On Mon, 2003-07-21 at 11:56, Bruce D'Arcus wrote:
> Followup observations/questions:
> (not sure if a conference goes
> as a title or a name, particularly when you consider they are often
> numbered (e.g. - "23rd Meeting of Whatever").

That depends on your cataloging rules. Library rules say that the
general conference name is the author (it's a special kind of author and
there is specific coding for conferences as authors), and the title of
that specific conference (usually it's "theme") is the title. (I think
of this as the difference between the conference "body" and the
conference publication.) Since conference publications are often
organized by conference volunteers, and therefor may lack full
information on the publication itself, there is a NISO standard for
title pages for conference proceedings that guides one to provide all of
the necessary information.

Here are some conference authors and titles from a library catalog:

International Conference on Medical Image Computing and
Computer-Assisted Intervention (5th : 2002 : Tokyo, Japan)

Medical image computing and computer-assisted intervention--MICCAI 2002
: 5th International Conference, Tokyo, Japan, September 25-28, 2002 :
proceedings / Takeyoshi Dohi, Ron Kikinis, (eds)

SIGGRAPH (Conference) (12th : 1985 : San Francisco, Calif.)

SIGGRAPH '85 conference proceedings, July 22-26, 1985, San Francisco,
California / edited by Brian A. Barsky ; sponsored by the Association
for Computing Machinery's Special Interest Group on Computer Graphics

> 4) This is a little extreme, but I've run into it:
> Does MODS have all the facilities to deal with an unpublished letter,
> with an unknown author, with an only vaguely known ("cerca") date?  I
> remember the issue of the last came up awhile back, but does it make
> sense to distinguish between "anonymous" authors and unknown authors,
> either in cataloguing or in MODS?

Again, that's a cataloging rules question. There is a difference between
an item that has no author (usually works of collective authorship) and
one that is written by an individual who chose to remain anonymous. If
your discipline cares about such things, and that would be the case in
archives of unpublished letters, then you will need a rule that
determines how you handle those items. Archivists use different
cataloging rules from libraries, and I'm not familiar with them. I have
seen records, though, where the cataloger made a best guess as to the
author and that was made clear through the coding, i.e. [Smith, John?].
Also, in texts from a few centuries ago, the author may actually be
listed on the title page as "Anonymous" -- in which case that becomes
the author in the library record since it is actually printed as such.

The real problem with anonymous is that it's not always the same
anonymous, and therefore it's not very good information. There is
actually an attempt to distinguish between the anonymous authors since
we can generally know that an author writing in 1823 as anonymous is not
the same one writing in 1990. So there are a lot of listings for
anonymous in a library catalog like that of the Library of Congress, and
they aren't all the same:

Anonymous, 1900-1973.
Anonymous, Brother, 1927- -
   See:  Desbiens, Jean-Paul, 1927-
Anonymous Colombian Woman
Anonymous conqueror -
Anonymous II -
   See:  Anonymous II, 13th cent.
Anonymous II, 13th cent. -
Anonymous investor.
Anonymous IV.
Anonymous IV, 13th cent.