[SAC-FAST] History as form subdivision

[Stephen Hearn]

For Eric--the record where "Fiction" (not History) appears as both 650 and 
655 is #2043. Mostly "History" and "Fiction" do go into 655. "History and 
criticism" in my sample generally follows a 650 topical subject and stays 
bound to it in a $x. "Criticism and  interpretation" on the other hand, 
which usually follows a personal name, splits off and moves into a 655. 
Record #3137 includes both these "criticism" treatments.

The other consistency issue I'm seeing involves geographic headings. Some 
have a higher level body inserted, e.g., "United States $z Chama Valley" 
(derived from "Chama Valley (Colo. and N.M.)"), "Eastern Hemisphere $z Rome 
(Empire)"; and others don't, e.g, "Ohio River Valley." It could be hard to 
guess which treatment one of these regional entities would get. Is there a 
rule of thumb for guessing the converted FAST form which doesn't require 
knowing the LCSH heading's form? If the FAST terms are being searched as 
keywords rather than browsed as headings, maybe that wouldn't matter.

Of course, how the FAST headings would be searched or browsed is another 
big question.

Stephen



At 12:58 PM 5/26/2005, you wrote:
>Content-class: urn:content-classes:message
>Content-Type: multipart/alternative;
>         boundary="----_=_NextPart_001_01C5621C.7EAB0BB1"
>
>
>Sherman, Stephen, thanks for the comments.  Let me respond very briefly 
>and invite discussion (and particularly responses from the FAST team -- 
>Ed's away from the office at the moment, but I'm sure he'll want to 
>respond; we've thought/talked around so many the free-floating topical vs. 
>genre cases  so many times that I'd happily have other members of the FAST 
>team correct/amend any mis-comments I may make :).
>
>"History" is one of those items that has consumed quite a bit of time for 
>the FAST team to ponder many options.  No option has struck us as entirely 
>satisfactory, but moving it from a free-floating topical subdivision to a 
>genre term seemed to offer a pragmatic solution.
>
>Stephen -- I wasn't able to find the record (I didn't look too hard I'll 
>admit ;) with it being treated two ways; can you specify which it is?  My 
>guess is that we intended to have "history" apppear solely as a genre 
>term, and the output might a case that the algorithm didn't handle as 
>intended.
>
>"History and criticism" and various other cases have been part of 
>conversations, but let me defer to others to respond specifically.  The 
>point that predictability for searchers is a practical requirement is 
>absolutely on target.  That's our intent.
>
>Thanks for the feedback -- this exactly what we'd love to get from SAC!
>
>Cheers,
>
>Eric
>
>Eric Childress
>Consulting Project Manager
>OCLC Research {<http://www.oclc.org/research>http://www.oclc.org/research}
>OCLC Online Computer Library Center, Inc.
>6565 Frantz Rd., Dublin, OH 43017 USA
>US: (800) 848-5878 or (614) 764-6000
>Fax: (614) 718-7361 email: eric_childress at oclc.org
>
>_______________________________________________
>SAC-FAST mailing list
>SAC-FAST at mailman.yale.edu
>http://mailman.yale.edu/mailman/listinfo/sac-fast

****************************************************
Stephen Hearn
Authority Control Coord./Database Mgmt. Section Head
Technical Services Dept.
University of Minnesota
160 Wilson Library               Voice: 612-625-2328
309 19th Avenue South              Fax: 612-625-3428
Minneapolis, MN 55455      E-mail: s-hear at tc.umn.edu 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.yale.edu/pipermail/sac-fast/attachments/20050526/3c278bdf/attachment-0001.htm