> -----Original Message-----
> From: Next generation catalogs for libraries
> [mailto:NGC4LIB_at_LISTSERV.ND.EDU] On Behalf Of James Weinheimer
> Sent: August 12, 2011 2:23 AM
> To: NGC4LIB_at_LISTSERV.ND.EDU
> Subject: Re: [NGC4LIB] Cataloging Matters Podcast #12
>
..
>
> How can we create something that people want? One way would seem to be
> by helping to clear away the irrelevant materials to a person's search
> for them, or in other words, by creating a bibliography.
We already are enriching the catalog with many services that expand on the traditional catalog and that do a lot of the groupings and suggestions that are comparable to bibliographies on the fly. We subscribe to Bowker's Syndetics (http://www.bowker.com/index.php/data-services/catalog-enrichment ) and NoveList Plus (http://www.ebscohost.com/novelist/default.php?id=3 ), which have widgets that integrate with the OPAC and offer services like "More Like This ..." and "Read-alikes" that can be either hand-crafted lists and analyses, or machine-generated matching search results. We don't yet use some of the social networking add-ons (review and rating services) and FRBRizing tools like Library for Libraries (also from Bowker).
We tend to get a lot of UK and Canadian imprints (so, yes, publisher does matter!) and that often means the same work under a different title. Those looking for Robert Ferguson's The Vikings : A History may be surprised to find we have it, but under the UK title "The hammer and the cross". A card catalog might have uniform titles and authority-controlled reference cards to guide the user from one title to the other. In an OPAC, these handy guides are often smothered in the indexing structure that doesn't replicate the solutions used in traditional catalogs (and keyword searching begins to break down when data is stuck in authority records not bibliographic records).
The catalog enrichment services can therefore be behind the times as well, as one record will have links to enriched services, but the same work in another record will not. A bibliography would have to capture the FRBR user tasks of finding and identifying identical works in order to be as useful and accurate as it can be. (Which also speaks to the enormous deduping efforts required in managing catalogs nowadays -- I spend a lot of time working with the deduping algorithms in our new system. It's these kinds of algorithms that would be helpful in further FRBRizing of our data).
We should never say that we need more of one kind of service (bibliographies, Google searching style everywhere) and less of another (explicit manifestation and work elements). We need both. Otherwise the catalog would die a death of a thousand cuts with this shortsightedness. And moreover, a lot of effort already goes into putting the FRBR data into catalogs-- only it goes into the wrong places (!), such as optional uniform titles and free-text notes instead of controlled vocabulary relationship designators.
Thomas Brenndorfer
Guelph Public Library
Received on Fri Aug 12 2011 - 10:16:58 EDT