<scribe> ScribeNick: fantasai
dsinger: Apologies for the late
agenda
... Any modifications ?
... hearing silence...
github: https://github.com/w3c/w3process/issues/461
florian: chaals seems to be
maintaining his position
... As far as I'm concerned ...
... Seems we all want the TAG to be able to issue documents
with AC approval
<plh> --> https://www.w3.org/TR/2020/NOTE-design-principles-20201110/ Web Platform Design Principles
florian: The question is are they
RECs or not
... A bunch of us are saying that they can't be because invokes
Patent Policy
... chaals says it has to be REC but doesn't say why
... idk if we have consensus or not
... If we go with a class of doc that is like a REC but no
Patent Policy, should only TAG be able to publish or should all
groups be able to?
dsinger: I would say let any
group make such things
... Might be useful for HRGs
jrosewell: Reading through thread, why is TAG different from other Interest Groups?
florian: TAG is an elected
body
... has a special status
... revisiting that is a whole new question
... and I'm not sure there's a lot of appetite for such
question, first time I'm hearing it
fantasai: That question is a completely separate issue. If it's an issue to pursue, file it separately. Shouldn't block this issue.
dsinger: Agreed.
... So next step would be to draft out the Process text for
such a W3C Consensus document
plh: Agree giving to other groups
than TAG
... Difference between TAG and other groups is that Director is
part of the TAG.
... Other groups, need Director to approve, but Director is
part of TAG so TAG approval includes Director approval
florian: That's true in theory,
not so much in practice these days
... Willing to draft the text
... but also, we also have issues of disentangling Notes and
RECs
... Might be easier to do this after untangling those also.
dsinger: Related question, can we create different visual identity for things which are group consensus and those that are W3C Consensus?
<Zakim> jeff, you wanted to address plh's issue
<florian> The issue I mentioned is https://github.com/w3c/w3process/issues/342
jeff: Issue about visual identity
of RECs and other documents, already an issue filed elsewhere.
Let's not entangle that here.
... Raised important issue of Director ruling in case of
conflict of interest
... Simple fix, unsure where to document it
... Extreme case, TAG has a Finding and Director is involved in
the discussions
... Goes to AC approval, and there's an FO
... Use case of plh's concern wrt conflict of interest.
... Easy fix would be that the Director acting on the FO needs
to recognize the conflict of interest and delegate the Director
function of handling the FO to someone else
dsinger: OK, task for editor to
draft up text
... Side-issue that nobody likes the name 'W3C Report'
... if anyone has a better name?
fantasai: "W3C Guidance"? Kinda like REC but not? :)
github: https://github.com/w3c/w3process/issues/455
dsinger: So are we making this change?
<jrosewell> No strong view.
<cwilso> should keep the same as IETF
<cwilso> (but I don't care that much, I just think it's an arbitrary difference)
fantasai: I don't see people
using "level" outside the context of the Process document, but
stages has been
... And while IETF might use levels, TC39 uses stages
dsinger: ok, we'll leave it open
github: https://github.com/w3c/w3process/pull/432
florian: Merged in various
suggestions
... Suggest looking at the diffs rather than the
conversation
https://github.com/w3c/w3process/pull/432/files
dsinger: Changes *should* be
editorial.
... Can fix up minor problems later also
<wseltzer> lgtm
lgtm
RESOLUTION: Accept PR
github: https://github.com/w3c/w3process/issues/324
RESOLUTION: Close no change
https://github.com/w3c/w3process/issues/38
https://github.com/w3c/w3process/issues/63
<wseltzer> [I had proposed to close #38, as its initial opener]
https://github.com/w3c/w3process/issues/38#issuecomment-658794831
florian: Homework for me
florian: "P2021:Priority" are the
ones we're trying to do, others are hopeful
... Pile of issues about registries
... several issues on simplification. Don't need to do all, but
doing a few is nice.
... and Wide Review, need to revisit
... If we want to make the progress on others, need to do it,
otherwise when we're done with the Priorities we're going to
cut for the year
jrosewell: I had expected to put
through some changes to 2021, is there a deadline for the
Priorities?
... How do we decide the Priorities?
dsinger: We triaged those at the
end of the previous cycle.
... of course if something urgent comes up, we can address
it
... We decided at the beginning of the year if we address the
Priorities, we've accomplisehd what we needed from this year.
If we do additional things also, that's great.
... Goal is to have the text of the Process ready well before
AC meeting in April, so we can discuss at the AC section what's
in it and then start the formal ballot shortly after
... That means we need actual text in March.
... so PRs need to be done in Jan/Feb
florian: Sequence of the reviews is slow. All valuable, but we hardly ever get comments during multi-month review. Offline I think it woudl be good to think about how to consolidate. But otherwise I agree.
<cwilso> probably worthwhile to note the AB is nominally in charge of the Process, and thus if there are major issues you should probably raise them with the AB
cwilso: AB is nominally in charge, so good to get them engaged in that earlier, as they have to sign off on the CG's work
jrosewell: I was advised to bring issues to this group
cwilso: yes
jeff: AB hasn't decided on its
spring meeting yet
... and is short on meeting time due to COVID
... AB meeting in January might be last chance for AB
engagement prior to the deadlines
... so getting in sooner than later is probably important
cwilso: Although that's true, we should be able to do async work
fantasai: We also have the 2hr AB meetings, that's equivalent to an F2F, and we have one of those in Feb as well
dsinger: Some major differences
between two proposal
... One major one is whether registry definition and registry
tables are published together
... IETF/IANA split, also ISO
... I think that's the biggest difference in the diff that I
linked in the agenda
florian: You're right this is the
biggest difference
... The other difference
... We both agree that can incorporate registry definition into
a REC
... But both also have ablity for registry definition to stand
alone
... Question there is whether this separate document is a REC
and invokes patent policy or not
fantasai: Interesting question of whether IETF/IANA and ISOC split is due to technological limitations, or at least historic ones. We have the ability to publish multiple-file publications.
florian: We have this point of
disagreement, I think it would be good to get more people
involved in the discussion
... One way to do this would be to issue a survey
... Other ways?
dsinger: Would be nice to have
more people turning up in the meeting
... This meeting is frighteningly small
<wseltzer> [I support the lighter-weight option, which permits publishing separate documents]
[that's not "lighter weight"]
dsinger: AC hasn't given us feedback
<wseltzer> [less restrictive/prescriptive]
<florian> [more moving pieces need more rules to define how they work]
<jeff> [No strong opinions here]
<Zakim> fantasai, you wanted to suggest https://github.com/w3c/w3process/issues/466
https://github.com/w3c/w3process/issues/466
fantasai: It's hard to get feedback when it's open-ended "please give us feedback". It's better for us to send out a structured query, and to the chairs, not the AC, because they're the ones who know what their WG needs
plh: Agree with fantasai, let's send a survey. Let's not send it out now
florian: fantasai and I drafted up a survey. Happy to accept suggestions to make sure it's not biased, but think sending survey is the right way forward
jrosewell: What's a registry and why is it important?
dsinger: Places in specifications
where they have a field that contains a restricted set of
values, where somebody can add a new value with a lightweight
process
... e.g. add new encryption method in TLS, there's an
encryption field in TLS, and you want to add xyz encrpytion
method, get assigned a code name... etc.
... It's a mess at W3C.
... People maintain them in wikis, in separate Notes, sometimes
inlined into REC
... sometimes only maitained in the ED, the official doc being
out of date
... So we have agreement that we go through full review on the
definition of the registry, defines the process of how to add,
what fields there are, etc.
... and then lightweight process (as defined by the definition)
to add/update fields
jrosewell: What's the contentious issue?
plh: Registry is a formal
extension point for our specifications, so they're
important
... The way we actually do the process has been a mess over the
years
... This is an attempt to formalize how we do that
jrosewell: As a new AC member. Clearly needs to be fixed. Not contentious. Just have Team fix it.
dsinger: We're 95% of the way
there, just in disagreement on a couple details of
publication/management process.
... I feel strongly one way, and fantasai+florian feel strongly
a different way
florian: We can't just throw it
to the Team, because once this process exists, these documents
will have the full weight of W3C Process behind them.
... They will be backed by the whole community. So the
community needs to back the process by which they are
created.
... We've mostly, but not entirely, solved that quesiton.
weiler: Where is the best discussion of these isssue for newcomer?
florian: Discussion is a mess,
suggest you start with actual text.
... Two branches that are very very similar
... Suggest starting by reading the proposals
weiler: Links?
dsinger: Action item for me to
provide background on remaining questions
... What are arguments for each way is spattered across many
issues
https://www.w3.org/2020/10/TPAC/talk/Process2021#talk
fantasai: There was a talk on this and the open questions at TPAC AC meeting
<florian> branch 1: https://w3c.github.io/w3process/registries
<florian> branch 2: https://w3c.github.io/w3process/registries-on-rec-track
weiler: Any discussion with IANA to maintain registries for us?
dsinger: Didn't seem
necessary.
... We can handle it already.
florian: We can refer to registries maintained at IANA, but people want to maintain them here at W3C.
weiler: So we haven't explored
that with IANA
... Might raise that
plh: We have the
infrastructure.
... Why would we ask IANA instead of doing it ourselves?
weiler: Thinking about WGs I've
been involved in, often ppl other than chairs are working on
things related to registries
... How would we reach them?
florian: We reach them by asking the chairs. Ask them to answer on behalf of their WG.
dsinger: not hiding from rest of
the community
... We should make sure people maintaining quasi-registries are
aware of the survey and ansewr it
... 3 action items, one is to look at this survey and make sure
it's as crisply worded and unbiased as possible
... action on me and others and others to provide background on
the open questions
... Also action item on me to look at all the open issues on
registries if we can close or combine them
florian: On action #2, isn't that
the presentation from the AC meeting?
... That's what it was meant to be. Whether I succeeded or not
is a separate question. :)
jrosewell: You're looking for feedback, summarizing in a paragraph is better than a presentation cuz' ppl don't have time
fantasai: presentation is slides,
transcript, and video. Can look at whichever. And it's
short
... also we're not asking AC, we're askign WGs. Because we
don't want theoretical purity answers, we want to know what's
needed from ppl who actually need and are working on
registries.
jrosewell: Question about
Membership Agreement
... Process is referenced from membership agreement
... Issue on simplification also needs to extend to the mosaic
of documents
... Membership Agreement isn't subject to this CG
<weiler> to the Q re: dates. 23rd is likely problematic. but could do next week, if you need another hour this month.
jrosewell: Very hard to
understnad exactly what happens when when you look at all these
documents from Membership Agreement
... Where does it go?
jeff: Membership Agreement text
is not my favorite text. Stems from other consortia from 25
years ago, and badly needs updating.
... But updating needs legal team, and is difficult
... spin-out to Legal Entity would be the right time to make
any revisions
... that's roughly 12 months away
... not so far away actually
jrosewell: So gets picked up through LE work rather than this group
jeff: LE gets considerable
attention from the AB
... so would be the right place for that
wseltzer: While we agree that the
Member Agreement has references that are no longer relevant, I
would disagree that it badly needs updating.
... continues to serve purpose of creating consortium
members
... and we are able to update its references, the Process and
Patent Policy, separately
... so hasn't been a priority to update the Agreement
itself
dsinger: agree it's confusing, and some hope to simplify as we move to LE
weiler: If you can simplify the Process document by removing half the words, that'd be great
dsinger: Making it easier to read and use would be wonderful
dsinger: formally, 23rd of
December. Do we want to shift?
... meeting after that would be 13th of January, which I think
is viable
<jrosewell> Would prefer to drop 23rd December meeting
dsinger: Could back it up a week
cwilso: would slide it til after the holidays
plh: Don't think we should send survey until after holidays anyway
dsinger: How about we move up January meeting until 6th?
<jrosewell> Agree. 6th would be best for me.
dsinger: Cancel 23rd, have one on the 6th instead or in addition to 13th?
florian: And let's work on survey asynchronously
dsinger: Yes please, so we can agree to send it out that day
fantasai: AB meeting last week of January, so do we want to shift both meetings up a week, to 6th and 20th?
dsinger: Makes sense to me
RESOLUTION: Next two meetings 6 January and 20 January
dsinger: Thanks everyone for productive meeting! Meeting adjourned.
This is scribe.perl Revision of Date Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00) Present: dsinger jrosewell wseltzer fantasai plh cwilso Found ScribeNick: fantasai Inferring Scribes: fantasai WARNING: No meeting chair found! You should specify the meeting chair like this: <dbooth> Chair: dbooth WARNING: No date found! Assuming today. (Hint: Specify the W3C IRC log URL, and the date will be determined from that.) Or specify the date like this: <dbooth> Date: 12 Sep 2002 People with action items: WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option. WARNING: IRC log location not specified! (You can ignore this warning if you do not want the generated minutes to contain a link to the original IRC log.)[End of scribe.perl diagnostic output]