Antwort: WOWG: Potential new issues

Dear all:

taking the items

- Is owl:class different from rdfs:class (relates to level 0 issue)

- Will RDFS be level 0 of OWL?  (i.e. is RDFS : Level 1 :: Level 1 : Level 2)

- Do we allow foo:bar owl:subPropertyOf rdf:type??

from our new issues list, I would like to make a (maybe technically  _and_  
politicall VERY NAIVE !!) suggestion concerning the relation between RDF and 
OWL:

If we consider RDF and RDF(S) as level-0 of a Semantic Web language then what 
we need from there is a kind of simple ABox (to use an old-fashioned term - 
i.e., an RDF graph) and a very restricted form of a TBox (primitive classes, 
subclass relationships, and relations). All the features going _beyond_ that in 
the current RDF(S) approach have a tendency to be used in a way which can not 
be semantically well restricted in RDF(S) - because that is something to be 
really done in OWL !! RDF(S) has too much freedom here !

An RDF/RDF(S) "re-defined" or understood in this way  as "Semantic Web level-0" 
would be a _very valuable_ thing - because it will allow us to treat a vast 
amount of what we want to do on the Semantic Web by relatively clear and simple 
means.

So, we can think about two alternatives:

A) to ask the RDF Core Group for a discussion about these issues; or

B) define OWL in a way which takes such restrictions on the RDF/RDF(S) level as 
_pre-conditions_ for semantical well-foundedness of OWL ontologies (so that 
everybody using the Semantic Web with OWL has to follow these restrictions in 
order to avoid inconsistencies etc.).

In contrast to "dark triples" etc. this seems to be a way which can easier and 
clearer be communicated and explained.

TOO NAIVE??

Ruediger

Received on Thursday, 11 April 2002 09:50:06 UTC