From: OXVAXD::LOU "Lou Burnard" 29-NOV-1996 15:19:49.87
To: MX%"[log in to unmask]"
CC: LOU
Subj: RE: RENAME "IDENTIFIER" ELEMENT
I think dropping IDENTIFER is a really stupid idea. And I don't see what's so
difficult with the name.
Its function is to provide one or more of the possibly many identifiers used
for the object to which it is attached. This might be a URL, a URN, an FPI,
an ISBN, or any number of other things. Which particularly domain this
identifier comes from is specified by the SCHEME attribute, as usual.
It corresponds rather neatly with the <idno> element defined in the TEI
header, but that's the not the main reason for wanting to retain it. IT'S
USEFUL! I can't see any good reason to drop it.
Likewise, I cannot understand what's the problem understanding the difference
between TYPE and SCHEME. TYPE is for subcategorization -- it's a free
slot for people who want to distinguish amongst categories identified by
the DC element itself. SCHEME is for definining the taxonomy within which
the (sub)categorizations are defined. For example, you might want to
use the UDC to define TOPIC (or whatever the thing is called now) or you might
want to use LC subject headings or you might want to use Uncle Bernies Patent
Subject Classification Scheme (UBPSCS). Or even all three. A browser that
can handle any one or more will be very grateful for the information as to
which of them you're actually using. That's what SCHEME is for.
I second (warmly) Jon and Paul's desire for a bit less boat rocking and bit
more clarity as to what the consensus is. Would someone like to specify
here which documents (exactly) we should consider as embodying the consensus?
I know of several ...
Lou
|