Revising W3C Process Community Group -- 27 May 2020

<scribe> scribe: jeff

David: Today we hope to wrap up the process
... doubt that we'll get more from the AC
... no new issues filed since agenda was proposed

<dsinger_> https://www.w3.org/Member/wiki/P2020-issues

David: only 404 added since AB discussion
... any other issues for today?
... anything else to be discussed?

Florian: We should discuss that which we will ask PSIG about
... do we have a consensus?

David: So we'll look at 403.
... anything else

#388

<dsinger_> https://github.com/w3c/w3process/issues/388

<dsinger_> https://github.com/w3c/w3process/issues/388#issuecomment-632730884

UNKNOWN_SPEAKER: I've tried to put together the final comment on 388
... responsive to comments

<tzviya> as chair of PWE - there has to be clear direction about how to move forward and clear communication to those involved in ancillary groups. I do not think simply decoupling cepc is sufficient because then there is no guidance

UNKNOWN_SPEAKER: Tzviya's comment ^^

Florian: My position is consistent with that
... there may exist a better way for CEPC, Ombuds, etc.
... we should spend time exploring
... in the interim, some clear guidance is helpful
... rules written here are consistent with our behavior
... we should adopt this and keep the issue open for next year
... I think that is consistent with Tzviya

Chris: Key is to realize that CEPC was painful
... authority came from AB, but AB was late with comments
... we can't have the AB delegate its authority entirely to an unelected group of people
... we need to do it better
... not leave PWETF out to dry
... the AB should delegate to some other body
... hence proposed changes are fine

<florian> +1 to cwilso

<tzviya> +1

Chris: should not do more this year

<chaals> [+1 to telling the AB to be more communicative and directly involved rather than delegate and then complaining]

Jeff: I'm okay with that if we also adopt the changes in issue #404.

<fantasai> https://github.com/w3c/w3process/issues/404

David: I agree with Florian

<cwilso> [+1 to Chaals' comment]

David: ultimately the AB's responsibility
... buck must stop somewhere

<chaals> [I prefer not to change the text for 2020 because I don't see it improves things. This won't be an objection to hold up consensus. I think the responsibility lines are already there]

David: but agree that we should delegate CEPC to experts

<chaals> [+1 that the next revision should actually consider a different mechanism for managing CEPC]

PLH: One minor improvement is to acknowledge where the documents are currently developed

Fantasai: As a note?

PLH: Could be a note

David: In a process doc?

Fantasai: Helps people find things

David: Anyone object

Chaals: I'm concerned that putting stuff in process doc related to groups with fixed charters is tricky.

PLH: Add language that it is "as of now"

Florian: I'm neutral. Might not be the ideal time since we need to discuss it still.

<dsinger_> "Note: as of June 2020 the Patent Policy is developed in the PSIG and the CEPC in ... CG"

PLH: Leave it to the editor.

David: Comments on text?
... no comments?

Florian: Proves it is the right language

Fantasai: I agree with your changes
... @@@ issue #404

David: Are accepting this with addition of note?

<cwilso> +1

<plh> +1

<chaals> [For the record I vote against, but without standing in the way of consensus]

scribe: so we can document that the process is done in this CG

<fantasai> +1

scribe: Chaals, why are you voting against?

Chaals: I don't think it is an improvement
... changes the lack of clarity
... rather leave it out and come back next year
... leave as is for now
... not a strong objection

David: Tzviya was asking for more clarity this year
... other thoughts on this?
... thanks Chaals. Understood and move on.

RESOLUTION: accept the language in teh comment and add a note documenting where the docuemnts are developed

#404

<fantasai> github: https://github.com/w3c/w3process/issues/404

<dsinger_> https://github.com/w3c/w3process/issues/404

David: Florian, explain your spelunking

Florian: I understand why Jeff wants to make sure that the Director is in the loop
... but the definition already involves the Director.
... unless it is unclear, I'm not strongly convinced.

<dsinger_> https://www.w3.org/2019/Process-20190301/#ReviewAppeal

[Florian reads from 7.1]

Florian: It doesn't say that the Director can decide something else.

Fantasai: So we should accept David Singer's working

<fantasai> dsinger's wording: A W3C decision is one where the Director decides, after exercising the role of assessing consensus of the W3C Community after an Advisory Committee review.

David: It is ambiguous. Is he ratifying the AC decision or is he deciding based on the AC?
... is he just a rubber stamp?
... can he decide in the absence of a consensus?

Florian: I'm OK with the clarification if people think it is needed.

<dsinger_> https://github.com/w3c/w3process/issues/404#issuecomment-634069934

Fantasai: It is obviously not clear right now.

David: Proposal is ^^

<fantasai> +1 to the proposal

David: Comments?

Fantasai: We should accept the proposal.

RESOLUTION: adopt the text in https://github.com/w3c/w3process/issues/404#issuecomment-634069934

David: No objections, we should accept.
... we are cooking along, this is great, I love it.

#402

Florian: This could take a while

#403

<dsinger_> https://github.com/w3c/w3process/issues/403

<fantasai> github: https://github.com/w3c/w3process/issues/403

Florian: Anyone here who wasn't at the AB?

Jeff: Pierre is.

Pierre: Hard stop at top of hour

All: All of us do

Florian: Summarizes #403
... emphasize that informative notes that announce a proposed change are effectively a WD
... but not qualified as such for the Patent Policy
... Member who leave the group will not make commitments as they would on a WD

<dsinger_> I agree that we need to document the 'amendment material' as a Working Draft

Florian: proposal to recognize these a WD's

Fantasai: We should accept the proposed changes

<dsinger_> https://github.com/w3c/w3process/issues/403#issuecomment-632285395

Fantasai: subject to PSIG being OK

David: Any objection

<fantasai> jeff__: I agree with fantasai that we should accept this

<fantasai> jeff__: But also if PSIG pushes back, don't need to block Process 2020

<fantasai> dsinger_: I hope they don't. Will need to discuss if they do

<fantasai> florian: I think it would be an unfortunate loophole if we don't fix this, but not fatal

<fantasai> dsinger_: If they disagree, we

David: If they throw a wrench we'll have a discussion
... we close; PSIG meets; wrap up meeting on the 10th

RESOLUTION: we support the text in https://github.com/w3c/w3process/issues/403#issuecomment-632285395 pending PSIG discussion and acceptance

#402

<dsinger_> https://github.com/w3c/w3process/issues/402

David: Pierre, can you introduce - nomenclature and practical

Pierre: The objective is to have a perpetual CR
... no objection to that concept
... name is misleading. It is not a Candidate REC - it will never be a REC
... SoTD does not communicate accuracy
... indicates that it does not have endorsement of Membership
... trying to clarify what those perpetual CRs mean
... external and internal groups know how to reference
... I have three questions
... 1. Does a perpetual CR represent consensus of Membership
... 2. Is it a WIP or a complete work?
... 3. Does it need to demonstrate any implementation experience

<dsinger_> chair encourages everyone to read a CR SOTD section, example is here https://www.w3.org/TR/2020/CR-ttml-imsc1.2-20200324/

Pierre: don't object to concept, but want clarity on those questions.

David: Please look at SoTD for a CR ^^

<fantasai> "has been reviewed by the W3C Community and has the endorsement of its sponsoring Working Group, but does not have the endorsement of the full W3C Membership"

<fantasai> would be an improvement

<fantasai> and more in line with what we need here

<plh> https://htmlpreview.github.io/?https://github.com/w3c/tr-pages/blob/master/p2020mockup/cr-6.2.8.1.html

<plh> https://htmlpreview.github.io/?https://github.com/w3c/tr-pages/blob/master/p2020mockup/cr-6.2.8.2.html

[David and PLH quibble about what a typical SoTD should say]

Pierre: Still says "does not imply endorsement"

David: Other background?

Fantasai: We should change SoTD to better describe what it is.
... should say that it has the endorsement of the WG
... still does not have endorsement of the membership (no AC review)
... can say reviewed by W3C community
... can say has support of sponsoring WG

Florian: Whether CR is an end state or not?
... any CR publication is not an end state
... even perpetual CRs will be revised forever
... as long as technology keeps moving, the CR is a moving target
... no antipathy to REC, but too much happens at CR
... some groups will move to REC
... others will be at CR for a long time
... should be retired if done and not a REC
... naming is hard
... It is not endorsed
... It is a WIP
... Some will move forward; some will not
... so process is OK

PLH: Pierre has a point that CR is a broad name
... keep issue open and push after some experience, for P2021
... Boilerplate SoTD needs to be improved
... Glad it is getting attention
... It is not a consensus of Membership. That requires moving to PR.
... Represents consensus of WG
... even on the content
... (as opposed to WDs which only have consensus to publish)
... LS community want to look at WIP
... does not need to demonstrate implementation experiene if a CR
... although they may have it
... In WHATWG; only has implementation commitment

<Zakim> chaals, you wanted to suggest given two kinds of thing we should have two status options

Chaals: Boilerplate is good enough
... the non-boilerplate stuff needs attention
... 2 approaches
... those that aim at RECs and those that aim at a document that the WG can mess around with at will - but not prepared to get formal W3C review
... In my world, the latter meets the criteria for a Note
... should make that distinction clear
... working towards an endstate (given that we have no versioning approach)
... and distinguish to the other type of CR

Pierre: The AC can object to a CR transition

Florian: Anyone can object to anything

Pierre: Is CR announced on AC list?
... if you did that, it would have the consensus of the AC

Fantasai: The AC is not paying attention.

<dsinger_> https://www.w3.org/2019/Process-20190301/#candidate-rec

Pierre: Can provide a formal opportunity to object

David: Grounds for objection are limited to meeting criteria for CR
... other than that you cannot refuse a CR
... not as strong as with PR

Pierre: Only exception is implementation experience

Florian: And AC Review

PLH: And that you have closed all of your issues

Pierre: True for CR

<dsinger_> "the question of whether the - specification is appropriate to publish as a W3C Recommendation."

Pierre: we are so close - why not add an AC review for a perpetual CR
... if we put it before we are just moving goalposts
... I have a real issue with a formal publication explicitly state that it does not have endorsement

Florian: It does not have endorsement

David: We only ask for endorsement at REC at PR

<Zakim> dsinger_, you wanted to talk about 'draft document'

<Zakim> florian, you wanted to comment on implementation experience

Florian: If we put an AC review before CR we would give them a month
... introducing a one month delay
... we will then have fewer CRs
... or push things to other places
... if a group wants W3C endorsement they need to ask for it.

Pierre: But it has a W3C logo, hence an implied endorsement

Florian: Also EDs, WDs

Pierre: Clear not final

Fantasai: CR is also clear

<fantasai> fantasai: CR is not definitively not going to REC. We have processes to maintain living standards as RECs as well as in CR.

<fantasai> fantasai: A group could decide it wants W3C endorsement 10 years later, and that its spec is stable enough that it's content to maintain the extra requirements of REC for that, and transition its LS from an CR to a REC

<florian> fantasai: I think we need to update the status section to say that it does have the endorsement of the WG

PLH: To respond to Pierre
... CR may have open substantive issues that get opened
... Director does not require that you address them to update the CR
... Move the AC review forward into the process - the LS people want updates at any time
... You are moving burden from PR to CR; but people don't want that burden
... there is a price to get the consensus of membership
... not all groups willing to pay the price

Pierre: Just trying to help perpetual CR more formal
... SoTD also says that it is inappropriate other than a WIP
... people will want to reference CSS

Florian: CSS will not be a perpetual CR

Fantasai: We should remove the sentence.

Florian: Really? It is a WIP

<cwilso> +1 to fantasai

Fantasai: Does not need to be cited as such.

<fantasai> "This document is a work in progress, and may be updated, replaced or obsoleted by other documents at any time."

<fantasai> instead of "This is a draft document and may be updated, replaced or obsoleted by other documents at any time. It is inappropriate to cite this document as other than work in progress. "

David: We often call this a "draft"
... CR has a lot more status

Fantasai: I proposed wording

David: I'm not convinced we should alter the process and change the name
... does anyone disagree?
... some docs get constant trickle of additions (e.g. DOM)

<florian> +1 for working on SoTD, not changing the Process

David: maybe perpetual CR is OK

<chaals> [I think we should alter the process, not just tweak the Status Section]

David: If people need a REC they should just ask.

<fantasai> -1 to editing the Process, +1 to improving SOTD

David: does anyone think we need to edit the process document

<fantasai> https://w3c.github.io/w3process/

<dsinger_> https://w3c.github.io/w3process/#maturity-levels

Florian: Can tweak the wording to clarify. Adding an AC review, though, would be a major change

David: The process text for CR should match the SoTD

PLH: Should we open an issue to iterate on boilerplate

David: Yes
... not hearing changes for process text; but need to work on boilerplate

Chaals: We need to change the process
... but I doubt you would be happy to do the necessary work and accept the concomitant delays for 2020
... will take some time

David: A hasty edit will be damaging, not helpful

<plh> I will open an issue for CR Draft boilerplate. Once we we narrow it down, we could open a separate one for CR Snapshot boilerplate.

[David puts note in issue to reflect above discussion.]

David: Anyone disagree - leave issue open, work on SoTD

<fantasai> plh, maybe do both simultaneously :)

David: no disagreement hence resolved

<fantasai> plh, we can handle two open issues. ^_^

David: big thanks to Florian and Elika for work to date
... close to finishing
... next meeting is June 10th

<cwilso> +1

Florian: Big thanks to chair and the crowd

David: I enjoy working with you all

Fantasai: And the project manager as well

<plh> --> https://github.com/w3c/w3process/issues/405 CR Draft boilerplate

David: And PLH

[adjourned]

RESOLUTION: we keep 402 without Process changes this year, but turn our attention to improving the SOTD

- DRAFT -

Revising W3C Process Community Group

27 May 2020

Attendees

Contents

#388

#404

#402

#403

#402

Summary of Action Items

Summary of Resolutions

Scribe.perl diagnostic output