Re: further comments on IPV

Luc

Agreed. You will see in follow on mails that (natural language) definitions on concepts is what we are committed to providing in time for the meeting. The model tf is working on it.
Paolo

-- 
p.m. scripsit. 
(yes, this was sent from a phone)

----- Reply message -----
From: "Luc Moreau" <L.Moreau@ecs.soton.ac.uk>
To: "public-xg-prov@w3.org" <public-xg-prov@w3.org>
Subject: further comments on IPV
Date: Mon, Jun 27, 2011 08:36


Hi Paolo, all,

I am catching up with emails sent for the last 5 days.

Your suggestion of formalizing definitions is good, but it is not what we were trying to do here.
We wanted to provide *natural language definitions* for all concepts, so that we begin to share
a common vocabulary.  Once we have this, we can then:
- define an ontology/schema for the terminology, so that we can represent provenance
- define a semantics
- iteratively, inform and refine the natural language definitions

Therefore, we need to be reasonable. We might argue for the next 6 months about the English definition
of a Thing.  What will it bring us?

In fact, we are in serious danger of making the face to face meeting unproductive, since we are not going
to have definitions of concepts.  I would argue that we should have "strawman" definitions for all
concepts, including version and collections (recognizing though that we will not have reached agreement by
then).

Regards,
Luc




On 06/23/2011 03:52 PM, Paolo Missier wrote:
Greetings,
  I am slowly catching up and I have decided to start from what is recorded in the wiki rather than trying to playback  long and winding threads.

sorry to vent, but I started with "Concept 'Invariant View or Perspective on a Thing'" and I am already utterly confused. For a member of the Model TF who is tasked with a synthesising and reconciling job, this is not good :-)
So I appended some comments of my own here, which are mostly questions:  http://www.w3.org/2011/prov/wiki/ConceptInvariantViewOnThing#Comment_by_PM__24_June_2011.3D

The main message is: I really feel the need for some precision, which doesn't mean formalising at all costs, but at least picking a reference framework for modelling: ER, objects, UML... something that has, er, a clear semantics that one can build upon! (ok, so perhaps UML does not qualify :-))

Regards, -Paolo



--
Professor Luc Moreau
Electronics and Computer Science   tel:   +44 23 8059 4487
University of Southampton          fax:   +44 23 8059 2865
Southampton SO17 1BJ               email: l.moreau@ecs.soton.ac.uk<mailto:l.moreau@ecs.soton.ac.uk>
United Kingdom                     http://www.ecs.soton.ac.uk/~lavm