table metadata and the registry

[Tony Linde]

Hi Ray,

I was out today - that's why it has been so quiet :)  Only, joking: I agree
it is winding down.

Just one point I wanted to clarify: I sensed you were proposing the
table/column retrieval method as 'extra' to the getCapability method. If
not, no problem, discussion over. If so, can you explain why we'd need both?

Cheers,
Tony.

> -----Original Message-----
> From: owner-registry at eso.org [mailto:owner-registry at eso.org] On Behalf
> Of Ray Plante
> Sent: 08 May 2007 22:58
> To: 'IVOA Registry WG'
> Subject: RE: table metadata and the registry
> 
> Hey Tony,
> 
> On Tue, 8 May 2007, Tony Linde wrote:
> > Hi Ray,
> >
> > Looks like we're converging. (See also response to Doug re VOSI.)
> >
> >> Do you agree that we should try to find a way to do this?
> >
> > I assumed that was what was being done. That was my understanding
> from
> > Victoria: that resources not stored with detailed metadata would be
> 'fixed'
> > by calling the individual services.
> 
> Well, I'm 90% sure that's not how it is being done for Vizier
> resources,
> but I'm not sure what the deal is for specific AstroGrid services.
> 
> The problem with going to the individual services is that for our
> current
> std. catalog services, the method for getting this info is different
> for
> each service.  Thus, there is strong incentive to get this information
> into the VOResource record up front, prior to harvesting.  If there
> were a
> common way to retrieve this information from a resource (service or
> otherwise) in a common format, it would not be necessary to pre-load it
> into the VOResource record.
> 
> That is what my URL provides: the ability to 'fix' your metadata
> without
> passing on that fix to us.
> 
> >> begin deprecating the explicit inclusion of table metadata.  There
> is
> >
> > I don't *think* we can do that. Unless we get feedback that every
> existing
> > and possible service can provide its own detailed metadata, I'm not
> sure we
> > can rule out people entering the detailed metadata directly into the
> > registry and leaving it there forever more.
> 
> That's fine.  We have to provide backward compatibility.  If we can
> convert most of the existing services to this alternate approach and
> discourage direct inclusion of table metadata in the future, we would
> be a
> lot happier.
> 
> I sense this thread winding down.  Let's see what comes out of Beijing.
> I'll try to keep you and Kevin informed by email.
> 
> cheers,
> Ray