Re: arsclist Cataloging

[<stevenc@xxxxxxxxxxxxxx>]

Since I'm in the midst of similar discussions both on 78-L and off-list with
several
people, let me make some general comments (which can then be discussed):

1) Basically, there are two separate (related to some extent, but not
identical)
two types of discographic databases, as well as two subtypes of the
second...
a) A discographic information archive. This is a database containing all, or
as
much as possible, information about a given set of records ("set" can
include
every known record). This could consist of a table (or set of related
tables)
which contain fields sufficient to cover every possible aspect of the
phonorecords
in question, leaving it up to the user to select the desired fields
he/she/it needed
to look at. Note that this does not represent, and therefore needs no
one-to-one
relation to, an actual set of individual phonorecords that exists anywhere;
what
is being archived is data, not recordings. Since this data can come from
published
sources as well as actual phonorecords, data records may be incomplete and
the accuracy of the data cannot be guaranteed. This type of archive can
be used to answer questions such as "Did <company> ever issue <song>?"
or "What is on <label><number>?"
b) A catalog of an extant collection. This is a database containing desired
(not necessarily all) data on a specific extant set of phonorecords...namely
the set held by a specific collector or institution, as follows:
b-1) Catalogs for individual collectors. Such a database is not likely to
have more than one user; therefore, it can contain fields, and abbreviations
or terms, of interest to, or comprehensible to, only that individual. As
well,
it will have fields referring only to the specific instance of the
phonorecord
in that collection, such as condition, price paid, date acquired, etc.
b-2) Institutional catalogs. These, like the card catalog in a library, can
and will be used by any or all of the library's users (and this is the
important part) *many of whom will be ignorant in the areas of
database usage or discographic terminology*. For this reason, more
attention must be paid to making this database, particularly its
"front end" (the program the users use to access the data) as simple
and user-friendly as possible. Note that the more general public access
needed for these means they can be, and often are, available on
the internet.

Both of these types answer one basic question: Do (I/the institution)
have a copy of this phonorecord, and if so where is it? There are
other questions which they can answer; some germane to the
specific copy of the phonorecord (condition, cost, label type,
sleeve or jacket appearance, etc.) and some to all examples of
the phonorecord (matrix number, date recorded, personnel, etc.);
note that the latter questions really should be answered via the
archival database, but may be asked of the catalog database
if it is in a more accessible location/medium.

Now, while the above types (a, b-1/b-2) of databases are distinct,
there can be connections. One way to simplify the construction of
either of the catalog-type databases is to copy the data from data
records in an archival database (given the possible inaccuracies and
the differences between copies of phonorecords, entries should be
verified against the actual phonorecords)...the other way, essentially
the opposite, is to compile archival databases by consulting catalog
databases where they are available and accessible. This means that
catalog databases might be better off if they include information
not likely to be used by their users (as noted, archival databases
should be as complete as possible). However, it also suggests a
need for (and this might be ARSC's job) standardization both
among database types and between individual databases, at
least in the areas of field names, sizes (or at least minimum sizes)
and types. Given the usage of various standard systems and
proprietary cataloguing programs, this may be a major problem
(or set of problems!)

What I would suggest is that we consider what we expect out of
our catalogs/catalog databases, as well as archival databases
insofar as those exist (they should, given the availibility of
computers and digital storage, at least in my opinion!), look
at whether what we are using to see how well it meets those
expectations (and why not) and to what extent compromises
are necessary (lack of time/funds/qualified personnel/ambition/etc.),
and then do some discussion on the matter.

Oh, yes...don't look to me as an example! I have 30,000 78's in
about 300 milk crates, and about a fourth of those are neither
in order or catalogued in any form...what NOT to do, eh?
Steven C. Barr

-
For subscription instructions, see the ARSC home page
http://www.arsc-audio.org/arsclist.html
Copyright of individual posting is owned by the author of the posting and
permission to re-transmit or publish a post must be secured
from the author of the post.