Re: datatyping in test for Pat's review

• From: Ian Horrocks <horrocks@cs.man.ac.uk>
• Date: Sat, 24 May 2003 14:21:44 +0100
• To: Jeremy Carroll <jjc@hplb.hpl.hp.com>
• Cc: pat hayes <phayes@ai.uwf.edu>, Jeremy Carroll <jjc@hpl.hp.com>, www-webont-wg@w3.org
• Message-ID: <16079.29160.503449.800438@excalibur.oaklands.net>

On May 22, Jeremy Carroll writes:
> 
> 
> 
> pat hayes wrote:
> 
> > (Sending this quickly, more comments later.  -Pat )
> > 
> > section 4.1.2.
> > model/satisfying interpretation (Neither the RDF nor OWL documents uses 
> > 'model' in this technical sense.)
> > 
> > 2nd 'consistent' is potentially misused (referring to an 
> > interpretation), suggest
> > consistent with the constraints ... /satisfies all the constraints ....
> >
> 
> 
> Proposed rewording of this section.
> Note: this too some extent goes beyond the remit I felt I had after the LC 
> vote, but I suspect it would be churlish of me to refuse. This new text 
> conforms more closely with what the WG discussed at the January f2f.
> 
> OLD TEXT:
> [[
> 4.1.2. Semantic Conformance
> An OWL document is consistent with respect to a datatype theory [OWL 
> Semantics and Abstract Syntax], if and only if there exists some model of 
> the document that is consistent with the constraints specified by the 
> relevant model theory (see [OWL Semantics and Abstract Syntax]: OWL Lite 
> and OWL DL, OWL Full).
> ]]
> 
> NEW TEXT:
> [[
> 4.1.2. Semantic Conformance
> An OWL Lite or OWL DL document is consistent with respect to a datatype 
> theory [OWL Semantics and Abstract Syntax], if and only if
> a corresponding collection of OWL DL ontologies in abstract syntax form 
> with a separated vocabulary is simulataneously
> <a 
> href="https://www.w3.org/TR/2003/WD-owl-semantics-20030331/direct.html#direct_consistent"
>  >consistent</a> with respect to the datatype theory.
> 
> An OWL Full document is consistent with respect to a datatype theory [OWL 
> Semantics and Abstract Syntax], if and only if it is a member of an
> imports closed collection of RDF graphs which is OWL Full consistent with
> respect to the datatype theory.
> ]]
> with "imports closed" and "OWL Full consistent" linked to their definitions in
> https://www.w3.org/TR/2003/WD-owl-semantics-20030331/rdfs.html
I prefer something closer to the original text (at least for OWL DL
consistency). The relevant words can be copied almost verbatim from
S&AS. (The only problem there is that S&AS currently says "if" - I
believe that this should be strengthened to "iff" - I will mention it
to Peter.) E.g.:
An OWL DL document D is consistent with respect to datatype theory T if
and only if there is some Abstract OWL interpretation I with respect
to T such that I satisfies D.
An OWL Full document D is consistent with respect to a datatype theory
T [OWL Semantics and Abstract Syntax], if and only if there is some
OWL Full interpretation I with respect to T such that I satisfies all
the RDF graphs in the import closure of D.
> 
> The second paragraph is slightly broken since the datatyping in the rdfs 
> part of OWL S&AS is conspicuous by its absence, but I believe Peter will 
> fix that.
> 
> The new text defers as much as I can to S&AS and all the technical terms 
> are taken from S&AS.
> 
>  
> > 4.2.2
> > Im still not happy with the way that conformance is stated.
> > 
> 
> <discussion snipped>
> 
> 
> > The cheapest way around this would be to add a remark when you give the 
> > definition of 'complete' to indicate that this sense is not the standard 
> > sense. For example
> > "This is stronger than the usual sense of completeness used in 
> > describing logical inference systems, which refers only to the detection 
> > of inconsistency."
> 
> 
> I will add this note, at the end of the section, but with "This" expanded 
> to be something like "The use of the word 'complete' in complete and 
> terminating and complete OWL Lite consistency checker"
I can live with this. I would prefer it if the text made it clear that
the use of complete here *is* consistent with the standard usage in
algorithms for decidable logics (e.g., in the modal, temporal and
description logic communities, amongst others).
Ian
> 
> 
> > 
> > A better way, IMHO, <snipped>
> 
> 
> I did not hear wg support for a better way.
> 
> Jeremy
> 
> 
> 

Received on Saturday, 24 May 2003 08:21:02 UTC