Re: PROV-ISSUE-232 (TLebo): activity start/end: direct time upgrades to instantaneous events [Ontology]

Although we've removed the 2006 namespace and any reference to its classes/properties, I still have a concern.

The demand to use prov:TimeInstant is unnecessarily verbose

I wrote [1] to contrast the options:
:activity
   a prov:Activity;
   # Unqualified start
   prov:startedAtTime    "2012-01-01T01:01:01-05:00"^^xsd:dateTime;
   # Qualified Start
   prov:startedAt [
      a prov:TimeInstant; 
      prov:atTime        "2012-01-01T01:01:01-05:00"^^xsd:dateTime; 
   ];

Practical users of prov-o will prefer the "unqualified start" modeling because it avoids minting rdfs:Resources just to state the time.
"Proper" provenance insists on the nuanced indirection (unless WD4 fixed this).
Having this extra level of indirection slows queries and can lead to duplicate results (unless we impose atTime being inverse functional, but that's a lot of overhead that could be avoided with a direct triple.)

In the example, I was using the pattern "atTime" to indicate that the range is a xsd:dateTime. This is seen in startedAtTime and atTime.

-Tim

[1] http://dvcs.w3.org/hg/prov/file/d9433150115d/examples/eg-14-unqualified-and-qualified-times/rdf/eg-14-unqualified-and-qualified-times.ttl




On Mar 6, 2012, at 2:33 PM, Daniel Garijo wrote:

> Hi Tim,
> now startedAt and endedAt can be bound to timeInstants (not using the 2006 namespace).
> Can we close this issue?
> 
> Thanks,
> Daniel
> 
> 2012/2/3 Timothy Lebo <lebot@rpi.edu>
> Per Ivan, we should not use the 2006 namespace because it was never a final document.
> We can mirror what we need into our own namespace.
> 
> TODO: Tim to change the owl ont.
> 
> -Tim
> 
> On Feb 3, 2012, at 3:43 AM, Provenance Working Group Issue Tracker wrote:
> 
> > PROV-ISSUE-232 (TLebo): activity start/end: direct time upgrades to instantaneous events [Ontology]
> >
> > http://www.w3.org/2011/prov/track/issues/232
> >
> > Raised by: Timothy Lebo
> > On product: Ontology
> >
> > per 9:40 Friday f2f2
> >
> >
> >
> >
> 
> 
> 

Received on Tuesday, 6 March 2012 19:56:48 UTC