Web Exclusive

Coverage of ASIS 1997 Annual Meeting

Then and Now: Information Science Problems and Systems


ASIS Annual Meeting Technical Session sponsored by SIGs/HFIS, ED, November 3, 1997

Speakers:
Ben-Ami Lipetz, State University of New York at Albany: What Were Those Big Old Extract Files, and Why Should Anyone Care Today?
Trudi Bellardo Hahn, University of Maryland Libraries: Online Information Retrieval: How Far Have We Come?
Donald Hillman, Lehigh University: The Leadermart System: Then and Now
Moderator:
Robert V. Williams, University of South Carolina

Session Abstract: This session continues the annual sessions looking at some classical problems of information science that will be of interest to a general audience and those with an historical interest in the field. (From the Final Program)

Online Information Retrieval: How Far Have We Come? Trudi Bellardo Hahn, University of Maryland Libraries

Author Abstract: The first online information retrieval systems appeared over 30 years ago and are continuing to evolve. However, their development has not been a long line of continuous improvement in functionality and usability. In our forthcoming book on the early development of online systems, Charles Bourne and I researched the milestones of invention and implementation of the basic capabilities of online retrieval systems to show when each capability was introduced and by whom. Some examples are:

Which of these capabilities are commonly found in Web search engines? Boolean and proximity operations are common but not universal. Sometimes the relationships are automatic, but it is hard for a user to determine which logical operations are being performed. In some search engines, users can create queries with natural language, but retrieval is sloppy. Truncation and automatic incorporation of synonyms are available. However, they are often automatic, which may result in unwanted retrieval. Date ranging, term weighting, and field limiting are available only in a limited way. Case sensitivity may be available, but it is hard to determine. Ranking and relevance feedback are commonly available, but based on a variety of criteria unknown to user. Citation searching is not found. Output options, iterative search, canned query, vocabulary browse, and displays of concept hierarchies or thesauri are seldom available. On the other hand, the World Wide Web shines in document delivery and instant world-wide access.

Many basic capabilities invented 30 years ago are not found regularly in Web search engines or they are done automatically without a user's knowledge or control. The Web searcher has given up a lot of control in exchange for simplicity of use, which causes frustration and wasted effort. Current designers should look back a few decades for inspiration about how to make their systems not only easy to use, but more under the real control of the searcher.