents in this survey were in their late teens or twenties, there will be a tendency for the results of the survey to be biased towards trends with the younger generation in Uganda (see Figure 7-3). 390 Figure 7-3: Age Range of Survey Respondents Education Level. The sample of survey respondents was well educated: 36% were undergraduates at university pursuing a bachelor’s degree; 24% had attained bachelor’s level education (indicated as Graduate in Figure 7-4), 4% had a diploma level education (Associates Degree); and 4% had a Master’s degree or higher. 28% of the sample was still enrolled in Secondary (High) School. Because 68% of the respondents had university level education, the results will be biased towards the more educated segment of Ugandan society (see Figure 7-4). Figure 7-4: Education Level of Survey Respondents Districts of Residence. 57% of the respondents reside in the Capital City, Kampala in the south eastern part of Uganda; 24% of them reside in Mayuge district, a small rural town in eastern Uganda. 8% of the respondents reside in Soroti, a rural town in eastern Uganda, while 5% in the surrounding rural towns of Kaberamaido, and Katakwi. 391 Holistically, 61% of the respondents are from urban towns, that is Kampala, Mukono, and Jinja, while 38% from rural towns, that is, Soroti, Mayuge, Iganga, Katakwi, and Kaberamaido. Because of the larger proportion of the sample being urban, the analysis will be biased towards trends in in urban areas of Uganda (see Figure 7-5). Figure 7-5: Districts/Towns of Residence of Survey Respondents 7.2.1.2 Indicators of Mobile Phone Trends Trends in the mobile phone market in Uganda vary in terms of (1) the number of handsets owned, (2) brand, (3) year of manufacture, and (4) technological capabilities of the mobile phones owned. Number of Mobile Phones Owned by an Individual 77% of the respondents owned only one mobile phone; 16% owned 2 mobile phones; 3% owned 3 mobile phones; 1% owned 4 mobile phones; and only 3% either did not own a 392 mobile phone or the number of phones owned could not be established. Thus, 97% of all respondents owned at least one mobile phone (see Figure 7-6). According to the data, one in five individuals owns at least two mobile phones. This might seem peculiar in the West, but this is a common trend in Uganda because mobile phone services in Uganda are mostly prepaid, and cross-network calling is prohibitively expensive. Post-paid monthly plans are only offered by telecommunication companies to corporate customers, mainly for fixed lines in offices. In an attempt to capture the market, the five telecommunication companies operating in Uganda, Mobile Telephone Networks (MTN), Orange, Warid, Zain (recently acquired by Bharti Airtel), and Uganda Telecommunications Limited (UTL), have made cross- network calling prohibitively expensive. This has led to customers buying two or more different phone lines corresponding to the networks of their friends and family members – this reduces cost by ensuring same-network calling. Moreover, the monthly cost based on the prepaid model is generally higher compared with the post-paid monthly billing model which is commonplace in United States and Europe (with a fixed of minutes per month). One of the major reasons for the lack of monthly plans in developing countries is the lack of infrastructure to support a small-scale credit-based economy. The infrastructure to support small-scale credit transactions simply does not exist, for example, there is no official street address system (gazetteer) in place to identify customer residences, and no infrastructure to support a credit card system. 393 Diffusion Analysis: As a result of these limitations, mobile phone users in Uganda have had to re-invent the perceived usage of mobile phones by owning multiple phones, one for each network that their friends and relatives are subscribed to. This allows for a huge amount in savings in voice call and text messaging (SMS) costs across networks. Re-invention through customization of mobile phone subscription patterns has led to greater diffusion of mobile phones in Uganda leading to an increase in handset sales and network subscriptions. Figure 7-6: Number of Mobile Phones owned by Each Survey Respondent Mobile Phone Brands 58% of the survey respondents owned a Nokia mobile phone, while all the other brands, such as Samsung, Motorola and Sony Ericsson, almost equally shared the rest of the handset market (see Figure 7-7). The reason for Nokia’s dominance in Uganda could be that the Finland-based company developed effective networks with local distribution 394 partners in Africa, and even does its own market research (Ewing, 2007) employing anthropologists to study market trends, for example Jan Chipchase (S. Corbett, 2008). Diffusion Analysis: The domination of Nokia handsets in the Ugandan market can be attributed mainly to two perceived attributes of the mobile phone as an innovation: (1) observability, and (2) relative advantage. Because of the early widespread use of Nokia phones in Uganda, the observability of this particular phone model was much higher than the other phone brands trying to penetrate the Ugandan market. Moreover, Nokias were perceived to have an advantage over the other phones in terms of user friendliness and a more intuitive user interface, as compared to some of the other brands, such as Ericsson and Motorola. Figure 7-7: Mobile Phone Brands used by Survey Respondents 395 Mobile Phone Year of Release 20-27% of the survey respondents owned one or more phone models released between 2006 and 2009, 9% owned a 2005 phone model, while 19% owned an older 2003 model (see Figure 7-8). This shows that there is a tendency for the respondents to own both older and newer models of cell phones; however, the majority owns a phone that is five years old or newer. The older cell phone models were typically found in the rural areas, and the newer models in the urban. The year of manufacture of the cell phones is significant with respect to LBS because that has an implication for the technological capabilities of the handsets, for example, the availability of GPS and Internet connectivity. Figure 7-8: Year of Release of Mobile Phone Models Owned by Survey Respondents 396 Diffusion Analysis: This trend in high percentage ownership of recent models of mobile phones seems to indicate a high level of innovativeness. Many mobile phone users tend to be innovators and early adopters of new cell phone models. This trend could be driven by the fact that the sample used in this research could have been strongly influenced by high education levels, socio-economic status and cosmopoliteness of the mobile phone users surveyed (see section 7.2.1.1 Background of Survey Respondents), which confirms Rogers (2003) theory on diffusion. In this survey, respondents were unfortunately not asked to provide the year they purchased each of their handsets rendering a plot of the S-shaped diffusion curve difficult. However, based on an assumption that mobile users could only have purchased a handset in the year of its release or after, a diffusion curve can be estimated as shown in Figure 7-9. This curve is based on cumulative percentages of125 handsets whose models were declared by 182 respondents during the survey. 397 Figure 7-9: Diffusion Curve of Mobile Phone Adoption in Uganda Due the approximation technique used, the actual percentage values cannot be used in a diffusion analysis. Despite this limitation, the shape of the curve still provides an insight into the diffusion trend with respect to mobile phones in Uganda. The curve approximately conforms to the early and middle-stages of Rogers’ (2003) diffusion curve, which indicates an increasing rate of cellphone adoption in the country from 2005 to 2009. Further, an inflexion point, which can be inferred as the take-off stage of diffusion, seems to be between 2004 and 2005. Based on this curve, one can conclude that mobile phone diffusion in Uganda is seems to be in full throttle. 398 Mobile Phone Technological Capabilities 100% of the cell phone models were 2 nd Generation GSM network, and SMS (Short Message Service) capable; 51% were capable of Internet connectivity (data capable). All phones capable of Internet connectivity featured Mobile WAP/HTML/XHTML browsers. The data connectivity for these phones was through GPRS technology, and in some cases (45%), EDGE technology. 32% of the phones were WiFi/WLAN capable. 3 rd Generation capability was present in 24% of the mobile phones, while GPS capability was only to be found in 13% of the phones. Internet connectivity and GPS capability are the two important ingredients for running location-based services applications on mobile handsets. Because of the diverse platforms of mobile phones, the presence of Java capability on 51% of the mobile phones in the sample is significant because it implies that the Java platform would be well suited to LBS application development in Uganda (see Figure 7-10). 399 Figure 7-10: Technological Capabilities of Mobile Phones Owned by Survey Respondents Diffusion Analysis: About half of the phone models (declared) were capable of advanced data connectivity (Internet/Web access) through 2.5G, and one quarter through 3G technology, while only about one in ten were had GPS location capability. This implies that the perceived attributes of mobile phones, in this case, (1) the relative advantage, and (2) compatibility with users’ mobile lifestyles, tends to encourage adoption of technologically advanced mobile handsets. Users are able to surf the Web wirelessly, which is more convenient than visiting an Internet café, as was the case in the past. However, the GPS functionality seems to be redundant as there is no evidence of a perceived relative advantage of a handset with GPS capability, over one without (see 7.2.3 Innovative Uses of Mobile Phones for Spatial Decision Making). However, this fact 400 supports the data in section 7.2.1.1 Background of Survey Respondents) that innovators and early adopters of innovations tend to be those in higher socio-economic status and cosmopoliteness. Comparison of Technological Capabilities by Year By 2005, the first smart phones were emerging on the global mobile phone market. Table 7-1 shows the year of release of some of the earliest smart phones carried by survey respondents in this research. Starting in 2005, mobile phones were capable of Internet connectivity, for example, the Nokia N70 that had a mobile WAP/HTML browser on board. This cell phone model was also capable of high speed data download speeds characterized by the 3 rd Generation (3G) mobile telecommunication standard. Between 2005 and 2010, more advanced cell phone models started emerging on the Ugandan consumer market, mainly Nokia and Blackberry phones. The internal memory, Central Processing Unit (CPU) speeds, and phone display sizes started increasing incrementally every year. GPS capability was a standard feature of smart phones starting in 2007. All these features support the development of LBS applications, especially for the standard Java mobile platform which is a standard feature on most smart phones. 401 Table 7-1: A Comparison of Technological Capabilities of Smart Phones owned by Respondents Based on Year of Release Year of Release Phone Brand and Model Internal Memory CPU - Speed Proce ssor OS GPS Bro wser Java 3G WLA N 2005 Nokia N70 22MB 220M Hz ARM 9 Symbian OS Yes Yes Yes 2006 Nokia N73 42MB 220M Hz Dual ARM 9 Symbian OS Yes Yes Yes Blackberr y 7130 64MB 312M Hz Intel Xscal e Blackber ry OS Yes Yes 2007 Nokia N95 160MB 332 MHz Dual ARM 11 Symbian OS Yes Yes Yes Yes Yes Blackberr y 8300 64MB 312 MHz Intel Xscal e Blackber ry OS Yes Yes Yes 2008 Nokia 5800 XpressMu sic 81 MB 434 MHz ARM 11 Symbian OS Yes Yes Yes Yes Blackberr y 8900 256MB 512 MHz Blackber ry OS Yes Yes Yes T-Mobile G-1 192MB 528M Hz ARM 11 Android OS Yes Yes Yes Yes Yes 2009 Nokia N97 128MB 434 MHz ARM 11 Symbian OS Yes Yes Yes Yes Yes Nokia E75 85MB 369 MHz ARM 11 Symbian OS Yes Yes Yes Yes Yes 2010 Nokia N8 256MB 680 MHz ARM 11 Symbian Yes Yes Yes Yes Yes 7.2.2 Access to ICT In the sub-sections that follow, all the percentage values in the tables are based on a sample of 101 and respondents, and not 182 like in the previous sub-section. This is because only 101 out of 182 respondents completed this part of the survey. Access to Information and Communication Technology (ICT) is analyzed in this sub-section. The relationship between LBS and ICT is that citizens’ access to digital information depends on a country’s ICT infrastructure; and this includes digital geographic data and 402 information. Data was collected in this survey regarding society’s access to hardware, software, and the Internet and Web. 7.2.2.1 History of ICT in Uganda The first mobile telephone network in Uganda was Celtel Uganda, a subsidiary of Celtel International, which started operations in 1995 (AfDevInfo, 2008). At this time, mobile communication was very expensive and unaffordable to the general public. Landline phones were the predominant mode of communication by that time, the landline network being managed and regulated by the government-owned Uganda Posts and Telecommunications Corporation (UPTC) (UNECA-NICI, 2001a). In 1997, the Uganda Communications Commission (UCC) was established under an Act of Parliament, the Uganda Communications Act 1997 (ICTRegulationsToolkit, 2006). This opened the Ugandan telecommunication market to private investors, and soon afterwards, in 1998, a second mobile telecommunication company started operations in Uganda, Mobile Telephone Networks (MTN) (UNECA-NICI, 2001a). Soon after, three other companies followed, Uganda Telecom Limited (UTL) in 2000 (partially owned by the Uganda government), Warid Telecom in 2007 (WaridTelecom, 2012), and Orange Telecom in 2009 (Orange, 2012). Celtel International was bought by the Kuwait-based Zain Telecom in 2005, and then sold to Bharti Airtel in 2010. As of 2010, Zain has been rebranded Airtel (Airtel, 2012). The first Internet Service Provider (ISP) in Uganda was InfoMail, licensed by the UCC (Infocom, 2012; UNECA-NICI, 2001b) in 1997. The second ISP in the country was STARCOM (UNECA-NICI, 2001b). These two companies merged by to form a new ISP 403 called Infocom. By the end of the 1990s, four other ISPs were licensed by UCC: Swift Global, Africa Online, Uganda Online and Bushnet (UNECA-NICI, 2001b). Today, one of the major ISPs in Uganda is iWayAfrica, which was formed by a merger in 2010 between Africa Online (AfricaOnline, 2012; Yeomans, 2012), Afsat Communications Africa, and MWEB Africa (iWayAfrica, 2012). iWayAfrica is a subsidiary of Telkom South Africa. 7.2.2.2 Hardware and Software 26% of respondents owned a desktop computer, 35% a laptop, and 9% both (see Table 7-2). Thus, a total of 52% of the sample owned at least one desktop or laptop computer. Because the sample of respondents consisted of students and professionals, it was not surprising that over 50% of them owned a computer. The most common operating system (OS) on desktop and laptop computers was Windows, and the most common brand was Dell and Toshiba (for laptops only). Another operating system in use was Linux, and other brands included Compaq, HP, Fujitsu Siemens and Gateway. 404 Table 7-2: Hardware and Software of Computers Owned by Respondents Hardware and Software Owned % Owned a Desktop 26 Desktop OS: Windows 15 Desktop OS: Linux 1 Desktop OS: Undeclared 10 Desktop Brand: Compaq 3 Desktop Brand: Dell 12 Desktop Brand: HP 1 Desktop Brand: Undeclared 10 Owned a Laptop 35 Laptop OS: Windows 30 Laptop OS: Linux 1 Laptop OS: Undeclared 4 Laptop Brand: Acer 3 Laptop Brand: Compaq 2 Laptop Brand: Dell 14 Laptop Brand: Fujitsu Siemens 1 Laptop Brand: Gateway 1 Laptop Brand: HP 2 Laptop Brand: Toshiba 7 Both Desktop and Laptop 9 7.2.2.3 Access to the Internet/Web 41% of the survey respondents accessed the Internet/Web in an Internet Café, while 48% accessed it through their University’s WLAN/WiFi network. Because a large portion of the respondents were university students, this result was a bit skewed towards WiFi access. However, WiFi access in urban areas in Uganda is not common place, and Makerere University is one of the few locations that provide WiFi access to its students. 22% of the respondents accessed the Internet at their places of work, while 2% accessed 405 it directly on their Mobile phones via the telecommunication network they were subscribed to via 2.5G technology (GPRS and EDGE). See Table 7-3. Table 7-3: Access of Survey Respondents to the Internet/Web Internet/Web Access % Internet Café 41 WLAN/WiFi (Public University) 48 Portable USB 3G Wireless Modem 19 Mobile Phone Prepaid 2 Office 22 Satellite Connection (Community Based Organization- BROSDI) 5 19% of the respondents accessed the Internet/Web through a portable USB 3G Wireless Modem from a telecom company, for example, through Orange Telecom (see Figure 7-11). This result is very significant as it shows a new trend in public access to the Internet/Web. This portable modem technology became an instant hit with the public upon the arrival of Orange Telecom into the country in 2009. It is capable of delivering data to customers at 3G+ speeds of up to 21 megabits/second (MBPS) (Orange, 2012). For the first time in Uganda, the public was able to access the Internet/Web at speeds comparable to those characteristic of DSL in industrialized countries for about $10 per month. This was unprecedented. MTN Uganda had offered high speed Internet before Orange, however, it was based on the arguably less popular Wi-MAX technology (MTN, 2012a), which typically requires an antenna to be installed on the roof of a user’s house. This technology is not particularly attractive to the mobile user because of its lack of portability, not to mention initial costs associated with hardware installation. The popularity of the wireless USB dongle-based mobile Internet can be explained by the fact that a high percentage of the 406 urban population now owns a laptop; 35% of the respondents in this survey (see Table 7-2). Portable high speed wireless modems from UTL (UTL, 2012), Orange (Orange, 2012), Airtel (Airtel, 2012) and Warid Telecom (WaridTelecom, 2012) are wildly popular with customers because of this boom in the laptop market. Figure 7-11: Portable USB Wireless 3G Modem (“Internet Everywhere”) from Orange Telecom, Uganda Source: (Orange, 2012) One limitation with 3G technology at the moment is that geographical coverage by most of the mobile telecommunication networks is limited to urban areas, unlike 2G which covers most of the entire country, for example, Figure 7-12 shows Airtel’s geographical coverage of their 2G and 3G networks. MTN’s national coverage is shown in Figure 7-13. The only company providing 4 th Generation (4G) wireless mobile technology in Uganda by August 2010 was Foris Telecom, owned by an Israeli Investment Group (ForisTelecom, 2012). This is an Internet Service Provider (ISP) that utilizes WiMAX broadband Internet Technology at 4G speeds. It first started operations in Uganda in 2009 407 (ForisTelecom, 2012; Matthews, 2010; Mizrachi, 2009). None of the respondents of this survey used services provided by this company, however, possibly because it was still a very new and little known company. Moreover, WiMAX technology would not appeal to mobile users, but rather, to home Internet users and corporate companies who were not the focus of this survey. 408 Figure 7-12: Airtel Telecommunication: 2G and 3G Coverage in Uganda Source: (Airtel, 2012) 409 Figure 7-13: MTN Geographical Network coverage of Uganda as of July 2009 Source: (MTN, 2012b) 410 7.2.2.4 Internet Service Providers 5% of the survey respondents indicated MTN as their ISP, 6% Orange Telecom, 8% Uganda Telecom, 6% Warid Telecom, and 1% Zain Telecom (today known as Airtel) (see Table 7-4). The type of Internet access used by respondents from these five mobile companies is mobile broadband wireless via a portable 3G portable modem. In a few instances, some respondents indicated that they accessed the Web directly on their phones, but this was only on rare occasions, for example, if there was an on-going promotion by a telecom company, such as Orange Telecom’s free access to Facebook promotion in 2010. 3% of the respondents indicated iWayAfrica Uganda as their ISP for fixed Internet access in the office. Table 7-4: Internet Service Providers Used by the Survey Respondents ISP % Mobile Telephone Networks (MTN) 5 Orange Telecom 6 Uganda Telecom (UTL) 8 Warid Telecom 6 Zain (Airtel) 1 iWayAfrica Uganda 3 7.2.2.5 Reasons for Accessing the Internet/Web 59% of the respondents accessed the Internet/Web for work-related reasons, 56% for information related to their school or university education, 26% for accessing their email, Facebook and surfing the Web for information, 22% for online games, 15% to listen to and download music, 14% to watch movies, and 1% to access free Short Message Service (SMS) or text message services (see Table 7-5). The highest percentages relate to work- and education-related activities which underscore the importance of the Internet/Web in supporting citizen livelihoods and intellectual development. 411 Table 7-5: Reasons for Accessing the Internet/Web Reasons for Accessing the Internet % Work 59 Study 56 Email/Facebook/Surfing the Web 26 Games 22 Music 15 Movies 14 Free SMS/Texting 1 7.2.2.6 Cost of Accessing the Internet per Month 11% of respondents indicated that they paid $1 - $10 per month for accessing the Internet/Web, 3% indicated $11 - $30, and 5% indicated $11 - $40 (see Table 7-6). These respondents access the Internet/Web via a portable 3G/3G+ wireless USB modem (on their desktop or laptop computers). The wireless USB modems are initially purchased from their provider for about $20, and then a monthly prepaid subscription is paid to the provider for a given amount of data per day or per month, for example, Orange Telecom offers one such package, 0.6GB of data for 1 month at a prepaid cost of $10; the cheapest package is 30MB of data for 1 day at a prepaid cost of $0.20 (Orange, 2012). Table 7-6: Cost of Accessing the Internet per Month for Survey Respondents Cost of Accessing the Internet Per month ($) % $1 - $10 11 $11 - $20 1 $21 - $30 2 $31 - $40 5 7.2.2.7 Affordability of Internet/Web Access 35% of the respondents were of the opinion that the cost of accessing the Internet/Web was too high and unaffordable, while 17% thought that it was affordable at the current mobile provider rates. 48% of the respondents did not respond to the question possibly 412 because they did not access the Internet/Web, or accessed it free of charge at their offices, or through a community based organization’s social center, such as the Busoga Rural Open Source Development Initiative (BROSDI) in Mayuge district. The fact that the majority of respondents who stated that the Internet was unaffordable was twice the number that said that it was affordable reflects the public’s sentiment toward the high cost of Internet services in Uganda. One of reasons for the high cost of Internet services to the general public in Uganda is the expensive prepaid model employed by all ISPs. However, this model is employed precisely because of a lack of a consumer credit system in the country, and the related infrastructure to support it, for example, a central citizen identification system based on a unique identifier, such as the social security number in the United States. Table 7-7: Opinion of Survey Respondents: Is the Cost of Accessing the Internet affordable? Opinion: Is the cost of accessing the Internet/Web Affordable? % Yes 17 No 35 No response 48 7.2.2.8 Mobile Internet/Web Usage Trends 23% of the respondents indicated that they access the Internet to access the social networking site, Facebook; 10% to perform Google searches for information; 9% to access their email; 9% to access political and sports news; 1% to access the Standard Chartered bank mobile banking website to pay utility bills to Umeme and National Water and Sewerage Corporation; and 7% access other websites, such as waptrick.com to download mobile ring tones. Facebook usage exhibited the highest percentage of mobile 413 Internet/Web users in this sample of respondents. This is because of the high interest in social networking among the 18-29 year olds in the sample: 20% out of the 23% that indicated they use the Internet to access Facebook were between ages 18-29 years. Those in their 30s, 40s, 50s, and 60s age ranges accessed the Internet primarily to perform Google searches and access political and sports news websites. Table 7-8: Mobile Internet Usage Trends of Survey Respondents Mobile Internet Usage Trends % Email 9 Facebook 23 Google 10 News/Sports 9 Mobile Banking (Standard Chartered Bank) 1 Other 7 7.2.2.9 Diffusion Analysis of ICT in Uganda A decade ago, most Ugandans frequented Internet cafés for their Internet/Web surfing needs (see Mwesige 2004). This trend has changed over the years with many owning their own laptops and desktop computers with Internet access. According to the data presented in sections 7.2.2.2 Hardware and Software) and 7.2.2.3 Access to the Internet/Web), over 52% of those sampled own at least a laptop or desktop computer, 48% have access to WLAN/WiFi, 19% have access through a wireless 3G modem, and 22% have access through an Internet connection in their offices. This indicates an increasing rate of adoption of ICT in Uganda. While in the late 90s there were only two telecom service providers, Celtel and MTN, today there are at least five, all of which provide wireless Internet connectivity (see 7.2.2.4 Internet Service Providers). The perceived attributes of ICT as an innovation, 414 such as observability, trialability and relative advantage, have increased tremendously, favoring sustained diffusion of ICT in the country. 59% of respondents indicated that they use the Internet/Web for work related information (see section 7.2.2.5 Reasons for Accessing the Internet/Web), which indicates a strong relative advantage of the Internet over previously used methods. 26% of respondents use the Internet for social networking, email and casual Web surfing, which reinforces the homophilous nature of the Internet/Web with regard to information exchange. Peer to peer homophilous communication leads to higher rates of diffusion mainly because interpersonal networks facilitate information exchange about an innovation. The cost of accessing the Internet/Web is still relatively high, as is evident in section 7.2.2.7 Affordability of Internet/Web Access) which could be a hindrance to ICT diffusion in Uganda. 7.2.3 Innovative Uses of Mobile Phones for Spatial Decision Making Mobile phone users in Uganda use their mobile phones to solve a number of problems in an innovative manner. This survey revealed several innovative applications of mobile phones for spatial decision making: voice-based spatial queries, mobile money transfer, SMS- or text-based Google search, pay-phone start-up businesses in rural areas, customer and supplier tracking in small business, and taking SMS or text-based orders for food delivery by local restaurants. In this section, these innovative uses are discussed under three categories: (1) voice-based spatial queries, (2) popular mobile phone-based applications, and (3) the use of SMS in small business. 415 7.2.3.1 Voice-based Spatial Queries 36% of the survey respondents indicated that they use their mobile phones to make voice calls to their friends, family, and business partners to inquire about their current locations, and possibly get directions. Only 3% of the respondents indicated that they make voice calls to establish the locations of certain mobile services of interest, such as an auto mechanic who travels to the location of his customers to fix a broken down vehicle (see Table 7-9). This shows that there is only marginal interest in location-based services at the moment, based on current location-related mobile phone usage trends exhibited in this data. The only interest exhibited is in the location of friends and family, which seems to reflect interest in some sort of a social networking LBS similar to Google Latitude (Google, 2012a). Table 7-9: Spatial Queries Performed by Survey Respondents using Mobile phones Spatial Query % Locate Friend/Get Directions 36 Locate Mechanic/other Service 3 7.2.3.2 Popular Mobile Phone-based Applications Respondents were asked to identify mobile phone based application that involved location of the user. They identified three categories of popular applications that in some way involved the location of the user: (1) Mobile money transfer, (2) the MTN Village Phone Service, and (3) Google SMS Search and Trader services. See Table 7-10. 416 Table 7-10: Popular Mobile Phone Based Applications Mobile Phone Service % MTN Village Phone (Grameen Foundation) 4 Mobile Money Transfer (MTN) 66 Mobile Money Transfer (Other Telecom) 1 Google SMS Search 32 Google Trader (Mobile Web) 6 Mobile Money Transfer 67% of them indicated that they used their mobile phones to transfer money to their relatives, friends and business associates. The transfer of money was mainly from urban towns and cities to villages in rural areas. MTN seems to have the largest share of the mobile money transfer market with 66% out 67% of the respondents indicating they use the MTN MobileMoney service. This service works in relatively simple fashion similar to the internationally focused MoneyGram and Western Union money transfer services. An MTN subscriber X at any location in Uganda approaches the nearest MTN MobileMoney agent and registers a MobileMoney account, providing valid Identification, such as a passport, driver’s license, company ID, tax certificate, a local council (local government) certificate, or a voter’s card. The MobileMoney agents operate out of custom-made semi-permanent kiosk offices (see Figure 7-14) located conveniently within trading centers in towns and villages, close to rural and urban populations. After registering an account, which is identified by the mobile phone number, subscriber X is able to authorize the transfer of a given amount of money to a relative Y, possibly located in another town or village. X does this by first loading credit onto his/her account at the MobileMoney kiosk, then accessing a special menu on his/her phone to send the money based on Y’s phone number. A small fee is charged on the 417 transaction, for example, according to one of my respondents, a fee of approximately $0.40 is incurred on a transfer of $40, which equates to a 1% charge. Upon the completion of the transaction, an SMS (text) message alert is received by both X and Y. For Y to receive the money, he/she has the option of going to any MTN MobileMoney kiosk in his/her town or village and presenting valid ID, a Personal Identification Number (PIN) to authenticate the withdrawal (if Y is a registered MobileMoney client), or a token number provided by X (if Y is an unregistered client). Y secretly enters his/her PIN or token number into the MobileMoney agent’s phone, and the transaction is authenticated and authorized. Cash is then issued to Y by the MobileMoney agent and the transaction is complete. (MTN, 2012c) Figure 7-14: MTN Mobile Money kiosk Source: (Hannes, 2011) 418 MTN Village Phone Service 4% of the survey respondents indicated that they used the MTN Village Phone service (see Figure 7-15). This is a rather low percentage of users that could be explained by the increase in personal mobile phone ownership in Uganda over the years. The MTN Village Phone is a mobile phone service for people in rural areas in Uganda that was introduced by Grameen Foundation (ICITD et al., 2010), a global non-profit organization based in Washington DC, in partnership with MTN Uganda in 2002 (GrameenFoundation, 2012). Figure 7-15: MTN Village Phone Source: (GrameenFoundation, 2012) It was modeled after a similar project implemented by the Grameen Bank in Bangladesh where small micro-loans were extended to women in rural areas of the country to alleviate poverty (Bayes et al., 1999; S. Corbett, 2008). The loans were used to purchase special mobile phones, so called Village Pay Phones (VPP), which were the basis for a 419 business model that involved delivering prepaid phone services to villagers for a small fee (Bayes et al., 1999; S. Corbett, 2008). The project led to a reduction in poverty in the villages in Bangladesh, and improved access to information and communication technology among the poor Google SMS Service 32% of the respondents indicated that they used the Google SMS service, which includes the Google SMS Tips, the General Google SMS Search service, and the Farmer’s Friend SMS service). 6% of the respondents indicated that they us the Google SMS Trader service. Users of the Google SMS services were from a broad spectrum of professional and educational backgrounds, and included students, teachers, lecturers, farmers, traders, social workers and many more. Half the respondents who accessed the Google SMS service lived in Kampala, while the other half lived in the rural areas of Mayuge, Kaberamaido, and Katakwi districts. Half the respondents who accessed the Google SMS services were either university or secondary school students, one-fifth of them were farmers and development workers in the rural towns (Mayuge, Kaberamaido, and Katakwi), and the remaining users of the SMS applications were a mix of professions, including surveyors, IT specialists, businessmen and administrators. This shows that there is interest in the Google SMS service among people of all socio-economic backgrounds in all towns in Uganda. 6% of the respondents indicated that they use the Google Trader SMS application. The Google SMS service was an SMS- (text message) based mobile application or service developed by AppLab Uganda, a technology-focused organization under the 420 Grameen Foundation, in collaboration with Google (the Search Engine provider), and MTN Uganda (the telecom carrier) (ICITD et al., 2010). This collaboration between Grameen Foundation/AppLab, Google, and MTN Uganda started in 2007 (AppLab, 2012a). The Google SMS application allowed a mobile phone user to send a query via SMS to the Google servers, and receive a response, also via SMS (AppLab, 2012a; Google, 2012b; Schaffner, 2009). Three types of applications were launched by AppLab in 2009 on the Ugandan market, (1) SMS Tips, (2) SMS Search, and (3) SMS Trader (AppLab, 2012a; ICITD et al., 2010; Schaffner, 2009). See Figure 7-16, Figure 7-17, Figure 7-18, and Figure 7-19. Figure 7-16: Google SMS Search Source: (Google, 2012b) 421 Google SMS Tips provides information on sexual and reproductive health issues; Google SMS Search allows a user to make a generic query about anything; while SMS Trader provides the user with relevant business and market related information, for example, the current price of farm produce in the local market, or the location of sellers and buyers for a given product (Google, 2012b). A related application to the SMS Tips service was the Clinic Finder application which allowed mobile users to locate their nearest health center (ICITD et al., 2010). A pilot program was implemented for an SMS application designed to provide information to farmers in rural areas about homemade pesticides and herbicides, the weather, and other location-specific agricultural information – this application is called the Farmer’s Friend application (Donner, 2009) and relies on weather information provided by the Department of Meteorology and local agricultural information from NGOs such as the Busoga Rural Open Source Development Initiative (BROSDI) (BROSDI, 2012a, 2012b; Schaffner, 2009). See Figure 7-20 for the architecture behind this SMS-based application. The Farmer’s Friend application morphed into the Community Knowledge Worker (CKW) application (GrameenFoundation, 2010; SMSInAction, 2011)which makes use of smart phones with Web browser capabilities, which is a different model from the SMS based approach of the Farmer’s Friend application (AppLab, 2012b). In addition, the CKW application relies on a network of “trained intermediaries” (GrameenFoundation, 2012), issued with smart phones, to work with farmers “who lack access to relevant information on sustainable farming practices, market conditions, pests and disease control, and the weather” (GrameenFoundation, 2010). The CKWs act as the go-between 422 farmers in remote villages, and agriculture-focused information sources such as NGOs, the Department of Meteorology, and public sector agriculture-support institutions. “The idea (with the CKW Initiative) basically is that if you have somebody on the ground, he/she is a member of the community who is trained on the use of a mobile phone to collect and disseminate information, then that person can bridge the digital divide. So you can use that person, for instance to provide information in the community, for example, if you wanted to do awareness, you can also use that same individual, because he/she is equipped with a smart phone. You can also use them (the CKWs) to conduct research. You can task them to gather basic information from the community using their smart phone. So, there are applications that we develop as part of the (CKW) project, and these run with the smart phone. We are currently building the network. We have 100 such people (CKWs – in Kapchorwa district), and over the 4 year period of this project, we hope to have maybe 4000 or so such community knowledge workers” (P. Ssengooba, AppLab Uganda, personal communication, 20 th July, 2010). The CKW Initiative was first piloted in Mbale, and Bushenyi districts, and by July 2010 was being implemented in Kapchorwa district. Ultimately, the objective of all Grameen Foundation/AppLab projects is to use ICT to fight poverty in rural areas around the world and improve the livelihood of the people. The CKW project was to be rolled out next to rural areas in Budaka, Bukedea, Palisa, and Gulu districts (P. Ssengooba, AppLab Uganda, personal communication, 20 th July, 2010). 423 Figure 7-17: Google SMS Poster 424 Figure 7-18: Google SMS Tips Figure 7-19: Google SMS – Farmer’s Friend 425 Figure 7-20: Google SMS – Farmer’s Friend Application – Architecture Source: (E. Karamagi, BROSDI, personal communication, 13 th July, 2010) 7.2.3.3 Use of SMS in Small Business Mobile phones have benefitted small business operators in increasing their profit margins. I provide an example of a unique SMS-based application that I witnessed being used by the owner of a small restaurant in Kampala. Open-are restaurants are a common sight in Kampala’s suburbs. Because of the make-shift nature of the restaurants, the cost of a meal is low, usually in the range of $1 - $2. These restaurants are usually found in food markets (similar to farmer’s markets in the US), and also next to other small businesses, such as auto mechanics workshops, and busy retail stores. Their menu usually consists of “local” food consisting of steamed green bananas (“matooke”), corn meal (“posho” or “kaunga”), sweet potatoes, Irish potatoes, millet meal (kalo), and cassava, 426 together with a meat or vegetable entrée such as beef, goat’s meat, chicken, peanut paste (“binyebwa”), cabbage, and leafy vegetables. The target market of these restaurants was blue-collar workers in the areas surrounding the restaurant – auto mechanics, small retailers, taxi conductors and drivers. Customers arrive at the restaurant, make an order, eat, pay their bill and leave. However, over the years, the clientele has grown to include corporate customers from large companies around the city, and not only the areas surrounding the restaurant’s location. Government vehicles, company cars and private luxury vehicles can often be found oddly parked outside these restaurants. This is an indication of the popularity of local food over western styled fast-food restaurants found in the city. Also, the lower prices are attractive. However, with the increase in traffic congestion in Kampala, customers have found it increasingly difficult to drive across the city to eat at these restaurants. Mobile phones have provided a solution to the traffic problem. Bulk orders for food are made by sending an SMS message to the restaurant operator in the suburb, and the food is delivered cheaply by motorbike taxis (“boda-boda”) directly to the customers in their corporate offices around the city. See Figure 7-21 for an illustration. The boda-boda bikes are ideal for cutting through heavy traffic and getting to their locations on time, thus the delivery of the “merchandise” is always on time, and at a very low cost to the restaurant operator. It should be noted that unlike in the US, there is no cost associated with receiving an SMS (text) message, so the only cost incurred, is by the sender of the message, which is approximately $0.10 per message. 427 Figure 7-21: Small Restaurant Business: Owner receives bulk orders for food from corporate customers on her mobile phone via SMS and delivers food packages to their offices by bike taxi, “boda-boda” The location of the customers is known by the restaurant operator a priori based on previous contact between the customers and the restaurant owner. However, there is potential for restaurant operators to grow their clientele if there was a more robust mobile phone based application, say a location-based service that could connect restaurant operators with new potential customers in various corporate offices around the city. 428 7.2.3.4 Diffusion Analysis of the Innovative Uses of Mobile Phones The most critical element in the innovative use of mobile phones in Uganda is re- invention – the customized use of an innovation to suit one’s needs. The small restaurant operator taking orders for food via SMS on her cell phone is a perfect example of re- invention, and confirms Rogers’ (2003) assertions that re-invention will lead to sustained adoption of an innovation. Mobile Money transfer, the Village Phone Service and Google SMS service are other concrete examples that confirm the effect of re-invention on the diffusion of mobile phone applications in Uganda. The perceived attributes of these innovative mobile phone based applications have also had a positive effect on diffusion of mobile applications in Uganda. For example, the relative advantage of Mobile Money transfer over wire transfers through banks cannot be ignored. Mobile phones are pervasive, while banks are not. The high compatibility, low complexity, high trialability and observability of all the innovative mobile phone applications (in section 7.2.3 Innovative Uses of Mobile Phones for Spatial Decision Makingreflect the hall marks of a classical diffusion process with respect to mobile applications in Uganda. Moreover, these applications inherently leverage interpersonal networks, which also tend to be homophilous. These factors have further contributed to the successful diffusion of these mobile phone applications in Uganda. Innovators and early adopters of the various mobile applications are concomitant with need and necessity. Farmers in rural areas tend to adopt the Google SMS Search application faster than people in urban areas because of the lack of ICT facilities in rural 429 areas, as opposed to urban areas. On the contrary, innovators of Mobile Money applications tend to be urban folk because of the need to periodically send money to financially assist relatives in rural areas. The Village Phone service was originally intended as a small business venture to alleviate poverty in rural areas, thus, innovators of this mobile application tend to be in villages. The role of opinion leaders and change agents has also been very important in the diffusion process. Change agents include employees of telecom companies in Uganda, mainly marketing and sales agents, and employees of local NGOs such as BROSDI. Opinion leaders include local government officials, from the lowest level (Local Council 1) to the district level (Local Council 5); these officials promote the use of innovative mobile applications, especially if they seem to have the potential to improve livelihoods in their towns and villages. The consequences of mobile phone applications have mainly been desirable and positive. For instance, the Village Phone application increased earnings of local villagers by allowing them to become small scale entrepreneurs; the Google SMS Search – Farmer’s Friend – SMS application has allowed local farmers to increase their crop yields and increase their incomes; and the Google SMS Trader application has allowed small business owners and farmers to find market for their goods and produce, and eliminate the middle man in business transactions. 7.2.4 Potential for Location-Based Services in Supporting Spatial Thinking Respondents were asked to check off possible LBS applications that they would be interested in using if made available. A summary of their responses is shown in Table 430 7-11.The highest level of interest (82%) was in an LBS application that informs users about available land parcels for sale around their current location. 75% of respondents were interested in LBS applications that answer query about whether a land parcel they were about to purchase is located in an area designated as a wetland by the National Environment Management Authority (NEMA). This is significant because NEMA forbids the construction of “unwise” use of wetlands, for example, for the construction of permanent structures. NEMA has the authority to demolish such structures without compensation, which can cause a serious financial loss to the land owner. 75% of respondents were interested in an LBS application that informs them about job availability in their areas, especially contract-based work. 74% were interested in an LBS application that alerts them about disease outbreaks in their localities, while 73% were interested in an application that could help them find a nearby health center. 69% were interested in an LBS application about weather forecasts, especially farmers. 68% were interested in receiving alerts about the environmental degradation in their localities. 68% wanted to use an LBS application to find out about current food prices in the country, especially local farmers who wanted to avoid being cheated by middlemen. 66% were interested in an LBS application that would allow them to fulfill their civic duties such as report the presence of a very bad pot hole on their roads that needed to be fixed. 66% wanted an LBS application that could help them report a crime to the nearest police post to improve response times of the police. 65% of the respondents were interested in an LBS application that would allow them to report suspected illegal encroachment of wetlands. The importance of wetlands has resonated with the local citizens because of 431 increased flooding in urban areas, which the government often attributes to wetland destruction and illegal encroachment (UgandaWetlands et al., 2009). Table 7-11: Survey Respondents’ Interest in Future LBS Applications for Various Spatial Queries LBS Application % Parcel Owner (Cadastre) 82 Parcel in Wetland? 75 Jobs 75 Disease Outbreak Alert 74 Health Center 73 Weather 69 Environmental Degradation Alerts 68 Food Prices 68 Report Potholes 66 Police Post 66 Report Wetland Encroachment 65 31% of the survey respondents indicated that they use the desktop versions of Google Earth and Google Maps which can be used as an indicator of the level of interest in geospatial information in Uganda. The importance of spatial thinking has only recently started receiving attention in the United States, for example, see the National Research Council’s publication on “Learning to Think Spatially” (NRC, 2006). Spatial thinking entails knowing about (1) space, (2) representation, and (3) reasoning (Manson et al., 2012; NRC, 2006) and has significance in various economic sectors, for instance, environmental conservation, citizen participation in government, and public health. LBS has the potential to instill spatial thinking in society because of the inherent use of the user’s current location in the mobile software application: “In order to empower people through Spatial Thinking, thus creating spatially literate people, LBS, like all other spatially-oriented problems, requires the three elements of Spatial Thinking: 432 ‘concepts of space, tools for representation, and processes of reasoning’. One who is ‘spatially literate’ is defined as ‘a person proficient in spatial thinking ...who can match the norms for what should be known about space, representation, and reasoning’… Spatial Thinking in LBS would lead to mastering many problem solving skills related to location/mobility and to the development of a general location/mobility perspective” (Karimi 2007, 88). Table 7-12: Survey Respondents that use Google Maps and Google Earth Google Maps/Google Earth Usage % Use Google Maps and Google Earth 31 At a mobile applications workshop in Kampala, Uganda (Brahma, 2010; OrangeTelecom, 2010; SMS-Media, 2010), telecom operators, content providers and application developers discusses the future of mobile phone application development in Uganda. This workshop had important pointers for the path LBS might take in Uganda. According to the presenter from Orange Telecom, “The mobile application is a link between the customer and content” (OrangeTelecom, 2010). The provider’s role is to allow customers to access mobile applications via services on the telecom’s network. These services can be delivered via voice, SMS, MMS (multimedia service), Unstructured Supplementary Services Data (USSD), Sim Application Toolkit (STK), or the Internet/Web, depending on how sophisticated a user’s mobile phone is. The relationship between the user- friendliness of these technologies and the degree of device dependency of each of these technologies is illustrated in Figure 7-22. In the 1990s, the 2G mobile phones were only capable of voice, SMS, and USSD communication, with no capabilities for displaying graphics on their screens. USSD is 433 similar to SMS, only that the former is an instant form of communication, while SMS is “store and forward” type of communication. Typically, the response to a USSD query is immediate, while, an SMS message can take anywhere from a few seconds to hours to be delivered. This is because SMS messages are first stored in the network servers, and then forwarded to their final destinations. This is due to congestion of data traffic on the network. If the forwarding of the text message does not succeed repeatedly, the message is simply discarded to save memory on the server. In the early 2000s, 2.5G phones with data capabilities emerged, and Wireless Access Protocol (WAP) browsers starting appearing on mobile phones. This was an exciting time in mobile application development because mobile applications could be delivered on mobile phones via the mobile WAP browser, and screen size, resolution and color were all improving for better graphical display. From the mid-2000s, 3G/3G+ mobile phones started emerging with capabilities approaching those of small computers; so called smart phones. The Limitation of WAP, which was basically a “compressed form of the Internet standard”, was no longer an issue, as the full power of the Internet could be leveraged on the 3G/3G+ smart phone. “This opens a new dimension,” according to Orange Telecom Uganda, in terms of mobile application development. A new channel of communication is suddenly possible with 3G/3G+ between the customers, ISP, Telecom Company, and application developers (service providers). High end smart phones allow a greater degree of user interactivity, unlike low end Global System for Mobile Communication (GSM) phones. Developers suddenly have the ability to build mobile applications with high levels of user 434 friendliness and interactivity. A telecom company’s role is to provide access to such applications using appropriate communication channels. Highly interactive mobile applications are targeted toward the high value customer. Access to such applications is made possible through 3G/3G+ mobile communication standards that allow the use of the Web, MMS, STK, and WAP. The downside of all this is the complexity in the development of such applications due to device dependency (see Figure 7-22). A developer considering developing an MMS-based application will be faced with the challenge that not all phones support MMS. An SMS-based application will probably allow for greater accessibility, however, SMS based applications are limited to text messages, and no graphics. High end smart phones would allow for applications with a rich user experience, however, there are few smart phones on the Ugandan market. (OrangeTelecom, 2010) Smart phones operate on various software platforms that need to be taken into consideration, for example, Android, Windows Phone, iOS, Symbian, BlackberryOS, and many others. In addition, User Interface (UI) are different, for example, a mobile application developer developing a mobile application for Nokia smart phones needs to consider the various UIs, such as, S30, S40. S50, and S60, and even possibly develop for all these UIs to maximize accessibility of the application. 435 Figure 7-22: Mobile Technologies: User Friendliness and Device Dependency Source: (OrangeTelecom, 2010) Application developers are the people that actually develop mobile applications to be accessed by users via SMS, MMS, USSD, STK, WAP or the Web. At the mobile workshop at the faculty of Computing and Informatics Technology, Dr. Ashish Brahma, Chairman of Mobile Monday Kampala chapter (MoMoKla) revealed that there were over forty eight mobile application development projects that were on-going in Uganda, all under the banner of mobile-Health (m-Health) (Brahma, 2010). A grassroots movement of volunteer mobile application developers, Mobile Monday Kampala, is behind the growing interest in mobile application development in Uganda. Mobile Monday (MoMo) started as a casual meeting of mobile developers in a bar in Finland, and by 2010, had spread globally with over 100 chapters in various countries around the world. 436 The major challenge facing the various m-Health projects in Uganda is the lack of coordination and communication among the developers (Brahma, 2010). Thus, MoMoKla provides a forum for developers and non-technical people with interest in mobile applications for development (M4D) to meet and have a conversation about their work. This avoids duplication of efforts and allows for sharing of ideas. A special interest group focusing on m-Health applications emerged out of these activities, m-Health Uganda. Three m-Health applications that were being developed in Uganda in 2010 were described at the Mobile workshop (Brahma, 2010). First is a mobile application that promotes maternal health and reduces the chance of death of mothers during birth. This project has been implemented in the villages surrounding the Bwindi Impenetrable National Forest and Park, which is also one of the only remaining Gorilla sanctuaries in the world. The GPS capability on smart phones was used by a small group of professionals who work with the local hospital in the area to map the locations of families living around the park. This data was mapped using GIS software, and used in the planning and coordinating of a maternal health program at nearby health centers. An SMS-based application was used to allow for communication between pregnant mothers and village health workers at the nearest health center at regular intervals throughout a mother’s pregnancy. Approximately two weeks to the projected time of birth, the mothers are relocated to the nearest health center so that the trained health worker performs a safe delivery. The mothers are monitored for a few days after giving birth, and then discharged. 437 Second is a mobile application designed by Swift River (SwiftRiver, 2012), modeled on the Ushahidi (Ushahidi, 2012) platform of crowd sourced information that was effectively used to monitor election violence in Kenya in 2008 (Okolloh, 2008), and to plan for humanitarian interventions after the Haiti earthquake in 2010 (Zook et al., 2010). This SMS-based application allows mobile users to text in health-related information, for example, “I think I have symptoms of the flu.” This information is crowd-sourced, verified, and plotted in a GIS so as to provide early warning of serious disease outbreaks in remote areas of the country. A third mobile application called TextToChange (TTC) carries out SMS-based surveys about a population’s level of knowledge about HIV/AIDS (TextToChange, 2012). To increase response rates, mobile users are entered into a draw to win small prizes, such as cash prizes, or a bicycle. Users are quizzed on health basics, especially on their knowledge about HIV/AIDS, for example, “Can HIV/AIDS spread by through a mosquito bite?” Responses from such a survey allow the Ministry of Health to gauge public knowledge about health issues, and plan interventions on health education around the country. There are over a hundred such mobile application development projects in Uganda in the areas of health and development. UNICEF Uganda’s UReport, a social monitoring tool for children that promotes community-led development and citizen engagement via SMS- based surveys built on a SMS-based platform called RapidSMS (RapidSMS, 2012; UNICEF, 2012a, 2012b). Social issues addressed by UReport include teacher absenteeism in primary and secondary schools, and water point functionality (S. 438 Blaschke, UNICEF Uganda, personal communication, 10 th July, 2010). UNICEF Uganda further uses the RapidSMS platform to develop mobile m-Health applications to “strengthen the health system in Uganda, particularly Integrated Community Case Management, tracking of supplies, improving rates of antenatal visits, transfer of medical results, and behavioral change messaging” (S. Blaschke, UNICEF Uganda, personal communication, 10 th July, 2010). Open source GIS and mapping software are used to analyze the results obtained. ChildCount+ is another RapidSMS-based m-Health platform developed by the Millennium Villages Project (MVP) of the Earth Institute, Columbia University (ChildCount, 2012; EarthInstitute, 2012). This SMS-based mobile application empowers communities to promote child survival and maternal health. The application facilitates and coordinates activities of field-based health care providers, known as Community Health Workers (CHWs). It allows CHWs to perform community health event reporting, send feedback and illness alerts about Tuberculosis, Malaria and non-communicable diseases. This information is collected by a central server that integrates with existing health management information systems to allow targeted interventions and timely action. A comprehensive list other m-Health projects in Uganda can be found on the Mobile Health Uganda website (M-HealthUganda, 2012). 7.2.5 Location in Mobile Applications and the Link to GIS Depending on the objective of the mobile application the locational accuracy and precision of a query can often determine the set of possible responses to the mobile user. 439 The exact origin of a query, in terms of its geographic coordinates or position, can be classified along a continuum: from low to high accuracy. The role played by the level of accuracy determines the type of mobile LBS applications that can be designed. High accuracy positioning can be obtained by leveraging the GPS capability of smart phones, which allows for up to 3-5 meter accuracy in the positioning of the mobile device. Such accuracy levels can be used for routing LBS applications to be built. Positioning obtained by referencing the cell ID of the closest telecom tower, or a triangulation of such towers provides an approximate low accuracy position of the mobile device. With such positioning, LBS applications with a wide area context can be built. Of the two types of mobile LBS applications, the ones currently being developed in Uganda are the latter. In the case of Uganda, it is evident from the sections above that by far the largest number of mobile applications being developed are in the health sector, followed by the social development and agricultural sectors. In all these sectors, location is inherently a part of the mobile application architecture. However, in many of the mobile applications discussed in the earlier sub-sections, location information would have to be explicitly provided by the user so as to receive a location-specific and context-aware response. For example, in the Google SMS Trader application, a mobile user who is looking for market for his agricultural produce would have to structure his or her SMS post to Google as follows: “sell sorghum Kampala, contact John, 0784 123456, got 100 bags.” A buyer would similarly post his or her query as follows: “buy sorghum Kampala, contact Mary, 0784 654321, need 10 bags.” The Google SMS Trader application uses the name of the district, town, city or locality, in this case Kampala, to connect buyers and sellers in a 440 way similar to Craigslist in the US. This is also the case with most of the m-Health and agricultural mobile applications described earlier, where the location of the mobile device has to be overtly stated in the SMS query. However, a few applications are leverage locational positions of higher accuracy levels, such as the GPS capability of the smart phone device, so as to capture 3-5 meter-level- accuracy-positions in the query sent over the network. Two examples of such mobile applications include (1) AppLab’s CKW application, and (2) Makerere University FCIT’s OpenXdata m-Health application. The CKW application was built for high end smart phones with mobile browser and GPS capabilities. The Community Knowledge Workers in Kapchorwa district, for example, were issued with HTC T-Mobile G1 Google phones running the Android Operating System (for detailed phone specifications, see GSM-Arena 2012). The GPS location of this device provided the high accuracy positioning needed in one of the projects executed by the CKWs, my survey respondent at AppLab explained. “In the test of concept for the CKW, we worked with National Agricultural Research Organization (NARO) to do banana disease monitoring, where we had community knowledge workers again equipped with some phones with GPS. So they would actually go and interview farmers, ask questions about banana disease, then they would go out to the farmers’ fields, and then working with the farmers, diagnose their various banana diseases. The CKWs had been trained on how to identify certain common banana diseases. So then they would map that information, and information would be collected, and then you could actually visually map the distribution of incidences. So over the 441 course of the pilot, we were able to develop a map where you could see the incidences of a given banana disease.” To be able to map the results of the banana disease research, a GIS was used to map the GPS locations sent by the CKWs over the GPRS (2.5G) network to AppLab’s central server. The CKW model moves away from the limitations posed by low-end mobile phones by using intermediaries as point sources of information for their communities. They also double as research assistants. The respondent from AppLab suggested that developers of mobile applications in the future may need to consider a hybrid solution that uses both SMS and the Web in the near future. Much as the largest number of phones in rural areas today is low-end GSM phones, the prices of smart phones seem to be reducing, and they will become more affordable and pervasive over time, eliminating the need for SMS-based solutions. Makerere University’s School of Computing and Informatics Technology, National Software Incubation Center (NSIC) was established in 2008 through funding from the Rockefeller Foundation to promote software development in Uganda (Kisambira, 2008; SCIT, 2012). The newly created institution embarked on a number of software development projects (SCIT, 2012). One of these projects was in collaboration with SNV Netherlands Development Organization; a project called CU@School (Ndiwalana, 2011). The main goal of this project is to monitor teacher and student attendance in primary and secondary schools. The project was initially piloted in Mbale and Kiboga districts, involving about 100 schools (Ndiwalana, 2011). The mobile application consists of mobile Web form on a smart phone into which the head teacher of a school can enter teacher and student attendance information and post it 442 using an Internet connection to central server located at the Ministry of Education and Sports. This data is stored in a MySQL database based on the IDs of the schools on a daily basis. The data is synchronized with servers located in individual districts on a weekly basis, where it is aggregated to produce reports and GIS thematic maps showing school attendance in each district. The actual coordinate locations of the schools were stored in a GIS based on previously acquired geospatial data from another source. The School IDs are the keys use to join the mobile form attribute data to the geospatial data of the school locations. GIS visualization is then carried out to establish attendance patterns in the schools. Based on the analysis, thematic maps and reports are generated. District Educational Officers (DEOs) can then take appropriate action based on the reports and thematic maps, for example, in the event that school attendance levels drop significantly in a given week – this could be due to a disease outbreak or some other cause that might need immediate action. The GIS software used to create thematic maps is GeoServer, an open source Web GIS server written in Java, compliant with the Open GIS Consortium (OGC) Web Feature Service (WFS) and Web Coverage Service (WCS) standards (Geoserver, 2012). See Figure 7-23 for the system architecture used. The mobile application is built on an open source software platform called openXdata (OpenXdata, 2012). The advantage with open source software is that it can be used for software development without any cost to purchase the software development platform. 443 Figure 7-23: FCIT’s OpenXdata School Attendance Project Architecture Source (P. Wakholi, Makerere University, Faculty of Computing and Informatics Technology (FCIT), National Software Incubation Center (NSIC), personal communication, 22 nd July 2010) According to a survey respondent at the NSIC, the cost of smart phones is not an issue that would affect future development of rich mobile internet applications in Uganda that leverage the Web and GPS capabilities of mobile devices. This is mainly because, one, the price of smart phones is projected to fall over the next few years, and two, institutions that need such mobile applications can afford to purchase smart phones for their employees as long as they are committed to solving a problem. “Right now, not many people have GPS enabled phones. As more capabilities come up and phones become cheaper, then people buy phones, not for GPS, they buy the phone for the Internet. 444 Originally, phones with Internet were very expensive, but now, you buy a phone, and Internet is one of things you buy and you did not intend to have Internet. You wanted something else. You wanted a nice camera, but the phone also has Internet. But also, one could say that people who are interested in such sophisticated mobile applications, like the police, would not be barred by that (the high prices) because they need it (the smart phones) for work. It depends on the capability. If you want to capture GPS locations, the phone should be able to have GPS capability, and a Nokia phone, I forget the series, costs about $150, and has that capability. Yes. And if you want to capture images, then you need to have a camera phone.” (P. Wakholi, NSIC, personal communication, 22nd July 2010). 7.2.6 The Role of Mobile Telecom Service Providers Mobile telecom service providers have been the cornerstone mobile application development in Uganda. The majority of the applications discussed in this paper have been SMS-based applications. The public’s access to these applications has been through the GSM network of five providers in Uganda, MTN, UTL, Zain (Airtel), Orange, and Warid telecom. The majority of the respondents to this survey were subscribed to the MTN network (91%), followed by Warid (30%), Zain (19%), UTL (15%), and Orange (14%), respectively in that order of popularity. The reason for the rather large market share enjoyed by MTN is that it was the first telecom company in Uganda to offer “affordable” cell phone subscription to local Ugandans, much as it was the second telecom company start operations in Uganda in October 1998. The first provider, Celtel, which had started operations in 1995, only 445 captured a small percentage of the market consisting of a small class of well-to-do Ugandans because its subscription charges were prohibitively high, upwards of $500, not to mention it had limited geographic network coverage, restricted to the major urban areas in Uganda. Celtel dropped its subscription and call rates, and expanded its network coverage due to competition from MTN, however, the damage had already been done – most Ugandans, the majority of whom are middle and lower class citizens, associated Celtel with the crème de la crème of society, and hence not the “people’s” network. In addition, MTN had an extensive network coverage including rural areas in Uganda. This was one of the conditions given to MTN and UTL when they were granted licenses to operate – to ensure that rural areas would also be covered by their networks (anonymous respondent, MTN, personal communication, 12 th July, 2010). Moreover, because cross network calls were prohibitively expensive, it made more economic sense to subscribe to the same network that the friends and family members of potential subscribers were on, which in most cases was MTN – hence, the rapid diffusion of MTN subscription among the Ugandan population. In 2010, MTN had about 6 million active subscribers on its network, which is about 19% of the total population of Uganda (anonymous respondent, MTN, personal communication, 12 th July, 2010). The government-owned Uganda Telecom (UTL) was the third telecom company to start operations in Uganda, almost at the same time as MTN, however, the poorer quality of service meant that MTN still won most of the subscribers over. The role of the Uganda Communications Commission in the early set up of Uganda’s telecom network cannot be understated. UCC offered an incentive to UTL and MTN to build infrastructure to cover 446 the entire country, especially the rural areas. In return, UCC would not allow any other telecom company into the Ugandan market for five years. The telecommunication infrastructure was needed to allow for better administration of the country, especially of rural areas that were geographically distant from the rest of the country. In fact, it was at this time that the idea of the Village Phone was borne – a wireless CDMA-based fixed line phone that would connect rural areas to urban areas (anonymous respondent, MTN, personal communication, 12 th July, 2010). Eventually, UCC did allow other telecom companies to enter the Ugandan market. Telecom companies such as Zain telecom (which bought Celtel, and was in turn bought by Bharti Airtel), Warid telecom, and Orange telecom opened up shop in Uganda, and started trying to compete against the dominant provider of voice and SMS services on the market, MTN, and to some extent UTL. So as to compete favorably, these companies started looking to target other niche market-dependent services, one of which was data services (mobile Internet services), and value added services such as mobile money, mobile banking, mobile advertising and so on. As of April 2010, there were two additional telecom companies operating in Uganda in addition to the five already mentioned. These include i-Telecom, and Smile Telecom (Mugabe, 2010). In total, there were 7 telecom companies operating in the country. 447 Table 7-13: Survey Respondents’ Subscription to Various Mobile Telecommunication Service Providers Telecom Company % MTN 91 UTL 15 Zain 19 Orange 14 Warid 30 Five in-depth personal interviews were carried out with respondents working for MTN, Orange Telecom, Zain Telecom, Nokia Siemens, and Ericsson. While MTN, Orange, and Zain are telecom network providers, Nokia Siemens and Ericsson are the suppliers of telecom network hardware equipment to the providers, allowing them to build their cell phone towers equipped with BTSs, and other network infrastructure. 7.2.6.1 Internet Infrastructure The telecom companies in Uganda are Internet Service Providers to various mobile users. The telecom companies’ Core Networks are connected to the global Internet infrastructure. See Figure 7-24 for an illustration of this infrastructure for Orange Telecom. A subterranean fiber optic cable from Saudi Arabia connects the entire East African region to the global Internet infrastructure. This cable connects to a fiber optic network in Kenya at Mombasa. This in turn connects to a fiber optic network in Uganda at the border town of Malaba, and then to Orange Telecom’s Core Network. The core network then connects to the Access Network that disseminates Internet services to mobile clients. 448 Fiber (Managed by TKL – Telecom Kenya)Fiber (Managed by UMEME – Via high-voltage power line) Kenya Uganda Mombasa U nd er -s ea F ib er c ab le - TE A M S Orange H/Q (Kampala) HSDPA/ HSUPA Modem PDA Wireless Microwave Microwave Microwave Malaba Access Network Figure 7-24: Internet Infrastructure of Orange Telecom Source: (anonymous, Orange Telecom, personal communication, August 2010) Internet Data Services Internet data services are available to mobile subscribers from MTN, Orange, UTL, Warid and Airtel (formerly Zain) Telecom. By 2010, MTN, Airtel, Zain, and UTL were all offering 2.5G GPRS/EDGE, and 3G data services to clients. The network infrastructure for 3G is different from 2.5G, and thus, new hardware had to be installed at all Base Transceiver Station (BTS) masts. 3G data speeds are much faster than 2.5G; 3.2 – 7.2 megabits per second, as compared to 128 kilobits per second. The decision to adopt 3G was driven by two factors: (1) the availability of 3G enabled phones on the Ugandan market, and (2) the projected return on investment for the telecom company ISP. According to a respondent at MTN, the decision to adopt 3G was a difficult one because there were only 200,000 – 300,000 3G enabled mobile phones 449 circulating on the Ugandan market amongst the MTN subscribers, which only represented 5% of their subscribers. This made it difficult to justify an investment in 3G at that time. 7.2.6.2 Telecom Network Positioning Techniques All telecom companies used the cell-ID positioning technique based on the cell in which a mobile user’s phone is located. This cell is identified by the strongest signal received by a mobile phone – the mobile device latches onto the BTS with the strongest signal. Cell- ID positioning is used to provide location-based discounts to mobile users. This technique is a very approximate method of geographical positioning. The denser the network of BTS towers, for example in urban areas, the smaller the cells, and thus, the more accurate the estimated position of the mobile phone. However, in rural areas, the network of BTS towers is quite sparse, and thus, the positioning is not as accurate. A network positioning technique that is also possible is the network triangulation technique. Here, the position of the mobile user is determined mathematically by triangulation based on three BTS towers that register the strongest signal at the mobile phone. The network server database stores the locations of these towers, and based on the radius of coverage of each station, a triangulated position can be calculated. Such positioning produces far more accurate results, however, the logic for such positioning would require the telecom company to install a location server specifically for this purpose. None of the telecom companies interviewed had any knowledge of such a location server within their network infrastructure. 450 The use of GPS positioning on the mobile devices would be a good alternative to an LBS developer, however, client based positioning would only support pull-based LBS applications, yet push-based applications make the most sense in the current mobile phone market. Pull-based LBS applications are those that allow a mobile client to initiate a query based on the GPS location of the mobile device, hence “pulling” data from the telecom network. Such a query might be, “What is the nearest gas station to my location?” A Push-based LBS application, on the other hand, is initiated by the network based on the current location of the mobile user. This position is usually determined by the network- determined cell-ID of the user. An example of such a service is location-based tariff discounts, and mobile advertising. Push-based services do not require a smart phone, and in fact, such services often apply to voice calls, or are SMS-based. Because of the shortage of smart phones on the network, it would make more economic sense for a third-party LBS developer to develop an SMS-based LBS application, in which case, the only available positioning technique would be that provided by the telecom network. This means that GPS capabilities on phones currently are not a determining factor for the diffusion of LBS applications because there are too few smart phones on the market. “It’s just all about how advanced the LBS is, because you know, LBS can be so many things, but it just depends on what idea you have and whether you are going to run it over GPRS, or SMS, or USSD, but in the market like ours, if you are going to start by doing something complex that requires you to have access to 3G, you are going to have to 451 target only 200,000-300,000 subscribers, instead of the 6 million that have access to SMS. But if you want something that is mass market, for example a trade in Kikuubo market, then you are going to have to go with SMS, or USSD” (anonymous respondent, MTN, personal communication, 12 th July 2010). 7.2.6.3 Value Added Services Based on Location Telecom companies leverage the location capabilities of their networks to launch services that add value to existing voice and SMS services that area already available to customers. These so called value added services (VAS) include locational tariff discounts, mobile banking, mobile money, and mobile advertising. Of these, locational tariff discount services are the only ones that are based on the current location of the mobile user. Users located in geographical areas (cells) that are characterized by low bandwidth usage, where the network is less congested, are offered lower charges per minute for voice calls and SMS. These services are marketed under various names, such as MTN Zone, and Zain Dynamic Discount Service. Mobile money applications use location at course scale in specifying the town or district in which the money is to be received. Cell-ID positioning makes sense in this case as there is no requirement for accurate positioning in providing such a service. This research could not establish actual examples of mobile advertising LBSs based on location, although interview respondents at MTN, Zain, and Orange made mention of the concepts involved in actualizing such a service. Nevertheless, the constraint with such an application would be the course resolution afforded by the cell-ID network positioning technique. The more accurate GPS positioning technique would not make sense in the 452 development of mobile advertising LBSs because the availability of GPS is device dependent, and there are too few GPS enabled smart phones on the market; such an LBS would not make business sense. Using VAS, Telecom companies hope to drive data usage up, especially among the rural and less educated populations. “It is one of those things where you have to say data is just data, but what can we put on top of that for some guy downtown Kampala to actually use it, because a simple trader is not just going to get on the Internet to just surf for no good reason! VAS involves building the products around data, and then causing the data usage to bump up, and then users go out to buy smart phones, and then with time, the data usage picks up” (Anonymous, MTN, personal communication, 12th July, 2010). 7.2.6.4 Third-party LBS Developers and the Revenue Share Model Third-party LBS developers seeking to leverage the telecom networks’ SMS servers to deliver location-relevant information to mobile users via SMS would need to plug in their servers into the network, usually via an Internet connection. To be able to do that, they would need to negotiate a revenue sharing deal with the network provider. However, the network provider would only agree to a deal if the application makes business sense; in other words, the application, or business idea, must have the potential to increase network traffic significantly to earn sizeable profits for both the provider and the developer. Revenue Share is the predominant business model used. Here, the provider and application developer, who is usually also the content provider, each take 50% of the profits at the end of a fiscal period. An LBS in this case qualifies as one of many possible Value Added Services (VAS). 453 Examples of current VAS in Uganda include bulk SMS services provided through third party content providers such as SMS-Media, that provide information to users such as sports and political news updates. Mobile users have to subscribe to receive these text messages, and they usually cost two or three times more than a regular SMS message. It is from the inflated cost of such value added services that content providers and telecom companies make a profit, and hence the revenue share model. Many of the VAS currently do not leverage location in their business logic, and thus do not qualify as LBS. They are simply subscription based bulk SMS push services characterized by the sending of periodic text messages to client mobile devices based on temporal events. However, there is quite a lot of potential for location-enabled VAS according to a respondent at MTN, “There is a lot of competition among content providers, and so you must have a winning idea to convince MTN to allow you to plug into their server network (via IP). The question is, can you help us retain subscribers on our network, and how much money can we make from the LBS application?” (Anonymous, MTN, personal communication, 12 th July, 2010). 7.2.6.5 LBS Mobile Applications in Uganda There were a few LBS applications in that were identified in this research in Uganda. Here, I discuss an application developed by Zain and Ericsson that helps lost fishermen on Lake Victoria to make a rescue call to the local coast guard (BalancingAct-Africa, 2010; Nakaweesi, 2009; Zain, Ericsson, & GSMA, 2010), and vehicle tracking applications being utilized by three commercial companies: 3D Tracking, Sentrack 454 Systems Limited (SSL) and Fleet Monitoring Solutions (FMS) (Katende, 2012; OrangeTelecom, 2012; Sentrack, 2012). Zain Telecom, in collaboration with Ericsson Uganda, GSMA Development Fund, and the National Lake Rescue Institute (NLRI) developed a mobile location server application to help fishermen lost on Lake Victoria contact the local coast guard on shore. It is estimated that there are over 5000 deaths a year on Lake Victoria resulting from capsized boats (Zain et al., 2010). By 2009, Zain Telecom had constructed 21 new mobile radio towers on the shores of Lake Victoria to increase GSM network coverage on the lake allowing voice and data communication; data transport is through EDGE 2.5G technology (Zain et al., 2010). In addition, Ericsson developed a Mobile Positioning System (MPS) so as to locate a rescue call from the lake, and reroute it to the Coast Guard on the shores. A fisherman “lost at sea” can make a call to 110 to the Rescue Center on the shores of Lake Victoria. The Center establishes the position of the fisherman, and alerts the nearest mobile rescue unit (coast guard) of his or her position (see Figure 7-25 for an illustration of the communication network). Position computation was done through two methods: (1) the mobile network based positioning method, Cell Global Identity Timing-Advance (CGI- TA), and (2) Assisted-GPS (A-GPS) positioning (Zain et al., 2010). The CGI-TA solution uses the known locations of cell phone towers, in combination with the time difference between the start and arrival time of the radio signals at BTS towers in the vicinity of the mobile device. The accuracy of this Positioning technique is between 20-50m on Lake Victoria, according to an interview respondent at Zain 455 Telecom. Accuracy with the CGI-TA positioning generally tends to be higher where there is a dense network of cell towers, for example in urban areas. General accuracy with this method is in the 10-500m range (Andersson, 2012). The second positioning technique, A-GPS, is based on the Global Positioning System, assisted with data from the cell network to allow for quicker Time-To-First-Fix (TTFF). The downside of this method is that it requires a GPS capable phone, which is unlikely among the fishermen living on the shores of Lake Victoria. Much as this innovation was created and the proof of concept demonstrated to the local authorities, the Government of Uganda “could not afford” to purchase the technology for the Police Coast Guard, and thus, this LBS mobile positioning solution is not currently being utilized (anonymous, Zain Telecom, personal communication, 23 July 2010). 456 Figure 7-25: Mobile Communication in the Lake Victoria Rescue Project’s Mobile Positioning System Source: (Zain et al., 2010) Fleet Management is the monitoring of a fleet of vehicles while in the field. One of the services offered by fleet management is vehicle tracking. Government agencies, NGOs, and private companies that own a fleet of cars benefit from vehicle tracking to ensure that official vehicles are not abused by their drivers to run personal errands. In Uganda, fleet management services are provided by companies such as Sentrack Systems Limited, 3D Tracking, and Fleet Monitoring Solutions (Katende, 2012). These systems use GPS to determine the location of the vehicles. The locations are sent over a telecom company’s 457 GSM/GPRS network to a central server. Customers are provided with a Web based interface to monitor their vehicles in real time on their PCs in their offices. Another company that provides fleet management services is Orange Telecom. Its system is called Fleet Live (OrangeTelecom, 2012). This system has been enhanced to furnish customers with live SMS feeds on mobile devices (OrangeTelecom, 2012). In addition, a mobile client installed on a customer’s device allows for the display of detailed text and graphical information by leveraging mobile mapping applications, such as Google Maps (OrangeTelecom, 2012). Apart from fleet management, other applications of the Fleet Live system also include recovery of stolen vehicles, and even tracking of children. 7.2.7 The Role of Content Providers Content providers are a critical component of the mobile application development and infrastructure. In Uganda, content providers include SMS Media, Blue Cube, and True African (MTN, 2012d). They provide mobile content as Value Added Services to mobile clients over USSD/SMS, for example, political and sports news, ring tones, caller ring back tones (cRBT) logos, national exam results, and bulk broadcast SMS, for example, invitations for a wedding meeting (BlueCube, 2012; SMS-Media, 2010, 2012; TrueAfrican, 2012). Information delivered to clients is usually in a bulk SMS push model, as opposed to a pull model. In the push model, a user simply subscribes via USSD to receive a certain number of bulk SMS messages within a given period. However, none of these VAS applications are location-enabled. The Google SMS Tips Farmer’s Friend application created by AppLab relied on several content providers. “For the Google SMS Tips, we work with content providers, for 458 instance for Farmer’s Friend, we worked with BROSDI, who provided the local agricultural tips, and they obtained those ones through their own network – farmer’s network. As part of Farmer’s Friend, there is weather information as well, so that is from the Department of Meteorology, and then for health information, that comes from Straight Talk Foundation. So we are not content experts, we look out for people, organizations and have good content that is provided by experts, and this is the content that we make available to our targeted people” (P. Ssengooba, AppLab Uganda, personal communication, 20 th July, 2010). 7.2.7.1 Case Study: Use of Google SMS by BROSDI Two in-depth interviews were carried out with members of a local NGO based in Kampala called the Busoga Rural Open Source and Development Initiative (BROSDI, 2012a). This NGO employs mobile and ICT technologies to improve the lives of rural populations in multiple districts in Uganda, concentrating mainly on children and farmers. Further, a focus group discussion was carried out with three coordinators of BROSDI in a small rural district in eastern Uganda, Mayuge district, to better understand the use of ICT and Mobile application technologies by youth and farmers in rural communities (BROSDI, 2012b). BROSDI’s vision is to facilitate the sharing of locally generated information and knowledge among the members of a social system so as to leverage that knowledge to improve livelihoods. In the Google SMS Tips – Farmer’s Friend – mobile application developed by Grameen Foundation’s AppLab, BROSDI served as content providers for local agricultural information. They made a unique contribution in terms of gathering and 459 providing to Google local knowledge on effective farming methods and practices used traditionally by rural African farmers. BROSDI collected local knowledge from farmers in various parts of Uganda about organic methods of disease, pest, and weed control that were passed down to tribal societies by their ancestors. These methods were tested on several crops on small plots of land, scientifically verified and documented, ready to be shared via ICT tools. “If you know something, that is knowledge to you, when you start to share it, it is information … All our programs have got to have the element of sharing information, information management, and the concept of ICT. And it has to have the concept of open source and development.” (E. Karamagi, BROSDI, personal communication, 13 th July 2010). Local knowledge is collected by BROSDI representatives from farmers in grass roots community meetings with elders in the village, and local government officials – this BROSDI project is called Collecting and Exchange of Local Agricultural Content (CELAC). The local knowledge obtained in such meetings is tested at BROSDI demonstration plots in the district, certified, cross checked with the National Agricultural Research Organization (a government agency), approved, formatted, packaged and uploaded as content to Google servers ready to be served back to the broader rural community via SMS messaging. Such content includes organic methods of producing fertilizer, pesticides, and herbicides. (E. Karamagi, BROSDI, personal communication, 13 th July 2010). An SMS query from a local farmer’s mobile phone, for example in Mayuge district, is made through the Google SMS Tips Farmer’s Friend application, which is simply 460 accessed by sending the SMS query to a special phone number, 6001 (see Figure 7-17). Such a query string might simply be, “how do I make fertilizer for my onions?” The response from the Google Search Engine via SMS back to the famer might be: “Collect human urine, ferment it for two weeks, mix it till it becomes a white color, mix 1 part of Urine with 9 parts of water, and spray around the crop at night”. Such information allows farmers to increase yields of their agricultural produce at almost zero cost since the remedies are based on local organic raw materials and ingredients. Such simple interventions have gone a long way in empowering rural communities at the grass roots level. “Our principles are based on the usage of the environment to address an existing agricultural problem. You don’t go and buy the chemicals from the shops. If you want to make manure, say you are a woman, and you want to make manure, because the other manure, the pit manure is very hard if you are a woman, because of the turning (mixing) – you need a man’s strength. So instead, you get cow dung, put it in a sack, mix it with grass, tie the sack at the top, hang it on a tree and make sure that it is suspended in a drum of water. Then after 1 week, you come and stir the water and put it in your backpack, and go and spray your crops. That is content from our great grandparents!” (E. Karamagi, BROSDI, personal communication, 13 th July 2010) Prior to BROSDI’s collaboration with Grameen Foundation’ AppLab, Google, and MTN, BROSDI was still able to share local agricultural content from the CELAC project meetings with farmers in the 17 districts that the project was operating in. The communication method used by BROSDI was bulk SMS. Every week, BROSDI would 461 send out bulk SMS messages to over 500 farmers in the 17 districts about crops – diseases, pesticides, herbicides, and fertilizers. The technology used to send such bulk SMS messages was free-ware developed by Ken Banks, FrontlineSMS (Banks, 2009; Banks & Burge, 2004; FrontlineSMS, 2012). This technology had been applied successfully in several developing countries as part of m-Health applications in Malawi, South Africa, Pakistan, Afghanistan, and the Philippines (Banks, 2009; FrontlineSMS, 2012). A critical problem occurred – some the farmers complained that the content was not location-relevant, for example, a farmer in Apac district receiving an SMS message about how to plant and nurture Irish potatoes. The climate and soils in Apac district in northern Uganda is not conducive to growing Irish potatoes. Such a message would be more relevant to a farmer in Kabale district, south western Uganda, which is at a high elevation, has a cooler climate and suitable soils. Thus, location provides context to the information, and to a query. It is in this light that a partnership with a telecom company and AppLab made sense so as to be able to provide location-specific information to farmers depending on the geographical origin of an SMS query from a farmer. BROSDI has a center in Mayuge district, Wainah village (see Figure 7-27); the population of this village is about 600 people (M. Walwaana, HCC, personal communication, 12 th August 2010). A focus group discussion was carried out with three administrators of BROSDI’s projects in the village, at the grass roots level (see Figure 7-26). Three projects were being implemented, Collecting and Exchange of Local Agricultural Content (CELAC – agricultural program), Hope Children’s Club (HCC – 462 educational program), and the Youth and HIV/AIDS Awareness Project (YoHAAP – health program). Figure 7-26: BROSDI staff in Mayuge district that took part in the Focus group discussion All three projects are implemented within BROSDI’s broad philosophy of leveraging ICT for development through sharing of local content in rural areas of Uganda. There are 30 computers at the center, 20 of which are connected to the Internet via a satellite connection (see Figure 7-27). The Internet allows local farmers and youth to access information on agriculture, sexual health, and social issues through BROSDI’s website (BROSDI, 2012a). There are about 30 computers at this center, 20 of which are connected to the Internet via a satellite broadband connection through an ISP called AfSat, under Telkom South Africa. Some of the farmers were provided with their own laptops with a mobile internet connection (see Figure 7-27 and Figure 7-28). Funding for 463 BROSDI’s projects in Uganda comes from HIVOS (The Humanist Institute for Development Cooperation), a Dutch International Organization. Figure 7-27: BROSDI center in rural Mayuge district, Wainah Village: satellite broadband Internet connection through AfSat Figure 7-28: Internet room at BROSDI center in Mayuge district, Wainah Village, provides Internet access to local farmers and youth 464 Because many of the farmers have not had a formal education, they cannot read and write in English, thus, some of the information is translated into their local languages. One of the earlier interview respondents does not regard the farmers are illiterate: “They are Illiterate according to English. You see, this westernization calls somebody who doesn’t speak and write English illiterate. But I say an illiterate person is a person who cannot speak and write their local language, because I think our local language is more important than the English language. So the farmers, it’s very interesting, they can write their local language, and they can read their local language, but they can’t read and write English. But there are some who cannot even read and write their local language. They don’t even know how to write their names. Those ones are still inclusive in the project (CELAC); we have no problem with them. But then what would happen, is they of course, they did not know how to type, so we gave them skills. Every year, we have an annual training and among the modules in the training are ICT skills, so they were able to learn how to type and now they have the computers. I think the majority of them have laptops, so they know how to use this (touch pad) mouse” (E. Karamagi, BROSDI, personal communication, 13 th July 2010). One of the focus group members trains local farmers in the village in effective organic farming methods. “My job is to train farmers in some small (inaudible) of agriculture, and mainly, we grow organic. We have a project called CELAC with some group members. We actually come every month for farm forums. We discuss a number of things concerning agriculture, soil management, organic soil and disease control. And I 465 make articles for other people to read concerning agriculture” (J. Muloopi, BROSDI, personal communication, 12 th August 2010). BROSDI’s Mayuge district center collects local knowledge from local farmers about traditional organic methods of pest and weed control, making fertilizer, and other agricultural information that would be useful for increasing crop yields of farmers all over the country. An intriguing narrative given by one of the farmers was on biological control of the caterpillar, a pest that can devour a whole field of sweet potatoes in a few days. The farmer harvested a few of the live caterpillars from her field of sweet potatoes and placed them into empty disposable mineral water bottles. She sealed the bottles to cut of the supply of oxygen to the caterpillars, which led to them dying after a few days. She then scattered the dead caterpillars in her pest infested field. In a few days, she noticed that her field was pest free – the caterpillars were gone. Her crop yields increased substantially much to the dismay of her neighbors. They were wondering why their gardens were not flourishing like hers, and so they started suspecting her of witchcraft. In fact, she was about to be lynched by the local villagers based on an accusation practicing heresy. BROSDI intervened in this escalating social matter just in time to save the local farmer’s life, and educated the local villagers on why the woman’s field was doing better than all the other fields: similar to humans, live caterpillars cannot be around dead decomposing members of their own species. “We explained to them. By the end of the day, they are using the method now as pest control, but they don’t know, and neither do I, why the caterpillars do not thrive in areas where their friends’ bodies have been dumped” (E. Karamagi, BROSDI, personal communication, 13 th July 2010). 466 The BROSDI office in Mayuge district goes a step further to test the local organic farming methods they discover through their group meetings with farmers. They have several small demonstration plots, 50 by 300 feet in size, where the various organic methods of farming are tested and demonstrated to rural farmers who attend their meetings. BROSDI’s activities with respect to local agricultural content and ICT for development made them a good fit for the Google SMS Tips mobile application, “Farmer’s Friend,” implemented by Grameen Foundation’s AppLab, and made accessible through MTN’s mobile network. BROSDI was tasked with providing local agricultural content to Google, with technical support from AppLab, and domain knowledge support from the National Agricultural Research Organization (NARO) (see Figure 7-20 for architecture). It is one thing to create a mobile application, and another to have this application actually put to use by the intended target group – rural farmers. The respondents confirmed that the Farmer’s Friend mobile application was indeed being used by the local farmers, and making an impact on their knowledge base on organic farming methods. The other Google SMS applications such as Google SMS Health Tips, Clinic Finder and Google SMS Trader were also being actively used by local villagers. “Farmer’s Friend ranks number 1, then health tips, and then Clinic Finder … at the moment, more people have phones. For those who don’t have phones, they come over to the center and access that information.” (M. Walwana, BROSDI, personal communication, 12th August, 2010). The focus group members provided a demonstration of how a farmer would use the Farmer’s Friend application. “So you go to messages, and send to 6001 (mobile phone 467 number). An example query could be ‘Newcastle disease poultry.’ I am going to send another query: I type into a message composition box: ‘control aphids cabbages (sic).’ The reply (response) time depends on the network. If the result is very good (query well- formed), you get a reply back instantly” (J. Muloopi, BROSDI, personal communication, 12 th August 2010). This message was sent using the respondent’s T-Mobile Google G1 smart phone (see Figure 7-29) provided by Grameen Foundation’s AppLab to the project facilitators. The message could simply have been sent by any low-end GSM phone, for example, the ones commonly used in rural areas of Uganda. Figure 7-29: T-Mobile G1 Google phone used by BROSDI staff in Mayuge district to access the Web, and the Google SMS mobile applications Within 5-10 seconds a response was received an SMS text message: “Aphids on cabbage/kale, Brassiccas. Several species of aphids attack brassicas in East Africa: The cabbage aphid, the falls cabbage aphid, the green peach aphid. In particular the cabbage aphid is a major pest in the region. Cool dry weather is most favorable for aphid 468 development. Large numbers of aphids may kill small plants. Their feeding can distort leaves of older plants and cause leaf curl. Insecticidal soaps provide control and spot sprays of pyrethrum or nyme can prevent build-up of large populations. Reply 1 for related tips… (Message truncated). Information by: www.infonet-biovision.org. Tip: get weather forecast, Send ‘help’ to learn more.” Because a smart phone was used to send and receive the message, the response was easier to read as one single aggregated text message, however, if a low-end mobile phone was used, this message would have been broken up into multiple text messages, each consisting of 160 characters. For farmers who do not own their own mobile phone, the BROSDI center provides a one-stop to cater for their information needs. A total of 8 people work as project facilitators at this center, all equipped with Google G1 smart phones, accessible to the local farmers in the village. Of course, there is a cost implication for sending SMS messages from one’s phone. In rural areas, the cost of multiple SMS messages could be an economic deterrent to the success of the utilization of the mobile application because the majority of the population is poor. In Uganda, there is no cost to receive SMS text messages on the prepaid service, unlike in the United States. However, there is a charge for sending an (outbound) SMS message; about $0.05 on most mobile networks, including MTN’s. To address this dilemma, and to encourage usage of the service, MTN developed a special SIM card for use in rural areas that incurs substantially lower voice and SMS tariffs as compared to the regular SIM cards used in cities. A special prefix is attached to these SIM card numbers (0392). These are, however, only made available to facilitators of developmental projects, such as BROSDI staff, in various villages, and not to the local 469 population in rural areas. This, at least, reduces the cost of SMS messaging by the facilitators as they help local farmers who come to the BROSDI center in search of information. This same concept of lower tariff SIM cards was also used in the MTN- Grameen Foundation’s Village Pay Phone project described earlier. The focus group members explained the impact of the local agricultural content disseminated to rural farmers in the village, and the importance of organic farming to improve livelihoods of the poor: “The reason why we do this (CELAC project) is that chemicals are expensive, but then our farmers are poor. Because if we wanted to buy chemicals to work on aphids, you would need something like 20,000/= (about $8). But this is farmer whose child is being chased from school because of school fees of 40,000/= (about $16), yet he has to buy chemicals worth 20,000/= (about $8), which is not practical. So what we say, let’s use the local environment, what we have, locally and it works. Save the 20,000/= for school fees, use what you have locally and enable your crops to grow” (M. Walwana, BROSDI, personal communication, 12th August, 2010). Google Trader has had a positive impact on increasing farmers’ revenues from the sale of their agricultural produce, explained one the respondents: “They came, it was a group (of local farmers); they had all this cabbage. In the local market they were being given very little money. We called them and told them you can come and use the Internet, ‘you surf and get market, you can find vendors.’ So, they came, they surfed and they got one vendor from Kampala who was giving them good money. So, they made a deal, they came and they sold to him. He paid them. Even Google trader is now helping because our farmers wanted … they are now doing group selling like they come together, so if he has 470 cabbage, I have cabbage, we come together and market together, so they use Google trader, send, find the buyer, and then they make a deal” (M. Walwana, BROSDI, personal communication, 12 th August, 2010). The focus group identified various challenges facing local farmers in the village with regard to ICT usage. One respondent explained that English was a communication barrier with respect to information on the Web: “People are still ignorant with ICT. More so, we have farmers who are not learnt, but when they come to the computers, the computers are just using English, which they don’t understand. There is a language barrier. And you find that some people have started buying these complicated phones, and they would like to know how to access the Internet on the phones, but they are still ignorant about it. So maybe we have a task of training people how to use computers and so on” (M. Malime, BROSDI, personal communication, 12 th August 2012). 7.2.8 Challenges to Location-based Services in Uganda There were various technological, socio-economic, political and institutional challenges identified in this research. A discussion of these challenges is presented in this section. 7.2.8.1 ICT Infrastructure Access to the Internet/Web is a necessary ingredient for location-based services. ICT infrastructure in Uganda is existent in many urban areas, but the rural areas need better coverage. Internet connectivity in Uganda today is mainly through mobile wireless access based on GPRS/EDGE (2.5G), 3G and 3G+ technology. 4G technology has only recently started diffusing into the country, for example, through Foris Telecom. The country has seven telecom networks that also act as wireless ISPs, for example, MTN, Airtel, Warid 471 Telecom, and Uganda Telecom. GSM-based Telecom infrastructure built by these companies covers about 80% of Uganda for voice and SMS services. It covers all urban areas and a large percentage of rural areas. Geographic coverage for data services is not as vast as for voice and SMS services, however, most urban centers are covered. In addition, there are ISPs that are not telecom companies, but pure data service providers, for example, iWayAfrica, offering various types of Internet connection, such as satellite wireless broadband, and WiMax. 7.2.8.2 Telecom Network Positioning Technology Network positioning technology in Uganda is predominantly based on cell-ID. This is a very approximate positioning technology that is suitable for low accuracy and low precision applications. It has been leveraged by network providers like MTN to offer value added services such as location-based discounts on Voice and SMS services. However, for medium and high accuracy LBS applications such as search and rescue, cell-ID is not adequate as a positioning technique. Zain telecom (today known as Airtel) developed a network based positioning technique that leverages both cell-ID and time advance features of the signal received at a BTS station; this method is called CGI-TA. Its accuracy for the Lake Victoria Search and Rescue project was between 20-50 meters. The government of Uganda, unfortunately, was not able to purchase this technology from the innovators, Zain Telecom and Ericsson Uganda. Nevertheless, this proof of concept demonstrates that the network infrastructure for more accurate network based positioning is in existence in Uganda, and could support robust LBS applications in the future. The 472 major challenge is to find a business model that would work for both the innovators of the technology, LBS developers, and the mobile clients. 7.2.8.3 Lack of Digital Street Data Digital data for the street network in both urban and rural areas is lacking in Uganda. Many streets do not have street signs indicating the names of the streets, and in fact, there is no conventional street address system in existence. An informal system based on land parcel numbers (plots) is in existence in the core urban and upper class areas of major cities and towns, but for the most part, there is nothing. This impacts LBS negatively as this eliminates the possibility of creating mobile applications that have a routing component in them. 7.2.8.4 Content Providers There are a number of content providers for mobile applications in existence in Uganda, for example, SMS Media, BlueCube, and True African. Much of the content is primarily provided via SMS, and not through Mobile applications. In fact, there are no LBS applications that provide content to subscribers based on location, unless the location is overtly specified in the SMS query string, for example, “Weather Kampala.” 7.2.8.5 Mobile Applications Developers Mobile application developers in Uganda are found in the International Organization, NGO, academic, and private sectors. UNICEF has developed applications on the RapidSMS Platform, BROSDI has leveraged the FrontlineSMS platform, and Makerere University’s NSIC has developed applications on the OpenXdata platform. These applications are mainly in the m-Health, and social development realm. There is a 473 popular movement to develop innovative mobile applications that has somehow diffused into Uganda called Mobile Monday Kampala (MoMoKla) amongst software hobbyists and volunteers, especially in the m-Health sector. The change agents in this diffusion have been the expatriate community in Uganda, mainly from Europe. Most of the mobile applications found in Uganda are not necessarily consumer based applications driven by economic forces and funding from the private sector; rather, they tend to be driven by international donor funding. Makerere University’s National Software Incubation Center is one of the institutions promoting the development of commercially viable mobile software development among former computer science and IT students of the School of Computing and Information Technology. Value Added Services from content providers such as SMS Media, and BlueCube are some of the few mobile applications that have enjoyed commercial success in the Ugandan market. There is a need for better integration between telecom companies and research institutions like Makerere University so as to promote commercial mobile application development from innovators in the academic sector. Most of the mobile applications developed are SMS based server-side applications, and not client-side applications running on mobile devices. Some sophisticated client-side applications have indeed been developed, but these are few due the lack of smart phones on the market. Such applications include the NSIC’s OpenXdata m-Health application CU@Work, AppLab’s Community Knowledge Worker applications. The most sophisticated LBS applications developed are for fleet management and vehicle tracking from private sector companies such as Sentrack Systems. 474 7.2.8.6 Smart Phone Penetration Low The level of penetration of smart phones is still quite low to support commercially viable LBS applications. Smart phones with GPS and Internet/Browser capabilities are a key ingredient for LBS. However, with ever decreasing prices of smart phones, it is projected that the number of smart phones will steadily increase in the market in the not too distant future, which could allow for an increase in the number of location-enabled mobile applications developed for the Ugandan market. 7.2.8.7 Cost of Telecom Services High The dominant model of service delivery in the telecom industry in Uganda is the prepaid “scratch card” model. This includes voice, SMS, and data services. This means that access to mobile services tends to be quite expensive as compared to the post-paid “monthly plan” system found in the US and Europe. This could be a hindrance to the diffusion of LBS in Uganda in the future. 7.2.8.8 LBS Business Model in the Private Sector For an LBS application to enjoy commercial success, it has to make business sense. For successful diffusion of LBS in Uganda, the private sector will need to find innovative ways of developing commercially viable location-enabled applications that will yield returns on investment. Telecom companies would only be willing to collaborate with content providers and 3 rd party developers in a revenue share model; however, only when an innovative idea makes business sense. 475 7.2.8.9 Privacy and Junk Messages Complaints from mobile subscribers about spam messaging have appeared in media reports, for example, see (BizCommunity, 2008). Tracking applications developed by Sentrack Systems and Orange Telecom (the Fleet Live tracking application) could possibly be used to compromise the privacy of unsuspecting subscribers, for example, members of opposition parties. However, because of the lack of a national personal identification system in Uganda, this issue is not foreseen to have immediate negative consequences. 7.2.8.10 Support from Government for LBS The biggest contribution of the Uganda government to LBS has been the promotion of the ICT sector through the setting up of a regulatory institution, the Uganda Communication Commission (UCC), and the Ministry of ICT. However, further support is needed to allow LBS to thrive. The lack of government interest in Zain/Ericsson’s Lake Victoria Search and Rescue mobile application is an indication that location- enabled mobile applications are not considered valuable enough the warrant financial support from the national treasury. 7.2.9 Weaknesses of the Analysis The analysis in this paper is based on two survey samples of individuals using mobile phones, five in-depth personal interviews with telecom companies, one in-depth interview with an NGO and one focus group discussion with the same NGO using mobile applications in their work. The first survey sample consisted of 182 individuals, and the second, 101 individuals. The first sample was used to analyze the level of penetration of 476 mobile phones in Uganda, while the second was used to assess the potential for LBS diffusion in the country. The in-depth interviews with telecom countries were used to assess the completeness of ICT infrastructure in Uganda. The interview and focus group discussion with the NGO, BROSDI, was used to perform a case study of the use of mobile technologies similar to LBS for development in developing countries. The analysis could have been deeper with a much larger sample of survey respondents, more interviews with the ICT sectors, and more focus group discussions. Further, 48% of the respondents of the first sample of182 individuals were either University or High School students. 53% of the respondents of the second survey sample of 101 individuals were university students. This could have biased the results of the analysis to reflect trends amongst the youth. However, 52% of this sample represented individuals that exhibited diversity in professions, age-groups and socio-economic status in Ugandan society. Mobile software applications in this paper are used to draw implications for LBS applications. In a strict sense, an LBS application has a location component inherent in its design. However, as this analysis has demonstrated, quite a number of mobile applications are existent in Uganda, however, few leverage the actual location of the device. In addition, the accuracy and precision of positioning methods vary for those applications that are more or less LBS applications by definition. 7.3 Diffusion Analysis of the Google SMS Application in Wainah Village Diffusion is “the process by which (1) an innovation (2) is communicated through certain channels (3) over time (4) among the members of a social system” (Rogers 2003, 11). 477 There are four main elements in diffusion theory including the innovation, communication channels, time, and a social system (Rogers, 2003). In this section, I analyze the diffusion of LBS in Uganda based on these four main elements of diffusion. 7.3.1 The Innovation The diffusion of LBS technology in Uganda is related to the diffusion of a number of technologies that enable LBS applications on mobile devices: the low-end GSM mobile phone, mobile telecommunication, the Internet/Web, GPS, and the smart phone. In this chapter, all mobile applications, deployed on the server-side, or client-side, are considered LBS applications as long as they leverage location in a query response. Such applications, as seen in earlier sections, include m-Health, social development, fleet management, and search and rescue applications. 7.3.1.1 The Technology Cluster The various elements of LBS technology all form a technology cluster. By definition, “a technology cluster consists of one or more distinguishable elements of technology that are perceived as being closely interrelated” (Rogers 2003, 14). To be able to understand the diffusion of LBS as an innovation in Uganda, consideration needs to be given to the diffusion of the elements in the technology cluster. 7.3.1.2 Perceived Attributes The characteristics of innovations, as perceived by individuals, can lead to different rates of adoption of an innovation (Rogers, 2003). There are five perceived attributes of any innovation: (1) relative advantage, (2) Compatibility, (3) complexity, (4) trialability, and (5) observability (Rogers, 2003). 478 In this section, I perform diffusion analysis of LBS technology in Uganda based mainly on the Google SMS applications developed by Grameen Foundation’s AppLab. This is because of the detailed case study of the NGO, BROSDI performed in Mayuge district, Wainah Village. In addition, I use the example of innovative location-based mobile applications in Kampala where a restaurant operator uses SMS messaging for taking location-specific orders for food on her low-end GSM Nokia mobile phone. Relative Advantage and Compatibility LBS applications, such as the Google SMS (Health) Tips, Farmer’s Friend, Clinic Finder, and Google Trader, have been perceived as being better than relying on hearsay information from friends and neighbors. These LBS applications are used mostly in rural areas and not so much in the urban areas mainly because people in urban areas have better access to information, such as through Internet cafés, Internet on personal computers and on office computers. Indeed, AppLab and Google developed these applications to target rural population with limited access to ICT services. The degree of relative advantage can be assessed in economic terms, social prestige factors, convenience, and satisfaction (Rogers, 2003). Economically, the lives of rural farmers have improved tremendously. The BROSDI case study shows that farmers in Mayuge district use the Farmer’s Friend application to obtain information about organic pest and weed control, and the weather, which eventually increases their crop yields. The Google Trader application allows farmers to inquire about market prices for their produce, so as to avoid being underpaid for their produce by middlemen. Moreover, the use of the LBS applications for obtaining information on 479 health, farming and other subjects empowers rural populations, which increases their social prestige. An application such as the clinic finder application is convenient as village folk are alerted of new health services in their area. The Google SMS application is compatible with the cultural norms of the farmers. There are no cultural barriers to the use of mobile phones and mobile applications. Complexity The level of complexity for users of the Google SMS technology is relatively low, and this has facilitated the diffusion of the technology. The Google SMS applications are all based on a simple user interface technology, the text message, also known as SMS. The reason SMS technology was chosen is because of the relatively high penetration rates of low-end GSM mobile phones that have limited capabilities. Voice calling and SMS are standard capabilities on these phones, hence the logical choice of technology. The mobile applications only require a user to be able type a 160 character message in English, or in their local language, and send it to a special number, 6001. A response in English is received in a few seconds, or minutes. The user must be able to read English, and if not, help can be sought from community workers at the local NGO offices, such as BROSDI in Mayuge district. Trialability and Observability The trialability and observability of the technology are high. Users are able to simply send a text message using their GSM phones to try the application, and receive an instant response. The results of using the technology are higher crop yields, which are easily observable by farmers in the village as greener and healthier crops in their neighbors’ 480 fields. This observation encourages discussion amongst peers in the village, which accelerates the diffusion process. Moreover, there are periodic community meetings organized by BROSDI. Farmers are invited to attend, ask questions about the technology, and see the results of using organic farming methods in the experimental demonstration plots at the BROSDI community center. 7.3.1.3 Re-invention Re-invention is important in the diffusion of a technology. In the diffusion of LBS, SMS technology was reinvented to deliver location-aware services to clients in addition to it being used a low-cost mobile communication technology. While doing this research, I found a restaurant operator using SMS to take orders for food from various clients around the city. Clients would specify a number of food orders, and indicate their location. The restaurant operator would package the food and have it delivered using public transportation, the “boda-boda” motorbikes. Such simple reinventions of innovation allow small business owners to increase their profit margins, and expand their business capacity. This allows for faster diffusion of the innovation. 7.3.2 Communication Channels Communication Channels are the second element of diffusion. “Diffusion is a particular type of communication where the message content that is exchanged is concerned with a new idea. … A communication channel is the means by which messages get from one individual to another” (Rogers 2003, 18). There are three common communication channels that allow for an innovation to diffuse from one individual to the other: (1) mass media, (2) interpersonal, and (3) interactive communication. All three played a role to 481 some degree; however, the relative degree of each channel cannot be empirically verified based on the data collected during this research. The Google SMS application was widely publicized in local newspapers. Posters and fliers about the application were made available through local MTN dealers, for example, see Figure 7-17. Because the target population was rural, advertising using electronic media would be ineffective. Most rural areas in the country are not connected to the electricity grid. Less than 2% of the rural population and 5% of the national population has access to electricity (Kaijuka, 2007). Wainah village in Mayuge district does not have grid-based electricity. In fact, the BROSDI offices in the village are powered by a diesel- generator so as to be able to run the computing facilities. 7.3.2.1 Heterophily and Homophily The transfer of ideas in a communication process occurs most frequently between two individuals who are similar in attributes such as beliefs, education, and socioeconomic status (Rogers, 2003). Homophily is the degree to which two or more individuals are similar in these attributes, while heterophily is the exact opposite. In the diffusion of the Google SMS application, both homophilous and heterophilous communication contributed to its diffusion. Ugandan society, especially in rural areas, is communal by nature, and very sociable. Farmers in the village who’ve used the innovation freely discuss their experiences with other farmers. Heterophilous communication occurs when these farmers assemble periodically for meetings at the BROSDI center in the village. Here, forums are conducted by the NGO’s representatives and agricultural experts where farmers are educated about organic farming methods, and 482 also informed about the Google SMS application, and how to use it. Farmer’s receive training in general ICT skills, and reading and writing in their local language for those who are illiterate. 7.3.3 Time Time is the third element in the diffusion process, and this dimension is manifest in the (1) innovation-decision process, (2) innovativeness of the individual, and (3) an innovation’s rate of adoption (Rogers, 2003). 7.3.3.1 Innovation-decision Process The innovation-decision process starts with an individual collecting some knowledge about an innovation, then being persuaded to adopt the innovation, and finally, making a decision to adopt or reject the innovation. Rogers (2003) conceptualizes this process as a five-step process: (1) knowledge, (2) persuasion, (3) decision, (4) implementation, and (5) confirmation. Through the meetings between farmers and BROSDI staff at their center in Wainah village, farmers gained knowledge about the Google SMS mobile application, and were persuaded by the BROSDI staff to use the technology. Farmers made a decision about adopting the technology, and followed this with put the innovation to use. Confirmation of the innovation occurred when farmers saw higher crop yields from their fields and obtained better market prices for their produce, in which case they decided to keep using the technology. 483 7.3.3.2 Innovativeness and Adopter Categories The innovativeness of an individual is measured by how early or late they adopt an innovation, in other words, are they (1) innovators, (2) early adopters, (3) early majority, (4) late majority, or (5) laggards (Rogers, 2003). This diffusion research did not collect data on the innovativeness of the individual. The rate of adoption of an innovation typically follows an S-shaped curve (Rogers, 2003). The rate of adoption starts slow with only a few innovators, then picks up steam when a critical mass of individuals has adopted the innovation (usually 10% of the population), at which stage the diffusion process occurs rapidly. The rate of adoption slows down after a period of time, and finally levels off as fewer individuals adopt the innovation (the laggards) (Rogers, 2003). This diffusion research did not collect data on the rate of adoption of LBS. 7.3.4 Social System The fourth and final element in the diffusion process is the social system (Rogers, 2003). 7.3.4.1 Social Structure The structure of a social system can facilitate or impede the diffusion of an innovation. Social structure is the patterned social relationships among the members of a system. As mentioned previously, rural societies in Uganda tend to be very closely-knit societies with high levels of social interaction. 7.3.4.2 Opinion Leaders and Change Agents Opinion leaders and change agents can drive the diffusion of an innovation. Opinion leadership is the degree to which an individual can influence attitudes of other individuals 484 in a social system in a desired direction with relative frequency (Rogers, 2003). A change agent is an individual that influences clients’ innovation-decisions in a direction that benefits a change agency (Rogers, 2003). Leadership is provided by the local government offices in the village. The head of the village is the Local Council Chairman, Level 1 (LC1). The LC1 is the smallest unit of government in Uganda, and is constituted by a handful of officials hierarchically under the Chairman. The LC1officials play an administrative role in their village, and are charged with intervening and officiating over social and civic matters concerning the residents in the village. The LC1 officials are opinion leaders in their villages. A successful diffusion model would need to involve the LC1 office of any village in helping to spread the news about a new diffusion in the village. People in rural areas are heavily influenced by opinion leaders in their villages, and these tend to be the local government officials. The change agent in the diffusion of the Google SMS application in Mayuge district is the local NGO, BROSDI. This organization runs three developmental programs in the village, one of which is CELAC. The objective of the CELAC program was to allow BROSDI officers to collect local knowledge on organic farming methods from the farmers, and then share it through the Google SMS application with rural communities in several districts around the country. 7.3.4.3 Consequences of the Innovation Consequences of an innovation can be both positive and negative. In the case of the Google SMS Farmer’s Friend and Google Trader applications, positive consequences 485 include the increase in access to information about organic farming methods amongst local farmers, higher crop yields, and higher prices for farm produce. Further, youth in the village have increased their knowledge about reproductive health and HIV/AIDS through the Google Health Tips application. Respondents in the BROSDI case study did not identify any negative consequences of the Google SMS applications. 7.4 Conclusion The potential for growth of LBS technology in the country is high. Mobile applications are actively being used in Uganda in myriad ways. These applications are mainly SMS- based, and are designed for low-end GSM phones, which constitute the majority of mobile phones owned by the population. The ICT infrastructure generally required to support LBS technology is firmly in place. GSM infrastructure covers both urban and rural areas of the country quite well. Broadband Internet, 2.5G and 3G wireless data technologies are available in all major towns and urban areas, and even in some rural areas, with coverage expanding every day. Telecom companies are currently the major drivers of ICT infrastructure in the country. Challenges to LBS diffusion in Uganda include the small number of smart phones currently in the market due to their relatively high prices; the relatively smaller geographic coverage of high speed Internet infrastructure, such as 3G and broadband; and the relatively high costs of mobile voice and data services in Uganda, mainly do the prepaid business model used by telecom companies in developing countries. Never the less, there is great potential for LBS technology in the future, especially with the decrease 486 in the prices of the smart phones in the future. LBS technology could be a major driver of the democratization of GIS in society in developing countries. It is recommended that future research focus on emerging high end smart phone based LBS applications in developing countries, and on negative consequences of the diffusion of these applications. 487 Chapter 8 : Summary and Conclusions 8.1 Aim and Research Questions The aim of the research presented in this dissertation was to analyze the evolution, current state, and future of Geographical Information Systems (GIS) technology in developing countries, based on a case study of Uganda. Such an understanding is important, first, to national governments in developing countries to assess how equipped they are to meet the challenges of physical planning and development, in light of the Millennium Development Goals (MDGs), and second, to Western international development agencies, such as the United States Agency for International Development (USAID) and the World Bank, to reexamine the objectives of their foreign aid policies with respect to technology transfer of software technologies to developing countries. There were three core research questions investigated in this dissertation. One, how did GIS evolve as a technology in developing countries? Two, what is the current state of GIS technology in the various sectors of the economy in developing countries? Three, what is the future potential for emerging mobile-device based geospatial technologies, known as location-based services (LBS), as alternative tools to GIS for spatial decision making in developing countries? Next, I present a summary of my research findings. 488 8.2 Summary of Research Findings In this section, I present a summary of the research findings based on each of the three research questions formulated in Chapter 1. Research Question 1: How did GIS evolve as a technology in developing countries? The introduction of GIS in Uganda coincided with the creation, in 1989, of an agency called the National Environment Information Center (NEIC) under the Ministry of Environment Protection. The United Nations Environmental Programme – Global Resource Information Database (UNEP-GRID) introduced Idrisi and PC Arc/INFO GIS software at NEIC in 1989. In 1995, NEIC morphed into the information unit of the National Environmental Management Authority (NEMA), a semi-autonomous agency “charged with the responsibility of coordinating, monitoring, regulating and supervising environmental management in the country” (NEMA, 2012). Around the same time that GIS was introduced by UNEP-GRID at NEIC, in 1989, another institution, the Department of Forestry under the Ministry of Environment Protection (Drichi, 2002), had GIS introduced into its workflow through a donor funded project supported by the Norwegian Agency for Development Cooperation (NORAD) in collaboration with the Norwegian Forestry Society (Drichi, 2002). In 1990, GIS was introduced at yet another institution in Uganda, the Wetlands Inspection Division, under the Ministry of Environmental Protection. This was under a program funded by the Dutch government, the Wetlands Management Program/Project, which was a project meant to protect wetlands in Uganda. 489 In the early 1990s, GIS was introduced for the first time in the academic sector. Makerere University Institute of Environment and Natural Resources (MUIENR) and the Department of Surveying, both at Makerere University Kampala, a government owned university, saw the introduction of GIS as a course in their respective curricula. A precursor to the project that introduced GIS at the Department of Surveys and Mapping in Uganda was the Kampala Mapping Project which had started in 1993, funded by a loan from the World Bank to the government of Uganda. Following this project, GIS was introduced at the Department by a French company under a project that started in December 1995. This was under a project called Computer Aided Mapping Project Uganda Surveys (CAMPUS), funded by the French Government, and implemented by a French company. The GIS software introduced was ArcView 3.1. From 1996 to 2002, various institutions had GIS introduced into their workflow, for example, the Famine Early Warning System Network (FEWSNET), the Department of Geography at Makerere University, the National Agricultural Research Laboratories (NARL), Makerere University’s Department of Forestry, Mobile Telephone Network (MTN), and Uganda Bureau of Statistics (UBOS). 2002 saw the first ESRI authorized reseller in Uganda open shop in Kampala, GeoInformation Communication (GIC). Most of the licenses for the ESRI family of GIS products are issued by GIC and ESRI Eastern Africa in Nairobi, Kenya. One of the major factors that supported the diffusion of GIS in Uganda was the academic institutions that trained staff of public sector institutions in developing countries in GIS, remote sensing and photogrammetry to equip them with the necessary skills needed to 490 support GIS workflows in their home countries. The International Institute for Aerospace Survey and Earth Science (ITC) in the Netherlands (Toppen, 1991), today known as the faculty of Geo-Information Science and Earth Observation, University of Twente, was one of the first academic institutions in the world to offer formal training to students from developing countries, mainly staff of public sector institutions. There were other institutions that played key roles in providing early GIS education and training to public sector employees from Uganda. The Royal Museum for Central Africa in Tervuren, Belgium, administered six-month long certificate level training in GIS to employees from the Wetlands Inspection Division and the Department of Geological Survey and Mines. The Regional Centre for Services in Surveying, Mapping and Remote Sensing (RCSSMRS) in Nairobi that had been set up by the United Nations Economic Commission for Africa (UNECA) offered short term training and workshops in GIS to employees of public sector institutions in Uganda, for example, staff from the National Agricultural Research Laboratories, and Kampala City Council. The surveying and mapping departments of the UK, Netherlands and Germany also trained staff from the Department of Surveys and Mapping, Uganda, in cartography and GIS. In Uganda, the first academic institution to provide GIS training and certificates was Makerere University Institute of Environment and Natural Resources (MUIENR) in the early 1990s. It was also arguably the first academic institution to obtain GIS. The Departments of Surveying, Geography, Forestry, Veterinary Medicine and Computer Science soon followed suite at Makerere University, some offering undergraduate courses, while others, graduate level courses and certificates in GIS. Kyambogo 491 University and Gulu University are the other public universities that offer GIS courses in their curricula today. A majority of the institutions in Uganda that use GIS in their workflows today employ the ESRI family of software. The reason for this trend was because Idrisi, PC Arc/INFO and ArcView were the first software introduced into Uganda by UNEP-GRID and other international development agencies. The second reason is the academic institutions that provided early GIS education to international students from Uganda; these universities and colleges used ESRI software in the training of their students. The most influential institution in this respect was ITC. A third reason could be because in 2002, GeoInformation Communication (GIC) was incorporated as the first ESRI value added reseller in Uganda. An analysis based on the diffusion of innovations perspective revealed that the major factors that allowed for the early adoption of GIS at institutions in Uganda include, (1) the type of tasks carried out at that institution based on its mandate, (2) government support for public sector institution activities, (3) and donor agency support for public sector institutions. Challenges faced by early adopters of GIS technology according to diffusion analysis included: (1) the cost of licensing the software, and related costs of maintenance, (2) insufficient training facilities within the country, (3) a lack of awareness about the importance of GIS technology, (4) the lack of a support network of GIS specialists in the country, (4) a culture of corruption and unethical institutional practice, (5) bureaucratic roadblocks in the acquisition of equipment and in the hiring of new GIS staff, (6) brain 492 drain (the loss of skilled personnel to developed countries), (7) a project-driven instead of problem-driven approach to GIS, and a (8) dependency on donor funding to support GIS activities. The most significant perceived attribute that led to the adoption of GIS technology was its relative advantage over old technologies such as paper, pen and ink cartography. Going from analog to digital methods helped expedite the map production process at the Department of Surveys and Mapping, for example. Heterophilous communication played a larger role than homophilous communication in the adoption of GIS in Uganda. Because of the lack of knowledge about GIS in the early 90s, homophilous communication among peers was not common place. In most cases, GIS experts from UNEP-GRID and other expatriates were the major agents as communication channels in a heterophilous relationship with Ugandan staff at public sector institutions. Social structure and norms were significant to the diffusion of GIS technology. Uganda’s public, private, academic and NGO sector institutions are part of a larger society where culture, norms and values play a major role in an individual’s decision making process. There is a culture of communal living, communal rights and ownership, and a respect for cultural values, norms, practices, and above all, a cautious attitude toward those in authoritative positions, for example, local politicians. There is a fear among local NGOs that adopting GIS might lead to controversial analytical results contradicting official government records, for example, those from the Uganda Bureau of Statistics (UBOS). 493 Research Question 2: What is the current state of GIS technology in the various sectors of the economy in developing countries? The diffusion of GIS in Uganda was investigated in five sectors: the public, academic, nongovernmental organization (NGO), International Organization (IO), and private. The public and IO sectors have seen the highest level of GIS penetration in Uganda as compared to the others. GIS in the public, academic, and NGO sectors is project-based, donor driven, and supported by agencies such as the World Bank, USAID, and NORAD. The IO sectors are well-funded by their parent governments in the West, while private sector GIS is mainly driven by funding from multinational corporations. ESRI ArcGIS software is dominant in all sectors. The licensing level of the software used is mainly the lowest level, “ArcView,” though the more powerful institutions like the multinational corporations use the next level of licensing, “ArcEditor.” Single-use licenses are used in most cases, as opposed to unlimited floating licenses. The high cost of software licensing is one of the major hindrances to the diffusion of GIS in the public, academic, NGO and private sectors. Primary geospatial data producers include: the Department of Surveys and Mapping, Uganda Bureau of Statistics, National Forestry Authority, and Wetlands Management Department. Most institutions obtain their data from these data producers. Data collection for GIS is mainly done by the use of handheld GPS receivers. Land surveying, remote sensing and photogrammetric methods are only used by public sector institutions that have such capacity, such as the Department of Surveys and Mapping. The sharing of data 494 among the various institutions is a major challenge due to bureaucracy, the lack of NDSDI, and the absence of geoinformation policy. An initiative to tackle this data sharing problem was undertaken by a group of NGOs; it is called the Uganda Clusters, and it equates to an operational, though unofficial, NSDI in Uganda. Training of employees in the use of GIS technologies is done both on-site and off-site. Usually week-long courses are offered in introductory desktop, mobile and server GIS courses. Because of the dominance of ESRI software, this training is often times provided by the ESRI business partner in Uganda, GIC. In some cases, employees acquire their training at ESRI Eastern Africa in Nairobi. Oftentimes public sector and NGO institutions do not offer GIS training to their employees citing high costs of training. Applications of GIS are in various fields including: environmental conservation, physical planning, topographic mapping, education, food security, telecommunication, public health, government, social rehabilitation, utility planning, water quality monitoring, and oil and natural gas prospecting. In most cases, the use of GIS is limited to mapping, inventorying, and the creation of thematic maps for presentation purposes, and only rarely is sophisticated GIS analysis performed. Government support for GIS as a field and discipline has historically been lacking. Only recently did the government recognize the importance of GIS in physical planning in its five year National Development Plan 2010/11-2014/15. The proposed intervention specifically prescribes the introduction of GIS technology into the institutional structure of central and local government, and for the training of staff in geospatial technology. 495 The most significant impact of GIS in the public sector on society is the contribution made by GIS to service delivery, for example, publicly owned utility parastatals, such as National Water and Sewerage Corporation, use GIS to manage various aspects of water delivery to communities. Other impacts of GIS include improved management of natural resources, such as forests and wetlands; better planning of peace and recovery projects in northern Uganda; and improved job prospects for university graduates. There are myriad challenges to GIS in the various sectors, and these can broadly be classified as political, institutional, economic, social, and technological. Political challenges include: bureaucracy in government leading to delays in acquiring GIS equipment and hiring of new GIS staff; a lack of political support for GIS activities; the lack of geoinformation sharing policy and an NSDI; political backlash for institutions whose analytical GIS results deviate from official government findings; and the decentralization of government leading to disunity among smaller departments within the same ministries. Institutional challenges include: the lack of decentralization of geospatial data collection activities at the various local government (jurisdiction) levels; the lack of a single GIS center at large universities such as Makerere University; the lack of unity among departments teaching GIS at university level; a shortage of educators in GIS due to low salaries at public universities; duplication of effort in data collection, and redundancy. Economic challenges to GIS include: the project-driven nature of GIS; the dependency on donor funding; poor funding for GIS activities from the government; high costs of GIS 496 licenses; lack of adequate equipment and hardware; high costs of training; and lack of support from the private sector. Social challenges to GIS include: low salaries for public servants; job insecurity; the threat of GIS to older professions of cartography and draughting; a culture of corruption, misappropriation of public funds, and impunity for perpetrators; poor attitudes towards work in all institutional sectors; brain drain; the lack of relevance of the current education system in the country; and the lack of awareness about GIS. Technological challenges include: a lack of adequate ICT infrastructure at public sector institutions; the lack of enforced standards for data quality monitoring and interoperability; mismatches in geospatial data from various institutions; the lack of a central data repository in the country; lack of NSDI; and lack of adequate technical support for open source GIS. Research Question 3: What is the future potential for emerging mobile-device based geospatial technologies, known as location-based services (LBS), as alternative tools to GIS for spatial decision making in developing countries? In Uganda, LBS systems and software applications have a high potential for success in the future as a tool for supporting spatial decision making and spatial thinking. Trends in mobile phone ownership indicate that a very high percentage of Ugandans own at least one mobile phone. This trend is across the board, irrespective of age, profession, socio- economic status, or education level of individuals in society. 497 The mobile phones owned by local Ugandans are mainly based on GSM technology, and capable of voice and SMS (text message) communication. Smart phone penetration is currently low. Smart phone ownership is characteristic of urban populations. Capabilities of available mobile devices vary with respect to GPS and Internet/Web capability, two important ingredients for LBS systems and applications. Capabilities related to the speed of data communication vary as well, the majority of smart the phones being 2.5G (GPRS/EDGE) capable, and only a few 3G/3G+ (faster data speeds) capable. 3G/3G+ infrastructure is limited to urban areas with little to no coverage in the rural areas. 3g/3G+ and 4G data speeds are more suited for LBS, as opposed to 2.5G. Wireless Internet access through the 2.5G and 3G telecom infrastructure has become increasingly affordable boosting subscriber numbers. Social networking, Google searches, and Email are the major drivers of Internet usage in Uganda. Location is increasingly playing an important role in usage patterns of mobile phones. Due the fact that the majority of handsets are low-end GSM phones, SMS is the most ubiquitous form of data communication. This explains the diverse SMS-based mobile applications currently available in Uganda, for example, (Grameen Foundation) AppLab’s Google SMS application. Such applications, so called Value Added Services (VAS), are the drivers of mobile application development in Uganda, and precursors to sophisticated LBS applications. Indeed, SMS-based applications that leverage cell-ID locations to provide a response to a query can indeed be classified as LBS applications, for example, the Google SMS Clinic Finder and Google SMS Weather. 498 Thus, I argue that LBS applications are already in existence in Uganda. Sophisticated LBS mobile applications on the Ugandan market include fleet management and Vehicle tracking applications from companies like Sentrack Systems, and Orange Telecom. The most popular mobile application in Uganda by far is MobileMoney from the telecom company, MTN. Other telecom networks also provide similar services, where money is transferred from location A to location B in a matter of minutes using a mobile phone. Other location-enabled mobile applications and services include the Community Knowledge Worker (CKW) developed by AppLab, various m-Health applications to link patients in rural areas with doctors in urban areas, and social development applications such as CU@School developed by Makerere University’s National Software Incubation Center. Software development platforms that have been used for the development of mobile applications include RapidSMS, FrontlineSMS, and OpenXdata, all of which are open source. There is a growing grass roots movement in Uganda called Mobile Monday Kampala (MoMoKla) whose objective is to foster mobile application development for development purposes. Volunteers meet periodically to brainstorm and initiate new mobile based projects, especially in the health and development sectors. Findings from this research indicate a substantial interest in LBS applications for location-based queries on land parcel ownership, environmental degradation, location- based reporting of illegal wetland encroachment, and reporting of crime incidences, illegal “tapping” of water and electricity, disease outbreaks and potholes in roads. 499 LBS applications have the potential to instill spatial thinking in society because of the inherent use of the user’s current location in a query. In fact, evidence from this research points to the fact that one in three people already use Google Maps and Google Earth regularly to make spatial decisions. The most common location determination technique used by Telecom companies in Uganda is the Cell-ID technique. Network-based location determination is currently not used on a commercial scale. Zain telecom (Airtel), in collaboration with Ericsson Uganda developed a location server for search and rescue operations on Lake Victoria based on the Cell Global Identity Timing-Advance (CGI-TA) technique, but this service remained a proof of concept due to a lack of government interest. GPS and Assisted-GPS (AGPS) positioning are some of the more accurate mobile positioning techniques (compared to Cell-ID), however, only the fleet management and tracking applications currently leverage this type of positioning in Uganda. There is an enormous business potential for third party software developers to enter the LBS market in Uganda as VAS developers. However, telecom companies only give audience to those applications that make business sense. Third party VAS developers and content providers currently in the Ugandan market include SMS Media and BlueCube. Coming up with an exciting and potentially successful business idea is a challenge to the future LBS industry. Other challenges to LBS in Uganda include: inadequate 3G infrastructure in rural areas; a lack of accurate mobile-positioning techniques, digital street data, private sector involvement, and government support for, and interest in, LBS; the absence of locational- 500 context in content disseminated by content providers in the market; low smart phone penetration rates; high costs of telecom services (voice, SMS and data) due to the prepaid “scratch card” model; user privacy issues and spam messaging. Next, discuss the contributions of this dissertation. 8.3 Contributions and Implications On the academic level, this research contributes to the literature on diffusion of innovations, and also to the body of literature known today as the GIS and Society discourse. Facets of diffusion theory have been tested in this mixed methods research, and for the most part, the findings are consistent with the theory. The author’s main critique of diffusion theory is the failure of the theory to specifically address the economic condition of adopters of an innovation. Well known criticisms of diffusion theory include its pro- innovation bias, the individual-blame bias, the recall problem, and the issue of equality (Rogers, 2003). This research makes a contribution to the growing body of literature, GIS and Society. This literature tends to have a Western focus, and thus, this research offers a developing countries’ perspective to each of the five major perspectives of this conceptual framework. In fact, the author suggests that a separate perspective be formally recognized in GIS and Society: the “developing countries perspective.” This is mainly because of the unique issues and concerns with respect to the impact of GIS on society, characteristic of countries in the developing world. 501 A contribution is also made to the literature on Information and Communication Technology for Development (ICT4D) and Mobile for Development (M4D). This is a growing body of literature that has attracted a lot of interest in recent scholarly work especially with the widespread diffusion of ICT and Mobile technology in the developing world. In particular, perspectives provided through the lens of diffusion theory in this dissertation research are considered a significant contribution to the field. This research would be useful to International organizations such as the World Bank, United Nations agencies and international development agencies, such as USAID, DFID, and NORAD. Findings from this research would allow these organizations to reassess their current policies with regard to technology transfer of software technologies, such as GIS, to developing countries. Finally, the findings in this research would be useful to national governments of developing countries that are in the process of establishing infrastructure to support physical planning activities, for example, the government of Uganda that plans to establish a national GIS center by 2015 according to the 2010-2015 national development plan (GoU, 2010). I make suggestions for future research in the next and final section. 8.4 Suggestions for Further Research In the future, it is suggested that comprehensive research be carried out to establish the exact rate of diffusion in each of the sectors of the Ugandan economy. This would require a very large and complete sample of institutions covered in the research. The exact shape 502 of the diffusion curve can then be determined to estimate the degree of adoption and the critical mass of adoption. It is also suggested that longitudinal case studies of GIS diffusion at major public, NGO and private sector organizations be carried out in the future to develop a deeper understanding of the innovation-decision process, and consequences of GIS diffusion in developing countries. Finally, it is recommended that future research focus on emerging high end smart phone based LBS applications in developing countries, and on positive and negative consequences of their diffusion. 503 References ACM. (2005). Cerf’s Up Again! A New Ubiquity Interview with Vint Cerf. ACM: Ubiquity - Information Everywhere. Retrieved September 17, 2012, from http://ubiquity.acm.org/article.cfm?id=1066345 Abbas, R. (2011). The social implications of location- based services : an observational study of users. Journal of Location Based Services, 5(3-4), 156–181. Adjibolosoo, S. B.-S. K., & Ofori-Amoah, B. (1998). Misconceptions about Africa’s Underdevelopment and Development. In Senyo B-S. K. Adjibolosoo & B. Ofori-Amoah (Eds.), Addressing Misconceptions About Africa’s Development (pp. 1–8). Lewiston, NY, USA: The Edwin Mellen Press. Adu-Febiri, F. (1998). Social Inequality and Africa’s Underdevelopment: A Critical Assessment. In Senyo B-S. K. Adjibolosoo & B. Ofori-Amoah (Eds.), Addressing Misconceptions About Africa’s Development (pp. 38–49). Lewiston, NY, USA: The Edwin Mellen Press. AfDevInfo. (2008). Organisation Record: Celtel Uganda. Retrieved August 14, 2012, from http://www.afdevinfo.com/htmlreports/org/org_69215.html AfricaOnline. (2012). Africa Online: Our History. Africa Online: Our History. Retrieved August 14, 2012, from http://www.africaonline.com/index.php/our-history Agaba, E., & Shipman, N. (2006). Public Procurement Reform in Developing Countries: The Uganda Experience. In G. Piga & K. V. Thai (Eds.), Advancing Public Procurement: Practices, Innovation and Knowledge-sharing (pp. 373–391). Academic Press. Retrieved from http://www.ippa.ws/IPPC2/BOOK/Chapter_16.pdf Ahsan, K., & Gunawan, I. (2010). Analysis of cost and schedule performance of international development projects. International Journal of Project Management, 28(1), 68–78. doi:10.1016/j.ijproman.2009.03.005 Aina, Y. A. (2009). Geomatics Education in Saudi Arabia: Status, Challenges and Prospects. International Research in Geographical and Environmental Education, 18(2), 111–119. doi:10.1080/10382040902861197 Airtel. (2012). Airtel: Internet on 3.75G - FAQ. Internet on 3.75G. Retrieved August 13, 2012, from http://www.africa.airtel.com/wps/wcm/connect/africaairtel/uganda/3g/home/faq/ Aker, J. C., & Mbiti, I. M. (2010). Mobile Phones and Economic Development in Africa. Journal of Economic Perspectives, 24(3), 207–232. 504 Akingbade, A., Navarra, D. D., & Georgiadou, Y. (2009). A 10 Years Review and Classification of the Geographic Information Systems Impact Literature ( 1998-2008 ). Nordic Journal of Surveying and Real Estate Research, 4, 84–116. Akinyemi, Felicia O. (2011). Evaluating Access to Spatial Data and Information in Rwanda. Retrieved September 5, 2011, from http://www.urisa.org/files/spatialdatainRwanda.pdf Akinyemi, Felicia O. (2007). Spatial Data Needs for Poverty Management. In H. Onsrud (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (pp. 261–278). Redlands, CA, USA: ESRI Press. Akinyemi, Felicia Olufunmilayo, Uwayezu, E., & Simbizi, M. C. D. (2011). An Assessment of the Current State of Spatial Data Sharing in Rwanda. International Journal of Spatial Data Infrastructures, (Under Review), 19. Akpan-obong, P., Thomas, C. A., Samake, K. D., Niwe, M., & Mbarika, V. W. (2009). An African Pioneer Comes of Age : Evolution of Information and Communication Technologies in Uganda. Journal of Information, Information Technology, and Organizations, 4. Retrieved from http://jiito.org/articles/JIITOv4p147-171Akpan413.pdf Alesina, A., & Dollar, D. (2012). Who Gives Foreign Aid to Whom and Why? Journal of Economic Growth, 5(1), 33–63. Retrieved from http://www.jstor.org/stable/pdfplus/40216022.pdf Alford, J. (2009). Metadata Challenges Faced by Producers and Users of Spatial Data in South Africa. University of KwaZulu-Natal, Pietermaritzburg, South Africa. AllAfrica. (2012, April). Uganda: National ID Card Project Rakes Billions in Losses. AllAfrica.com. Kampala. Retrieved from http://allafrica.com/stories/201204130995.html Amadra, O.-O. (2003a). GIS - Challenges and Opportunities for Effective Data Sharing. Kampala, Uganda. Amadra, O.-O. (2003b). Geographic Information Systems: Challenges for Effective Data Sharing. Kampala, Uganda. Andersson, C. (2012). Mobile Positioing - Where You Want to be? Wirelss Developer Network. Retrieved August 23, 2012, from http://www.wirelessdevnet.com/channels/lbs/features/mobilepositioning.html AppLab. (2012a). About Grameen Foundation AppLab: History. About Grameen Foundation AppLab: History. Retrieved August 16, 2012, from http://www.grameenfoundation.applab.org/about-us.html AppLab. (2012b). Grameen Foundation Applab in Action. Grameen Foundation AppLab in Action. Retrieved August 16, 2012, from http://www.grameenfoundation.applab.org/AppLab-Ag.html 505 Arminen, I. (2007). Review Essay: Mobile Communication Society? Acta Sociologica, 50(4), 431–437. doi:10.1177/0001699307083983 Arnaud, A. M. (1993). Imapact of a GIS on a Southern European Country: The Case of Portugal. In I. Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (pp. 245–260). Dordrecht, The Netherlands: Kluwer Academic Publishers. Arnaud, A. M., Vasconcelos, L. T., & Geirinhas, J. D. (1996). Portugal: GIS Diffusion and the Modernization of Local Government. In I. Masser, H. J. Campbell, & M. Craglia (Eds.), GIS diffusion: the Adoption and Use of Geographical Information Systems in Local Government in Europe (pp. 111–124). London, UK: Taylor & Francis. Asiimwe, D., & Musisi, N. B. (2007). Decentralization and Transformation of Governance in Uganda (p. 347). Kampala, Uganda: Fountain Publishers Ltd. Assimakopoulos, D. (1996). Greece: The Development of a GIS Community. In I. Masser, H. J. Campbell, & M. Craglia (Eds.), GIS diffusion: the Adoption and Use of Geographical Information Systems in Local Government in Europe (pp. 147–162). London, UK: Taylor & Francis. Ayogu, M. D. (1999). Case Studies: Private Sector Participation in Infrastructure in Uganda , Ghana and Nigeria. BBC. (2012). Costa Concordia disaster: Crew urged “return to cabins.” BBC News Europe. Retrieved April 17, 2012, from http://www.bbc.co.uk/news/world-europe-16641592 BISS. (2011). Information System Definition. Business Intelligence Secret Solutions. Retrieved from http://www.business-intelligence-secrets.com/articles-directory/definition-of- information-system/ BROSDI. (2012a). Busoga Rural Open Source Develoment Initiative - BROSDI. Busoga Rural Open Source Develoment Initiative - BROSDI. Retrieved May 16, 2012, from http://www.brosdi.or.ug/ BROSDI. (2012b). Collecting and Exchange of Local Agricultural Content - CELAC. Collecting and Exchange of Local Agricultural Content - CELAC. Retrieved May 16, 2012, from http://www.celac.or.ug/ Babbie, E. (2007). The Practice of Social Research (11th ed., p. 511). Belmont, CA, USA: Thomson Higher Education. Bagozzi, R. P. (2007). The Legacy of the Technology Acceptance Model and a Proposal for a Paradigm Shift . Journal of the Association for Information Systems (JAIS), 8(4), 244–254. Retrieved from http://aisel.aisnet.org/cgi/viewcontent.cgi?article=1406&context=jais BalancingAct-Africa. (2010). GSM Association launches Save and Rescue scheme with operators and vendors for Lake Victoria. Balancing Act Africa: GSM Association launches Save and 506 Rescue scheme with operators and vendors for Lake Victoria. Retrieved August 22, 2012, from http://www.balancingact-africa.com/news/en/issue-no-488-0/telecoms/gsm- association-laun/en Balogun, E. E. (2004). Milestones and Challenges of ARCSSTE-E. 47th Session of the United Nations Committee on the Peaceful Uses of Outer Space (COPUOS), June 2-11, 2004, Vienna, Austria (p. 33). Vienna, Austria: United Nations Committee on the Peaceful Uses of Outer Space (COPUOS). Retrieved from http://www.oosa.unvienna.org/pdf/sap/centres/ARCSSTE-E.pdf Bandura, A. (1976). Social Learning Theory (p. 247). Prentice-Hall, Inc. Bandura, A. (1986). Social Foundations of Thought and Action: A Social Cognitive Theory (p. 544). Prentice-Hall, Inc. Bandura, A. (2001a). Social Cognitive Theory of Mass Communication. Media Psychology, 3(3), 265–299. doi:10.1207/S1532785XMEP0303_03 Bandura, A. (2001b). Social cognitive theory: an agentic perspective. Annual review of psychology, 52, 1–26. doi:10.1146/annurev.psych.52.1.1 Banks, K. (2009). On the Frontline: Technology, Design and Social Change. Technology. Kiwanja.net. Banks, K., & Burge, R. (2004). Mobile Phones : An Appropriate Tool For Conservation And Development? Cambridge, UK. Retrieved from http://www.kiwanja.net/media/docs/Appropriate-ICT-Report.pdf Barrett, M., Sahay, S., & Walsham, G. (2001). Information Technology and Social Transformation: GIS for Forestry Management in India. The Information Society, 17(1), 5– 20. Basaza, R., Criel, B., & Van der Stuyft, P. (2008). Community Health Insurance in Uganda: Why does Enrolment Remain Low? A View from Beneath. Health policy (Amsterdam, Netherlands), 87(2), 172–84. doi:10.1016/j.healthpol.2007.12.008 Basheka, B. (2008). Procurement Planning and Local Governance in Uganda : A Factor Analysis Approach. International Journal of Procurement Management, 2(2), 191–209. Retrieved from http://www.irspm2008.bus.qut.edu.au/papers/documents/pdf/Basheka - PROCUREMENT PLANNING AND LOCAL GOVERNANCE IN UGANDA A FACTOR ANALYSIS APPROACH - IRSPM - 2008.pdf Batungi, N. (2008). Land Reform in Uganda: Towards a Harmonised Tenure System (p. 261). Kampala, Uganda: Fountain Publishers Ltd. 507 Bayes, A., von Braun, J., & Akhter, R. (1999). Village Pay Phones and Poverty Reduction: Insights from a Grameen Bank Initiative in Bangladesh. Bonn, Germany. Retrieved from http://www.zef.de/fileadmin/webfiles/downloads/zef_dp/zef_dp8-99.pdf Bennet, S., Creese, A., & Monasch, R. (1998). Health Insurance Schemes for People Outside Formal Sector Employment (p. 74). Geneva, Switzerland. Retrieved from http://libdoc.who.int/hq/1998/WHO_ARA_CC_98.1.pdf Beshah, G. (2003). Developing Guidelines for Institutional Arrangements of the Federal Governmental Geo-Information Provider Organizations In Ethiopia. Geo-Information Science. INTERNATIONAL INSTITUTE FOR GEO-INFORMATION SCIENCE AND EARTH OBSERVATION ENSCHEDE, THE NETHERLANDS. BizCommunity. (2008). Uganda: SMS marketing irks MTN subscribers. BizCommunity. Retrieved August 23, 2012, from http://www.bizcommunity.com/Article/220/78/28068.html Blankenbach, J., & Norrdine, A. (2011). Building information systems based on precise indoor positioning. Journal of Location Based Services, 5(1), 22–37. doi:10.1080/17489725.2010.538016 Blaschke, S., & Weikel, T. (2010). T4D Vision. Kampal, Uganda: UNICEF. BlueCube. (2012). About BlueCube. BlueCube. Retrieved August 23, 2012, from http://www.bluecube.co.ug/about-bluecube Blumer, H. (1969). Symbolic interactionism: perspective and method (p. 208). University of California Press. Retrieved from http://books.google.com/books?hl=en&lr=&id=HVuognZFofoC&pgis=1 Borges, K. A. de V., & Sahay, S. (2000). GIS for the Public Sector : Experiences from the City of Belo Horizonte , Brazil. Information Infrastructure and Policy, 6, 139–155. Braa, J., Mohamed, W., Hanseth, O., Shaw, V., & Heywood, A. (2007). Developing Health Information Systems In Developing Countries: T he Flexible Standards Strategy. MIS Quarterly, 31(August), 1–22. Braa, J., Monteiro, E., & Sahay, S. (2004). Networks of Action: Sustainable Health Information Systems Across Developing Countries. Management Information Systems Quarterly (MIS Quarterly), 28(3), 337–362. Braa, J., & Muquinge, H. (2004). Building Collaborative Networks in Africa on Health Information Systems and Open Source Software Development– Experiences from the HISP/BEANISH Network. MIS Quarterly, 28(3), 337–362. Brahma, A. (2010). Prospects of Mobile Monday Kampala (MoMoKla ). Workshop on Mobile Applications, Faculty of Computing and Informatics Technology, Makerere University, 508 Kampala , 30th July 2010. Kampala: Faculty of Computing and Informatics Technology, Makerere University, Kampala, Uganda. Brunn, S. D., Dahlman, C. T., & Taylor, J. S. (1998). GIS Uses and Constraints on Diffusion in Eastern Europe and the Former USSR. Post-Soviet Geography and Economics, 39(10), 566–587. Budhathoki, N. R., & Nedović-Budić, Z. (2007a). Expanding the Spatial Data Infrastructure Knowledge Base. In H. Onsrud (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (Vol. 5, pp. 7–31). Redlands, CA, USA: ESRI Press. Budhathoki, N. R., & Nedović-Budić, Z. (2007b). Expanding the Spatial Data Infrastructure Knowledge Base. In H. Onsrud (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (pp. 7–32). Redlands, CA, USA: ESRI Press. Businge, J. (2012, August). Phones to Overtake People. The Independent. Retrieved from http://independent.co.ug/business/business-news/6219-phones-to-overtake-people Campbell, H. J. (1993). GIS Implementation in British Local Government. In I. Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (pp. 117–146). Dordrecht, The Netherlands: Kluwer Academic Publishers. Campbell, H. J. (1996). Theoretical Perspectives of the Diffusion of GIS Technologies. In I. Masser, H. J. Campbell, & M. Craglia (Eds.), GIS diffusion: the Adoption and Use of Geographical Information Systems in Local Government in Europe (pp. 23–45). London, UK: Taylor & Francis. Campbell, J., Hardy, T. E., & Barnard, R. C. (2010). Emerging Markets for Satellite and Aerial Imagery. In J. D. . Bossler, J. B. . Campbell, R. B. . Mcmaster, & C. Rizos (Eds.), Manual of Geospatial Science and Technology (2nd ed., pp. 423–437). Boca Raton, FL, USA: CRC Press. Cartwright, T. J. (1991). Information Systems for Urban Management in Developing Countries: The Concept and Reality. Computer, Environmental and Urban Systems, 15, 7–21. Cartwright, T. J. (1993). Geographic Information Technology as Appropriate Technology for Development. In I. Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (pp. 261–274). Dordrecht, The Netherlands: Kluwer Academic Publishers. Cavric, B. I., Nedović-Budić, Z., & Ikgopoleng, H. G. (2003). Diffusion of GIS Technology in Botswana. The Interdisciplinary Design and Research e-Publication (IDRP), 25(2). Cerf, V. (2009). Tracking the Internet into the 21st Century. Youtube.com: Vint Cerf at U Maryland, 17th April 2009. Retrieved September 17, 2012, from http://www.youtube.com/watch?v=gTfFB7L85Hc 509 Chaminama, N. (2009). Analysis of Public Sector Cooperation and Geoinformation Sharing: A Resource Dependence Prespective. Geo-Information Science. International Institute for Geo-information Science and Earth Observation, ITC, Enschede, The Netherlands. Chan, T. O., Feeney, M., Rajabifard, A., & Williamson, I. (2001). The Dynamic Nature of Spatial Data Infrastructures: A Method of Descriptive Classification. Geomatica, 55(1), 451–462. Chan, T. O., & Whitworth, R. (2003). SDI Development: Roles of Local and Corporate SDIs. In I. Williamson, A. Rajabifard, & M.-E. F. Feeney (Eds.), Developing Spatial Data Infrastructures: From Concept to Reality (pp. 165–182). Boca Raton, FL, USA: CRC Press. Chan, T. O., & Williamson, I. P. (1999a). The Different Identities of GIS and GIS Diffusion. International Journal of Geographical Information Science, 13(3), 267–281. Chan, T. O., & Williamson, I. P. (1999b). A Model of the Decision Process For GIS Adoption and Diffusion In A Government Environment. Urban and Regional Information Systems Association (URISA) Journal, 11(2), 7–16. Chan, T. O., & Williamson, I. P. (2000). Long Term Management of a Corporate GIS. International Journal of Geographical Information Science, 14(3), 283–303. doi:10.1080/136588100240859 ChildCount. (2012). ChildCount+: Overview. ChildCount+: Overview. Retrieved August 19, 2010, from http://www.childcount.org/about/ Cho, G. C. H. (2008). Geographic Information Science , Personal Privacy , and the Law. In J. P. Wilson & S. A. Fotheringham (Eds.), The Handbook of Geographic Information Science (pp. 519–539). Malden, MA, USA. Cho, Y. J. (1997). Credit Policies and the Industrialization of Korea: Lessons and Strategies. In K. B. Staking (Ed.), Policy-based Finance and Market Alternatives: East Asian Lessons for Latin America and the Carribean. Washington D.C., USA: Inter-American Development Bank. Retrieved from http://books.google.com/books?hl=en&lr=&id=ukd2uEOxyNEC&oi=fnd&pg=PA17&dq= Credit+supporting+private+sector+&ots=fd4arAQJhD&sig=uprb7trQEs1dm6pzNd33V1mu 8TM#v=onepage&q=Credit supporting private sector&f=false Chrisman, N. R. (1998). Academic Origins of GIS. In T. W. Foresman (Ed.), The History of Geographic Information Systems: Perspectives from the Pioneers (pp. 33–43). Upper Saddle River, NJ, USA: Prentice-Hall, Inc. Chrisman, N. R. (2006). Charting the Unknown: How Computer Mapping at Harvard Became GIS (p. 218). Redlands, CA, USA: ESRI Press. Chu, W.-L., Liu, X., & Wu, F.-S. (2010). Diffusion of Telecommunication Technologies: A study of Mobile Telephony. In J. Tidd (Ed.), Gaining Momentum: Managing the Diffusion of Innovations (pp. 283–312). London, UK: Imperial College Press. 510 Churchill, W. (1908). My African Journey. London, UK: Hodder & Stoughton. Chuttur, M. (2009). Overview of the Technology Acceptance Model: Origins , Developments and Future Directions. Bloomington, Indiana. Retrieved from http://sprouts.aisnet.org/785/1/TAMReview.pdf Ciancarella, L., Craglia, M., Ravaglia, E., Secondini, P., & Valpreda, E. (1996). Italy: GIS and Administrative Decentralization. In I. Masser, H. J. Campbell, & M. Craglia (Eds.), GIS diffusion: the Adoption and Use of Geographical Information Systems in Local Government in Europe (pp. 87–110). London, UK: Taylor & Francis. Cipriano, P. (2006). SITAD: Building a Local Spatial Data Infrastructure in Italy. In M. Campagna (Ed.), GIS for Sustainable Development (pp. 489–500). Boca Raton, FL, USA: CRC Press. Clarke, D., Hedberg, O., & Watkins, W. (2003). Develpment of the Australian Spatial Data Infrastructure. In I. Williamson, A. Rajabifard, & M.-E. F. Feeney (Eds.), Developing Spatial Data Infrastructures: From Concept to Reality (pp. 131–146). Boca Raton, FL, USA: CRC Press. Clements, P. (1993). An approach to poverty alleviation for large international development agencies. World Development, 21(10), 1633–1646. doi:10.1016/0305-750X(93)90098-T Commonwealth-Network. (2012). Uganda: Government Ministries. Commonwealth Network. Retrieved July 5, 2012, from http://www.commonwealth-of- nations.org/Uganda/Government/Government_Ministries Compeau, D., Higgins, C. A., & Huff, S. (1999). Social Cognitive Theory and Individual Reactions to Computing Technology: A Longitudinal Study. MIS Quarterly, 23(2), 145– 158. Compeau, D. R., & Higgins, C. A. (1995). Application of Social Cognitive Theory to Training for Computer Skills. Information Systems Research, 6(2), 118–143. Retrieved from http://www.jstor.org/stable/pdfplus/23011006.pdf Comte, A. (1851). A General View of Positivism. Paris: Hertford: Stephen Austin. Retrieved from http://books.google.com/books?id=SgaHpaeZAewC&dq=“A+General+View+of+Positivis m”&pg=PP1&ots=iubuk2ztrh&source=bn&sig=gYv9EMnFcGza5T8BlOivnQ0aREM&hl= en&sa=X&oi=book_result&resnum=4&ct=result#v=onepage&q=“A General View of Positivism”&f=false Corbett, J., Rambaldi, G., Kyem, P., Weiner, D., Olson, R., Muchemi, J., Mccall, M., et al. (2006). Mapping for Change - The Emergence of a New Practice. Participatory Learning and Action, (54), 155. 511 Corbett, S. (2008). Can the Cellphone Help End Global Poverty? New York Times. Retrieved August 10, 2012, from http://www.nytimes.com/2008/04/13/magazine/13anthropology- t.html?pagewanted=all Corbin, J., & Strauss, A. (2008). Basics of Qualitative Research (3rd ed., p. 379). Thousand Oaks, CA, USA: SAGE Publications Ltd. Cottarelli, C., Dell’Ariccia, G., & Vladkova-Hollar, I. (2005). Early birds, late risers, and sleeping beauties: Bank credit growth to the private sector in Central and Eastern Europe and in the Balkans. Journal of Banking & Finance, 29(1), 83–104. doi:10.1016/j.jbankfin.2004.06.017 Couclelis, H., Nyerges, T. L., & McMaster, R. (2011). GIS and Society Research: Reflections and Emerging Themes. In T. L. Nyerges, H. Couclelis, & R. McMaster (Eds.), The SAGE Handbook of GIS and Society (pp. 531–541). London, UK: SAGE Publications Ltd. Craig , W. J. (1995). Why We Can’t Share Data: Institutional Inertia. In H. J. Onsrud & G. Rushton (Eds.), Sharing Geographic Information (pp. 107–119). New Brunswick, New Jersey: Center for Urban Policy Research, Rutgers, The State University of New Jersey. Cremona, G., & Ciancarella, L. (2006). GIS Application to Support Water Infrastructures Facilities Localization in Particularly Valuable Environmental Areas: The Eolian Islands Case Study. In M. Campagna (Ed.), GIS for Sustainable Development (pp. 403–416). Boca Raton, FL, USA: CRC Press. Creswell, J. W. (2012). When Should I Choose a Mixed Methods Approach? Methods. Sage Publications, Inc. Crompvoets, J., Bouckaert, G., Vancauwenberghe, G., Vandenbroucke, D., Orshoven, J. V., Janssen, K., Dumortier, J., et al. (2008). Interdisciplinary Research Project: SPATIALIST; Spatial Data Infrastructures and Public Sector Innovation in Flanders (Belgium). GSDI 10 - 10th International Conference for Spatial Data Infrastructure, 25-29 February 2008, St. Augustine, Trinidad (p. 24). St. Augustine, Trinidad: GSDI. Curran, K., Furey, E., Lunney, T., Santos, J., Woods, D., & McCaughey, A. (2011). An evaluation of indoor location determination technologies. Journal of Location Based Services, 5(2), 61–78. doi:10.1080/17489725.2011.562927 Curran, K., & Hubrich, S. (2009). Optimising mobile phone self-location estimates by introducing beacon characteristics to the algorithm. Journal of Location Based Services, 3(1), 55–73. doi:10.1080/17489720902776738 Câmara, G., Fonseca, F., Monteiro, A. M., & Onsrud, H. (2005). Networks of Innovation and the Establishment of a Spatial Data Infrastructure in Brazil. Information Technology for Development, 12(4 (Special issue, part I: Implementation of spatial data infrastructures in transitional economies)), 255–272. doi:10.1002/itdj.20047 512 Câmara, G., Fonseca, F., Monteiro, A. M., & Onsrud, H. (2006). Networks of Innovation and the Establishment of a Spatial Data Infrastructure in Brazil. Information Technology for Development, 12(4), 255–272. doi:10.1002/itdj Câmara, G., Fonseca, F., Onsrud, H., & Monteiro, A. M. (2004). Efficacious Sustainability of GIS Development within a Low Income Country: the Brazilian Experience. Image Processing. Brazil. Dahunsi, F., & Dwolatzky, B. (2012). An Emprirical Investigation of the Accuracy of Location- based Services in South Africa. Journal of Location Based Services, 6(1), 22–34. Dale, P. (1991). Education in Land Information Management. Cartographica, 28(3 (GIS Education and Training)), 23–30. Dangermond, J. (1995). Public Data Access: Another Side of GIS Data Sharing. In H. J. Onsrud & G. Rushton (Eds.), Sharing Geographic Information (pp. 331–339). New Brunswick, New Jersey: Center for Urban Policy Research, Rutgers, The State University of New Jersey. Dangermond, J., & Smith, L. K. (1988). Geographic Information Systems and the Revolution in Cartography: The Nature of the Role Played by a Commercial Organization. The American Cartographer, 15(3), 301–310. doi:10.1559/152304088783886964 Davis, C. A., & Fonseca, F. (2011). Considerations from the Development of a Local Spatial Data Infrastructure. In Z. Nedović-Budić, J. Crompvoets, & Y. Georgiadou (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 181–202). Boca Raton, FL, USA: CRC Press. Davis Jr., F. D. (1985). A Technology Acceptance Model for Empirically Testing New End-User Information Systems: Theory and Results. Massachusetts Institute of Technology, MIT. Retrieved from http://dspace.mit.edu/handle/1721.1/15192 De Man, E. W. H. (2007). Are Spatial Data Infrastructures Special? In O. Harlan (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (pp. 33–54). Redlands, CA, USA: ESRI Press. De Vries, W. T., & Lance, K. T. (2011). SDI Reality in Uganda: Coordinating between Redundancy and Efficiency. In Z. Nedović-Budić, J. Crompvoets, & Y. Georgiadou (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 105–119). Boca Raton, FL, USA: CRC Press. Denzin, N. K., & Lincoln, Y. S. (1994). Part II: Major Paradigms and Perspectives. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of Qualitative Research (pp. 99–104). Thousand Oaks, CA, USA: Sage Publications. Dessers, E., Hootegem, G. V., Crompvoets, J., & Hendriks, P. H. J. (2011). Integrating Spatial Information and Business Processes: The Role of Organizational Structures. In Z. Nedović- 513 Budić, J. Crompvoets, & Y. Georgiadou (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 49–67). Boca Raton, FL, USA: CRC Press. Dewey, J. (1929). The Quest for Certainty. New York, New York, USA: G. P. Putnam. Dewey, J. (1938). Logic: The Theory of Inquiry (p. 546). New York, New York, USA: Henry Holt and Co. Dholakia, N., & Kshetri, N. (2003). The Global Digital Divide and Mobile Business Models : Identifying Viable Patterns of e-Development The Global Digital Divide and Mobile Business Models : Identifying Viable Patterns of e-Development. In S. Krishna & S. Madon (Eds.), The Digital Challenge: InformationTechnology in the Development Context (pp. 1– 25). Burlington, VT: Ashgate Publishing Company. Retrieved from http://books.google.com/books?hl=en&lr=&id=GVhMzNSMsLYC&oi=fnd&pg=PA237&o ts=lhNgUYi2zs&sig=HwhJC0NEi_Ltlllnzp5JO4on1Aw#v=onepage&q&f=false Diallo, A., & Thuillier, D. (2004). The success dimensions of international development projects: the perceptions of African project coordinators. International Journal of Project Management, 22(1), 19–31. doi:10.1016/S0263-7863(03)00008-5 Djankov, S., McLiesh, C., & Shleifer, A. (2005). Private Credit in 129 Countries. Cambridge, MA, USA. Retrieved from http://www.nber.org/papers/w11078 Dobson, J. E., & Fisher, P. F. (2003). Geoslavery. IEEE Technology and Society Magazine, 47– 52. Retrieved from http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=1188276 Donner, J. (2009). Mobile-based Livelihood Services in Africa: Pilots and Early Deployments. In M. Fernández-Ardèvol & A. R. Híjar (Eds.), Communication Technologies in Latin America and Africa: A Multidisciplinary Perspective (pp. 37–58). Barcelona: Universitat Oberta de Catalunya. Retrieved from http://www.mobileactive.org/files/file_uploads/Chapter_01_Donner.pdf Donner, J. (2010). Framing M4D: The Utility of Continuity and the Dual Heritage of “Mobile and Development.” The Electronic Journal on Information Systems in Developing Countries - EJISDC, 44(3), 1–16. Retrieved from http://www.ejisdc.org/ojs2/index.php/ejisdc/article/viewFile/746/342 Donner, J., & Tellez, C. A. (2008). Mobile banking and economic development: linking adoption, impact, and use. Asian Journal of Communication, 18(4), 318–332. doi:10.1080/01292980802344190 Donovan, K. (Georgetown U., & Donner, J. (Microsoft R. (2010). A Note on the Availability and Importance of Pre-paid Mobile Data in Africa. Africa. Kampala, Uganda: Mobile 4 Development, M4D. Drichi, P. (2002). National Biomass Study. Population (English Edition) (p. 134). Kampala, Uganda. 514 Dunn, C. E., Atkins, P. J., Blakemore, M. J., & Townsend, J. G. (1999). Teaching Geographical Information Handling Skills for Lower-income Countries. Transactions in GIS, 3(4), 319– 332. doi:10.1111/1467-9671.00025 Dunn, C. E., Atkins, P. J., & Townsend, J. G. (1997). GIS for Development: A Contradiction in Terms? Area, 29(2), 151–159. doi:10.1111/j.1475-4762.1997.tb00017.x EB. (2012). Lake Victoria. Encyclopedia Britanica. Retrieved August 16, 2012, from http://www.britannica.com/EBchecked/topic/627661/Lake-Victoria EMG, AVSI, Agrisystems, C., & Foundation, S. T. (2008). Geographic Assessment: Stability, Peace and Reconciliation in Northern Uganda (SPRING) PRoject (pp. 1–16). Gulu, Uganda. ESRI. (2011). How will Esri make itself more visible to universities and institutions of higher learning in Africa and sub-Saharan countries? ESRI International USer Conference 2011. Retrieved July 1, 2012, from http://events.esri.com/uc/QandA/index.cfm?fuseaction=answer&conferenceId=DD02CFE7- 1422-2418-7F271831F47A7A31&questionId=4076 EarthInstitute. (2012). Millenium Villages. Millenium Villages. Retrieved August 19, 2012, from http://www.earth.columbia.edu/articles/view/1799 Ein-Dor, P., & Segev, E. (1993). A Classification of Information Systems: Analysis and Interpretation. Information Systems Research, 4(2), 166–204. doi:10.1287/isre.4.2.166 Eisenhardt, K. M. (1989). Building Theories from Case. The Academy of Management Review, 14(4), 532–550. Retrieved from http://www.jstor.org/stable/pdfplus/258557.pdf?acceptTC=true Ekman, B. (2004). Community-based health insurance in low-income countries: a systematic review of the evidence. Health Policy and Planning, 19(5), 249–270. doi:10.1093/heapol/czh031 Elmes, G. A., Epstein, E. F., McMaster, R. B., Niemann, B. J., Poore, B., Sheppard, E., & Tulloch, D. L. (2005). GIS and Society: Interrelation , Integration , and Transformation. In R. B. McMaster & L. E. Usery (Eds.), . Boca Raton, FL, USA: CRC Press. Elwood, S. (2006a). Beyond Cooptation or Resistance: Urban Spatial Politics, Community Organizations, and GIS-Based Spatial Narratives. Annals of the Association of American Geographers, 96(2), 323–341. Elwood, S. (2006b). Critical Issues in Participatory GIS: Deconstructions, Reconstructions, and New Research Directions. Transactions in GIS, 10(5), 693–708. doi:10.1111/j.1467- 9671.2006.01023.x 515 Elwood, S. A. (2002). GIS use in Community Planning: A Multidimensional Analysis of Empowerment. Environment and Planning A, 34(5), 905–922. doi:10.1068/a34117 Ernest, U. (2010). Analyzing fee and free spatial data sharing policies in public sector in Uganda. Geo-Information Science. International Institute for Geo-information Sciences and Earth Observation,ITC, Enschede, The Netherlands. Espeter, M., & Raubal, M. (2009). Location-based decision support for user groups. Journal of Location Based Services, 3(3), 165–187. doi:10.1080/17489720903339668 Evans, L. (2011). Location-based services: transformation of the experience of space. Journal of Location Based Services, 5(3-4), 242–260. Ewing, J. (2007, September). Exteme Telecom. BusinessWeek, 2. Retrieved from http://www.kiwanja.net/database/article/article_extreme_telecom.pdf Ezigbalike, C., Selebalo, Q. C., Faïz, S., & Zhou, S. Z. (2000). Spatial Data Infrastructures: Is Africa Ready? Fourth Spatial Data Infrastructure Conference, Cape Town, South Africa, March 13-15, 2000 (pp. 1–9). Capetown, South Africa. Fadahunsi, J. T. (2010). A Perspective View on the Development and Applications of Geographical Information System ( GIS ) in Nigeria . The Pacific Journal of Science and Technology, 11(1), 301–308. Farlex. (2012). The Free Dictionary. The Free Dictionary. Retrieved February 15, 2012, from http://www.thefreedictionary.com Foresman, T. W. (1998). GIS Early Years and the Threads of Evolution. In T. W. Foresman (Ed.), The History of Geographic Information Systems: Perspectives from the Pioneers (pp. 3–17). Upper Saddle River, NJ, USA: Prentice-Hall, Inc. ForisTelecom. (2012). Foris Telecom: The Fore Group. Foris Telecom. Retrieved August 13, 2012, from http://www.foristel.com/home/artdetails.aspx?mCatID=61016&artID=7460 Foth, M., Bajracharya, B., Brown, R., & Hearn, G. (2009). The Second Life of Urban Planning? Using NeoGeography Tools for Community Engagement. Journal of Location Based Services, 3(2), 97–117. doi:10.1080/17489720903150016 FrontlineSMS. (2012). FrontlineSMS: About us. FrontlineSMS. Retrieved August 14, 2012, from http://www.frontlinesms.com/about-us/ Fuller, R. M., Groom, G. B., Mugisha, S., Ipulet, P., Pomeroy, D., Katende, A., Bailey, R., et al. (1998). The Integration of Field Survey and Remote Sensing for Biodiversity Assessment: A Case Study In The Tropical Forests and Wetlands of Sango Bay, Uganda. Biological Conservation, 86(3), 379–391. doi:10.1016/S0006-3207(98)00005-6 516 Furuholt, B., & Matotay, E. (2011). The Developmental Contribution from Mobile Phones Across the Agricultural Value Chain in Rural Africa. The Electronic Journal of Information Systems in Developing Countries - EJISDC, 48(7), 1–16. Retrieved from http://www.ejisdc.org/ojs2/index.php/ejisdc/article/viewfile/849/377 GOU The Local Governments (Amendment) Act, 2005 (2005). Government of Uganda. GOU. (2012). Republic of Uganda: Budget Speech - Financial Year 2012/13. Kampala, Uganda: Government of Uganda. Retrieved from http://www.finance.go.ug/ GSDI. (2011, August). Uganda Lauches Centralized Data Centre. Spatial Data Infrastructure - Africa Newsletter, 10(8), 6. GSM-Arena. (2012). T-Mobile G1. T-Mobile G1. Retrieved August 20, 2012, from http://www.gsmarena.com/t_mobile_g1-2533.php Georgiadou, Y., Puri, S. K., & Sahay, S. (2005). Towards a Potential Research Agenda to Guide the Implementation of Spatial Data Infrastructures — A case study from India. International Journal of Geographical Information Science, 19(10), 1113–1130. doi:10.1080/13658810500286950 Georgiadou, Yola, Bana, B., Becht, R., Hoppe, R., Ikingura, J., Kraak, M., & Lance, K. (2011). Sensors , empowerment , and accountability : a Digital Earth view from East Africa. International Journal of Digital Earth, 4(4), 285–304. Georgiadou, Yola, Budhathoki, N. R., & Nedović-Budić, Z. (2011). An Exploration of SDI and Volunteered Geographic Information in Africa. In Z. Nedović-Budić, J. Crompvoets, & Y. Georgiadou (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 203–218). Boca Raton, FL, USA: CRC Press. Georgiadou, Yola, Harvey, F., & Miscione, G. (2007). A Bigger Picture: Information Systems and Spatial Data Infrastructure Research Perspectives. In M. Wachowicz & L. Bodum (Eds.), AGILE 2007 : proceedings of the 10th AGILE international conference on geographic information science : the European Information Society : leading the way with geoinformation, Aalborg, Denmark, 8-11 May 2007 (p. 6). Aalborg, Denmark. Geoserver. (2012). Welcome. Geoserver: Welcome. Retrieved August 21, 2012, from http://geoserver.org/display/GEOS/Welcome Getis, A., Anselin, L., Lea, A., Ferguson, M., & Miller, H. (2005). Spatial Analysis: Modelling in a GIS Environment. In R. B. McMaster & L. E. Usery (Eds.), A Research Agenda for Geographic Information Science (Vol. 164, pp. 157–196). Boca Raton, FL, USA: CRC Press. Ghose, R. (2011). Politics and Power in Participation and GIS Use for Community Decision Making. In T. L. Nyerges, H. Couclelis, & R. McMaster (Eds.), The SAGE Handbook of GIS and Society (pp. 423–438). London, UK: SAGE Publications Ltd. 517 Gibson, R. (1998). Informatics Diffusion in South American Developing Economies. Management, 6(2), 35–42. Giddings, L. S. (2006). Mixed-methods research: Positivism dressed in drag? Journal of Research in Nursing, 11(3), 195–203. doi:10.1177/1744987106064635 Gilfoyle, I., & Thorpe, P. (2004a). The Development of GIS in Local Government. Geographic Information Management in Local Government (p. 15). Boca Raton, FL, USA. Gilfoyle, I., & Thorpe, P. (2004b). Case Study — London Borough of Harrow. Geographic Information Management in Local Government (p. 7). Boca Raton, FL, USA: CRC Press. Glaser, B. G., & Strauss, A. L. (1967). The Discovery of Grounded Theory: Strategies for Qualitative Research. London, UK: Wiedenfeld and Nicholson. Retrieved from http://books.google.com/books?hl=en&lr=&id=rtiNK68Xt08C&oi=fnd&pg=PA1&dq=Glas er+and+Strauss+1967&ots=UUxY_jZKXK&sig=eMaOJDzBXJRikZcOMnCcixBx4QU#v= onepage&q=Glaser and Strauss 1967&f=false GoU. (1995). Constitution of the Republic of Uganda, 1995. Kampala, Uganda: Government of Uganda. GoU. (2006). The National Integrated Monitoring and Evaluation Strategy Framework (NIMES): Final Document. Components (p. 57). Kampala, Uganda. GoU. (2008). Land Act 1998 (Ch 227). Kampal, Uganda: Government of Uganda. GoU. (2010). National Development Plan (2010/11-2014/15). Development (p. 417). Kampala, Uganda. GoU, & UNFPA. (2010). The State of Uganda Population Report 2010. Population (English Edition) (p. 111). Kampala, Uganda. Retrieved from http://www.popsec.org/publications_3_1868272171.pdf Goodchild, M. F. (1987). A Spatial Analytical Perspective on Geographical Information Systems. International Journal of Geographical Information Systems, 1(4), 327–334. doi:10.1080/02693798708927820 Goodchild, M. F. (1988). Stepping Over The Line: Technological Constraints And the New Cartography. The American Cartographer, 15(3), 311–319. doi:10.1559/152304088783886973 Goodchild, M. F. (1992). Geographical Information Science. International Journal of Geographical Information Systems, 6(1), 31–45. Goodchild, M. F. (1995). Sharing Imperfect Data. In H. J. Onsrud & G. Rushton (Eds.), Sharing Geographic Information (pp. 413–425). New Brunswick, New Jersey. 518 Goodchild, M. F. (2008). Geographic Information Science: The Grand Challenges. In J. P. Wilson & S. A. Fotheringham (Eds.), The Handbook of Geographic Information Science (pp. 596–608). Malden, MA, USA: Blackwell Publishing. Goodchild, M. F. (2009). NeoGeography and the Nature of Geographic Expertise. Journal of Location Based Services, 3(2), 82–96. doi:10.1080/17489720902950374 Goodchild, M. F., Johnson, D. M., Maguire, D. J., & Noronha, V. T. (2005). Distributed and Mobile Computing. In R. B. McMaster & L. E. Usery (Eds.), A Research Agenda for Geographic Information Science (pp. 257–286). Boca Raton, FL, USA. Goodchild, M. F., Mark, D. M., & Sheppard, E. (1999). Introduction to the Varenius Project. International Journal of Geographical Information Science, 13(8), 731–745. Google. (2012a). Google Latitude. Google Latitude. Retrieved August 14, 2012, from https://www.google.com/latitude Google. (2012b). Google SMS Applications. Google SMS Applications. Retrieved August 16, 2012, from http://www.google.co.ug/intl/en_ug/mobile/sms/ Gow, D. D. (1988). The Notorious Nine: Critical Implementation Problems in Project Implementation. World Development, 16(12), 1399–1418. Gowa, E. K. (2009). Best Practices in Environmental Information Management in Africa. (T. E. Muramira, C. Sebukeera, & F. Turyatunga, Eds.)Network (p. 30). Birkeland Trykkeri, Norway: UNEP/GRID-Arendal. GrameenFoundation. (2010). Community Knowledge Worker Pilot Report. Knowledge Creation Diffusion Utilization (pp. 1–113). Kampala, Uganda. GrameenFoundation. (2012). Uganda. Uganda. Retrieved May 15, 2012, from http://www.grameenfoundation.org/sub-saharan-africa/uganda Gray, J., & Peters, C. (1960). Anglo-German Relations in Uganda , 1890-1892. The Journal of African History, 1(2), 281–297. Grus, L., Crompvoets, J., & Bregt, A. K. (2008). Spatial Data Infrastructures as Complex Adaptive Systems. International Journal of Geographical Information Science, 24(3), 439– 463. Guba, E. G., & Lincoln, Y. S. (1994). Competing Paradigms in Qualitative Research. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of Qualitative Research (pp. 105–117). Thousand Oaks, CA, USA: Sage Publications. Guptill, S. C., & Eldridge, D. F. (1998). Spatial Data Policy and Pricing in the United States. In F. D. R. Taylor (Ed.), Policy Issues in Modern Cartography (pp. 19–28). Oxford, UK: ELSEVIER SCIENCE Ltd. 519 Gyamfi-Aidoo, J., Schwabe, C., & Govender, S. (2005). Determination of the Fundamental Geospatial Data Sets for Africa Through a User Needs Analysis - A Synthesis Report. Africa (pp. 40–94). Pretoria, South Africa. Retrieved from http://www.eis-africa.org/EIS- Africa/Chapter 2 - Synthesis report (English).pdf HPDMH, DWD, MWE, UBOS, ILRI, & WRI. (2009). Mapping a Healthier Future: How Spatial Analysis Can Guide Pro-Poor Water and Sanitization Planning in Uganda. (K. Bennett, H. Billings, P. Ghazi, & G. Mock, Eds.)Statistics (p. 48). Nairobi, Kenya: The Regal Press Kenya Ltd. Hall, B. G. (1999). Guest Editorial: GIS Education and Infrastructure - Challenges and Problems in Emerging Countries. Transactions in GIS, 3(4), 311–317. Hallward, P. (2011). Egypt’s Popular Revolution will Change the World. Guardian UK. Retrieved September 16, 2011, from http://wwwhttp//www.guardian.co.uk/commentisfree/2011/feb/09/egypt-north-africa- revolution.guardian.co.uk/commentisfree/2011/feb/09/egypt-no... Hannes. (2011). Mobile Banking. Mobile Banking. Retrieved May 15, 2012, from http://mbanking.blogspot.com/2010/11/implications-of-using-mobile-payments.html Harjo, L. (2006). GIS Support for Empowering Marginalized Communities: The Cherokee Nation Case Study. In M. Campagna (Ed.), GIS for Sustainable Development (pp. 433–450). Boca Raton, FL, USA: CRC Press. Harris, T. M., & Weiner, D. (1996). GIS and Society: The Social Implications of How People, Space and Environment are Represented in GIS. Science (p. 193). Koinonia Retreat Center, South Haven, Minnesota. Retrieved from http://www.ncgia.ucsb.edu/Publications/Tech_Reports/96/96-7.PDF Harris, T. M., Weiner, D., Warner, T., & Levin, R. (1995). Pursuing Social Goals Through Participatory GIS: Redressing South Africa’s Historical Political Ecology. In J. Pickles (Ed.), Ground Truth: The Social Implications of Geographic Information Systems (pp. 196– 222). New York, NY, USA: The Guilford Press. Harvey, F. (2001). Constructing GIS: Actor Networks of Collaboration. Journal of the Urban and Regional Information Systems Association (URISA), 13(1), 29–37. Harvey, F. (2011). Spatial Data Infrastructure for Cadastres: Foundations and Challenges. In T. L. Nyerges, H. Couclelis, & R. McMaster (Eds.), The SAGE Handbook of GIS and Society (pp. 287–303). London, UK: SAGE Publications Ltd. Harvey, F., & Chrisman, N. R. (2004). The Imbrication of Geography and Technology: The Social Construction of Geographic Information Systems. In S. D. Brunn, S. L. Cutter, & J. W. Harrington Jr. (Eds.), Geography and Technology (pp. 65–80). Dordrecht, The Netherlands: Kluwer Academic Publishers. 520 Harvey, F., & Tulloch, D. L. (2004). How do Local Governments Share and Coordinate Geographic Information? Issues in The United States. 10th EC GI & GIS Workshop, ESDI State of the Art, Warsaw, Poland, 23-25 June 2004 (pp. 23–25). Warsaw, Poland. Harvey, F., & Tulloch, D. L. (2006). Local Government Data Sharing: Evaluating the Foundations of Spatial Data Infrastructures. International Journal of Geographical Information Science, 20(7), 743–768. doi:10.1080/13658810600661607 Hayes, D. (2011). The Arab Spring: Protest, Power, Prospect. Open Democracy. Retrieved September 16, 2011, from http://www.opendemocracy.net Hazeltine, B., & Bull, C. (1999). Appropriate Technology: Tools, Choices and Implications (p. 364). San Diego, CA, USA: Academic Press. Hellstrom, J. (2010). The Innovative Use of Mobile Applications in East Africa. (P.-E. Troeften, Ed.)Africa (pp. 1–104). Stockholm, Sweden: Edita; Swedish International Development Cooperation Agency (SIDA). Heywood, I. D., & Petch, J. R. (1991). GIS Education: A Business Perspective. Cartographica, 28(3 (GIS Education and Training)), 10–22. Hirabe, N. O. (2009). Globalization and the Challenge of Urban Development in Uganda: Implications on Land Use Planning in Kampala. The University of British Columbia, Vancouver, Canada. Hofmann, B. (2001). The Determinants of Private Sector Credit in Industrialised Countries: do Property Prices Matter? Frankfurt, Germany. doi:http://dx.doi.org/10.2139/ssrn.847404 Hofweber, T. (2011). Logic and Ontology. Stanford Encyclopedia of Philosophy. Retrieved December 25, 2011, from http://plato.stanford.edu/archives/fall2011/entries/logic-ontology/ Howe, K. R. (1992). Getting Over the Quantitative- Qualitative Debate. American Journal of Education, 100(2), 236–256. Retrieved from http://www.jstor.org/stable/10.2307/1085569 Hudson-Smith, A., Crooks, A., Gibin, M., Milton, R., & Batty, M. (2009). NeoGeography and Web 2.0: Concepts, Tools and Applications. Journal of Location Based Services, 3(2), 118– 145. doi:10.1080/17489720902950366 Huxhold, W. E. (1993). The Application of Research and Development from the Information Systems Field to GIS Implementation in Local Government: Some Theories on Successful Adoption and USe of GIS Technology. In I. Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (pp. 59–60). Dordrecht, The Netherlands: Kluwer Academic Publishers. ICITD, MUBS, & PC-TECH. (2010). Mobile Phones in Health care in Sub-Saharan Africa – The Case of AppLab Uganda (pp. 1–6). Kampala, Uganda. Retrieved from 521 http://www.pctechmagazine.com/index.php?option=com_docman&task=cat_view&gid=49 &Itemid=151 ICTRegulationsToolkit. (2006). Uganda Communications Act 1997. Uganda Communications Act 1997. Retrieved August 14, 2012, from http://www.ictregulationtoolkit.org/en/Publication.1431.html ITU. (2012). Mobile-Cellular Telephone Subscriptions. Geneva, Switzerland: International Telecommunication Union (ITU). Retrieved from http://www.itu.int/ITU- D/ict/statistics/material/excel/Mobile-cellular2000-2011.xls Infocom. (2012). Infocom: Products and Services. Infocom: Products and Services. Retrieved August 13, 2012, from http://www.infocom.co.ug/product-services Investopedia. (2012). Private Sector. Dictionary. Retrieved April 27, 2012, from http://www.investopedia.com/terms/p/private-sector.asp#axzz1tHvQb2RS InvestorWords. (2012). Public Sector. Glossary. Retrieved April 27, 2012, from http://www.investorwords.com/3947/public_sector.html Isomursu, M., Ervasti, M., Kinnula, M., & Isomursu, P. (2011). Understanding Human Values in Adopting New Technology—A Case Study and Methodological Discussion. International Journal of Human-Computer Studies, 69(4), 183–200. doi:10.1016/j.ijhcs.2010.12.001 Jackson, M., Schell, D., & Taylor, F. D. R. (2009, April). The Evolution of Geospatial Technology: Calls for Changes in Geospatial Research , Education and Government Management. Directions Magazine, 13. Jacoby, S., Smith, J., Ting, L., & Williamson, I. (2002). Developing a Common Spatial Data Infrastructure Between State and Local Government — an Australian Case Study. International Journal of Geographical Information Science, 16(4), 305–323. doi:10.1080/13658810110096001 Jagoe, A. (2003). Mobile Location Services: The Definitive Guide (p. 444). Upper Saddle River, NJ, USA: Pearson Education Inc. Retrieved from http://books.google.com/books?id=8VX5_RH8oc8C&printsec=frontcover&source=gbs_ge_ summary_r&cad=0#v=onepage&q&f=false James, J., & Versteeg, M. (2007). Mobile phones in Africa: how much do we really know? Social indicators research, 84(1), 117–126. doi:10.1007/s11205-006-9079-x Janelle, D. G., & Goodchild, M. F. (2011). Concepts, Principles, Tools, and Challenges in Spatially Integrated Social Science. In T. L. Nyerges, H. Couclelis, & R. McMaster (Eds.), The SAGE Handbook of GIS and Society (pp. 27–45). London, UK: SAGE Publications Ltd. Janssen, K., Crompvoets, J., & Dumortier, J. (2011). When is Providing Spatial Information a Public Task? A Search for Criteria. In Z. Nedović-Budić, J. Crompvoets, & Y. Georgiadou 522 (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 3–20). Boca Raton, FL, USA: CRC Press. Janssen, K., & Dumortier, J. (2007). Legal Framework for a European Union Spatial Data Infrastructure: Uncrossing the Wires. In H. Onsrud (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (pp. 231–244). Redlands, CA, USA: ESRI Press. Johnson, R Burke, & Onwuegbuzie, A. J. (2004). Mixed Methods Research : A Research Paradigm Whose Time Has Come. Educational Researcher, 33(7), 14–26. Johnson, R. B., Onwuegbuzie, a. J., & Turner, L. a. (2007). Toward a Definition of Mixed Methods Research. Journal of Mixed Methods Research, 1(2), 112–133. doi:10.1177/1558689806298224 Junius, H., Tabeling, M., & Wegener, M. (1996). Germany: a Federal Approach to Land Information Management. In I. Masser, H. J. Campbell, & M. Craglia (Eds.), GIS diffusion: the Adoption and Use of Geographical Information Systems in Local Government in Europe (pp. 67–86). London, UK: Taylor & Francis. Kaijuka, E. (2007). GIS and Rural Electricity Planning in Uganda. Journal of Cleaner Production, 15(2), 203–217. doi:10.1016/j.jclepro.2005.11.057 Kalande, W., & Ondulo, K. J. D. (2006). Geoinformation Policy in East Africa. Shaping the Change, XXIII FIG Congress, October 8-13, 2006 (pp. 1–26). Munich, Germany. Karatunga, A. M. (2002). Spatial Data Infrastructure in Uganda. Addis Ababa, Ethiopia. Karatunga, A. M. (2003). GIS Data for the Development of Karamoja. Kampala, Uganda. Karikari, I., & Stillwell, J. (2005). Applying Cost/Benefit Analysis to Evaluate Investment in GIS: The Case of Ghana’s Lands Commission Secretariat, Accra. Transactions in GIS, 9(4), 489–505. doi:10.1111/j.1467-9671.2005.00231.x Karikari, I., Stillwell, J., & Carver, S. (2003). Land Administration and GIS: the Case of Ghana. Progress in Development Studies, 3(3), 223–242. doi:10.1191/1464993403ps050ra Karikari, I., Stillwell, J., & Carver, S. (2005). The Application of GIS in the Lands Sector of a Developing Country: Challenges Facing Land Administrators in Ghana. International Journal of Geographical Information Science, 19(3), 343–362. doi:10.1080/13658810412331280149 Karimi, H. (2007). Editorial. Journal of Location Based Services, 1(2), 87–88. doi:10.1080/17489720701874494 Katende, J. (2012). Fleet management saves companies undue stress. The New Vision. Kampala, Uganda. Retrieved from http://www.enteruganda.com/brochures/cartrack0609_c.html 523 Kemp, K. K., & Goodchild, M. F. (1991). Developing a Curriculum in GIS: The NCGIA Core Curriculum Project. Cartographica, 28(3 (GIS Education and Training)), 39–54. Kennedy, E. (1998). A Private Sector Perspective. In F. D. R. Taylor (Ed.), Policy Issues in Modern Cartography (pp. 125–140). Oxford, UK: ELSEVIER SCIENCE Ltd. Kerski, J. J. (2008). Geographic Information Systems in Education. In J. P. Wilson & S. A. Fotheringham (Eds.), The Handbook of Geographic Information Science (pp. 540–556). Malden, MA, USA: Blackwell Publishing. Khan, M. A. (2010, January 1). Information Systems. Pakistan Library Automation Group. Retrieved November 26, 2011, from http://www.paklag.org/resources.htm Kiib, H. (1996). Denmark: Local Autonomy, Register Based Information Systems and GIS. In I. Masser, H. J. Campbell, & M. Craglia (Eds.), GIS diffusion: the Adoption and Use of Geographical Information Systems in Local Government in Europe (pp. 125–143). London, UK: Taylor & Francis. Kincheloe, J. L., & McLaren, P. L. (1994). Rethinking Critical Theory and Qualitative Research. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of Qualitative Research (pp. 138–157). Thousand Oaks, CA, USA: Sage Publications. Kisambira, E. (2008). Uganda University Opens Software Incubation Center. PC World. Retrieved August 20, 2012, from http://www.pcworld.com/businesscenter/article/144815/uganda_university_opens_software _incubation_center.html Kisembo, S. W. (2006). Handbook on Decentralization in Uganda (p. 81). Kampala, Uganda: Fountain Publishers Ltd. Kitutu, K. M. G. (Research O. N. (2003). The Role of GIS in Environmental Management. Kampala, Uganda. Kiyaga, A. (2012, July 14). 43% of Ugandans Use Mobile Money Transfers. The Monitor Daily Newspaper, pp. 9–10. Kampala, Uganda. Retrieved from http://www.monitor.co.ug/News/National/43++of+Ugandans+use+mobile+money+transfers /-/688334/1453496/-/view/printVersion/-/9trru4z/-/index.html Kizza, J. M., Lynch, K., Nath, R., & Aisbeth, J. (2010). Strengthening the Role of ICT in Development. In J. M. Kizza, K. Lynch, R. Nath, & J. Aisbeth (Eds.), 6th Annual International Conference on Computing and ICT Research, Kampala, Uganda, 01-04 August, 2010 (VI., Vol. VI). Kampala: Fountain Publishers Ltd. Retrieved from http://cit.mak.ac.ug/iccir/downloads/ICCIR10.pdf Koerten, H., & Veenswijk, M. (2011). Thinking in Circles: How National Geo-Information Infrastructures Cannot Escape from the Temptation of Technology. In Z. Nedović-Budić, J. 524 Crompvoets, & Y. Georgiadou (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 137–153). Boca Raton, FL, USA. Koti, F. T. (2004). Same Place Yet Different Worlds: A GIS and Society Perspective on Kenyan Peri-Urbanization. West Virginia University. Kruse, D. (1997). Innovation in Financial Markets: The Experience of Japan from the Perspective of a Private Sector Financial Institution. In K. B. Staking (Ed.), Policy-based Finance and Market Alternatives: East Asian Lessons for Latin America and the Carribean. Washington D.C., USA: Inter-American Development Bank. Retrieved from http://books.google.com/books?hl=en&lr=&id=ukd2uEOxyNEC&oi=fnd&pg=PA17&dq= Credit+supporting+private+sector+&ots=fd4arAQJhD&sig=uprb7trQEs1dm6pzNd33V1mu 8TM#v=onepage&q=Credit supporting private sector&f=false Kuhn, T. (1962). The Structure of Scientific Revolutions (1st ed.). Chicago, IL: University of Chicago Press. Retrieved from http://www.aoni.waseda.jp/sidoli/Kuhn_Structure_of_Scientific_Revolutions.pdf Kuhn, T. (1977). The Essential Tension: Selected Studies in Scientific Tradition and Change. Chicago, IL: University of Chicago Press. Retrieved from http://www.amazon.com/Essential-Tension-Selected-Scientific- Tradition/dp/0226458067/ref=sr_1_1?ie=UTF8&qid=1324859012&sr=8- 1#reader_0226458067 Kyem, P. A. K. (2000). Embedding GIS Applications into Resource Management and Planning Activities of Local and Indigenous Communities: A Desirable Innovation or a Destabilizing Enterprise? Journal of Planning Education and Research, 20(2), 176–186. doi:10.1177/0739456X0002000204 Kyem, P. A. K. (2004a). Of Intractable Conflicts and Participatory GIS Applications: The Search for Consensus Amidst Competing Claims and Institutional Demands. Annals of the Association of American Geographers, 94(1), 37–57. Kyem, P. A. K. (2004b). Power, Participation, and Inflexible Institutions: An Examination of the Challenges to Community Emprowerment in Participatory GIS Applications. Cartographica, 38(3 & 4), 5–17. Kyem, P. A. K., & Saku, J. C. (2009). Web-Based GIS and the Future of Participatory GIS Applications within Local and Indigenous Communities. The Electronic Journal on Information Systems in Developing Countries, 38(7), 1–16. Laituri, M. (2011). Indigenous People’s Issues and Indigenous Uses of GIS. In T. L. Nyerges, H. Couclelis, & R. McMaster (Eds.), The SAGE Handbook of GIS and Society (pp. 202–221). London, UK. 525 Lance, K., & Bassole, A. (2006). SDI and National Information and Communication Infrastructure ( NICI ) Integration in Africa. Information Technology for Development, 12(4), 333–338. doi:10.1002/itdj Latour, B. (1987). Science in Action (p. 274). Cambridge, MA, USA: Harvard University Press. Laudon, K. C., & Laudon, J. P. (1988). Management Information Systems. Lauriault, T. P., & Taylor, F. D. R. (2007a). Geospatial Data Infrastructure for Sustainable Development of East Timor. In H. Onsrud (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (pp. 175–200). Redlands, CA, USA: ESRI Press. Lauriault, T. P., & Taylor, F. D. R. (2007b). Geospatial Data Infrastructure for Sustainable Development of East Timor. In H. J. Onsrud (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (Vol. 5, pp. 175–199). Redlands, CA, USA: ESRI Press. Lawal, G. (2007). Corruption and Development in Africa: Challenges for Political and Economic Change. Humanity and Social Sciences Journal, 2(1), 1–7. Leitner, H., McMaster, R., Elwood, S., Mcmaster, S., & Sheppard, E. (1998). Models for Making GIS Available to Community Organizations: Dimensions of Difference and Appropriateness. Paper presented to the NCGIA specialist meeting on Empowerment, Marginalization and GIS, Santa Barbara CA, October 1998 (pp. 1–24). Santa Barbara, CA, USA: NCGIA. Letsie, M. (2008). Spatial Data sharing for Sustainable Development in Landlocked Countries : The Case of Lesotho. GSDI 10 - 10th International Conference for Spatial Data Infrastructure, February 25-29, St. Augustine, Trinidad (pp. 1–10). St. Augustine, Trinidad. Liang, T., & Chen, D. (2003). Evolution of Information Systems Research. 7th Pacific Asia Conference on Information Systems (PACIS), 10-13 July 200, Adelaide, South Australia (pp. 834–842). Adelaide, Australia. Lo, C. P., & Yeung, A. K. W. (2007). Concepts and Techniques in Geographic Information Systems (p. 532). Upper Saddle River, NJ, USA: Pearson Education Inc. Loh, P., Marshall, C., & Meadows, C. J. (1998). High-Tech/Low-Tech: Appropriate Technologies for Developing Nations. Journal of Global Information Management, 6(2), 5– 12. Longley, P. A., Goodchild, M. F., Maguire, D. J., & Rhind, D. W. (2010). Geographic Information Systems and Science (p. 560). John Wiley and Sons. Retrieved from http://books.google.com/books?id=zyFR8uxHM9oC&pgis=1 526 Ludford, P. J., Frankowski, D., Reily, K., Wilms, K., & Terveen, L. (2006). Because I Carry My Cell Phone Anyway: Functional Location-Based Reminder Applications. CHI 2006, April 22–27, 2006, Montréal, Québec, Canada (pp. 1–10). Monteal, Canada: ACM. Luszczynska, A., & Schwarzer, R. (1996). Social Cognitive Theory. In M. Conner & P. Norman (Eds.), Predicting Health Behaviour (2nd ed., Vol. 24, pp. 127–169). Buckingham, UK: Open university press. doi:10.1016/S0925-7535(97)81483-X Lwanga-Lunyiigo, S. (1987). The Colonial Roots of Internal Conflict in Uganda. International Seminar on Internal Conflict. London, UK: Institute of Social Research, Makerere University, Kampala. Retrieved from http://opendocs.ids.ac.uk/opendocs/bitstream/handle/123456789/1412/ISIC 17-The colonial roots of internal conflict in Uganda - 331408.pdf?sequence=1 Lwasa, S. (2006). The Role of GIS in Monitoring and Evaluation of Poverty and Performance of Poverty Eradication and Action Plan (PEAP) Policy in Uganda. GSDI 9 - 9th Global Spatial Data Infrastructure Conference, 15-19 June, 2009, Santiago, Chile. Santiago, Chile. Retrieved from http://www.gsdidocs.org/gsdiconf/GSDI-9/abstracts/TS16.4abstract.pdf Lwasa, S., Bamutaze, Y., Karatunga, A., Muhwezi, B. J., Mukwaya, P., Sjaak, D., & Picton, G. (2006). National Spatial Data Infrastructure Workshop. National Spatial Data Infrastructure Workshop, 7th June 2006, Faculty of Computing and Information Technology, Makerere University, Kampala, Uganda (pp. 1–29). Kampala, Uganda. Lwasa, S., Nasirumbi, S., Amadra, O.-O., Muhwezi, B. J., & Dielmann, S. (2005). Development of the Spatial Data Infrastructure in Uganda. Kampala, Uganda. M-HealthUganda. (2012). Mobile Health Projects. Mobile Health Projects. Retrieved August 19, 2012, from http://mobilehealthuganda.wikispaces.com/Mobile+Health+Projects MLHUD. (2011a). Land Sector Strategic Plan 2001-2011: Utilizing Uganda’s Land Resources for Sustainable Development. Sustainable Development. Kampala, Uganda: Ministry of Lands, Housing and Urban Development, Uganda. Retrieved from http://www.mlhud.go.ug/index.php?option=com_docman&task=cat_view&gid=36&trm=34 5&Itemid=60&limitstart=10 MLHUD. (2011b). The Uganda National Land Policy. Kampala, Uganda: MLHUD. Retrieved from http://www.mlhud.go.ug/index.php?option=com_docman&task=cat_view&gid=36&trm=34 5&Itemid=60 MSE. (2009). Institute of Survey and Land Management (Formerly Survey Training School - STS). Kampala: Ministry of Sports and Education. Retrieved from http://www.education.go.ug/Uganda_Rwanda_Exhibition/MINISTRY OF EDUCATION AND SPORTS (ENTEBBE UGANDA).pdf 527 MTN. (2012a). MTN Broadband. MTN: Broadband. Retrieved August 13, 2012, from http://www.mtn.co.ug/MTN-Internet/MTN-Broadband.aspx MTN. (2012b). MTN National Coverage. MTN: National Coverage. Retrieved August 13, 2012, from http://www.mtn.co.ug/Coverage/MTN-Coverage.aspx MTN. (2012c). MTN MobileMoney. MTN MobileMoney. Retrieved August 15, 2012, from http://mtn.co.ug/MTN-Services/Mobile-Banking/MTN-MobileMoney.aspx MTN. (2012d). Content Providers. MTN: Content Providers. Retrieved August 23, 2012, from http://mtn.co.ug/Partners/Content-Providers.aspx MUK-FCIT. (2009). Makerere University Faculty of Computing and Informatics Technology (CIT): Annual Report 2009 (p. 18). Kampala, Uganda. Retrieved from http://cit.mak.ac.ug/annual-report/148-annual-report-2009.html MUK-FCIT. (2010). Mobile Based ATM Fault Detection. 6th Annual International Conference on Computing and ICT Research, Kampala, Uganda, 01-04 August, 2010 (Vol. 4). Kampala, Uganda. Retrieved from http://cit.mak.ac.ug/iccir/downloads/Conference Programme final_revised.pdf MWE. (2012). About Ministry of Water and Environment. Structures of the MWE. Retrieved May 21, 2012, from http://www.mwe.go.ug/index.php?option=com_content&view=article&id=102&Itemid=145 Makanga, P., & Smit, J. (2010). A Review of the Status of Spatial Data Infrastructure Implementation in Africa. Free and Opensource software for Geospatial (Foss4g) Conference, 2010, Cape Town, South Africa. (Vol. 45, pp. 357–365). Capetown, South Africa: FOSS4G. Makumbi, H. (2010). Investigating the Influence of Resource Dependencies on Compliance to National Policies of Geo-Information: A Resource Dependence Perspective . A Case Study Of Geo-Information Sector In Uganda. Geo-Information Science. International Institute for Geo-Information Science and Earth Observation, Enschede, The Netherlands. Manson, S. M., Kne, L., Dyke, K., Shannon, J., & Eria, S. (2012). Using Eye Tracking and Mouse Metrics to Test Usability of Web Mapping Navigation. Cartography and Geographic Information Science, 39(1), 48–60. Mark, D. M., Usery, L. E., & Mcmaster, R. B. (2005). Postscript on the UCGIS and Research. In R. B. McMaster & L. E. Usery (Eds.), A Research Agenda for Geographic Information Science (pp. 389–392). Boca Raton, FL, USA: CRC Press. Martin, B., & Abbott, E. (2008). Development Calling: The Use of Mobile Phones in Agriculture Development in Uganda. Makerere University Kampala Publications. Retrieved August 14, 2012, from http://mak.ac.ug/documents/IFIP/DevelopmentCalling.pdf 528 Marx, L., & Smith, M. R. (2011). Introduction. In M. R. Smith & L. Marx (Eds.), Does Technology Drive History? (p. ix–xv). Cambridge, MA, USA: MIT Press. Retrieved from http://www.open.ac.uk/blogs/sirg/wp-content/uploads/2011/05/1-Marx+Smith- DoesTechnologyDriveHistory-Intro2.pdf Masser, I. (1993). Diffusion of GIS in British Local Government. In I. Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (pp. 99–115). Dordrecht, The Netherlands: Kluwer Academic Publishers. Masser, I. (2005). GIS Worlds: Creating Spatial Data Infrastructures (pp. 1–312). Redlands, CA, USA: ESRI Press. Masser, I. (2007). Building European Spatial Data Infrastructures (1st ed., pp. 1–91). Redlands, CA, USA: ESRI Press. Masser, I. (2010). Building European Spatial Data Infrastructures (2nd ed., p. 95). Redlands, CA, USA: ESRI Press. Masser, I. (2011). Emerging Frameworks in the Information Age: The Spatial Data Infrastructure (SDI) Phenomenon. In T. L. Nyerges, H. Couclelis, & R. McMaster (Eds.), The SAGE Handbook of GIS and Society (pp. 271–286). London, UK: SAGE Publications Ltd. Masser, I., Borrero, S., & Holland, P. (2003). Regional SDIs. In I. Williamson, A. Rajabifard, & M.-E. F. Feeney (Eds.), Developing Spatial Data Infrastructures: From Concept to Reality (pp. 59–78). Boca Raton, FL, USA: CRC Press. Masser, I., & Campbell, H. J. (1995). Information Sharing: The Effects of GIS on British Local Government. In H. J. Onsrud & G. Rushton (Eds.), Sharing Geographic Information (pp. 230–249). New Brunswick, New Jersey. Masser, I., & Campbell, H. J. (1996). Great Britain: The Dynamics of GIS Diffusion. In I. Masser, H. J. Campbell, & M. Craglia (Eds.), GIS diffusion: the Adoption and Use of Geographical Information Systems in Local Government in Europe (pp. 49–66). London, UK: Taylor & Francis. Masser, I., & Craglia, M. (1996). A Comparative Evaluation of GIS Difusion in Local Government in Nine European Countries. In I. Masser, H. Campbell, & M. Craglia (Eds.), GIS diffusion: the Adoption and Use of Geographical Information Systems in Local Government in Europe (pp. 211–228). London, UK: Taylor & Francis. Masser, I., & Onsrud, H. J. (1993a). Diffusion and Use of Geographic Information Technologies: An Introduction. In I. Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (pp. 1–8). Dordrecht, The Netherlands: Kluwer Academic Publishers. 529 Masser, Ian, & Onsrud, H. J. (1993b). Diffusion and Use of Geographic Information Technologies. (Ian Masser & H. J. Onsrud, Eds.) (p. 349). Dordrecht, The Netherlands: Kluwer Academic Publishers. Masuki, K. F. G., Kamugisha, R., Mowo, J. G., Tanui, J., Tukahirwa, J., Mogoi, J., & O., A. E. (2010). Role of Mobile Phones in Improving Communication and Information Delivery for Agricultural Development: Lessons from South Western Uganda. ICT and Development - Research Voices from Africa. International Federation for Information Processing (IFIP), Technical Commission 9 – Relationship Between Computers and Society. Workshop at Makerere University, Uganda. 22-23 March 2010 (pp. 1–13). Kampala, Uganda: Makerere University Uganda. Retrieved from http://mak.ac.ug/documents/IFIP/RoleofMobilePhonesAgriculture.pdf Matheson, A., Shall, A., & Ogeda, M. (2008). Independent Evaluation of Uganda’s Poverty Eradication Action Plan (PEAP): Final Results and Performance Theme Paper. Oxford, UK. Matthews, P. (2010). Turning the Light on Ugandan Power Sector Investment. Ugandan Investor, (01), 29–32. McDonald, A. B. (2011, March). Can a Phone Become Your Digital Wallet? PC World: Consumer Watch. Retrieved October 9, 2011, from http://www.pcworld.com/article/217267/can_a_phone_become_your_digital_wallet.html McDougall, K., Rajabifard, A., & Williamson, I. P. (2007). A Mixed-Method Approach for Evaluating Spatial Data Sharing Partnerships for Spatial Data Infrastructure Development. In H. Onsrud (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (pp. 55–74). Redlands, CA, USA: ESRI Press. McGrath, G. (1976). The Surveying and Mapping of British East Africa 1890 to 1946. Cartographica (Monograph., pp. 1–30). Toronto, Canada: University of Toronto Press. McMaster, R. B., & Harvey, F. (2010). Geographic Information Science and Society. In J. D. . Bossler, J. B. . Campbell, R. B. . Mcmaster, & C. Rizos (Eds.), Manual of Geospatial Science and Technology (2nd ed., pp. 653–665). Boca Raton, FL, USA: CRC Press. McMaster, R. B., Leitner, H., & Sheppard, E. (1997). GIS-based Environmental Equity and Risk Assessment: Methodological Problems and Prospects. Cartography and Geographic Information Science, 24(3), 172–189. McMaster, R. B., & Manson, S. M. (2010). Geographic Information Systems and Science. In J. D. . Bossler, J. B. . Campbell, R. B. . Mcmaster, & C. Rizos (Eds.), Manual of Geospatial Science and Technology (2nd ed., pp. 513–523). Boca Raton, FL, USA: CRC Press. Miellet, P. (1996). France: A Historical Perspective on GIS Diffusion. In Ian Masser, H. J. Campbell, & M. Craglia (Eds.), GIS diffusion: the Adoption and Use of Geographical 530 Information Systems in Local Government in Europe (pp. 163–182). London, UK: Taylor & Francis. Miller, R. P. (1995). Beyond Method, Beyond Ethics: Integrating Social Theory into GIS and GIS in to Social Theory. Cartography And Geographic Information Systems, 22(1), 98–103. Miscione, G., & Vandenbroucke, D. (2011). Spatial Data Infrastructures in Context: North and South. In Z. Nedović-Budić, J. Crompvoets, & Y. Georgiadou (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 221–232). Boca Raton, FL, USA: CRC Press. Mizrachi, Y. (2009). Foris Telecom: Broadband Wireless Solutions for Digital Inclusion - The Case of Mobile WiMAX. WSIS Forum, E-Government and Public Private Partnerships for Better Public Service Delivery and MDGs Implementation, Geneva, Switzerland, 21-22 May 2009 (p. 40). Geneva, Switzerland: United Nations: World Summit on the Information Society (WSIS). Retrieved from http://unpan1.un.org/intradoc/groups/public/documents/un/unpan035408.pdf MoLHUD. (2012). MoLHUD Departments. Ministry of Lands, Housing and Urban Development, Uganda. Retrieved May 21, 2012, from http://www.mlhud.go.ug/index.php?option=com_content&view=article&id=5&Itemid=2 MobileHealthUganda. (2010). Mobile Health Uganda. Kampala, Uganda: Mobile Health Uganda. Mooneyhan, W. D. (1998). International Applications of GIS. In T. W. Foresman (Ed.), The History of Geographic Information Systems: Perspectives from the Pioneers (pp. 349–366). Upper Saddle River, NJ, USA: Prentice-Hall, Inc. Morawczynski, O. (2008). Surviving in the “Dual System”: How M-PESA is Fostering Urban-to- ural Remittances in a Kenyan Slum. In A. O. Bada & P. Musa (Eds.), Proceedings of IFIPWG 9.4, University of Pretoria Joint Workshop: Towards an ICT Research Agenda for African Development, 23-24 September, 2008, Pretoria, South Africa (pp. 110–127). Pretoria, South Africa. Retrieved from http://researchspace.csir.co.za/dspace/bitstream/10204/2501/1/Phahlamohlaka_2008.pdf#pa ge=72 Moyo, D. (2009). Dead Aid: Why Aid is not Working and How there is a Better Way for Africa (p. 188). Great Britain: Allen Lane, Penguin Books. Retrieved from http://books.google.com/books?hl=en&lr=&id=2T_RbtTslzEC&oi=fnd&pg=PP1&ots=ybX dlVZWpZ&sig=etrD3FLM2iN0YXO-_uMhuBU-C_8#v=onepage&q&f=false Moyo, R. M., & Chuba, B. (2001). Development of National Digital Spatial Database Standards and Topographic Base Maps For Environmental and Natural Resource Management in Zambia. International Conference on Spatial Information for Sustainable Development, 2–5 October 2001, Nairobi, Kenya (pp. 1–2). Nairobi, Kenya. 531 Mugabe, D. (2010, April). Bharti Airtel’s Entrance to Shake up Uganda's Telecom Industry. The New Vision. Kampala. Retrieved from http://www.newvision.co.ug/D/8/220/716082 Muhwezi, B. J. (2005). Solving Complexities in Implementation of SDI in Uganda: A Case of Uganda Bureau of Statistics. Development (pp. 1–8). Kampala, Uganda. Muhwezi, B. J. (2006). Solving Complexities in Implementation of SDI in Uganda. Development. Kampala, Uganda. Retrieved September 11, 2011, from http://www.ugandaclusters.ug/geo- im.htm Muriithi, N., & Crawford, L. (2003). Approaches to project management in Africa: implications for international development projects. International Journal of Project Management, 21(5), 309–319. doi:10.1016/S0263-7863(02)00048-0 Musinguzi, M. (2003a). Spatial Data Infrastructure - A new Challenge for GIS Application in Uganda. Mukono, Uganda. Musinguzi, M. (2003b). GIS Activities in the Wetlands Inspection Division (WID). Kampala, Uganda. Musinguzi, M. (2003c). SDI - A new challenge for Wetlands Inspection Division. Kampala, Uganda. Musinguzi, M., Bax, G., & Tickodri-Togboa, S. S. (2004). Opportunities and Challenges for SDI Development in Developing Countries - A Case Study of Uganda. Geoinformatics 2004, Proc. 12th Int. Conf. on Geoinformatics - Geospatial Information Research: Bridging the Pacific and Atlantic, University of Gävle, Sweden, 7-9 June 2004 (pp. 789–796). Gaevle, Sweden: University of Gaevle, Sweden. Musinguzi, M., Tickodri-togboa, S. S., & Bax, G. (2010). Assessment of GIS Data Interoperability in Uganda. International Journal of Spatial Data Infrastructures, (under review). Muturi, D. (2008). Zain/Ericsson Internal Lake Victoria MPS Solution Description (pp. 1–44). Kampala, Uganda. Mwape, A. (2010). Geographical Information Systems and Natural Resource Management in Zambia. Massey University, Palmerston North, New Zealand. Retrieved from http://mro.massey.ac.nz/bitstream/handle/10179/1657/02_whole.pdf?sequence=1 Mwenda, A. (2007). TED Talk: Let’s Take a New Look at African Aid. TED Talks (Technology, Entertainment, Design). Retrieved from http://www.youtube.com/watch?v=RfobLjsj230 Mwesige, P. G. (2004). Cyber Elites: a Survey of Internet Café Users in Uganda. Telematics and Informatics, 21(1), 83–101. doi:10.1016/S0736-5853(03)00024-8 NAPE. (2009, June). Oil Economy. NAPE Lobby, (June), 1–24. 532 NCGIA. (1990). National Center for Geographic Information and Analysis Annual Report, Year 1 (December 1, 1988 - November 30, 1989). International journal of geographical information systems (Vol. 6, p. 43). Santa Barbara, CA, USA. doi:10.1080/02693799208901894 NEMA. (2009). Uganda: Atlas of Our Changing Environment (p. 255). Kampala, Uganda: United Nations Environment Programme. Retrieved from http://www.grida.no/publications/the-uganda-atlas/ NEMA. (2012). National Environmental Management Authority - NEMA Uganda. About NEMA. Retrieved March 26, 2012, from http://www.nemaug.org/about_NEMA.php NFA. (2009). National Forestry Authority: Business Plan 2009-2014. Kampala, Uganda. Retrieved from http://www.nfa.org.ug/pdf/NFA Business Plan 30062009.pdf NGN. (2012). Definition of NGOs. NGO Global Network. Retrieved July 20, 2012, from http://www.ngo.org/ngoinfo/define.html NIMES. (2004). The National Integrated Monitoring and Evaluation Strategy. NORAD. (2012). About the Norwegian Agency for Development Cooperation (NORAD). About NORAD. Retrieved July 21, 2012, from http://www.norad.no/en/about-norad NRC. (2002). Down to Earth: Geographical Information for Sustainable Development in Africa. Sustainable Development (p. 155). Washington D.C., USA: National Academies Press. NRC. (2006). Learning To Think Spatially (p. 313). Washington D.C., USA: National Academies Press. Retrieved from http://books.google.com/books?hl=en&lr=&id=TJsmcnvrWVMC&pgis=1 NTV. (2011). Voters to Use Biometric Machines. Retrieved from http://www.youtube.com/watch?v=TrgEBX86Qg8 NWSC. (2012). National Water and Sewerage Corporation. National Water and Sewerage Corporation: Our Profile. Retrieved May 22, 2012, from http://www.nwsc.co.ug/about.php Najar, C., Rajabifard, A., Williamson, I., & Giger, C. (2007). A Framework for Comparing Spatial Data Infrastructures: An Australian-Swiss Case Study. In H. Onsrud (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (pp. 201–214). Redlands, CA, USA: ESRI Press. Nakaweesi, D. (2009). Uganda: Zain Rescue Project Starts on Lake Victoria. The Monitor: Uganda - Zain Rescue Project Starts on Lake Victoria. Retrieved August 22, 2012, from http://allafrica.com/stories/200907300043.html Nasirumbi, S. (2006). Towards Strategy of Spatial Data Infrastructure Development with Focus on the Private Sector Involvement: A case Study in Uganda. Geo-Information Science. 533 International Institute for Geo-Information Science and Earth Observation, ITC, Encschede, The Netherlands. NatureUganda. (2012). About Nature Uganda. Nature Uganda. Retrieved July 25, 2012, from http://www.natureuganda.org/index_.php#about Ndiwalana, A. (2011). Transformation-Ready: The Strategic Application of Information and Communication Technologies in Africa: Final Education Sector Study, Annex 8, Uganda Case Study. Kampala, Uganda. Retrieved from http://etransformafrica.org/sites/default/files/eTA - Education - Final report - Supplementary - Annex 8 - Uganda case study.pdf Nebert, D., Reed, C., & Wagner, R. (2007). Proposal for a Spatial Data Infrastructure Standards Suite: SDI 1.0. In H. Onsrud (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (pp. 147–160). Redlands, CA, USA: ESRI Press. Nedović-Budić, Z., Crompvoets, J., & Georgiadou, Y. (2011). Conclusion: Implications for Future Research and Practice - Toward Scientific Pragmatism. In Z. Nedović-Budić, J. Crompvoets, & Y. Georgiadou (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 233–241). Boca Raton, FL, USA: CRC Press. Nedović-Budić, Z., & Godschalk, D. R. (1996). Human Factors in Adoption of Geographic Information Systems: A Local Government Case Study. Public Administration Review, 56(6), 554–567. doi:10.2307/977254 Nedović-Budić, Z., Pinto, J. K., & Warnecke, L. (2011). GIS Database Development and Exchange: Interaction Mechanisms and Motivations. In Z. Nedović-Budić, J. Crompvoets, & Y. Georgiadou (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 69– 100). Boca Raton, FL, USA: CRC Press. Nedović-budić, Z., Pinto, J. K., & Warnecke, L. (2004). GIS Database Development and Exchange: Interaction Mechanisms and Motivations. Urban and Regional Information Systems Association (URISA) Journal, 16(1), 15–29. NewAgriculturist. (2009). Uganda’s GPS stock-take. NewAgriculturist: Developments. Retrieved August 11, 2011, from http://www.new-ag.info/en/developments/devItem.php?a=934 Nkurunziza, E. (2006). Two states, One City? Conflict and Accommodation in Land Delivery in Kampala, Uganda. International Development Planning Review, 28(2), 159–180. Noble, D. F. (1984). Forces of Production: A Social History of Industrial Automation (p. 409). New York, NY, USA: Knopf. Noongo, E. N. (2007). The Implementation of Geographic Information Systems in Namibia. Joensuu, Sweden: Social Science Publications, University of Joensuu, Sweden. Retrieved from 534 http://joypub.joensuu.fi/publications/frameappl/pdf.php?pdf=dissertations/noongo_impleme ntation/noongo Nyakana, J. B., Sengendo, H., & Lwasa, S. (2007). Population, Urban Development and the Environment in Uganda : The Case Of Kampala City and its Environs. PRIPODE Workshop on Urban Population, Development and Environment in Developing Countries, Nairobi, Kenya, 11-13 June, 2007 (pp. 1–22). Nairobi, Kenya: Committee for International Cooperation in National Research in Demography, CICRED. Nyemera, B. W. (2008). Evaluation of Redundancy in the Geo-information Community in Uganda. Geo-Information Science. International Institute for Geo-information Science and Earth Observation, ITC, The Netherlands. Nyerges, T. L., McMaster, R., & Couclelis, H. (2011). Geographic Information Systems and Society: A Twenty year Research Perspective. In T. L. Nyerges, H. Couclelis, & R. McMaster (Eds.), The SAGE Handbook of GIS and Society (pp. 3–21). London, UK: SAGE Publications Ltd. OAG. (2004). Investigation into Alleged Computer Fraud at Kampala City Council. City (pp. 1– 25). Kampala, Uganda. OECD. (2012). International Organizations. Organization for Economic Co-operation and Development (OECD): Glossary of Statistical Terms. Retrieved July 21, 2012, from http://stats.oecd.org/glossary/detail.asp?ID=1434 OPM. (2008). Independent Evaluation of Uganda’s Poverty Eradication Action Plan (PEAP): Final Synthesis Report. Evaluation (p. 156). Oxford, UK. Obermeyer, N. J. (1993). GIS in a Democratic Society: Opportunities and Problems. Proceedings of the Geographic Information and Society Workshops, NCGIA, Friday Harbor, Washington, Nov. 11-14, 1993. Friday Harbor, Washington: NCGIA. Obuah, E. (2010). Combatting Corruption in Nigeria : The Nigerian Economic and Financial Crimes (EFCC). African Studies Quarterly, 12(1), 17–44. Ofori-amoah, B. (2004). The Effectiveness of LoGICS in the Uganda Local Governments: A Case Study. Mauritius. Ofori-amoah, B. (2010). Building Capacity to Use Geospatial Technology for Development in Africa: Lessons from the Uganda GIS Project. Okia, Y. (Departmet of S. and M. U., & Kitaka, J. (Department of S. and M. U. (2010). The Uganda Triangulation Network: Establishment and Current Status. Africa. Kampala, Uganda. Okolloh, O. R. Y. (2008). Ushahidi, or ‘testimony: Web 2.0 tools for crowdsourcing crisis information 9. Participatory Learning and Action, (January), 65–70. 535 Okuku, J. A. (2006). The Land Act 1998 and Land Tenure Reform in Uganda. Africa Development, 31(1), 1–26. Onah, C. C. (2009). Spatial Data Infrastructures Model for Developing Countries: A Case Study of Nigeria. Analysis. Westfaelische Wilhelms Universitaet, Muenster, Germany. Onsrud, H. J. (1991). Diffusion of Geographical Information. International Journal of Geographical Information Systems, 5(4), 447–467. Onsrud, H. J. (1998). Tragedy of the Information Commons. In F. D. R. Taylor (Ed.), Policy Issues in Modern Cartography (pp. 141–158). Oxford, UK: ELSEVIER SCIENCE Ltd. Onsrud, H. J. (2004). The US National Spatial Data Infrastructure: Legal & Economic Issues and Developments. In B. Van Loenen & B. C. Kok (Eds.), Spatial Data Infrastructure and Policy Development in Europe and the United States (pp. 87–100). Delft, The Netherlands: Delft University Press. Onsrud, H. J., & Craglia, M. (2003). Introduction to the Special Issues on Access and Participatory Approaches in Using Geographical Information. Urban and Regional Information Systems Association (URISA) Journal, 15(1), 5–8. Onsrud, H. J., & Pinto, J. K. (1991). Diffusion of Geographic Information Innovations. International Journal of Geographical Information Systems, 5(4), 447–467. Onsrud, H. J., & Pinto, J. K. (1993). Evaluating Correlates of GIS Adoption Success and the Decision Process of GIS Acquistion. Urban and Regional Information Systems Association (URISA) Journal, 5(1), 18–39. Onsrud, H., Poore, B., Rugg, R., Taupier, R., & Wiggins, L. (2005). The Future of the Spatial Information Infrastructure. In R. B. McMaster & L. E. Usery (Eds.), A Research Agenda for Geographic Information Science (pp. 225–255). Boca Raton, FL, USA: CRC Press. OpenXdata. (2012). OpenXdata. OpenXData. Retrieved August 20, 2012, from http://www.openxdata.org/ Orange. (2012). Internet Everywhere. Orange Telecom: Internet Everywhere. Retrieved August 13, 2012, from http://www.orange.ug/mobile-plans/internet-everywhere.php OrangeTelecom. (2010). Mobile Innovations from the Perspective of the Service Operator. Workshop on Mobile Applications, Faculty of Computing and Informatics Technology, Makerere University, Kampala , 30th July 2010. Kampala, Uganda: Faculty of Computing and Informatics Technology, Makerere University, Kampala, Uganda. OrangeTelecom. (2012). Fleet Live. Orange Telecom: Fleet Live. Retrieved August 23, 2012, from http://www.orange.ug/orange-business/fleet-live.php 536 Owoeye, J. S., & Oyebade, S. A. (2011). Higher Education Research in Uganda: Problems and Prospects. Tallahassee, FL: Association of Institutional Research. Retrieved from http://www3.airweb.org/images/herpnet_v2_no2.pdf Paay, J., & Kjeldskov, J. (2008). Understanding the User Experience of Location-based Services: Five Principles of Perceptual Organisation Applied. Journal of Location Based Services, 2(4), 267–286. doi:10.1080/17489720802609328 Peng, Z.-R., & Tsou, M.-H. (2003). Internet GIS: Distributed Geographic Information Services for the Internet and Wireless Networks (p. 668). New Jersey, NJ: John Wiley and Sons, Inc. Retrieved from http://books.google.com/books?id=sk5UHK- FJM8C&printsec=frontcover&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false Perry, J., & Bratman, M. (1999). Introduction to Philosophy: Classical and Contemporary Readings (3rd ed., p. 824). Oxford, UK: Oxford University Press. Pfaff, J. (2010). A mobile phone: Mobility, Materiality and Everyday Swahili Trading Practices. Cultural Geographies, 17(3), 341–357. doi:10.1177/1474474010368606 Phillips, D. C., & Burbules, N. C. (2000). Postpositivism and Educational Research. Lanham, Maryland, USA: Rowman & Littlefield Publishers, Inc. Retrieved from http://books.google.com/books?id=0H40ZTPQvQAC&printsec=frontcover&source=gbs_ge _summary_r&cad=0#v=onepage&q&f=false Pickles, J. (1995). Representations in an Electronic Age: Geography, GIS, and Democracy. In J. Pickles (Ed.), Ground Truth: The Social Implications of Geographic Information Systems (pp. 1–30). New York, NY, USA: The Guilford Press. Pinto, J. K., & Onsrud, H. J. (1993). Correlating Adoption Factors and Adopter Characteristics with Successful use of Geographic Information Systems. In Ian Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (pp. 165–194). Dordrecht, The Netherlands: Kluwer Academic Publishers. Pinto, J. K., & Onsrud, H. J. (1995). Sharing Geographic Information Across Organizational Boundaries. In H. J. Onsrud & G. Rushton (Eds.), Sharing Geographic Information (pp. 44– 64). New Brunswick, New Jersey: Center for Urban Policy Research, Rutgers, The State University of New Jersey. Poiker, T. (1995). Special Isssue: GIS and Society. (E. Sheppard & T. Poiker, Eds.) (p. 104). Friday Harbor, Washington: Cartography and Geographic Information Systems. Popper, K. R. (1959). Prediction and Prophecy in the Social Sciences. In P. Gardiner (Ed.), Theories of History (pp. 276–285). The Free Press. Retrieved from http://keidahl.terranhost.com/Fall/HIS3104/Popper Prediction and Prophecy.pdf 537 Puri, S. K. (2006). Technological Frames of Stakeholders Shaping the SDI Implementation: A Case Study from India. Information Technology for Development, 12(4), 311–331. doi:10.1002/itdj Puri, S. K., & Sahay, S. (2004). The Politics of Knowledge in Using GIS for Land Management in India. In B. Kaplan, D. P. Truex III, D. Wastell, T. A. Woodharper, & J. I. DeGross (Eds.), Information Systems Research: Relevant Theory and Informed Practice (IFIP Advances in Information and Communication Technology) (pp. 597–614). Dordrecht, The Netherlands: Kluwer Academic Publishers. Puri, S. K., Sahay, S., & Georgiadou, Y. (2007). A Metaphor-Based Sociotechnical Perspective on Spatial Data Infrastructure Implementations: Some Lessons from India. In H. Onsrud (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (pp. 161– 174). Redlands, CA, USA: ESRI Press. Pursell, C. (1993). The Rise and Fall of the Apprpriate Technology Movement in the United States, 1965-1985. Technology and Culture, 34(3), 629–637. Retrieved from http://www.jstor.org/stable/10.2307/3106707 RTI. (2010). Transforming Participatory Governance in Uganda. Manager (p. 2). Washington D.C., USA. Rajabifard, A., Binns, A., Masser, I., & Williamson, I. (2006). The Role of Sub-National Government and the Private sector in future Spatial Data Infrastructures. International Journal of Geographical Information Science, 20(7), 727–741. doi:10.1080/13658810500432224 Rajabifard, A., Feeney, M.-E. F., Williamson, I., & Masser, I. (2003). National SDI Initiatives. In I. Williamson, A. Rajabifard, & M.-E. F. Feeney (Eds.), Developing Spatial Data Infrastructures: From Concept to Reality (pp. 95–110). Boca Raton, FL, USA: CRC Press. Ramasubramanian, L. (1999). GIS Implementation in Developing Countries: Learning from Organisational Theory and Reflective Practice. Transactions in GIS, 3(4), 359–380. doi:10.1111/1467-9671.00028 Ramasubramanian, L. (2011). PPGIS Implementation and the Transformation of US. In T. L. Nyerges, H. Couclelis, & R. McMaster (Eds.), The SAGE Handbook of GIS and Society (pp. 400–422). London, UK: SAGE Publications Ltd. Rambaldi, G., Mccall, M., & Weiner, D. (2006). Participatory Spatial Information Management and Communication in Developing Countries. The Electronic Journal on Information Systems in Developing Countries - EJISDC, 25(1), 1–9. Rana, S., & Joliveau, T. (2009). NeoGeography: An Extension of Mainstream Geography for Everyone Made by Everyone? Journal of Location Based Services, 3(2), 75–81. doi:10.1080/17489720903146824 538 Raper, J., Gartner, G., Karimi, H., & Rizos, C. (2007a). Applications of location–based services: a selected review. Journal of Location Based Services, 1(2), 89–111. doi:10.1080/17489720701862184 Raper, J., Gartner, G., Karimi, H., & Rizos, C. (2007b). A Critical Evaluation of Location Based Services and their Potential. Journal of Location Based Services, 1(1), 5–45. RapidSMS. (2012). RapidSMS. RapidSMS. Retrieved August 19, 2012, from http://www.rapidsms.org/ Raubal, M. (2011). Cogito Ergo Mobilis Sum: The Impact of Location-based Services on Our Mobile Lives. In T. L. Nyerges, H. Couclelis, & R. McMaster (Eds.), The SAGE Handbook of GIS and Society (pp. 159–173). London, UK: SAGE Publications Ltd. ReCAPP. (2011). Social Learning Theory. Social Learning Theory: Resource Center for Adolescent Pregnancy Prevention (ReCAPP). Retrieved from http://www.etr.org/recapp/index.cfm?fuseaction=pages.TheoriesDetail&PageID=380 Ribot, F. (1989). Télédétection en Afrique subsaharienne, bilan régional. Télédétection en Francophonie, (AUPELF-UREF, John Libbey Eurotext, Paris), 17–34. Retrieved from http://www.francophonie.org/ Richter, C., Miscione, G., & De, R. (2011). Enlisting SDI for Urban Planning in India: Local Practices in the Case of Slum Declaration. In Z. Nedović-Budić, J. Crompvoets, & Y. Georgiadou (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 157–180). Boca Raton, FL, USA: CRC Press. Rizos, C. (2007). A Traveller’s Tale … Lost in Transit … Tango in Turin … A Streetcar Named Padang … An Australian in Bandung … (make up your own travel-themed movie titles). Journal of Location Based Services, 1(4), 231–236. doi:10.1080/17489720802232527 Rocco, T. S., Bliss, L. A., Gallagher, S., & Pérez-Prado, A. (2003). Taking the Next Step: Mixed Methods Research in Organizational Systems. Information Technology, Learning, and Performance Journal, 21(1), 19–29. Roche, S., & Humeau, J.-B. (1999). GIS Development and Planning Collaboration: a Few Examples from France. Urban and Regional Information Systems Association (URISA) Journal, 11(1), 5–12. Rogers, E. M. (1993). The Diffusion of Innovations Model. In Ian Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (pp. 9–24). Dordrecht, The Netherlands: Kluwer Academic Publishers. Rogers, E. M. (1995). Diffusion of Innovations (4th ed., pp. 1–519). New York, NY, USA: The Free Press, A Division of Macmillan, Inc. 539 Rogers, E. M. (2003). Diffusion of Innovations (5th ed., p. 551). New York, New York, USA: Free Press. Rumor, M. (1993). The Use of Geographic Information Technology in the City of Padova. In Ian Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (Vol. 1, pp. 229–244). Dordrecht, The Netherlands: Kluwer Academic Publishers. Rundstrom, R. a. (1995). GIS, Indigenous Peoples, and Epistemological Diversity. Cartography and Geographic Information Science, 22(1), 45–57. doi:10.1559/152304095782540564 Ryan, B., & Gross, N. (1950). Acceptance and Diffusion of Hybrid Corn Seed in Two Iowa Communities. Agricultural Experiment Station, Ames, Iowa - Iowa State College of Agriculture and Mechanic Arts, Research B(January), 661–708. SCIT. (2012). Welcome. School of Computing and Informatics Technology, Makerere University. Retrieved August 20, 2012, from http://cit.mak.ac.ug/nsic/index.php SMS-Media. (2010). Innovative Value Added Services: Future Trends Versus Customer Demands. Workshop on Mobile Applications, Faculty of Computing and Informatics Technology, Makerere University, Kampala , 30th July 2010. Kampala, Uganda: Faculty of Computing and Informatics Technology, Makerere University, Kampala, Uganda. SMS-Media. (2012). SMS-Media: Products. SMS Media: Products. Retrieved August 23, 2012, from http://www.smsmedia.info/ SMSInAction. (2011). Farmer’s Friend - Grameen AppLab/Google SMS. Farmer’s Friend - Grameen AppLab/Google SMS. Retrieved May 17, 2012, from https://smsinaction.crowdmap.com/reports/view/61 Sahay, S. (1998). Implementing GIS technology in India : Some Issues of Time and Space. Accounting, Management, and Information Technology, 8, 147–188. Sahay, S., & Walsham, G. (1996). Implementation of GIS in India: Organizational Issues and Implications. International Journal of Geographical Information Systems, 10(4), 385–404. Sahay, S., & Walsham, G. (1997). Use of Geographical Information Systems in Developing Countries: Social and Management Issues. Vienna, Austria: UNIDO and the International Centre for Science and High Technology. Sale, J. E. M., Lohfeld, L. H., & Brazil, K. (2002). Revisiting the Quantitative-Qualitative Debate : Implications for Mixed-Methods Research. Quality and Quantity, 36, 43–53. Salem, B. B. (1994). Scientific Applications of GIS in Egypt: A Commentary on the Current Status. In M. Ehlers, D. Steiner, & J. Johnston (Eds.), ISPRS Workshop: Requirements for Integrated Geographic Information Systems (Vol. 5, pp. 29–34). Ann Arbor, MI, USA: Environmental Research Institute of Michigan. 540 Samatar, A. I. (1999). An African Miracle: State and Class Leadership and Colonial LEgacy in Botswana Development (p. 217). Portsmouth, UK: Heinemann. Sanderson, G. N. (1963). The Anglo-German Agreement of I890 and the Upper Nile. The English Historical Review, 78(306), 49–72. Schaffner, C. (2009). Africa: Google and Grameen Foundation Launch AppLab. Afrik-News: Africa - Google and Grameen Foundation Launch AppLab. Retrieved May 16, 2012, from http://www.afrik-news.com/article15974.html Schmid, F. (2008). Knowledge-based Wayfinding Maps for Small Display Cartography. Journal of Location Based Services, 2(1), 57–83. doi:10.1080/17489720802279544 Schumacher, E. F. (1973). Small is Beautiful: Economics as if People Mattered (1st ed.). London, UK: Blond and Briggs. Retrieved from http://books.google.com/books?hl=en&lr=&id=ME1_WnYYW6UC&oi=fnd&pg=PR11&d q=E+F+Schumacher+Small+is+Beautiful+1973&ots=aefQGfs0rn&sig=AYuN6VngtK4mz vtwGt-FOQ4ozhQ#v=onepage&q=E F Schumacher Small is Beautiful 1973&f=false Schuurman, N. (2009). An Interview With Michael Goodchild: GIScience and Social Reordering in the New Millennium. The Information Society, 25(5), 360–363. doi:10.1080/01972240903213159 Schwabe, C. A. (2005). The GeoInformation Industry in Africa: Prospects and Potentials. Fourth Meeting of the Committee on Development Information (CODI IV - UN Economic and Social Council, UN Economic Commission for Africa), 23-28 April 2005, Addis Ababa, Ethiopia (pp. 1–16). Addis Ababa, Ethiopia: UN Economic Commision for Africa. Schwabe, C. A. (2010). Getting Geoinformation and SDI to work for Africa – Part 2. GIS Technical. Retrieved August 10, 2011, from http://eepublishers.co.za/images/upload/part2.pdf Schwandt, T. A. (1994). Constructivist, Interpretivist Approaches to Human Inquiry. In N. K. Denzin & Y. S. Lincoln (Eds.), Handbook of Qualitative Research (pp. 118–137). Thousand Oaks, CA.: Sage Publications. Schwartz, M. (2011). The Egyptian Uprising: The Mass Strike in the Time of Neoliberal Globalization. New Labor Forum, 20(3), 33–43. Sentongo, P. M. (Assistant C. C. and M. O. of the P. M. (2003). The Role of Spatial Data Infrastructure (SDI) in the Development of a National Integrated Monitoring & Evaluation System (NIMES) Capacity in Uganda. Kampala, Uganda. Sentrack. (2012). Vehicle Tracking. Sentrack Uganda. Retrieved August 23, 2012, from http://sentrackuganda.com/fleetmanagement.php 541 Setel, P. W., Macfarlane, S. B., Szreter, S., Mikkelsen, L., Jha, P., Stout, S., & AbouZahr, C. (2007). A Scandal of Invisibility: Making Everyone Count by Counting Everyone. Lancet, 370(9598), 1569–1577. doi:10.1016/S0140-6736(07)61307-5 Sey, A. (2008). Mobile Communication and Development: A Study of Mobile Phone Appropriation in Ghana. University of Southern California. Retrieved from http://digitallibrary.usc.edu/assetserver/controller/item/usctheses-m1318/etd-Sey- 20080707.pdf Sharma, N. (2012). LBS India 2012: Operators Seeking Govt. Subsidy on Monetary Basis to Deploy LBS Infra. Insight VAS. Sheppard, E. (1993). GIS and Society: Ideal and Reality. Proceedings of the Geographic Information and Society Workshops, NCGIA, Friday Harbor, Washington, Nov. 11-14, 1993. Friday Harbor, Washington: NCGIA. Sheppard, E. (1995). GIS And Society: Towards a Research Agenda. (E. Sheppard & T. Poiker, Eds.)Cartography and Geographic Information Science, 22(1 (Special Issue: GIS and Society)), 5–16. Sheppard, E., Couclelis, H., Graham, S., Harrington, J. W., & Onsrud, H. (1999). Geographies of the Infomation Society. International Journal of Geographical Information Science, 13(8), 797–823. Shrum, W., Mbatia, P. N., Palackal, A., Dzorgbo, D.-B. S., Duque, R. B., & Ynalvez, M. A. (2011). Mobile Phones and Core Network Growth in Kenya: Strengthening Weak Ties. Social Science Research, 40(2), 614–625. doi:10.1016/j.ssresearch.2010.09.015 Shupeng, C. (1987). Geographical Data Handling and GIS in China. International Journal of Geographical Information Systems, 1(3), 219–228. Sieber, R. E. (1993). GIS Implementation in the Grassroots. Urban and Regional Information Systems Association (URISA) Journal, 12(1), 15–29. Silva, L. (2011). Institutionalization Does Not Occur by Decree: Institutional Obstacles in Implementing a Land Administration System in a Developing Country. In Z. Nedovic- Budic, J. Crompvoets, & Y. Georgiadou (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 21–47). Boca Raton, FL, USA: CRC Press. Sim, J. (1998). Collecting and analysing qualitative data: issues raised by the focus group. Journal of Advanced Nursing, 28(2), 345–52. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/9725732 Sipe, N. G., & Dale, P. (2003). Challenges in Using Geographic Information Systems (GIS) to Understand and Control Malaria in Indonesia. Malaria Journal, 8, 1–8. 542 Smit, J., Makanga, P., Lance, K., & Vries, W. D. (2009). Exploring Relationships Between Municipal and Provincial Government SDI Implementers in South Africa. Proceedings of the GSDI 11 World Conference: Spatial Data Infrastructure Convergence - Building SDI Bridges to Address Global Challenges, June 15-19, 2009, Rotterdam, The Netherlands (p. 18). Rotterdam, The Netherlands: GSDI. Smith, J. K. (1983). Quantitative versus Qualitative Research: An Attempt to Clarify the Issue. Educational Researcher, 12(3), 6–13. Retrieved from http://www.jstor.org/stable/10.2307/1175144 Smith, J., & Kealy, A. (2003). SDI and Location-Based Wireless Applications. In I. Williamson, A. Rajabifard, & M.-E. F. Feeney (Eds.), Developing Spatial Data Infrastructures: From Concept to Reality (pp. 263–279). Boca Raton, FL, USA: CRC Press. Smith, R. G. (2007). Biometric Solutions to Identity-related Cyber Crime. In Y. Jewkes (Ed.), Crime Online (pp. 44–59). Devon, UK: Willan Publishing. Retrieved from http://books.google.com/books?hl=en&lr=&id=- QzLySW8KtYC&oi=fnd&pg=PA44&dq=National+Identity+Card+UGanda&ots=QkBapH qmM-&sig=vPmKn63xGC-f4Jd13wp2Iqn3Jck#v=onepage&q=National Identity Card UGanda&f=false Snider, E. A., & Faris, D. M. (2011). The Arab Spring: U.S. Democracy Promotion in Egypt. Middle East Policy, 1–9. Retrieved from http://www.mepc.org/journal/middle-east-policy- archives/arab-spring-us-democracy-promotion-egypt? Steup, M. (2011). Epistemology. Stanford Encyclopedia of Philosophy. Retrieved December 25, 2011, from http://plato.stanford.edu/archives/win2011/entries/epistemology Strauss, A. (1995). Notes on the Nature and Development of General Theories. Qualitative Inquiry, 1(1), 7–18. doi:10.1177/107780049500100102 Sui, D. Z. (2011). Legal and Ethical Issues of Using Geospatial Technologies. In T. L. Nyerges, H. Couclelis, & R. McMaster (Eds.), The SAGE Handbook of GIS and Society (pp. 504– 528). London, UK: SAGE Publications Ltd. Sui, D. Z., & Morril, R. (2004). Computers and Geography: From Automated Geography to Digital Earth. In S. D. Brunn, S. L. Cutter, & J. W. Harrington Jr. (Eds.), Geography and Technology (pp. 81–109). Dordrecht, The Netherlands: Kluwer Academic Publishers. SwiftRiver. (2012). SwiftRiver. SwiftRiver. Retrieved August 19, 2012, from http://www.swiftly.org/beta/ Tashakkori, A., & Teddlie, C. (Eds.). (2003). Handbook of Mixed Methods in Social & Behavioral Research (p. 768). SAGE. Retrieved from http://books.google.com/books?hl=en&lr=&id=F8BFOM8DCKoC&pgis=1 543 Taylor, F. D. R. (1986). Computer Assisted Cartography in Developing Nations. AutoCarto - Digital Mapping and Spatial Information Systems: Proceedings of the International Conference on the Acquisition, Management and Presentation of Spatial Data, 2, 462–471. Retrieved from http://mapcontext.com/autocarto/proceedings/auto-carto-london-vol- 2/pdf/computer-assisted-cartography-in-developing-nations.pdf Taylor, F. D. R. (1998). Modern Cartography, Policy Issues and the Developing Nations: Rhetoric and Reality. In F. D. R. Taylor (Ed.), Policy Issues in Modern Cartography (pp. 185–214). Oxford, UK: ELSEVIER SCIENCE Ltd. Taylor, F. D. R. (2004). Capacity Building and Geographic Information Technologies in African Development. In S. D. Brunn, S. L. Cutter, & J. W. Harrington Jr. (Eds.), Geography and Technology (pp. 521–546). Dordrecht, The Netherlands: Kluwer Academic Publishers. Taylor, F. D. R. (2005). The History and Development of Global Map: A Spatial Data Infrastructure for Sustainable Development in the Middle East and North Africa. GSDI-8, From Pharaohs to Geoinformatics - The Role of SDI’s in an Information Society, April 16- 21, 2005, Cairo Egypt. Cairo, Egypt: GSDI. Taylor, F. D. R. (2006). A Geospatial Data Infrastructure for Sustainable Development in Latin America. GSDI-9 Conference Proceedings, Geospatial Information: Tool for Reducing Poverty, 6-10 November 2006, Santiago, Chile. Santiago, Chile: GSDI. Taylor, M. (2011). The Difficult Bit: the Arab Spring After Libya. Middle East. Retrieved September 16, 2011, from http://www.opendemocracy.net Tchindjang, M., Menga, V. F., Makak, J. S., Nghonda, J. P., Nziengui, M., & Edou, M. (2005). Remote Sensing in Africa: Mapping Approach and Perspectives. Proceedings of the 22nd International Cartographic Conference (ICC), Coruna, Spain, 9-16 July, 2005 (p. 10). Coruna, Spain. Retrieved from http://icaci.org/documents/ICC_proceedings/ICC2005/htm/oral.htm Teferra, D., & Altbach, P. G. (2004). African higher education: Challenges for the 21st century. Higher Education, 47, 21–50. TextToChange. (2012). TextToChange. TextToChange. Retrieved August 19, 2012, from http://www.texttochange.org/about-ttc Thompson, D. (1991). G.I.S. A View from the Other (Dark?) Side: The Perspective of an Instructor of Introductory Geography Courses at University Level. Cartographica, 28(3 (GIS Education and Training)), 55–64. Tian, L., Shi, J., & Yang, Z. (2009). Why Does Half the World’s Population Have a Mobile phone? An Examination of Consumers' Attitudes Toward Mobile Phones. Cyberpsychology & behavior: the impact of the Internet, multimedia and virtual reality on behavior and society, 12(5), 513–6. doi:10.1089/cpb.2008.0335 544 Timko, I. (2006). Capturing and Querying Complex Location-Based Data. Aalborg University. Retrieved from http://www.inf.unibz.it/~itimko/thesis/thesis_fulltext.pdf Toffler, A. (1980). Third Wave (p. 544). New York, New York, USA: Morrow. Retrieved from http://calculemus.org/lect/07pol-gosp/arch/proby-dawne/materialy/waves.htm Tomlinson, R. F. (1987). Current and Potential Uses of Geographical Information Systems: The North American Experience. International Journal of Geographical Information Systems, 1(3), 203–218. Tomlinson, R. F. (1988). The Imact of the Transition from Analogue to Digital Cartographic Representation. The American Cartographer, 15(3), 249–262. Tomlinson, R. F. (1998). The Canada Geographic Information System. In T. W. Foresman (Ed.), The History of Geographic Information Systems: Perspectives from the Pioneers (pp. 21– 32). Upper Saddle River, NJ, USA: Prentice-Hall, Inc. Toppen, F. J. (1991). GIS Eduction in the Netherlands: A Bit of Everything and Everything About a BIT? Cartographica, 28(3 (GIS Training and Education)), 1–9. TourGuide. (2010a). Travel Map of Uganda. Kampal, Uganda: TourGuide Publications, an Imprint of Fountan Publishers Ltd. TourGuide. (2010b). Kampala A-Z. Kampala, Uganda: TourGuide Publications, an Imprint of Fountan Publishers Ltd. TourGuide. (2012). TourGuide Publications: Maps. TourGuide Publications. Retrieved July 30, 2012, from TourGuide Publications Touzani, A. (2006). Activités du Centre R Régional Africain des Sciences et Technologies de l’Espace en Langue Française (CRASTE CRASTE-LF). 49ème Session du Comité des Utilisations Pacifiques de l’Espace Extra - AtmosphAtmosphérique, Vienna, Austria, 7-16 June 2006 (pp. 1–35). Vienna, Austria: United Nations Committee on the Peaceful Uses of Outer Space (COPUOS). Retrieved from http://www.oosa.unvienna.org/pdf/pres/copuos2006/10.pdf Transparency International. (2011). Corruption Perceptions Index 2011. Corruption Perceptions Index 2011. Retrieved July 5, 2012, from http://cpi.transparency.org/cpi2011/results/ TrueAfrican. (2012). TrueAfrican: Company Profile. TrueAfrican. Retrieved August 23, 2012, from http://www.trueafrican.com/about-us Tuhumwire, J. T. (Department of G. S. and M. (2009). Recent Mineral Sector Interventions and On-Going Activities in Uganda. Symposium on Uganda Airborne Geophysical Surveys, Kampala, Uganda, 16-17 July 2009 (Sustainable Management of Mineral Resource Project (SMMRP)) (pp. 1–26). Kampala, Uganda: Department of Geological Survey and Mines, Ministry of Energy and Mineral Development. 545 Tukugize, C. (2004). The Role of Public and Private Organizations in the Growth of Geoinformation Market in Uganda. Kampala, Uganda. Tukugize, C. (2005). Evaluation of Geoinformation Market Environment in East Africa. Africa. International Institute for Geo-information Science and Earth Observation, ITC, Netherlands. Turyareeba, P., & Drichi, P. (2001). Plan for Development of Uganda’s Biomass Energy Strategy. Energy Policy (p. 39). Kampala, Uganda. U-Consult. (2004). GIS Capacity Development for the Uganda Bureau of Statistics: Second Economic and Financial PRoject, Final Report (p. 122). Entebbe, Uganda. UBOS. (2006). 2002 Uganda Population and Housing Census Analytical Report: Population Composition. Ethnicity (p. 47). Kampala, Uganda. Retrieved from http://www.ubos.org/onlinefiles/uploads/ubos/pdf documents/2002 CensusPopnCompostionAnalyticalReport.pdf UBOS. (2007). Projections of Demographic Trends in Uganda, Volume 1. Applied Economics (Vol. I, p. 37). Kampala. UBOS. (2010). Uganda National Household Survey 2009/2010. Distribution (pp. 1–221). Kampala, Uganda. UBOS. (2012). Uganda Bureau of Statistics: About US. Uganda Bureau of Statistics. Retrieved May 22, 2012, from http://www.ubos.org/index.php?st=page&id=1&p=Facts about Uganda Bureau of Statistics UCGIS, & AAG. (2006). Geographic Information Science and Technology: Body of Knowledge. (D. DiBiase, M. Demers, A. Johnson, K. Kemp, & A. T. Luck, Eds.) (1st ed., p. 162). Washington D.C., USA: AAG. Retrieved from http://books.google.com/books?id=YevhAQAACAAJ&dq=Geographic+Information+Scien ce+and+Technology+Body+of+Knowledge&source=bl&ots=w8Z_OVOI3x&sig=7hJE_wK 3pfBuMjJnnlCE67qfDE4&hl=en&sa=X&ei=SvwBUNaFKIjPqgHJ9JGxDA&ved=0CDoQ 6AEwAA UNECA-NICI. (2001a). Uganda: NICI Infrastructure. Uganda: NICI Infrastructure. Retrieved August 14, 2012, from http://www.uneca.org/aisi/nici/country_profiles/uganda/uganinfra.htm UNECA-NICI. (2001b). Uganda: Internet Connectivity. Uganda: Internet Connectivity. Retrieved August 13, 2012, from http://www.uneca.org/aisi/nici/country_profiles/uganda/uganinter.htm UNEP. (1972). Report of the United Nations Conference on the Human Environment. United Nations Environmental Programme - Environment for Development. Retrieved December 13, 2011, from http://www.unep.org/Documents.Multilingual/Default.asp?documentid=97 546 UNEP. (2008). Africa: Atlas of Our Changing Environment (p. 374). Nairobi, Kenya: United Nations Environment Programme. Retrieved from http://www.unep.org/publications/search/pub_details_s.asp?ID=3993 UNEP-DEWA. (2007). Africa Environment Outlook: Policy Analysis Guidelines for Integrated Environmental Assessment and Reporting (p. 77). Nairobi, Kenya. UNICEF. (2008). Challenges in UNICEF´s Mobile Technology Implementations. In W3C (Ed.), Workshop on the Role of Mobile Technologies in fostering social development, Sao Paulo, Brazil, June 2-3, 2008. Sao Paulo, Brazil: W3C. Retrieved from http://www.w3.org/2008/02/MS4D_WS/papers/ UNICEF. (2012a). UReport: Community Empowerment Via RapidSMS - Uganda. UReport: Community Empowerment Via RapidSMS - Uganda. Retrieved August 19, 2012, from http://unicefinnovation.org/case-studies/ureport-community-empowerment-rapidsms- uganda UNICEF. (2012b). U-Report: Revolutionary Technology for Development (T4D) from UNICEF Uganda. U-Report. Retrieved August 6, 2012, from http://vimeo.com/38458029 URISA. (2003). GIS Code of Ethics. GIS Code of Ethics. Retrieved December 17, 2011, from http://www.urisa.org/about/ethics USAID. (2010). Spring Project. Spring. Gulu, Uganda: USAID. USAID. (2012). USAID: FAQs. USAID: Frequently Asked Questions. Retrieved July 21, 2012, from http://transition.usaid.gov/faqs.html UTL. (2012). 3G Broadband. UTL: 3G Broadband. Retrieved August 13, 2012, from http://www.utl.co.ug/internet/broadband-services/3g-broadband/ UgandaClusters. (2012). Uganda Clusters. Uganda Clusters. Retrieved May 22, 2012, from http://www.ugandaclusters.ug/ UgandaPeople. (2011). Uganda: Geography and Climate. Uganda People. Retrieved November 6, 2011, from http://ugandapeople.com/sitemap.htm UgandaWetlands, UBOS, ILRI, & WRI. (2009). Mapping a Better Future: How Spatial Analysis can Benefit Wetlands and Reduce Poverty in Uganda. Wetlands (pp. 1–39). Kampala, Uganda and Washington, DC, USA. Unwin, T. (2010). ICTs, Citizens, and the State: Moral Philosophy and Development Practices. The Electronic Journal on Information Systems in Developing Countries - EJISDC, 44(1), 1–16. Retrieved from http://www.ejisdc.org/ojs2/index.php/ejisdc/article/viewFile/744/337 547 Usery, L. E., & Mcmaster, R. B. (2005). Introduction to the UCGIS Research Agenda. In R. B. McMaster & L. E. Usery (Eds.), A Research Agenda for Geographic Information Science (pp. 1–16). Boca Raton, FL, USA: CRC Press. Ushahidi. (2012). Ushahidi. Ushahidi. Retrieved August 19, 2012, from http://ushahidi.com/ Uwayezu, E. (2010). Evaluating Trends of Spatial Data Sharing Policies in Uganda. GSDI 12 World Conference, Realising Spatially Enabled Societies, 19-22 Oct, 2010, Singapore (p. 16). Singapore: GSDI. Valente, T. W., & Rogers, E. M. (1995). The Origins and Development of the Diffusion of Innovations Paradigm as an Example of Scientific Growth. Science Communication, 16(3), 242–273. doi:10.1177/1075547095016003002 Van Loenen, B., & Kok, B. C. (2004). National Spatial Data Infrastructure in the Netherlands: Legal & Economic Issues and Developments. In B. Van Loenen & B. C. Kok (Eds.), Spatial Data Infrastructure and Policy Development in Europe and the United States (pp. 71–86). Delft, The Netherlands: Delft University Press. Vancauwenberghe, G., Crompvoets, J., & Vandenbroucke, D. (2011). Social Network Analysis of the SDI in Flanders. In Z. Nedović-Budić, J. Crompvoets, & Y. Georgiadou (Eds.), Spatial Data Infrastructures in Context: North and South (pp. 121–135). Boca Raton, FL, USA: CRC Press. Venkatesh, V. (2000). Determinants of Perceived Ease of Use : Integrating Control, Intrinsic Motivation, and Emotion into the Technology Acceptance Model. Information Systems Research, 11(4), 342–365. Venkatesh, V., & Bala, H. (2008). Technology Acceptance Model 3 and a Research Agenda on Interventions. Decision Sciences, 39(2), 273–315. Venkatesh, V., & Davis, F. D. (2011). A Theoretical Extension of the Technology Acceptance Model: Four Longitudinal Field Studies. Management Science, 46(2), 186–204. Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User Acceptance of Information Technology: Towards a Unified View. MIS Quarterly, 27(3), 425–478. Volman, Y. (2004). The Reuse of Public Sector Information in the EU: Increasing the Scope for Cross-Border Exploitation. In B. Van Loenen & B. C. Kok (Eds.), Spatial Data Infrastructure and Policy Development in Europe and the United States (pp. 59–70). Delft, The Netherlands: Delft University Press. Wac, K., & Ragia, L. (2008). LSPEnv: Location-based Service Provider for Environmental Data. Journal of Location Based Services, 2(4), 287–302. doi:10.1080/17489720802612710 Warf, B. (2011). Myths , Realities , and Lessons of the Arab Spring. The Arab World Geographer, 14(2), 166–168. 548 WaridTelecom. (2012). Mobile Data. Warid Telecom: Mobile Data. Retrieved August 13, 2012, from http://www.waridtel.co.ug/index.php/personal/mobile-data Wegener, M., & Junius, H. (1993). “Universal” GIS Versus National Land Information Traditions: Software Imperialism or Endogenous Developments. In Ian Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (pp. 213–228). Dordrecht, The Netherlands: Kluwer Academic Publishers. doi:0-7923-2190-1 Weiner, D., & Harris, T. M. (1999). Community-Integrated GIS for Land Reform in South Africa. GISOC’99: An International Conference on Geographic Information and Society, The University of Minnesota, Minneapolis, MN, 20-22, 1999 (pp. 1–24). Minneapolis, USA. Wicander, G. (2010). M4D Overview 1.0 - the 2009 Introduction to Mobile for Development. World Wide Web Internet And Web Information Systems. Karstad, Sweden. Wiggins, L. L. (1993). Diffusion and USe of Geographical Information Systems in Public Sector Agencies. In Ian Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (pp. 147–163). Dordrecht, The Netherlands: Kluwer Academic Publishers. Williamson, I., Rajabifard, A., & Binns, A. (2007). The Role of Spatial Data Infrastructures in Establishing and Enabling Platform for Decision Making in Australia. In H. Onsrud (Ed.), Research and Theory in Advancing Spatial Data Infrastructure Concepts (pp. 121–132). Redlands, CA, USA: ESRI Press. Willoughby, K. W. (1990). Technology Choice: A Critique of the Appropriate Technology Movement (p. 350). Boulder, CO, USA: Westview Press, Inc. Wise, S., & Burnhill, P. (1991). GIS: Models of Use and Implications for Service Delivery on Higher Education Computing Campuses. Cartographica, 28(3 (GIS Education and Training)), 31–38. Woldai, T. (2002). Geospatial Data Infrastructure: The Problem of Developing Metadata for Geoinformation in Africa. Proceedings of the 4th International Conference of the African Association of Remote Sensing of the Environment, AARSE: GeoInformation for Sustainable Development in Africa, 14-18 October, 2002, Abuja, Nigeria (p. 15). Abuja, Nigeria: African Association of Remote Sensing of the Environment (AARSE). Wood, R., & Bandura, A. (1989). Social Cognitive Theory of Organizational Management. The Academy of Management Review, 14(3), 361–384. WorldBank. (2010). World Development Indicators. Google Public Data Explorer. Retrieved November 7, 2011, from http://www.google.com/publicdata/explore?ds=d5bncppjof8f9_&met_y=ny_gdp_mktp_cd &idim=country:UGA&dl=en&hl=en&q=gdp+uganda 549 WorldBank, & InfoDev. (2012). Information and Communications for Development 2012: Maximizing Mobile. 2012 Information and Communications for Development (p. 222). Washington D.C., USA: International Bank for Reconstruction and Development / The World Bank. doi:10.1596/978-0-8213-8991-1 Yapa, L. S. (1991). Is GIS Appropriate Technology? International Journal of Geographical Information Systems, 5(1), 41–58. Yapa, L. S. (1998). Why GIS Needs Postmodern Social Theory, and Vice Versa. In F. D. R. Taylor (Ed.), Policy Issues in Modern Cartography (pp. 249–273). Oxford, UK: ELSEVIER SCIENCE Ltd. Yeomans, M. (2012). The other AOL - Africa Online. The other AOL - Africa Online. Retrieved August 14, 2012, from http://www.hrea.org/lists/huridocs-tech/markup/msg00413.html Yue, L., Zhang, L., Xiaogang, C., & Yingming, Z. (1991). The Establishment and Application of the Geographic Mapping Database by City/County Unit in China. International Journal of Geographical Information Systems, 5(1), 73–84. Zain, Ericsson, & GSMA. (2010). Life Lines at Lake Victoria. Kampala. Retrieved from http://www.ericsson.com/res/thecompany/docs/corporate- responsibility/2010/lake_victoria_project.pdf Zandbergen, P. A. (2012). Comparison of WiFi Positioning on Two Mobile Devices. Journal of Location Based Services, 6(1), 35–50. Zanello, G., & Maassen, P. (2011). Strengthening Citizen Agency and Accountability Through ICT. Public Management Review, 13(3), 363–382. doi:10.1080/14719037.2011.553265 Zook, M., Graham, M., Shelton, T., & Gorman, S. (2010). Volunteered Geographic Information and Crowdsourcing Disaster Relief: A Case Study of the Haitian Earthquake. World Medical & Health Policy, 2(2), 7. doi:10.2202/1948-4682.1069 Zwart, P. R. (1993). Embodied GIS - A Concept for GIS Diffusion. In Ian Masser & H. J. Onsrud (Eds.), Diffusion and Use of Geographic Information Technologies (pp. 195–204). Dordrecht, The Netherlands: Kluwer Academic Publishers. iWayAfrica. (2012). About Us: The iWayAfrica Story. iWayAfrica: The iWayAfrica Story. Retrieved August 14, 2012, from http://iwayafrica.com/about-us/about-us.jsp 550 Appendix 551 Appendix A: Institutions Interviewed Number Name Sector 1 Department of Surveying, Faculty of Technology, Makerere University Academic 2 Makerere University Institute of Environment and Natural Resources (MUIENR) Academic 3 Department of Information Systems, Faculty of Computing and Informatics Technology, Makerere University Academic 4 Faculty of Forestry, Makerere University Academic 5 Department of Geography, Faculty of Arts, Makerere University Academic 6 Institute of Computer Science, Mbarara University Academic 7 Department of Computer Science, Kyambogo University Academic 8 Department of Computer Science, Gulu University Academic 9 US Agency for International Development (USAID) IO 10 Infectious Disease Institute (IDI) IO 11 Center for Disease Control (CDC); Uganda Virus Research Institute (UVRI) IO 12 Office for the coordination of Humanitarian Affairs (OCHA), Gulu Office IO 13 Spring Project, US Agency for International Development (USAID) IO 14 Worldwide Fund for Nature (WWF) Uganda IO 15 UNOCHA IO 16 United Nations Office for the Coordination of Humanitarian Affairs (UNOCHA) IO 17 ACODE NGO 18 FXB Uganda Program NGO 19 Uganda Land Alliance NGO 20 Famine Early Warning System Network (FEWSNET) NGO 21 National Association of Professional Environmentalists (NAPE) NGO 22 Northern Uganda Malaria, Aids and Tuberculosis (NUMAT) Project, US Agency for International Development (USAID) NGO 23 Nature Uganda NGO 24 MAP (U) Ltd Private 25 BIN-IT Services Ltd Private 552 26 GeoInformation Communication (GIC) Private 27 Umeme (Electricity Utility Company) Private 28 Mobile Telecom Network (MTN Uganda) Private 29 TourGuide Publications, an imprint of Fountain Publishers Limited Private 30 Nokia Siemens Private 31 WE-Consult: Water, Environment and GeoServices Private 32 Mitland (U) Ltd Private 33 Tullow Oil Uganda Private 34 GeoMaps Africa Ltd Private 35 Wetlands Management Department, Ministry of Water and Environment Public 36 Department of Surveys and Mapping (DSM), Ministry of Lands, Housing and Urban Development: Cartography and Surveying Sections Public 37 Uganda National Health Research Organization (UNHRO) Public 38 National Environmental Management Authority (NEMA), previously: National Environment Information Center (NEIC) Public 39 National Forestry Authority (NFA), Ministry of Water and Environment Public 40 Uganda Bureau of Statistics (UBOS), Ministry of Finance Public 41 Northern Uganda Data Center (NUDC) Public 42 Uganda National Roads Authority (UNRA) Public 43 Uganda Wildlife Authority (UWA) Public 44 Computer Aided Mapping Project Uganda Survey (CAMPUS), Department of Surveys and Mapping, Ministry of Lands, Housing and Urban Development Public 45 Ministry of Agriculture, Animal Industries and Fisheries: Cartography Office Public 46 National Agricultural Research Laboratories/Organization (NARL/NARO) Public 47 Department of Geological Survey and Mines, Ministry of Energy and Mineral Development Public 48 Department of Petroleum Exploration and Production, Ministry of Energy and Mineral Development Public 49 National Planning Authority, Ministry of Finance Public 50 Central Police Station, Kampala, Ministry of Internal Affairs Public 51 Kampala City Council (KCC), Ministry of Local Government Public 553 52 Mbarara Police, Ministry of Internal Affairs Public 53 National Water and Sewerage Corporation (NW&SC); Kampala Water Ltd Public/Private Note:  IO: International Organization  NGO: Non-governmental organization 554 Appendix B: Interview Questions - Academic Sector 1. Name of academic institution __________________________________________________________________ 2. Location of institution: City/Town/Locality __________________________________________________ 3. Respondent information: a. Name ______________________________________________ b. Position/Post _______________________________________________ c. Department ________________________________________________ d. Job description _____________________________________________________ 4. Is GIS used or applied at your institution? _______________________ a. Which GIS software is used? _____________________________ b. Who decides on the type of GIS software used? ______________________________________________________ c. What criteria affect the selection of GIS software? ________________________________________________ d. How is the software licensed? _________________________________________ Is it one of the following?:  Proprietary _____________________________________  Who is your supplier of GIS software and GIS business partner? ________________________________________________ ______  Cost of license? ________________________________________________ __  Have you ever had contact with the ESRI East Africa business partner in Nairobi, Kenya? ________________________________________________ _____  Open-Source ______________________________________________________  Where did you obtain the open-source software? ________________________________________________ _____  ERDAS _________________________________ 5. In which year was GIS first introduced at your institution? ___________________ 6. What courses did GIS replace or complement when it was introduced? _________________________________________________ 555 7. Are GIS courses taught in your Department? ________________________________________________________ 8. At what level: undergraduate, masters, or PhD? ______________________________________________________ 9. Course names and credit hours: __________________________________________________________ 10. What is the enrollment for a given GIS course in a semester in your department? _____________________________________ 11. How many faculty members use/teach GIS in their course offerings? 4 - Dr. Musinguzi, Dr. Gidudu, Otukei, Lydia (did Master’s at ITC in Netherlands, and is now doing her PhD in Sweden). 12. To what extent is GIS important in a student’s academic experience? ____________________________________________ 13. How is GIS related to the development of Information and Communication Technology infrastructure at this institution? ______________________________________________________ 14. What problems have you experienced in the implementation of GIS at your institution? ______________________ 15. Would you say that the availability of GIS data in Uganda is a problem? _________________________ 16. How can the use of GIS and Mapping be further encouraged in this academic institution? _______________ 17. What impact has GIS had on planning and environmental monitoring in this city/town? __________ 18. To what extent has GIS penetrated into the institutional framework of the city/town? ___________ 19. To what extent do public and private institutions include the views of the local people (society) when making spatial decisions based on GIS? _________________________________ 20. What is the relationship between cartography, land tenure, land surveying and GIS in Uganda? ____________________________________ 21. How can the use of GIS by public and private sector institutions be promoted in Uganda? _______________________ 22. How can public participation in GIS decision making be increased in Uganda? ___________________ 23. In light of the fact that mobile devices are becoming the most ubiquitous computing platform in developing countries, what are your thoughts on the use of location based services to encourage public participation in GIS based decision making? So you think we can leverage mobile phones for getting feedback from the public, or sending messages to them; do you think Uganda is ready for this sort of a move? Because everybody has a mobile phone, however, not everyone has a laptop… ____________________________________________________________ 556 24. Any other comments on the subject of GIS in Uganda and its impact on society? ___________________ 557 Appendix C: Interview Questions – Public, Private, NGO, IO Sectors 1. Name of institution or organization ___________________ 2. Location of institution: City/Town/Locality _________________________ 3. Respondent information: a. Position/Post _________ b. Department ____________ c. Job description ____________________ 4. Is GIS used or applied at your institution? ______________________________________ a. Which GIS software is used? ____________________________________________________________ ______ b. Who decides on the type of GIS software used? ____________________________________________________________ ______ c. What criteria affect the selection of GIS software? ____________________________________________________________ ____________________________________________________________ ____________________________________________________________ __________________ d. How is the software licensed? Is it one of the following:  Proprietary __________________________________________________  Who is your supplier of GIS software and GIS business partner? ________________________________________________ ______  Have you ever had contact with the ESRI East Africa business partner in Nairobi, Kenya? ________________________________________________ ______  Open-Source ______________________________________________________ ______  Where did you obtain the open-source software? ________________________________________________ _____ 5. In which year was GIS first introduced at your institution? __________________________________________________________________ 6. What methods or technologies were used for spatial decision-making before the introduction of GIS? __________________________________________________________________ 558 __________________________________________________________________ ____________ 7. How is GIS used in your institution; specifically for which tasks? __________________________________________________________________ __________________________________________________________________ ____________ 8. Do new employees receive further training in the use of GIS while at your institution? ______________________ a. Where is the training carried out?  On-site ________________________________  Off-site ________________________________ b. Who are the GIS instructors?  Internal employees ________________________________  External instructors ________________________________ c. How long does the training last? _________________________________ d. Do employees take courses or attend workshops/seminars through external (local and foreign) accredited organizations, e.g.  ESRI ______________________________________________________  GeoInformation Communication Ltd __________________________  Makerere University  Which Department? _____________________________________  Other organizations ___________________________________________ 9. To what extent is GIS important in carrying out tasks at your institution? __________________________________________________________________ __________________________________________________________________ ____________ 10. In what ways does GIS help in decision making at your institution? __________________________________________________________________ __________________________________________________________________ ____________ 11. Do the decisions made consider the views of the society that are impacted by your decisions? _______________________________________________________________ 12. How can the participation of local people (society) in GIS spatial decision making be further encouraged? __________________________________________________________________ __________________________________________________________________ ____________ 559 13. Any other comments on the subject of GIS and mobile technologies in Uganda and their impact on society? ________________________________________________________________ 560 Appendix D: Interview Questions – Telecom Industry 1. Name of Telecom Company_______________________________________________________ 2. City/Town/Locality ___________________________________________________________ 3. Respondent information: a. Position/Post______________________________________________________ _______ b. Department________________________________________________________ ______ c. Job description________________________________________________________ _______ 4. When did your company start operation in Uganda? ____________________________________________ a. How many subscribers are on your network? _________________________________ b. How many districts in Uganda (out of 80) does your network cover? _______________ c. Is this only for voice communication or also for data services (mobile Internet)? _______ 5. What types of technologies are used for: a. Voice communication ____________________________________________________ b. Data Services__________________________________________________________ __ c. SMS Services _____________________________________________________________ 6. What percentage of your users uses each type of service? a. Voice _________________________________________________________________ _ b. Data _________________________________________________________________ _ c. SMS _________________________________________________________________ __ 7. Does your telecom company provide Mobile Internet/Data Access to clients? ______________ a. What is the speed of the wireless internet connection ____________________________ b. What is the percentage geographical coverage of your wireless Internet in Uganda; how many districts (out of 80) are covered? _______________________________________ 561 8. Does your current network infrastructure support Location Based Services, and if so, how? ____ 9. Does your network currently support any location-based services in Uganda, and if so please provide examples ______________________________________________________________ 10. What adjustments in infrastructure set up would have to be made for LBS to become a reality in Uganda? _____________________________________________________________________ 11. What other factors could affect the development of LBS in Uganda? ______________________ 12. Do you think that LBS could become a success in Uganda? ________________________ 13. Is spatial decision making an important aspect of your subscribers’ needs, for example, have they ever expressed interest in finding location-specific information based on the location of their devices? ____________________________________________________________ 14. What mobile positioning techniques (network based) can your network support and what interfaces would be available to 3 rd party LBS developers and content providers? ____________ a. Would these be free or at a cost? ______________________________________ 15. What other mobile services does your company provide, for example, UTL’s m-Pesa, MTN’s Mobile Money, Google SMS Search, SMS Trader? ____________________________________ 562 Appendix E: Survey Questionnaire – Mobile Phone Usage Research Question: How do Ugandans use mobile phone technology in a spatial way and what are the implications for location-based services, LBS, in Uganda? 1. Your profession_______________________________________________________________ ___ 2. Age Range (in your 20’s, 30’s, 40’s, etc…) ___________________________________________ 3. Education Level (Primary, Secondary School, Undergraduate, Graduate) ____________________ 4. General location of where you live/work, e.g. City, suburb ________________________________________________________________________ ______ 5. Your home-town/district/village ____________________________________________________ 6. Do you have a mobile phone(s)? ________________How many? _________________________ 7. What Make? (check as many as necessary) □ Nokia □ Samsung □ Sony Ericsson □ Motorola □ Blackberry □ Chinese □ Kabiriti □ Other ____________________________________ 8. What Models, e.g. Nokia N97, Nokia E-67? ___________________________________________ 9. Who is/are your network provider (s) – please check as many as necessary. □ MTN – Mobile Telecom Network □ Orange □ UTL – Uganda Telecom □ Warid □ Zain □ Other_______________ 10. Is your Mobile phone(s) Internet capable? ____________________________________________ 11. How much do you pay for mobile Internet per month? __________________________________ 12. Is mobile Internet affordable? ______________________________________________________ 13. Is your phone GPS (Global Positioning Systems) capable? ________________________________ 14. Have you ever used your mobile phone for any of the following (please tick)? 563 □ Grameen Foundation’s Village Phone via a Call Center? ___________________________ □ MTN Mobile Money Transfer? _______________________________________________ □ MTN/Google SMS Tips? ____________________________________________________ □ MTN/Google SMS Search? __________________________________________________ □ MTN/Google SMS Trader? __________________________________________________ □ Google Trader? ___________________________________________________________ □ Mobile Internet. Which websites? ___________________________________________ □ Mobile application, e.g. LBS? _________________  Name of Application? _________________ □ Any other use of the mobile phone to obtain spatial information – to answer the question: WHERE IS SOMETHING/SOMEONE? Examples include:  SMS to someone asking for directions to some place,  Voice call about the whereabouts of your friends on the weekend,  Use of mobile Internet search to locate a nearby mechanic using Google Maps …  Please list any others here… __________________________________________________________ __________________________________________________________ __________________________________________________________ _____________________ 15. Would you be inclined to use a mobile LBS application that addresses the following problems (please check the box and/or answer Yes/No): □ Allows you to report pot holes and bad city roads in need of repair to the local government? __________________________________________________ □ Alerts your about environmental degradation in your neighborhood, e.g. deforestation, wetland destruction, water and air pollution? __________________________________ □ Allows you to report on encroachment on wetlands, forests, and nature reserves e.g. for human settlement, illegal dumping of waste? _________________________________________ □ Allows you to inquire whether a plot of land is in a designated wetland before buying it? _________________________________________________________________ _______ □ Allows you to inquire ownership of a plot of land prior to committing to buying it? 564 _________________________________________________________________ _______ □ An application that gives you local weather forecasts, especially the availability of rainfall, to help you make better decisions when planting crops? _________________________________________________________________ _______ □ Availability of nearby health and immunization centers near your village? _________________________________________________________________ _______ □ Alerts on an outbreak of a contagious disease, e.g. Cholera, Meningitis? _____________ □ Availability of nearby police posts near your village/town? ________________________ □ Current buying /selling prices of food crops in nearby local markets for your produce? _________________________________________________________________ _______ □ Availability of jobs in your local area? _________________________________________ □ Please list any other way you would be interested in using a mobile phone to answer spatial questions , that is questions about objects in your surrounding e.g., where is my nearest friend, where is my nearest auto mechanic, is this plot of land in a wetland? _________________________________________________________________ _________________________________________________________________ ______________ 16. What is your understanding of the term location-based services (LBS)? ________________________________________________________________________ ______ 17. Do you own one or more of the following: □ A Laptop Computer?  Make/model? _____________________________________________________  Operating System __________________________________________________ □ A desktop computer?  Make/model ______________________________________________________  Operating System __________________________________________________ 18. What do you use your computer for? □ For Work/study purposes □ To play Games □ To access the Internet □ Other, please specify ______________________________________________________ 19. How do you usually access the Internet? □ In an Internet café □ In your office or University Computer Lab 565 □ At home  Who is your service provider? _________________________________________ □ Other, please specify _________________________________________________________________ _________________________________________________________________ ______________ 20. While Surfing the Internet, which of the following websites do you frequently visit? □ Email websites, e.g. Yahoo Mail, Google Mail, Hot Mail □ Facebook □ Google □ Bing □ Google Maps □ Bing Maps □ Environmental Conservation Websites, e.g. NEMA’s website □ News Websites, e.g. CNN, New Vision □ Entertainment Websites, e.g. Gaming Websites □ Technology and Science websites □ Others, please specify _________________________________________________________________ _________________________________________________________________ ______________ 21. Have you ever used Google Earth? ________________________________________________________________________ ______ 566 Appendix F: Social Consent Form The State of GIS in Developing Countries: A GIS and Society Analysis of Uganda and the Potential for Location-Based Services You are invited to be in a research study of the state of Geographical Information Systems and Science in Uganda, with a focus on GIS and Society Issues. You were selected as a possible participant because you work in a private, public or academic sector institution that might or might not use GIS. We ask that you read this form and ask any questions you may have before agreeing to be in the study. This study is being conducted by: Sami Eria, Department of Geography, University of Minnesota – Twin Cities, USA. Background Information The purpose of this study is: to establish the current state of GIS in Uganda from a GIS and Society perspective, and to gauge the appropriateness of this computer technology in developing countries from a technology transfer point of view. In addition, this study also explores the potential for emerging mobile phone technologies, called location-based services, for spatial decision making and spatial awareness creation in Ugandan society. Procedures: If you agree to be in this study, we would ask you to do the following things: Participate in a survey or answer questions in a semi-structured interview. The survey/interview will last approximately 30-45 minutes and will be recorded using a digital voice recorder. Risks and Benefits of being in the Study: The study has the following risk: First, you might not feel comfortable answering certain questions about GIS technology in Uganda; Second, you might not feel comfortable answering questions about how you use your mobile phone in a “spatial” way to support spatial decisions, however, the likelihood of both risks is minimal and has very low psychological impact. The benefits to participation are: society will benefit from the findings of the research in terms of better information regarding the use of Geographical Information Systems and Science, and location-based services, in developing countries to improve the livelihoods of society at large. Compensation: You will receive no payment for participation, although a reward “in kind” might be offered for your time, for example, a soft drink or a cookie! Confidentiality: The records of this study will be kept private. In any sort of report we might publish, we will not include any information that will make it possible to identify a subject. Research records will be stored securely and only researchers will have access to the records. Any 567 tape recordings or video recording will be used for the sole purpose of statistical and social science analysis by the principle investigator and this information will be destroyed after a period of two years. Voluntary Nature of the Study: Participation in this study is voluntary. Your decision whether or not to participate will not affect your current or future relations with the University of Minnesota or with Makerere University Kampala, Uganda; I am collaborating with the Department of Surveying under supervision of Dr. Moses Musinguzi. If you decide to participate, you are free to not answer any question or withdraw at any time without affecting those relationships. Contacts and Questions: The researchers conducting this study are: Sami Eria and Robert B. McMaster, Ph.D. (Advisor from University of Minnesota-Twin Cities, USA). In Uganda, I am collaborating with Dr. Musinguzi Moses, Head, Dept of Surveying, Faculty of Technology, Makerere University, Kampala. You may ask any questions you have now. If you have questions later, you are encouraged to contact them at: - Sami Eria, 414 Social Sciences Building, 267 19th Ave South, Minneapolis, MN 55455, USA; Cell Phone (Uganda) 0784-325-292, email: eriax001@umn.edu. - Robert B. McMaster, 414 Social Sciences Building, 267 19th Ave South, Minneapolis, MN 55455, USA, Tel: 1-612-626-9425, email: mcmaster@umn.edu - Musinguzi Moses, Dept of Surveying, Faculty of Technology, Makerere University, PO Box 7021, Kampala, Uganda; email: musinguzim@tech.mak.ac.ug If you have any questions or concerns regarding this study and would like to talk to someone other than the researcher(s), you are encouraged to contact the Research Subjects’ Advocate Line, D528 Mayo, 420 Delaware St. Southeast, Minneapolis, Minnesota 55455; (612) 625-1650. You will be given a copy of this information to keep for your records. Statement of Consent: I have read the above information. I have asked questions and have received answers. I consent to participate in the study. Signature: _____________________________________________ Date: ___________________ Signature of Investigator:_______________________________________ Date: ___________________ 568 Appendix G: Situating GIS within Information Systems Literature Related to Information Systems is Information technology (IT). The relationship between IS and IT is illustrated in Figure 0-1 below. Figure 0-1: Relationship between IT and IS Source: (Khan, 2010) Information Technology consists of the hardware and software technology needed to build an Information System. An Information System can be broadly defined as: “Any computerized system with a user or operator interface is an information system, provided the computer is not physically embedded” (Ein-Dor & Segev, 1993). In this definition, a distinction is made between systems that function without human intervention, for example, a self-guided missile, from systems that require human 569 intervention, for example, a computer system for automation of office tasks; another is a decision support system. The purpose of an Information System is to convert data into information. This process of conversion is supported by the following four resources (BISS, 2011): People: this consists of IT specialists and end-users Hardware: this consists of machines and media Software: this consists of programs and procedures Data: this consists of form data, models and knowledge base data The end product of an Information System is information products for end users (BISS, 2011). The process of conversion of data into information can be diagrammatically represented as shown in Figure 0-2. Figure 0-2: Process of conversion of data into information in an Information System Source: (BISS, 2011) 570 Data stored in a repository is used as input into an Information System. Processing of the data using software programs and procedures, residing on hardware, and human intervention by IT specialists, converts this data into information products for end-users. Ein-Dor & Segev (1993) identified seventeen different types of information systems in IS literature, classified based on either their attributes (components) or functionality (what they do). Table 0-1: A classification of Information Systems; Source: (Ein-Dor & Segev, 1993) Type/Name Description Year First Mentioned in Literature Early Computation 1946 Early DP Early Data Processing 1956 MIS Management Information Systems 1967 OIS/OA Office Information Systems/Office Automation 1975 Mature DP Mature Data Processing 1976 Manufacturing Robots 1978 DSS Decision Support Systems 1979 MRP Manufacturing Resource Planning 1980 Expert Systems 1985 MRP II Manufacturing Resource Planning II 1988 CAD Computer Aided Design 1988 CAM Computer Aided Manufacturing 1988 C 3 I Command, Control, Communications and Intelligence 1988 EIS Executive Information Systems 1990 GDSS Group Decision Support Systems 1990 CAD/CAM 1990 Scientific Computation ? 571 By the 1980s, a pyramid model was used to capture the main types of Information Systems at the time. This model is illustrated in Figure 0-3. Figure 0-3: The pyramid model categorizing Information Systems in the 1980s Source: Based on (K. C. Laudon & J. P. Laudon, 1988, 44) It is obvious from this model that Geographical Information Systems were not explicitly considered as a special category of Information Systems in the pyramid model, and this could be because they were still relatively an emerging technology by 1988. Another way of looking at it is that what we have come to know as GIS today could be considered synonymous in functionality and character, to some extent, with Decision Support Systems (DSS), which emerged (according to Table 0-1) in about 1979. It is interesting to observe that one would be hard pressed to find mention of “GIS” in the Information Systems literature even by the end of the 20 th Century. 572 Laudon and Laudon (1988) classify Information Systems based on functionality of the systems and the corresponding level within an organization in which that system is used. There are four major types of information systems (see Figure 0-3), 1) Transaction Processing Systems (TPS), 2) Management Information Systems (MIS), 3) Decision Support Systems (DSS), and 4) Executive Support Systems (ESS). I will briefly define each of these types of systems based on definitions provided by Laudon and Laudon (1988). TPS are basic business systems that serve an organization at the basic level. “A TPS is a computerized system that performs and records the daily routine transaction necessary to conduct business. Examples include sales order entry, hotel reservation systems, payroll, employee record keeping, and shipping” (K. C. Laudon & J. P. Laudon, 1988, 44). Management Information Systems “serve the management level of the organization providing managers with reports, and in some cases, with online access to the organization’s current performance and historical records” (K. C. Laudon & J. P. Laudon, 1988, 46). The function of MIS is mainly for planning, controlling, and decision making at the management level, and they rely on the TPS for their data, which is a level lower in the pyramid. MISs are oriented almost exclusively to internal events, not environmental or external. Decision Support Systems also serve the management level of an organization. They help managers make decisions that are unique, rapidly changing and not easily specified in advance. Much as they input information from the TPS and MIS, lower down in the pyramid, they often bring in external sources of data, for example, current stock prices 573 (Laudon & Laudon, 1988). Because they are more of an analytical tool, they are interactive, have more analytical power, and usually include user-friendly software. An example of a DSS may be a voyage-estimating system of a shipping company. This system calculates the financial and technical details of a voyage, and answers questions such as what is the optimum speed of travel of a vessel to ensure timely delivery of goods, and at the same time maximize profits and keep customers happy? (Laudon & Laudon, 1988) Executive Support Systems support senior managers in making decisions. They serve the strategic level of an organization by addressing decisions, which are non-routine, and require judgment, evaluation and insight because there is no single agreed-upon procedure for arriving at a solution (Laudon & Laudon, 1988). They draw upon information from the internal MIS and DSS, then filter and compress this information. There is a particular emphasis on reducing the time and effort required by executives to obtain information from this system, thus these systems employ advanced graphics software that can deliver graphs and data from many sources to the executive’s boardroom. An example of an ESS is a system that provides a summary of activities important to a given business in a graphical display on a desktop computer, for example, a firm’s financial performance measured by working capital, accounts receivable/payable, cash flow and inventory (Laudon & Laudon, 1988). Of these four types of systems, Geographical Information Systems could be equated to Decision Support Systems because of the analytical power of GIS, the interactivity, and complexity of the systems. The only difference between DSS and GIS is that GIS 574 exclusively deal with spatial data, while DSS may or may not have a spatial component. Also, GIS are often times associated with environmental applications and public sector projects, as opposed to DSS which are closely linked to the world of business and management in organizations in the private sector, with a strong drive towards maximizing profits. Laudon and Laudon (1988) categorize GIS as a “special category of DSS that use data visualization technology to analyze and display data for planning and decision making in the form of digitized maps” (K. C. Laudon & J. P. Laudon, 1988, 425). 575 Appendix H: Theories for Understanding the Acceptance and Use of Information Systems 1. Technology Acceptance Model (TAM) The Technology Acceptance Model (TAM) is an Information Systems theory that models the ways in which users come to accept a technology. This model was developed by Fred D. Davis Jr. in 1985 (Davis Jr., 1985). It was the subject of his Ph.D. dissertation submitted to the Sloan School of Management at Massachusetts Institute of Technology. The model is illustrated in Figure 0-4 below. Figure 0-4: Technology Acceptance Model Source: (Davis Jr., 1985, 24) According to this model, a potential user’s overall attitude towards using a given system is a major determinant of whether he/she actually uses it. This user attitude, “Attitude Toward Using” in Figure 0-4, is a function of two major beliefs, 1) Perceived Usefulness 576 and 2) Perceived Ease of Use. “Perceived Ease of Use has a causal effect on Perceived Usefulness” (Davis Jr., 1985, 24). Design features of a system, for instance, an Information System, fall into a category of external variables and do not have a direct effect on attitude or behavior of the system user. Design features are theorized to have only an indirect effect through perceived usefulness and perceived ease of use. TAM has been extended over the years by many researchers to account for several variables that may affect perceived ease of use and perceived usefulness of Information Technologies and Information Systems. Viswanath Venkatesh (2000) proposed an extension of TAM that accounts for how the perception of ease of use forms and changes over time. He proposed three determinants of perceived ease of use including control (computer self-efficacy and facilitating conditions), intrinsic motivation (computer playfulness), and emotion (computer anxiety). These are “anchors” that determine early perceptions about the ease of use of a new system. “With increasing experience, it is expected that system-specific perceived ease of use, while still anchored to the general beliefs regarding computers and computer use, will adjust to reflect objective usability, perceptions of external control specific to the new system environment, and system- specific perceived enjoyment” (Venkatesh, 2000). TAM is a dynamic theory and further extensions to this theory are still being proposed, for example, Venkatesh and Davis (2011) proposed an extended TAM model, called TAM2, that adds two significant influences to technology acceptance and usage: 1) social influence processes (subjective norm, voluntariness and image), and 2) cognitive 577 instrumental processes (job relevance, output quality, result demonstrability and perceived ease of use) (Venkatesh & F. D. Davis, 2011). 2. Social Cognitive Theory (SCT) Social Cognitive Theory is an extension of Social Learning Theory (Bandura, 1976). SCT was formulated by Albert Bandura in 1986 and published as Social Foundations of Thought and Action: A Social Cognitive Theory (Bandura, 1986), a seminal piece of work in the field of Psychology. SCT provides an overview of human cognition within the context of social learning. It is a theory of human motivation and action that analyzes the cognitive, self-regulatory, vicarious, and self-reflective processes in psychosocial functioning (Bandura, 1986). SCT is a Social Learning Theory that describes how behaviors are learned. “SCT emphasizes reciprocal determinism, or the interactive and dynamic processes by which behavioral, personal (cognitive) and environmental factors affect each other, and are affected by each other” (Bandura, 1986). These factors are illustrated in Figure 0-5 below. 578 Figure 0-5: Social Cognitive Theory – triadic reciprocal causation in the causal model of SCT Source: (ReCAPP, 2011) The personal (cognitive) and environmental factors form the constructs of SCT (Bandura, 1986; ReCAPP, 2011), and include (1) psychological determinants of behavior, (2) observational learning, and (3) self-regulation. “Psychological determinants are cognitive factors which influence behaviors. They include outcome expectations, or the perceived value of the consequences associated with the behavior, and the self-efficacy, or the perceived belief of one’s ability to perform a particular behavior, say promoting health” (Bandura 1986, 23). “Observational Learning is the ability to learn a new behavior by exposure to interpersonal or media displays of it, especially through peer modeling” (Bandura 1986, 23). “Environmental determinants are external and physical factors which influence behavior, such as incentive motivation and facilitation. Incentive motivation is the use of rewards or punishment to modify behavior, and facilitation is the process by 579 which tools, resources and environmental changes are used to make new behaviors easier to perform” (Bandura 1986, 24). “Self-regulation is one’s personal ability to control oneself through self-monitoring, goal-setting, feedback, self-reward, self-instruction and enlistment of social support” (Bandura 1986, 24). SCT has been used in various studies in multiple disciplines to understand human behavior with respect to myriad disciplinary foci, for example in public health (Luszczynska & Schwarzer, 1996), management (Wood & Bandura, 1989), psychology (Bandura, 2001b), mass communication (Bandura, 2001a), and Information Systems/Technology (D. Compeau, Higgins, & Huff, 1999). 3. Unified Theory of Acceptance and Use of Technology (UTAUT) A unified theory that marries together several theories on user acceptance and use of technology in general is the Unified Theory of Acceptance and Use of Technology (UTAUT) (Venkatesh et al., 2003). The components of the UTAUT model are illustrated in Figure 0-6, and is the result of the combination of eight theoretical models: 1) The theory of reasoned action, 2) the technology acceptance model, 3) the motivational model, 4) the theory of planned behavior, 5) a model combining the technology acceptance model and the theory of planned behavior, 6) the model of PC utilization, 7) the diffusion of innovations theory, and 8) the social cognitive theory. 580 Figure 0-6: Universal Theory on Acceptance and Use of Technology Source: (Venkatesh et al., 2003, 447) Venkatesh and colleagues (2003) note that the basic conceptual framework underlying most user acceptance models is as shown in Figure 0-7. Figure 0-7: Basic Conceptual framework underlying user acceptance models Source: (Venkatesh et al., 2003, 427) In this conceptual framework (Figure 0-7), actual use of Information Technology (IT) by an individual is affected directly by the intentions to use IT, which in turn depends on individual reactions to using IT. 581 In formulating UTAUT, Venkatesh and others (Venkatesh et al., 2003) theorized that in the eight models of acceptance and usage of technology (mentioned earlier), there are between two and seven determinants of acceptance, for example, in one of the models, TAM, these include 1) perceived usefulness, and 2) perceived ease of use. For a full list and comparison of these determinants, readers are referred to the MIS Quarterly article by Venkatesh and others (2003). The researchers compared these determinants and reduced them to four key moderating variables that are significant throughout the eight theoretical models (see Figure 0-6): 1) experience, 2) voluntariness of use, 3) gender and 4) age. Further, the researchers’ proposed that there are four constructs that “play a significant role as direct determinants of user acceptance and usage behavior: 1) performance expectancy, 2) effort expectancy, 3) social influence, and 4) facilitating conditions” (Venkatesh et al., 2003, 447). The researchers theorized that attitude towards using technology, self-efficacy, and anxiety are not direct determinants of intention. 582 Appendix I: Theories for Understanding the Diffusion of Technology 1. Technological Determinism Technological determinism is a theory that hinges on the premise that innovations will diffuse if they are technologically more advanced than others (H. J. Campbell, 1996). The technological aspects of an innovation are solely responsible for its adoption. Another way of looking at it is that technology is the driving force of history (Marx & Smith, 2011). Technology is seen as a “crucial agent of change” (Marx & Smith, 2011, ix), and holds an important and special place in the culture of a modern society. Indeed, technology has shaped history; the invention of machines for factory production led to the agricultural and industrial revolutions in Europe, for instance, at different times in the history of that continent. So, all that is needed for a society to adopt an innovation is for a technology to be as advanced as possible. With respect to the adoption of GIS technology in the developing world, this theoretical perspective suggests that if “more powerful and user-friendly GIS applications” (Campbell, 1996, 27) are created, they will automatically be adopted by society. However, proponents of this line of argument would find it hard to explain why the QWERTY keyboard was adopted instead of the more efficient and user-friendly Dvorak keyboard (see Rogers (2003, 8) for an interesting discussion of the “nondiffusion of the Dvorak keyboard”). Also, some have questioned the appropriateness of GIS as a technology for developing countries (for example, see Yapa, 1991). Drawing upon the work of Schumacher (1973), Yapa (1991) argues that GIS is not appropriate technology 583 for developing countries because of the high costs associated with acquiring, and maintaining the technology. For a history of the appropriate technology movement, see Pursell (1993). 2. Economic Determinism Economic determinism argues that technology drives economic development, and that computerization is the most recent wave, the so called third wave, of technology to influence the path of global economic development (H. J. Campbell, 1996). Campbell (1996) draws upon the work of Toffler (1980) to explain that the first two phases of technological development are, one, the agricultural revolution, and two, the industrial revolution. This led to a transition from agricultural to industrial societies. Information technology, which is the driver of the third wave (Toffler, 1980) related to the diffusion of computational technology, is the foundation for a more prosperous society (H. J. Campbell, 1996). Thus, from the economic determinist perspective, there is a need for the private and public sector to adopt information technology if these institutions are to be economically viable, especially with increasing scale of operations. There is a strong link between technological advancement and economic progress in economic determinism. Skeptics have pointed out that despite the obvious rationale that capitalist forms of production and business have a lot to gain from the benefits provided by automation through information technology, questions need to be raised regarding, among other things, the “progressive subordination of the working class as jobs are both simplified and made more tedious, whilst surveillance and control over individual employees increases” (Campbell, 1996, 29; Noble, 1984). Downsides of economic determinism (and 584 the related notion of technological determinism) according to critics include, 1) over- simplification due to a one dimensional view of the world, 2) questions about “prosperity for who, mass society or just a few?”, 3) ignorance about the institutional context within which technology determines economic outcomes, and 4) an extreme focus on equipment and machinery instead of society (H. J. Campbell, 1996). 3. Social Interactionism The Social Interactionism perspective states that technology is socially constructed and, thus, “the diffusion of innovations is therefore the interaction between the technology and potential users within particular cultural and organizational contexts” (Campbell, 1996, 31). It follows that the simplistic arguments of technological and economic determinism do not hold in the social interactionist perspective on technology diffusion mainly because of the complexity inherent in social interactions, not to mention the messiness. Societal interaction with technology is problematic, and no claims are made that computers will be automatically adopted by society, or how beneficial such an outcome will be for society. Emphasis is placed on social values and norms leading to adoption of a technology, as opposed to the technical superiority of a technological innovation. Social values, norms and politics cannot be divorced from the eventual adoption of a technology, because these all play a role in the perception of the utility of a given technology. “For the social interactionist approach questions of ethics and values are not external to the technology but inherent within the system. Consequently, the diffusion of innovations is dependent upon the interaction between the technology and the setting in which it is located” (Campbell, 1996, 32). Thus, the implementation of the technology 585 adopted by society is dependent upon context within which such an innovation diffused into that particular society (H. J. Campbell, 1996). 586 Appendix J: Philosophical Perspectives on Research 1. Ontology, Epistemology, Methodology and Method Epistemology is defined as “the theory of knowledge, the inquiry into its possibility, nature and structure” (Perry & Bratman, 1999, 111, 806). It can also be defined as “the study of knowledge and justified belief” (Steup, 2011). Some of the questions it addresses include: “What are the necessary and sufficient conditions of knowledge? What are its sources? What is its structure, and what are its limits?” (Steup, 2011) “How do we know the world? What is the relationship between the inquirer and the known?” (Denzin & Lincoln, 1994, 99) Some have treated epistemology and method as synonymous terms (Howe, 1992, in R. B. Johnson et al., 2007). Ontology is a branch of metaphysics. It is the study of “what there is” (Hofweber, 2011). Questions in ontology address whether or not certain things, broadly called entities, exist, for example, is there a God? (Hofweber, 2011) “Ontology raises basic questions about the nature of reality” (Denzin & Lincoln, 1994, 99). If epistemology is the science of knowing, then methodology could be defined as “the science of finding out” (Babbie, 2007, 4). Another definition of the term methodology is “the particular ways of knowing a reality” (Sale, Lohfeld, & Brazil, 2002, 44). “Methodology focuses on how we gain knowledge about the world” (Denzin & Lincoln, 1994, 99). Social science methodology can be thought of as the ways in which social scientists find out about human social life, which is usually done based on a predisposition of humans to 587 use causal and probabilistic reasoning. Patterns of cause and effect are usually probabilistic in nature. It is the case that science makes concepts of causality and probability more explicit and provides techniques for dealing with them more rigorously than causal human inquiry does (Babbie, 2007). Method can be defined as the specific ways in which an investigation is undertaken within the framework of a given methodology. Schwandt (1994, 119) defines method as “the techniques for gathering and analyzing data.” Method is more specific than methodology; it refers to the actual steps taken to collect store and analyze data, for example, use of questionnaires, or telephone surveys, or personal interviews, or ethnographic studies. As mentioned previously, there are three research paradigms: qualitative, quantitative and mixed methods (R Burke Johnson & Onwuegbuzie, 2004). Each of these research paradigms are defined in turn by paradigms in philosophy such as positivism, postpositivism, constructivism, interpretivism and critical theory (Denzin & Lincoln 1994). In addition, there are also related perspectives alongside these paradigms that provide different lenses through which each of the three types of research paradigms are viewed and conceptualized, such as feminism, ethnic models of inquiry and cultural studies (Denzin & Lincoln 1994). I use the term paradigms loosely in this chapter to refer to both the “research paradigms” and “philosophical paradigms” in which the latter defines the former. I discuss these research and philosophical paradigms next. 588 2. Constructivist, Critical Theory, and Postmodernist Perspectives “Constructivism adopts a relativist (relativism) ontology, a transactional epistemology, and a hermeneutic, dialectical methodology. The inquiry aims of this paradigm are oriented to the production of reconstructed understandings, wherein the traditional positivist criteria of internal and external validity are replaced by the terms trustworthiness and authenticity” (Denzin & Lincoln, 1994, 100). Constructivist perspectives are opposed to positivism (explained later under the quantitative research paradigm), and are committed to the study of the world from the point of view of the interacting individual. Hence, constructivist approaches are idiographic in nature, as opposed to positivist approaches, which are nomothetic (Denzin & Lincoln 1994; Schwandt 1994). The Critical theory paradigm can be articulated as having an ontology based on historical realism, an epistemology that is transactional, and a methodology that is dialogic and dialectical (Denzin & Lincoln, 1994). It sprung out of the philosophies and social thought nurtured by German philosophers Karl Marx, Immanuel Kant, Georg Wilhelm Friedrich Hegel, and Karl Emil Maximilian (Max) Weber (Kincheloe & McLaren, 1994). These philosophies were employed at the Frankfurt School in Germany by Max Horkheimer, Theodor Adorno, and Herbert Marcuse during, and especially after World War I, to shape critical theory as a paradigm. Marcusian thought was adopted by social theorists in the United States in the 1960s as a philosophical foundation for leftist politics, a platform for a cultural and psychological revolution, and for political emancipation (Kincheloe & McLaren, 1994). 589 More recently, in the 1970s, a form of social theory developed that rejects the notion that a “reality” actually exists (Kincheloe & McLaren, 1994). This is called postmodern social theory. It can be regarded as a type of critical theory that succeeds modernism, the latter’s conviction being that it accepts the diversity of views about reality among different societies as legitimate; in essence acknowledging that none of the views is right or wrong, and that different people have different ideas (Babbie, 2007). According to postmodernism, in contrast, what different societies believe, diverse as they may be, are all correct because reality is only an image in the mind of the observer, and this image is subjective depending on the circumstances and contexts under which different observers view the world (Babbie, 2007). “Whereas the modern view acknowledges the inevitability of human subjectivity, the postmodern view suggests there is actually no ‘objective’ reality to be observed in the first place. There are only our several subjective views” (Babbie 2007, 10). In postmodern social theory, the view of is that “reality is socially constructed or semiotically posited” (Kincheloe and McLaren 1994, 143). 3. Interpretive Perspectives, Symbolic Interactionism Interpretivist approaches to research are an alternative to logical empiricist epistemology (Schwandt, 1994). Proponents of the interpretivist (and also, constructivist) persuasions “share the goal of understanding the complex world of lived experience from the point of view of those who live it” (Schwandt 1994, 118). In the interpretivist perspective, the world of lived reality and situation-specific meanings that constitute the general object of investigation is thought to be constructed by social actors – particular actors, in particular places, at particular times (Schwandt, 1994). Meanings are derived from events and 590 phenomena in the actors’ lives through prolonged and complex processes of social interaction involving history, language and action (Schwandt, 1994). Thus, the interpretivist (and constructivist) believes that to understand this world, one must interpret it. “The inquirer must elucidate the process of meaning construction and clarify what and how meanings are embodied in the language and action of social actors. To prepare an interpretation is itself to construct a reading of these meanings; it is to offer the inquirer’s construction of the constructions of the actors one studies” (Schwandt 1994, 118). These approaches to thinking are more concerned with matters of knowing and being, and do not have very clearly spelled out methods, and specifically deemphasized in their epistemological justifications. Methods might include participant observation, informant interviewing, archival research; however, these are only a “means to an end.” More importantly, all interpretive inquiries watch, listen, ask, record, and examine (Schwandt, 1994). An example of the interpretivist persuasion is symbolic Interactionism (Schwandt, 1994) used and developed intellectually by George Herbert Mead and John Dewey (Blumer, 1969). Symbolic Interactionism rests of three simple premises, according to Blumer (1969). First, human beings act towards things on the basis of the meanings that the things (e.g. trees, chairs, institutions, friends) have for them. Second, the meaning of such things is derived from, or arises out of, the social interaction that one has with one’s fellows. Third, these meanings are handled in, and modified through, an interpretive process used by the person in dealing with the things he encounters. 591 4. Positivism The French philosopher Auguste Comte (1998-1857) coined the term sociologie in 1822, a philosophy that proposed that society was a phenomenon that could be studied scientifically (Babbie, 2007). Before Comte’s time, the understanding of society had gone through two major stages of history, one, the theological stage, based on God and religion, up to 1300 A.D., followed by a metaphysical stage that replaced God with philosophical ideas such as “nature” and “natural law” (Babbie, 2007). Comte’s “positive philosophy” was the start of the third stage in which Science would replace religion and metaphysics by basing knowledge on observations through the five senses rather than on belief or logic alone (Babbie, 2007). According to Comte, society could be observed, after which it could be explained logically and rationally. He offered that sociology could be as scientific as biology or physics. “In his optimism for the future, he coined the term positivism to describe this scientific approach, in contrast to what he regarded as negative elements in the Enlightenment” (Babbie 2007, 34). It is only until recently that positivism has received some serious criticism from scholars. Take for example, positivism is based on an epistemology that humans always act rationally, for example, models in economics assume that humans will always choose the highest-paying jobs, pay the lowest price and so on (Babbie, 2007). However, this assumption ignores the power of tradition, loyalty, image, and other factors that compete with reason, calculation and rationality in determining human behavior (Babbie, 2007). 592 5. Postpositivism Postpositivism is a critique of positivism (Phillips & Burbules, 2000), and not a total rejection of the original philosophy of positivism first theorized by Auguste Comte (1851). The most well-known critics of positivism were John Dewey (1938) and Karl Popper (1959). According to postpositivism, there is no absolute truth, which is opposition to the doctrine of positivism. It is not meant to be in direct opposition, though, as postpositivists believe that the seeking the truth is an on-going process, and does not really have an end. For instance, knowledge about certain diseases one hundred and fifty years ago in the field of medicine has changed in comparison to what we now call the truth about those same diseases. Thus, there is no need for a commitment to “absolute” truth or its attainability. Instead, Dewey suggested that (Phillips & Burbules, 2000) such a term (“truth”) be replace with the term “warranted assertibility” to emphasize the point that “absolute” truth can never be obtained by human beings (Phillips & Burbules, 2000; Popper, 1959).