W3C home > Mailing lists > Public > public-prov-wg@w3.org > May 2012

Re: PROV-ISSUE-371 (junzhao): timestamped provo.owl [PROV-O HTML]

From: Jun Zhao <jun.zhao@zoo.ox.ac.uk>
Date: Wed, 09 May 2012 12:43:20 +0100
Message-ID: <4FAA5858.6040807@zoo.ox.ac.uk>
To: Stephan Zednik <zednis@rpi.edu>
CC: Timothy Lebo <lebot@rpi.edu>, Provenance Working Group WG <public-prov-wg@w3.org>
Tim and Stephan,

What Stephen outlined here will provide a very satisfactory solution to 
my requirement.

How far you want to go probably depends on your momentum, and time and 
resources. Getting as far as sorting out the content negotiation will 
make me super happy. But I can live with some the following implemented, 
if not all:)

Cheers,

Jun


On 09/05/2012 00:45, Stephan Zednik wrote:
>
> On May 8, 2012, at 5:07 PM, Timothy Lebo wrote:
>
>> Stephan,
>>
>> On May 8, 2012, at 5:59 PM, Stephan Zednik wrote:
>>
>>> I thought the OWL2 spec stated that the object of owl:versionInfo was supposed to be a literal.
>>>
>>> from http://www.w3.org/TR/owl-guide/
>>>
>>> "the object of owl:versionInfo is a literal and the tag can be used to annotate classes and properties in addition to ontologies."
>>
>>
>> Thanks for pointing that out.
>>
>>
>>
>>>
>>> and that is why owl:versionIRI was created in OWL2 to specifically refer to IRI (and also to be functional) where the specific version of the ontology can be found.
>>
>> I'll change it to versionIRI (now that I found it in http://www.w3.org/2007/OWL/refcard )
>>
>> Do you have a pointer to the documentation for versionIRI and how it should be used?
>
> There are descriptions at http://www.w3.org/2007/OWL/wiki/New_Features_and_Rationale#Imports_and_Versioning and http://protegewiki.stanford.edu/wiki/How_Owl_2.0_Imports_Work.
>
> The idea is that the version tagged ontology be available at the versionIRI.
>
> from my understanding it should be as simple as
>
>      <owl:Ontology rdf:about="http://www.w3.org/ns/prov#">
>          <rdfs:label xml:lang="en">PROV Ontology</rdfs:label>
> 	<owl:versionIRI rdf:resource="http://dvcs.w3.org/hg/prov/raw-file/WD-prov-o-2012MMDD/ontology/ProvenanceOntology.owl"/>
>      </owl:Ontology>
>
> Question: Did we mean to tag the ontology with 'MMDD' for month and year?
>
> As for content negotiation, I think it would be nice if http://www.w3.org/TR/2012/WD-prov-o-20120503/ was configured to return the tagged ontology file as RDF/XML for requests made with accept headers specifying "application/rdf+xml".  This would be very clean and would not result in the ontology specifying a HG dependent URL.
>
> Our WD tagged ontology would have:
>
>      <owl:Ontology rdf:about="http://www.w3.org/ns/prov#">
>          <rdfs:label xml:lang="en">PROV Ontology</rdfs:label>
> 	<owl:versionIRI rdf:resource="http://www.w3.org/TR/2012/WD-prov-o-20120503/"/>
>      </owl:Ontology>
>
> This will work with the RDF tools I am familiar with (Protege and topbraid) since they use accept headers to specifically ask for RDF/XML responses.
>
> If some tools aren't using accept headers and get HTML back by mistake, then I would suggest http://www.w3.org/TR/2012/WD-prov-o-20120503.owl or http://www.w3.org/TR/2012/WD-prov-o-20120503/prov.owl, as long as its related to the working draft URL and does not reference the hg URL.
>
> --Stephan
>
>> I'm having trouble navigating the specs :-)
>>
>>>
>>> If this IRI references that specific version of the OWL file, I would suggest using owl:versionIRI since that seems to exactly match our intention.
>>
>> Sounds good.
>>
>> -Tim
>>
>>>
>>> --Stephan
>>>
>>> On May 8, 2012, at 3:39 PM, Timothy Lebo wrote:
>>>
>>>> Currently:
>>>> <owl:versionInfo rdf:resource="http://www.w3.org/TR/2012/WD-prov-o-2012MMDD"/>
>>>> points to the HTML, which has a link to the OWL file.
>>>> Is that adequate?
>>>>
>>>
>>
>
>
Received on Wednesday, 9 May 2012 11:43:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 11:43:52 GMT