Community and Business Groups: Plenary Discussion

31 Oct 2012

See also: IRC log


Josh_Soref, Ian_Jacobs, Daniel_Burnett, Martin_Alvarez, HJ_Lee, Alexandre_Morgant, Yuan_Ji, AlanB, Naomi_Yoshizawa, Coralie_Mercier, Jim_Barnett, Debbie_Dahl, Karima_Boudaoud, Milan_Young, Kaz_Ashimura, Richard_Cyganiak, Suresh_Chitturi, Kiyoshi_Tanaka, dom, Judy_Brewer, Kenji_Baheux, Jean-Michel_Rouger, Stephanie_Tabary, Dave_Singer


<koalie> scribe: Josh_Soref

<koalie> scribenick: timeless_scribe

<koalie> Slides: Community and Business Groups: Plenary Discussion

<scribe> meeting: CommunityZ

Ian: the most time consuming part for me
... is talking to them about the CLA
... what it means to join
... we can improve the UI
... and make it easier for non-members to join

<koalie> [currently, 1934 individual participants]

Ian: examples of successful groups

<cygri> stats on participation: http://www.w3.org/2012/Talks/ij-tpac2012-plenary/#(3)

Ian: Responsive Images
... this CG represents the vision we had for CGs
... you had a guy who had never done standards work
... he went into whatwg
... bounced off
... he formed a CG
... they developed an idea
... it was pitched to WG
... and was used to form idea
... can WGs take up things from CGs?
... they can if they're already in charter
... there are some CGs which have totally stalled
... we've created a CG Council CG
... to document issues
... "Chair Selection"
... in WGs, there's a way to handle Chairs not working w/ participants
... one theme for change is: what's the minimal change needed
... what can we fix through Good Practice v. Rules
... questions about chairs listening to participants
... maybe we can have a default charter
... at first blush, it sounds ok

<koalie> Community Council

Ian: but let's not have it be the default, but let's have them actively choose it

Milan: what was the reason?

Ian: I want people to make a conscious decision
... to adopt the rules

Milan: if we give groups a charter
... and "if you don't like, please describe what you want"

Ian: it could be that
... but people don't necessarily know what they want
... or how they want to run things
... it requires a lot of training
... i don't know that they know everything in advance
... people respond "we're not going to pick our chairs yet"
... i'm like "pick your chair"
... i don't want to force anything

ddruta: Dan Druta, AT&T
... how do you deal w/ overlap?

Ian: CG process says "Groups choose their own chairs"
... process could be strengthened on how one chooses chairs
... there's risks in any approach
... if we vote and 80% say Joe should be chair

<scribe> ... done

UNKNOWN_SPEAKER: problem, there's a denial of service capability
... you could the next day have another vote, and another vote
... how do we address chair selection to help avoid creating a DoS capability
... e.g. only can vote for chair N-weeks after last vote

Ian: today, people can create a group, they're free and fast
... but it was suggested to strengthen chair selection
... another question was Transition of Work from CG/BG to WG
... Patent Policies/Copyright Policies are designed to make this easy
... Contributor Agreement say this:
... if you have a contribution, and part of the spec goes to WG track
... then your commitment follows on to that WG track
... so stuff goes to WG+standard easily
... but not in opposite direction
... CG produces spec, hands off, goes to do other things
... another way to make it possible is Invited Expert policy
... we've made that easier for CGs
... because we want continuity
... we still seek your Membership
... but we make it easier
... it's a grace period
... next, there's WGs finish v1.0
... CG is launched for v2
... but there's no commitment from the WG into the CG
... CGs are allowed to have overlapping

ddruta: XXA

Ian: we recommend charters
... people typically give us short overview
... some give us detailed charter, like CoreMob CG

ddruta: not a full verbose description
... there are 3 specs here, they don't address features A,B,C, we're here to address that

Ian: I love that
... not all CGs are in that position, but yes
... CGs are working on something, want to hand off v1, but want to enhance for v2
... we think that's ok
... Transition problem
... WG could want to take up CG
... or CG might want WG to take it up
... process: WG tech reports must be in English (US)
... but other IGs/CGs could work in another language
... for Patent policy
... we want you to get Final Spec agreement from the 10 most important contributors
... it's a question of community being convinced they have sufficient coverage

Suresh: who decides that?

Ian: in the case, where a WG wants to take up a spec
... the WG decides

Suresh: in resp images
... when it was moved to HTML WG
... was a commitment gathered before the move?

Ian: Yes
... This was for the case where the WG wanted this
... politeness says to talk to the CG before they do this
... we don't have rules against taking up w/o dialog
... but we expect it
... if no WG has scope
... directory proposes to membership
... we'd ask CG to help draft charter

Kaz: Is this for BG too?

Ian: yes, CG/BG are identical
... skipping licensing
... We have CG Council CG
... it doesn't meet regularly
... we're not using it
... infrastructure to build, is Tracking/Dashboard
... what work should i be doing
... we're working on a redesign ofUI

[ time check : 20 minutes ]

Milan: Milan Young
... I understood why you didn't want to impose requiremment for operating agreement
... if a group elects not to have an operating agreement
... a group should post a warning indicating it's doing this
... you said you were looking for an algorithm for Chair selection
... what if you could have an invisible vote of confidence on chair
... if 60% of chair is listed as unhappy
... maybe an email could be sent out

Ian: like France vote of no confidence
... what if you get new members?
... and stack the vote?
... Today, if you go into a group
... produce a report
... "final specification"
... if you go through XG process
... staff/fair principled leadership
... if you get something that sounds weaker
... whatever specification should be signed by the operating agreement
... if that agrement is "wild west"
... so can you reflect the policy of the group for creation into the thing

Judy: appreciate your comment about naming
... w3c has for a long time had problem with the outcome documents
... when we started CG process
... we wanted to avoid having CG process confused with WG process
... for WAI of W3C
... we're ramping up horizontal review
... it wasn't reflected in AC discussion yesterday
... part of W3 process is WG products have cross review
... complaints about CG is
... one CG has "living standard"
... which creates confusion
... your point about "final specification" sounding "more final" is an interesting point

Ian: the document isn't called that
... we have a legal agreement

burn: it doesn't prohibit it

Judy: it would be nice to get meaning clearer

<Zakim> dsinger, you wanted to raise point about needing default/baseline

dsinger: I think CGs need a 'default/baseline' process, as many don't want to invent their own, and I think we need some filtering of proposed CGs to make sure that proposals for CGs are well described etc. -- reduce noise and proliferation. just 2 cents

<koalie> [Josh reading dsinger's note]

Ian: I do that filtering up front
... we sometimes adjust their descriptions dynamically

burn: some have seen this email
... half as a threat, half as a joke, half seriously
... steps in Speech API
... "why can't this occur"
... i create a new CG "HTML 6"
... with me as chair, requires a few people, not members
... I nominate them (Milan, Jim) as editors
... I believe i can create following press release

[ burn reads press release ]

burn: i didn't say "CG", i didn't say "Standard"
... world at large doesn't understand the distinction
... Milan had constructive suggestions
... i don't usually like to complain w/o constructive suggestions
... when people behave properly, you don't need laws/robert's rules
... when you have people who don't see the need to behave
... how do you stop this chain of events from occuring
... with html
... if i did that, there'd be so many people saying "what, this is a crock"
... but this happened in Speech
... there's a narrower area
... fewer people to raise an issue

Ian: great question
... 1. we could have started CGs w/ Robert's Rules of Order
... it was anithetical to what we wanted to do
... it was a conscious choice for minimally constraining
... ok, can be tweaked
... i can't guarantee it'll prevent every bad scenario
... index of reports is different
... can't prevent that by choice
... incrementally add bits to improve situation
... i don't expect a giant process

burn: this is a sequence of events, break it somewhere
... i don't see a way to stop that sequence

Ian: there are some mechanisms

<scribe> ... new proposals

UNKNOWN_SPEAKER: if person is from trusted set
... "member/team"
... it goes to proposed state
... non trusted, requries me to approve

ddruta: a value of CGs is non-member participants
... example of happy was non member starting process
... if you have members starting CGs
... i'd argue that's not trusted
... because they have other opportunities

Ian: that's the one way without Staff interaction

Suresh: we've been in a couple of CGs
... clearly there should be a best practices document
... what is clear scope
... ideally set in stone at beginning
... can't have evolving strategy

Ian: you can

Suresh: there's an individual commitment

Ian: you only commit to your commitments
... good practices don't surprise people

Suresh: be specific
... Scope
... What gaps is this trying to fill
... which WG is this targeting?
... could be a form to fill out
... comment about Living Standard
... set a timeline of expiration for a CG
... set 12/18 months expiration if you don't deliver something w/in time

Ian: we need ways to measure activity of a group
... we've had people who say "don't close us down, we have 500 people"
... we have a plan to contact them
... we need tools for this

ddahl: question of naming
... it's strange
... that document w/ light level of process should be called specification

Ian: why?

ddahl: it carries a lot of meaning
... especially with W3C branding

<Suresh> rename "Final specification" to "Final Report"

ddahl: i read the boilerplate
... i'm not typical
... but people says hey look, i can do X
... the word specification is misleading
... very easy to rename

[ the document Ian showed used Specification ]

[ elsewhere Ian showed places with "Report" ]

jim: call it this report
... and stick expiration date

Ian: we don't want things to disappear from web
... we could have "this is not a standard"

jim: getting "specification" out
... and have votes be by Organization
... what burn proposed for html6
... had more companies behind it than what happened

Ian: spectrum of ways groups operate
... we should explain how they operated
... but people say they don't read
... i don't want blanket statements

jim: ideally it should go to a WG
... we want it to go to a WG
... don't make it look like a WG product
... some guys' smart ideas

Josh: There's a problem with documents in WGs+CGs being non final and lasting for a long time but not being updated to say "this is obsolete"

Ian: there's a better process in CGs
... than in WGs
... we have less maintenance to do
... thank you all, i got a strong message
... join Community Council to help further

[ Applause ]

burn: i wonder if we could have a thing in the logo itself

Ian: or you could have it in a watermark

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.137 (CVS log)
$Date: 2012/10/31 11:10:39 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.137  of Date: 2012/09/20 20:19:01  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/XXX:/Milan:/
Succeeded: s/Burn/Druta/
Succeeded: s/burn:/ddruta:/
Succeeded: s/CG/WG/
Succeeded: s/all/al/
Succeeded: s/XXA/Kaz/
FAILED: s/identicall/identical/
Succeeded: s/didn't think what the result could be//
Succeeded: s/creea/crea/
FAILED: s/broadcasters are complianing having limitation to get slides or irc on time//
Succeeded: s/have as/half as/
Succeeded: s/broadcasters are complaining having limitation to get slides or irc on time//
Succeeded: s/QQ/There's a problem with documents in WGs+CGs being non final and lasting for a long time but not being updated to say "this is obsolete"/
Succeeded: s/Coraline/Coralie/
FAILED: s/Alfred_Newman/AlanB/
Succeeded: s/)/_/
Succeeded: s/gama/zawa/
Succeeded: s|s/identicall/identical/||
Succeeded: s|s/broadcasters are complianing having limitation to get slides or irc on time//||
Succeeded: s|s/Alfred_Newman/AlanB/||
Succeeded: s/Alfred_Newan/AlanB/
Succeeded: s/JoshSoref/Josh_Soref/
Found Scribe: Josh_Soref
Found ScribeNick: timeless_scribe

WARNING: No "Topic:" lines found.

Present: Josh_Soref Ian_Jacobs Daniel_Burnett Martin_Alvarez HJ_Lee Alexandre_Morgant Yuan_Ji AlanB Naomi_Yoshizawa Coralie_Mercier Jim_Barnett Debbie_Dahl Karima_Boudaoud Milan_Young Kaz_Ashimura Richard_Cyganiak Suresh_Chitturi Kiyoshi_Tanaka dom Judy_Brewer Kenji_Baheux Jean-Michel_Rouger Stephanie_Tabary Dave_Singer
Got date from IRC log name: 31 Oct 2012
Guessing minutes URL: http://www.w3.org/2012/10/31-community-minutes.html
People with action items: 

WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report

[End of scribe.perl diagnostic output]