Re: R2RML ontology bug? constant shortcut property definition not according to spec.

> On 24 Feb 2016, at 14:47, Martin G. Skjæveland <martige@ifi.uio.no> wrote:
> 
> 
> 
> On 24. feb. 2016 14:19, Richard Cyganiak wrote:
>> 
>>> On 24 Feb 2016, at 12:32, Ivan Herman <ivan@w3.org> wrote:
>>> 
>>>> I can confirm that the domain definitions for rr:subject, rr:predicate, rr:object and rr:graph in https://www.w3.org/ns/r2rml.ttl <https://www.w3.org/ns/r2rml.ttl> do not match (or even resemble) the specification.
>>>> 
>>>> There are numerous other problems with the OWL representation of this namespace document.
>>>> 
>>>> I’m afraid the document is unfit for purpose and best ignored.
>>> 
>>> Or, alternatively, the document is updated and changed…
>> 
>> Yes, this would of course be preferable.
>> 
>> But there are numerous issues with the current document, so this would be a slightly bigger project.
> 
> Not knowing what these issues are, would it not be possible to do a first quick fix by removing the causes of the inconsistencies and let the next version just be a mere vocabulary listing?
> 
> I would be interesting in joining (also) the slightly bigger project of making a richer version of the ontology.

Let us separate the issues. As I said, there is no problem to update the file to handle bugs. However, I would be uneasy to change the namespace document with a different ontology. That would require a community consensus of some sort, which is a different story.

Ivan


> 
> Martin
> 
>> Is it still known how the document was produced? There are HTML+RDFa, Turtle and RDF/XML variants (and possibly others?). Which one is the master copy? What toolchain was used to produce the others?
>> 
>> Richard
>> 
>> 
>> 
>> 
>>> 
>>> The point is: while it is not possible to change a file on /TR, I do not see any problem changing the namespace document in case there is a documented bug.
>>> 
>>> Ivan
>>> 
>>> 
>>>> 
>>>> Richard
>>>> 
>>>> 
>>>>> On 24 Feb 2016, at 11:08, Martin G. Skjæveland <martige@ifi.uio.no> wrote:
>>>>> 
>>>>> Hi all,
>>>>> 
>>>>> I think I have identified problems with the constant shortcut properties rr:subject, rr:predicate, rr:object and rr:graph in the R2RML ontology (http://www.w3.org/ns/r2rml). It seems at their domain definitions are wrong and not according to the specification text.
>>>>> 
>>>>> The attached test ontology contains two examples which are equivalent according to the recommendation text, but inconsistent according to the R2RML ontology. The attachment explains this in more detail.
>>>>> 
>>>>> Regards,
>>>>> Martin G. Skjæveland
>>>>> <r2rml-test.ttl>
>>>> 
>>>> 
>>> 
>>> 
>>> ----
>>> Ivan Herman, W3C
>>> Digital Publishing Lead
>>> Home: http://www.w3.org/People/Ivan/ <http://www.w3.org/People/Ivan/>
>>> mobile: +31-641044153
>>> ORCID ID: http://orcid.org/0000-0003-0782-2704 <http://orcid.org/0000-0003-0782-2704>
>> 


----
Ivan Herman, W3C
Digital Publishing Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
ORCID ID: http://orcid.org/0000-0003-0782-2704

Received on Wednesday, 24 February 2016 13:55:50 UTC