Re: reification

>From: "Pat Hayes" <phayes@ai.uwf.edu>
>
>RE:  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0257.html
>
>>  >(exists (?x)
>>  >      (and
>>  >           (rdf:type ?x  rdf:Statement)
>>  >           (rdf:predicate ?x :pred)
>>  >           (rdf:subject ?x :subject)
>>  >           (rdf:object ?x :obj)
>>  >))
>>  >
>>  >So our interpretation is any of the things which qualify according
>>  >to the above formula that exist in our universe.
>>
>>  OK, that's fine. That just treats a reification as a piece of
>>  ordinary RDF with the standard MT reading. But that means
>>  that the  reification "of" a piece of RDF now has absolutely no
>>connection  with, or particular relationship to, the piece of RDF
>>  that it is a  reification "of".
>
>I think 'absolutely no connection' is a bit of an over statement; since we
>certainly have enough data in the reification syntax to retrieve any triple
>that it reified.

Well, that depends on what reification is supposed to mean, which is 
exactly what we are arguing about.  If reification refers to 
propositional content, then 'any triple that it reifies' means one 
thing; if it refers to syntactic form (ie is a kind of quotation) 
then it means something else. If it refers to the actual syntactic 
token, then 'any triple that it reifies' is meaningless, and we have 
to speak of the one particular triple (token) that it reifies. The 
wording in the M&S suggests the last interpretation to me, but others 
read it differently. In general, there seem to be several different, 
and incompatible, interpretations all in use in different contexts.

>  This means that there is a mapping from ?r to ?s (call it
>k ) which is entirely specified by the syntax;  see my diagram below.  Of
>course getting the quantification right on those variables is the job of the
>working group.
>
>http://robustai.net/mentography/reifyRDF.gif
>
>>That thing that exists is in the semantic universe:
>>  the piece of RDF that got reified was in the syntax. Nothing
>>  establishes any kind of connection between them. To get such a
>>  connection we need to extend the MT in some way.
>
>Ok I can see that.  If ?y and ?x are variables that can quantify over the
>set of things that exist corresponding to the triple and it's reification
>respectively,  then it seems to me that the job of the working group is to
>specify the mappings below and provide some interpretation for them:
>
>i : ?s -> ?y
>j : ?r -> ?x
>k : ?r -> ?s
>l: ?x -> ?y
>
>Informally I would think that the mapping from the reified node to the
>triple could be interpreted as naming, representing, or describing.  This
>strangely enough has the same interpretation as I would place on the mapping
>of a triple to its object in the semantic plain.  Is that interpretation
>totally off the wall ?

No, not at all. BUt there are other interpretations, and every 
interpretation has its defenders and detractors. Which, for a 
proposed standard language for world-wide machine communication, is 
not good. Any one of the above can be given a model theory pretty 
easily, the problem is to choose the right one.

The WG is working on this, as you know. Im sure we will come up with 
a coherent story eventually.

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

Received on Tuesday, 29 January 2002 17:28:09 UTC