Jacques Hugo, Bob Day
In the developed world, humancomputer interactions (HCI) professionals have had success spreading the word about user-centered design (UCD) as something that companies need to embrace for success of their products in the marketplace. If this pitch is validand we certainly believe it isthen market forces alone should drive adoption of UCD practices.
What happens, though, when the context is a developing economy instead of a developed consumer society? How does UCD fit in not just with the "product development lifecycle," but also with national initiatives aimed at broad social development? Such programs, typically endorsed, sponsored, or even mandated by government agencies, tend to have a top-down character that often leads them to operate in ways that conflict with the fundamental principles of UCD. As with consumer products, if technological development does not really meet the needs of the user, it will fail. UCD seems like a natural antidote to this tendency, but what are the dynamics of getting a non-market-driven entity to adopt or promote UCD practice so that these principles can help guide national technological initiatives toward success?
This article is a case study of one such effort in progress, involving a collaboration of government, business, and academic stakeholders. The context is South Africa, a nation whose economic potential can be released only if political enfranchisement for masses of people is backed up by economic and technological enfranchisement. Economic and technological development in South Africa may face obstacles beyond what most of us in HCI are familiar with. However, a group of dedicated HCI visionaries sees a potential for familiar UCD principles to be a key element in addressing those problems and possibly an antidote to the all-too-common failure paths of well-intended, top-down technology planning. Susan Dray
South African society is undergoing a transformation shaped both by indigenous socioeconomic forces and cultural practices and by the forces of globalization. In South Africa, as in many other countries, a digital divide exists between the technological haves and have-nots. For the haves, computer-based commerce and dynamic networked communication are not only possible but also an indispensable tool for everyday business and personal activities. For the have-nots the situation is radically different. In South Africa, this group is especially diverse. In fact, the population on the fringes of information and communications technology (ICT) seems to grow culturally, economically, and educationally more diverse. Users from these communities that have not yet been brought into the global technology revolution have vastly different expectations of technology. These expectations are colored by their frame of reference, educational level, cultural prejudices, career expectations, income, and many other variables that are poorly understood by software developers, nor accommodated within current design practices in South Africa.
Because this population of potential users is so diverse, and because the needs are so great, human factors has a crucial role to play, not only in the design of computing products for diverse user groups, but also in our efforts to build a healthy information and communications technology industry that can enrich people's economic, social, cultural, and political lives. Without proper understanding of the cultural variables involved, design of technology threatens to expand, rather than resolve, the divide between haves and have-nots, something our society cannot tolerate.
Although the forces of globalization make it more urgent to resolve this problem so that South Africa can emerge as a full participant in technological development, globalization also tends to conflict with the goal of making technology accessible to the broader population. The blind acceptance of imported products threatens to impose a typical "one-size-fits-all" marketing strategy on South African users that will inevitably result in excluding various sectors of society from access to, and benefiting from, ICT. As everywhere else in the world, ICT is both a product and an agent of change. When we look at the impact of globalization on development in South Africa in particular, it is clear that ICT can clash with cultural values that emphasize traditional community and plays an important role in hastening change toward more individualistic values. South African developers therefore must understand how these technologies (whether in the form of device, interface, or the institution represented) might improve or undermine this and other aspects of the indigenous cultural systems of South African society. Human factors inevitably must address the need for sensitivity to indigenous realities and the design of technology. In fact, to any HCI practitioner it is obvious that user-centered design (UCD) is needed to leverage modern information technology for South African society.
Unfortunately, software and systems development in South Africa is still characterized by the application of engineering techniques to technical problems, with little attention to human factors. As elsewhere in the world, this results in poor application usability, poor user performance, low productivity levels, and users' dependence on support and training. Where the notion of UCD has reached the IT community at all, it is fraught with all the common misconceptions: "too expensive, irrelevant, unnecessary..." The predominant attitude in South African mainstream information technology (IT) is to regard human-computer interaction (HCI), UCD, and usability as "fringe" disciplines and therefore "nice to have." This attitude is exacerbated by the fact that there are only a handful of practitioners and they have no official standing, no matter how professional or well-qualified they may be. The problem is clearly one of awarenessdevelopers simply "don't know what they don't know" and they will either fail to recognize the failure of an application or attribute problems to technical difficulties or users' lack of training.
Fortunately, two stakeholders seem to show some promise of contributing to a change in this state of affairs and make it more likely that technological development may be oriented to the needs of particular communities. These are the South African government and the budding HCI community.
The South African government has recognized that research and development, both applied and basic, in the natural sciences and in the social sciences, are crucial to social and economic development, innovation, and national competitiveness. Government has several legitimate missions for development that make it potentially a supporter of HCI. First, it needs to advance democratization by breaking down the gulf between technological haves and have-nots, thereby promoting the development of a just and stable society. Second, as an important stakeholder in the development of national resources, the government must thus ensure that state funds are applied effectively when services are contracted from the private sector. This can mean holding suppliers accountable for ensuring that their technology solutions are truly usable by their target populations. Finally, as promoter of the overall development of South African industry and economy, the government also has an interest in reducing dependence on foreign sources and encouraging the development of indigenous industry to meet indigenous needs
Having neither the expertise nor the resources, the government depends on industry and academia to execute initiatives that it may promote in these areas. As is typical with most top-down initiatives, this often creates unwieldy bureaucracies and complex structures that make it very difficult for small organizations and individual specialists to contribute. It also increases the risk that design will be driven from the top down rather than in a user-centered manner, unless government recognizes the need for UCD and promotes UCD as part of the solution.
Another factor makes the government a potentially powerful sponsor of UCD. Because of the government's strong influence on economic and industrial development, fringe initiatives often gain high-profile status overnight when the government is somehow associated with or "endorses" them. This creates the possibility for demonstration projects that will be highly visible on the national level. This phenomenon will be exploited in our solution discussed toward the end of this article.
South African HCI Community
A small number of educators are introducing HCI into computer science curricula, and practitioners are working with application developers to integrate UCD with the overall product development life cycle. In addition, the local HCI special interest group (CHI-SA, the South African ACM SIGCHI chapter) is actively working with several stakeholders, including academics and nongovernment organizations (NGOs) to raise awareness of the impact of UCD in usability and overall organizational performance. HCI practitioners are now beginning to pay attention to the processes of enculturation, acculturation, and cultural identities in the localization of software. They also have recognized the need to encourage developers to understand the role of the many cultural factors at work in the design of computing products, as well as the issues involved in inter- and multicultural software design and internationalization, and how they affect the bottom line for organizations. In a very few organizations, some progress has been made toward institutionalizing UCD.
These efforts are certainly praiseworthy, but the awareness gap is still large, the technology accessibility gulf seems to be widening, and poor usability is very much in evidence in most sectors. Qualified practitioners and educators are still scarce, a condition related to a lack of awareness and implementation at industry level and an isolation and fragmentation between academia, industry, private research and development (R&D), and government.
A large part of the challenge lies in the fact that, despite the recent chartering of CHI-SA, South Africa does not yet have a well-established HCI communitythere are simply not enough people active in HCI to ensure that the message is spread widely enough to reach a large number of lecturers, students, and developers. Even the few who have committed themselves to the field experience many practical difficulties. Several factors such as economy, language, infrastructure, and sheer ignorance among potential sponsors or employers prevent them from experiencing the realities of the field first-hand or to learn from their counterparts in the international community. The handful of HCI practitioners in South Africa are painfully aware that the research that will lead to user interfaces for the computing devices of tomorrow is happening now at universities and corporate research labs all over the worldalmost everywhere else except in South Africa.
Clearly neither technology nor social constructs alone can provide the answers to these challenges. By definition, the issues are multidisciplinary and will therefore require multidisciplinary approaches to the design of software for the masses.
Clearly, in view of the needs and limited resources, something rather dramatic needs to be done to mobilize all available resources to reduce ignorance in this country. This is the challenge that CHI-SA and its partners in government, industry, government and NGOs hope to address in a bold venture that will create the right number of resources (in all possible terms: ideologically, economically, technologically, educationally or politically) necessary to help us cross the chasm and develop innovative ways to grow sustainable indigenous developments through identification, development, and exploitation of local HCI skills and resources. This venture is called an indigenous HCI cluster, a term used to describe a virtual organization that is a collection of real and virtual resources, concepts, skills, and organizations with the mission of identifying and promoting the appropriate conditions and projects that will leverage HCI knowledge and UCD practices into the South African ICT context.
The concept of an indigenous HCI cluster originated in the similarly named "indigenous ICT clusters" that resulted from a number of high-level projects launched by the Department of Arts, Culture, Science and Technology (DACST) and the Department of Trade and Industry (DTI), in collaboration with advisers from industry and academia. Notable among these initiatives is the Technology Foresight Exercise, which investigated the potential technological trends and trajectories of significance to the social and economic development of South Africa and served to inform R&D investments by the public and private sectors in this country. However, apart from the specific technological insights from Foresight, the conduct of such an exercise had in itself paid dividends in other countries by highlighting certain characteristics among the groups who participated. In the South African Government's white paper on science and technology (see www.gov.za/whitepper/1997/sc&tecwp.htm) these characteristics were referred to as the six Cs of research foresight:
- Communication, bringing together people in a novel forum so that they can interact.
- Concentration on the longer term, so that participants can look further into the future than they might otherwise.
- Coordination, so that people not accustomed to working together can form productive R&D partnerships.
- Consensus, so that a clear picture of likely future scenarios is generated.
- Comprehension, so that those involved gain an understanding of changes that are happening in their businesses or professions, at a global level, and attempt to exert some control over these events.
- Commitment. so that people participate fully and are able and willing to implement changes in the light of the Foresight exercise.
Recommendations and implementation plans that have emerged from Foresight have centered on ICT industrial clusters. These have mostly been modeled on experiences with successful clusters formed in the developed world and satellite clusters formed in other developing countries such as India and Sudan. Such satellite clusters promote ICT capacity and capability and merit further development. However, they usually rely on developed- world markets and technologies and tend to play a secondary role to developed-world clusters. Because they are not intended to address the needs of the disadvantaged, satellite clusters tend to involve only the developed elites in the developing world and have little impact on the vast majority of the population in these countries.
This is where the indigenous ICT clusters come inthey are intended to focus on how ICTs in the developed world might be customized or new ICTs might be locally developed to satisfy specifically the needs of the poor, disadvantaged groups in remote and rural areas. It is anticipated that some of the resultant ICT applications and services will not only satisfy the needs of our disadvantaged communities, but might also find significant markets in the developed world. Hence, by addressing our own burning needs, it is likely that in the medium term we might also create areas of high growth and foreign revenue generation from these indigenous ICT clusters.
One of the most promising areas for such local ICT development is an indigenous HCI cluster that would include, and initially emphasize, cultural diversity in ICT, for example in the form of voice technology for African languages. The key objectives of this HCI Cluster would be
- Investigation of cultural issues in the design of computing products.
- Training of practitioners in the application of UCD principles.
- Dissemination of UCD knowledge and benefits.
- Development of methods, resources, and tools for system designers working in multicultural environments.
- Management of special projects.
The cluster could become the virtual headquarters of HCI resources and skills for the IT industry, offering specialized skills such as ethnographic analyses, multicultural design, and many more. As an indigenous cluster, it could also stimulate investment through tax incentives and access to scare resources.
The HCI cluster seems the most practical way to prevent fragmentation of scarce resources.
The success of this model depends on acceptance by all stakeholders of the bidirectional dynamics of the cluster: being essentially a top-down initiative, the cluster provides the vital framework within which the bottom-up processes of domain-specific principles and resources (such as user-centered design) must be deployed. At this early stage it seems to be the most practical way to prevent fragmentation of our scarce resourcesthe fledgling HCI community simply cannot sustain the development of independent HCI organizations and uncoordinated resources. The cluster provides such a strong framework for the optimization and consolidation of our scarce usability resources that we expect that this will be compelling enough to become accepted as an operating model for business, government, and academia.
Although CHI-SA plays an important role in the direction and activities of the cluster, it should not be mistaken for simply a SIG on steroids. Through its alliance with computer societies and related initiatives like UsabilityNet (www.usabilitynet.org), it is intended to combine the best attributes of private sector flexibility, initiative, and entrepreneurship, with tertiary sector research and the needs and enthusiasm of the individual practitioner (an often-repeated dream!).
Hence, thought leadership on HCI theory and practice would be provided by the HCI community, universities, and other research organizations like the Human Sciences Research Council (HSRC) and the Council for Scientific and Industrial Research (CSIR), via a high-interaction mechanism, but with no direct control. This would be implemented by
- Allowing leading academics to allocate a portion of their time (for example, through secondment) to membership of the HCI Cluster Board, or to each cluster project's advisory committee, or both.
- Allowing local academics with a "burning project" in mind to work full-time for a fixed period (3, 6, or 12 months) to champion and co-manage such a project.
- Making provision for international experts (academic and nonacademic) to spend appropriate time in the cluster, in a variety of flexible research, advisory, design, or development roles.
- Ensuring the full-time involvement of several private sector players from a range of South African companies (five or six companies).
The major components of the cluster and their relationships are illustrated in the diagram on page 31.
The diagram indicates that the primary stakeholders are those who overlap the cluster's own infrastructure (CHI-SA, the cluster's own contributors, the HCI community, the IT industry, its own related societies, and the user community). All other stakeholders, contributors, and suppliers lie outside the cluster's own infrastructure. All supporting functions such as policy, funding, and public relations are also located outside the cluster.
The cluster will have a staff of approximately 40 people, of which about two-thirds will be made up of seconded researchers, students, and other temporary project members. We expect to accommodate an average of 20 postgraduate student interns per year. The remainder of the members will be full-time HCI specialists, project managers, developers, and two administrators. The cluster will be managed by a young, energetic, nonbureaucratic champion with a lot of energy and drive, idealism, and excellent communication skills. The total operating cost of the cluster is estimated at $200,000 for the first year.
The intent is that real HCI work will be performed inside the cluster, but most of the outputs will be ploughed back straight into the community (for example through multipurpose community centers [MPCCs]), into both the IT and the HCI communities as new job opportunities and best practice, or into tertiary education as industry experience. Perhaps more important, a portion of the outputs will be spun off in the form of new small and medium-size enterprises (SMEs) where HCI and usability best practice will almost automatically form part of their core operations.
Although key sponsors and partners have yet to be confirmed, a number of critical projects have already been identified:
- Development of a multilingual (African languages) advisory system on HIV/AIDS for MPCCs, employing speech technologies. Although this is a private initiative, it will directly support the Department of Health's AIDS awareness program. Currently HIV/AIDS counselors have to travel great distances to reach people in remote communities. This consumes large amounts of resources, is often available only in English, and can at best only provide limited and inconsistent support to the community. A facility that employs natural language and speech technology will make critical information more accessible to more people, in time for them to act on it. The design and development of such a facility will require a range of specialized skills and technologies. The cultural, task, and interaction components alone are enough to make this a daunting task while making it an ideal demonstration project.
- National investigation of cultural issues in the design of computing products (this will form the basis of CHI-SA's contribution to the Development Consortium at CHI 2002).
- Collaboration with local and international academics and project teams in developing resources for South African HCI practitioners and students.
These initiatives are interdependent and potentially mutually reinforcing, emphasizing the need to coordinate and consolidate a large number of resources.
Developing local or national indigenous HCI clusters will consolidate all resources needed to help ensure the support of the objectives of all stakeholders involved in economic growth and democratization of technology and that user-centered design in South Africa will include a cultural dimension. If not, we will remain at the receiving end of foreign computing developments. Software will never reflect and acknowledge our economic, political, and cultural diversity, nor will it accommodate the true needs of our upcoming generations. We need structures like the HCI cluster to ensure the development of interfaces for the Rainbow Nation!
Usability Sciences Consulting
Pretoria, South Africa
Executive Director of Information & Communications & Technology
University of South Africa
Business Column Editor
Dray & Associates, Inc.
2007 Kenwood Parkway
Minneapolis, MN 55405, USA
©2001 ACM 1072-5220/01/0700 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2001 ACM, Inc.