Thoughts on FRBR sparked by a discussion on RDA-L (archives for month in progress not online, sadly) about FRBR modelling of moving pictures (thanks Martha Yee), which itself was sparked by Diane Hillman’s excellent stab at creating some RDA cataloging scenarios (which I can’t find the url for having no archive of RDA-L to search. You know, I should go and subscribe a google list to it and make my own archive henceforth).
1) Even though FRBR with it’s four Group 1 entities is already considered too complicated by some people (who think we only need three, or two), it’s actually still only a modelled approximation of the complexity of our actual bibliographic/information universe. What we represent will neccesarily be one model, and an approximation. But that’s okay–our systems can and should draw relationships between what the users need not limited to the Group 1 Entity hieararchy anyway! (That is to say, just becuase the Group 1 sets don’t capture what you want, doesn’t mean it can’t or shouldn’t or won’t be captured by the system).
I continue to think that the FRBR model is a pretty good skeleton for us, based on an evolving and evolved cataloging tradition. [But don’t get me started on the FRAD component, which I still think is fatally flawed. ]
2) The work of coming to general agreements about conventions for how we are going to model the messy real world in our model to have reasonable consistency–is still a work in progress. Which is fine, as long as it actually progresses. Discussions like this help. Ideally I would hope that RDA will give guidance on this: Here is how to decide work boundaries, here is how to decide expression boundaries, etc.–that’s the function I would see RDA playing. I’m not certain if it does or not.
Certainly AACR2 does to some extent (with directives as to when two items in hand belong on the same bib record [regarding manifestation boundaries]; with directives as to how to create uniform titles [regarding work and expression boundaries–sort of]), but not neccesarily in the right ways, or in language that is clear as to the principles and purposes at stake (RDA keeps saying it means to be principles based, and have those principles be transparent, which is good). What do people think, is RDA on it’s way to succesfully giving guidance here?
3) While it is, I agree, useful to have some conventions and guidelines for consistency in a document like RDA, exactly how lock-step this consistency needs to be is something that needs to be investigated and analyzed (and which there is and will be significant controversy on; one of the poles of cataloging thought is definitely to try and enforce
absolute uniformity, although to my reading that’s never been that
But no matter how much uniformity we try to enforce, and even if we are succesful within our ‘own’ community, we will, like it or not, be integrating data from entirely different sources which do not follow those uniform rules too.
I actually think we can likely be succesful in getting these other sources to follow the Work-Expression-Manifestation-Item model (because it is useful, and there is nothing competing), or else succesfully translate foreign data to this model. But we aren’t going to be succesful in forcing everyone everywhere to make the same decisions about exactly where work boundaries or expression boundaries are. So an unresolved question and useful research program is: How do we build systems that deal with integrating records from sources that have made _different_ decisions about where work boundaries or expression boundaries are?
(or even manifestation boundaries; ISBN can be seen as a different decision about how to draw manifestation boundaries, although one that isn’t as problematic as it could be because their manifestation sets end up being strict subsets of ours; I think. Would be worse if they were overlapping rather than strict sub/superset..) (*)
[*] I continue to find both entity-relational model and a set theory model to be useful and complementary, not contradictory, in understanding the bibliographic/information universe and FRBR. I think that the FRBR document would be better served to explain the definition of the Group 1 entities in terms of set theoretic concepts (ala Svenonius and Yee among others). I think catalogers, metadata technicians, and librarians working with metadata ought to make sure they have a basic understanding of set theory/set algebra–which is not such a hard thing to do, especially for the cataloging mindset.