The FRBR Blog

Almost two years ago I posted OpenFRBR Manifesto Number One. Then nothing happened. Now I’m going to take another stab at it.

Over July and August, mostly while on the train back and forth to the IFLA conference and then while I was on holiday, I hacked up OpenFRBR 2.0. It’s a simple partial implementation of FRBR done in Ruby on Rails. You can go in and fool around; in fact, you can add or delete anything. I’ll reset it back to the initial test data when it needs it. Or you can download it and run it at home (I think you’ll need Rails 2.1.0 exactly).

I hope to:

• Polish up some of the big missing bits (like work-to-work relationships), then get into the smaller things that need to be done. Or maybe I’ll do it the other way around.
• Learn more about Ruby and Rails as I go. There’s much to be improved in this code.
• Make it all available through Git so it’s easy for other people to get involved.
• Write here and elsewhere about how it goes.
• Work FRAD into it.
• Use Robert Maxwell’s insights in FRBR: A Guide for the Perplexed to improve the data modelling.
• Use the Library of Congress and OCLC algorithms with xISBN and thingISBN and other sources of data so that someone can enter in an ISBN and lots of related manifestations will get FRBRized and slotted into place.
• Use the Harry Potter bibliographic universe as the test domain.
• Show a working example of what a FRBRy catalogue could look like.
• Have library geek fun.

Leaving a comment is probably best if you have any questions. I’ll be at the Access 2008 conference the rest of this week and will be glad to have a chat about this.

Dan Brickley popped into #code4lib t’other day to point this out.

Here’s another Flickr picture tagged with frbr.

Norberto Manzanos announced that an English translation of his Spanish paper “El Impacto de FRBR en Argentina: Implementación de un Modelo de Objetos Basados en FRBR, CRM y FRBRoo en CAICYT-CONICET” is now online.

• El Impacto de FRBR en Argentina: Implementación de un Modelo de Objetos Basados en FRBR, CRM y FRBRoo en CAICYT-CONICET (356 KB PDF)
• The Impact of FRBR in Argentina: Implementation of a model based on FRBR-ER, CRM and FRBRoo at CAICYT-CONICET (101 KB PDF)

This article presents a computer science design for the registry of documentary material based on models FRBR-ER, CIDOC CRM and FRBR-OO that is being developed in CAICYT-CONICET. The proposal leaves from the first group of entities defined by the model FRBR-ER (Work, Expression, Manifestation and Item) and incorporates some of the proposals of the other two models. In general, it rescues of these the use of the paradigm of objects, which derives in a more rigorous definition of concepts. It takes from them, among other things, the modelization of events, which allows to represent documents in its temporary process. It to deepen in other aspects that not yet have been treated by FRBR-OO: the lack of exhaustive classification, ontológical status of the Item and its relation with the hardware, the problem of the responsibility and the function of the documentary person in charge and the problem of the names. Since the presented product is at the moment in use in CAICYT-CONICET’s data base of argentine ISSN, it is detailed which is the approach of the model to the problem of the series. Finally, some particularitities of the implementation are mentioned: use of the Smalltalk language, his dialect of opened code, Squeak, and frameworks independent that has been developed: Atón and Smallfaces.

Squeak! Cool.

Two Books About FRBR, Compared, by Christine Schwartz (who blogs at Cataloging Futures), appears in The Code4Lib Journal, Issue 4. (See the code4lib web site and Planet Code4lLib if you haven’t already.)

This article reviews 2 books on FRBR published in the past year. Although both books aim to be introductions to FRBR, their approaches are very different. One is sort of a FRBR study with commentary, the other a collection of essays. Robert Maxwell’s book, FRBR: A Guide for the Perplexed, takes the study guide approach. Arlene Taylor edited Understanding FRBR: What It Is and How It Will Affect Our Retrieval Tools, a book of essays about FRBR and FRAD, written by cataloging experts, aimed at a broader audience, not just the cataloging specialist. The first seven chapters lay out the basics: introductions to FRBR and FRAD, FRBR research, FRBR and the history of cataloging, FRBR and RDA. These chapters provide an excellent introduction for those new to FRBR. The last seven chapters each look at different types of resources in relation to FRBR.

I continue to catch up, and today point out a thinky piece Karen Coyle posted on her blog the week before last (!): Thinking About Linking.

In the FRBR vision that RDA has embraced, there is something called the “relational/object oriented model.” I have some basic problems with this because I perceive relational and object oriented designs to be quite distinct. This concept of relational/object-oriented gives me one of those “blank brain” moments — when something sounds like it should make sense but I just can’t make sense out of it.

… I can’t shake the feeling that there are at least two distinct kinds of relationships: those that fill in what otherwise would be gaps in a metadata record, and those that inform relationships between bibliographic items. I also wonder about links with and between complex entities.

Online Audiovisual Catalogers (“The Internet and AV Media Catalogers Network”) has a Cataloging Policy Committee, and the OLAC CAPC formed a working group called the Moving Image Work-Level Records Task Force. And lo, they issued a report (195 KB Word) describing the findings of some subgroups.

Due to the large nature of the task, after initial discussion, the task force split up into subgroups to work on different aspects of our charge. In addition, it quickly emerged that there was not complete consensus on the definition of a moving image work nor on where to draw the boundaries between moving image works. Therefore, an additional task, that of defining a moving image work and examining some test cases to see where boundaries might usefully be drawn, was added. After each subgroup completed its work, the task force as a whole discussed the results. This paper consists of a draft of our recommendations based on the work of the first two subgroups.

I haven’t read the whole report, but at a glance there’s a lot of interesting, thoughtful, and useful stuff in it, including lists of examples and explanations of whether this is a new work based on that other work, or what:

Foreign feature film dubbed and reedited with added sequences for the American market with actors not in the foreign version (e.g., the original Godzilla)

Expression/version based on one WPE [practical work/primary expression].

Foreign feature film with the original soundtrack removed and dubbed with dialogue which has little or nothing to do with the original dialog and changes the plot (e.g., What’s up Tiger Lily?, in which Woody Allen uses dubbed dialogue to spoof a Japanese action film or the Firesign Theater DVD entitled Hot Shorts in which new dialogue is put to episodes of old movie serials, e.g., Spy Smasher becomes Revenge of the Non-Smokers)

New WPE with link to original moving image work. In most cases, changes in soundtracks are either translations or commentary and do not make a new work. However, in this case, the new plot presented in the soundtrack fits the FRBR criterion of “significant degree of independent intellectual or artistic effort” and therefore should be considered a new work. Most works of this sort are parodies, which are explicitly defined in the FRBR report as new works (p. 18).

Helpful for anyone FRBRizing movies. Martha Yee was one of the advisors.

I didn’t get to the pre-IFLA conference about Resource Description and Access, but now I can see what I missed: the presentations are up on the Joint Steering Committee web site.

Gordon Dunsire’s RDA Vocabularies and Concepts (114 KB PDF) will be of particular interest to people interested in RDF, the Semantic Web, SKOS, etc. Have a look.

My colleague Tim Knight posted his impressions of the event on the York University cataloguers’ blog after he got back.

Kathy Glennan, chair of the Music Library Association’s Bibliographic Control Committee, last week announced on the RDA-L mailing list:

The Music Library Association’s Bibliographic Control Committee (BCC) is pleased to announce the availability of the Final Report of The BCC Working Group on Work Records for Music (available at: http://www.musiclibraryassoc.org/BCC/BCC-Historical/BCC2008/BCC2008WGWRM1.pdf). As BCC chair, I constituted this short-term working group in April, charging the seven members to look at the issues surrounding the question of what elements and attributes of musical works should be included in a work record, once the cataloging community moves beyond AACR2 and MARC21 to a new descriptive model that stores data in a relational or object-oriented database.

It’s a short report, nine pages plus a two page introduction. From the Discussion section towards the end:

As most musical works in the Western canon contain an internal structure (movements, arias, etc.), it is desirable, and indeed necessary, to describe this structure in the work record. Current authority record practice sanctions the creation of separate records for parts of larger works, when necessary for creating access points. The user then infers the whole/part relationship based on the syntax of these access points (e.g. “Zauberflöte. Hölle Rache” is a part of “Zauberflöte”).

In a relational database environment, such whole/part relationships can better be represented by dynamic relationships (expressed as links) between larger works and their parts. This is especially desirable when the parts have discrete attributes of their own (e.g. Variant titles, Keys).

Hey. How’s it going? I was on vacation for a while there, then taking a bit of a break while I got caught up on things at work, so I have a few things in the backlog here. Hope you had a good summer. (Or winter, if you’re in the southern hemisphere.) I’m reading Neal Stephenson’s Anathem, in case you were wondering. Look for some news about a revived project soon.

Jenn Riley sent out this announcement late last week. Congratulations to them on getting funding for this project! If they do what they want to do, we’ll all benefit.

Indiana University is pleased to announce funding for a National Leadership Grant from the Institute of Museum and Library Services for a project entitled “Variations as a Testbed for the FRBR Conceptual Model” <http://www.imls.gov/news/2008/091008a_list.shtm#IN>. This project will build on Indiana University’s expertise in digital music libraries and the well-known Variations digital music library system < http://www.dlib.indiana.edu/projects/variations3/index.html>, and provide a concrete testbed for the Functional Requirements for Bibliographic Records (FRBR) conceptual model. This project is focused on testing FRBR in a real-world environment, and on providing data, code, and system design specifications that can be re-used by others interested in FRBR.

These are our primary planned project activities:

• Convert the production Variations system to use a FRBR-compliant data model
• Adjust our existing algorithm for use of MARC records to conform to the FRBR-compliant data model.
• Load FRBR Group 1, 2, and possibly 3 records for all score and recording holdings in the IU William and Gayle Cook Music Library (approximately 80,000 bibliographic records for audio recordings and 105,000 records for scores) into the redesigned system.
• Make FRBRized records available for community use via OAI-PMH, SRU, and batch download.
• Design and implement a new, openly-accessible search interface for discovering FRBRized data
• Design and implement a new cataloging system for FRBRized data that takes advantage of the distinction between the FRBR entities yet supports efficient data entry.
• Perform usability testing on the new end-user and cataloger interfaces to evaluate their effectiveness.

Our project expects to have the following concrete work products:

• A published FRBRization algorithm that operates on multi-Work Manifestations, and
  evaluation of its effectiveness
• A formal data model for FRBR, if none is available before our project begins
• FRBRized data made available to the community for further testing and analysis
• An openly-accessible system for searching FRBRized music data for community testing and analysis
• Usability evaluations of FRBR-based end-user discovery and cataloging systems
• Figures on the costs of creating FRBRized bibliographic data by both automated and
  manual means
• Source code for the Variations FRBR-based discovery system.

The three-year project will begin October 1, 2008. We look forward to starting work on this project, and sharing our progress widely.