W3C home > Mailing lists > Public > www-rdf-interest@w3.org > May 2003

RE: Answers to URIQA Questions + Appologies

From: <Patrick.Stickler@nokia.com>
Date: Fri, 23 May 2003 10:16:17 +0300
Message-ID: <A03E60B17132A84F9B4BB5EEDE57957B01B90E39@trebe006.europe.nokia.com>
To: <drepchick@intellidimension.com>, <www-rdf-interest@w3.org>



> -----Original Message-----
> From: ext Derrish Repchick [mailto:drepchick@intellidimension.com]
> Sent: 23 May, 2003 01:51
> To: www-rdf-interest@w3.org
> Subject: Answers to URIQA Questions + Appologies
> 
> 
> 
> I'm sorry about the multiple postings of my original 
> questions. I was having
> a bad email day.
> 
> >
> > Q1) Will proxy servers present problems w.r.t. the URI in the HTTP
> request?
> >
> > GET http:://example.com/document.html ----> GET /document.html
> > URI-Resolution-Mode: Description
> 
> After looking over the HTTP1.1 spec I think URIQA is fine with proxy
> servers. It looks like the proxy will turn the request URI 
> into an absolute
> path but host information is preserved in the header field 
> Host (ex: Host:
> example.com).
> 
> GET /document.html
> Host: example.com
> 
> So the original URI can be reconstructed by the server.

Right. This is what the URIQA servlet does.

> >
> > Q2) How are equivalent URLs dealt with?
> >
> > GET http:://example.com/document.html
> > URI-Resolution-Mode: Description
> >
> > GET http:://127.0.0.1/document.html
> > URI-Resolution-Mode: Description
> >
> > GET http:://example.com:80/document.html
> > URI-Resolution-Mode: Description
> >
> > GET http:://example.net/document.html
> > URI-Resolution-Mode: Description
> >
> > Should the server map these requests onto the same resource?
> >
> 
> After thinking about this some more it seems to me from a 
> trust standpoint
> its important that the subject of the statements returned to 
> the agent have
> exactly the same URI that was used in the request. So if an 
> agent requests
> information about http://example.net/document.html then the 
> server must
> respond with statements about that URI and not some internal 
> URI used to
> store the information ( http://example.com/document.html). So 
> if internally
> the server performs some mapping before querying for the 
> information then it
> must reverse the mapping in the response.

I agree that it is better to err on the side of caution and
uphold any distinctions between URIs, even if they ultimately
resolve to the same set of representations. Just because they
resolve to the same set of representations does not necessarily
mean that they denote the same thing.

Patrick
Received on Friday, 23 May 2003 03:16:38 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:59 GMT