W3C home > Mailing lists > Public > www-webont-wg@w3.org > September 2003

RE: Chair's Concerns re: Test Cases and CR

From: Ian Horrocks <horrocks@cs.man.ac.uk>
Date: Tue, 30 Sep 2003 19:35:29 +0100
Message-ID: <16249.52465.881809.344858@merlin.horrocks.net>
To: "Smith, Michael K" <michael.smith@eds.com>
Cc: Frank van Harmelen <Frank.van.Harmelen@cs.vu.nl>, webont <www-webont-wg@w3.org>

On September 29, Smith, Michael K writes:
> 
> I agree with most of what has been said in this thread.
> 
> But, I would argue that use of oneOf is not really gratuitous.  We
> needed at least one example in order to present the concept.  And the
> selected uses are very similar and natural.  The example ontologies
> must be DL and because they were contrived to use all of the language
> it should not be surprising that they are a little odd.
> 
> That it has implications for efficiency for certain classes of
> reasoners is important to know, and alternative, more efficient
> formulations would be good to present.
> 
> Ian, what is the preferred, more efficient formulation?  Do you make
> the constants into classes and define WineFlavor as a union rather
> than collection?  And then what?  How do you rewrite 
> 
>   <WhiteWine rdf:ID="StGenevieveTexasWhite">
>     <locatedIn rdf:resource="#CentralTexasRegion" />
>     <hasMaker  rdf:resource="#StGenevieve" />
>     <hasSugar  rdf:resource="#Dry" />
>     <hasFlavor rdf:resource="#Moderate" />
>   </WhiteWine>
> 
> Create an instance of a class Dry, to plug into the hasSugar
> relation?

No, make "StGenevieveTexasWhite" an instance of the intersection of
WhiteWine and someValuesFrom restrictions "locatedIn
CentralTexasRegion" etc. (e.g., see below).

Another point here is that I would also question the use of an
individual for StGenevieveTexasWhite. Why not make it a sub class of
WhiteWine rather than an instance? This seems more intuitive and is
much more flexible as we can now add, say,
Vintage2001StGenevieveTexasWhite and/or StGenevieveTexasWhiteReserva
as a subClasses of StGenevieveTexasWhite. Instead, by choosing to make
it an individual, you have fixed a priori the level of detail that can
be represented in the ontology, seriously restricting its
applicability.

Here is the rewriting - don't blame me for the verbosity of the
syntax! (In practice you would probably have classes such as
"DryWine", "CentralTexasWine" etc, which would make this much easier
to write.)


<WhiteWine rdf:about="StGenevieveTexasWhite">
    <rdf:type>
        <owl:Class>
            <owl:intersectionOf rdf:parseType="Collection">
                <owl:Restriction>
                    <owl:onProperty rdf:resource="#locatedIn"/>
                    <owl:someValuesFrom>
                        <owl:Class rdf:about="CentralTexasRegion"/>
                    </owl:someValuesFrom>
                </owl:Restriction>
                <owl:Restriction>
                    <owl:onProperty rdf:resource="#hasMaker"/>
                    <owl:someValuesFrom>
                        <owl:Class rdf:about="StGenevieve"/>
                    </owl:someValuesFrom>
                </owl:Restriction>
                <owl:Restriction>
                    <owl:onProperty rdf:resource="#hasSugar"/>
                    <owl:someValuesFrom>
                        <owl:Class rdf:about="Dry"/>
                    </owl:someValuesFrom>
                </owl:Restriction>
                <owl:Restriction>
                    <owl:onProperty rdf:resource="#hasFlavor"/>
                    <owl:someValuesFrom>
                        <owl:Class rdf:about="Moderate"/>
                    </owl:someValuesFrom>
                </owl:Restriction>
            </owl:intersectionOf>
        </owl:Class>
    </rdf:type>
</WhiteWine>

Regards, Ian




> 
> - Mike
> 
> Michael K. Smith, Ph.D., P.E.
> EDS - Austin Innovation Centre
> 98 San Jacinto, #500
> Austin, TX  78701
> 
> phone: +01-512-404-6683
> email: michael.smith@eds.com
>   
> 
> 
> -----Original Message-----
> From: Ian Horrocks [mailto:horrocks@cs.man.ac.uk] 
> Sent: Monday, September 29, 2003 10:52 AM
> To: Frank van Harmelen
> Cc: webont
> Subject: Re: Chair's Concerns re: Test Cases and CR
> 
> 
> 
> On September 29, Frank van Harmelen writes:
> > 
> > 
> > 
> > Ian Horrocks wrote:
> > 
> > > the extensive use of oneOf in the wine and food
> > > ontologies is largely gratuitous (it seems to be mainly the result of
> > > their origin in a language that supported this constructor but did not
> > > support unions of classes), and is setting a bad example to
> > > prospective users - it encourages the use of statements that are, in
> > > most cases, stronger than is needed/intended, and that are known to be
> > > difficult to reason with.
> > >
> > > One further point. Given the elimination of oneOf, then the wine and
> > > food ontologies could even be transformed into OWL Lite, although this
> > > would result in some mangling of the syntax (in order to capture
> > > negation and disjunction).
> > 
> > Jeremy Carroll wrote:
> > 
> > > I have code for that if needed.
> > 
> > I think all these points:
> > 
> > 1) be careful with oneOf's
> > 2) without oneOf's, an ontology can often be transformed from DL into
> Lite
> > 3) that step can even be done automatically
> > 
> > are all very good points to make in whatever "style/how-to" guide we  will
> be 
> > writing.
> 
> Agreed, but w.r.t. point 1, our advice would carry more weight if we
> were seen to be taking it seriously in our own example ontologies!
> 
> Ian
> 
> 
> 
> > 
> > Frank.
> >    ----
> > 
> 
> 
> -- 
> Ian Horrocks, Department of Computer Science,
> University of Manchester, Oxford Road, Manchester, M13 9PL, UK.
> Tel: +44 161 275 6133/6248  Fax: +44 161 275 6211
> Email: horrocks@cs.man.ac.uk
> URL: http://www.cs.man.ac.uk/~horrocks
> 
Received on Tuesday, 30 September 2003 14:39:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:58:02 GMT