Mar 25, 2003: Wee Report on the IA Summit
Just back from the IA Summit in lovely Portland, Oregon. I think Christina Wodtke and the rest of the program committee really outdid themselves; these Summits just keep getting better and better. This year's edition offered a keynote by Stewart Brand, followed by two days of three concurrent session tracks, not to mention evenings of much boozy jollity. Spirits were definitely higher than the past two years.
There were often two and sometimes three sessions per time slot that I wanted to attend. I'm hoping to catch up on what I missed from the spate of trip reports that will likely crop up over the coming week on blogs near you. In fact, there was a blog set up for live reporting from the conference that you might want to visit.
In the meantime, I'll contribute just a wee bit of coverage here. Former Argonaut Amy Warner's session on thesaurus design really charged me up. Amy is responsible for updating the NISO standard for monolingual thesaurus design (the standard's NISO code is Z39.19-1993, if you're into that sort of thing). By the way, are you one of those folks who's totally bored by discussion of thesauri? Then stop reading, but realize that the standard is currently being downloaded about a thousand times per month (and trust me, it's not exactly light reading).
Despite its surging popularity, the standard is showing its age. The last version was released in 1993, just before the Web changed everything. The old standard was designed for those days of nasty expensive online systems, where you might pay hundreds of dollars per hour for the privilege of hunting in Dialog, Lexis-Nexis, and their ilk.
Because these systems were so expensive, it was more common to design a search query offline and then do your search as quickly as possible. Exploratory browsing, naturally, was uncommon. So were many niceties we take for granted in contemporary searching, like spell-checking and automatic handling of plurals. So the old thesaurus standard was designed to support those rapid-fire searches, typically by trained specialists who already knew exactly what they were looking for and how to express those needs as search queries.
But thesauri no longer need to be quite so exacting in how they specify term development; today's search engines already handle spell-checking, stemming, and other tricky things that were done by trained searchers in the past. And something else has changed: many contemporary designers are trying to exploit the semantic relationships between thesaurus terms to improve browsing, both in terms of top-down taxonomy design and bottom-up contextual navigation. Yet the old thesaurus standard wasn't intended to assist designers in designing the semantic relationships that make browsing work better.
Well, the good news is that the new standard will. One of Amy's chief goals is to provide more guidance in specifying semantic relationships. Term development will shed much of its past inflexibility. And other new guidelines will be added, including user testing for thesauri, thesaural content display for digital contexts like the Web, and guidance on how new technologies can improve thesaurus interoperability, construction, and maintenance.
It's a tall order, but Amy is shooting for having this work done later this year. That's ridiculously fast, but she's got some great advisors helping out, including two you may already know: Peter Morville and Vivian Bliss.
Of course, I'd love for her to take on one more Big Issue: how to design and maintain thesauri in distributed environments like, for example, large, political, messy enterprises.
email this entry
// so comments are closed on this entry...
©2008 Louis Rosenfeld LLC. All rights reserved.