> Date: Fri, 18 Jun 2004 10:59:15 +0100
> From: Robert Sanderson <[log in to unmask]>
>
> > I meant just in the profile, but I could be wrong.
>
> Hrm. I'll at least make the note in the profile more extensive, but
> as it's really the only way to do a search using index and set, it
> could be used in other profiles, so I think it belongs at least in
> the context set documentation as well.
OK.
> > So don't we also need zrx.uri?
>
> Nope, as cql has a uri relation modifier already.
Oh yes, so it does.
http://www.loc.gov/z3950/agency/zing/cql/context-sets/cql.html
> Of course that should probably be noted as useful in the profile.
Yup.
> > Also, don't these two relation modifiers (really structure
> > attributes, of course) belong in a more generic set than ZeeRex?
>
> URI definitely belongs where it is, but where would OID go?
Same place. The core set
> I guess that it could go into the Network set.
I don't think so. Z39.50 happens to be a network protocol that uses
OIDs, but there is nothing essentially networkish about them, any more
than there is with (generalised) URIs.
> Or we could rename ZeeRex to something more schema neutral.
Nope, its name is right.
> But I don't think it's of broad enough applicability to go into CQL.
> Apart from Z39.50, where are oids used nowadays?
Who can say? Any number of ANSI/NISO standards might use OIDs for all
sorts of reason we know nothing about. If people want to build CQL
profiles to do with searching for those objects, it doesn't seem
unreasonable to me that they should find the structure right there in
the core set, along with URI which is precisely analogous.
_/|_ _______________________________________________________________
/o ) \/ Mike Taylor <[log in to unmask]> http://www.miketaylor.org.uk
)_v__/\ "No time to lose!" -- Monty Python's Flying Circus.
--
Listen to free demos of soundtrack music for film, TV and radio
http://www.pipedreaming.org.uk/soundtrack/
