- From: Timothy Lebo <lebot@rpi.edu>
- Date: Fri, 9 Mar 2012 08:47:36 -0500
- To: Stian Soiland-Reyes <soiland-reyes@cs.manchester.ac.uk>
- Cc: Daniel Garijo <dgarijo@delicias.dia.fi.upm.es>, Provenance Working Group WG <public-prov-wg@w3.org>
On Mar 9, 2012, at 4:43 AM, Stian Soiland-Reyes wrote: > On Fri, Mar 9, 2012 at 04:29, Timothy Lebo <lebot@rpi.edu> wrote: >> I left TimeInstant and prov:inXSDDateTime around so that some _may_ use it if they wish, but it is not a principal (simple) modeling construct. > >> But I don't think this prevents Stian from just associating the temporal entities directly (activities, usages, etc). No? > > Now that they are all InstantaneousEvent, then no, that should still > be kind-of fine, as I can do: > > :activity1 a prov:Activity ; > prov:qualifiedStart :activity1Start ; > prov:qualifiedEnd :activity1End . > > :entity1 prov:qualifiedGeneration :entity1Gen . > > :activity1Gen ex:after :activity1Start . > :activity1End ex:after :activity1Gen . > > and in fact, if I as an asserter still like OWL Time, I can make > :activity1Gen etc. instances of time:Instant and use time:after > instead of my own ex:after - even make a property ex:started as > subproperty of both prov:qualifiedStart and time:hasBeginning (and > equivalent for prov:atTime/time:inXSDDateTime) and talk about > :activity1 as an time:Interval. > I added this as https://www.w3.org/2011/prov/track/issues/309 on the best practices doc. -Tim
Received on Friday, 9 March 2012 14:01:03 UTC