The FRBR Blog

Last December, the Bibliographic Services Task Force of the University of California Libraries (a group under their Systemwide Operations and Planning Group (SOPAG!)) published a report titled Rethinking How We Provide Bibliographic Services for the University of California.

• Executive summary (112 KB PDF)
• Rethinking How We Provide Bibliographic Services for the University of California (396 KB PDF)

Other library bloggers have discussed this and the new NSCU catalogue together, for example Lorcan Dempsey in his post Thinking About the Catalog and Karen G. Schneider in The Revolution Will Be Folksonomied (where she says: “FRBR in a nutshell: a user shouldn’t be confused by multiple records for the same item”).

Here are the recommendations in report’s “Enhancing Search and Retrieval” category. I only quote the FRBR-related one in full, but the others are all explained in the full document.

I.1 Provide users with direct access to item

I.2 Provide recommender features

I.3 Support customization/personalization

I.4 Offer alternative actions for failed or suspect searches

I.5 Offer better navigation of sets of large results

I.5a: Implement FRBR concepts to present related works hierarchically, pulling together all records related to a particular work (e.g., Moby Dick), diverse expressions of that work (e.g., translations into German, Japanese and other languages), different versions of the same basic text (e.g., the Modern Library Classics vs. Penguin editions), and particular items (a copy of Moby Dick on the shelf).

I.5b: Follow all of the linking fields in serial records to present all of the variant titles to users in a “family tree.”

I.5c: Implement faceted browsing based on sophisticated analysis of the contents of the records.

I.6 Deliver bibliographic services where the users are

I.7 Provide relevance ranking and leverage full-text

I.8 Provide better searching for non-Roman materials

That’s just part one of the list of recommendations. Go read the rest. They’re very good.

FRBR comes up throughout the report as one of the things that will make users’ lives easier, including in a use case and in this recommendation:

III.3 Manually enrich metadata in important areas

Manual metadata creation is by definition both expensive and time-consuming and is an activity that should judiciously be applied where it yields the most benefit. There are a number of areas where the application of intellectual effort in the creation of metadata justifies the high cost.

The enhancement of FRBR relationships through the manual addition or correction of name, main title, series titles, and uniform title, especially for prolific authors in the fields of music, literature, and special collections is one such area. The collocation of materials and the concomitant search and retrieval improvements in these fields more than justify the cost.

Additional attention to serials holdings would likewise have a major positive impact of effective search and retrieval. If serials holding were better structured, services to users would be much more reliable and major efficiencies could be reached through automated record matching and processing.

III.3a: RECOMMENDATION: Enhance name, main title, series titles, and uniform titles for prolific authors in music, literature, and special collections.

III.3b: RECOMMENDATION: Implement structured serials holdings format.

That’s a good line: “The collocation of materials and the concomitant search and retrieval improvements in these fields more than justify the cost.”

This report and the NCSU catalogue show that FRBR has become a basic part of new thinking about catalogues — overtly in the Group 1 (work, expression, manifestation, item) hierarchy that everyone wants to see, and more subtly in the way that the user tasks (find, identify, select, obtain) are helped by the other tools for searching and browsing that the catalogues have. I think this is great, and it’s all the more impressive given that it’s unknown what Resource Description and Access will end up looking like in the end, how authority records will work, if MARC and other standards will change, and so on. People are demanding FRBR, and making it work, before it’s fully settled. (It’ll be interesting to see how new implementations handle serials, collections and anthologies, and other forms of aggregates.) The number of RFPs mentioning FRBR will be increasing, I’m sure.