W3C home > Mailing lists > Public > www-rdf-logic@w3.org > July 2003

Re: Blank Node Equality Question

From: Peter F. Patel-Schneider <pfps@research.bell-labs.com>
Date: Mon, 21 Jul 2003 07:28:54 -0400 (EDT)
Message-Id: <20030721.072854.74194729.pfps@research.bell-labs.com>
To: jjc@hplb.hpl.hp.com
Cc: jimbobbs@hotmail.com, www-rdf-logic@w3.org

From: "Jeremy Carroll" <jjc@hplb.hpl.hp.com>
Subject: RE: Blank Node Equality Question
Date: Mon, 21 Jul 2003 12:08:17 +0200

> > 2/ If two blank nodes are connected to identical nodes via identical
> > predicates then can any inference made about the first also be made about
> > the second?  In RDF and RDFS this is true.  In many stronger languages,
> > including OWL, this is not true.   For example
> >
> > 	_:a rdf:type ex:Class .
> > 	_:b rdf:type ex:Class .
> >
> > entails in OWL
> >
> > 	_:a owl:sameAs _:a .
> >
> > but not
> >
> > 	_:a owl:sameAs _:b .
> >
> 
> Completely exchanging _:a for _:b (and vice versa) will always provide a
> proof for statements about _:b that are similar to proven statements about
> _:a, so that
> 
>  _:b owl:sameAs _:b .
> 
> follows with a proof identical to that for
> 
>   _:a owl:sameAs _:a .
> 
> Jeremy

Sure.  So there is a fourth way of reading the question.

In fact there are many more ways of reading the question if you allow
changing the notion of being connected to identical nodes via identical
predicates to a notion of indistinguishability.

Peter F. Patel-Schneider
Bell Labs Research
Lucent Technologies
Received on Monday, 21 July 2003 07:30:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:47 GMT