RE: Correction Re: The case against URNs

> -----Original Message-----
> From: Tim Bray [mailto:tbray@textuality.com]
> Sent: Sunday, October 06, 2002 11:40 PM
> To: Champion, Mike
> Cc: www-tag@w3.org
> Subject: Re: Correction Re: The case against URNs

 
> the Web already contains plenty 
> of physical  robots and cameras and so on addressible with HTTP URIs whose

> state can  be manipulated with PUT & POST & so on, and when you do a GET
you're 
> getting a representation which might well be a document but 
> the resource  just plain isn't.

Good point!  Thanks for expanding my world view.

So I concede the point about "documents" but am not sure I understand how
this relates to the original query from Micah Dubinko:

"From the viewpoint of a web developer, it makes sense to differentiate
between network-accessible and non-network-accessible resources. A trivial
transform ('now:' -> 'http:') can provide additional details on the abstract
thing-that-means-whatever-the-DNS-owner-defines-it-to-be."

So, the TAG is saying that there is no reason to make the distinction
between network-acessible resources and non-network accessible resources,
because the REST resource/representation distinction covers all practical
use cases?  And the "car in the driveway vs network-referenceable
information about the car" discussion that went on forever was all a big
misunderstanding?  I'm afraid I get more and more confused.  Perhaps the
next draft of the Web Architecture document will clarify it.

Or perhaps we're all agreeing that the "car problem" is something that the
Semantic Web people will just have to figure out for themselves with
metadata about the resources and representations rather than baking their
worldview into the definition of a URI... and that the whole thing didn't
have much practical significance for the Web as it currently exists?
Likewise the namespace URI flamefest will be resolved with son-of-RDDL
metadata retrieveable via the HTTP namespace URI?  Sounds like the
time-honored tradition of solving computer science problems by adding yet
another layer of meta-ness has been followed :-)

Received on Monday, 7 October 2002 00:36:22 UTC