----- Original Message -----
From: "Green Richard" <[log in to unmask]>
To: <[log in to unmask]>
> This is an old (never-ending) debate - quick and (sometimes) dirty versus
> complete and detailed.  Even if you make the cataloguing as complete as
> possible, as we do on the Virtual Gramophone site
> people will
> - there is too much detail or we didn't check all the sources for whatever
> specific detail they are interested in, be it subject headings, session
> information, or whatever.  (Those libraries that want to do copy
> from the Virtual Gramophone can now do so through the LACs main catalogue
> AMICUS - this is a step forward.)
> My experience tells me that whatever depth of cataloguing you go for it
> needs to be consistent. Which means that record collectors do not always
> make the best cataloguers - as consistency is not always a strong point.
> You need good instructions, good training, and a patient person with a
> meticulous streak in them to spot errors and inconsistencies.  Being a
> little retentive helps.
> The super catalogue  is a dream.  Yes there is lots of data out there in
> various computer formats, but there are so many variables in that data,
> nothing is really standardized, that bringing it altogether would result
> another series of Rigler Deutsch type complaints.  It should be noted that
> Rigler Deutsch does what it was intended to do - provide inexpensive
> to large collections that were previously inaccessible at any level.
> NAXOS itself has had a debate about cataloguing levels and is dealing with
> complaints about errors, inconsistencies and omissions.   They are even
> hiring librarians.  The cycle begins again.
Basically, there are at least two completely different rationales behind
cataloguing...and this means there are two different forms of the beast!

The first is cataloguing simply to provide a record of the phonorecords
held in a particular collection (or, as in RDI, a group of collections.
At a minimum, this requires some data on each phonorecord (artist credit(s),
title credit(s), label, country, catalog number) as well as the location
where the phonorecord is stored if this is not obvious from the label/
number data.

The second is cataloguing to provide a discographic data archive. Note
that a "type I" catalog can be maintained including all, or as much as
possible, of the applicable discographic data for each phonorecord in
the collection(s) being catalogued.

Insofar as collection catalogs include discographic data, they become
more useful to researchers accessing the data for purposes rather
than accessing the listed phonorecords. The best example is (are?)
the Abrams files; these provide only 160 bytes of data (including
spaces to fill out fields) per phonorecord side, and probably
include about 5% of the existing phonorecords (with a fair amount
of inaccuracy, it might be noted)...but they are consistently
used, and cited on 78-L, to answer questions about phonorecords!

Since I am both collector and discographer, my collection catalog
includes considerable discographic data. I created it using MS
Access and a 3 level relational database. The problem is that
data entry is a long and tedious procedure, requiring 10-15 minutes
per catalogued phonorecord; so far, I have less than 200 phonorecords
catalogued, and have serious doubts as to whether I will ever be
finished with the task! However, if I ever do it will be a valuable
data source for posterity.

As far as the usefulness and/or use of a fairly complete
discographic data archive...note how many times Brian Rust's
discographies are used!

Steven C. Barr