GURU INTERVIEW: SUE FELDMAN
After listening to Sue's presentation, "The Answer Machine," at Nylink, I wanted to find out more and share her answers with you.
Marylaine: How do you stay current with the technology and information resources? Specifically, what do you read, what sites do you routinely visit, what listserves or discussion groups do you participate in?
Sue: I try to read all or some of the following: JASIS, Searcher, Online
Magazine, Communications of the ACM, Information Advisor, Cyberskeptic,
the New York Times, Discover Magazine. I also read textbooks on
information retrieval and natural language processing. And mystery and
historical novels for relief.
Alerting services from: DR-LINK, Ingenius Technologies NetBrief and
EgoSurf, DIALOG, various grants and proposals services such as NASA, IRIS,
NIH GUIDE. I also get press releases from the web search engines and the
major onlilne vendors.
Listservs: DIGLIB, SIGMETRICS, AIIP, EDUPAGE, TIDBITS, search engine
Favorite sites: D-LIB Magazine, www.searchenginewatch.com, www.ingetech.com,
botspot.com/. I search frequently for various subjects of current interest,
and keep up on new developments in the search engines that way.
One of the invaluable sources I have is friends and colleagues who alert me
to various articles I might not see otherwise. I try to do the same for
Marylaine: Of all the new developments in data mining and information visualization and natural language processing, which do you think has the greatest capacity to completely change what librarians do and how we do it? In other words, what should we learn next?
Sue: Librarians have so many choices these days. I do think that the regular jobs such as cataloging or reference won't suddenly disappear. If
anything, being able to make sense of the piles of information available to
everyone is even more valuable. But if librarians want to move up to a
higher level of abstraction, they will need to develop analytical skills
and learn to market them. Most of us analyze materials for content and
value already. The trick is to make our clients understand that we have a
talent for analysis and organization..
You asked about which new developments I feel are the most vital to learn
next. There are two big developments just over the horizon. One is tools
that will help people find patterns in information and data. Some of these
are analytical tools like data mining or rule-based systems for decision
support. Others are visual tools which sit on top of the analytical tools
to allow a user to understand patterns at a glance. I think a basic, high
level understanding of how search technologies work, and how Natural Language Processing works is
crucial to understanding the technologies built on top of them. For
instance, you can understand most of the new visualization tools if you
know what the vector space model is. That sounds fancier than it is, by
The second big development is intelligent agents. Agents will change
information systems from static to dynamic systems. They will adapt and
change as your interests or the field changes. They will work in the
background and improve at each iteration so that more and more materials
that would not be of interest to you are filtered out.
Both of these areas are aimed at making sense of too much information, and
that is the major problem facing knowledge workers these days.
Marylaine: One of the things you talked about at Nylink is the visual display of information. Do you think librarians should be doing more with visual display to enhance our information services?
Sue: Absolutely. In fact, this is an area we have been deficient in for as long as I've been a librarian (31 years). How many displays have you seen in libraries that have taken your breath away? More to the point, how many libraries were easy to navigate 30 years ago? The problem is no different today. You need to be able to go into a real library or a cyber library
and know what's there at a glance. Where are the novels? How do I get a
journal article? How do I get back to the entrance? We need sign posts
and cyber-sign posts, and they must be visual because there is too much
text and it takes too long to read and digest it.
Marylaine: In a world where patrons want and expect full-text when they sit down at a computer, what do you think will happen with traditional databases which have only citations and abstracts?
Sue: I wish I knew the answer to that one. You aren't the first to ask it. In fact, many of the abstract and indexing services are asking themselves that exact question. What the A & I services can add to full text is better precision in searching. They also create a product which gives a summary, and summaries save people time. Most of the researchers I know would like a good summary as well as the option to see the full text. The problem is that humanly created indexing and abstracts add months to the publishing cycle. Researchers aren't willing to wait that long. Machine aided indexing and automatic categorization is one of the hottest development
areas in information science these days. Possibly this can speed up the
process. Good automatic summarization is much further in the future.
I do believe that traditiional abstracting and indexing services can only
survive if they can at least point to full text on the publisher's site.
This must be done so easily that it just pops up with a single click,
though. Technically it's feasible. Will publishers be able to develop a
seamless method for collecting payments and distributing royalties? More
to the point, will they see that if they don't, newer electronic
publishing ventures like e-Biomed will make them extraneous? I don't know
the answer to that one.
Marylaine: Do you think publishers will continue to offer small, highly targeted databases, or do you think the future belongs to large aggregated databases?
Sue: In the near future, I expect--and dread--continued fragmentation. Publishers see a new market, and are intent on making sure that they have a recognizable "brand name" on everything they own. They want to grab the electronic distribution venues.
In the long run, I would hope that users will become frustrated enough so
that they will make aggregation into larger information services a
possibility. This can only happen if the pricing is clear, predictable and
affordable. And if the payment mechanism is not onerous. Probably
everyone is watching Northern Light http://www.northernlight.com/) right now to see if their model works.
Certainly, it satisfies all my criteria except for the fact that I would
like to be able to find even more materials in one place.
One final point about the future and librarians: we need to know what we
do well. It really isn't finding information. What we do well is to ask
good questions, and be able to organize and cluster information
meaningfully so that we can draw useful and unexpected conclusions.
Other guru interviews:
- Tara Calishain
- Jenny Levine, part I
- Jenny Levine, Part II
- Reva Basch
From one of my all-time favorite rock bands, The Church. It's from the song "Volumes," on the album Remote Luxury. It's a nice reminder that it's not just information, internet and databases.
Volumes have secrets. Take them on holidays.
Book them a room, save them a moment.