On Tue, 2004-09-07 at 10:22, Bruce D'Arcus wrote:

> One problem is that it then introduces an inconsistency between
> mods:name and mods:titleInfo.

But it's not an inconsistency when using the library cataloging rules,
which allow multiple forms of a title in a bibliographic record, but
only one form of each author's name.

What I think you're saying is that you want to apply a different set of
rules for author. In this case, MODS clearly has a bias toward the
library rules (which it inherits from MARC). Under those rules it isn't
necessary to distinguish between multiple authors and multiple forms of
an author's name (except with the transliteration case).

 If one allows more than one titleInfo
> element, I see no logical reason why not to allow more than one name.

Right. It's not a question of logic. It's a question of how far MODS
should deviate from the MARC record. This is a question that has puzzled
me for a while and not only do I not have an answer, I'm not sure what
criteria one would use to come up with an answer. I believe that some
people expect to be able to go from MARC to MODS and MODS to MARC. In
that case, adding multiple forms of an author's name complicates the
MODS to MARC unless there is specific coding that would identify what
librarians call the "authoritative name." In fact, the entire issue of
MODS to MARC hinges on the content of the fields and the cataloging
rules that were used to create the original metadata. My gut feeling is
that unless you've used library cataloging rules to create the original
bibliographic data, the odds of being able to recreate a MARC field from
any non-MARC metadata is very low. If we could abandon all hope of a
MODS to MARC mapping, then the MODS record could probably gain a lot of
flexibility. Even so, I would want to clearly mark parallel
transliterated/vernacular names as a special case (as opposed to two
forms of the author's name, like T.C. Boyle and T. Coraghessan Boyle).
Karen Coyle
Digital Library Specialist
Ph: 510-540-7596 Fax: 510-848-3913