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

Re: PROV-ISSUE-225 (objects-in-universe-of-discourse): What are the objects in the universe of discourse? [prov-dm]

From: Graham Klyne <GK@ninebynine.org>
Date: Tue, 24 Jan 2012 13:08:13 +0000
Message-ID: <4F1EAD3D.9020103@ninebynine.org>
To: James Cheney <jcheney@inf.ed.ac.uk>
CC: Satya Sahoo <satya.sahoo@case.edu>, Luc Moreau <L.Moreau@ecs.soton.ac.uk>, public-prov-wg@w3.org
FWIW, I broadly agree with the technical view James reports here - as I 
interpret it - with one possible exception: see below.

On 24/01/2012 10:41, James Cheney wrote:
> Hi,
>
> We discussed this issue at the call yesterday, and came to some agreement about which things in Luc's list are in the "universe (domain) of discourse".  I believe Satya is planning to summarize this.
>
> To simplify terminology, for the moment I will call something a DomainObject if it is in the domain of discourse.
>
> One concern I had was what exactly the consequences are of these decisions, because it seemed like different people had different notions in mind for what it means to be in the universe of discourse.  (Or at least, everyone seemed to have a different notion than I do.)
>
> * To me (with respect to PROV-SEM), the distinction is:
>
> DomainObjects are things that exist independently of observers or descriptions, but can be observed or described.
> Thus, syntax (e.g. entity records) are not DomainObjects; Entities, Events, Activities and Agents are.  I would argue that Things might also be DomainObjects.  The consequence of saying that they are not would be that we can't easily relate different Entities that are intuitively "alternate, fixed perspectives on the same changing Thing".
>
> * With respect to PROV-DM, I believe Luc/Paul were saying that we already have Entities to represent Things in the domain of discourse, so Entites are DomainObjects and Things are not.  The consequence is that we don't need to add a "thing(id,...)" syntax to declare or identify Things.
>
> * With respect to PROV-O, I believe Satya and others have a similar view, and the consequence is that we don't need to add a new class of Things.
>
> I think these perspectives are mutually reconcilable, as long as it's OK for the formal semantics to talk about Things as a model of the underlying reality that Entities are trying to describe/surrogate.

It seems to me:  if entities subsume things, then by talking about entities we 
also talk about things.  I'd hold off distinguishing things unless there's a 
clear need (e.g. like the now-discontinued discussion of a axiomatic model for 
altermnativeOf to be transitive).

#g
--

> On Jan 19, 2012, at 10:48 PM, Satya Sahoo wrote:
>
>> Sure we will put it on our agenda for the Monday call.
>>
>> Thanks.
>>
>> Best,
>> Satya
>>
>> On Thu, Jan 19, 2012 at 5:38 PM, Luc Moreau<L.Moreau@ecs.soton.ac.uk>  wrote:
>> Dear Satya, Mike, Prov-o team
>> May I invite you to discuss this at your call on Monday.
>> Thanks,
>> Luc
>>
>>
>> On 19/01/12 22:35, Provenance Working Group Issue Tracker wrote:
>> PROV-ISSUE-225 (objects-in-universe-of-discourse): What are the objects in the universe of discourse? [prov-dm]
>>
>> http://www.w3.org/2011/prov/track/issues/225
>>
>> Raised by: Luc Moreau
>> On product: prov-dm
>>
>>
>> On Jan 19 2012, we agreed on a proposal related to identifiers and objects in the universe of discourse [1]:
>>
>> "*All* objects of discourse ("entities") MUST be identifiable by all participants in discourse. Object descriptions ("entity records" and otherwise) SHOULD use an unambiguous identifier (either reusing an existing identifier, or introducing a new identifier) for the objects described." (intent)
>>
>> To make progress on the front of identifiers (ISSUE-183, ISSUE-215), we agreed that we would discuss what the objects of the universe of discourse are. The purpose of this issue is to address this question.
>>
>> Here are a list of object types we can find in the document. Note that some have not been exactly named as such. Feel free to add any missing here.
>>
>>
>> Thing
>>
>> Entity (including its subtypes Agent, Plan, Collection, ...)
>>
>> Activity
>>
>> Note
>>
>> Events: Entity Usage event, Entity Generation Event, Activity Start Event, Activity End event
>>
>> Derivation
>>
>> Association (between activity and agent, cf wasAssociatedWith)
>> and its subtypes:
>>    StartAssociation (wasStartedBy)
>>    EndAssociation (wasEndedBy)
>>
>>
>> "ResponsibilityChain" (between agent and agent, wrt activity, cf actedOnBehalfOf)
>>
>>
>> Specialization (specializationOf)
>>
>> Alternative (alternateOf)
>>
>> Annotation (hasAnnotation)
>>
>> Traceability (tracedTo)
>>
>> ActivityOrdering (wasInformedBy and wasStartedBy)
>>
>> Revision (wasRevisionOf)
>>
>> Attribution (wasAttributedTo)
>>
>> Quotation (wasQuoteFrom)
>>
>> Summary (wasSummaryOf)
>>
>> Original Source (hadOriginalSource)
>>
>> CollectionAfterInsertion/CollectionAfterRemoval
>>
>> Event ordering constraint (e.g  start event precedes the end event)
>>
>> Records
>>   e.g. Entity Record (as opposed to Entity listed above)
>>
>> In particular:
>>
>> Account Record
>>
>> Is there a notion of Account (as opposed to account record)
>>
>> Record container
>>
>> Attribute
>>
>>
>>
>>
>>
>> [1] http://www.w3.org/2011/prov/meeting/2012-01-19#resolution_2
>>
>>
>>
>>
>>
>>
>
>
>
>
> The University of Edinburgh is a charitable body, registered in
> Scotland, with registration number SC005336.
Received on Tuesday, 24 January 2012 13:11:04 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:58:11 UTC