Although I liked your original idea, after the discussion I prefer your
first option below with a very broad default context set that implies
dc, bath and whatever we add. This would also imply Ray's utility.
The idea of different scopes is in my view not a way to save on the
number of indexes and I wouldn't mind to give different things a
different name and have an for example a separate identifier index and
record_id index.
Ralph asked:
How is dc.creator any/cql.scope=metadata "sanderson"' any better than
metadata.creator any "sanderson"?
I would ask why is metadata.creator better than metadatacreator ?
However the concept of scope remains interesting with respect to FRBR.
Theo
>>> [log in to unmask] 31-8-04 14:16:57 >>>
On Mon, 30 Aug 2004, Ray Denenberg, Library of Congress wrote:
> I believe SRW needs a utility index set.
> An OAI record includes an identifer for the resource it describes.
It's
> proper to search that identifier as dc.identifier. The oai record
itself has
> an identifier. I don't think it's proper to cast that as a dc
identifier,
> scoped or not.
Then we'll have two identifier (etc) indexes?
There seems to me to be several options:
* We could abandon the dublin core context set as being too
restrictive
with regards to semantics and not iso11179 compliant. Then we create a
new
context set with everything from DC, 11179ified, but because we control
it
we can add in new fields, or scope things as we want (metadata/frbr)
* We keep DC but expand its description to be obvious that it's the
simple dc schema but that all we're using is the base semantics of the
field (eg title, creator, date) and that we reserve the right to do
un-dc
like things with them (eg scope to record metadata rather than record)
* Status quo. The status quo actually isn't -that- bad in practice.
While you can have multiple metadata scopes and multiple FRBR scopes,
in
practice these are fudged together.
And while we might end up with lots of indexes here and there, it's
what
everyone is used to and the DC semantics are fairly pervasive. If DC
can't do it, we add a new context set that can.
Rob
,'/:. Dr Robert Sanderson ([log in to unmask])
,'-/::::. http://www.o-r-g.org/~azaroth/
,'--/::(@)::. Special Collections and Archives, extension 3142
,'---/::::::::::. University of Liverpool
____/:::::::::::::. L5R Shop: http://www.cardsnotwords.com/
I L L U M I N A T I
