Martha Yee wrote:
> Thanks for your comments, James...
>
> I have argued elsewhere (in the FRBR-ization paper that I cited in my
> initial posting) that the FRBR tables do not correspond to the FRBR entity
> definitions. If you are going to define expression as any change in content
> that does not lead to a new work, and manifestation as mere change in format
> or distribution information, it is absurd to map the statement '2nd rev.
> ed.' to manifestation. It is clearly a statement about
> expression. We must
> be willing to trust publishers' statements on title pages unless proven
> erroneous. The Yee cataloging rules represent, among other things, my
> attempt to remap the elements of a bibliographic description to the correct
> FRBR entities, thereby keeping the original FRBR definitions but discarding
> the flawed mapping tables at the back of FRBR. There is the question,
> though, of whether catalogers always have enough information to map all data
> elements to either expression or manifestation. I suspect that the
> original
> designers of the Anglo-American cataloging rules judged that catalogers did
> not always have enough information, so limited their collocation work to the
> work level (with main entries).
Again, this is very important work.
I agree that the expression/manifestation/item is messed-up in FRBR, and while edition statements should certainly go into expression, I'm not so sure that publication information should as well. I realize that this has always been the practical way of determining expressions. But still, if we put the publication information into the expression, it seems to me as if it will have serious consequences in the future, as different formats with the same expression will come up more and more often, especially as different agencies will have different rules determining manifestations, such as the differences between LCRI 1.0 vs. "Differences between, changes within" from ALA. Many other rules will probably have their own formulations of manifestation, e.g. ONIX with differences in ISBN/hardcover/softcover; when I worked for the UN, we had different rules, and so on. I think one of the purposes of FRBR should be to attempt to bring these different records together, as much as possib!
le.
Additionally in the future, we will be able to measure extents of expressions by word counts, check sums, or simple file compares, something we could never before think of doing except through an incredible amount of tedious work (nobody except antiquarian book dealers or textual scholars check different copies word for word), and we do "trust" the publishers on this (too much sometimes)).
I still believe that the problem is with the concept of the manifestation, which is only a "virtual view" of the work/expression/items, depending on how you want to define the manifestation. This is quite different from the other entities. To me, publication information should definitely go into the item record(!), while dates should be expanded much more than they are now. Catalogers are told to either add dates or ignore dates based on all sorts of reasons. All of this could wind up being more accurate than it is now.
The "manifestation view" could be generated from the RDF very efficiently.
James Weinheimer
Received on Tue Dec 04 2007 - 11:08:15 EST