Re: organizing the state of the art document

Hi Paul, Jun, all,

Last minute comments before the telecon...

I agree that it sounds a good idea to focus on how existing
technologies meet the specific requirements we've enumerated.

While the concept of the matrix is helpful, I suspect that the matrix
might have a lot of (most?) entries showing "partial fulfilment" of
requirements by technologies - it would not be even close to black and
white unless we have a very large number of very specific technical
requirements.

Another way to present it may be to say, for each technology, which
requirements illustrate its strengths and which illustrate its
weaknesses.

thanks,
Simon

On 23 April 2010 09:51, Paul Groth <pgroth@gmail.com> wrote:
> Hi Jun and all,
>
> I think you're right that defining the matrix is difficult but I also
> think we already have quite a few technical requirements to work with. I
> think the task forces should be responsible for taking those technical
> requirements making them consistent with the dimension and as you say
> including information from the literature.
>
> +1 for Paolo's technology gap phrase.
>
> Looking forward to the discussion this afternoon,
> Paul
>
>
> Jun Zhao wrote:
>> Hi Paul, Paolo, and all,
>>
>>
>> Paolo Missier wrote:
>>> Hi Paul & all
>>>
>>> I agree on the focus, in view of the ultimate goal of the group which
>>> is, in my understanding, to provide recommendations on the
>>> opportunity for a W3C WG
>>
>> Or even to people who are interested in searching for technologies to
>> suit their provenance requirements.
>>
>>
>> -- and IMO this can only be based on technology gap
>>> analysis wrt requirements.
>>>
>>> Associating a task force to each /dimension/ may be a bit too
>>> fine-grained, maybe? the idea is good but management can be a
>>> concern.  Maybe dimension group to start with, to see where it goes?
>>
>> Paolo, what do you mean by /dimension group/? Do you mean the top
>> three dimensions, i.e. content, management and use? The initial idea
>> of having task forces for each dimension meant exactly this level of
>> granularity. Sorry if the message became misleading in the telecom.
>>
>> A good thing to use wiki for creating such document is that nobody is
>> excluded from anything. One can also chip in and check the progress
>> and direction of others.
>>
>> As we are all volunteers to the group, I felt it was absolutely
>> necessary to have someone dedicating their time to think carefully how
>> to organize and coordinate things. This someone can be different
>> people at different development stage, given the time scale of
>> individuals. Sub-tasking the writing of the document could increase
>> focus and reduce efforts, to some extent.
>>
>> [...]
>>
>>>> Given that goal, I would suggest to merge the two approaches that
>>>> were brought up on the last telecon.
>>
>> +1
>>
>>>> We should aim to develop a matrix that shows how different
>>>> technologies meet the technical requirements that we've gathered.
>>
>> Matrix is a good idea. Defining the matrix is the most difficult.
>>>> I would suggest that each provenance dimension have a task force
>>>> that is responsible for listing the technologies and requirements
>>>> for each dimension and then recruiting people expert in each
>>>> technology to
>>
>> It sounds like that before reviewing the state-of-the-art, we will
>> need to start from reviewing the requirement gathering and tease out
>> the technical requirements?
>>
>>>> write how this technology fits with the requirements. For example, I
>>>> would expect someone like Paulo to give a good response for PML, or
>>>> James on DBNotes. We could also approach people outside the group to
>>>> fill out a form about their technology.  It would be important that
>>>> some text would be attached to give justification for how the
>>>> technology meets a requirement. The task force would then be
>>>> responsible for integrating this text.
>>
>> It is a group/community effort, and this is how it should be done.
>>>> Anyway, that's my proposal: focus on technology meeting
>>>> requirements, task forces ensuring that the matrix gets filled out
>>>> with proper justfication.
>>
>> Are the existing provenance reviews only getting combined in when
>> filling in the matrix? Would they also help us with defining the matrix?
>>
>> cheers,
>>
>> Jun
>
>
> ______________________________________________________________________
> 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 Computer Science
Kings College London, WC2R 2LS, UK
+44 (0)20 7848 1166

Received on Friday, 23 April 2010 14:25:22 UTC