Re: PROV-ISSUE-331: feedback on PROV-Dm WD5

Paul,

On 09/04/2012 10:03, Paul Groth wrote:
> Graham,
>
> I understand that we can always vote. But I would prefer to have a
> conversation and identify particular issues and ways to resolve those
> issues, so that we can proceed in a consensus bases fashion. That's
> why some sort of prioritization from your side is helpful. If we can
> address your concerns, in particular, if the changes are
> straightforward we should try to do that.

Agreed that consensus is best.

I'm not sure what counts as "straightforward" here.  For someone coming to this 
from outside our little echo-chamber, I think the document is hard to follow and 
understand, and as such needs quite comprehensive sub-editing and tightening up. 
  I find it hard to prioritize any particular issue, because I find it hard to 
go more than a couple of paragraphs without running into text that I find 
potentially confusing, or worse, mainly because it seems to me to draw in too 
much unnecessary detail.

In making my comments, I'm trying to put myself in the place of someone who has 
not been party to all our discussions, and who has not prior familiarity with 
provenance models.  Maybe the best approach is to get a real outsider to review 
this - and the best timing for that would probably be *after* this PWD round is 
released.

> ... This is less about
> addressing issues (which will happen) but what is the best time for
> that occur.

Of course.  Which is why I indicated that I would abstain rather than oppose for 
this coming round.  I may feel differently if there were proposed as a LC document.

> Also, will you be on the call Thursday?

That's difficult to know for sure.  I'm travelling this week and next, with 
intermittent Internet connection.  It's probably safest to make sure you have 
needed inputs from me beforehand.

#g
--

> On Sun, Apr 8, 2012 at 11:47 PM, Graham Klyne<Graham.Klyne@zoo.ox.ac.uk>  wrote:
>> Paul,
>>
>> I appreciate your problem.  But I have to call things as I see them - I don't
>> think we should sacrifice quality to schedule.  And, in any case, if I'm right,
>> you'll get all these issues raised in last call by people outside the WG.
>>
>> If the rest of the group think it's OK to go ahead, then as chair you need to go
>> with the group consensus, which doesn't have to be unanimous.  (FWIW, I'd
>> probably vote to abstain (-0) rather than oppose (-1) - but I'd want my concerns
>> to remain on record.)
>>
>> #g
>> --
>>
>> On 08/04/2012 08:48, Paul Groth wrote:
>>> Graham,
>>>
>>> Just to say that in terms of schedule, we have this release and then
>>> the last call. At least that was the plan.
>>>
>>> We'll have to see what other reviewers and the group as a whole stay.
>>> Also, we need to see what the editor's have to say in terms of time to
>>> respond to your comments.
>>>
>>> I would like to figure out how we can divide the work so that we can
>>> keep on our schedule i.e. what needs to be addressed for this WD and
>>> what can be left till the revisions before last call. My general
>>> feeling is:
>>>
>>> - WD5: address any issues with constructs and "easy" editorial issues now an
>>> - LC: all open issues
>>>
>>> Cheers
>>> Paul
>>>
>>> On Sat, Apr 7, 2012 at 9:20 AM, Graham Klyne<Graham.Klyne@zoo.ox.ac.uk>    wrote:
>>>> Paul,
>>>>
>>>> Yes, it's largely a document/text quality thing - I feel it doesn't entirely lay
>>>> things out clearly enough for its target audience, and in some cases is actively
>>>> confusing.  This may be "editorial", but I think it's important enough to need
>>>> addressing to move forwards towards LC.  There are a few points of substance
>>>> (mainly stuff that feels superfluous to me), but I wouldn't be surprised to be
>>>> lone voice on that.
>>>>
>>>> I've indicated a number of specific points points in the "details" part of my
>>>> email, with suggested alternative phrasing, though there are many more (similar
>>>> to those I detail) that I've skipped over in passing.
>>>>
>>>> #g
>>>> --
>>>>
>>>>
>>>> On 06/04/2012 21:36, Paul Groth wrote:
>>>>> Hi Graham,
>>>>>
>>>>> Just for clarification, given that you think prov-dm is not ready for
>>>>> release, it's important to understand what exactly could be done to
>>>>> get it to the point where it is.
>>>>>
>>>>> Reading through your points, it seems to me that your comments are
>>>>> primarily editorial, in that it's the explanation, definition and
>>>>> organization of the terms that is the issue. Is that a correct
>>>>> interpretation?
>>>>>
>>>>> If not, can you identify the specific things that would need to be
>>>>> addressed for us to move forward on prov-dm?
>>>>>
>>>>> Regards
>>>>> Paul
>>>>>
>>>
>
>
>

Received on Monday, 9 April 2012 10:31:45 UTC