W3C home > Mailing lists > Public > www-tag@w3.org > February 2003

RE: Proposed issue: site metadata hook

From: <Patrick.Stickler@nokia.com>
Date: Tue, 11 Feb 2003 18:13:55 +0200
Message-ID: <A03E60B17132A84F9B4BB5EEDE57957B01B90B13@trebe006.europe.nokia.com>
To: <www-tag@w3.org>


> ----- Original Message -----
> From: "Jeffrey Winter" <JeffreyWinter@crd.com>
> To: "Tim Berners-Lee" <timbl@w3.org>; <www-tag@w3.org>
> Cc: <tag@w3.org>
> Sent: Tuesday, February 11, 2003 8:39 AM
> Subject: RE: Proposed issue: site metadata hook
> 
> 
> >
> >
> > Why limit this approach to just site-level
> > metadata?  Shouldn't a similar approach be
> > adopted to bind metadata to any resource
> > under the control of the "publisher"?

This is the gist of MGET.

The same authority that names a resource and provides
representations for it, also provides descriptions of
it, all by the very same URI.

No matter what the resource.

And descriptions can be provided even for resources
for which no representations are available (and visa
versa of course, per the present state of the GET only
web...)

> > I can see how this would benefit an RPC-style
> > gateway as a means of (for example) standardizing
> > how to obtain a WSDL document, but what about
> > REST-style applications where each resource
> > may (and probably will) have its own specific
> > metadata?

Exactly.

Cheers,

Patrick
Received on Tuesday, 11 February 2003 11:14:03 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:16 GMT