W3C home > Mailing lists > Public > public-w3process@w3.org > October 2013

Comments re 7.6.2 and LCCR [Was: New draft of Chapter 7 Process proposal]

From: Ian Jacobs <ij@w3.org>
Date: Thu, 17 Oct 2013 22:48:44 -0500
Cc: "public-w3process@w3.org" <public-w3process@w3.org>
Message-Id: <6F42FC94-9D27-4007-93F2-D696829F91D1@w3.org>
To: Charles McCathie Nevile <chaals@yandex-team.ru>

On Oct 17, 2013, at 8:44 PM, Charles McCathie Nevile <chaals@yandex-team.ru> wrote:

> Hi folks,
> 
> as always at https://dvcs.w3.org/hg/AB/raw-file/default/tr.html
> 
> changes were mostly related to Ivan Herman's comments and subsequent discussion:
> - tweaked and poked at Section 7.6.2 on modifying a recommendation

Please revert that edit that you made to change:

>> "Corrections that do not affect conformance"
>> to:
>> "Corrections to references"

Here are several reasons:

 1) In June at the AB meeting we decided import the classes of change as-is. I am surprised at this change
     given previous discussion.
 2) The change in question narrows the scope of that category but there is no clear rationale why that
     category of change needed to be narrowed. I had understood that a goal of this process document
     revision was to make many things easier; by narrowing category 2, it becomes more difficult to make
     some corrections that do not affect conformance.
 3) References are important enough to be reviewed by the Director as part of the Rec Track process.
     I think that a corrections policy that has "no technical review of the proposed changes" does not align
     with the way the references were reviewed in the first place. (In the case where a reference needs 
     to be updated because of a broken link, it can be done in place as a category 1 fix.) Therefore, I believe
     this edit removes from the Director an important review role.

> - added requirements for status to be unique, and explicitly identify if a spec is published as a provisionally approved Recommendation (technically, there is no requirement to republish in this case) or a W3C Recommendation (Yes, these really need to be republished still :) ).
> - Some editorial stuff
> 
> Full changelog at https://dvcs.w3.org/hg/AB/
> 
> Your comments on the changes, or the document as is, are as always very welcome.
> 
> There are now 7 issues pending review. Issues 2 and 37 are old and I think only pending review because we forget them. 44-46 are from the last draft (there has been no meeting of the Ch7TF since then). Issue 47 is dealt with in this draft, but I am not sure if it is ready for closing, and would really appreciate a close look. Issue 48 is also dealt with in this draft, and while of course I appreciate feedback I am happier that it has been handled reasonably.
> 
> There is also the open issue 39 (transition) and accompanying discussion on this list.
> 
> Finally, Ivan Herman points out that the name Last Call Candidate Recommendation is horrid. I agree, but think we need "Last Call" in the title to help us clarify that it is the stage referred to in the Patent Policy as "Last Call". Anyone have a good idea for this?

I think it is challenging to find one short label to convey three signals:

   * time for broad review
   * time for (more) implementation experience
   * start formal AC review

If I had to favor one of the three signals, I would favor "time for broad review" (since I think there may be
more reviewers than implementers at this point).

With that in mind, here are some ideas:

 * Mature Draft                   (This conveys info about the maturity)
 * External Review Draft    (This conveys the action we want done, and that we are looking beyond the WG)
 * RIPE Draft. This stands for: Review and Implementation, Particularly Externally! :)

I would not use:

 * Stable Draft (since that may oversell its stability)

I do not think we need to keep either LC or CR, and recognize that whatever term we use we will need to 
bind it to the term "Last Call Working Draft" in the Patent Policy.

On the other hand, if forced to choose between "Last Call" or "Candidate Recommendation" (rather than the combo),
I would favor "Candidate Recommendation" since that avoids situations like "Third Last Call".

Ian

--
Ian Jacobs <ij@w3.org>      http://www.w3.org/People/Jacobs
Tel:                                          +1 718 260 9447
Received on Friday, 18 October 2013 03:48:47 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:35:09 UTC