How about this scenario? You like a book by John Smith. You retrieve it by title Book X. The record includes a hotlink with message: More by this creator. If all John Smiths are controlled by one undifferentiated record, you will retrieve books by all of the undifferentiated John Smiths. If each currently undifferentiated John Smith has a separate authority record (a separate ID) linked to all titles created by John Smith the author of Book X, I think that would be a great service for the user.
Catalog Librarian for Training & Documentation
Catalog & Metadata Services, SML, Yale University
P.O. Box 208240 New Haven, CT 06520-8240
(203)432-8286 [log in to unmask]
From: Program for Cooperative Cataloging [mailto:[log in to unmask]] On Behalf Of Benjamin A Abrahamse
Sent: Wednesday, April 04, 2012 2:41 PM
To: [log in to unmask]
Subject: Re: Undifferentiated personal names: call for community discussion
Respectfully, I wasn't trying to dismiss your comment, I honestly could not determine your tone.
I don't disagree with you that we should be looking at the costs vs. benefits of cataloging practices. Nor do I feel that RDA has been evaluated rigorously enough in this respect. Like catalogers everywhere, I suspect, I sort of feel like I'm about to hop on a train to an unknown destination, and the only reassurance that I have that it's where I want to go is, "Well, everyone else is getting on this train..."
But with respect to undifferentiated NARs in particular, or even NARs in general, while I would love to see some empirical evidence regarding their usefulness (or lack thereof) I'm honestly not sure how one would go about measuring it. The value of authorities practice is primarily to provide an infrastructure that users don't necessarily see, or interact with directly. (A lot of the newer, so-called "next gen" discovery interfaces I've looked at, such as WCL, Summon, and EDS, don't make the authorities index visible to users at all.)
Like real-world infrastructure, authorities are only apparent to users when they're not there, or not functioning properly. We naturally notice the bridge that collapses, not the thousands that don't. How does one measure that in an empirical fashion? (Not speaking rhetorically here, if someone has an idea how to do this I'd be interested in it.)
Acquisitions, Metadata and Enterprise Systems