Re: OWL-S in protege?

On Jan 27, 2004, at 9:59 AM, Florian Probst wrote:

> hi,
> I am searching for reasons and explanations why most of the OWL-S 
> examples provided on the net can not be validated with an OWL 
> validator.

Both because there are direct OWL Fullisms in the core example 
documents, and because there are OWL Fullisms in the  core ontologies.

As I wrote before, part of the issue is legacy (conversion from 
DAML+OIL), there are some issues where the OWL-S coalition things that 
an OWL Fullism is the right choice for modeling something. (I tend to 
dissent ;)) Also, some of it is still just that not all members of the 
coalition are using OWL aware tools.

But your question is a bit careless, I'm sure you *could* validate most 
of the ontologies with an OWL Full validator.

> I would consider it as highly useful to be able to import OWL-S based 
> descriptions of services into the protégé OWL plugin. What are the 
> reasons for this incompatibility? Is this incompatibility intended?

The Protege OWL plugin only supports the OWL DL species. So if the 
coalition decides to go with (certain at least) OWL Full constructs, 
there will be a strong incompatibility. Whose "fault" that is, well, 
it'll be the result of the Protege guy and us making different 
decisions.

My goal is to eliminate all the purely gratuitous OWL Fullisms. Any 
that remain will have to be clearly justified. I believe that I can 
eliminate the rest in a fairly reasonable way.

As I have recently written, it's an action item on me to to this for 
the next incremental release, current scheduled for Feb. 15th.

Cheers,
Bijan Parsia.

Received on Wednesday, 28 January 2004 08:05:08 UTC