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

Re: Definitions and provenance and invariance

From: Pgroth <pgroth@gmail.com>
Date: Sun, 26 Jun 2011 22:51:15 +0200
Message-Id: <75010C30-5EC6-4C52-885D-D4FA54252291@gmail.com>
Cc: "public-prov-wg@w3.org" <public-prov-wg@w3.org>
To: Paolo Missier <Paolo.Missier@ncl.ac.uk>
Yes this would be good to clarify.

My understanding was that for a particular invariant view from an observer, the invariant properties and the values those properties take must remain the same.

Is that correct or am I missing something?

Thanks
Paul

Sent from my iPad

On Jun 26, 2011, at 8:06 PM, Paolo Missier <Paolo.Missier@ncl.ac.uk> wrote:

> once again, I think I need a clarification of the difference between a /property/ being invariant (valid within a scope) and its /value/ being constant (within a scope, interval, etc).
> 
> In the File example  http://www.w3.org/2011/prov/wiki/ConceptInvariantViewOnThing#Comment_by_Stian__24_June_2011
> Stian wrote:  "The Printout content is an invariant property, but a variant property of the Document."
> 
> This seems to refer to the value of property "content" (which will not change for the Printout), rather than to the property itself?
> 
> -Paolo
> 
> On 6/26/11 3:00 PM, Paul Groth wrote:
>> 
>> Hi Simon,
>> 
>> Thanks for the clarification. That was what I thought but just wanted to 
>> check.
>> 
>> cheers,
>> Paul
>> 
>> Simon Miles wrote:
>>> Paul - As I understood the intent, the provenance of a thing would be
>>> assertions about the invariant properties of that thing, yes. You
>>> couldn't say much about the variant properties other than that they're
>>> variant. However, it may be unnecessary to explicitly "declare"
>>> properties to be invariant.
>>> 
>>> Paolo - I agree that we shouldn't mix up fixed assertions on the
>>> values of properties with those properties being invariant. Though I
>>> don't think anyone has suggested the former - even if properties were
>>> declared to be invariant (as I think Paul's question implied), that
>>> would not be the same as declaring what their values were, let alone
>>> not allowing those declarations to change.
>>> 
>>> Thanks,
>>> Simon
>>> 
>>> On 26 June 2011 12:12, Paolo Missier<Paolo.Missier@ncl.ac.uk>  wrote:
>>>> I would say you can talk about the provenance of that view. But, to be clear, the fact that some properties are "declared" to be
>>>> fixed (constant), for example by saying that "the creator is {invariant, immutable, constant,...}",  should have nothing to do with
>>>> provenance.  It's in the semantics of the "thing" and its properties, I hope we agree on this? In a data model, one would enforce a
>>>> constraint to avoid "creator" being updated, to reflect the intended semantics of the property. Is this what we mean by "fixed"?
>>>> And even then, that's assuming too much already: updates to "creator" may be acceptable if the original value is found to be wrong,
>>>> for example.
>>>> 
>>>> -Paolo
>>>> 
>>>> On 6/26/11 9:56 AM, Pgroth wrote:
>>>>> For my clarification, once an asseter  has an invariant view of some stuff the asserter can talk about the stuff's provenance with respect to the set of properties that have been declared fixed?
>>>>> 
>>>>> Thanks
>>>>> Paul
>>>> 
>>>> ______________________________________________________________________
>>>> This email has been scanned by the MessageLabs Email Security System.
>>>> For more information please visit http://www.messagelabs.com/email
>>>> ______________________________________________________________________
>>>> 
>>> 
>> 
> 
Received on Sunday, 26 June 2011 20:33:16 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 13:06:32 GMT