W3C home > Mailing lists > Public > www-rdf-comments@w3.org > January to March 2004

Re: Media types and assertions

From: Graham Klyne <GK@ninebynine.org>
Date: Thu, 11 Mar 2004 09:22:57 +0000
Message-Id: <5.1.0.14.2.20040311091331.0259c1c0@127.0.0.1>
To: Mark Baker <distobj@acm.org>, www-rdf-comments@w3.org

At 22:47 10/03/04 -0500, Mark Baker wrote:
>I believe that the latest application/rdf+xml registration draft[1] is
>missing some very important information; it fails to declare that the
>use of this media type *asserts* the triples encoded in the document.
>Previous versions of the draft included some discussion about exactly
>this topic[2] (sec. 4), though it didn't come right out and say what was
>asserted, if anything.

[I'm replying in a personal capacity, though I believe this broadly 
reflects the consensus of where we stand.]

We did spend a considerable amount of effort trying (and failing) to come 
up with a form of words that would convey this intent, without getting 
caught up in a range of other difficulties.  In the end, the RDFcore group 
decided to stick with the technical specification and the broader issue was 
passed to the SWCG for further consideration:
   http://www.w3.org/2000/03/rdf-tracking/#rdfms-assertion

Meanwhile, the MIME registration draft:

(1) refers to the published formal semantics document:
[[
        Published specification: see RDF/XML Syntax Grammar [1] and RDF:
        Concepts and Abstract Syntax [2] and the older RDF Model and
        Syntax [4]
]]
-- 
http://www.ietf.org/internet-drafts/draft-swartz-rdfcore-rdfxml-mediatype-04.txt
which provides a technical treatment of assertions in RDF, and

(2) alludes to the broader issues under the security considerations heading:
[[
           Security considerations include many of those described in
           section 10 of RFC 3023 [5] and more, due to the semantic nature
           of RDF.  RDF documents may make assertions about anything and
           it is expected that future work with Digital Signature and "Web
           of Trust" will make it more clear how to build secure RDF
           systems.
]]
-- 
http://www.ietf.org/internet-drafts/draft-swartz-rdfcore-rdfxml-mediatype-04.txt

Under the circumstances, I feel this is as far as the current consensus 
extends, and that we really don't want to delay registration of the MIME 
type any further.

#g
--

At 22:47 10/03/04 -0500, Mark Baker wrote:

>Hi,
>
>I believe that the latest application/rdf+xml registration draft[1] is
>missing some very important information; it fails to declare that the
>use of this media type *asserts* the triples encoded in the document.
>Previous versions of the draft included some discussion about exactly
>this topic[2] (sec. 4), though it didn't come right out and say what was
>asserted, if anything.
>
>There's been a lot of discussion recently about the topic of assertion,
>and AFAICT, most folk seem to be assuming that every RDF/XML document
>asserts triples.  I don't believe that's the case, since I can't find
>any mention of triple assertion in the spec (except perhaps indirectly
>in the definition of parseType="literal"?).  As Dan Connolly correctly,
>IMO, points out[3], an RDF/XML document described as "application/xml"
>(or "text/plain", or ...) is not communicating any triples (let alone
>asserting them).  This also seems consistent with the TAG finding on
>authoritative metadata[4].
>
>The text I'd like to see in there might go like this;
>
>   Use of this media type not only indicates the specification to use
>   to extract "triples" from the document, it also indicates that these
>   triples are asserted.
>
>(any reasonable definitions of "asserted" out there?)
>
>Aside; this also suggests that another RDF/XML media type might be
>needed in order to publish RDF/XML documents without asserting them
>(e.g. test cases).  But that's a whole other discussion. 8-O
>
>  [1] 
> http://www.ietf.org/internet-drafts/draft-swartz-rdfcore-rdfxml-mediatype-04.txt
>  [2] 
> http://www.aaronsw.com/2002/draft-swartz-rdfcore-rdfxml-mediatype-02.html
>  [3] http://lists.w3.org/Archives/Public/www-webont-wg/2002Oct/0162.html
>  [4] http://www.w3.org/2001/tag/doc/mime-respect.html
>
>Mark.
>--
>Mark Baker.   Ottawa, Ontario, CANADA.        http://www.markbaker.ca

------------
Graham Klyne
For email:
http://www.ninebynine.org/#Contact
Received on Thursday, 11 March 2004 04:54:21 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 18 February 2014 13:20:08 UTC