[165] in ad-lib
B.Index meeting, my take on it: Part 1 of 2
daemon@ATHENA.MIT.EDU (sbyrd@MIT.EDU)
Wed Mar 15 17:03:44 1995
From: sbyrd@MIT.EDU
To: ad-cat@MIT.EDU, tat@MIT.EDU
Cc: m.blake@geac.com
Date: Wed, 15 Mar 1995 17:03:23 EST
B.Index----results of our meeting today with Miriam Blake at Geac:
we're all confused. No, just kidding! We've got it down now!
Really!!
Here is part of what we decided: (PLEASE let me know if any of this
contradicts what you thought we decided!)
I. Miriam will make us a printout of the B.Index in the form of the
examples on the handout she gave us. This should be easier for us
to read (and thus understand).
II. Currently, series statements are only searchable with "series
title" searches (indicated by P in the "dcode" column/value of the B.Index).
We want series to also be searchable with "title" searches (indicated
by T in "dcode"). In order to get this, we need to request the change
for the 2nd GMA. (I have added this to our 2nd GMA request
clipboard action module.) (And of course this is something we'll have
to test after the 2nd GMA...)
III. Problem: Series volume numbers are not indexed. (This
means subfield v of the 400, 410, 411, 440, 800, 810, 811, 830 fields.
From now on I'll refer to all of these as "440/v" .) We want them
indexed so a user with a citation for a specific volume of a series
can search for it by that specific volume number.
Michel Ridgeway says that if we indexed subfield v along with subfield
a, etc. that provisional authority records would then be created for
each individual volume in a series. This would clearly be
unacceptable.
At BOG, Michel then told us the way we can do this search now is to
search by the series first, and then modify the search by using the
Change limits option and choosing "with series volume."
Problems with "modify search" option:
1. It's an extra step the user has to take, and
clumsy to teach (although a detailed help screen for search choice "search
for particular volume in a series" is an option Miriam suggested).
2. It currently doesn't work properly. If you modify
a search to get volume 34, for example, you get a record for v. 34,
but you also get records for volumes 534, 634, 347, etc.)
[I need to write an error report for this. I haven't done it
as of 3/15/95--Sam]
Our possible suggestions, solutions:
1. Index 440/v in WORDS, as a keyword search ("files"
column in B.Index). Problem with this: we'd still have to explain to
the user that in order to search for a specific volume of a series,
would have to do something like:
sew = lecture notes and w = 489
2. Index 440/v as T, in title index (value T in "dcodes"),
following the model of 240 in the B.Index. In B.Index, 240 is indexed
3 different ways: 1. as a name/title authority string using the
special JOIN 240 program found in "prep" column of 100, 110;
2. as a 130 authority heading, subject to
authority control on its own;
3. as a title (dcode = T), not subject to authority
control.
We suggested indexing series in 2 different ways, following
the 240 model: 1. as it is set up now, as 130 in authority file,
subject to authority control (dcode M), excluding subfield v [with
addition of T to dcode--see II above].
2. as a title only, including subfield v, not subject
to authority control (so with dcode being only TP but not M, the way
240 is only T in the 3rd way it's indexed in B.Index). This would
render possible this single search: tw = lecture notes 489
Decision: Miriam Blake will work with Grant to effect one or both of
these changes (which one first?) in B.Index so we can test them,
including studying their potential impact on regular title searching,
etc.
We need to come up with some test records which contain
440/v, which we will then have to refile after the B.Index has been
changed so that they'll be re-indexed the new way.
It is my understanding that these changes will be made so
we can test them BEFORE the 2nd GMA. Is that right?
Next part: questions! questions! questions!
--file under ASUB:
Sam
