I've just received the following privately from a well known cataloguer
who does not wish to be named. This follows an earlier communication
from another of our more famous cataloguers who did wish to "make
enemies".
It leads one to wonder just how cut throat things are at the upper
levels of our profession :-{)} . Perhaps it should make me worry about
having become known as a !@#$ disturber, in that I'm the one they tell
:-{(} .
In fact, I think they are wisely reserving their strength for the really
big upcoming decision making moments.
------------------------------------------------------------------
Tuesday, October 9, 2001
Mac,
With over 30 years experience, I think it's time MARC studied how
many fixed fields are seldom or never used for retrieval. I don't mean
just assigned during cataloging. For example, has the indication of
the presence or absence of an index in 008/31 ever been used? ...
--------------------------------------------------------------------
I've felt for sometime that if we wish to address redundancy in MARC. we
should not remove explicit information in favour of fixed field
information (gmd for example), but rather reduce fixed fields
which date from the time it was too expensive to search variable
fields as we now do routinely. Fixed field 008/31 (Index), if wanted,
could be automatically generated from the presence of the word "index"
in 5XX, usually 504, but occasionally 500 (those who thought "Includes
index" should be a 504 were probably correct). Wouldn't most patrons
just use keyword in the first place? Do any OPACs actually have patrons
use a 008/31 based Boolean search? Do those libraries have logs which
show if that option has been used? How many actual recorded uses are
required to justify the time we spend coding it?
Isn't it time for a ranking of fixed fields in the order of actual
*recorded* use in retrieval? Shouldn't this then lead to their drastic
reduction in number? Into whose ball court does that bounce?
Please note that I am not talking about 008/07-10 (Date 1) which should
be used to create inverse order of hits under subject searches, nor
008/35-37 (Language), which has been used for creating monolingual
bibliographies, and possibly even related to initial articles for
filing. But many fixed fields are simply not used in OPACs. Others
we've made less useful by ambiguous use (LDR/6 "a" language material now
coded for most computer files for example). The usefulness of *all*
fixed fields should not be defended on the basis of the usefulness of a
*few*.
Another thing which bothers me about JSC (Joint Steering Committee for
the Revision of Anglo-American Cataloguing Rules) discussing the
omission of gmd (or any explicit information) in favour of fixed fields,
is the cross standards nature of the proposal (found in a report* to JSC
by Barabara Tillett). The gmd is a matter for AACR; fixed fields are a
matter of MARC, and not even *mentioned* in AACR. Should the JSC even
be talking about fixed fields? Jean Weihs' survey* concerning gmds
showed the rejection of the idea of its being dropped in favour of fixed
fields by most respondents. Since then it has been pointed out that
cataloguers working with raw MARC records would find it time consuming
to consult a fixed field to identify the approprate record for the
format in hand, that fixed field based icons might not work with one
line title lists, nor translate across Z39.50 based intercatalogue
searches.
Shouldn't the powers that be for MARC, based on a usage study, make
some fixed fields at least optional, if not obsolete? Shouldn't JSC
stay away from fixed fields as being outside their mandate?
__ __ J. McRee (Mac) Elrod ([log in to unmask])
{__ | / Special Libraries Cataloguing HTTP://www.slc.bc.ca/
___} |__ \__________________________________________________________
*http://www.nlc-bnc.ca/jsc/current.html#GMD
|
