Issue
Article
Vol.30 No.4, October 1998
Article
Issue
Issue |
Article |
Vol.30 No.4, October 1998 |
Article |
Issue |
On April 19 and 20, for the seventh consecutive year, a group of researchers from the CHI community gathered for a symposium devoted to fundamental issues in research. It can be both invigorating and intimidating to attempt to capture two days worth of vigorous discussion in a few pages. We know we cannot fully succeed, but it revives the spirit of the event to try. In this article, we attempt to relate some of the background of the event, its structure and formal content, and a glimpse at the discussions and interactions that filled two days.
Each time we mention the Basic Research Symposium (BRS), we are asked this central question. Indeed, if CHI is a conference that publishes the premier research in HCI, than what is the point of having an additional two-day research symposium?
There are many answers, but two of them most directly motivated us.
First, while CHI is a wonderful venue for learning about all sorts of research, it cannot provide the time and close contact needed for interdisciplinary research discussions to develop and mature. One key feature of the BRS is that it offers time -- time for a collection of researchers to meet each other, learn from each other, and teach each other. Researchers from computer science and psychology and sociology and information science; researchers from all ranks of academia and from industry, researchers from all parts of the world; some researchers just starting their careers and others well-established -- all together for two days. Indeed a common point made in feedback from participants is that the BRS provides them with an opportunity (in some cases, even "forces them") to do what they'd like to do at CHI, but cannot for fear of missing too many important activities in their own specialty.
Second, the BRS serves as a venue for presenting early research ideas and results to a collection of colleagues while there is still time for feedback to help steer or strengthen the research. While much of the work presented at prior years symposia has subsequently been published at CHI and other SIGCHI conferences, an early presentation at the BRS has often helped direct the researcher to interesting results, methods, techniques, and prior work known in parts of the HCI community, if not in each specialty. As part of this practice of sharing early research, the BRS does not distribute proceedings to non-participants. We feel it is important to leave control over the dissemination of these early works in the hands of the researcher. Accordingly, this article summarizes themes and discussions, but not particular research presentations or papers.
Thus, the BRS serves as a double-incubator of sorts. In the process of incubating new research, the participants interact, disagree, and even argue, bringing the whole group to a healthy boil. At the same time, the participants learn from and about one another, leading to mutual respect and better understanding among several disciplines/fields.
The basic structure of the BRS remains relatively stable from year to year. We were quite happy to have inherited a very successful organization from Leon Watts and Susanne Jul, giving us little reason to tinker.
Some of the key ongoing features are:
One way to ensure that the BRS participants are serious about research, and have serious research to discuss is to invite participants based on a thorough review process. Each position paper submitted was reviewed by three to five program committee members, with detailed reviews returned to the author. The focus of the review process is not to judge the work for publication, but to assess whether the work contains seeds of interesting research that can be developed at the BRS. Reviewers also were asked to suggest the appropriate format for presentation and to assess strengths and weaknesses of the submitted work. As a result, reviews tend to be detailed, and they are often cited as one of the benefits of the event. We also invited short position statements from researchers who were interested in participating but did not have a paper to present. As a whole, the review process ensured a serious and interesting group of participants and a provocative set of papers and presentations.
We would like to express particular appreciation to the members of the 1998 BRS program committee for their hard work on short deadlines: Matt Conway, Prasun Dewan, Suzanne Jul, Satoshi Matsuoka, Dag Svanes, Lisa Tweedie, Yvonne Waern, and Leon Watts.
This year's program was divided into 30- and 45-minute presentation and discussion slots, small-group activities, and whole-group discussions. Having learned well from previous chairs, the program contained substantial sections of changeable time, to allow the group to pursue a topic of interest in greater depth.
Increasingly, the BRS has been organized using much of the same technology that we study in the HCI community. The submission and review process was entirely electronic, and the program and proceedings were distributed through the web. As the sophistication of our tools grows, we also find that we learn more from the failures and successes in using them.
For many years, an important part of the BRS has been scheduling large group meals for lunch and dinner. Overcoming the difficulty in finding restaurant space at noontime on a Sunday in downtown Los Angeles, the group ate together for four meals over two days. At these often lengthy meals, we continued discussions from the day, learned more about each other, and built relationships and friendships. The power of these relationships was illustrated repeatedly through the CHI conference as we observed participants, even first-timers, walking and talking together during the rest of the week.
The topics of this year's presentations were very varied. Sunday sessions revolved around issues of methodology and broader understanding of HCI, with a mix of theoretical and applied research talks. The morning session brought us discussions on understanding people, computers, and the layers of HCI. These were followed by a compelling case for a more literate development process. The afternoon brought forward discussion of the use of dynamic systems, particularly cellular automata, for exploring human decision-making and user interfaces. We also learned about interesting research on evaluating the conditions that alter the effectiveness of different knowledge elicitation techniques.
Monday focused more on HCI systems research, with discussion of visualization spreadsheets, haptic interfaces, and interface ideas revolving around large image databases. We also had a paper reporting on an evaluation of different evaluation methods.
As symposium chairs, we had three remaining responsibilities:
The first of these responsibilities was unexpectedly pleasant. Even during the usual post-meal lulls, participants were exceedingly eager to discuss and provide feedback. At one time we made a point of noting participation, and found that almost all participants were contributing to the discussion regularly.
Keeping things on-time was a challenge. We were forced to cut-off interesting discussions of several research efforts to be fair to the researchers wiith later presentation slots. This led to our third, and most critical responsibility -- finding more time.
It would be fair to say that the group work and discussion did not follow precisely the path we planned. Over the course of two days, the group found its own topic of interest: breaking down the barriers between the subfields of HCI. Motivated by the broad talks on the first morning, we set about exploring the barriers, and possible solutions. The following list is an attempt to re-capture some of the sense of these discussions, which occurred both in small groups and among the larger group as a whole.
It is illustrative that the group encountered two serious problems that prevented the work from being completed. First, there was substantial disagreement on whether concepts or the systems that illustrate them deserved greater coverage on the timeline. (The former group gained the upper hand, with the latter group still believing that systems such as the Macintosh and Windows deserve a place for their impact.) Second, we realized how extensive a project this was, potentially requiring substantial research into the history behind several ideas. Some of this work exists in published form, but little is in a form that can be easily connected to the psychology, human factors, and other timelines of the development of HCI.
We are wasting no time in preparing for the CHI 99 Basic Research Symposium, to be held at CHI 99 in Pittsburgh. Yvonne Waern and John F. McGrew will chair the eighth BRS. For further information, please contact them at yvowa@tema.liu.se or jfmcgre@ptss.com or see the CHI 99 BRS Call for Participation at http://www.cs.umn.edu/~konstan/BRS99.
While many people helped make the 1998 BRS a success, we particularly like to acknowledge two people on the CHI 98 conference committee who undertook great efforts to help make the CHI 98 BRS a success: Susan Dumais, workshops co-chair and Arnold Lund, conference co-chair.
Joseph A. Konstan is an assistant professor of computer science and engineering at the University of Minnesota. His specialty is HCI with a focus on collaborative information filtering, multimedia systems, and visualization applications. He has been a member of the BRS program committee for 1996 and 1997.
Jane Siegel is a Senior Systems Scientist at the Human Computer Interaction Institute at Carneigie Mellon University. Her research focuses on applying empirical research methods to measure complex group behavior and in assessing the impact of alternative technologies on communication, team performance, and organizational memory. She has been a member of the BRS program committee for 1996.
Joseph A. Konstan
Department of Computer Science and Engineering
University of Minnesota
Minneapolis, MN 55455 USA
+1 612 625-1831
konstan@cs.umn.edu
Jane Siegel
HCI Institute
Carnegie-Mellon University
Pittsburgh, PA 15213 USA
+1 412 268-6764
Jane.Siegel@cs.cmu.edu
Issue |
Article |
Vol.30 No.4, October 1998 |
Article |
Issue |