EJISDC (2009) 38, 9, 1-18

APPROPRIATE ICT AS A TOOL TO INCREASE EFFECTIVENESS IN ICT4D: THEORETICAL CONSIDERATIONS AND ILLUSTRATING CASES

Victor van Reijswoud

Divine Word University Papua New Guinea

vvanreijswoud@dwu.ac.pg ABSTRACT The need to bridge the digital divide is no longer a point of discussion and therefore focus has shifted to the design and implementation of programs that have the potential to close the information and knowledge gap between the developing and developed nations. Unfortunately, the majority of these programs are small and mimic what has been successful in the developed world. It has become increasingly clear that these successes do not necessarily translate well in the context of developing nations. This paper develops the hypothesis that Information and Communication Technology (ICT) projects in developing countries will become successful only when they are adapted to local conditions. The general concept of Appropriate Technology (AT) will be explored for the field of ICT. AT has already been embraced by fields like architecture, building technology and agriculture, but has not yet been rooted in ICT.

The paper proposes a preliminary theory of Appropriate ICT along the lines of existing theories in AT and System development. The theory identifies Appropriate Technology principles at three levels: hardware, software and ICT change management. By means of real life mini cases in the ICT for Development context in Africa, the guiding principles for Appropriate ICT are illustrated. The paper will conclude with an agenda for further research in the three identified levels. The research agenda targets academia, governments, NGO's and industry. Keywords: ICT for Development, hardware design, software design, Africa, appropriate technology 1. INTRODUCTION It sounds pretty normal: when you plan a mountain hike you ensure that you wear strong boots and a pullover against the cold at higher altitudes; in case you go to the tropics you choose a light, well ventilated tropical out-fit and a hat or cap against the merciless sun. You have been taught that you need to adapt to the local circumstances. In disciplines such as architecture, civil technique and industrial design, identifying, selecting and introducing appropriate and suitable technology is well recognized, but in the field of ICT (being a young discipline) this process is still in its infancy.

Not only computer hardware and software, but also methods and techniques for design and implementation of information technology, are almost without exception invented/developed in the developed countries (Europe and North America). The contextual and cultural elements of these countries are ingrained in the design. These elements limit the transferability of the technology to other, different, environments (Collins, 1992; Evans & Collins, 2007). Tacit assumptions become clear in case of breakdown of operations (Winograd & Flores, 1986) and will form the start of an explanatory of problem solving-discussion or discourse (Habermas, 1985). In the field of ICT for Development (ICT4D), this discussion on the limitations of commercial-off-the-shelf (COTS) ICT tools, softwares and methodologies in the context of developing countries has been initiated (Gurstein, 2003, Gairola et al., 2004, Dymond & Oestmann, 2004, Reijswoud & Topi, 2005) but is still young.

The field of ICT4D has grown dramatically in size and importance over the past decade

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 2

(McNamara, 2003, Levey & Young 2002). ICT4D is based on the premise that ICT is able to bridge the digital divide between the developed and developing countries and thereby contribute to equal distribution of wealth. Two dimensions are identified to achieve this goal: increasing access to ICT and rationalization of work procedures to increase transparency and accountability (Krishna & Madon, 2003). ICT is considered to be vital for the improvement of governance and production resources (Sciadas, 2003).

The importance of ICT for poverty alleviation has been recognized at the highest international levels when the United Nations Development Programme (UNDP) dedicated their Annual Human Development Report to the role of information and communication technologies (UNDP, 2001). At present, most of the large development organizations have programs with a substantial ICT component and a growing number of smaller development initiatives have started targeted projects in the field of ICT.

In spite of all the efforts, the digital divide has not been bridged and well-documented success stories of the application of ICT for poverty alleviation are hard to find (Krishna & Madon, 2003, Curtain, 2004, Osama, 2006, Walsham et al., 2007). Evaluation of ICT projects often reveals underutilization of the resources because the newly introduced ICT has not been well integrated within the local context (Kozma, 2005), in the worst case as a result of 'dump-and-run' approaches (Volsoo, 2006, Reijswoud et al., 2005), and lack of local ownership in the receiving communities (Vaughan, 2006). Also technical (hardware and software) problems resulting from the 'hostile' conditions in which the ICT was introduced put a strain on the actual impact (Gichoya, 2005). High rates of hardware breakdown combined with the low locally available technical problem-solving skills have lead to underutilized and even abandoned projects. Finally, high maintenance (recurring) costs for hardware, software and internet connectivity put a financial burden on the projects rendering them financially non-sustainable.

In spite of the enormous challenges in the context of the development of especially the Least Developed Countries (LDCs), the academic literature to date on ICT for Development is relatively sparse (Walsham et al., 2007). Most of the literature is produced in the public domain by development agencies. Furthermore, little attention is paid to general frameworks to improve the success of ICT implementations in LDCs.

There are many reasons that ICT projects in LDCs fail (Heeks, 2003) and they have been reported from the start (Moussa & Schware, 1992, World Bank, 1993). Failure may be caused by selection of inappropriate hardware, software and/or design and implementation approaches. This article starts from the premise that many ICT projects in LDCs fail to properly take into account the local context in LDCs. Building on this premise we will develop a theory for the design and implementation of ICT projects in LDCs that takes into account local conditions. We develop this approach along the lines of existing theories in AT in other fields of science and general theories in information systems design.

It is apparent that ICT projects cannot be adequately understood and addressed as technical/rational initiatives (Avgerou, 2003) and therefore, like other technical solutions, the design and implementation of ICT solutions must be carried out in relation to the culture (Westrup et al., 2003), the environment, the organization, the available resources, the economic and political circumstances, and the desired impact (Avgerou, 2003). As Avegrou states (2003: 57-58) there is need for a situated approach where IT innovation is understood in ‘their’ complex context. We propagate an integration of the discipline 'Appropriate Technology' that aims at devising 'suitable' technological solutions. Our theory identifies principles to do so at three levels: hardware, software and ICT change management. The theory is described in section 2 and 3. By means of real life mini cases in the ICT for Development context in Africa, the guiding principles for Appropriate ICT are illustrated in section 4. In section 5 we will conclude with an agenda for further research.

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 3

2. APPROPRIATE TECHNOLOGY In order to understand better how we can improve the effectiveness of design and implementation of ICT projects in the LDCs we will explore the field of appropriate technology (AT). Since the concepts of AT has not yet gained much ground in the area of ICT, we will start our exploration with a short discussion of AT.

As a general definition we adopt the idea that AT is technology that is suitable for the environmental, cultural and economic conditions in which the technology is intended to be used. The opposite of AT is the 'one-size-fits-all' concept that builds on the premise that well designed technology can be used under all circumstances – the universal model. The best example of this line of thinking is the Swiss Army Knife. The Swiss Army Knife is designed with the idea that it will help you out in whatever environmental, cultural and economic condition you are in. Avgerou and Walsham (2000) characterize these two attitudes as paralyzing anti-technology and techno-enthusiasm.

Darrow and Saxenian (1986) provide 10 criteria in the Source Book for Appropriate Technology that we take as starting point. These criteria have been formulated to act as a basic set of guidelines for a broad spectrum of several technologies in developing countries. ICT is not considered explicitly by the authors. The following criteria are proposed in Darrow and Saxenian (1986):

1. It should be possible to implement/realize technological solutions with limited financial

resources. 2. The use of available resources must be emphasized to reduce the costs and to guarantee the

supply of resources e.g., for maintenance 3. Technologies may be relatively labor-intensive, but must have a higher output than the

traditional technologies. 4. The technology must be understandable for people without specific or academic training 5. Small rural communities should be able to produce and maintain the technology 6. The technology must result into economic and/or social progress. 7. The technology must be fully understandable for the local population, the end-users

resulting into possibilities for them to become involved in the possible innovation and extension of the use of the technology

8. The technological solutions must be flexible and easily to be adapted to changing circumstances.

9. The technology must contribute to the increase of productivity 10. The technology should not have a negative impact on the environment.

The guiding idea for these criteria is that technologies have a good chance to be effective if they are appropriate to the needs, expectations and limitations of the surroundings in which they will be applied. In other words, the selected solution should be in harmony with local standards and values and build on existing skills and techniques. A new technology will not be embedded in a sustainable manner into an organization or community if the dependence on the developers of the solution is high and the available resources (financial as well as human) for maintenance are expensive and scarcely available.

Development needs that are met through community education and development tend to be sustainable. New technology needs to address the local community as the main stakeholder. Only then we observe self-sustainability and expanding reservoirs of skills in the communities (Tharakan, 2006).

Although the criteria proposed by Darrow and Saxenian (1986) will result in an appropriate design of the technology, they fail to highlight the implementation process. Even appropriate technology can be rejected by the potential end-users if the implementation process

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 4

does not address the needs, expectations and limitations of the community and/or when the invoked changes are not guided in an appropriate manner. We will develop this aspect in more detail in the next section where we concentrate on Appropriate ICT.

Understanding AT is an important element in the education of engineers. In order to operate in a development context they are required to have a good understanding of general development issues, philosophy and ethics, and AT in development (Tharakan, 2006). In the education of computer and information systems engineers, this knowledge has received little attention. 3. APPROPRIATE ICT AND THE DIGITAL DIVIDE The use of ICT in developing countries is increasing and the expectations of its role in accelerating the socio-economic development in these countries are high (Walsham et al., 2007, McNamara, 2003). Until recently the use of ICT in Africa and other developing countries was reserved to large international organizations and foreign NGOs (non-governmental organizations). Foreign ICT experts were contracted for the installation and maintenance (Bruggink, 2003) and for conducting training beyond the basic level of Office applications (Heeks, 1998). Local implementers had to fly to Europe or North America to receive training. Over the last couple of years this is changing rapidly (Levey & Young, 2002). The digital gap made it to the international agenda (and is often strongly linked to amongst others the eight Millennium Development Goals (MDG)) and with foreign support the first programs and projects have been set up.1

Although some progress has been booked, the penetration of computers is still very low in comparison with Western countries (Jensen, 2002). Even at this moment the majority of people in LDCs had never touched a computer and most small and medium sized businesses operate without them. A major challenge concerns the need to bridge the so-called ‘digital divide’ between those people with the ability to access and use information technologies effectively, and those without. The challenge remains to tackle such difficulties and to resolve them (Walsham et al., 2007)

In order to bridge the digital gap information technology must be available in the life of the 'ordinary man'. Like in Europe, North America and increasingly in Asia, the ultimate goal is that everyone should have access to computers and information everywhere and always (Universal Access). Not only in the large cities, not only the rich class, but information technology must also be available for the population in the rural areas and for people with a lower level of education. This creates formidable challenges to the developers and implementers of ICT solutions.

The application of ICT as a tool in bridging the digital divide is not as straight forward as many international development organizations claim. More technology does not necessary result in development when collaboration between industry, government and development organizations is not involved (Avgerou, 2003). Moreover, the implementation of ICT without local support and no contributing to a local demand/need will not lead to sustainable development. The technology will be rejected after an initial period of euphoria, as there is no added value in it. Small technical failures will easily result in complete break down when there is not technical support. For a multitude of reasons these aspects of the implementation of ICT in developing countries are often overlooked by supporting donor agencies. A theory of appropriate ICT aims to address these issues and support sustainable design and implementation of ICTs in general and particularly in LDCs.

1For more details on the role of ICT in realizing the MDG, see: http://wbln0018.worldbank.org/ict/projects.nsf

/WebPages/AA097228395769ED85256DF0007A4065?OpenDocument/*/

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 5

3.1. Appropriate ICT In line with the approach of Darrow and Saxenian (1986) we aim at the development of a framework for Appropriate ICT that will guide designers, implementers and maintainers of ICT to design and implement effective and sustainable solutions that address the needs, expectations and limitations of the targeted communities and allows the 'ordinary man' in LDCs to get connected to the information and knowledge society. This is an enormous challenge and requires careful consideration to limit partial and complete failure and increase the overall effectiveness of ICT for Development (Avgerou & Walsham, 2000).

Appropriate ICT should be perceived from two perspectives: the product and process perspective. The product perspective is concerned with the design of the ICT systems that will be used to offer information and communication services. This covers all aspects from computers (and other connected electronic equipment), servers, network and connections. For example in our approach, a computer setup that is to operate in a community in the African desert is not considered to be appropriate when it is not well protected against heat, sand and dust. The product perspective is very much in line with the guidelines that were developed by Darrow and Saxenian.

The process perspective is just as vital but has not received adequate attention up to this point. A mere techno-centric approach, even when rooted in the principle of AT, will not deliver effective community-embedded ICT that will be appreciated and used by the potential end-users. The emerging, interdisciplinary fields of Social Informatics (Kling, 1999) and Community Informatics (Gurstein, 2000, 2003) provide good references for our process perspective. Social Informatics concentrates research in three areas: 1. Theories and models: The development of models and theories that explain the social and

organizational uses and impacts of ICT.

2. Methodologies: The development of methodologies that address the social impacts of the design, implementation, maintenance and use of ICT.

3. Philosophical and ethical issues: The study of philosophical and ethical issues that arise in the use of ICT in social and organizational contexts.

Community Informatics (CI) constitutes a subset of Social Informatics with a focus on

communities (McIver, 2003). Where Social Informatics has a stronger research focus, CI is more suitable for the development and implementation of ICT in developing countries (see: Vaughan, 2006). The community itself is involved in the adaptation of ICT to their purposes including advocacy, local information on community resources and services available, community mapping for community planning and development (demographics, geography) (Gurstein, 2000).

The process perspective of Appropriate ICT needs to encompass a community oriented and participatory focus to address the needs, expectations and limitations in which the technology is to be used. So, if computers are introduced in the African context, people should be empowered on the devastating effect of humidity, power fluctuations, dust and sand and learn how to open and clean the computers from the dust door maintain the computers under these circumstances.

In order to better serve development needs, IT-related projects and systems in the developing world must improve their capacity to address the specific contextual characteristics of the organization, sector, country or region within which their work is located (Avgerou & Walsham, 2000). We consider five variables relevant to address in the context:

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 6

1. Culture: Societies, or groups in a society vary in their sets of shared attitudes, values, goals, and practices. Culture deserves careful attention when ICTs are introduced in the development context (Westrup et al., 2003). We consider culture a central variable.

2. Environment: Physical conditions (heat, cold, dust, humidity etc) need to establish an important ICT solution design.

3. Organization: The structure of the organization (in the broadest sense of the word) determines the implementation strategy of systems both in the developed as developing world.

4. Economy: The current and future economic situation of a country, sector or organization should serve as a determinant in the ICT investment decisions.

5. Political climate: Some governments are more restrictive in their ICT guidelines than others. Openness is not always appreciated and some governments have ‘partnerships’ with hardware and software suppliers.

In conclusion we define Appropriate ICT as follows:

The integrated and participatory approach that results in tools and processes for establishing Information and Communication Technology (ICT) that is suitable for the cultural, environmental, organizational, economic and political conditions in which it is intended to be used.

4. AN APPROPRIATE ICT FRAMEWORK In this section we will go beyond the definition of Appropriate ICT and propose a framework that enables a more effective and appropriate design and implementation of ICT in LDCs. The framework is founded in the traditional Systems Development Life Cycle (SDLC) that is used in Information Systems development, but extends it with tools and approaches that will guide the ICT solution to appropriateness. The SDLC comes in many types and flavors (Brandon, 2006) but we will adopt a basic five-phase model2: 1. Definition: Determine the goals, scope and requirements of the ICT solution 2. Design: Resolution of technical issues, selection of architecture and standards 3. Construction: Implementation of the design, testing and documentation of the system. 4. Installation: Roll-out of the services offered by the systems to the end-users, training. 5. Operation/maintenance: problem solving, user support, and incremental improvement

through monitoring an evaluation focusing on the use of the services by the end-users.

In the Appropriate ICT Framework, the SDLC is merged with the principle of Appropriate ICT and methods are proposed. Figure 1 shows the foundation of the A-ICT Framework: the SDLC on the outer ring and the five criteria that should be considered to design and implement appropriate ICT solutions on the inner ring. Culture is displayed in a central role.

2We deliberately deal very briefly with the SDLC since there are many textbooks that deal with this aspect in

detail. For more information the reader is referred to Hoffer et al., 2006; Dennis et al., 2001; Laudon and Laudon, 2005.

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

7

Figure 1: The Foundation of the Appropriate ICT Framework

As mentioned above, Appropriate ICT encompasses two perspectives: the product and process. In the framework this is expressed in three aspects: Hardware, Software and Change Management. Hardware and software result in a product, an ICT artifact. Change management establishes the process for the design, development and implementation of the ICT artifact. In order to guide the development of an appropriate ICT solution the framework provides a range of guiding questions. Table 1 lists the key questions that need to be answered to address the issues relevant in the Appropriate ICT Framework. The questions in the table integrate the 10 rules for AT and the focus areas of Appropriate ICT: culture, environment, organization, economy and politics. The questions are structured along the phases of the System Development Life Cycle.

Several tools and methodologies are in use by professionals in the ICT for Development arena. A systematic inventory of approaches still needs to be made. In Figure 2 the A-ICT model is supplemented with some of the tools that are currently in use or are recommended. More tools can be included to assess or support cultural, environmental, organizational, economical and political dimensions of the ICT project.

SDC (2009) 38, 9, 1-18

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

Table 1: Key guiding questions for Appropriate ICT Development Phase Hardware Software Change Mgnt Definition Specific requirements to

hardware in terms of climatological and environmental conditions? What are the possibilities in terms of enabling factors (Internet connectivity, electricity)?

What are the needs? What are the expectations?

What ICT knowledge levels? What the financial constraints? What is the cultural context? What added value is created? How is the economic equilibrium affected? What new ways of working are introduced? What will the impact be of the system in terms of organizational change? What is the involvement in the idea generation of key decision makers (political leaders, religious leaders)?

Design What is offered on the local market? What are physical constraints? What the financial constraints?

What interoperability needs? What localization is needed? What flexibility is expected?

What are the information needs of the various target groups? How will these needs evolve? How do the expectations change?

Construction

What local skills are available? Is the equipment protected against physical conditions?

What local skills are available?Are features in line with skills?Are free and open source alternatives considered? Are the systems well documented?

Are local skills and knowledge being developed? Are stakeholders actively involved? What new ways of working are introduced? What will the impact be in terms of organizational change?

Installation Is all the equipment well protected?

Has the system been tested with all stakeholders?

Are all stakeholders involved in training program? Is the added value made clear?

Operation / maintenance

Is local capacity sufficient? Are spare parts easily available?

Are software maintenance skills available?

Is a support organization in place? Is the support organization able to support all stakeholders (e.g. Gender issues)

EJI

EJISDC (2009) 38, 9, 1-18

Figure 2: Some tools and methodologies supporting the design and implementation of Appropriate ICT Solutions 5. ILLUSTRATING THE APPROPRIATE ICT FRAMEWORK In this section we illustrate the use of the Appropriate ICT Framework with some examples. The examples in section 5.1 and 5.3 are based on projects that were designed and implemented with the help of the International Institute for Communication and Development (IICD) 3. The author was involved on several occasions as external advisor (between 2003 and 2005) and as evaluator for the projects in the period September – December 2006. Section 5.2 describes a project in which the author participated as technical adviser. The project was designed in the period 2002-3. The implementation of the project started in 2003 and continued through 2006.

All information was collected in an ‘action research’ mode (Baskerville 1999, Argyris et al., 1990). The information for these case studies was collected through (face-to-face and online) interviews with program/project managers and users, attendance of project meetings (especially in the case of the project described in section 5.2) and through document analysis (progress reports, meeting minutes and other organizational documentation at side of the projects as well as the donor). During the monitoring of the projects several discussion meetings were initiated to assess the appropriateness of the proposed solution and ways to improve it. The meetings were structure along the criteria for AT coined by Darrow and Saxenian (1986). These discussions served as a steppingstone to start the learning process of the project partners (second loop).

In the subsequent sections we provide a general description of the situation and explain how an Appropriate ICT perspective provides changes to the ICT solution. Therefore the purpose of the examples and short case studies is explorative and aiming at achieving a better understanding of the framework and its implications for the design and implementation of ICT in developing countries. 3 The author would like to thank Arjan de Jager and Deem Vermeulen of the IICD for sharing their experiences.

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 10

5.1. Determine Appropriate ICT Requirements Determining requirements for ICT solutions has been a challenging issue in ICT for development projects. The most important reason being that the majority of the users and policy makers in LDCs have very limited exposure to the actual use of computer to support their work processes. Their understanding of the capabilities of ICT is based on the media and hear-say in other projects. Expectations are high and often unrealistic. Project managers from development agencies, donors or ICT consultants end up prescribing the solution based on their experiences in other projects. As a result, a larger number of the resulting ICT solutions are not based on the needs of the target group and the context.

Lack of proper knowledge sharing between projects results in almost identical projects initiated and executed in the same geographical area and/or the same thematic field. In many cases similar ‘mistakes’ are made resulting in an unnecessary number of failures. All over developing world organisations are supporting the set-up of telecenters or informationcenters to improve the access to information and communication in poor neighbourhoods. At the same time, many of these initiatives fail to deliver services that are appreciated by the clients and almost suffer from operational problems because of limited technical maintenance skills. Within an Appropriate ICT Framework, these problems need to be identified, addressed and avoided. 5.1.1 Determine Appropriate ICT Requirements in the Definition Phase: ICT for Health in

Tanzania In recognition of the fact that ICT can offer important advantages to the Health sector in developing countries and the fact that large number of projects has been initiated over the past years, the start of a new ICT for Health program in Tanzania was faced with a challenge. The chances of unwanted duplication of initiatives were high and the need gaps for the sector unclear. At the same time it was noticed that the local ICT skills were minimal, especially in the sub-urban and rural areas. In order to address this challenge it was decided to start the program with a health sector-wide initiation phase.

To determine the needs and expectations of the sector and to avoid duplication of projects, a participatory idea generation process 4 was initiated. This participatory idea generation process goes beyond that traditionally used consultation processes and is more suitable to generate the requirements in the context of an Appropriate ICT Framework. It uses the method of scenario development to allow participants to identify priority areas for the development of their sector or organization. This avoids a technology focus. In a later stage during the workshop the question is asked how ICT can be used for these priority areas. To bring all participants to the same starting level, a reference report (situational analysis) is produced on the current status of the sector prior to the workshop. As such the Round Table Process is an example of a useful method during the initiation phase allowing participants to answer the questions of this phase way. The workshop is just one part of the overall process and gets a follow-up allowing the actors to develop and implement their own ICT projects.

During the workshop in Tanzania the stakeholders were not only requested to express their needs and expectations of the program, they were also facilitated in the generation of preliminary project and policy proposals that were shared with the all the stakeholders involved. Through a process of feedback and support the stakeholders were able to generate ideas that were new, in line with the needs of the sector and feasible in the context of the available skills.

4The process was facilitated by the International Institute for Communication and Development (IICD). The

organization uses the term 'Round Table' to label a participatory idea generation process. For more details on ideas behind this process, see: IICD, 2004.

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 11

Through this approach, the sector has been able to generate a coherent ICT program for the Health sector of Tanzania. The program now includes the following interrelated project areas: ! ICT Policy for the Health sector, supported by all the stakeholders in the sector; ! Capacity development through a continuous medical education program; ! Development of ICT capacity in both the health sector as well as the private sector to fill

the capacity; ! Development of ICT tools, based on Free and Open Source Software5 and Open Standards,

to support the hospital management; ! Information portal and low-costs Internet connectivity services and support for hospitals.

The participatory idea generation approach that was selected has been able to naturally structure a Definition phase that takes into account the needs and expectations of the stakeholders. Political, cultural and organizational issues were addressed in the process through elaborate peer and reality checking of the preliminary project proposals that were generated. Economical issues were less covered in the initiation phase, but financial feasibility studies were conducted before the design phase of the projects. Environmental issues were insufficiently covered in this phase for this program and were addressed in the Construction phase.

Within the Appropriate ICT Framework the local needs are central in the design and more innovative solutions are anticipated. 5.2. Designing Appropriate ICT Designing hardware, software and network configurations that fit within the development context has been a major challenge over the past years. Unfortunately in most cases the solutions are founded in the European or American contexts and are not necessarily suitable for the development context. For example, hardware that is not protected against the dust, sand and heat in African countries, software that needs an internet connection to be activated or is not supporting local languages. Also theories and tools for system development like e.g. information modelling with Unified Modelling Language (UML) have proven to be difficult to use in an African context.

Where a carefully conducted project Definition phase results in an outline of the needs, expectations and possibilities for the ICT solution, the design phase needs creativity to determine the specifications of the new ICT tool. The specifications have to suit the characteristics of the context and not the other way around. We do not have to implement the complexity of Microsoft Word when the characteristics of a simple word processor like AbiWord suffice. Similarly, WiFi equipment that is suitable for the European context where large numbers of people operate in a small area are not a suitable choice for Timbuktu where a very limited number of people are using the connection in a large geographical area (Wireless Internet Institute, 2003). 5.2.1. Designing a Campus Wide ICT Infrastructure in Rural Uganda6 Uganda Martyrs University is located 80 km from Kampala, the capital of Uganda. The university operates on a relatively small budget, but has been able to develop itself as the leading private university in the country. In spite of the financial limitations, the university

5See Dravis, 2003; Reijswoud & Topi, 2004; Reijswoud & Jager, 2008; on the advantages of the use of Free and

Open Source Software in the developing world. 6For an in-depth discussion of this case study see: Reijswoud and Mulo, 2007.

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 12

administration decides on a policy that states that access to information technology and Internet is essential for their students and staff. Against this background the ICT Department is requested to design an ICT infrastructure that is low-cost (both in terms of initial and recurrent costs), easy to maintain but allows staff and students to access academic information and equips students with good ICT skills.

With the requirements in mind the department of the university developed, in collaboration with external expertise, a plan for an ICT infrastructure based on: ! Free and Open Source Software and Open Standards for server and desktops to keep costs

for software low and avoid future license costs; ! Re-use and upgrade of existing infrastructure through the implementation of a Linux

terminal server project (Martindale, 2002); ! Elaborate capacity development to train knowledgeable ICT support staff and

well-informed students and lecturing staff; ! Bandwidth management to provide maximum access to research information while

limiting non-academic information (pornography, music and film downloads, sports, chat etc).

Staff development for the ICT department as well sensitization of users (students and

staff) proved key in the development of these plans. Participatory learning and awareness creation slowly guided ICT staff and users towards the conceptualization of the new ICT infrastructure plan.

The ICT infrastructure was successfully implemented over the period 2003 – 2005 and serves as a point of reference for other universities in the East African region that operate under similar conditions. The infrastructure has proven to be low-cost, and through the capacity development program administrator skills were developed to maintain the new system. The managed Internet connection provides the required information for staff and students while keeping maximized performance against minimized operational costs.

The project is a good example of the Design and Construction phases of Appropriate ICT since it has taken into account the specific local conditions of the university and the context in which the ICT was to be implemented. The economic requirements were addressed as well as the organizational aspects. The academic culture that was considered important was covered in the design of the network.

5.3. Implementing Appropriate ICT Introducing ICT in a rural setting has its challenges from a product perspective (e.g. robustness of hard- and software) but also from a process perspective. E.g. introducing a computerized information system up-country requires organization and sector-wide consultation and/or participation. While citizens of urban centers are more or less used to the idea of modern ICT's and increasingly understand the advantages, the rural based projects require a step-by-step approach to engage stakeholders and beneficiaries to counteract resistance. 5.3.1. Constructing and Implementing a Health Management Information System in Uganda The Uganda Catholic Medical Bureau (UCMB) is the health office of the Roman Catholic Church (RCC) in Uganda and is amongst others supporting 27 rural hospitals. From 2004 UCMB established a system aiming at improving the use of the national Health Management Information System (HMIS). Improved health facility management is a necessity in a situation in which the funds for the health sector in most LDCs are decreasing while at the same time health care delivery funding is changing to a cost-based system. Therefore more effective and

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 13

efficient ways of working are the only solution to keep the Health services at an acceptable level. The most prominent problems that were to be addressed by the project were:

1. The information collection using the lengthy HMIS forms were unreliable (incomplete and inaccurate) due to the labor intensive information collection procedures: The hard-copy forms had to be filled in manually demanding hospital managers to spend hours to correct for mistakes while all the data of the 27 hospitals had to be retyped at UCMB HQ etc.

2. Slow feedback by UCMB to the inaccurate information in hard-copy form provided by the hospitals (Backlogs of three to six months were not uncommon).

3. Lack of human capacity (technical as well as managerial) in order to implement and use a proper HMIS.

The proposed technical solution was simple and straightforward. At first problem 1 was

addressed: The hard-copy forms were replaced by standardized Spreadsheets (which were sent to UCMB using normal mail). The organizational impact of this step was still small: The manual data entry was now supported by ICT applications. However, careful training and awareness creation for the hospital staff was needed to avoid the computer to be locked away and to be used as tool emphasize hierarchical differences.

After the first step the communication links between the up-country hospitals and UCMB HQ data were set-up to address problem 2 while problem 3 was addressed. The percentage of hospitals submitting complete and accurate HMIS forms in time moved from 48% to 96% during project implementation. The UCMB HQ can now concentrate on the analysis and feedback to hospital managers to support informed decision-making. The feedback mechanism enables hospital managers to finalize their planning and budgeting processes in a timely manner. The impact of these steps from an organizational perspective was much bigger. Due to better information the health facility managers could develop clearer guidelines. At the same time, the decision of traditional hierarchical leaders was challenged. Administrators were now able to challenge the management decisions of the medical staff, in some occasions resulting in doctors sabotaging the data collection process. Only through education and training a schism between administration and clinical staff could be avoided.

The change management process took three years and a carefully planned step-by-step approach where the following results were achieved:

1. From hard-copy forms which had to be filled in manually to standardized spread sheets which could be filled in using a computer: Focus on quality of data.

2. From sending these Excel forms by normal mail to sending it by email (forms as attachments): Focus on timeliness of data.

3. A system which made it possible to link and analyze the HMIS output with a cost-based financial system allowing hospital managers to implement an activity based costing model in their hospitals: Focus on the use of data at managerial level.

4. From sending Excel forms by email (as attachment) to filling in on-line forms to allow UCMB and the 27 hospitals to generate quality information on trends, statistics etc.: Focus on use of data a macro level. The project illustrates the importance of careful implementation processes in ICT

project in the development context. A strong traditional hierarchical structure (also described in Stoops et al., 2003) has to be recognized and managed with an appropriate change process. From a product perspective is needs to be mentioned that the economic (the hospitals took over the recurrent costs after a period of two years) requirements were addressed as well as the

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 14

political aspects (involvement of religious leaders and Ministry of Health). 5.4. Discussion In the first study we observe a situation that a participatory and consultative process guides the definition and design of the ICT for Health program. Through this process users and other stakeholders were engaged in a learning and knowledge sharing experience. The outcome of the process was driven by the needs and knowledge of the users as well as constraints emerging from the context. The resulting ICT solutions were low-tech, but appropriate. Traditional approaches could have resulted in state-of-the-art solutions in line with available technical paradigms in developed countries but that lack ownership of the users and fail to find support in the local ICT community.

The design of the new ICT infrastructure at Uganda Martyrs University in Uganda shows a design based on the requirements of the organization and the, mainly financial, limitations of the context. Where many ICT solutions in Africa are based on donated and pirated software, the described solution fits the situation, is legal, low-cast and has built new and sustainable capacity. By re-using written-off hardware, increased access to ICT for students and staff was created at very limited costs. Unfortunately, designs like these are hardly seen in Africa and appropriate designs are hardly promoted by donor organizations.

The implementation of the Health Management Information Systems in Uganda addresses challenges that are common in project implementation in developing countries. No understanding of ICT, superstition and impenetrable hierarchical relations constitute barriers to implementation that are not accommodated in modern IT project management approaches. Most of these approaches are founded in a rational understanding of the possible advantages of ICT and are not geared to first-time users. Long and careful implementation processes are required that often look inefficient from a developed world perspective. 6. CONCLUSION There is little discussion that ICT can play an accelerating role in bridging the digital divide between developed and developing countries. The biggest challenge at the moment is to develop an ICT infrastructure, knowledge and skills that provide opportunities for sustainable development. As we have discussed in this paper this requires a change in perspective. The ICT for development community has to shift away from traditional one-size-fits-all solutions to solutions that fit the context in which the technology is to be used. The hardware, software and change management needs to be aligned with the local cultural, environmental, organizational, economical and political conditions in order to realize effective and sustainable development.

The Appropriate ICT Framework as presented in this article provides a guiding model to develop appropriate ICT solutions for LDCs. The model is based on existing theories in Appropriate Technology, Social and Community Informatics and the System Development Life Cycle. As was illustrated in the examples, an Appropriate ICT perspective promotes a thinking process guiding ICT4D experts to reconsider the solutions that are being designed, developed and implemented.

The high number of failures suggests that the complexity of ICT for Development projects is often underestimated. We often observe that the ICT component in development projects is being handled by general project managers that have limited understanding of the possibilities of ICT and resort to the implementation of standard solutions that have proven their effectiveness in the developed world. In the end, the recipients in the ICT projects remain with hardware that breaks down because it is not designed for the harsh conditions, no available skills to maintain the software or clients that are not using it since they do not see the added value. The initial success and the euphoria disappear as soon as the donor retreats. Donor agencies need to recognize that ICT projects needs qualified expertise and cannot be left to

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 15

general project managers. An important step will be when computer science and information systems curricula

start to adopt (elective) courses in ICT for Development and Appropriate ICT issues in line with Tharakan (2006). Students aiming at a career in ICT for Development will be provided with a solid understanding of the challenges and limitation of standard ICT solutions and available alternatives.

Education needs to find its foundation in research. So far, the academic world has shown very little interest in the development of theories and models that suit the developing world. Too often academics seem to believe that they are developing universal theories or models. Experiences with the applications of these theories and models in the developing world show that they are far from universal.

Governments and policy makers in developing countries need to be educated that theories, models and technology that are developed in the developed countries do not provide a guarantee for success. Many do believe this. Suppliers of software and hardware as well as project managers should be questioned critically on how the products fit in the local context. The A-ICT Framework will give them guidance in assessing their needs and query the solutions that are presented.

The A-ICT framework we have presented in this paper forms the foundation for a larger project to provide a flexible approach for the design and implementation of ICT in developing countries. A more detailed inventory of tools and methodologies will be developed and case studies will be reviewed more thoroughly. The A-ICT framework will have to be tested as a prescriptive methodology. At present the framework first test have started in a project focusing on the design and implementation of a nation-wide Academic Registration Information System in Mozambique (Pscheidt et al., 2009). REFERENCES Argyris, C., Putnam, R. and McLain Smith, D. (1990) Action Science: Concepts, Methods, and

Skills for Research and Intervention. Jossey-Bass, San Francisco. Avgerou, C. and Walsham G. (Eds.) (2000) Information Technology in Context: Studies from

the Perspective of Developing Countries. Ashgate. Avgerou, C. (2003) IT as an Institutional Actor in Developing Countries, in: Krishna, S. and

Madon, S. The Digital Challenge: Information Technology in the Development Context. Ashgate.

Baskerville, R. (1999) Investigating Information Systems with Action Research, Communications of the Association of Information Systems, 2, 9.

Brandon, D. (2006) Project Management for Modern Information Systems. IRM Press/Idea Group.

Bruggink, M. (2003) Open Source in Africa: A Global Reality; Take It or Leave It? IICD Research Brief – No UICT01. International Institute for Communication and Development, http://www.iicd.org

Collins, H.M. (1992) Artificial Experts: Social Knowledge and Intelligent Machines. MIT Press.

Curtain, R. (2004) Information and Communications Technologies and Development: Help or Hindrance? Australian Agency for International Development (AusAID), http://www.developmentgateway.com.au/jahia/webdav/site/adg/shared/CurtainICT4DJan04.pdf

Darrow K. and Saxenian, M. (1986) The Complete Appropriate Technology Sourcebook. http://www.villageearth.org/atnetwork/atsourcebook/index.htm

Dennis, A., Wixom, B. and Tegarden, D. (2001) Systems Analysis and Design, Wiley.

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 16

Dravis, P.J. (2003) Open Source Software: Perspectives for Development. Worldbank/InfoDev, http://www.infodev.org/en/Document.21.pdf

Dymond, A. and Oestmann, S. (2004) A Rural ICT Toolkit for Africa. Worldbank/InfoDev. http://www.infodev.org/projects/telecommunications/351africa/RuralICT/Toolkit.pdf

Evans, R. and Collins, H.M. (2007) Expertise: From Attribute to Attribution and Back Again?, in: Hackett, E.J., Amsterdamska, O., Lynch, M. and Wajcman, J. (Eds.). The Handbook of Science and Technology Studies, 3rd Edition, MIT Press.

Gairola, B.K., Chandra, M., Mall, P., Chacko, J.G., Sayo, P. and Loh, H. (2004) Information and Communications Technology for Development: A Sourcebook for Parliamentarians. Elsevier.

Gichoya, D. (2005) Factors Affecting the Successful Implementation of ICT Projects in Government. Electronic Journal of e-Government 3, 4. http://www.ejeg.com/volume-3/vol3-iss4/GichoyaDavid.pdf

Gurstein, M. (2000) Community Informatics: Enabling Communities with Information and Communications Technologies, Idea Group Publishing.

Gurstein, M. (2003) Effective Use: A Community Informatics Strategy beyond the Digital Divide, First Monday, 8, 12, http://firstmonday.org/issues/issue8_12/gurstein/index.html

Habermas, J. (1985) The Theory of Communicative Action, Volume 1: Reason and Rationalization of Society. Beacon Press.

Heeks, R. (1998) Getting the Most from IT Training Courses for Africa, IDPM, University of Manchester, UK.

Heeks, R. (2003) Most eGovernment-for-Development Projects Fail: How Can Risks be Reduced?, IDPM, University of Manchester.

Hoffer, J.A., George, J.F. and Valacich, J.S. (2006) Modern Systems Analysis and Design. Prentice Hall,

IDRC (2001) Outcome Mapping: Building Learning and Reflection into Development Programs. International Development Research Center. http://www.idrc.ca/booktique

IICD (2004) The ICT Round Table Process: Lessons Learned from Facilitating ICT-Enabled Development. International Institute for Communication and Development. Den Haag. http://www.bcoalliance.org/node/88

Jensen, M. (2002) The African Internet: A Status Report. http://demiurge.wn.apc.org/africa/afstat.htm

Kling, R. (1999) What is Social Informatics and Why Does It Matter? D-Lib Magazine, 5, 1. http://www.dlib.org/dlib/january99/kling/01kling.html

Kozma, R.B. (2005) Monitoring and Evaluation of ICT for Education Impact: A Review, in: Wagner, D.A., Day, B., James, T., Kozma, R.B., Miller, J. and Unwin, T. (Eds), Monitoring and Evaluation of ICT in Education Projects: A Handbook for Developing Countries. Worldbank/InfoDev. http://www.infodev.org/en/Publication.9.html

Krishna, S. and Madon, S. (2003) Challenges of IT in the Development Context, in: Krishna, S. and Madon, S. The Digital Challenge: Information Technology in the Development Context. Ashgate.

Laudon, K. and Laudon, J. (2005) Essentials of Management Information Systems, Prentice Hall.

Levey, L. and Young, S. (Eds.) (2002) Rowing Upstream: Snapshots of the Pioneers of the information Age in Africa. Sharp Media. http://www.piac.org/rowing_upstream

Martindale, L. (2002) Bridging the Digital Divide in South Africa. http://www.linuxjournal.com/article/5966

McIver, W. (2003) A Community Informatics for the Information Society. UNRISD briefing Paper. http://www.unrisd.org/unrisd/website/document.nsf/%28httpAuxPages

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 17

%29/96F7F7BE45C6A0F6C1256E55005A3A62?OpenDocument&panel=additional McNamara, K. (2003) Information and Communication Technologies, Poverty and

Development: Learning from Experience. Worldbank/InfoDev. http://www.infodev.org/en/Publication.17.html

Moussa A. and Schware, R. (1992) Informatics in Africa: Lessons from World Bank Experience. World Development, 20, 12, 1737-52.

Osama, A. (2006) ICT for Development: Hope or Hype? Science and Development Network. http://www.scidev.net/Opinions/index.cfm?fuseaction=readOpinions&itemid=537&language=1

Pscheidt, M., Reijswoud, V. van and Weide, Th. van der, (2009) Assessing Appropriate ICT with the ARIS case in Mozambique. Proceedings of the 5th Annual International Conference on Computing and ICT Research SREC09. Kampala.

Reijswoud, V. van and Jager, A. de (2008) Free and Open Source Software for Development: Exploring expectations, achievements and the future. Polimetrica. http://arxiv.org/ftp/arxiv/papers/0808/0808.3717.pdf

Reijswoud, V. van, Jager, A. de, Mulder, J.B.F. (2005) 8 lessen voor ICT in de Derde Wereld: Raamwerk voor ICT projecten. Informatie, 47, 4, 18-22.

Reijswoud, V. van, C. Topi. (2004) Alternative Routes in the Digital World: Open Source Software in Africa. In: P. Kanyandago, L. Mugumya (Eds), Mtafiti Mwafrika (African Researcher), African Research and Documentation Centre, Uganda Martyrs University Press.

Reijswoud, V.E. van, Mulo, E. (2007) Free and Open Source Software for Development – Myth or Reality: Case Study of a University in Uganda, in: Amant, K. St. and Still, B. (Eds.). Handbook of Research on Open Source Software: Technological, Economic and Social Perspectives, IDEA Press.

Sciadas, G. (Ed.) (2003) Monitoring the Digital Divide and Beyond. NRC Press, Montreal. http://www.infodev.org/en/Publication.20.html

Stoops, N., Williamson, L. and Braa, J. (2003) Using Health Information for Local Action: Facilitating Organisational Change in South Africa, in: Krishna, S. and Madon, S. The Digital Challenge: Information Technology in the Development Context. Ashgate.

Tharakan, J. (2006) Educating Engineers in Appropriate Technology for Development, World Transactions on Engineering and Technology Education, 5, 1, 233-235.

United Nation Development Program – UNDP (2001) Human Development Report 2001: Making New Technologies Work for Human Development, United Nations. http://hdr.undp.org/reports/global/2001/en/

United Nations Conference on Trade and Development – UNCTAD (2007) The Least Developed Countries Report 2007: Knowledge, Technical Learning and Innovation for Development. United Nations. http://www.unctad.org/Templates/webflyer.asp? docid=8674&intItemID=4314&lang=1&mode=downloads

Vaughan, D. (2006) ICT4D – Linking Policy to Community Outcomes. Partners in Micro-Development. http://www.microdevpartners.org/documents/ICT4D LinkingPolicytoCommunityOutcomesPDF.pdf

Vosloo, S. (2006) A Quick Guide to Implementing ICT for Development Projects. Digital Divide Network. http://www.digitaldivide.net/articles/view.php?ArticleID=742

Walsham., G., Robey, D. and Sahey, S. (2007) Special Issue on Information Systems in Developing Countries, MIS Quarterly, 31, 2, 317-326.

Westrup, C., Al Jaghoub, S., El Sayaed, H. and Liu, W. (2003) Taking Culture Seriously: ICTs Culture and Development, in: Krishna, S. and Madon, S. The Digital Challenge: Information Technology in the Development Context. Ashgate.

Winograd, T. and Flores, F. (1986) Understanding Computers and Cognition: A New

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org

EJISDC (2009) 38, 9, 1-18 18

Foundation for Design. Addison Wesley. Wireless Internet Institute (2003) The Wireless Internet Opportunity for Developing Countries.

Worldbank/InfoDev. http://www.w2iorg World Bank (2003) Turkey: Informatics and Economic Modernization, World Bank.

The Electronic Journal on Information Systems in Developing Countries http://www.ejisdc.org