Re: [poe] ODRL Ontology in non-perfect state

> The current ODRL ontology fails to be loaded by Protègè v.5.2 (the up to date version).
> 
> I cannot openly denounce any specific mistake, but I raise the warning as most of our readers will use Protègè. Or am I the only one with such a problem? I have been able to open the ontology with specific software (owlgred, swoop and older versions of Protègè once transformed the ontology into RDF/OWL by converters like EasyRDF -http://www.easyrdf.org/converter- <http://www.easyrdf.org/converter->)
> 
> My own wrapper around OWL API (available here http://owlprofiler.appspot.com/ <http://owlprofiler.appspot.com/>) actually determines the ontology is not OWL DL because it is using several undefined resources.
> For example, we speak about vcard:Organization, but the reasoner does not have any clue about it being a class, and consistently throws an error. The same applies to the metadata properties we use ("Use of undeclared object property"). We can solve them, I think. For example, once I simply add the axiom:
> 
> vcard:Organization a owl:Class .
> The error promptly disappears. I believe the same policy should be followed by each of the external elements.
> 
> The inconsistencies are also recognized by Pellet, but not by the online validator of the Univ. of Manchester. http://mowl-power.cs.man.ac.uk:8080/validator/ <http://mowl-power.cs.man.ac.uk:8080/validator/>
> The ultimate goal I have in mind is that nobody can claim the ontology has errors/inconsistencies. Any opinion about this?
> 

While I think having our ontology recognized by these tools is a fine goal, I am a bit uneasy. To take the vcard example above: as a naïve user I would expect a reasoning tool to pick up the ontology for vcard and add the resulting axioms to the ODRL ones, and not expect everything to be in one place. Otherwise the tools breaks the distributive nature of the Web.

Of course, if the vcard (OWL) ontology does not exist, or is not accurate than we may think of doing this but then… again thinking in an idealistic semantic web way, we add a new statement on an ontology that we do not own and that statement will be available for everyone on the (Semantic) Web. It is not really o.k., is it? (I believe community referred to that as 'ontology hijacking':-)

Bottom line: I am tempted not to add these external elements.



-- 
GitHub Notification of comment by iherman
Please view or discuss this issue at https://github.com/w3c/poe/issues/189#issuecomment-306162909 using your GitHub account

Received on Monday, 5 June 2017 11:09:52 UTC