No. It is not an issue of patrons and display. I believe Karen's "users" in this context are librarians and library systems. She is saying that BibFrame development is going to go off the tracks if it continues in a fashion that is willfully ignorant of the realities of the "bibliographic data marketplace" -- the assumption of a one-to-one relationship involving the ISBN and the "Instance" is not tenable. As articulated by others, despite the specification that ISBNs should be uniquely assigned, they aren't; conversely, different ISBNs are assigned to manifestations that librarians assess as equivalent (e.g. different bindings, same content). Then there are resources (contemporary and historical) to which an ISBN was never assigned and cases where individual and "collective" ISBNS also apply to a single item. All of these failures of one-to-oneness, while perhaps a small number in the overall set of records, are nonetheless sufficiently significant (in both their number and the specific resources they impact) that a system predicated on a one-to-one arrangement will a) fail in the technical sense and b) be rejected, thereby failing in the commercial sense.
There's a joke that goes around in Navy circles: An aircraft carrier radios a distant radar contact that they are on a collision course and directs the radar contact to change course (since the carrier is bigger, more powerful, etc.). The radar contact radios back that it is a lighthouse and the carrier can choose accordingly.
BibFrame is being told it's the aircraft carrier in this case.
John F. Myers, Catalog Librarian
Schaffer Library, Union College
807 Union St.
Schenectady NY 12308
[log in to unmask]
From: Bibliographic Framework Transition Initiative Forum [[log in to unmask]] on behalf of Ford, Kevin [[log in to unmask]]
Sent: Friday, May 17, 2013 3:24 PM
To: [log in to unmask]
Subject: Re: [BIBFRAME] What's an instance?
> What your potential users (who may choose to become non-users pretty
> quickly) are telling you is that they have chosen this method because
> they do NOT want to present two different options to users where those
> options are not meaningful to the user.
-- By "user," I take it you to mean "patron" (this is fine, just want to be clear who the user is). In that case, this is a display issue, and not directly within the the scope of BIBFRAME. Do I think displays should find a graceful way to deal with such a skimpy Instance? Absolutely. Is it within the scope of BIBFRAME to solve that problem the issue of UI design? No.
More importantly, it should be borne in mind that this is first and foremost about data representation and exchange. The BIBFRAME model need not determine how underlying systems store, crunch, or manage the data.