Last Wednesday I flew up to Boston for ALA Midwinter and a meeting of the NASIG Executive Board, which I attended as Member-at-Large. Although NASIG is an independent organization, we piggyback our board meeting on the front-end of the conference. So Thursday, I attended our all-day meeting to discuss the on-going business of NASIG in a conference room at MIT’s main library, followed by dinner with my fellow board members, and a late-evening drink with Jim Galbraith, who was in town for the conference as an OCLC rep.
On Friday, the highlight of my conference day was the ALCTS FRBR Interest Group. Jennifer Bowen of the University of Rochester discussed their library’s project to build the eXtensible Catalog, a faceted, FRBR-compliant, customizable, open-source catalog. The catalog is still in development, but they have completed a metadata services toolkit, which allows libraries to automate the processing of batches of metadata, and the creation of FRBR-compliant records.
As you’ve probably heard before, FRBR is a conceptual framework for describing bibliographic information at various levels (work, expression, manifestation, item). The clearest way to explain what FRBR is may be to explain the process that Bowen and her team are using on catalog records. The bibliographic records that we are used to seeing are essentially manifestation level records. For example a bib record for “Tom Sawyer” will describe one particular edition of the book, printed in a given year, with a certain number of pages. This is a manifestation of the book, to use FRBR terminology. Above the manifestation level is the expression level, which is more abstract and would be “Tom Sawyer” in English. All editions of “Tom Sawyer” in English would be related to this one expression of the book. Above the expression level is the work level, which is an even more abstract concept. This would be the novel “Tom Sawyer” in any language (Spanish, French, German, etc.), in any edition, published in any year. Bowen and her team are taking catalog records and other metadata and running batches of this data through their toolkit to turn the records into MARC-XML. These MARC-XML records are then run through the tool kit and parsed into records at the Work, Expression, and Manifestation levels. Each Manifestation level has an 004 field with a number that links it to an Expression level record above it. And each Expression level record has an 004 field that link it to a Work level record. The tool kit includes an aggregation service that aggregates records that represent the same manifestation and prevents duplication of records. Because of their fairly thorough development of FRBR-compliance, Bowen would like to see the University of Rochester become a test bed for the implementation of RDA, the new cataloging and access code, which encourages the use of FRBR principles.
Several questions occurred to me after the session. Namely, will all libraries have to go through this process of creating Expression and Work level records? Would Expression and Work level records be added to OCLC? If they are, who would create these records, or would they be generated automatically? Would Expression and Work level records function as a new type of authority record? These are issues to consider as RDA moves forward.
On Saturday, I attended the Technical Services Manager in Academic Libraries Interest Group. The meeting actually was based around table discussions, and I joined the table discussing the topic “Re-tooling technical services staff to meet evolving needs.” The conversation was interesting and lively, although I have to confess I didn’t pick up many new suggestions. I did share some of the stuff we’ve been doing in Resource Services, and it seemed pretty well received. It made me feel like Lauren’s got us ahead of the curve.
Later that morning I attended the CCS Copy Cataloging Interest Group Meeting, where there was discussion of the recent Library of Congress report on the MARC record marketplace. This report, if you haven’t read it, argues that there are plenty of catalogers available, but not enough catalog records are being created, that libraries are relying on vendor-produced records, which are not freely available and sharable, and that there are not sufficient incentives in the current system to encourage original cataloging. The discussion prompted a sharp retort from an audience-member that the same consulting company that wrote this report for LC had advised their institution some years ago to use vendor records, wait for other libraries to catalog titles, and to generally do less original cataloging, in order save money. I found this to be an interesting illustration of one of the risks of using consultants, namely that they will tell a client what is in the client’s best interest, rather than looking at the field as a whole. What is best for the individual library is not necessarily what’s best for the library field as a whole and vice versa.
At this session, I met up with Jennifer Roper and we had lunch together and caught up. Her daughter Maddie is nearly three years old already. Time flies!
That afternoon, Jennifer and I attended a session that shall remain nameless, and saw a presentation by a cataloger and systems librarian from a university that shall remain nameless. They described the remarkably convoluted and, frankly, wrong-headed process that they use to batch-load records for electronic monographs (they use a single record for both the print and electronic versions of monographs, which is a technical nightmare). It made me enormously grateful that we use the process we developed here.
That evening, I had dinner with one of my old college roommates and his wife, who live up in Boston, and made an early night of it. I had to get up at 3 am to make my 5:45 am flight early Sunday morning. Although it was fairly rough, I was able to get home before noon, which was nice.