B  U  L  L  E  T  I  N

Annual Meeting Coverage

Information Science Fiction

by Howard D. White

Editor’s Note: Howard White delivered the following address at the Awards Luncheon at the Annual Meeting in Providence, Rhode Island, on November 16, 2004, where he received the ASIS&T Award of Merit.

Howard D. White is affiliated with the College of Information Science and Technology at Drexel University in Philadelphia (PA 19104). He can be reached by email at whitehd@drexel.edu

Thank you, everyone. This is a very untypical day for me, and I deeply appreciate the work of all those who made it possible.

If I had to pick a situation that typifies my life, it would be me standing in front of a bunch of books I haven’t read. Wherever I go and however long I live, I always seem to be doing this; for some reason I like standing there, taking books down and looking into them and putting them back. I have done it at home, at school, in my office, in bookstores and in libraries. I suppose that this fondness for browsing is why I trained at one point to be a librarian: they get paid to stand in front of a bunch of books they haven’t read. But I never worked as a professional librarian; instead, I became an academic, and this allowed me to characterize my relation to books in more stately fashion as “the human-literature interface.” I don’t apologize for that phrase; one has to set the broader context when one is doing newish things, like mapping authors in intellectual space based on their cocitation counts. But as I looked at these maps over time, I came to realize that all I had created was a miniature version of a bunch of books I hadn’t read.

            Even so, whether I’m browsing in library stacks or looking at cocitation maps, I’m reminded of the central concern of information science – that is, effective intermediation between people and literatures. People have a sharply limited capacity for absorbing recorded information. They can increase this capacity only by extending it over time – that is to say, over chunks of their lives, which are finite. People are also sharply limited in where they can be at a given moment. Literatures, in contrast, contain vast amounts of information, they extend globally over highly fragmented space, and they are timeless. They and people could hardly be more different. Yet people want to avail themselves of the powers and pleasures found in literatures, despite the mismatch at the point where they and literatures meet.

One way of conveying this mismatch is to speak of “you versus the literature.” The phrase suggests overload – the fact that, from childhood on, you have too many things to read or otherwise attend to. But it also suggests resistance – the fact that you learn to filter even the good stuff and restrict what you take in. Beyond that, you can sometimes mean more than one person – it can imply you the individual plus all the others who cut literatures down to size on your behalf. Sometimes they do this by singling out just those writings that address your interests or your specific questions. Other times they shorten what has to be read through abridgement and synthesis. One way or another, you as reader go from glut to gist. Or at least that’s how it’s supposed to be.

Let me pretend I’m a science fiction writer who is coining names for two imaginary sciences that do R&D on the activities just mentioned. Both overlap present-day information science, but they do not limit us to it, because both can make use of research from many other disciplines. The first is a behavioral science that deals with filtering. It is the study of how people choose what they will and will not read; how they ask for writings and how they judge what they get. It is the study of selection from diverse information sources – more broadly, the social psychology of information seeking. I’ll call this Eclectics. The second science is concerned with text. It deals with the resources language affords for making overviews, condensations and indexing schemes, for reducing messages to briefer compass, for putting much into little, perhaps with pictures rather than words. I’ll call this Synoptics. The unit of analysis in Eclectics is people, in Synoptics, it is writings and other graphic records. Eclectics and Synoptics are both motivated by the condition of information overload; they focus on how you – whether singular or plural – cope with ever-growing literatures.

Needless to say, the two fields are highly interrelated. Eclectics defines you as a bundle of interests and questions. Synoptics defines the literature as bundles of satisfactions of your interests and answers to your questions. The engineering task is to effect a match, or reduce the mismatch, between you and your literature-based satisfactions and answers. The scientific task is to find out enough about you and about the micro- and macro-structure of literatures to make this possible. Of course, the great bridge between people and literatures should be language that they have in common, but the major finding of Eclectics to date is that many people don’t know the right language to get what they want from literatures. All the intermediating systems that Synoptics puts up can be considered proposals as to what the right language should be. They may also inadvertently show that the right language doesn’t yet exist – that it remains to be created.

The ultimate goal of information science is an artificial intelligence that can furnish answers from literatures like an ideal human conversationalist. A preview of this in a science fiction movie is Vox, the holographic reference librarian in the 2002 remake of H. G. Wells’s The Time Machine. As an interviewer, Vox has flaws – he can jump to conclusions just like a real reference librarian – but he is an object lesson in what the human-literature interface might look like as Synoptics and Eclectics converge.

I think Vox’s brain will exploit the core-and-scatter distributions of terms that figure so prominently in information science. Those distributions concentrate meanings that are otherwise hidden across texts, and a Vox could marshal them to recommend some writings over others. My own work has suggested what this might mean with terms as diverse as cited authors and the titles of items held by OCLC-member libraries. In my current project, I hope to push this work, which is a form of Synoptics, closer to user studies, which are a form of Eclectics.

If these names seem like something out of early L. Ron Hubbard, you can blame my boyhood love of science fiction. I assure you that if I live to speak to Vox about information science, I won’t use them. I’ll use the same terms you would and then take only his top recommendations. I’ll filter. In fact, my final words to him will probably be “Just browsing.”

Thank you.

Copyright © 2005, American Society for Information Science and Technology