I think we're providing places for there to be multiple instances in the data, so that catalogers who wish to may further describe the paperback, although in practice you wouldn't unless there was something special to capture, or you have a rare book.

The database is going to be cluttered with a lot of data that users never see, but which can be drawn upon to show FRBR-like relationships between and  among the various ways we have of describing a book.

Here we're talking about transforming out from existing MARC records, so it's just a machine transformation. No one will be required to create a new bibframe record with multiple instances for each ISBN. But by splitting them out, you make it possible for as much further description as is desired.

I can see a use case for data transfer where someone requests a work record and the first (probably hardback) instance as a package, as opposed to the whole ball of wax, since that fits their library best, and in fact that's the item in hand most of the time when cataloging was done.


-----Original Message-----
From: Bibliographic Framework Transition Initiative Forum [mailto:[log in to unmask]] On Behalf Of J. McRee Elrod
Sent: Tuesday, January 29, 2013 12:30 AM
To: [log in to unmask]
Subject: Re: [BIBFRAME] Bibframe and translations from MARC

Kevin Ford asked:

>It might not *always* be correct (perhaps it does boil down to a
binding >difference only), but from what we're seeing creating Instances based on >differing ISBNs is right nearly all of the time.

It seems to me a great disservice to patrons to clutter the database with multiple records for the same content.  It would complicate holds and ILL.  Binding differences are better suited to item records.

We go too far in ignoring differences between manifestations with Provider Neutral (PN) records for e-books.  We don't use differing
DOIs* as the basis for a different record.  But we would be going too far in the other direction to create different records for hardback, paperback, library, deluxe binding, etc.

The AACR2 definition of an edition is all copies produced from the same type image.  That seems a good principle to me, and based on long term experience.

It seems to me Bibframe is forgetting patrons convenience in this regard, as well as in omitting punctuation.

*Another important access point not yet in Bibframe.

   __       __   J. McRee (Mac) Elrod ([log in to unmask])
  {__  |   /     Special Libraries Cataloguing   HTTP://
  ___} |__ \__________________________________________________________