I think Expression/Expression relationships probably _can_ handle
connecting different performances of the same arrangement.
We don't actually need entities for everything, and shouldn't assume
that the entities in the model are the only things that will ever be
presented to the user. Other things can be presented to the user based
on relationships and attributes of the data. (Karen will like that
agreement, even though I still disagree that we can get by without
recording work and expression relationships in the data!).
Still, the four entity model seems a reasonable compromise to give us a
basic framework, and is a lot better than if we had a model that had
nothing _but_ "things" and "things related to things." The four-entity
model serves to make clear certain types of relationships and sets which
100 years of cataloging have shown are especially useful to the general
purpose bibliographic universe--and recent experimentation shows apply
fairly well to digital items too.
Some people think the four items is already too many (suggesting we get
rid of Expression, or if I understand Karen right, that we get rid of
Expression AND Work, and have nothing but manifestations and
relationships!). Others have found that the four entities may be too
few, and more would be more convenient to them (Jenn's music
experience). To me the four entities are a reasonably good general
purpose model framework on which we can hang everything else. As a
conceptual model, it is not meant to provide the One True Description of
Reality, just a _convenient_ generalized model which gives us a common
frame of reference.
Jonathan
Riley, Jenn wrote:
>> Jean, so if performances are expressions, does that mean that
>> a new arrangement is a different work? Because otherwise you
>> have no place for the performance of a new/different
>> arrangement, right?
>>
>
> Indiana University has done some work applying FRBR principles to music, and we definitely ran up against problems using Expression to define both the arrangement and the performance. It may be that to model the performing arts we'd really need yet another entity. But we wanted to see how we'd do it using FRBR as it was, so we for the most part defined arrangements as new Expressions (not new Works), and assumed Expression/Expression relationships could handle connecting different performances of the same arrangement together. More information (*much* more information) on our thoughts is available from this report: <http://www.dlib.indiana.edu/projects/variations3/docs/v3FRBRreport.pdf>
>
> Jenn
>
> ========================
> Jenn Riley
> Metadata Librarian
> Digital Library Program
> Indiana University - Bloomington
> Wells Library W501
> (812) 856-5759
> www.dlib.indiana.edu
>
> Inquiring Librarian blog: www.inquiringlibrarian.blogspot.com
>
>
--
Jonathan Rochkind
Digital Services Software Engineer
The Sheridan Libraries
Johns Hopkins University
410.516.8886
rochkind (at) jhu.edu
Received on Wed Dec 05 2007 - 16:59:52 EST