Re: PROV-ISSUE-84 (namespace-for-properties): What should namespace for properties be? [Formal Model]

Hi Luc,
My responses are interleaved:

>shows properties such as <cf:isGeneratedBy>. I was hoping to see
<po:isGeneratedBy> (and >similarly for other relations)

That is an error - I have updated the example on the repository with "po"
namespaces for the "isGeneratedBy", "isControlledBy", "isDerivedFrom", and
"Used".

> I don't have access to the crimefile ontology, but I suppose it's domain
specific.

That is a good point - we have not created a crimefile ontology schema. It
might serve as an example to illustrate the extension mechanism to create a
domain-specific provenance ontology from the WG provenance ontology.

> While this may be good practice for owl ontology designers, the resulting
RDF representation is >not processable by an application that is not OWL
aware.
>I was hoping to see <po:isGeneratedBy> (and similarly for other relations)
so that the resulting >RDF is intelligible by applications that are not owl
aware.

I did not understand this point - applications that process RDF
representation but are not "OWL aware"?

>  It's more a lightweight vocabulary, than a full ontology with all bells
and whistles.
Any domain-specific ontology that extends the WG provenance ontology will
not be a vocabulary. If we want the provenance model (and the ontology) to
facilitate provenance interoperability on the Web they have to conform to WG
provenance model.

Thanks.

Best,
Satya


On Fri, Aug 26, 2011 at 3:35 AM, Provenance Working Group Issue Tracker <
sysbot+tracker@w3.org> wrote:

>
> PROV-ISSUE-84 (namespace-for-properties): What should namespace for
> properties be? [Formal Model]
>
> http://www.w3.org/2011/prov/track/issues/84
>
> Raised by: Luc Moreau
> On product: Formal Model
>
> The owl2 encoding of the file example at
>
> http://dvcs.w3.org/hg/prov/raw-file/default/ontology/ProvenanceFormalModel.html#owl2-encoding-of-the-provenance-model
>
> shows properties such as <cf:isGeneratedBy>.
>
> I don't have access to the crimefile ontology, but I suppose it's domain
> specific.
>
> While this may be good practice for owl ontology designers, the resulting
> RDF representation is not processable by an application that is not OWL
> aware.
>
> I was hoping to see <po:isGeneratedBy> (and similarly for other relations)
> so that the resulting RDF is intelligible by applications that are not owl
> aware.
>
> In my eyes, that's what we meant by 'lightweight ontology' at F2F1. It's
> more a lightweight vocabulary, than a full ontology with all bells and
> whistles.
>
>
>
>
>
>
>
>