Kari Kuutti on the Lost User:
The concept of usability is directly connected with the concept of user. This has, however, meant very different things in different times. The paper traces the changes in the conception what the user is and what is user's role in system development through the last 25 years in the areas of information systems design, human-computer interaction and computer-supported cooperative work. This development has not been linear or straightforward, on the contrary, but a certain evolution from a passive informant in the process towards an active actor and cooperative constructor of one's own lifeworld can be identified.
Kari Kuutti is professor of Human-Computer Interaction and Group Technology in the Department of Information Processing Science at the University of Oulu. His research interest lies in the relationship between humans, computer systems, and organizations. He has been working mainly in the areas of Human-Computer Interaction, Computer-Supported Cooperative Work and Information Systems helping to construct methods to develope usable and useful computer technology.
Hunting for the lost user:
From sources of errors to active actors and beyond.
Kari Kuutti, Department of Information Processing Science, University of Oulu
Paper written for the Cultural Usability seminar, Media Lab, University of Art and Design Helsinki, 24.4.2001
See slides for the presentation (pdf)
Introduction: The evolving concept of the user in information technology
The idea of what actually is the "user" of a technological system has obviously a strong influence in the systems to be designed. Different communities of research and design have made different interpretations of the term at different times, and thus the idea has undergone a number of changes during the last decades it has even been a subject of contrast and controversy. Additionally, the old ideas do not completely disappear, but sediments of them tend to remain and to influence our thinking even long after the initial rationale has been found misleading. Thus it may be useful to take a closer look at the evolution of the idea of the user.
The term "user" itself has often been criticized because of its implicit technocentrism: as a user a person is defined with respect to a system, but computer games nothwithstanding humans usually do not use a system for the sake of using; the use is just a means to achieve something else that is much more characteristic and important to a person using a system. As users we are sort of reduced to appendixes of the machines we are using. This criticism has a point, but currently we seem not to have a better term; when one eventually emerges, we may be much wiser in this respect...
The purpose of this paper is to give a short tour through some of the interpretations that have been developed by some communities. There has been a certain enlargement and enrichment of the concept of user during the last decades, but some important aspects of human nature seem still to be missing from the picture of the user.
Quite a lot has been happening in this respect at the interface between humans and information technology during the last quarter of the 20th century, and there is some danger to be drowned in the details. To create some recognizable overall structure I have grouped the different viewpoints into four successive, larger phases or waves of influence from human and organizational sciences into the design of technical systems. It must be admitted that this broad classification is not very accurate in many ways, and that for example many views located in different phases have actually existed parallelly, and there have been important developments of ideas within the phases as well. Some of these reservations are discussed in the text. However, I believe that the classification grasps something essential at the overall level.
The four phases I have named as follows: User as a rational "cog in the organizational machine", user as a source of errors, users as partners in social interaction, and user as a consumer. Each will be discussed more accurately in the following sections, and in the end of the paper there is a section for some discussion and reflection.
1970s: User as a cog in a rational machine the influence from organization theory
The first wave of influence came from organizational theory, and from the perspective of management. The organization was conceived as a rational machine, and to function properly the machine needed predictable and well-behaving human parts. Computer systems were replacing a part of the machinery and making it more efficient, and it was necessary to make the human parts fit well with the computer parts to ensure efficient functioning. During the 1970s the dominant view of a user of a Management Information System (MIS) was that of a holder of a bunch of "information needs", derived from work tasks and position in the decision-making hierarchy. Davies and Olsons Management Information Systems (Davis & Olson 1983), widely used as a textbook also in Finnish universities, is a good example of this thinking.
For anybody who has spent some time inside any organization it is quite clear that the idea of rational machine is a dangerously far abstraction of what is actually happening in organizations. Eventually this difference between the ideal and the real became too hard to hide, and the then dominant MIS school generated criticism, counter-reaction and opposition movements. Here I will discuss two of them.
One reaction was the increasing recognition that whatever changes when a computer system is designed and taken in the use, the change cannot be limited only to the technical system. Something else changes as well, and often this something else is more important for the success than the technical system itself. The something else has been called with various names like social implications, social aspects, social part of the system or social system. Thus the users are not only holders of information needs, but also in social relations with each other, and this of course complicates the picture quite a bit. The opposition started as a debate on suitable methodology to study information technology in development and use, which surfaced in the famous "Manchester conference" (Mumford et al. 1985), and the discussion has continued within information systems for example in (Nissen et al 1991).
Another reaction was even more clear with respect to users, even militant in defending their position. This movement was sc. Scandinavian participatory design, with promoters in Norway, in Denmark and in Sweden. The members of this tradition saw that users are like skilled craftsmen: they have vital tacit knowledge that can radically improve the quality of a system to be designed; at the same time their interests must be protected in the process of design. Both of these goals can best be achieved by taking users into the design process as active participants. Pelle Ehn's dissertation (Ehn 1988) can be seen as an intellectual exposition, the book Computers and Democracy (Bjerknes et al. 1987) as a political manifesto of the movement. The ideas of participation and democracy had a larger, mainly European, audience at that time, as documented in the books System design for, with, and by the users (Briefs et al. 1985) and Systems design for human development (Docherty et al. 1987), for example.
Later, in early 1990s participatory design gained renewed interest, this time in U.S. and a new tradition has been formed around a series of Participatory Design conferences. The U.S. variant is pragmatic and practical, not interested in theoretically reflecting the nature of users, use situations, or organizations, and the militant aspect has been watered down. (Schuler & Namioka 1993) is a good overview of this movement.
1980s: User as a source of error the influence from human factors and psychology
The second wave of influence came from ergonomics and from information-processing cognitive psychology. The idea of user as a source of error has similarities with the previous one, and some of their roots are common, but the main channel of influence has been another. The development that led to the emergence of the field of Human-Computer Interaction has two roots: the first started in the WW II, when the complexity of rapidly developing military technology exceeded the capabilities of humans operating it, and the second in process automation, where humans had to fill the not-so-easily automated gaps in the automated processses. The field where these issues were studied was called "ergonomics" in Europe and "human factors" in the U.S., and initially it was a study of human limitations and fallibility. The initial areas of study were human physical and physiological capabilities, but later, when computers came in the picture, the study was expanded into the cognitive realm: users were seen as similar information processors as computers, only hopelessly slower and more unreliable. This was the starting point when the area of Human-Computer Interaction really took off in 1980s.
The human factors perspective on users leaves off quite a few human features, so it wasnt a wonder that the view also generated intellectual reaction and opposition. These did surface both in the area of ergonomics itself and in human-computer interaction. In ergonomics a demand that ergonomics should be a more holistic study of humans instead of their limitations was voiced already in 1960s, and the discussion has continued since then. The traditional human-computer interaction based on the idea of a cognitive machine got increasingly criticized during the 1980s from two directions: on one hand the narrowness of the concept of a user was criticized from a theoretical and psychological point of view: as users humans are much more than only cognitive machines, and this should be reflected in the concepts and theories used. "From human factors to human actors" is the name of Liam Bannon's polemic, insightful paper (Bannon 1991), which became a battlecry against the traditional way of conceiving HCI my own battlecry as well, in fact.
Another direction of criticism was coming from practitioners, because the traditional academical HCI was quite incapable of developing any practical tools for design. The demand for efficient methods to ensure at least minimal quality of use for software - developed for the expanding personal computer market - was increasing rapidly. Eventually, the practical people did not wait for the academy to solve the problem they started solving it themselves, and this marked the birth of the usability movement. The usability movement was not at all interested in problematizing and theorizing on the nature of the user. It was and is purely practical in its interests, and it adopted the technocentric view described in the beginning of the paper: users are those who sit next to our systems, and they are interesting, because they use our systems. In fact, it is the usability movement that has made the term user popular: "know the user" was and is their battlecry and has been my own battlecry as well, as a practicing designer and teacher. Gould's "How to design usable systems" (Gould 1988) is a good articulation of the goals of the movement and still an adequate practical guide for design as well.
From the early 1990s on the usability movement took the HCI community by storm, because it was able to deliver what was needed: working methods to design better systems. It became rapidly the dominant paradigm of HCI and it still reigns unchallenged. One of the results has been that the theoretical and methodological discussion on what actually is the user, the object of the study, and how this object should be studied, has almost completely died down, and the frontier of this discussion has moved into another field, computer-supported cooperative work (CSCW).
1990s: Users as partners in social interaction - the influence from anthropology and microsociology
The next enrichment of the user concept was a result of influence from anthropology and microsociology within the new field of computer-supported cooperative work (CSCW). After its introduction in the later part of the 1980s, the field solidified and started to influence other diciplines during the1990s. There are a number of reasons why CSCW emerged, and a couple of them can be mentioned here. On one hand, although cooperation is why organizations exist and thus all multi-user systems have from the beginning been supporting cooperation in a form or another, the traditional database-oriented systems did their best to hide the interactions between different users and isolate them from each other. With the increasing use of local networks and independent workstations this hiding was no more possible, because the interactions between users became suddenly visible and often they had to be explicitly designed.
On the other hand, there was a certain void in the research on development and use of technology in organizations, a void waiting to be filled. MIS research had, in its search for a key to understand the mystery of the social, been mostly turning to "grand" social theories which, because of their very grandness, were quite weakly armed in dealing with the details of everyday life, which is the level where systems are both designed and functioning. Thus there was a need for research approaches capable of dealing with small enough grain size, and that was what anthropology and everyday sociology had been doing since long. Thus researchers with backgrounds in anthropology and ethnometodologically oriented sociology have got influential positions in the CSCW community, organized around the Journal of CSCW and the series of European and U.S. CSCW conferences.
Within CSCW, users are not only seen as active actors acting purposefully in contexts, theview advanced within the HCI field, but also partners in an ongoing and continuous social interaction with various communities. Moreover, their thinking is shaped by their personal histories and by their membership in those various cultural communities. The interaction enables user to make sense on situations and to give meanings to things. It makes them capable not only to use a system, but to use it in unanticipated ways, to work around the system, and to make it a tool or environment for all kinds of ongoing organizational games.
The CSCW idea of the user is probably the richest currently available within the field of information technology. It has had some influence on the research in HCI and related fields, but it has not yet influenced practical design in a remarkable way, like the HCI ideas have already done. Towards the end of 1990s still one more way to see the user emerged in the discussions, that of seeing the user as a consumer, influenced by design professions and marketing.
2000s: Users as consumers - the influence from design and marketing
Until the later part of the 1990s the user was dominantly considered to be the user of a system needed in work or at least in a work-related activity. This started to change with the increasing demand to design computer software or smart products for consumer markets. The change was initiated already in the1980s with computer games, but games have carried such a unfavouring stigma within the HCI community, for example, that game developers were efficiently isolated from the rest of the interface design community, although games were continuously pushing the leading edge of interface development in many ways. The threshold between the communities started to get lower only in the end of the last decade of the millenium.
Perhaps the main factor in the development towards a consumer orientation has been the introduction of mobile phones, which have developed during the1990s into true consumer products, with ever stronger associations with clothing accessories. When people started to use the phone as a means for self-expression, a new concept of the user was needed a user who besides rationality and reason has also emotions and needs for pleasure and self-expression. Such an idea of the user was already well-established in other areas of consumer products, and thus the fourth wave of influence has been from marketing and industrial design the latter discipline, interestingly, having so far been quite absent from any HCI related discussions.
This current wave of influence, which is clearly gaining momentum, is not by any means homogenous: on one hand there is a marketing-oriented tendency to treat users as faceless figures belonging to various market segments, on the other hand there is a real interest in drawing from experiences of the design professions in a more holistic sense, as manifested in Terry Winograd's Bringing Design to Software (Winograd 1996).
What is still missing?
Although the concept of the user has indeed undergone a number of changes, and although in general there has been a tendency towards a richer and more multidimensional interpretation (altough it is still to be seen whether the last wave of influence will enrich the earlier concepts or try to replace them), some perspectives are still missing. I will highlight a few of them here.
Users as learners. One perspective that is very important but badly neclegted is the perspective of learning and dynamics in the situation of use. When designers think about users, they rarely think of learning and dynamics, neither in the small users as learning to master the technology device itself nor in the large users as learning something through the technology. The first issue is addressed a bit in HCI, but with emphasis on not obstructing learning rather than supporting it. A tradition of pedagogically oriented software that has tried to address the latter question, but yet without very convincing and conclusive results that support for learning in the large can be designed. Learning being a central human characteristics, the omission of it from the picture of the user is interesting.
Users as shapers and changers of their environment. In its interest in social interaction, CSCW research has often had a tendency to overlook the fact that users do not only interact socially and make sense and create meanings, but they are in a continuous interaction with their material environment as well, as makers and creators of the material aspects of their world. And often these changes in the material world are what matters, why the use of technology is needed in the first place. The idea that users have bodies and that they move around has been infiltrating into design consciousness already for a while, not least because of the increasing interest in mobility. But the users active, transformative existence in the material world has still got less notice.
Users as becoming something else by using a device or a system. A more radical idea on learning and change, the idea of functional organs (Leontjev 1978), questions the dividing line between a user and the thing being used. When our relationship to the world surrounding us is remediated by a new tool, system, or service, we not only use it, but the use changes us as well. The tool, system or service becomes a part of us in a very literal sense. The more we use it and the more skilfully, the more essential part it becomes. Like one of my colleagues said a couple of days ago, having forgotten his electronic calendar at home: "I feel like missing a limb". The border between us and the world is not our skin, the border is where our mediated relations with the world can take us, the limits of our ability to reach out, contact and change the world through tools, systems and services. There is nothing new in this, of course: since the invention of the first stone-axe an ever larger part of "us" has been in this sense artificial, assimilated from the culture we live in we are all "cyborgs" by nature, so to say. But these new information technology tools, systems and services are going to transform us over and again into different cyborgs, and it would be good to be able to give this some thought already in the time of design.
There is a bit more discussion about this idea in (Kaptelinin and Kuutti 1999). I have a feeling, that this is the most promising venue of thought to proceed to get rid of the term of a "user"...
Bannon, L. J. (1991). From Human Factors to Human Actors: The Role of Psychology and Human-Computer Interaction Studies in System Design. In J. a. M. K. Greenbaum (Eds.), Design at Work: Cooperative Design of Computer Systems (pp. 25-44). Hillsdale: Lawrence Erlbaum.
Bjerknes, G., P. Ehn, M. Kyng (1987) Computers and democracy : a Scandinavian challenge. Aldershot : Avebury.
Briefs, U. , C. Ciborra, L. Schneider (1983) Systems design for, with and by the users : proceedings of the IFIP WG 9.1 Working Conference on Systems Design For, With and By the Users, Riva del Sole, Italy, 20-24 September . Amsterdam : North-Holland
Davis, G. B. , M. H. Olson (1985) Management information systems : conceptual foundations, structure and development , 2. ed. New York, N.Y. : McGraw-Hill.
Docherty, P. et al.(eds.) (1987) System design for human development and productivity: participation and beyond. Amsterdam : North-Holland.
Ehn, P. (1988). Work-Oriented Design of Computer Artifacts. Stockholm: Arbetslivscentrum.
Gould, J. D. (1988). How to Design Usable Systems. In M. Helander.(ed.) Handbook of Human.Computer Interaction. Amsterdam, Elsevier/Norh-Holland: 757-789.
Kaptelinin, V., Kuutti, K. (1999) Cognitive Tools Reconsidered: From Augmentation to Mediation. In Marsh, J., B. Gorayska, and J.L. Mey (eds.) Humane Interfaces: Questions of Methods and Practice in Cognitive Technology. Elsevier/North-Holland, Amsterdam, pp. 145-160.
Leontiev, A. N., 1978. Activity, Consciousness, and Personality. Prentice Hall. Englewood Cliffs, NJ
Mumford, E. & Hirschheim, R. & Fitzgerald, G & Wood-Harper, A.T. (1985) Research Methods in Information Systems. North-Holland, Amsterdam.
Nissen, H.-E. , H.K. Klein, and R. Hirschheim (eds.) (1991) Information Systems Research Arena of the 90´s., North-Holland: Amsterdam.
Schuler, D., A. Namioka (eds.) (1993) Participatory design : principles and practices. Hillsdale (NJ) : Lawrence Erlbaum Associates.
Winograd, T. (1996) Bringing Design to Software. New York; ACM Press.