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

Jun,

On May 8, 2012, at 5:16 PM, Jun Zhao wrote:

> Hi Tim,
> 
> Your question made me wonder whether I've missed something. Having missed a few telcoms in a row does not help either.
> 
> My requirement is very simple and reflects my concern that currently people might not be able to access a specific version of the ontology that corresponds to each version of the draft specification. If you say your proposed solutions can satisfy this simple need, then please feel free to close the issue.

Please review

http://www.w3.org/mid/887B0652-9B9C-4087-9A47-8123C7B5879F@rpi.edu
and
http://www.w3.org/mid/E3C07901-9C5C-4D20-97F2-6A636D3990A1@rpi.edu

and let me know if this satisfies your needs.


> 
>>> 
>>> I am happy to help settle the versioning issues with the OWL ontology,
>>> but I'm afraid that I'm not sure what your concerns are.
>>> 
>>> We have three options:
>>> 
>>> 1) hg tag the OWL file
> 
> Am I right to say that with this solution someone can find a specific version of an ontology by accessing the mercurial repository?

mercurial allows anybody to access all versions of the ontology.
Adding a tag corresponding to a particular "release" will allow anyone to distinguish the released ontology from the many smaller and irrelevant changes.

> It is /a/ solution, provided that everybody who has a need to access a specific or previous version of an ontology is happy to use the hg.

I hope they are happy to use hg, since it is a URL request like any other URL request:

http://dvcs.w3.org/hg/prov/raw-file/WD-prov-o-2012MMDD/ontology/ProvenanceOntology.owl



> 
> (Please don't ask me why I want people to access any specific or previous version of an ontology. We are a provenance WG, and we deal with history, no?:) And of course I have other reasons if you want me to list them. But you are asking what I want, not why I want, right?)

Correct. I'm just trying to satisfy your concern so that we may close this issue.
If you are not happy yet, perhaps a little "why" would help.

> 
> 
>>> 2) add to the prov-o html automation to reference the OWL version that it is documenting
> 
> Yes, that's what we did with OPMV and the Provenance Vocabulary. Whether you want to use tags or different OWL files to achieve this, it's up to your choice.
> 
>>> 3)<>  owl:versionURI
> 
> Also helps, as long as the different versions of OWL files that are identified by the versionURI can be accessed.

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?

Thanks,
Tim



> 
> Does it help? I really hope it does!
> 
> cheers,
> 
> Jun
>>> 
>>> Could you please clarify your concerns so that we can scope the effort to address them?
>>> 
>>> Thanks,
>>> Tim
>>> 
>>> 
>>> On May 5, 2012, at 7:48 AM, Paul Groth wrote:
>>> 
>>>> This seems good. Stian can you add it?
>>>> 
>>>> Thanks
>>>> Paul
>>>> 
>>>> On Thu, May 3, 2012 at 1:26 PM, Jun Zhao<jun.zhao@zoo.ox.ac.uk>  wrote:
>>>>> On 03/05/2012 11:02, Stian Soiland-Reyes wrote:
>>>>>> Don't you think the OWL should contain something like
>>>>>> 
>>>>>> <>   owl:versionIRI
>>>>>> <www.w3.org/TR/2012/WD-prov-o-20120501/ProvenanceOntology.owl>    ?
>>>>> 
>>>>> Stian, yes, we should also have that!
>>>>> 
>>>>> -- Jun
>>>>> 
>>>>>> 
>>>>>> I would +1 that as people like myself will download the OWL locally
>>>>>> for processing with say Sesame-Elmo, and it later will be important to
>>>>>> know which one it is based on.
>>>>>> 
>>>>>> We just need to know the magic date to add it in advance.
>>>>>> 
>>>>>> 
>>>>>> On Wed, May 2, 2012 at 9:39 AM, Jun Zhao<jun.zhao@zoo.ox.ac.uk>   wrote:
>>>>>>> Hi Tim,
>>>>>>> 
>>>>>>> I am happy with what we will do with the public release.
>>>>>>> 
>>>>>>> And dealing with versioning for internal releases can wait if you are
>>>>>>> overwhelmed by other commitment at the moment.
>>>>>>> 
>>>>>>> -- Jun
>>>>>>> 
>>>>>>> 
>>>>>>> On 02/05/2012 00:27, Tim Lebo wrote:
>>>>>>>> 
>>>>>>>> Jun,
>>>>>>>> 
>>>>>>>> The prov.owl will be "copied" to the official w3c website directory when
>>>>>>>> the WD2 is published on Thursday, so there will be no question about what
>>>>>>>> OWL file the HTML is talking about.
>>>>>>>> 
>>>>>>>> Hopefully, the "dereferencability problem" (which paul took on and we
>>>>>>>> asked Daniel to help with) will be addressed soon, which will provide the
>>>>>>>> latest OWL when requesting the terms' URIs.
>>>>>>>> 
>>>>>>>> If we want to be explicit about what version of the ontology the HTML is
>>>>>>>> taking about, I can look into exposing that within every compiled draft up
>>>>>>>> to LC that is due in a few weeks. But generally, these are always in sync
>>>>>>>> because the ontology changes less frequently and the HTML is generated much
>>>>>>>> more frequently.
>>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>>>> Please let me know which aspects you need most, so that we can address the
>>>>>>>> right issues soon.
>>>>>>>> 
>>>>>>>> Regards,
>>>>>>>> Tim
>>>>>>>> 
>>>>>>>> Sent from my iPhone
>>>>>>>> 
>>>>>>>> On May 1, 2012, at 11:38, Provenance Working Group Issue
>>>>>>>> Tracker<sysbot+tracker@w3.org>     wrote:
>>>>>>>> 
>>>>>>>>> PROV-ISSUE-371 (junzhao): timestamped provo.owl [PROV-O HTML]
>>>>>>>>> 
>>>>>>>>> http://www.w3.org/2011/prov/track/issues/371
>>>>>>>>> 
>>>>>>>>> Raised by: Jun Zhao
>>>>>>>>> On product: PROV-O HTML
>>>>>>>>> 
>>>>>>>>> Can we talk  about when or whether we will have snapshots for our
>>>>>>>>> ontology, like  ProvenanceOntology-20120430.owl? Or achieve similar
>>>>>>>>> functionality via other mechanisms?
>>>>>>>>> 
>>>>>>>>> Because our ontology is still work in progress, it is important to have
>>>>>>>>> the right ontology content associated with each prov-o spec public release
>>>>>>>>> or even work draft.
>>>>>>>>> 
>>>>>>>>> I think this would be something really nice to have at least for this
>>>>>>>>> upcoming public release.
>>>>>>>>> 
>>>>>>>>> I am happy to discuss more on this.
>>>>>>>>> 
>>>>>>>>> -- Jun
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>>> 
>>>>>>>> 
>>>>>>> 
>>>>>>> 
>>>>>> 
>>>>>> 
>>>>>> 
>>>>> 
>>>>> 
>>>> 
>>>> 
>>>> 
>>>> --
>>>> --
>>>> Dr. Paul Groth (p.t.groth@vu.nl)
>>>> http://www.few.vu.nl/~pgroth/
>>>> Assistant Professor
>>>> Knowledge Representation&  Reasoning Group
>>>> Artificial Intelligence Section
>>>> Department of Computer Science
>>>> VU University Amsterdam
>>>> 
>>>> 
>>> 
>>> 
>>> 
>> 
> 
> 
>