Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[protege-owl] Layered Ontologies?

Johann Petrak johann.petrak at chello.at
Tue Mar 3 10:37:38 PST 2009


Pitonyak, Andrew D wrote:
> Do you know the term Reification? The typical definition is to create a resource to act as a proxy for an object that you can not address. 
> 
> As this relates to RDF and OWL, you want to reference a statement (for example, who stated that fact?).
> 
> _:55445 rdf:type       rdf:Statement .
> _:55445 rdf subject    a:productA
> _:55445 rdf:predicate  a:weight 
> _:55445 rdf:object     "1.2"
> _:55445 dc:creator     a:Bob
> 

Yes, I know about reification, but I do not like it ;)
As has been pointed out in a previous email, any reified
statement leads a strange "double" life as far as
e.g. SPARQL Queries are concerned. So one cannot
just use the reified version of the triple alone and
as it seems, one has to make sure to always have both
versions in the store. Also this creates issues when
maintaining a  store that has both the original and
the reified versions of triples.

> 
>> If I understand things correctly, there is no way to use
>> a class as an individual in OWL2 and there is no way to
>> just "use" a class as an individual in OWL1 without
>> making the ontology Full and getting problems with
>> reasoning.
> 
> If you intend to use a class as an individual (ie, the target of an ObjectProperty) then yes, you have OWL Full.
> 
I am not sure what the implications are for reasoning when
the ontology becomes OWL full, but I think there are
severe implications, no?

But even worse, it seems to be entirely
impossible with OWL2, no? At least it seems to be
impossible to even create such a triple with
Protege4-beta.




More information about the protege-owl mailing list