[1014] in Public-Access_Computer_Systems_Forum
Subject Headings for Gopher Virtual Library
daemon@ATHENA.MIT.EDU (Public-Access Computer Systems For)
Wed Aug 19 09:19:23 1992
Date: Wed, 19 Aug 1992 08:16:00 CDT
Reply-To: Public-Access Computer Systems Forum <PACS-L%UHUPVM1.BITNET@ricevm1.rice.edu>
From: Public-Access Computer Systems Forum <LIBPACS%UHUPVM1.BITNET@ricevm1.rice.edu>
To: Multiple recipients of list PACS-L <PACS-L%UHUPVM1.BITNET@ricevm1.rice.edu>
5 Messages, 124 Lines
*-----
From: li40163@prism.gatech.edu (Mary A. Axford)
Subject: Re: Subject Headings for Gopher Virtual Library
The biggest problem I see right now with Library of
Congress subject headings is that it does not at all seem
adequate in some areas. There is the NLM scheme that breaks
out medicine. Working in a science and technology library,
the LC scheme seems to me to be sorely inadequate in the field
of computer science. For one thing, the call number ranges
are too small (call numbers may or may not be part of your
subject indexing scheme). Then you have some headings that
are similar, but not exact, or are out-of-date but still in
use or used for older records.
So a suggestion might be to use LC as the basis, but to
make the first task figuring out in what areas LC is
inadequate and needs expansion.
*-----
From: slides@ced.berkeley.edu (Arch Slide Library)
Subject: Re: Subject Headings for Gopher Virtual Library
As an architecture slide librarian (ie no books) I suggest
that you not use Library of Congress. Do ou really need
long stings, many of which are inconsistent and overlapping?
Many of which are actualoy too general to pin point access?
There has been considerable discussion in some segments of
the library and information management world that LC
subject headings should be de-coordinated, that is,
broken apart from strings into more discrete words.
Then a cataloger, indexer, or person of any ilk could apply
the most specific terms, and if they so desired they could string
them. It is important that the individual terms be searchable
whether they are in strings or not. We are using the ARt and
Architecture Thesaurus (AAT). This is currently a controlled, hierarchic
vocabulary of 40,000 terms. In a year or two the second edition
will be released in about 100,000 terms of art and architecture.
The terms are arranged in 7 facets (groupings larger than hierarchies)
and 23 hierarchies. Broader and narrower terms are easily searchable
because of the hierarchic arrangement. Many terms have scope notes
and the AAT supports alternate terms, usually the singular form.
The cataloger-indexer is free to use individual terms, lots of
individual terms (wood, Victorian, houses), combine terms into
expressions (wood Victorian houses), or make complex expressions
known as strings (wood Victorian houses--pattern books). It was the
development of the AAT with its atomized terms that prompted
serious discussion about decoordinating LCSH. The problem with
the AAT is that it is too specialized for your material. Nevertheless,
its usefulness lies in its atomistic nature. It is easier to apply
individual terms that pre-existing strings. If LC catalogers cannot
apply the same LCSH string consistently, why should you try?
Maryly Snow
Architecture Slide Librarian
UC Berkeley
slides@ced.berkeley.edu
*-----
From: suef@theory.TC.CORNELL.EDU (Susan Feldman)
Subject: Re: Subject Headings for Gopher Virtual Library
I'm very interested in your subject approach. It's exactly what will
make the Internet usable. I'm a librarian by training and would be
very interested in helping, if I can. I run a library service, and
spent last year comparing various classification schemes for two
computing oriented libraries. LC is a disaster for computers,
especially for new subjects such as supercomputing. The numbers
are limited mainly to QA76 and TK 7885 or therabouts, and books
are scattered all over in various subject classifications, but
unpredictably. Subject headings are peculiar. For instance,
UNIX and other operating systems last year were given a new
heading--"computer file". You can't find them under operating
system any more. Computer file of course means many different
things, and is not very precise. There are other vagaries, and
new terminology is added slowly.
On the other hand, since the Internet covers all subjects, you may have to stick
with LC. Otherwise I'd recommend using the INSPEC thesaurus and classification.
Sue Feldman, Datasearch
*-----
From: jahb@Lehigh.EDU (JENNIFER A. HEISE)
Subject: Re: Subject Headings for Gopher Virtual Library
The LoC Subject Headings make a LOUSY hierarchical organizational scheme for
information. They are too diffuse, too broad, and have absolutely NO
organization. If you tried to diagram the LoC SubHead. relationships, you'd
get something that looked like an extremely perverted nest of snakes.
On the other hand, there is something to be said for checking out the Dewey
Decimal System's division headings. Start with the 10, then the 100, then the
1000 divisions. It ain't the best of all possible schemes, but it IS a start.
Perhaps the more modern European variants of Dewey might help...
Jennifer Heise Net: jahb@lehigh.edu
Reference Dept., Phone: (215) 758-3072
Fairchild-Martindale Libraries #8A, Lehigh University, Bethlehem, PA 18015
****"You know I don't know so many things, but I know what's been goin' on:
****They put in a nickel, and they want a dollar song"--Melanie Kafka
My opinions are my own. No one else would HAVE them anyway.
*-----
From: SOLIBJA@GSVMS2.CC.GASOU.EDU (JULIUS ARIAIL)
Subject: RE: Subject Headings for Gopher Virtual Library
If you really mean the LC "subject headings" list rather than the LC
"classification schedule," then I certainly would not use the "subject
headings." They are not well organized in a hierarchical sense; if you
have a basic subject, it's hard to link it all the way up and down a
scale of subjects from narrower to broader. At best, you would have a
set of references for one subject which would move you a step up or a
step down.
The LC "classification schedule" would be a better choice, since most
academic libraries in the US already use it and many individuals are
familiar with it. It's old and clunky, but it's infinitely expandable
and the Library of Congress does all the work in coming up with new
numbers for, let's say, the "biology of Mars." And most libraries have
a complete set of the printed class schedules handy which would be
easy to consult.
Julius Ariail, Director of Libraries solibja@gsvms2.cc.gasou.edu
Georgia Southern University Statesboro, GA 30460
