Jonathan Rochkind wrote,
> ... we _desperately_ need what FRBR is trying to do---a formal and explicit
> schematic of how we
> model the 'bibliographic' (or 'information resource') universe.
> ...
> So to me, what we need is to convince people that indeed FRBR is an
> unfinished product, ...
Many people need to be better informed first about FRBR and its goals,
and problems of its implementation.
I see one big problem, and that is the linking of records. Up until now,
there are no record links other than the "textual" ones in MARC records
as they are. Using control numbers for linking presents a much bigger
challenge, esp. in data exchange and sharing. But if this challenge
cannot be met, I'm afraid there can be only half-hearted FRBR
implementations, far from the full potential of the concept.
Another big problem is that we need to be more specific about indexing.
What is to be indexed and precisely how? A distinction must be made
(and it presently is not being made) between an index(1) that is never
displayed as such but only consulted by search software, and an index(2)
that can be displayed and used for browsing. For headings (names and
subjects) and also titles the latter has much potential but too little
prominence in existing systems. See Martha Yee's work.
Some background and perspective around FRBR, and the differences
between catalogs and search engines, are elaborated upon in an
updated version of my paper "On the theory of library catalogs
and search engines":
http://www.allegro-c.de/formate/tlcse.htm
Regards, B.Eversberg
Received on Thu Nov 29 2007 - 04:26:26 EST