Monday, April 10, 2006

The Future of the Catalog

Well worth reading and considering, The Changing Nature of the Catalog and Its Integration with Other Discovery Tools. Final Report. March 17, 2006. Prepared for the Library of Congress by Karen Calhoun. A Critical Review by Thomas Mann.
According to the Calhoun report, library operations that are not digital, that do not result in resources that are remotely accessible, that involve professional human judgement or expertise, or that require conceptual categorization and standardization rather than relevance ranking of keywords, do not fit into its proposed "leadership" strategy. This strategy itself, however, is based on an inappropriate business model -- and a misrepresentation of that business model to begin with. The Calhoun report draws unjustified conclusions about the digital age, inflates wishful thinking, fails to make critical distinctions, and disregards (as well as mischaracterizes) an alternative "niche" strategy for research libraries, to promote scholarship (rather than increase "market position"). Its recommendations to eliminate Library of Congress Subject Headings, and to use "fast turnaround" time as the "gold standard" in cataloging, are particularly unjustified, and would have serious negative consequences for the capacity of research libraries to promote scholarly research.

7 comments:

Anonymous said...

Wow... very interesting, Mr. B. On top of the BSTF report, an intriguing front-end tackle. When I saw the review (rebuttal?) was from a union, I rolled my eyes.

K.R. said...

I really liked Thomas Mann's review. It is very well-written and he obviously has enough experience on the front lines to back up his opinions.

Anonymous said...

I haven't read the LC report yet, but I have read the UC report and I think these "death of the catalog" worries are greatly overstated and aren't necessarily leading us in the right direction.

Here's one example. The bibliographic information for our cylinder website (cylinders.library.ucsb.edu) lives in our ILS and the website queries it through Z39.50. This allows the contents of the collection to appear in a virtual catalog (and be indexed in Google). The website will get its 5,000,000th pageview this afternoon (after four months online) and has had enough traffic that our ALEPH system has ocasssionally been overwhelmed with traffic. The catalog and cataloging aren't irrelevant, but we do need to rethink how we get bibliographic information about the contents of our collections into people's hands.

I think this is a great example of how important it is to maintain catalogs and how they can still be relevant. No, nobody is finding these cylinders directly through the catalog but the catalog is providing the backbone to the whole system. And since it is already integrated into the library structure, it is easier to maintain.

David Seubert - UCSB

Anonymous said...

This is a link to "Draft 2B" of the report -- where is the final report, and does it differ significantly from this draft version?

Thanks.

Anonymous said...

Note that in industry, among people who don't know or care what libraries do, the idea that controlled vocabulary is important is really on the rise. Taxonomies and all that. Even the (mis-named) "folksonomies", while not actually a _controlled_ vocabularly, are a topic of excitement precisely becuase simple keyword searching of source text is not sufficient.

So now all the internet people are realizing what the library people have always known, but the library people, _still_ 10 years behind, are thinking like the internet people thought 10 years ago, still behind the curve.

Now, LCSH itself---it's got a whole lot of problems. It is not well suited to the digital environment. It does not allow systems to support certain kinds of browsing and retrieval that they ought to be able to support. The cost-benefit analysis of _LCSH_ as a _specific system_ may indeed be a losing bet. But the only two choices aren't LCSH or nothing---the question is, what would a controlled vocabulary that _was_ optimized for online environments, that _did_ have a cost-benefit outcome that worked--what would that look like? Let's get rid of LCSH for something _better_, not for nothing.

Incidentally, I feel the same way about cataloging in general (descriptive and access; not just subject classification)---the way it's currently done now is pretty damn broken for the current environment. But that doesn't mean the better alternative is a completely uncontrolled environment. The solution is _fixing_ the controlled environment, coming up with the right controlled environment. I am continually frustrated with these debates that put "What we do now is just fine!" on one side and "the solution is uncontrolled keyword searching" on the ohter, with people vociferously arguing both sides. Both sides are wrong. I see the Calhoun report as sadly representative of this problem.

--J

Anonymous said...

Final report right where it should be.

Anonymous said...

I'd just like to second J's comment that there is a better choice than either continuing what we do now or doing nothing. I found both the Calhoun and the Mann reports to be rather polarizing. It seems to me that the problem is not so much with cataloging theory or what we are trying to do, but with the *how* which still seems to be optimized for the card catalog environment. A faceted LCSH (perhaps with more structure than FAST such as the ability to cluster facets or use some sort of grammar to relate them) with better hierarchical navigation and better integration of the cross-reference structure (both hierarchy and x-refs could be expanded and improved in LCSH, but this is also a huge shortcoming of OPAC user interfaces) would be a much more powerful solution in a digital environment.