Karen ended with "That [what is an edition] is something we humans infer
from the data provided."
This reminds of how author-title references worked so well with cards. We
depended on humans to make the inferences necessary to go to the right
place in the drawers.
Then when we moved to online catalogs, somehow it became extremely
difficult for systems to make the same leap. Right now I have an OPAC that
will link you to the author or link you to a title in an author/title
reference in a bib record, but it can't do both. (Other systems may have
succeeded in this.)
In short, I thought we were supposed to move to a world where we were
going to be coding relationships *more* explicitly, not less so. That this
was going to make it easier for computers to handle our data. That is,
humans will no longer be asked to infer anything, right? ("If we have to
teach our users how to use the catalog, we have failed.") One of the
reasons OCLC's FRBR algorithm is less than optimal is that they are having
to infer things from existing legacy data.
I guess I need to see a mocked up example of how records can be "complete"
and contain data that would support a variety of FRBR relationships,
enabling people to draw the boundaries where they will, but at the same
time no where explicitly codes information about such relationships. I
just can't imagine it at all.
************
Diana Brooking (206) 685-0389
Cataloging Librarian (206) 685-8782 fax
Suzzallo Library [EMAIL PROTECTED]
University of Washington
Box 352900
Seattle WA 98195-2900
On Wed, 5 Dec 2007, Karen Coyle wrote:
Jonathan Rochkind wrote:
On Dec 5, 2007, at 9:04 AM, Karen Coyle wrote:
Well, I disagree with your basic assumption that "edition of Hamlet =
work."
I don't think it's any easier to decide whether a particular
manifestation is or isn't "an edition of Hamlet" than it is to decide
whether a particular manifestation is a member of the "Hamlet"
workset. The "is an edition" relation is just as contextual and
community dependent as the "is a member of work" relationship, so I
don't think we actually get as far as you think into easy or
objective decisions by focusing on the former rather than the latter.
Jonathan, you are leaping to the "is an edition" decision, which I never
mentioned. I say that you record the title, the edition statement from
the piece, the number of pages, illustrations, size, the date of
publication, the publisher. Whether or not that is an edition of Hamlet
will then depend on your defintion of "is an edition." I don't think
that today our records (with the exception of the uniform title which is
hardly ever used when it should be) actually state "is an edition." That
is something that we humans infer from the data provided.
kc
--
-----------------------------------
Karen Coyle / Digital Library Consultant
[EMAIL PROTECTED] http://www.kcoyle.net
ph.: 510-540-7596 skype: kcoylenet
fx.: 510-848-3913
mo.: 510-435-8234
------------------------------------
