W3C home > Mailing lists > Public > public-prov-wg@w3.org > July 2011

Re: PROV-ISSUE-30 (name-for-bob): What name do we use for the BOB construct? [Conceptual Model]

From: Luc Moreau <L.Moreau@ecs.soton.ac.uk>
Date: Thu, 21 Jul 2011 23:34:44 +0100
Message-ID: <EMEW3|2140ebe4f1d149cbdba065b10a5013a9n6KNYx08L.Moreau|ecs.soton.ac.uk|4E28A984.4050500@ecs.soton.ac.uk>
To: Jim McCusker <mccusj@rpi.edu>
CC: Paul Groth <p.t.groth@vu.nl>, "public-prov-wg@w3.org" <public-prov-wg@w3.org>
Hi Jim,

To echo Paul, could you illustrate what you need with the example in the 
document.

What is pil:describes?

It's also a first to me that we have both pil: Entity *and* pil:BOB.
Again, can their need be justified in the example?

Regards,
Luc


On 21/07/11 22:33, Paul Groth wrote:
> Hi Jim,
>
> I think the key thing to do would be to formulate your point in terms of the document put out.
>
> Paolo and Luc have tried to rectify definitions and in the process made a particular choice in how to describe the model. This may differ from prior assumptions about the model.
>
> Thanks
> Paul
>
> On Jul 21, 2011, at 22:56, Jim McCusker<mccusj@rpi.edu>  wrote:
>
>    
>> I guess what I'm arguing (and have been) is that the concept of Entity
>> needs a corresponding class in the model. Anything that is described
>> by a BOB is an Entitiy. Let's say that
>> http://tw.rpi.edu/person/JamesMcCusker represents me on the semantic
>> web, and a description of me might exist somewhere. Let's say it's an
>> RDF document that sits out on the web, and is called
>> http://tw.rpi.edu/instances/JamesMcCusker. That document refers to my
>> URI, and can be said in some way to describe me. We should be able to
>> say the following:
>>
>> http://tw.rpi.edu/instances/JamesMcCusker pil:describes
>> http://tw.rpi.edu/person/JamesMcCusker.
>>
>> That would in turn imply the following:
>>
>> http://tw.rpi.edu/instances/JamesMcCusker a pil:BOB.
>> http://tw.rpi.edu/person/JamesMcCusker a pil:Entitiy.
>>
>> We can now assert provenance based on that BOB, which is tied to a
>> pil:Entity that represents me.
>>
>> Jim
>>
>> On Thu, Jul 21, 2011 at 4:23 PM, Luc Moreau<L.Moreau@ecs.soton.ac.uk>  wrote:
>>      
>>> Jim,
>>> Can you relate to the document. What is a pil:entity?  This construct does not exist.
>>>
>>> What do you mean by Bob does not represent ...? The definition says it's a representation.
>>>
>>> I am confused.
>>>
>>> Luc
>>>
>>> Professor Luc Moreau
>>> Electronics and Computer Science
>>> University of Southampton
>>> Southampton SO17 1BJ
>>> United Kingdom
>>>
>>> On 21 Jul 2011, at 21:10, "Jim McCusker"<mccusj@rpi.edu>  wrote:
>>>
>>>        
>>>> I think we're still going around in circles.
>>>>
>>>> Entity: A thing in the world, can be represented by, for instance, a
>>>> URI. That URI, in PIL, is a pil:Entity.
>>>>
>>>> BOB: A description of an entity constrained by context (including time
>>>> and place). The description is not the entity, even within our
>>>> information representation. A BOB must be able to refer to something.
>>>> That BOB is a description of an entity, but does not REPRESENT the
>>>> entity.
>>>>
>>>> Jim
>>>>
>>>> On Thu, Jul 21, 2011 at 3:45 PM, Paolo Missier<Paolo.Missier@ncl.ac.uk>  wrote:
>>>>          
>>>>> Hi,
>>>>>
>>>>> I would advise against using the same terms with different typographical
>>>>> convention :-)
>>>>>
>>>>> what the document perhaps needs to clarify more upfront is that there is a
>>>>> "real world" and then a model of it, and the constructs of the language are
>>>>> about the model. It does say that but perhaps not strongly enough.
>>>>> - Characterized entity belongs in the world
>>>>> - BOBs belong in the data model that is a representation of the world.
>>>>> These two levels are never conflated.
>>>>>
>>>>> The good old "record linkage" community (data quality in databases) never
>>>>> had any qualms about using "real-world entities", as in "reconciling
>>>>> different records (BOBs?) that represent the same real-world entity".
>>>>> In their world (pun intended :-)), a record is a very concrete data
>>>>> structure that sits in a data store and you can display on a screen.
>>>>> Now, we cannot use "record", we have ruled out "information(al) resource"...
>>>>> but isn't that basically the territory?
>>>>> Entity representation?
>>>>>
>>>>> -Paolo
>>>>>
>>>>>
>>>>> On 7/21/11 8:33 PM, Luc Moreau wrote:
>>>>>
>>>>> Hi Khalid,
>>>>> OK.  This said we have over 20 occurrences of "characterized entity" in the
>>>>> text.
>>>>>
>>>>> We can't simply use the "expansion" everywhere. Having some terminology is
>>>>> desirable.
>>>>>
>>>>> Do you have a suggestion?
>>>>>
>>>>> We could also go for a typographic difference:
>>>>>    BOB ->  CharacterizedEntity
>>>>> and we keep 'characterized entity' elsewhere.
>>>>>
>>>>> Luc
>>>>>
>>>>> On 21/07/2011 20:27, Khalid Belhajjame wrote:
>>>>>
>>>>> Hi Luc,
>>>>>
>>>>> I guess I used the wrong term, "interchangeable". I guess that what I meant
>>>>> is that "Characterized Entity" can be considered as a candidate to replace
>>>>> "BOB". Of course, in that case, we will need to avoid the usage of the the
>>>>> term "characterized entity" in the core of the definition. E.g., we can use
>>>>> the following definition:
>>>>>
>>>>> A "Characterized Entity" is a description of the situation of an entity in
>>>>> the world.
>>>>>
>>>>> Or something in these lines.
>>>>>
>>>>> Thanks, khalid
>>>>>
>>>>> On 21/07/2011 19:54, Luc Moreau wrote:
>>>>>
>>>>> Hi Khalid,
>>>>>
>>>>> As far as I know, they are *not* interchangeable. One is the language
>>>>> construct, the other is "in the world".
>>>>>
>>>>> cf. definition:
>>>>>
>>>>> A BOB represents an identifiable
>>>>> characterized entity.
>>>>>
>>>>> Should we go for "Characterized Entity", we need a typographic
>>>>> convention to distinguish between
>>>>>
>>>>> the construct and the world-thing, otherwise, the reader will never
>>>>> know whether this is language construct
>>>>>
>>>>> or not.
>>>>>
>>>>>
>>>>> Luc
>>>>>
>>>>> On 21/07/2011 19:45, Khalid Belhajjame wrote:
>>>>>
>>>>> In the Provenance Model initial draft, the terms "Bob" and "characterized
>>>>> entity" are used interchangeably.
>>>>> Characterized entity seems then to be a candidate for replacing BOB.
>>>>>
>>>>> Thanks, khalid
>>>>>
>>>>> On 21/07/2011 19:30, Provenance Working Group Issue Tracker wrote:
>>>>>
>>>>> PROV-ISSUE-30 (name-for-bob): What name do we use for the BOB construct?
>>>>> [Conceptual Model]
>>>>>
>>>>> http://www.w3.org/2011/prov/track/issues/30
>>>>>
>>>>> Raised by: Luc Moreau
>>>>> On product: Conceptual Model
>>>>>
>>>>> How do we call the construct referred to as BOB.  "BOB" was introduced as a
>>>>> placeholder at F2F1. Before F2F1, we use to refer to it as thing.
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> -----------  ~oo~  --------------
>>>>> Paolo Missier - Paolo.Missier@newcastle.ac.uk, pmissier@acm.org
>>>>> School of Computing Science, Newcastle University,  UK
>>>>> http://www.cs.ncl.ac.uk/people/Paolo.Missier
>>>>>
>>>>>            
>>>>
>>>>
>>>> --
>>>> Jim
>>>> --
>>>> Jim McCusker
>>>> Programmer Analyst
>>>> Krauthammer Lab, Pathology Informatics
>>>> Yale School of Medicine
>>>> james.mccusker@yale.edu | (203) 785-6330
>>>> http://krauthammerlab.med.yale.edu
>>>>
>>>> PhD Student
>>>> Tetherless World Constellation
>>>> Rensselaer Polytechnic Institute
>>>> mccusj@cs.rpi.edu
>>>> http://tw.rpi.edu
>>>>
>>>>          
>>>
>>>        
>>
>>
>> -- 
>> Jim
>> --
>> Jim McCusker
>> Programmer Analyst
>> Krauthammer Lab, Pathology Informatics
>> Yale School of Medicine
>> james.mccusker@yale.edu | (203) 785-6330
>> http://krauthammerlab.med.yale.edu
>>
>> PhD Student
>> Tetherless World Constellation
>> Rensselaer Polytechnic Institute
>> mccusj@cs.rpi.edu
>> http://tw.rpi.edu
>>
>>      
>    
Received on Thursday, 21 July 2011 22:35:36 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 13:06:37 GMT