Re: resources and URIs

On 17 Jul 2003 10:13:35 -0500
Dan Connolly <connolly@w3.org> wrote:

> On Thu, 2003-07-17 at 09:40, Williams, Stuart wrote:
> > Indeed... "RDF/XML Syntax Specification Revised" [1] gives number of
> > examples that contain what I think Tim would regard as 'illegal' references.
> > 
> > Figure 1 contains the following URI References as property names:
> > 
> > 	http://www.example.org/terms/editor
> > 	http://www.example.org/terms/homePage
> > 	http://purl.org/dc/elements/1.1/title
> > 	http://www.example.org/terms/fullName

<snip/>

> > Well... I wish I understood the case for it being so wrong... 'illegal' in
> > Tim's words.

I agree with Stuart - I don't see any case for these being wrong.

In particular, Dublin Core is not going to change it's namespace or
URIs, so we have to use that to use the Dublin Core title element.
(Although I think Eric arranged DC to use the server trick of
redirecting from a slash URI to a hash URI)

> > To be frank... I don't understand it... and don't see the need for the
> > constraint... and I know I'm addressing a champion of the 'minimal
> > constraint' principle.
> > 
> > Last time you and I spoke about it you spoke in less black and white terms
> > and about an 'economic' rationale - it is more 'costly' done this way...
> 
> yes.
> 
> >  but
> > I have failed to recreate that discussion... did you write it
> > down anywhere?
> 
> umm... I don't think so.
> 
> Sandro and I and others have written some relevant stuff
> in/near http://esw.w3.org/topic/HashURI
> 
> but I haven't actually figured out how to write down my
> intuitions about economic justification for HashURIs.

I've read the wiki stuff and remain unconvinced that putting # at the
end of namespace URIs will have any technical benefit.  Maybe only
socially if people somehow feel they are "different" URIs.

<snip/>

Dave

Received on Thursday, 24 July 2003 10:50:13 UTC