Re: need to determine what RDF is

>At 09:15 30/05/2002 -0400, Dan Brickley wrote:
>
>[...]
>
>>I'd like to hear from Eric and Brian w.r.t. whether they'd prefer RDF Core
>>or SW CG to take a first crack at this. Having it lodged in the RDF Issue
>>List seems worthwhile, regardless.
>
>I'm not clear what the problem is, so have no opinion at this point. 
>I believe there has been previous discussion which I haven't read 
>(yet).  It would be helpful to have a succinct statement of the 
>problem to capture in the issues list.

It arose from the discussion arising in response to your email to Tim 
asking for clarification of what it means to publish some RDF. The 
central 'deep' issue is what some RDF actually means, and whether it 
can be assigned meaning by virtue of the human-readable content of 
English comments in an RDF document. Peter and I think not, Tim and 
Dan C. apparently think so.

You can see the main discussion at 
http://lists.w3.org/Archives/Public/www-rdf-logic/2002May/0046.html
and the subsequent thread.

>At a first glance I have no problem with Peter's definitions of RDF 
>and RDFS entailment, but I don't follow why they imply that we have 
>to change the text in the RDFS spec describing rdfs:comment.  I take 
>it there is a deep issue here, not just one of wordsmithing.

There is a deep issue, but I think that we can skirt around it by 
wordsmithing. The text in question says:

"rdfs:comment

       The <code>rdfs:comment</code> property is used to provide a
       human-readable description of a resource.

       A textual comment helps clarify the meaning of RDF classes
       and properties. Such inline documentation complements the use
       of both formal techniques (Ontology and rule languages) and
       informal (prose documentation, examples, test cases). A
       variety of documentation forms can be combined to indicate
       the intended meaning of the classes and properties described
       in an RDF Schema.

       Multilingual documentation of schemas is supported at the
       syntactic level through use of the <code>xml:lang</code>
       language tagging facility. Since RDF schemas are expressed as
       RDF graphs, vocabularies defined in other namespaces may be
       used to provide richer documentation."

and all the issue is over that word 'meaning' in the second 
paragraph. If that were changed to 'intended meaning' then I think 
that Peter would have no quarrel with this. As it stands, however, it 
could be interpreted as saying that the *actual* RDF meaning of some 
RDF uriref is assigned by the content of an rdfs:comment, which is 
unfortunate.  The next sentence might also be modified, as I'm not 
sure what it is supposed to mean ('complements' ? Isnt this in-line 
documentation just as 'informal' as prose? In fact, isn't it prose?) 
It might also be useful if a sentence could be added to emphasize the 
point, eg "The actual meaning of any piece of RDF is defined by the 
formal semantics in use, and may differ from the intended meaning." 
at the end of the second paragraph. (That sentence is deliberately 
vague about which formal semantics is in use, by the way.)

Pat
-- 
---------------------------------------------------------------------
IHMC					(850)434 8903   home
40 South Alcaniz St.			(850)202 4416   office
Pensacola,  FL 32501			(850)202 4440   fax
phayes@ai.uwf.edu 
http://www.coginst.uwf.edu/~phayes