W3C home > Mailing lists > Public > www-rdf-interest@w3.org > October 2000

RE: Circular references in RDF data model

From: McBride, Brian <bwm@hplb.hpl.hp.com>
Date: Thu, 5 Oct 2000 13:39:31 +0100
Message-ID: <5E13A1874524D411A876006008CD059F239417@0-mail-1.hpl.hp.com>
To: "'David Allsopp'" <dallsopp@signal.dera.gov.uk>, www-rdf-interest@w3.org
There is in general no prohibition as you suggest, so in theory
the same resource can be the subject and object of a statement,
an even also the predicate.

I have built two RDF implementations and haven't yet found this
to be a problem.

There are some restrictions on the creation of loops in rdf schema.
The subClassOf and subPropertyOf properties are constrained to 
not form loops - so a class can not be a subClass of itself.  It
has been suggested that these restrictions be lifted, but they are
in the current spec.

Brian McBride
HPLabs


> -----Original Message-----
> From: David Allsopp [mailto:dallsopp@signal.dera.gov.uk]
> Sent: 05 October 2000 11:38
> To: www-rdf-interest@w3.org
> Subject: Circular references in RDF data model
> 
> 
> Hello all,
> 
> A quick question abou the RDF data model:
> 
> I can't see any prohibition in the RDF Specification on circular
> references in RDF triples (such that the subject and object 
> are the same
> resource).  Is this correct? It looks like a pitfall for 
> implementors (a
> likely cause of infinite loop bugs), but perhaps I've missed 
> something?
> 
> Regards,
> 
> David.
> 
> 
> -- 
> Dr David Allsopp
>                  Quidquid latine dictum sit, altum viditur.
> 
Received on Thursday, 5 October 2000 08:39:35 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:44 GMT