Thanks for this feedback. Not that you disagree with this, but it bears repeating that BIBFRAME is not a only FRBR or RDA cataloging module, so we are not necessarily saying that a BF:Work has intention.  The definition from      is “Resource reflecting a conceptual essence of the cataloging resource.”  I think there’s a lot of room for describing security tapes or recorded birdsong in that definition. We would expect FRBR works to be expressed in BF using bf:Work, but so would RDA expressions. It’s a similar concept to using MARC to store your AACR2 cataloging.  





Nate Trail


LA308, Mail Stop 4402

Library of Congress

Washington DC 20540





From: Bibliographic Framework Transition Initiative Forum [mailto:[log in to unmask]] On Behalf Of Kelley McGrath
Sent: Tuesday, September 02, 2014 12:03 AM
To: [log in to unmask]
Subject: [BIBFRAME] Moving images, FRBR & Bibframe


It was suggested to me that there might be some interest on this list in the presentation on FRBR and moving images that I gave at the FRBR Interest Group meeting at ALA:

It is interesting, although not surprising, that many of the topics I discussed in the presentation overlap with concerns raised by the BIBFRAME AV Modeling Study. The BIBFRAME AV Modeling Study clearly articulates the many challenges faced by the moving image cataloging community when trying to work with existing generalized models. It does a good job of inventorying a variety of materials and different requirements, some of which lie outside the scope of resources commonly collected by libraries.

The problems that led to the proposal of an event entity are clear, but I haven't wrapped my head around it enough to be sure what I think of this approach. I wonder if some different types of events are being conflated. There is the existing model of events as subjects. Many models take into account events in the creation of a resource. The report also discusses events depicted by the resource, which seems similar to what would be represented by the dpc MARC relator term. For audiovisual resources, events and people as subjects, as depictions and as agents often overlap and exist on a continuum. Information associated with events in the creation of a resource are most wanted in archives, but even when dealing with commercial materials, there are obvious use cases. People are often interested in the filming location of feature films and may want to know the date and place where a filmed performance was recorded.

I am not completely convinced that there is no work in a collection of recordings of birdsongs or congressional hearings. From one point of view, a FRBR work is an intellectual or artistic creation that is bound up with intention and that can be contrasted with a straight act of recording. On the other hand, I think the idea of intellectual or artistic creation could be interpreted broadly enough to include essentially all human artifacts. Even with a plain recording, someone set up the camera at a certain angle and chose a start and stop point.

I wonder how the suggestions in this study relate to the FRBRoo model, which includes a recording work and performance work. In FRBRoo, the Recording Work is defined as "the artistic and technical decisions made about the recording material to be used, the location of microphones and/or cameras, the use of filters, lighting."

The study makes many useful suggestions. The inclusion of an explicit Collection entity seems helpful. The problem of aggregates and moving images makes my head hurt and perhaps this would help. Are things like publisher's series also a kind of collection? Even commercially-published DVDs often include outtakes, bloopers and alternate scenes. TV series present many problems. Consider a DVD compiling the best episodes of a TV series. As mentioned in the paper, reconciling the descriptions of TV series based on DVDs of seasons versus descriptions based on individual episodes is difficult. If the information in each is limited to the level being described, the data needed to make connections is lacking.

I would welcome support for explicitly distinguishing different types of dates, for describing more technical attributes at all levels, and for identifying sequences and time stamps, as suggested by the study. I am glad to see the Library of Congress and the developers of Bibframe take the unique requirements of describing audiovisual materials seriously, as many library standards merely squeeze them into a text-centric model. I hope that this will result in Bibframe being more hospitable to audiovisual materials.


PS If you are so inclined, why not take a few minutes to help OLAC out with our project to annotate movie credits in MARC records so we can try to teach a computer to do this: