In article <[log in to unmask]>, Barbara Palmer [BP]
<[log in to unmask]> writes
>Some people may not know that there is a difference between subject
>indexing, which is used in libraries, and object name indexing, which can
>also be used in museums.  Some indexing lists used by small museums may be
>combinations of the two.
>
and then in article <[log in to unmask]>,
Orycteropus afer [O a] <[log in to unmask]> writes
>You make an excellent point about differentiating between subject
>classification and object naming.  This gets confused all of the time and
>one sees museums trying to apply the Library of Congress Subject Headings
>inappropriately.  Further, there has been very little discussion about
>terminology controls in manual systems versus terminology controls in
>automated systems.

I think that we have to be careful here not to mix two different
problems:

1. The difference between three related but distinct approaches:

        (a) A hierarchically-structured thesaurus of indexing terms,
        where each term represents a single concept and where terms can
        be combined at search time (post-coordinate indexing). This is
        most appropriate for use in an automated search system.

        (b) A list of subject headings, each of which may include
        several concepts combined together at the time of indexing (pre-
        coordinate indexing). This is most appropriate for displaying
        lists which will be browsed or scanned visually, either on
        screen, on paper, or as headings for files and index cards.

        (c) A subject classification, which is a way of grouping
        compound subjects so as to bring related material together. The
        most useful grouping is usually not alphabetical, so a symbolic
        notation is generally used to make it easy to maintain items in
        the agreed order.

2. The difference between indexing objects and subjects. These are just
two facets, which may occur in all three types of system noted in 1
above. Other facets which may also be needed are places, materials,
events, activities, and so on. "Subject" is rather ambiguous, as it may
be taken to refer to all of these or just to names of disciplines or
areas of study or activity.


The mechanism for assigning consistent indexing terms to a record is the
same whether the record is for an object, a person, a place or anything
else. There should be a thesaurus or authority file which lists all the
variant forms by which a concept may be sought, and specifies one of
these forms as a "preferred term" which will be used to represent that
concept for indexing purposes. Each of the concepts by which the record
needs to be retrievable should first be identified and then a link made
between that record and the preferred term for each of these concepts in
the thesaurus.

When we link an object record to an object name category such as "axes"
or "woodworking tools", or a subject category such as "forestry" or
"joinery" we are not necessarily saying that that these are precise,
specific or exclusive terms for the object; we are saying that someone
seeking objects which fall into those categories should look at this
record. Use of a thesaurus of object names is no different in this
respect from use of any other type of thesaurus.

[BP:]
>At the Powerhouse Museum, Sydney, we are using an object name thesaurus,
>which we have published, and we are currently developing a subject
>thesaurus.

I have seen the object name thesaurus, and it looks good and well-
structured. I hope that the subject thesaurus will be closely integrated
with it, so that they can be combined, and so that we can have links
between them such as "shoes RT shoemaking".

[BP:]
>Museums Australia's Victorian branch have published "The Small
>Museums Cataloguing Manual" which includes a simple classification of
>historical objects.

This sounds interesting. Can you give us bibliographic details and
availability information, please?

[O a:]
>I think within the structure of an ANSI (AMerican National Standards
>Institute) thesaurus, you can get pretty close to a universal terminology.

If everyone compiling a thesaurus sticks to the principles set out in
standards such as this (ANSI/NISO Z39.19-1993; BS 5723:1987; ISO
2788-1986) it will certainly be much easier to combine thesauri compiled
by specialists in different subjects.

Leonard Will

--
Willpower Information       (Partners: Dr Leonard D Will, Sheena E Will)
Information Management Consultants                 Tel: +44 181 372 0092
27 Calshot Way, Enfield, Middlesex, EN2 7BQ, UK    Fax: +44 181 372 0094
[log in to unmask]           [log in to unmask]
------------------ http://www.willpower.demon.co.uk/ -------------------