Revising W3C Process Community Group Teleconference -- 19 Feb 2014

<trackbot> Date: 19 February 2014

<koaliie> Minutes and summary of 3 February 2014 Chapter 7 Revision Task Force teleconference

<scribe> scribe: CoralieMercier

<scribe> scribenick: koalie

[chaals has audio only]

Jeff: purposes of the call today

<Ralph> [Ralph: regrets -- conflicting meeting]

Jeff: We want to have a TF report to the AB for the F2F meeting on March 4-5
... In order to issue a second last call of the document to the Membership
... We should agree on that.
... Also there are a bunch of issues that chaals has addressed and are pending review
... Are we comfortable closing them all?
... From the TF today (Mike, Chaals and myself), are we comfortable with closing the issues?
... I haven't seen complains from others in the TF (SteveZ and Ralph)

Confusion between 7.4 CR and 7.4.1 Revising a CR

Jeff: re: my issues with Revising CR and revising PR, chaals revised that but I am not yet convinced

chaals: I saw your e-mail

http://lists.w3.org/Archives/Public/public-w3process/2014Feb/0055.html

chaals: There are 2 issues
... revising CRs and publishing revised CRs
... you pointed an issue that is valid
... and wondered if two sections are necessary
... Ian asked the same question
... I think we do
... this was a piece of issue-59

issue-59?

<trackbot> issue-59 -- The 24-Oct-2013 Draft of Ch7 has some organizational issues and readability suffers -- closed

<trackbot> http://www.w3.org/community/w3process/track/issues/59

Chaals: Fantasai raised issue-59
... the process is very similar but is not the same
... so it's useful to know what it takes to know how to publish a revised CR

Jeff: If you are laborious as to have them in separate sections, I think we need to be really laborious and add what the Director needs to do, etc.
... I wouldn't object, if that were your editorial decision, but it would have to be complete.

chaals: That would be my editorial decision
... The review is ongoing, we would have to say the document exists.

Jeff: Another question:
... Would publishing a revised CR also trigger an additional CfE?

chaals: It typically does, it's addressed in the draft.
... Beyond removing things at risk, you're likely to trigger a CfE, hence you need the Director's decision.

Jeff: This is why I think the lack of parallel instructions is problematic
... As it's covered by the Patent Policy, the info there is is informative

chaals: It's a normative patent requirement
... I'll continue that

Jeff: I'm fine with this in 7.4, but not in 7.4.1

chaals: I'm not sure what's missing

Jeff: the lack of parallel instructions is confusing
... In 7.1 you have a declarative statement
... in 7.4.1 it's part of a discussion.

chaals: It doesn't create a CfE
... formally that call belongs to the team contact who's supposed to make the call
... it's not necessarily the case that even with substantive changes a revised CR will trigger a CfE.
... Would you like this statement to be stronger?
... What statement would you like?

Jeff: I don't know; totally confused now.
... In 7.4.1, the reason you're explaining why we need a revised CR is because of a trigger of CfE
... and now you're telling me it doesn't necessarily.

chaals: We could copy and paste the text from the Patent Policy; Ian's preference is to refer and link.

Jeff: the text says if there are substantive changes made, you need a Director's decision to publish a revision because this will be an exclusion opportunity.
... and if it might not?

chaals: Section 4 of the patent policy describes this. @@@@missed the start of point.

Jeff: I'll leave it up to your editorial decision, if you can find a way to clarify this

<scribe> ACTION: chaals to amend 7.4.1 to add actions that need to be taken upon revising a CR including the requirement that the Diretors announces publication to W3C groups and the public. [recorded in http://www.w3.org/2014/02/19-w3process-minutes.html#action01]

<trackbot> Created ACTION-30 - Amend 7.4.1 to add actions that need to be taken upon revising a cr including the requirement that the diretors announces publication to w3c groups and the public. [on Charles McCathie Nevile - due 2014-02-26].

Section 7.5 PR seems to say that we still MAY remove features

Jeff: Did you see my confusion about 7.5

chaals: To get from CR to PR, it's ok to publish with features at risk
... WG may remove features at risk in CR w/out republishing a revised CR

Jeff: Why is that bullet in 7.5? it is already in 7.4 and 7.4.1

chaals: You can do it in CR, it's the practice

Mike: It seems we're keeping PR in its form
... in the new process PR was just a 30-day warning on AC review

chaals: This is analogous to revising a CR
... the practice is if it's the last revision of CR, you won't publish an updated CR if you publish a PR the next day
... PR is basically carved in stone.
... In the transition, the one class of substantive change allowed is drop things that were already marked at risk
... It's not a full blown PR @@@
... You don't have to publish a CR and a PR to @@@@

Mike: I understand

Jeff: There are other changes as well
... I don't mind that we reintroduce PR, but it's different in CR

Mike: What we came up with is not dramatically simpler than the current process.
... Major innovation is allowing AC review much earlier in the process.

Jeff: The 4th bullet of 7.5 PR [Jeff reads]
... CR review hasn't possibly closed when we enter PR

chaals: Yes. CR reviews are defined.

Jeff: Is that different from the AC review?

chaals: Yes

Mike: What's the difference then?

chaals: They're different reviews
... in 7.4 CR has a deadline for comments
... you might accept late comments for some reasons
... this happens in reality but not frequently
... either you fix spec or convince Tim you don't need to address

Mike: The existing process says you can get around for 5 months, and this is what we're trying to address.
... in 5 months the world has moved on
... look at the shadow DOM kerfuffle

chaals: This requirement ensures a decision is based on each comment received

Mike: You've convinced me, thank you.

Any other concern?

Jeff: Mike, any other concern?

Mike: I'm mildly dissatisfied with the way we put one sentence in the intro that new steps can be added in the chartering process
... Chap7 doesn't talk about chartering
... When do we address the rest of the Process?
... This round is focused only about Chap7?
... The sentence is at the bottom of 7.1 as I recall
... That was in response to complains by Paul and I
... I wonder if it's sufficient genuflexion to the concerns

Jeff: They wanted to adopt tighter processes?

chaals: So long as it doesn't conflict with the Process, go for it
... Will we address things outside Chap7? Jeff said no, focus is on Chap7 for now.
... I'm ready to make proposals for those
... but we need to finish Chap 7 first.

Jeff: I don't understand what Paul's comment have to do with Chapter 7

chaals: Half-fixed. We ruled out of scope for Chapter 7.

Mike: Paul's problem is that for major cat-herding like HTML and TP, LC is a valuable mechanism
... in Shenzhen, Paul proposed LC is optional
... Our proposal to make it explicit that WG can update their charters to meet certain needs
... Perhaps an extra sentence to explain might be appropriate.

Jeff: I'd use this as a Chapter 7 issue. If the community wanted to have spelled out
... that there's a concept of LC
... and this is how it works, and this is optional,
... that would be in chapter 7, not in chapter 5
... I don't accept that the restriction from the Team is that we're only doing Chapter 7.
... My impression overall was that the community had rejected that requirement.

Mike: I agree
... I will inform Paul that the TF thinks the current text is sufficient.

chaals: Paul can chime in at last call

[fantasai arrives]

Mike: It might be worth to add to the sentence and provide background.
... It would be a small change that could address Paul's question

Jeff: I'd suggest you write again to Paul that the current consensus is to address the issue with this sentence,
... that it's going to AB approval, and that he should speak soon lest it's too late to raise it again.

chaals: Last Call means many things to many people.
... To a11y, means forcing group to address a11y issues before going to CR.
... To some other groups, means "we think we're done, want to make sure".
... To HTML, means "please review this document".
... Point of not having LC is to not have all these differing interpretations.

Jeff: fantasai, do you have any comment?

fantasai: From the last time I looked, main concern I had was not understanding why a group would ship CR, since it's easier to keep in WD form
... CR is a signal to the world in the current Process

chaals: We introduced a section on revising CR
... to back-up fantasai's comment,
... The bar for revision is higher. Assumption is you want to get to Rec.
... You go to CR because you want to go to Rec
... You don't want to be revising and revising CR
... You're not speculating. Some groups will.

fantasai: in CSS WG we use CR as the signal to the wider community that the spec is stabilizing.
... I haven't looked enough to make further intelligent comments

Next steps

Jeff: Now, even after we say we're done, it will go for AC review
... and there will be opportunity to go back
... Do we need to meet again before the AB meeting?

chaals: I'd send a call for consensus about the proposed resolutions for issues
... and in a week, I'd send declare consensus and call the issues

Jeff: OK, I'll wait till you push a draft tonight
... and send a call for consensus tomorrow.
... Any other comment? question? before we adjourn for today?

[none]

Jeff: Thanks all
... Next actual conversation will take place at the AB f2f
... fantasai, it is on the West Coast
... if you would like to join, I invite you as a member of this TF
... Tuesday Mar 4 at 8:30 am in SF

fantasai: I can't object to the time and location. Thanks.

[adjourned]

Revising W3C Process Community Group Teleconference

19 Feb 2014

Attendees

Contents

Confusion between 7.4 CR and 7.4.1 Revising a CR

Section 7.5 PR seems to say that we still MAY remove features

Any other concern?

Next steps

Summary of Action Items

Scribe.perl diagnostic output