Human-Computer Interaction

Volume 15 (2000), Numbers 2 & 3

Introduction to this Special Issue on

New Agendas for Human-Computer Interaction

Wendy A. Kellogg, Clayton Lewis, and Peter Polson

The five articles contained in this special issue of Human-Computer Interaction were originally presented at a workshop at Snow Mountain Ranch, Colorado, in February, 1999. That workshop marked both the tenth anniversary of the Human-Computer Interaction Consortium (HCIC) and the end of the millennium. Accordingly, the workshop theme, Human-Computer Interaction In the 21st Century:  Prospects And Visions, called on participants to take stock and look forward to new directions for HCI. The results we think are thought provoking.

To get a perspective on the evolution of the field, it is useful revisit the analog of this special issue – an earlier Human-Computer Interaction special issue on the “Foundations of Human-Computer Interaction” (Card & Polson, 1990), which originated from the first meeting of HCIC in 1989. In doing so we are struck by two observations. First, some of the work in the earlier issue has proved to be foundational; indeed some of the papers are cited in this issue. Second, the comparison reveals how the foundations of the field have shifted, in that issues of major focus in the present issue were barely hinted at in 1990. In particular, the two collections clearly show the growing awareness that design for the isolated individual user is inadequate to produce systems of practical value in supporting the reality of people working together.

All of the papers in this special issue reflect the field’s evolution towards more complex and more contextualized views of interactive systems and their use.  This consideration of more realistic tasks, models, and situations of use can be seen in the various domains of inquiry represented here, from task-action models in the GOMS family to support for collaborative work.  It is also an impetus for more sweeping proposals to rethink how HCI research is organized and how it relates to other areas of science. The papers herein collectively offer a set of proposals for the future:  new agendas that can take research in human-computer interaction and computer-supported cooperative work to the next level.

These papers are at the same time conservative and radical. They are conservative in that they are skeptical of the popular hype prevailing at century’s end, the grandiose visions of an all-powerful technological transformation of existence. They are radical in that all sharply question prevailing methods and approaches in HCI and CSCW.

The conservative theme can be heard clearly in Ackerman. Contrary to millennial optimism, he argues that we remain unable to develop successful technological support for social processes. Similarly, Olson and Olson argue that the media-touted view of communication technology erasing distinctions of time and space is seriously in error. At another level, Bhavnani and John warn that harnessing new technology requires far more than creating a usable or learnable interface and putting it in front of people who will then absorb and master it with little effort.

The radical theme is also discernable across these papers. Whittaker, Terveen, and Nardi are strongly critical of prevailing research practice in HCI, building on the earlier critique of Newman (1994) that called attention to the disproportionate emphasis on radical innovation rather than evolutionary improvement in the field. Bhavnani and John’s attack on the incompleteness of learnability and usability as goals for system design strikes not only at attitudes towards technology, but also at common thinking within the HCI research community. Both the Olsons and Ackerman call for significant reorientations of research; techno-hype must be supplanted by hard work on a deep intellectual agenda. Finally, Furnas boldly outlines a theoretical framework that challenges HCI researchers to think seriously and systematically about the multifarious contextual interactions that will define the success or failure of systems.

While these themes tie the papers together, each also presents a specific critique of the field and an agenda for moving the work forward in new ways. Here is a preview of their arguments.

Whittaker, Terveen, and Nardi urge that the HCI research community should borrow the reference task approach that has been effective in organizing research in speech recognition. They argue that HCI research so far has suffered from dispersion, with each worker or research group setting an agenda only weakly related to what others are doing. Under the Whittake,r Terveen, and Nardi proposal, HCI workers would propose and agree on a relatively small set of common problems, embodied in sample user tasks, and would focus effort on providing improved support for these. Over time, the standard of accomplishment in the field would rise, as groups succeed in bettering the measured quality of earlier solutions.

There is no question that, if this proposal is adopted by a quorum of workers, the HCI landscape of the future will be very different from today’s. But questions abound. Do we understand any important interactive tasks well enough to propose performance measures adequate to guide progress over an extended period?  Will tasks remain comparable over time, given the rapid evolution of technologies?

Bhavnani and John show that users of complex applications do not spontaneously use strategies that make use of operations unique to computers (e.g., defining aggregations of objects and then using a single command to operate on all objects in the aggregate).  They use GOMS models to show that these strategies define methods that have very different structures from typical GOMS methods (e.g., a method to delete a word). 

These results have fundamental implications.  Landauer (1995) argued that computer applications have not improved the productivity of knowledge work because flawed product development processes produce applications with poor user interfaces.  While his conclusion is certainly justified, Bhavnani and John show that even if the interface to a complex application is relatively good, users do not spontaneously learn and use strategies that would make them far more productive. In addition, strategies useful in complex applications are not described in training and reference documentation. The dominant paradigm for user training is to provide a limited introduction to the user interface and basic functionality of a complex application.  Users are expected to acquire advanced skills as they gain more experience with an application.  Minimalist documentation (Carroll, 1990) and user interfaces that support learning by exploration are well-known, successful methodologies designed to support this training paradigm. Bhavnani and John's results show that this paradigm has fundamental limitations; it will not support the development of the kinds of powerful strategies they describe.  However, users can be trained to use these strategies and they present preliminary evidence that these strategies are retained and generalized. Thus, they argue that the agenda for HCI needs to shift away from support for learning by exploration to support for training users on productive strategies.

Widespread deployment of organizational intranets and access to the Internet has led to expectations that individuals and teams will be able to seamlessly collaborate remotely.  In “Distance Matters,” Gary and Judith Olson argue that this simplistic view is wrong.  The Olsons review over ten years of field and laboratory investigations of collocated and non-collocated synchronous group collaborations, providing a significant new analysis of factors contributing to a group’s ability to collaborate over distances and an outline of what may be possible in the future as technology evolves and users’ and organizations’ sophistication in the use of technology and collaboration increases.  In particular, they identify four key concepts:  common ground, the coupling of work, collaboration readiness, and collaboration technology readiness.  This analysis frames the discussion of distant collaboration in terms that range from the artifacts and groupware that support collaboration to complex and subtle factors including context and trust, time zones, and cultures.  Although the Olsons remain skeptical about the potential for effortless remote collaboration – no matter how good the technology becomes – their focused review of a large body of empirical evidence and analysis of the critical factors influencing distance collaboration sets the stage for future work in this area.

Ackerman takes skepticism to the next level by pointing out that while technical systems are still too brittle and rigid to accommodate the subtlety of social processes, this is no longer due to ignorance on the part of technologists about the relevant social processes to support.  He calls into question whether any amount of understanding of social requirements can lead to the creation of sufficiently responsive systems.  Dubbing this the “social-technical gap,” Ackerman argues not only that it may be impossible to abolish, but also that embracing and addressing it is the heart and soul of CSCW’s mission – its primary intellectual challenge.  After reviewing seminal findings in CSCW, he considers the gap in more detail through an example that focuses on privacy in information systems.  This leads to a discussion of potential resolutions for the social-technical gap, and a proposal to adapt Simon’s (1996) idea of a science of the artificial as a framework for future work in CSCW.  The proposal is provocative, suggesting that the way forward is to recognize the social-technical gap as fundamental, and the prospect for building a science that bridges the gap as the central intellectual mission of the field.

Going even further, Furnas outlines an intellectual framework that ties HCI research into broad themes that run through much or all of science. He develops an analysis of interacting systems that permits issues in HCI to be identified with issues general to adaptive systems of all kinds, in biology, economics, or management. The key abstraction is the MoRAS, a mosaic of responsive adaptive systems. In the mosaic the success or failure of one subsystem is tied to the pattern of interactions with other subsystems. Information technology has to be understood not only as supporting some kinds of subsystems, but also as altering the way subsystems are coupled. Understanding the coupling is crucial to successful design.

Situating HCI in the MoRAS means not only that, for example, economic perspectives are relevant to HCI, but also that some problems studied in HCI can be seen as fundamentally the same as some problems studied in economics. This is a bold program, and one that provides a fruitful basis for a whole range of new collaborations for HCI.

Taken together, these papers do a remarkable job of taking stock of where we are and looking forward to what comes next.  We applaud the authors for their willingness to do the hard work of looking back and forward in the same moment, and for their courage in recommending these new directions to all of us for the benefit of the field.  We hope you will find their thoughts as stimulating as we have and find yourself reflecting on how their proposals could affect your own area of work.

 References

Carroll, J.M. (1990).  The Nurnberg funnel:  Designing minimalist instruction for practical computer skill.  Cambridge, MA:  MIT Press.

 Landauer, T. (1996).  The trouble with computers:  Usefulness, usability, and productivity.  Cambridge, MA:  MIT Press.

 Card, S.K., and Polson, P. (Eds.) (1990). Special Issue on Foundations of Human-Computer Interaction, Human-Computer Interaction, 5, 119-344. Mahwah, NJ: Erlbaum.

 Newman, W. (1994).  A preliminary analysis of the products of HCI research, using pro forma abstracts.  Proceedings of the CHI’94 Conference on Human Factors in Computing Systems, 278-284. New York: ACM.

 Simon, H.A. (1996).  Sciences of the artificial, 3^rd edition.  Cambridge, MA:  MIT Press.

 Articles In This Special Issue

Ackerman, M. S. (2000). The intellectual challenge of CSCW:  The gap between social requirements and Technical Feasibility. Human-Computer Interaction, 15, xxx-xxx.

 Bhavnani, S. K., and John, B. E. (2000). The strategic use of complex computer systems. Human-Computer Interaction, 15, xxx-xxx.

 Furnas, G. W. (2000). Future design mindful of the MoRAS. Human-Computer Interaction, 15, xxx-xxx.

 Olson, G. M., and Olson, J. S. (2000). Distance matters. Human-Computer Interaction, 15, xxx-xxx.

 Whittaker, S., Terveen, L., and Nardi, B. A. (2000). Let’s stop pushing the technology envelope and start addressing it. Human-Computer Interaction, 15, xxx-xxx.