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

Re: PROV-ISSUE-549: XML Schema Bundle vs. NamedBundle [XML Serialization]

From: Luc Moreau <l.moreau@ecs.soton.ac.uk>
Date: Thu, 13 Sep 2012 09:23:50 +0100
Message-ID: <EMEW3|fad2789254eca1d750b845f055ba1e73o8C9Nr08l.moreau|ecs.soton.ac.uk|50519816.2000700@ecs.soton.ac.uk>
To: "Hua, Hook (388C)" <hook.hua@jpl.nasa.gov>
CC: "public-prov-wg@w3.org" <public-prov-wg@w3.org>

Hi Hook,

Good question!  Long history.

Bundles (used to be call accounts) could be arbitrarily nested in an 
earlier design.
There was a strong push back, in part due to the fact that arbitrary 
nesting would
cause extra complexity for the mapping to rdf and a flavor of named graphs.


Luc

On 09/13/2012 09:11 AM, Hua, Hook (388C) wrote:
> Hi Luc,
>
> I'm curious what was the rationale for not wanting arbitrary levels of
> nesting of bundles?
>
> In scientific workflows, nested workflows provide a useful mechanism to
> hide the level of detail at arbitrary levels. Wouldn't that be
> conceptually similar to arbitrary levels of nested bundles?
>
> Following example 39 in
> http://www.w3.org/TR/prov-dm/#anexample-provenance-of-provenance
> I can see that not having arbitrary levels of bundle nesting would keep
> the provenance of provenance simpler, say, to capturing first order
> observations.
>
>
> --Hook
>
>    
>
>
>
>
> On 9/10/12 9:36 AM, "Luc Moreau" <l.moreau@ecs.soton.ac.uk> wrote:
>
>> Hi Curt,
>>
>> Following discussion in prov-constraints/prov-n, we are going to rename
>> the toplevel bundle
>> a prov document, wheres the bundle (named set of statements) should
>> remain bundle.
>>
>> If we follow the same terminology, old:NamedBundle should become
>> new:Bundle
>> and old:Bundle should become new:Document.
>>
>> BTW Your suggestion would allow arbitrary levels of nesting of bundles,
>> which we don't want.
>>
>>
>> Cheers,
>> Luc
>>
>>
>> On 09/10/2012 05:32 PM, Provenance Working Group Issue Tracker wrote:
>>> PROV-ISSUE-549: XML Schema Bundle vs. NamedBundle [XML Serialization]
>>>
>>> http://www.w3.org/2011/prov/track/issues/549
>>>
>>> Raised by: Curt Tilmes
>>> On product: XML Serialization
>>>
>>> What is the rationale for distinguishing Bundle vs. NamedBundle in the
>>> current XSD Schema?
>>>
>>> Why not just have an optional prov:id on Bundle and just have one type?
>>>
>>>
>>>
>>>
>> -- 
>> 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
>>
>>

-- 
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 Thursday, 13 September 2012 08:24:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 13 September 2012 08:24:33 GMT