W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > January 2002

Re: No model theory for reification?

From: Patrick Stickler <patrick.stickler@nokia.com>
Date: Fri, 11 Jan 2002 08:44:38 +0200
To: ext Graham Klyne <Graham.Klyne@MIMEsweeper.com>, Dan Brickley <danbri@w3.org>
CC: RDF Core <w3c-rdfcore-wg@w3.org>
Message-ID: <B8645676.B617%patrick.stickler@nokia.com>
On 2002-01-11 1:07, "ext Graham Klyne" <Graham.Klyne@MIMEsweeper.com> wrote:


> One thing that I can imagine doing, but something that I think is beyond
> 
> our current charter, is to have a way of relating an rdf:Statement
> instance 
> to a truth value;  i.e. a way of saying that an rdf:Statement
> corresponds 
> to an asserted triple.  A simple example might be similar to the style
> of 
> RDFS-entailment, such that:
> 
>    s a rdf:Statement ;
>      rdf:predicate pred ;
>      rdf:subject   sub ;
>      rdf:object    obj ;
>      rdf:hasValue  rdf:True .
> 
> statement-entails:
> 
>      sub pred obj .
> 
> (I don't think this particular approach is especially useful;  e.g. for
> provenance we'd want to be able to make the entailment also depend in
> some 
> way on expressed trust in the source of the statement, but I think
> that's 
> still an area for experimentation.)

This is precisely what I had suggested some time ago with my X proposal,
which also takes into account issues such as trust/authority, as well
as source, authentication, scope, and all other qualifications of
statements. It's just an issue of (a) having only one representation
for statements and (b) shifting matters of assertion, trust, etc. to
a suitable ontology (or ontologies) for qualifying statements. E.g.

    s a rdf:Statement ;
      rdf:predicate      pred ;
      rdf:subject        sub ;
      rdf:object         obj ;
      rdf:assertedBy     <auth://abc.com/john.doe> ;
      rdf:source         <http://www.abc.com/~jdoe/what-I-know.rdf> ;
      rdf:scope          dilusions ;
      rdf:authentication <esl:SHA1:i8...a82> .

and the triple <s,p.o> is just a resource-centric view of that knowledge
derived from the qualified statement, as Graham points out.

All statements are "reified", per se. Assertion is taken to be an issue
of relevance to any given process -- and selection of relevant statements
is based on a statement's qualification properties. I.e. assertion is
a preliminary query on the statements in a knowledge base to select
knowledge of relevance to a subsequent query or operation.

C.f.

http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Nov/0303.html
http://www-nrc.nokia.com/sw/draft-pstickler-auth-00.html

Of course, changes of a magnitude such as suggested above are totally
beyond the scope of the present charter, which is why I dropped discussion
of the X proposal (for now ;-)

Cheers,

Patrick

--
               
Patrick Stickler              Phone: +358 50 483 9453
Senior Research Scientist     Fax:   +358 7180 35409
Nokia Research Center         Email: patrick.stickler@nokia.com
Received on Friday, 11 January 2002 01:43:58 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:43:53 EDT