Re: What is an entity?

Dear Jacco,

Thanks for your quick and precise feedback! We have created ISSUE-462
for your comment which you can find at:

https://www.w3.org/2011/prov/track/issues/462

We will now discuss in the Working Group how to address your comment.
To not swamp you with email, we'll do this on our internal mailing
list and get back to you with either a resolution or clarification
questions.

You can always check where we are at by looking at the issue above.

Thanks,
Paul


On Tue, Jul 24, 2012 at 7:16 PM, Jacco van Ossenbruggen
<Jacco.van.Ossenbruggen@cwi.nl> wrote:
> Dear WG,
>
> Congrats on publishing your last call documents!
>
> I've one comment, which is about the definition of what and entity is,
> and especially, what it is not, in the context of PROV.
> The current definition is:
>
> "An entity is a physical, digital, conceptual, or other kind of thing
> with some fixed aspects; entities may be real or imaginary."
>
> I'm confused about the "some fixed aspects" part.  What does that mean?
> In general, this definition is quite liberal, while the rest of the text
> suggests that you assume a much more constrained definition of what an
> entity should be.
>
>  From the remainder of the text, I derive that the "some fixed aspects"
> should at least include that
> - an entity is something with a fixed, permanent ID
> - an entity is something of which all (not some!) aspects that the
> provenance records make claims about are assumed to remain fixed
>
> E.g. the 5.2.2 revision example suggest that
> - if Alice makes a file which is later changed by Bob, and
> - if both contributions are to be recorded in the PROV records, that
> - than the two revisions of the file _need_ to be modelled by two
> different entities.
>
> Your current entity definition leaves sufficient room for assuming that
> entities may change over time without changing ids.
>
> I think that for people to decide how to use PROV in their applications,
> the crucial decision is to decide which local application objects should
> be mapped to PROV entities.  I do not think the current liberal
> definition is very helpful in this process. For example from your
> definition, a database table that is dynamically updated could be an
> entity, while I think the assumption of PROV is that in such
> applications, a specific snapshot of the table would be an entity.
>
> What about this def:
>
> "An entity is a physical, digital, conceptual, or other kind of thing
> with a fixed id and a fixed set of properties; entities may be real or
> imaginary, but are assumed not to evolve over time".
>
> Hope this helps,
>
> Jacco van Ossenbruggen
>



-- 
--
Dr. Paul Groth (p.t.groth@vu.nl)
http://www.few.vu.nl/~pgroth/
Assistant Professor
- Knowledge Representation & Reasoning Group |
  Artificial Intelligence Section | Department of Computer Science
- The Network Institute
VU University Amsterdam

Received on Wednesday, 25 July 2012 07:23:03 UTC