RE: PROV-ISSUE-651: 3 comments on prov-o (wasDerived, wasAttrib, actedOnBehalf)

Hello,

The response looks good. I might add:

An activity having used an entity and generated another, does not necessarily mean that the latter entity was derived from the former. So adding a dummy activity actually conveys less information than wasDerivedFrom.

It is important that provenance describes the past, regardless of the level of detail expressed. actsOnBehalfOf would incorrectly imply that the delegation relationship holds currently. This might or might not be the case, but is not part of provenance, so not expressible in PROV.

thanks,
Simon

Dr Simon Miles
Senior Lecturer, Department of Informatics
Kings College London, WC2R 2LS, UK
+44 (0)20 7848 1166

Efficient Multi-Granularity Service Composition:
http://eprints.dcs.kcl.ac.uk/1396/
________________________________
From: Timothy Lebo [lebot@rpi.edu]
Sent: 14 March 2013 13:57
To: Provenance Working Group
Subject: Re: PROV-ISSUE-651: 3 comments on prov-o (wasDerived, wasAttrib, actedOnBehalf)

prov-wg,

I've drafted a response to Jacobo's comments at:

http://www.w3.org/2011/prov/wiki/ResponsesToPublicCommentsPR#ISSUE-651

Any comments welcome.

Regards,
Tim

On Mar 14, 2013, at 9:13 AM, Provenance Working Group Issue Tracker <sysbot+tracker@w3.org<mailto:sysbot+tracker@w3.org>> wrote:

PROV-ISSUE-651: 3 comments on prov-o (wasDerived, wasAttrib, actedOnBehalf)

http://www.w3.org/2011/prov/track/issues/651

Raised by: Timothy Lebo
On product:

http://www.w3.org/mid/CABC+A3LHXh1r1BHqZR7eXqes8dRFKVg3A6oBB0F=mgUimL6r8g@mail.gmail.com

Hello,

I have been looking at PROV-O and I intend to use or extend it for a
project where we need provenance metadata for named rdf graphs. I have
a “triple” of suggestions that come from vocabulary restrictions I
have thought of for my own use, but since I see the vocabulary is
still in a CR stage, I have decided to expose them for the case they
might be included in the base vocabulary. However, I must say I am
quite a newcomer into RDF and related technologies, so I might well be
very wrong.

1. I wonder if instead of the wasDerivedFrom property, a dummy
instance of Activity could always be used to connect the original and
obtained entities, even without further properties. This would make
modeling more homogenous, which might make things easier for automated
tools, and would be straightforward to add information about the
activity if it was discovered in a later stage, without the need of
removing triples. I think these advantages and the reduction of the
vocabulary make up for the overhead in extra nodes.

2. The existence of wasAttributedTo seems unnecessary to me, as we
already can express the same with wasAssociatedWith from the Activity
that led to the Entity. I am aware that without cardinality
constraints between Activity and Entity, an activity can generate
several Entities and therefore an Agent involved in an Activity is not
necessarily involved in one of its generated Entities. But maybe this
would be a reason to consider introducing cardinality constraints, as
activities that generate several Entities can usually be divided into
more specific Activities that only lead to one Entity. So Activities
that generate several Entities could be modeled as a higher level
resource that aggregates several activities. I know in this way you
remove a term to introduce another one, but you get rid of the
semantic overlapping and possible redundancy between wasAttributedTo
and wasAssociatedWith.

3. The unqualified relation actedOnBehalfOf, since it is independent
of the activity, it becomes a general or "atemporal" property of the
agent and should be better named actsOnBehalfOf.

Best regards,
Jacobo Rouces.