This is a draft. Comments and criticisms are very welcome. Chris Johnson
The Social Implications of Future Forms of Human Computer Interaction

Chris Johnson,
Department of Computing Science, University of Glasgow.
http://www.dcs.gla.ac.uk/~johnson

This is a position paper that provides a personal view about the future of Human Computer Interaction (HCI). I believe that the challenges that face HCI will not be radically different from those that we find today. This view is justified by the argument that the future of HCI will be determined more by our social motivations than by technological innovation. Jefferson's emphasis on a social contract promoting "life, liberty and the pursuit of happiness" is used to provide a structure for this partial and preliminary analysis.


I believe that the biggest challenge for HCI continues to be its struggle for credibility. Many of the tools and techniques that have been developed within HCI research cannot be scaled up to meet the challenges posed by industry and commerce. Empirical studies often rely upon biased samples of undergraduate psychology students. Their results cannot easily be applied to the multi-threaded, mobile, distributed environment of modern computation. Similarly, vast areas of interaction involving the elderly or the very young, are neglected. As a result, most research in HCI has little or no practical value.

The failure of HCI...

Recent years have seen a rapid growth in the number of "usability consultants", of "web gurus" and "information architects". The job descriptions that accompany adverts for these positions often resemble the curricula for many HCI courses; a "user centred focus", "a knowledge of human perceptual and cognitive abilities" and so on. These job descriptions also introduce requirements that relate more narrowly to graphic design and marketing. These are subjects that cannot easily be reconciled with many of the underlying assumptions in human computer interaction. Visual impact is currently more important than comprehension or error rates in many commercial contexts.

The difficulty of using existing HCI techniques to support commercial and industrial enterprises will, I believe, have a number of consequences. In particular, we will see a new form of HCI that blends the traditional techniques of scientific empiricism with increasing amounts of pragmatism. HCI practitioners will have to abandon usability labs and follow users from their offices into the pubs, the clubs and out onto the streets. I also believe that we will see similar effects on the observational techniques that have been proposed to replace empirical studies. Anthropological and ethnographic qualifications are unlikely to be common job requirement until it is possible to derive pragmatic design requirements from these techniques. It is not enough simply to provide increasingly detailed analyses of previous failures.

The importance of education...

The greatest benefits of HCI research have never been the methods and tools that are described at conferences such as the present one. Very few of these techniques are ever used by people other than those who developed them. The biggest contribution of HCI has always been made through teaching and through the promotion of certain attitudes towards the products that we must all use. Most of our graduates never use the tools and techniques that we introduce. However, many remember the importance of finite perceptual and cognitive resources, of legibility and font selection, of appropriate feedback and so on. It remains to be seen whether or not we have provided them with sufficient preparation for the challenges that lie ahead.

The future biases of marketing and politics...

Our view of the challenges that face HCI can easily be distorted by the "advertising" that companies use to promote their view of future markets. Similarly, it is too easy to accept the technological "soothsaying" that government agencies use to encourage long term investment and academic research. These predictions are more successful as commercial marketing devices and as tools of political policy making than they are as oracles of future development.

I believe that the challenges that face HCI will not be radically different from those that we find today. This view is justified by the argument that the future of HCI will be determined more by our social motivations than by technological innovation. These motivations are often summarised as "life, liberty and the pursuit of happiness". It is important to contrast Jefferson's view with the economic and political imperatives that dominate the daily lives of most of the world's population. However, this cliche' can help to structure a cursory analysis of what might lie ahead for HCI in the next decade.

Life...

It is possible to identify a number of ways in which technological innovation affects an individuals "inalienable right" to life. In particular, it seems clear that technological literacy will continue to affect personal income and economic prosperity. If we can design devices that are "easier to use" and therefore reduce the burdens of technological literacy then we may both widen access to individual rewards and improve our economic competitiveness. On the other hand, it seems unrealistic to expect that access to technology through improved interface design will really reduce the social and economic distinctions that exist in Western society. It is ironic that so much attention is placed on on technological literacy at a time when many societies are facing increases in more conventional forms of illiteracy. In a more cynical view, improved interface design will reinforce existing economic and social distinctions. We can already see clear signs of this. Access to the Internet is no longer restricted to a technological elite but is widely available to those who can afford the associated costs. The growing "myth of free Internet access" has done little to hide the economic correlation of Internet access with domestic income.

Liberty...

Jefferson's second imperative, the promotion of individual liberty, has a clear resonance with the immediate problems facing the future of human computer interaction. Many people have focussed on the threats that context aware devices and electronic monitoring tools pose for civil liberties. However, much of this previous work under-emphasizes the important trade-off that exists between expected utility and the associated technological threats. For example, the cost savings that can be obtained over the web have convinced many consumers to accept the risks associated with disclosing personal information, such as credit card details and records of previous purchases, over the Internet. HCI plays an interesting role in all of this. By reducing the technological barriers to the exploitation and use of information technology, we may actually be making people more likely to accept the risks imposed by modern surveillance techniques. Arguably, we are helping people to obtain the benefits of information technology without necessarily educating them about the associated risks.

...and the pursuit of happiness

The final component of Jefferson's injuction is the "pursuit of happiness". HCI has had relatively little to say about this most fundamental of human motivations. A vast number of dour and worthy papers have focussed on task-directed interaction in offices, in banks and on shop floors. Relatively little has been said about the ways in which people have used computers for "fun" since their inception. The creation of new devices and markets is steadily changing this previous bias. Recent years have seen a growing interest in interface design for entertainment and leisure systems. There is even an annual workshop series on "Computers and Fun". This work had largely been driven by digital television and by the games industry. Much of the research in this area typifies the argument that was made at the start of this paper. It is largely design-driven and pragmatic in nature. It is not primarily informed by the lab based studies of empirical enquiry, although it occasionally makes use of these techniques to answer limited questions. It makes use of social and contextual observations but often ignores many of the interpretive processes and guards that are stressed by ethnography and anthropology.


Chris Johnson, May 2000.