Cataloging Transformed: From Traditional to Emerging Models of Use

Cataloging Transformed: From Traditional to Emerging Models of Use

RUSA_RSS “Catalog Use Committee”Co-sponsored by RUSA-MARS “User Access to Services Committee”

Sunday June 25, 2006 4:00-5:30PM

Morial Convention Center Rooms 295-296

I was impressed by the size of the room and the attendance. There was an air of expectation and excitement bubbling quietly though the room as people filed in from their previous sessions, or perhaps the exhibit floor. Certainly many people had made the trek from the Questioning Authorities session further along the river in the Hilton Riverside.

This was a very good program and full of rich information. As such, I’m breaking it up by speaker. So you know what to look forward to the speakers, in order, were:

  • Andrew K. Pace Head of Systems, North Carolina Ste University, Columnist, “Technically Speaking,” American Libraries.
  • Cindy Levine Reference Librarian for the Humanities, North Carolina State University
  • John Blyberg Network Administrator and Lead Developer, Ann Arbor District Library
  • Jina Wakimoto Head of Cataloging, University of Colorado Boulder.

All materials will be made available on RUSA-RSS page http://www.ala.org/RSSmainTemplate.cfm?Section=RSS. They will be setting up a blog. As of today’s writing, I do not see it available.

Andrew Pace http://www.lib.ncsu.edu/staff/apace/ was the first speaker, and seemed eager to start. (You can see many of the slides included in his presentation, “Dis-Integrated Library Systems and the Future of Searching,” Amigos Annual Meeting, 2006, Dallas, TX.

He outlined his slides describing the classic and disintegrated library systems. Andrew stated that the classic system does a decent job of inventory control and known item searching. He was quick to point out that “Google is good at known item searching; it sucks like the OPAC in this respect.”

Classic ILS systems do a poor job of searching anything other than known items or anything other than books. They lack logical groupings of results (FRBR), faceted browsing (if you like this…), robust relevance ranking, and what he terms sideways searching (expansion of searches and search topics).

The dis-integrated library system is envisioned to include ILL, ERMS, Collection Management, Websites, e-books, e-journals and databases. Andrew was quick to point out that it isn’t so much the electronicness of an item that makes it a challenge as its seriality. These items just aren’t widely or easily available in the current ILS systems.

These problems are well represented by the library puzzle. The four pieces include Serials, the Catalog, the Web, Abstracts/Indexes/Databases. What is needed is a way to bring all these pieces together to form a coherent whole.

In working to put together the library puzzle, one of his favorite metaphors, NCSU created their e-matrix (a proposed marketecture/ system architecture) http://www.lib.ncsu.edu/ematrix/ as a way to leverage the data as defined by DLF-ERMI. http://www.diglib.org/standards/dlf-erm02.htm

Andrew then asked, “What hath disintegration wrought?” Pointing at a slide with myriad of vendor trademarks and logos he pointed out the vast choices available in today’s library marketplace. Then he pointed out that our confusion in this world is minimal compared to our users day to day dealings with vendors and other services. We have library specific sites to deal with; they have everything that the great wide digital world can come up with.

He then quoted Roy Tennent (California Digital Library), “Most integrated library systems, as they are currently configured should be removed from public view.” A light bulb went off in my mind when we were all reminded that MARC was created so we could move data around from place to place with creating catalog cards the end goal. (There seemed to be general crowd consensus on this point.)

This general lay-of-the-land regarding ILSs is what lead NCSU to its decision to pursue Endeca as the way to present their library holdings. (I hesitate to use the world OPAC, it looks so very different). NCSU went out of the normal library space and into the rest of the digital world and chose the solution that is used by Walmart, John Deere & Circuit City, among many others.

He brought out an interesting conundrum at this point, how can you justify a slow library catalog search to a student when they are used to google running through millions of records in seconds.

Endeca brought several longed for features to the table:

  • Speed
  • Relevance Ranking
  • Faceted Searching
    You type in laser printer, it asks, do you want to search by manufacturer HP, Lexmark or do you want to search by price? under $100, over $100.
  • Suggested Searches

To see the NCSU catalog please go here: http://www.lib.ncsu.edu/searchcollection/

More information is available here: http://www.lib.ncsu.edu/endeca/

All of the search and suggestions to the left and at the top are selected by the library. Andrew mentioned 10 dimensions by which search results were refined. LC Classification, Subject Headings (split up into component parts), Topic, Genre, Time Period, and Geographic Location. One of the things that impressed me the most is that they’ve been able to mine the fixed fields to their advantage.

This is just one of many “Next Generation” library catalogs. Others included on the slide were:

  • Redlight Green
  • OCLC FictionFinder
  • Vivisimo clustered search
  • Aquabrowser visual context
  • Endeca Guided Nagication/Information Access Platform
  • Innovative Interfaces OPAC Pro and Encore
  • ExLibris Primo
  • Polaris AJAC-Enabled OPAC
  • SirsiDynix EPS, FAST and something “top secret”
  • OCLC Custom WorldCat (This generated an amusing comment, “The OPAC is dead, buy our new OPAC.”)
  • Georgia Pines Oubesm Koha and the Library 2.0 Bandwagon

NGC4LIB was promoted as a lively listserv to discuss the future of the catalog. Having joined the listserv as soon as I saw it announced, I agree.

Andrew then brought us back to the library puzzled and stated that Endeca has been misrepresented as a short term solution. He posited that it is part of a long term solution to get the pieces of the puzzle back together. At NCSU they’re talking about exposing Endeca to other technologies like Ex-Libris and Vivisimo to get them into the results.

His future plans for Endeca include:

  • Continued usability testing and more bling! (Table of Contents, Cover Art, Reviews…)
  • Relevance ranking algorithims and a spell correction threshold
  • Additional browsing options.
  • Endeca 2.0 FRBRized display
  • Natural language entry options
  • Patron generated refinements of folksonomies. (This is in a maybe, maybe not state)
  • Death of authority searching (not authority work.)
  • More integration with quicksearch and other data repositores.

To top it all off he also gloated (slightly) that re-indexing Endeca only takes 3 hours. Amazement and envy washed over the crowd and everyone mentally ticked off how long their last re-indexing project took.

On that note he handed over the podium to his colleague Cindy Levine. Check back for the post detailing her presentation later.