RE: SEM: circular primitive

Oh yes, that looks right (i.e. like what I thought you were trying to say). 

I find KB4 entailing KB2 quite difficult on the circularity basis. 
If we name the class _:1 as <Circular> then we are back to saying things like 

  John rdf:type Circular .

iff
 
 John rdf:type Circular .

which is reminiscient of sets like the set of all sets that *do* belong to themselves. Does it belong to itself or not?

Jeremy

> -----Original Message-----
> From: www-webont-wg-request@w3.org
> [mailto:www-webont-wg-request@w3.org]On Behalf Of Peter F.
> Patel-Schneider
> Sent: 23 April 2002 14:01
> To: jjc@hplb.hpl.hp.com
> Cc: connolly@w3.org; jonathan@openhealth.org; www-webont-wg@w3.org
> Subject: RE: SEM: circular primitive 
> 
> 
> On further reflection here is my understanding of this case
> 
> KB1:
>      John rdf:type Person .
>      Bill rdf:type Person .
>      John child Bill .
> 
> KB2:
>      John rdf:type _:1 .
>      _:1 rdf:type daml:Restriction .
>      _:1 rdf:onProperty child .
>      _:1 rdf:hasClass :_1 .
> 
> KB3:
>   Bill rdf:type _:1 .
> 
> KB4:
>   John child John .
> 
> I do not believe (contrary to my previous assertion), that KB1 should
> entail KB2, precisely for the reasons mentioned by Jeremy.
> 
> However, KB4 should entail KB2.
> 
> Sorry for any confusion my confusion caused.
> 
> 
> Here is a further examples that may prove illustrative
> 
> KB5:
>   John rdf:type _:2 .
>   _:2 daml:onProperty child .
>   _:2 daml:cardinality "0" .
> 
> KB6:
>   John rdf:type _:3 .
>   _:3 daml:onProperty child .
>   _:3 daml:toClass _:3 .
> 
> I believe that KB5 should entail KB6.
> 
> peter
> 
> 

Received on Tuesday, 23 April 2002 09:59:52 UTC