> > http://www.loc.gov/z3950/agency/zing/cql1-1/context-sets.html

> You mean the bit that says "The prefix of 'cql' is reserved for the
> default CQL context set, but authors may wish to recommend a short
> name for use with their set."  If I'd spotted that before, I'd have
> called it.  I don't see any reason to special-case this.  If I want to
> map the "cql" prefix to my private set and have the meanings of all
> the relation modifiers change, that's surely my decision.

Right.  Hence my following comment that we should make it clear that
_unless otherwise assigned_ CQL is the CQL set, and shouldn't generate an
unknown context set diagnostic.


> > And by extension, making it also mandatory to declare the DC schema
> > if supported, a reversal of what we said earlier.
> >   <schema name="dc" .../>
> This is a separate discussion, right?  Or have I missed a connection?

Yes, a separate discussion.  The connection being that if you have to
explicitly say that you support the CQL set, then to be consistent, we
should explicitly say that we support the DC schema, even though support
for it is mandatory.
Especially as it makes generating interfaces from the ZeeRex more
consistent (do I add a DC schema or is it already listed check)

On the other hand, we shouldn't have the Diagnostic schema or ZeeRex
listed, as those are required for protocol reasons, rather than
minimum level interoperability reasons.  Right?


> > The context set will live at the LC site, where its identifier
> > is. And (as yesterday) the srw.o-r-g.org site will still exist as
> > there's no reason to get rid of the host name, but I'd prefer
> > srw.cheshire3.org. The Identifier will definitely change.

> Why not put the RM set on the LC site, too?  It is very much part of
> the core group of context-sets, along with CQL, the Network set ...
> especially true after the OCLC debacle where the URL of the
> Cross-Domain Attribute Set changed.

(Yes! That was ... troublesome ... though debacle is probably too strong)

Last time I offered, we decided to just leave it with me. But I'm
still quite happy to have Net and Rec live on the LC site, with a loc.gov
identifier.

A 'credits' page might then attribute them (and/or the rest of the
documentation writing) back to me, but I'm not concerned.  (The University
on the other hand likes publications, which was my only reason for not
insisting that they be core sets last time round, I admit)

Rob

--
      ,'/:.          Dr Robert Sanderson ([log in to unmask])
    ,'-/::::.        http://www.o-r-g.org/~azaroth/
  ,'--/::(@)::.      Special Collections and Archives, extension 3142
,'---/::::::::::.    Nebmedes:  http://nebmedes.o-r-g.org:8000/
____/:::::::::::::.
I L L U M I N A T I