[Ex-plain] OIDs And Name
mike at tecc.co.uk
Tue Apr 2 18:31:21 CEST 2002
> Date: Sat, 30 Mar 2002 17:29:25 -0500
> From: Ray Denenberg <rden at loc.gov>
> > I want an OID that a client can specify as preferred record syntax
> > when it wants an Explain-- record. Should that just be the XML
> > OID? Maybe - I lost track of that discussion a long time ago. If
> > so, then how can the client specify that it wants the Explain--
> > DTD rather than some other DTD? I think it wouldbe bad if the
> > answer was "you have to use compSpec".
> This is a prime example of why I'm proposing compSpec-2. The record
> syntax is just XML, and yes, that's not sufficient to indicate
> "explain", so the answer right now would be that you have to use
> compspec. What I'm proposing is a minature compspec that might
> include nothing more than a character encoding and schema. So if
> the answer is "you have to use compspec-2" would that be so bad?
Er. I guess not. I suppose I am a bit unhappy about having Explain--
dependent on this sort of cutting-edge stuff, my fear being that it
will be a barrier to implementation. But in practice, you don't
really need a record-syntax OID to search an Explain-- database, since
every record in the IR-Explain---1 database is Explain---compliant XML
anyway. So that's OK then.
Thanks for the attribute-set OID, BTW. Now mentioned at
/o ) \/ Mike Taylor <mike at miketaylor.org.uk> www.miketaylor.org.uk
)_v__/\ "The secret of success is sincerity. Once you can fake that
you've got it made" -- Jean Giraudoux.
More information about the Ex-plain