Hi,
Stu has stated that he does *NOT* believe having two elements, one
called subject and the other called description, represents the
consensus of this list. Determining consensus on lists is pretty
tricky, but let me take the opposing position and offer reasons to
support it.
Most importantly, we have had a number of people express their
support for two such elements on this list. Other than Stu's
message, I am unaware of any messages expressing dissent with
this requested change. Since Stu will soon be leaving town and
has stated that the document will not be changed once he does
so, I urge people who feel strongly about this issue to send
their opinion to the list.
Stu expressed concern that havingtwo elements was not defensable.
Let me outline a possible course of justification here.
I think it is fairly easy to make naive users understand the
difference between "description" and "keywords". Descriptions
are pieces of prose on the order of a paragraph in length, and
"keywords" are a word or three describing major concepts addressed
in the work or setting its context. I think that the pitch to
make is that "subject" is a generalization of the "keywords"
concept that allows uncontrolled keywords or, through the
use of the scheme qualifier, controlled terms.
While there is some similarity of intent between "description"
and "subject", they are sufficiently important to maintain
the distinction. This is not the only case where we distinguish
between broadly similar concepts. In particular, we could have
made "Creator", "Publisher", and "Contributor" all fit into a
"ResponsibleParty" element with scheme qualifiers. We did not
do so because we believe that the concepts are distinct enough
and important enough that the distinction should be made at the
highest level possible.
There are a couple of ancillary justifications as well. First,
"subject" matches the intuition of users raised on the use of
tripartite card catalogs - Author, Title, Subject. Second, use
of the "description" element matches current use of a few
important web indexing services, with the immediate payoff
of making simple metadata useful to a wide audience.
I'm at the IETF this week so I can't guarantee to make prompt
responses to messages. However, I reiterate my statement from
last week: If I have to spend all my capital on one change to
Stu's current document, this is it.
Regards,
Ron Daniel Jr. email: [log in to unmask]
Advanced Computing Lab voice: (505) 665-0597
MS B-287 TA-3 Bldg. 2011 fax: (505) 665-4939
Los Alamos National Lab http://www.acl.lanl.gov/~rdaniel/
Los Alamos, NM, 87545 obscure_term: "hypernym"
|