W3C home > Mailing lists > Public > www-rdf-logic@w3.org > June 2001

Re: properties with classes as values?

From: pat hayes <phayes@ai.uwf.edu>
Date: Thu, 7 Jun 2001 13:04:12 -0500
Message-Id: <v04210158b74570bd126d@[205.160.76.219]>
To: "Sean B. Palmer" <sean@mysterylights.com>, "David Martin" <martin@ai.sri.com>
Cc: www-rdf-logic@w3.org
> > What I want is that the value of myProperty can be
> > *any* class.
>
>Then the range of it is rdfs:Class.
>
>   :myProperty rdfs:range rdfs:Class .
>
> > >    :x rdfs:range
> > >       [ daml:intersectionOf
> > >           (:Class [ rdfs:subClassOf :Animal ]) ] .
> > > XML RDF of these things available upon request...
> >
> > I would very much appreciate seeing the last sample above
> > in DAML+OIL.
>
><daml:Class rdf:ID="x">
> <rdfs:range>
>  <rdf:Description>
>   <daml:intersectionOf parseType="daml:collection">
>     <daml:Class rdf:ID="Class"/>
>     <rdf:Description>
>       <rdfs:subClassOf rdf:resource="#Animal"/>
>     </rdf:Description>
>   </daml:intersectionOf>
>  </rdf:Description>
> </rdfs:range>
></daml:Class>
>
>In prose: the range of class "x" is an intersectionOf "#Class" and
>something which is the subClassOf "#Animal"; which most likely isn't
>what you really want to say.
>
> >  (a) assert that something is a subclass of Animal
> >  (b) denote the set of all subclasses of Animal?
>
>[ rdfs:subClassOf :Animal ] should be read as "something which is a
>subClassOf Animal". Not all of the subClasses, one specific sub class.
>So, (a).
>
> > Your example seems to rely on (b).
>
>Nope.

There is a way that one might be led to think of (b), however. In 
general, an expression denoting an 'anonymous' entity can be 
understood as denoting a property (the one shared by all the possible 
entities that could be the anonymous entity), so that with this 
understanding, [rdfs:subClassOf :Animal] would refer to the property 
of being a subclass of Animal. And then, given that this is a 
class-based language where a property is understood to be a class, 
this might easily be taken to refer to the class of all subclasses of 
Animals. In sum, there is a natural conceptual path from 'anonymous 
thingie' to 'class of thingies'.

So the upshot is, that we probably need to be extra-sensitive to this 
natural slope and put up guard rails in the DAML documentation. 
Thanks for bringing this up, guys.

Pat Hayes

---------------------------------------------------------------------
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 Thursday, 7 June 2001 14:04:08 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:40 GMT