On Thu, 18 Aug 2005, George Brock-Nannestad wrote:
> It is reasonable to use the term "catalogue" to refer to a description of a
> particular collection (and let us just discuss shellacs - they give plenty of
> problems). Preferably the description should be to a level that when
> comparing a disc held at a different location, it would not be necessary to
> see the actual object held by the particular collection.
As you further point out...takes, discs made from different masters...etc.
and, if I dare to mention it, completely different takes on early
Also you mention variance in groove pitch, which I have frequently
encountered, and the difference it will make in the sound quality...
> Entering all this information into a catalogue or discography is strictly an
> afficionado task, however the community at large would be helped tremendously
I also think about the thickness of the disc, especially in the case of a
One could also address the somewhat subjective notion of wear...
I think of the Institute for Jazz Studies and their remarkable work at
indentifying players on the various recording sessions, information not
taken from the disc, but found in external sources. This would seem to me,
to be critical since in jazz, each musician has a substantive role,
through improvisation and through their own highly subjective
interpretation...for me, more so than say a performance of a string
quartet by Beethoven. However, I also find it curious that very few MARC
records include the names of the members of a quartet, even if their
personnel have changed over the years...for example, the Juilliard
Quartet. Also, should not individual players be cited in the performances
of rock and roll?
I believe that the recent BMG release of the Saint-Saens Third with Munch
and the Boston Symphony is the first time one could hear the three
separate channels, yet this is not reflected on the OCLC record.
Clearly there would be a potential for spending several days on a disc.
For me, I wonder if the performance is the primary concern and its
manifestation (the object) secondary? Also, as we move
towards files being stored and then refreshed, migrated, etc. we aren't
talking about a manifestation (object).
Should the "metadata" (term that for me in its current usage, is
a bit of a misnomer) for files created from objects, include as much detail
as possible as to the source object, as well as the description of its
It seems to me that to fully address all of these concerns is not
practical, or possible from a fiscal perspective.
It makes me wonder if one could make a list, in priority order, all of the
possible elements that could be used in cataloging a sound recording, or
is this what the project you (George) described is attempting to do?