W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > June 2001

Re: General and specific resources? (every document is in the Web)

From: Graham Klyne <Graham.Klyne@Baltimore.com>
Date: Sat, 16 Jun 2001 20:35:29 +0100
Message-Id: <5.0.2.1.2.20010616202716.03099150@joy.songbird.com>
To: Aaron Swartz <aswartz@upclink.com>
Cc: Dan Connolly <connolly@w3.org>, rdf core <w3c-rdfcore-wg@w3.org>
At 03:35 PM 6/15/01 -0500, Aaron Swartz wrote:
>On Friday, June 15, 2001, at 02:33  PM, Dan Connolly wrote:
>
>>And as Roy Fielding pointed out, RFC2396 itself
>>uses 'resource' only in the stricter sense;
>>he (among others) haven't bought in to the
>>way the RDF spec uses the term.
>
>I think you are misrepresenting the issue. The issue is not that resource 
>is only used in the stricter sense (in fact, the URI RFC makes it clear 
>that resources can be anything "with identity"). The problem is that 
>URI-refs (URIs plus fragments) are not defined to represent or relate to 
>resources in any ways.
>
>Roy and others were aware of this problem at the time they drafted the 
>spec -- they did it on purpose. It's a feature, not a bug. The problems is 
>that if you allow fragments to represent resources, you run into all sorts 
>of problems (many of which RDF is already experiencing).
>
>RDF needs to reconcile itself with web architecture somehow, and I feel 
>that this is a serious issue for the Working Group to address.
>
>I wrote up a summary of the issue at:
>http://lists.w3.org/Archives/Public/www-rdf-interest/2001May/0137.html

When I raised this issue in the RDFIG session of the W3C tech plenary 
earlier this year, the response (as I understood it) that TimBL gave at the 
time was that the sets of Web resources and RDF resources are not 
identical.  My reading is that Web resources (i.e. those identified by URIs 
-- without fragments) are a subset of RDF resources (roughly, anything 
denoted by a URI+optional fragment ID).

It seems to me, then, that an RDF resource identified by UTI+fragment must 
be distinct from a view of a Web resource similarly identified, because the 
latter must depend on the MIME type of a retrieved entity.  (Though in some 
cases there may be a strong correlation between these.)

#g


------------------------------------------------------------
Graham Klyne                    Baltimore Technologies
Strategic Research              Content Security Group
<Graham.Klyne@Baltimore.com>    <http://www.mimesweeper.com>
                                 <http://www.baltimore.com>
------------------------------------------------------------
Received on Saturday, 16 June 2001 16:19:49 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:37:10 EDT