- From: Stian Soiland-Reyes <soiland-reyes@cs.manchester.ac.uk>
- Date: Thu, 12 Jan 2012 09:42:02 +0000
- To: Daniel Garijo <dgarijo@delicias.dia.fi.upm.es>
- Cc: Luc Moreau <L.Moreau@ecs.soton.ac.uk>, public-prov-wg@w3.org
On Wed, Jan 11, 2012 at 14:54, Daniel Garijo
<dgarijo@delicias.dia.fi.upm.es> wrote:
> Hi Luc,
> I don't think it would affect the ontology: we could create an instance of a
> "Generation" and say that the qualified
> entity is the generated entity. Something like:
Also we already have a specialised property on the entity,
prov:wasGeneratedAt (which we are not sure if should stay):
http://dvcs.w3.org/hg/prov/raw-file/tip/ontology/ProvenanceFormalModel.html#wasGeneratedat
ex:generatedEntity prov:wasGeneratedAt [ time:inDateTimeXSD
"2011-10-21T09:21:31Z" ] .
Note:
I would argue that all QualifiedInvolvements and Entitiies can have a
duration, ie. both a start and end time (as the OWL currently permits)
- but this should be reflected back in PROV-DM. prov:generatedAt can
then be made a specialisation of prov:startedAt.
> ex:g1 a wfprov:Generation;
> prov:hadQualifiedEntity ex:generatedEntity;
> ext:generatedAt [ time:inDateTimeXSD "2011-10-21T09:21:31Z" ]. #Or
> whatever the user wants to use for asserting time
>
> However, I think that Stephan was against using orphan generations like in
> the example. IMO it is a bit tricky to find
> a generation without activity, but it covers the use case.
We know it was generated by an activity, we just don't know its name..
so you could link to the generation using bnodes:
ex:generatedEntity prov:wasGeneratedBy [
a prov:Activity;
prov:hadQualifiedGeneration [
a prov:Generation;
prov:hadQualifiedEntity ex:generatedEntity;
prov:startedAt [ time:inDateTimeXSD "2011-10-21T09:21:31Z" ]
]
] .
This is can be inferred from the shorter prov:wasGeneratedAt.
--
Stian Soiland-Reyes, myGrid team
School of Computer Science
The University of Manchester
Received on Thursday, 12 January 2012 09:43:02 UTC