Andrew Dillon
Associate Professor of Information Science/Program in Cognitive Science
Indiana University
Why does so much Information Retrieval (IR) work not link with Human-Computer Interaction (HCI) research (and vice-versa)? Even though HCI is taught in many LIS environments, the division remains in the minds of many researchers and students. Yet IR is in part an ideal unit task for HCI type analyses, full of the variables that engage researchers interested in cognition, problem-solving, skill development etc., and focused enough to enable generalization and theory-building. Bridging these disciplines will challenge us on many levels, not only intellectually but academically, politically, personally. Are we up to it?
We all think we know what we mean when we employ terms such as "usability" or "user-centered" in reference to information technology, but ask two people to define those terms, or to clearly state the real differences between user-centered and supposedly traditional models of systems design, and the divisions can quickly appear. While we may realize that meaning is culturally constructed, such a position needs to inform, not excuse our terminological differences.
Those in the information science domain have long understood the importance of information access and use. Now the rest of the world is waking up to the possibilities afforded by new technology. We are in a position of influencing our own development in a manner that was inconceivable with other tools. Can we meet that challenge?
All of the above challenges could (and should) be seen as opportunities, but in the spirit of the original call, I will add a further selection of explicit opportunities for us:
Digital libraries bring together a critical mass of researchers with interests in both the human and the technological aspects of system design, which coupled with the funding possibilities, provides us with the means of performing real user-centered design as never before. No more excuses, freezing out of sub-groups, setting user requirements in stone, leaving evaluation to the end etc. Now we have the chance to get it right in a new domain, with the lessons of the last 20 years of UCSD to guide us.
Attempting to design digital libraries for optimum human use provides us with a testing ground for our theories and models. Unlike any laboratory, real-world scenarios are where we find out how well we understand users, their tasks, and our conceptions of use. Every artifact we create is an embodiment of our theories, their can be no better test of a science than this.
Digital libraries provide us with all of the above, plus the opportunity to create a new form of discipline - one that is part design, part science; part cognitive and part computer science; part engineering, part art; part sociology, part education. The opportunity lies in grasping this and nourishing it, not to sidetrack it with concerns for labels or intellectual territoriality.