Re: Workflow Example in Formal Model HTML document


You say that "A provenance container is a house-keeping construct of PROV-DM", 
which might be a way of saying that a provenance container is a construct of the 
provenance ASN, which does not directly denote anything in the real world.

Entities can be (probably often are) constructs that convey information encoded 
to some language or format (an HTML page, a Word document, etc.).  So an entity 
can *be* a a provenance ASN construct (e.g. represented as RDF).

In this way, a provenance expression can refer to some other provenance 
expression as an entity, without having to explicitly expose the container as a 
special case in the domain ontology.

Or to put it another way, a provenance container is a specialization of Entity 
to exactly the same extent that a web page or a word document is a 
specialization of Entity.

Does this help?


I think this is consistent with luc's response.


On 29/09/2011 03:10, Satya Sahoo wrote:
> Hi Luc,
> We were not able to reach an agreement on how ProvenanceContainer is not a
> specialized type of Entity during our ontology call on Monday due to time
> constraints.
> To help better understand the differences and similarities, I copied the two
> definition from PROV-DM to two documents and tried to compare them
> side-by-side. The following are the two definitions:
> ===Entity====
> In PROV-DM, an entity expression is a representation of an identifiable
> characterized thing.
> ===ProvenanceContainer===
> A provenance container is a house-keeping construct of PROV-DM, also capable
> of bundling PROV-DM expressions. A provenance container is not an
> expression, but can be exploited to return all the provenance assertions in
> response to a request for the provenance of something ([PROV-PAQ]).
> According to the two definitions, a provenance container can be an
> "identifiable characterized thing" (not being an expression is not a
> conceptual constraint). Also, the ability to return all provenance
> assertions in response can be applied to an Agent also - similar to a
> software agent returning the current stock market quotes.
> Further, if an Entity "contains" provenance assertions it can still be an
> "identifiable characterized thing" thereby satisfying our current definition
> of Entity.
> During our ontology telcon today Paolo explained that the primary difference
> between Entity and Provenance Container is that Provenance Container can
> "contain" provenance assertions while Entity are assumed not to contain
> assertions. But, this seems to be an application-specific requirement.
> For example, for a person writing a 3-page letter the three pages will be
> instances of Entity and the envelope containing the three pages will be a
> container. But for the postal service personnel, who deal with thousands of
> envelopes per day, the envelope is an Entity (and a sack for transporting
> the envelopes will be a container).
> Hence, I believe the difference between what thing is a ProvenanceContainer
> or an Entity is an application-specific perspective/requirement and there is
> no fundamental difference between the two terms - except that Provenance
> terms seems to be a specialized form an Entity in the sense that Provenance
> Container contains provenance assertions, while an Entity may or may
> notcontain provenance assertions.
> Paolo suggested that we should bring up this issue to the WG mailing list -
> hence I am cc'ing the mailing list also.
> Thanks.
> Best,
> Satya
> On Wed, Sep 28, 2011 at 3:58 AM, Luc Moreau<>wrote:
>> **
>> Hi,
>> I thought we had already discussed this, but I see location as subtype of
>> entity.
>> Same issue as with provenance container.  This is not a subtype of entity.
>> Luc
>> --
>> Professor Luc Moreau
>> Electronics and Computer Science   tel:   +44 23 8059 4487
>> University of Southampton          fax:   +44 23 8059 2865
>> Southampton SO17 1BJ               email:
>> United Kingdom           

Received on Thursday, 29 September 2011 10:50:20 UTC