Terry Allen argues in favour of distinguishing publishers
from "other agent"s :
|... I think that Publisher merits its own field not only because
|it is sensibly different from Editor, Illustrator, and other
|things likely to fall into OtherAgent (the language of the June 1995
|DC report,
| http://www.oclc.org:5046/oclc/research/conferences/
| metadata/dublin_core_report.html
|makes a very clear distinction between intellectual contribution and
|distribution) but also because it is a traditional piece of publication
|metadata that is actually used (whether or not it is indexed) and is
|ikely to be useful on the Internet.
I think he is quite right to say that there are (at least) two ways in
which responsibilities may be assigned to a published resource: there is the
responsibility for its intellectual content, and there is the responsibility
for making it available/publishing it. (Sometimes they are the same agency
but that is irrelevant). For purposes of resource discovery, it is
convenient to prioritize the various responsibilities within each of these
two major divisions: we might have "principal author" and "original publisher"
for book-y things. But I am more persuaded than I was that both should be
present in the core.
In TEI terms, the DC.author corresponds to whichever of the various
responsibility elements (<author>, <editor>, <respStmt> etc) appears
within the <fileStmt>; and the new proposed DC.publ corresponds to
whichever of the various responsibility elements appears within the
<publicationStmt>.
A decisive factor for me is the possible encouragement this should give people
to admit as to who is responsible for formally distributing something
on the web!
My position is now that non-principal statements of responsibility,
non-principal statements of distribution should *not* appear in the core.
What is in there should be lean, mean, and precise.
Lou
|