W3C home > Mailing lists > Public > public-prov-wg@w3.org > February 2012

Re: PROV-ISSUE-242 (TLebo): generated twice? [prov-dm]

From: Paolo Missier <Paolo.Missier@ncl.ac.uk>
Date: Tue, 07 Feb 2012 14:00:40 +0000
Message-ID: <4F312E88.7010808@ncl.ac.uk>
To: Stian Soiland-Reyes <soiland-reyes@cs.manchester.ac.uk>
CC: Provenance Working Group <public-prov-wg@w3.org>
Stian, all

sorry, my email client has been acting up. trying again...

I support the use of symbolic events (what you call "instant resources", same thing), and the separate mapping of events to time. So 
from:

 > :e prov:wasGeneratedAt :t1 .
 > :e prov:wasGeneratedAt :t2 .

you can readily infer :t1 == :t2  which is what matters -- and possibly try and reconcile clock discrepancies separately

-Paolo


On 2/7/12 1:09 AM, Stian Soiland-Reyes wrote:
>> Is the following valid?
>>
>> wasGeneratedBy(e,,2001-10-26T21:32:52)
>> wasGeneratedBy(e,,2011-11-27T21:23:34)
> You can't write time like that in ASN, they must be quoted literals or
> kind-a-CURIEs. (And we should choose one of those to use for time in
> ASN - it seems we already like to talk about t1,t2 etc, but elsewhere
> claim it is XSD Date time. Given discussions in F2F I would suggest
> the former.)
>
>
> My interpretation of this is that 2011-11-27T21:23:34 and
> 2001-10-26T21:32:52 is the same event (the generation event). I don't
> know what that means, perhaps that the generation is a very slow
> thing, that the observer's clock has moved but the entity's clock has
> not (say a frozen computer simulation), that the two times are the
> same, or that the asserter is plain simply wrong.
>
> And so I don't think we should say much about it.
>
> It becomes of course much easier to narrow down the possibilities if
> we separate times and events, as currently done in PROV-O with using
> the (no longer time:) Instant resources - here you would just say:
>
>
> :e prov:wasGeneratedAt :t1 .
> :e prov:wasGeneratedAt :t2 .
>
> :t1 prov:inXSDDateTime "2001-10-26T21:32:52"^^xsd:dateTime .
> :t2 prov:inXSDDateTime "2011-11-27T21:23:34"^^xsd:dateTime .
>
> (or equivalent long-version with prov:Generation and prov:hadTemporalValue)
>
> prov:wasGeneratedAt is functional, because there is only one
> generation event of an entity. Thus we know that :t1 == :t2 and can
> limit our trouble-search there - in this case we could get an OWL
> error because prov:inXSDDateTime is a functional datatype, or not,
> depending on how strongly we choose to believe the owl:sameAs.
>
>
>


-- 
-----------  ~oo~  --------------
Paolo Missier - Paolo.Missier@newcastle.ac.uk, pmissier@acm.org
School of Computing Science, Newcastle University,  UK
http://www.cs.ncl.ac.uk/people/Paolo.Missier
Received on Tuesday, 7 February 2012 14:06:04 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 13:06:56 GMT