[
Lists Home |
Date Index |
Thread Index
]
John Cowan wrote:
>
> "Emmanuil Batsis (Manos)" scripsit:
>
> > The usability of anonymous nodes can be overrated as they don't have a
> > URI serving as a global unique identifier (and potentialy locator) for
> > them.
>
> Indeed. One would have to impose a higher-order rule that says "two
b-nodes
> with identical subjectIndicator properties are to be treated as
identical",
> as indeed they do in the TM world.
Yes, actually this is where owl:FunctionalProperty and
owl:InverseFunctionalProperty come in.
http://www.w3.org/TR/owl-ref/#FunctionalProperty-def
In the above the property "subjectIndicator" would be defined as
<owl:FunctionalProperty rdf:ID="subjectIndicator" />
and hence if you had the following:
#a :subjectIndicator #foo .
#b :subjectIndicator #foo .
you could conclude that:
#a owl:sameIndividualAs #b
[...]
>
> > The right thing to do would be to use a non-retreivable scheme URI
> > (perhaps a URN?) (or an rdf:ID) to denote Shakespeare, as one cannot
> > simply download him. His picture should be mentioned as just that, not
> > as the person itself. Anonymous nodes are not needed to avoid confusion:
> >
> > <rdf:Description rdf:about="nonRetreivable://Persona/Shakespeare">
> > <foo:picture rdf:resource="http://URL/to/pic"/>
> > </rdf:Description>
>
> The rdf:ID to which you refer is the b-node/anonymous node I was talking
> about in the first place.
for example:
<owl:FunctionalProperty rdf:ID="ssn" >
<rdfs:domain rdf:resource="#Person" />
</owl:FunctionalProperty>
then the following:
<Person>
<ssn rdf:resource="#000-11-1234" />
<haircolor rdf:resource="#blond" />
</Person>
and somewhere else:
<Person>
<ssn rdf:resource="#000-11-1234" />
<nosering rdf:resource="#large"/>
</Person>
you could conclude that these two (anonymous) persons are the same
individual. If you think about it, this is a totally powerful facility --
and rather useful.
Jonathan
|
