For the past several years, I've been mostly an in-name-only member of the Library & Information Technology Association's "Top Technology Trend Experts." Since I was actually at ALA this year, however, I participated in the panel discussion. The first time I did this, 4-5 years ago, it was a lot like a committee meeting, and there were more experts in the meeting room than audience members. This year, however, we filled our 300-person meeting room in the convention center, then opened the panels to the room next door and filled it, as well. Guess people are hungry for information on new technologies!
As predicted, there was no wifi to be found, so I didn't bother bringing my computer. Instead, I took notes the old-fashioned way, on tiny pads of paper stolen from hotel meeting rooms.
The way the program works is each of the "trendspotters" participating gets 5-6 minutes to talk about what they see as the important or interesting technology trends in the library world, and then we answer questions, argue amongst ourselves, etc.
First to speak was Karen Coyle, from the California Digital Library project at UC. Not unexpectedly, she had one of the most radical suggestions for the audience...calling not just for the end of MARC cataloging (something others have been saying in library land lately), but for "the death of alphabetical order." Given the power of current searching tools, she said, we don't need to organize materials using predetermined or universal "known order." Interesting food for thought. Not sure I agree, but it's a great conversation starter!
Next up was public library consultant Joan Frye Williams. She talked about the way some public sites (I think she used a museum as an example) are "annotating" their physical spaces with information online, allowing people to combine physical and virtual "hunting and gathering" of information. That's made easier by the number of people now "packing heat" in the form of portable electronic devices with wireless connectivity. The underlying trend she identified was the move towards information providers interfacing directly with user equipment--users no longer draw a line between the device and the content.
After Joan came Cliff Lynch, director of the Coalition for Networked Information. Cliff remarked on the impact of "last-mile" wireless changing network dynamics--the last mile used to be the most difficult, slowest piece, but WiFi changes that. He noted the change in Federal funding of science and technology infrastructure, which now includes a "third leg" of data curation and management--an obvious area for libraries to play a key role in. He ended by questioning the value of descriptive metadata (a bit of a sacred cow for libraries, for obvious reasons) because of the growing promise of computational linguistics for discovering information resources.
Cliff was followed by Walt Crawford, from RLG. Walt's a member of a decade-old informal group I'm in that meets for dinner at each ALA conference. When he started talking, I started thinking about this year's dinner, which got me distracted, and I missed his main points. :( Sorry, Walt! (I know you're reading this, so I feel doubly bad about spacing out here. Feel free to supplement in the comments!) I did hear him toss off one line that stuck in my brain, however...it was when he awarded the "Deader than DIVX" award to eBook technology.
Next up was Tom Dowling from OhioLink. He talked about the power of Bayesian filtering for catching spam in email, and speculated about the use of such filters ("trainable" tools) for determining usefulness of search results. (Cliff noted later that this is in fact being looked at in some places...he didn't say where.) He also talked about the growing need for libraries (and publishers who supply libraries) to do a better job on identifying, authenticating, and authorizing users of online resources. He suggested Shibboleth as a technology that libraries should explore.
I was next, and my two words to the audience were "decentralization" and "weblogs." (Obviously, SuperNova 2003 is much on my mind these days!) Among other things, I talked about how the combination of decentralized (wifi-based) internet access and personal publishing through blogs can change the entire conference experience for attendees. After my presentation, I was approached about possibly doing a blog-related presentation at the Internet Librarian conference this fall in Monterey. Am hoping that will work out!
After me was my old friend--and the new LITA president--Tom Wilson from the University of Maryland. He followed up on Joan's discussion of information provision to user devices by talking about the need to be aware of the increasingly smaller "window" onto content that these small devices offer and ensure that information services work in that environment. He then talked about the growth of "web services," and encouraged librarians to think about what they're actually "buying" when they get information in this form. Are they paying for the data? Or only for temporary access to that data? (This is an issue with electronic access generally. If you subscribe to a print journal, even if your subscription lapses you still retain the issues you purchased. With electronic resources, how can comparable persistent access be preserved?
The last speaker was Marshall Breeding, from Vanderbilt University. He talked about his perception that there's nothing "new and exciting" happening in library automation these days. He sees this as a result of the trend towards more and more integration of functions into large-scale automated systems, and the library RFP process forcing vendors to replicate each other's features rather than innovating. He also mentioned technologies like OpenURL and meta-searching as important things to watch in the library world.
Marshall's comments led to a discussion among panel members about the possible need for integrated library systems to become less integrated and more modular, allowing for "plug and play" components from different vendors. This "dis-integration" of library systems is an interesting concept--it will be interesting to see whether it can happen. The relatively small size of the ILS market makes it fairly cut-throat; there's not a lot of incentive for vendors to work in a way that allows their customers to use their competitors' systems, even in part.
So, that's my not-so-brief summary of the panel. Later tonight, or tomorrow, I'll post impressions and information from the scholarly publishing presentation I attended yesterday; I know that's one that several people are waiting for.
Search may work for the users, but is it going to work equally well for developers (both techie and info-sci-y) and maintainers?
I wouldn't toss the alphabet juuuust yet, though I'm as excited about faceted metadata and the like as anyone.
When you get a chance, ask Walt Crawford if there's a trophy or a ribbon or something associated with that award. I'll display it proudly. :)
Thanks again for sharing your detailed notes.
Several notes:
Dorothea--no trophy, and the honor really goes specifically to dedicated ebook appliances (REB and whatever), not to ebook technology in general.
Dr. Lawley--(yeah, I know, but...):
Great summary. I tossed off a few items, but none of them were very detailed. An acquaintance already asked whether I was going to summarize TTT in the next Cites & Insights (due out right around this weekend); I'm not, because the July issue is almost entirely material written in late May. I might do a summary in August--but meanwhile, I'll point people to this posting as a first-rate summary [better than my own notes would support].
Had there been enough time, I would have pushed Karen C. on the "death of the alphabet" point. Given real-world infelicities in cataloging and data entry, browsable headings lists in alphabetic order are a real service to users. I can demonstrate from Eureka logs that, overall, users encounter a browse list at least once in roughly 60% of sessions--and in roughly two-thirds of those sessions, the user's result is modified (and presumably improved) because of the browse list. As usual, it's an "and, not or" situation. Reverse chronological order is probably the most sensible ordering method for most article citation databases; when there's enough text for meaningful "relevance" ranking (which there generally isn't in catalog/index records), "relevance" is a good choice--but there are still lots of cases where alphabetic lists make sense for the user.
Maybe more later (when I retrieve the piece of paper containing my scribbled topics, written down half an hour before the session--as you might expect, my points aren't in *my* notes either). Maybe not. You captured the heart of the session. (And I see your estimate of the room's capacity--300 originally, 600 as expanded--is the same as mine. We filled both spaces, which is little short of astonishing.)