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

Re: PROV-ISSUE-589: scope of prefixes in prov-n documents [prov-n]

From: Luc Moreau <l.moreau@ecs.soton.ac.uk>
Date: Tue, 06 Nov 2012 14:30:51 +0000
Message-ID: <EMEW3|6e374563d89b7e0ef6bbf96dbceb5f07oA5EUq08l.moreau|ecs.soton.ac.uk|50991F1B.80100@ecs.soton.ac.uk>
To: public-prov-wg@w3.org
Dear all,

In addition to the changes made on Friday, I have made the following
changes, avoiding the sentence "bundles are self-contained".

http://dvcs.w3.org/hg/prov/rev/029c773b81a3


I am closing this issue pending review. Feedback away of the F2F meeting 
appreciated.

Regards,
Luc

On 11/02/2012 04:45 PM, Luc Moreau wrote:
> Hi all,
>
> I have committed some changes related to the scoping of 
> prefix-namespace declarations.
>
> http://dvcs.w3.org/hg/prov/rev/9a123c41e1b5
>
> Feedback is welcome!
> Regards,
> Luc
>
> On 01/11/2012 17:02, James Cheney wrote:
>> Hi,
>>
>> Just to clarify: There is some mention in the discussion of issue-496 
>> that the "excluding" scoping rule (where bundles must be 
>> self-contained) helps simplify prov-constraints.  I think they are 
>> orthogonal issues, since prov-constraints implicitly assumes that 
>> co-reference has been resolved before we start checking validity and 
>> so on.
>>
>> I think it is useful to illustrate the effect of the resolution today 
>> on a simple example:
>>
>> document
>>    prefix ex http://www.example.com
>>    bundle ex:b1
>>      entity(ex:e1)
>>    endBundle
>>    bundle ex:b2
>>      prefix ex http://www.example2.com
>>      entity(ex:e1)
>>    endBundle
>> endDocument
>>
>> With the excluding semantics, every bundle has to be self-contained, 
>> and so the example above is *illegal* in prov-n with "excluding" 
>> scoping, because b1 doesn't (re)declare namespace ex.
>>
>> After today's resolution, the above document is *legal*, provided we 
>> are clear that redeclarations at inner scopes are allowed and take 
>> precedence over any prior declarations.
>>
>> However, either way it is irrelevant to prov-constraints, which 
>> doesn't say anything about namespaces; implementations need to expand 
>> out prefixes in the usual way first.  The above document (if legal) 
>> is really shorthand for
>>
>> document
>>    bundle http://www.example.com/#b1
>>      entity(http://www.example.com/#e1)
>>    endBundle
>>    bundle http://www.example.com/#b2
>>      entity(http://www.example2.com/e1)
>>    endBundle
>> endDocument
>>
>> thus, avoiding the potential name clash between the two ex:e1's.
>>
>> Perhaps this should be clarified in prov-n, prov-constraints, or both.
>>
>> --James
>>
>> On Nov 1, 2012, at 4:42 PM, Provenance Working Group Issue Tracker 
>> wrote:
>>
>>> PROV-ISSUE-589: scope of prefixes in prov-n documents [prov-n]
>>>
>>> http://www.w3.org/2011/prov/track/issues/589
>>>
>>> Raised by: Luc Moreau
>>> On product: prov-n
>>>
>>>
>>>
>>> Change the scope of document prefixes in prov-n to include inner 
>>> bundles (unless they are re-declared).
>>>
>>> This is a follow-up of ISSUE-496
>>>
>>> See http://www.w3.org/2011/prov/meeting/2012-11-01#resolution_5
>>>
>>>
>>>
>>>
>>>
>>>
>>
>

-- 
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
United Kingdom                     http://www.ecs.soton.ac.uk/~lavm
Received on Tuesday, 6 November 2012 14:31:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 6 November 2012 14:31:23 GMT