:: Digital Libraries Columns


Library Journal "Digital Libraries" Columns 1997-2007, Roy Tennant

Please note: these columns are an archive of my Library Journal column from 1997-2007. They have not been altered in content, so please keep in mind some of this content will be out of date. :: Digital Libraries Columns

Fixing Library Discovery


   Regular readers of this column know that I've often taken library
   catalogs to task for being the "wrong solution" for most user needs
   ("[145]Library Catalogs: The Wrong Solution," LJ 2/15/03). I also
   haven't taken kindly to making small changes to what I believe to be
   systemically broken ("[146]Lipstick on a Pig," LJ 4/15/05). We are now
   finally seeing some major new interface initiatives by libraries and
   vendors to address this problem.

   NCSU blows the doors off

   To recent major notice, North Carolina State University (NCSU),
   Raleigh, announced a new catalog that departs in significant ways from
   the typical catalog. The NCSU catalog is just a front end--not an
   acquisitions system, circulation system, or indeed anything other than
   a finding tool.

   To avoid the pitfalls of the typical integrated library system, as well
   as to experiment with enterprise-grade software for clustering and
   faceted browsing, NCSU extracted all one million-plus records from the
   ILS and gave them to the Endeca software company. Endeca, whose clients
   include major online retailers, processed the information, and NCSU
   staff created a faceted browsing depiction of holdings so powerful you
   can browse the entire collection using Library of Congress (LC)
   Classification Numbers.

   You can click on "D - World History," for example, and see how many
   books NCSU owns in each D subclassification, which is easily viewed as
   53 items classified in "DX101-DX301 History of Gypsies" (my 13-year-old
   daughters would insist that should be "History of the Roma," but I will
   let them take that up with the Library of Congress). Clicking on that
   heading displays those 53 results. In only three clicks from the main
   page, I have honed in on 53 items of interest out of a collection of
   well over a million.

   All the various facets of those 53 records are displayed in a left
   sidebar, aggregated and exposed so I can simply click on a selection
   such as "Dictionaries" to see a historical dictionary of the Gypsies.
   It's quite powerful to expose the richness of the underlying metadata
   in an easy-to-understand interface.

   NCSU has set the bar high, and it will be interesting to see how many
   other institutions and/or vendors attempt such a radical makeover of
   our finding tools.

   Vendors get with the program

   Ex Libris is charting a dramatic new course in library finding tools.
   Set to unveil Primo at the American Library Association annual
   conference in June in New Orleans, Ex Libris is trying to create a
   next-generation finding tool. It isn't a library catalog, but it is
   touted as an "enterprise-level solution for the discovery of
   institutional content."

   You may think that vendors are creating these systems to sell yet
   another application. But it does solve an existing
   problem--library-based information is simply too difficult to find.

   Primo and similar systems are really the tip of an iceberg that may
   unify searching of your integrated library system, your metasearch
   application, indexes created from metadata harvested from remote
   repositories, and other methods that modern libraries use either to
   aggregate content or provide access to it. At first this sounds like a
   standard metasearch application, but products like Primo seek to offer
   added-value services such as preprocessing of local metadata to provide
   clustering, faceted browsing, and other user interface enhancements.
   (See "[147]Ex Libris To Debut Project Primo," LJ 3/1/06)

   Interesting alternatives

   Emerging products such as Primo are also trying to present an
   alternative, since the library catalog simply cannot be the main
   finding tool we introduce to our users. Libraries allow access to a
   wide array of resources--many of them not within their walls. We need
   tools that can logically and usefully encompass that wide world and
   present search results from this diverse universe in a way that users
   can comprehend.

   Whether you are an early adopter such as NCSU of new opportunities, or
   are content to wait for library vendors to provide the next generation
   of finding tools, it's clear that how library users will find
   information at a library is in a period of rapid change. This is a good
   thing, since library finding tools are mostly broken, particularly when
   compared to finding tools offered by companies such as Google and
   Amazon. We must fix them--and soon.

   For more on the wired library, see the [148]netConnect supplement
   mailed with the January, April 15, July, and October 15 issues of LJ

Link List

     * Endeca
     * NCSU Catalog
     * Primo