Issue
Article
Vol.29 No.1, January 1997
Article
Issue
Issue |
Article |
Vol.29 No.1, January 1997 |
Article |
Issue |
Are you a seasoned conference goer? You know what you want to see. You carefully select the sessions which cover your area. You get annoyed at those mixed-up sessions where some papers are in your sub-field but others are in some different field of HCI and bore you to tears. You know what you want and you want what you know.
Well, if that sounds like you, you'll be glad you did not attend the Basic Research Symposium (BRS) at CHI 96.
In our vision statement for the BRS we said:
We wish to create a place where the fundamentals which underlie the study of human-computer interaction can be discussed in an informal yet challenging environment; a place which is truly a common ground whereon researchers from different disciplines can meet to present new developments and insights from their own field, receive stimulation from others, and celebrate their diversity.
And we certainly got what we asked for! Participants came from academia and industry, from psychology and computing, engineering and sociology. Methodology included experiment and observation, mathematical analysis and professional opinion. Topics ranged from media spaces to medical practice, from animated algorithms to agent-based architectures, from emotion and initiative in human--computer communication to educational applications, from bicycle repairs to Boeing aircraft maintenance, Petri nets to paradigms of interaction.
There were some themes which emerged during the symposium, especially in the chosen application domains, for example the use of media and (not surprisingly) the web. However, the real common thread was the focus on underlying issues, principles and theory: the why questions which lead you from the complexity of breadth (which bedevils all real design), towards deeper understanding.
This basic knowledge is not divorced from practicalities, as the industrial attendees at the symposium will verify. The gap between research and practice in HCI is often very close: theory is being used to drive design decisions and practical experience is distilled to critique and develop theory. However, immediate practical benefit is certainly not a criteria for the symposium! Who would have guessed that the study of finite groups, a very pure part of pure mathematics for 100 years, should suddenly prove so important in the development of quantum mechanics and so ultimately the design of semi-conductor devices on which our own field depends. In HCI we should expect no greater presentiment.
Also it is basic knowledge which transfers between domains and, most important, between technologies. In a technology determined field like HCI we are in danger of having to start from scratch each time a new development appears: as if the builders of Alexandria were asked to build modern Atlanta. It is the underlying concepts and principles which allow one to take knowledge and re-apply it in a different context. GOMS was developed in the context of text-based interfaces, but is still being used today in graphical interfaces, audio interfaces and virtual reality. The method has been developed and modified, but the central concepts are still in use because they are based on underlying cognitive theory.
Having a technically diverse programme is not enough. In our vision statement for the symposium we said:
This will not be a market place where we each sell our own projects, products or personalities, but instead a meeting place where the successes and failures, joys and frustrations of our ongoing work can be openly offered in the expectation that critique rather than criticism will be offered in return.
Again, our hopes were more than fulfilled. The discussion was direct and dynamic, but we recall no occasion where a questioner attacked a piece of work for not employing the techniques of the questioner's own field. Instead, critique was based on the understanding that criteria and methodology differ between fields. This acceptance of differences was paralleled by an openness and frankness: most presentations ended with questions rather than final answers: "do you know of similar work", "do you have any ideas on", "I'm not sure the right way to continue this, can you help"; in one case the technical work touched very directly on the presenter's personal life, at another time, the entire room stood up and shook their legs in time to emphasise a point. We are all vulnerable at conferences and workshops both as we present our work and as we ask questions. The combative style of many conferences can neither allow new researchers to develop, nor allow innovative, but speculative, ideas to emerge.
Although the symposium was clearly forging a common ground at the level of personal and social acceptance, we all wondered whether this was just a "warm fuzzy" feeling. Perhaps we were just communicating at a superficial level, simplifying our own areas to make them accessible to others, but eliding the areas of difference. Our group sessions used the conference theme of "Common Ground" to explore this, to see where we could really find common ground between different academic disciplines and between academia and industry.
There was general agreement that this warm fuzzy feeling, although not a sufficient end-point, was certainly the proper foundation on which to build -- actually being together and entering into deep discussion with people of different backgrounds. However, various barriers were also discussed.
As always language featured high on the list of the difficulties. What do words like metaphor, context or task mean? But we realised that it is not just the names, but the conceptual and cultural frameworks which differ so markedly between disciplines. Some previous symposia have featured `scenario sessions' where several participants attack a similar problem from different perspectives. In general, the use of concrete examples allows us to meet in the real world and avoid some of the misunderstandings that arise through different use of words and ideas. However, it was also noted that common representations were needed to make an effective bridge with industrial practice. We have so many notations and esoteric word uses that, even if we can do the translations, there is little hope for someone wanting to actually use the knowledge.
Groups also considered the very notion of common ground between disciplines. Is it somewhere we come to from our own disciplines, or somewhere we stay, founding a new discipline. This is an issue which resurfaces again and again in HCI, for example Brian Shackel's plenary at INTERACT'87, and has certainly been discussed at previous symposia. However, it is not just a philosophical question, but one at the heart of HCI education, critical to those doing doctoral research in HCI and to those implementing usability strategies within industry. Do we want HCI experts, or do we want experts in their own specialist disciplines who are able to come to that common ground and discuss HCI issues together? Furthermore, are the examiners of doctoral theses ready to accept that the methodologies of their own discipline may not be applied as rigorously or as purely in work which bridges this common ground?
The basic research symposium must surely be one of the most international parts of CHI. Only a little over half the participants were from the US, the rest coming from at least 12 different countries. For several years, the CHI conference committees have been aware of the implicit US cultural bias in the conference and have made efforts to address this including the international representatives on the committees, the mentoring programme and the coming Development Consortium at CHI 97. Why is it that the BRS is so successful in this respect?
Three reasons spring to mind. First, the very fact that BRS attendees are willing (and eager) to understand the contributions of diverse disciplines means that they are also willing to accept those from their own discipline with different cultural approaches. Second, related to the first, is that the emphasis on the fundamental theory of HCI means that, although we all believe that methodology is important, we are focused on the outcomes and deeper results. The third reason is more prosaic, but probably most critical. It is that theoretical research tends to be less dependent on state-of-the-art equipment and is thus more financially accessible. If reviewers are more swayed by glitsy interfaces than by the design principles they embody, then there will always be a bias towards richer countries and richer institutions. This is surely an issue which needs to be addressed by all the technical sub-programme chairs within CHI, perhaps consulting the past chairs of the East-West HCI conference who have valuable experience in this respect.
Before we become too pleased with ourselves there were some limits to the international flavour of the BRS. First, the majority of the non-North-American participants came from Europe. Second, when giving their contact details only 2 of the US participants added US or USA at the end of their addresses (the rest of the world are supposed to guess that MS doesn't mean Mid-Sahara) and only one gave an international telephone prefix. Clearly as with other parts of the CHI programme the BRS and its attendees have lessons to learn in this respect.
Given the diversity of material in the symposium, this article has focused more on the spirit of the symposium than the content. Of course, the full impact of the technical discussion cannot be captured on paper. However, the final programme and position papers for CHI 96 BRS are available at:
http://www.soc.staffs.ac.uk/~cmtajd/conf/brs96/
Those of you who didn't recognise themselves at the beginning of this article may be interested in attending the CHI 97 basic research symposium. The co-chairs are Susanne Jul (University of Michigan, USA) and Leon Watts (University of York, UK) and full details can be found on the CHI 97 web page (http://www.acm.org/sigchi/chi97/).
If you did recognise yourself, perhaps you ought to come along anyway to broaden your perspectives!
Lots of thanks to the BRS organising committee who gave their time and support: Berardina (Nadja) De Carolis, Hilary Johnson, Joseph Konstan, Janni Nielsen, Marilyn Panayi, Dag Svanes, Leon Watts and Cathleen Wharton. Also thanks to all those who attended the symposium and made it such an enjoyable and exciting event.
Alan Dix is Associate Dean at the School of Computing, Staffordshire University, Stafford, UK. His research interests include CSCW, aspects of time in interface design, applications of formal methods in HCI and just about anything.
Francesmary Modugno received her PhD in computer science from Carnegie Mellon univesity and is currently a research associate at the University of Washington. Her interests include human-computer interaction, software engineering, empirical studies of programmers.
Alan Dix
Staffordshire University
PO Box 334, Stafford,
ST18 0DG, UK
A.J.Dix@soc.staffs.ac.uk
http://www.soc.staffs.ac.uk/~cmtajd/
Francesmary Modugno
University of Washington
Box 352350, Seattle,
WA 98115, USA
Issue |
Article |
Vol.29 No.1, January 1997 |
Article |
Issue |