RE: Proposed issue: site metadata hook (slight variation)

> -----Original Message-----
> From: ext Jeffrey Winter [mailto:JeffreyWinter@crd.com]
> Sent: 12 February, 2003 17:59
> To: Stickler Patrick (NMP/Tampere); miles@milessabin.com; 
> www-tag@w3.org
> Subject: RE: Proposed issue: site metadata hook (slight variation)
> 
> 
> 
> > > > IFF you want to posit another URI to denote that body of 
> > knowledge,
> > > > you should be able to, but the architecture should not 
> demand it.
> > > 
> > > Well, I think we've reached an impasse here.  I think the 
> > architecture
> > > should absolutely demand it.  
> > 
> > Then I guess we're going to have to figure out a way to convert
> > all the blank nodes in every RDF graph to URIrefs...
> 
> I was speaking specifically about the "document" that would 
> be returned
> when doing an MGET (or GET+Meta:); that document should have 
> a distinct
> URI.

Fair enough. But I disagree. It *might* have a distinct URI. But
it shouldn't *have* to have one. No more so that a given variant
representation returned by GET *has* to have one.

Patrick