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

Re: Definitions and provenance and invariance

From: Simon Miles <simon.miles@kcl.ac.uk>
Date: Mon, 20 Jun 2011 13:45:14 +0100
Message-ID: <BANLkTin=e4rhSEEVbxU-abgCFuzHsO_EmQ@mail.gmail.com>
To: Provenance Working Group WG <public-prov-wg@w3.org>
Luc,

I agree with Graham here. I think it's just a matter of wording,
nothing fundamental, but could lead to people misunderstanding the
model.

I think the example you give of "identity" is what most people (or at
least I) would call a "distinguishing feature". Being integral to
identity is, colloquially, a stronger requirement: it suggests that
the property is important to understanding what that thing is, not
merely one that distinguishes the thing from other things. You could
argue that all distinguishing features are important from some
perspective, but I suggest that is not how "identity" is commonly
understood.

As with Graham, I think we're more likely to get correct usage of the
model if we say nothing at all. I don't see a case for separating
distinguishing features from invariant properties in general.

Thanks,
Simon

On 20 June 2011 12:52, Luc Moreau <L.Moreau@ecs.soton.ac.uk> wrote:
> You may be right,  can you explain how you understand it?
> Luc
>
> On 06/20/2011 11:22 AM, Graham Klyne wrote:
>> It seems we understand the phrase "integral to identity" somewhat
>> differently, so that's a different reason not use it as part of the
>> definitions of "things" and "invariant views".  The more you say, the
>> more room there is for disagreement ...
>>
>> #g
>> --
>>
>> Luc Moreau wrote:
>>> Hi Simon and Graham,
>>>
>>> I added a response to Graham's comment on invariant property and
>>> identity.
>>> See
>>> http://www.w3.org/2011/prov/wiki/ConceptInvariantViewOnThing#Comments
>>>
>>> Cheers,
>>> Luc
>>>
>>> On 06/19/2011 12:18 PM, Simon Miles wrote:
>>>> Graham,
>>>>
>>>> OK, thanks for the clarification. I agree with your point, and am also
>>>> sympathetic to your discomfort with everything invariant being
>>>> "integral to identity".
>>>>
>>>> Thanks,
>>>> Simon
>>>>
>>>>
>>>> On 17 June 2011 23:00, Graham Klyne<GK@ninebynine.org>  wrote:
>>>>> Simon Miles wrote:
>>>>>> Graham, Stian, all,
>>>>>>
>>>>>> I might be confused, but this seems a more complex model than the one
>>>>>> proposed by Jim and Luc.  Why do we need to both a Dynamic Resource
>>>>>> and a View Resource?  I can't see any meaningful difference between
>>>>>> them either in Graham's definition or Stian's (helpful) concrete
>>>>>> example.  What is the point of saying anything about a mutable
>>>>>> property, e.g. "content of DynamicResource i0", when any assertion of
>>>>>> a mutable property's value will not always hold anyway?
>>>>> Speaking for myself... I used the terms "Dynamic" and "View" as
>>>>> labels to
>>>>> distinguish their roles in the structure given.  I would not choose
>>>>> to model
>>>>> them as different types.
>>>>>
>>>>> My point, expressed in terms of Stian's example, is that the notion
>>>>> we have been
>>>>> calling IVP is present in the viewOf relation rather than inherent
>>>>> in the
>>>>> resources themselves.  This was my point, which I think is also at
>>>>> the heart of
>>>>> the proposal by Jim and Luc.
>>>>>
>>>>> I happen to think that the definition as proposed in the wiki at
>>>>> http://www.w3.org/2011/prov/wiki/ConceptInvariantViewOnThing#ACCEPTED_at_teleconference_2011-06-16
>>>>>
>>>>> is over-specified (I've added some comments there).  But having
>>>>> expressed that
>>>>> reservation, I'm content to let them stand pro tem for the purposes
>>>>> of discussion.
>>>>>
>>>>> #g
>>>>> --
>>>>>
>>>>>
>>>>>> On 16 June 2011 15:39, Stian Soiland-Reyes
>>>>>> <soiland-reyes@cs.manchester.ac.uk>  wrote:
>>>>>>> On Tue, Jun 14, 2011 at 12:09, Graham Klyne<GK@ninebynine.org>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> Suppose that the "Dynamic resource has a number of different
>>>>>>>> observable
>>>>>>>> properties, some of which do not change over time, and others
>>>>>>>> which do.
>>>>>>>>   Then the View resource would be a resource for with a similar
>>>>>>>> set of
>>>>>>>> properties such that do not change over time, but correspond to
>>>>>>>> the dynamic
>>>>>>>> resource properties at a given time (including properties that
>>>>>>>> do not change
>>>>>>>> over time).  If the Dynamic resource does not change over time,
>>>>>>>> then it may
>>>>>>>> also serve as its own view resource:  the has view property can be
>>>>>>>> reflexive.
>>>>>>>>
>>>>>>>> The provenance resource is an assertion about the properties of
>>>>>>>> the view
>>>>>>>> resource.  I believe the key requirement that we try to capture
>>>>>>>> is that the
>>>>>>>> properties about which the provenance resource makes assertions are
>>>>>>>> invariant - there is no assertion in the provenance resource
>>>>>>>> which is not
>>>>>>>> always true of the view resource.
>>>>>>> This is a very beautifully simple model which I think we should keep
>>>>>>> in mind before digging too much into the exciting discussions.
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> "simplified" for the File example:
>>>>>>>
>>>>>>>
>>>>>>> :i0 a :DynamicResource ;
>>>>>>>   :name "/home/towns.txt" ;
>>>>>>>   :content [ :bytes "" ] ;
>>>>>>>   :creator :Alice .
>>>>>>>
>>>>>>>
>>>>>>> :i0View a :ViewResource ;
>>>>>>>    :viewOf :i0 ;
>>>>>>>    :name "/home/towns.txt" ;
>>>>>>>    :creator :Alice .
>>>>>>>
>>>>>>>
>>>>>>> # Metadata stored in filesystem
>>>>>>> :i0Provenance a :ProvenanceResource ;
>>>>>>>    :provenanceOf :i0View ;
>>>>>>>    :account :FileSystem ;
>>>>>>>    :processes (
>>>>>>>       [  :agent :Alice ;
>>>>>>>          :location :server1 ;
>>>>>>>          :process :fileCreation ;
>>>>>>>          :time "2011-06-15 18:00:01 UTC"  ]
>>>>>>>     ) .
>>>>>>>
>>>>>>> # however the log file claims the file was created on her
>>>>>>> workstation
>>>>>>> (not server), and 1 second later (clocks out of sync?)
>>>>>>>
>>>>>>> :i0Provenance2 a :ProvenanceResource ;
>>>>>>>    :provenanceOf :i0View ;
>>>>>>>    :account :ServerLogFile ;
>>>>>>>    :processes (
>>>>>>>       [  :agent :Alice ;
>>>>>>>          :location :AliceWorkstation;
>>>>>>>          :process :fileCreation ;
>>>>>>>          :time "2011-06-15 18:00:02 UTC"  ]
>>>>>>>     ) .
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> ### New graph - Content changed
>>>>>>>
>>>>>>> :i0 a :DynamicResource ;
>>>>>>>   :content [ :bytes "New York\nLos Angeles\n"  ] ;
>>>>>>>   :name "/home/towns.txt" ;
>>>>>>>   :creator :Alice ;
>>>>>>>   :readBy (:Alice :Bob :Charles :David)
>>>>>>>
>>>>>>>
>>>>>>> :i2 a :ViewResource ;
>>>>>>>   :viewOf :i0 ;
>>>>>>>   :name "/home/towns.txt" ;
>>>>>>>   :creator :Alice ;
>>>>>>>   :content [ :bytes "New York\nLos Angeles\n"  ] .
>>>>>>>
>>>>>>> :i2Provenance a :ProvenanceResource ;
>>>>>>>   :provenanceOf :i2 ;
>>>>>>>    :account :FileSystem ;
>>>>>>>    :processes (
>>>>>>>       [  :agent :Alice ;
>>>>>>>          :location :server1 ;
>>>>>>>          :process :fileCreation ;
>>>>>>>          :time "2011-06-15 18:00:03 UTC"  ]
>>>>>>> # Lost as :FileSystem metadata only keeps last-modified
>>>>>>> #     [   :agent :Alice ;
>>>>>>> #         :location :server1 ;
>>>>>>> #         :process :fileWrite ;
>>>>>>> #         :time "2011-06-15 18:00:03 UTC"  ]
>>>>>>>      [
>>>>>>>          #  :agent :Bob;  - not recorded as only owner/creator is
>>>>>>> kept
>>>>>>>          :location :server1 ;
>>>>>>>          :process :fileWrite ;
>>>>>>>          :time "2011-06-15 18:14:12 UTC"  ]
>>>>>>>     ) .
>>>>>>>
>>>>>>>
>>>>>>> So say there are additional mutable properties such as :readBy
>>>>>>> above -
>>>>>>> would you consider those propagating into the view as mutable
>>>>>>> properties? There could be another view over :i2 for the file before
>>>>>>> it was read by Charles, where :readBy is an immutable property.
>>>>>>>
>>>>>>> The example graph above does not distinguish between mutable and
>>>>>>> immutable properties - perhaps we shouldn't as they could be
>>>>>>> difficult
>>>>>>> to find, identify and measure.
>>>>>>>
>>>>>>> Here :readBy is not kept by neither the log file or file system
>>>>>>> and is
>>>>>>> a kind of conceptual property - it could be discovered by simply
>>>>>>> asking everyone who could have read it, or inferred from traced file
>>>>>>> usage, like if its sent in an email.
>>>>>>>
>>>>>>> --
>>>>>>> Stian Soiland-Reyes, myGrid team
>>>>>>> School of Computer Science
>>>>>>> The University of Manchester
>>>>>>>
>>>>>>>
>>>>>>> ______________________________________________________________________
>>>>>>>
>>>>>>> This email has been scanned by the MessageLabs Email Security
>>>>>>> System.
>>>>>>> For more information please visit http://www.messagelabs.com/email
>>>>>>> ______________________________________________________________________
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>> ______________________________________________________________________
>>>>> This email has been scanned by the MessageLabs Email Security System.
>>>>> For more information please visit http://www.messagelabs.com/email
>>>>> ______________________________________________________________________
>>>>>
>>>>
>>>>
>>>
>>
>
> --
> 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
>
>
>
> ______________________________________________________________________
> This email has been scanned by the MessageLabs Email Security System.
> For more information please visit http://www.messagelabs.com/email
> ______________________________________________________________________
>



-- 
Dr Simon Miles
Lecturer, Department of Informatics
Kings College London, WC2R 2LS, UK
+44 (0)20 7848 1166
Received on Monday, 20 June 2011 12:45:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 13:06:31 GMT