Resolving LC comments (was: pfps-16, proposed resolution)

At 22:52 01/04/2003 +0300, Jeremy Carroll wrote:

>Hmmm

OK, there are a number of points to tease out here.   Although this is in 
direct response to comments by Jeremy, I think it addresses the entire 
working group about how we might proceed to wrapping up this job.

First, I'll respond point by point, then I'll try to stand back and 
consider the larger picture, as I see it...

>Peter did not like one sentence.
>You deleted the sentence.
>
>Peter did not request that, he appeared to want a correct sentence.
>However, I would not object to the fix of deleting the sentence.

It would have been nice to come up with a statement that was brief, 
technically accurate and accessible to an audience of developers/users 
rather than logicians, but I found that beyond my ability... hence what I 
offered.

>How about.
>
>Propose: resolve pfps-16 by deleting the sentence:
>[[
>The expressive power of RDF corresponds to the existential-conjunctive (EC)
>subset of first order logic [Sowa].
>]]
>(and possibly in consequence the reference to Sowa).

This strikes at the heart of one of the problems I am having with this 
process... I find that focusing on issues in isolation from the wider 
context of the document to which they refer is not helpful, as I find it 
difficult to deal with issues in isolation without considering their 
relationship to each other and to the document as a whole.

Consequently, I am trying to work through the outstanding issues and deal 
with them as I understand them in the context of the document, and present 
proposals in terms of references to proposed document revisions.  My hope 
is that WG members (and, subsequently, commenters) can read the revised 
document sections (URIs provided with the proposed resolution) and agree or 
disagree with the text provided, thereby focusing attention on the end 
rather than the means.

>You seem to have forgotten that we have stopped doing new work except in
>response to issues raised.

When we went to last call, I thought it was with a clear understanding that 
there remained a number of document presentation issues that needed to be 
fixed.  I seem to remember discussion that the agreement to go to last call 
meant that we believed that the normative technical material was 
substantially correct, not that the documents were without need for further 
refinement.  If I had understood then that the last-call was to be a 
straitjacket to prevent the kind of refinement that (to me) was clearly 
still desirable, then I would have opposed doing so.

[[
15:44:13 [daveb-scr]
     ericm: gives LC process overview
15:44:54 [daveb-scr]
     ... WG believes this is pretty much ready for a REC
15:45:36 [daveb-scr]
     ... we are the polishing stage and formally addressing comments - "we 
think we are done"
]]
-- http://www.w3.org/2003/01/17-rdfcore-irc

(I note here the words "pretty much..." and "polishing stage".  We may 
reasonably disagree on how much change can fall under this, but I think it 
clearly allows more than making minimal changes to address comments.)

In many ways, I would have preferred more document refinement before 
LC.  Part of the reason that I supported going to last call when we did was 
the combination of time pressure, and the fact that it takes so long to get 
any new working draft published.

>I am at a lost to see why the section title changed, why a new paragraph on
>Clark Kent appeared, etc. etc.

The new paragraphs were additional elements in response to issues 
macgregor-01, macgregor-02.  Maybe they're unnecessary, but I was trying to 
respond to the spirit of the issues raised as well as the letter of the 
issue resolution.  If the group so agrees, they can easily be removed.

As for the section title changes, I viewed these were editorial refinements 
that reflected the change of emphasis of the section contents.

>Possibly I am seeing the result of many proposed issue resolutions all at
>once. However, I am nervous that you appear to have taken out one sentence
>that might have been objectectional and added five or six that might be!

Well, I took out one sentence that *was* objected to.  And rearranged some 
other material.  Apart from that explicitly highlighted in my commentary 
accompanying the proposal, I don't believe I added any new *content*.

>For instance, the change in the section title breaks W3C guidelines on case,
>and so one of the reviewers who was fairly positive would be less so now.

Oh, grumble.  That change may have been wrong to make, but the 
capitalization of section headings is something that grates quite severely 
on me.  (And FWIW my wife, who is a professional technical sub-editor, 
agrees.)  I didn't realize there were W3C rules governing such trivial 
things;  it was never an issue when I did the CC/PP document.  If it's 
important, that can be put back, though I find it hard to see it being 
somehow critical to the document's technical acceptability.

...

So here's my view of where we are:

The reduction of comments to specific issues is certainly useful as 
guidance and a checklist for the resolution of comments, but I fear the 
tail may be wagging the dog here.  We've already had some reviewer pushback 
on proposals that narrowly addressed the distilled issue rather than the 
original comment.

I note that the ultimate deliverables here are the documents, not the list 
of issue resolutions.

I also wonder that we seem to be trying to avoid a second last call at all 
costs.  I submit that it might conceivably be quicker to bite the bullet 
and accept a second last call than to agonize over every change in case it 
makes a last call necessary.  (I'm not arguing that we should plan for 
this, just raising the idea.)

...

I am becoming concerned that I may be out of step with the expectations of 
the working group here.  Rather than plough ahead, I shall suspend my work 
on this and turn my attention to the semantics document review that has 
been requested of me, in the hope that the consensus way forward may become 
clearer.

#g


-------------------
Graham Klyne
<GK@NineByNine.org>
PGP: 0FAA 69FF C083 000B A2E9  A131 01B9 1C7A DBCA CB5E

Received on Wednesday, 2 April 2003 05:08:48 UTC