Community Council teleconference

29 Nov 2012


See also: IRC log


Ian Jacobs, Coralie Mercier, Josh Soref, Suresh Chitturi, Milan Young, Sorin Stefan, Wayne Carr
Josh Soref, Ian Jacobs, Coralie Mercier


<scribe> scribe: Josh_Soref

Ian: I sent an agenda for the blog
... and koalie forwarded it along
... the agenda has a summary, a bit about policy, a bit on infrastructure
... Milan has suggested the template make clear group practices
... any other agenda items people want to go over?

<Ian> agenda: http://www.w3.org/community/council/2012/11/14/scheduling-november-2012-teleconf-on-tpac-2012-discussion/


Ian: immediately following the breakout section during TPAC

<Ian> http://www.w3.org/community/council/wiki/Chair_selection

Ian: i set about having discussions with a variety of people about the Chair selection algorithm
... we were trying to address a situation where some participants in the group felt the chair wasn't listening to some of the group's views
... the extant mechanism intended to give participants control of the chairs
... was underspecified
... so clarifying group members choose their chairs
... the result is a chair algorithm discussion
... at a high level
... Group starts in "Trusted Mode"
... minimal process/no process
... as long as group is able to manage itself
... if there's disagreement
... and a vote of no confidence
... and the next phase kicks in
... where there's some sort of election
... if there continues to be strife, then the Team steps in
... independent of the details
... if you on IRC think that that's a useful phasing (phase 1-2-3) of things

<Milan> +1

Ian: i'd like to see a +1
... Does anyone think that's a bad approach in principle?

[ None ]

<Michael> Why not create an umbrella election system for all groups from the start?

<Ian> Michael, if you wish to participate, please join the call

<koalie> this is Michael Cooper who is on IRC only

<Ian> But to answer your question: to allow groups to operate differently.

Ian: nothing prevents a group from jumping to phase 2
... but i don't want to force anything
... if i can avoid it
... this is a transition to the template tool
... we should give people tools, but i don't want to force them
... Now comes the question
... how do you describe an appropriate phase 2 (election)
... without doubling the current size of the CG process
... i did the process in the long form to try to avoid gaming
... it was hard to do in a small amount of space
... i tried to do something in a wiki
... and then reached a simple algorithm
... and then Wayne Carr proposed a simple algorithm
... I think Phase 2 is
... if we can keep it to a paragraph
... "participants choose their chairs"
... and a new section to the document
... the question is whether people have input on the selection algorithm
... which tries to prevent scenarios

"I'm not happy with the chairing"

half the group says "we're not happy"

chair gets 50 new people to join the group

Ian: there's text to prevent this
... there's text for ties

Josh_Soref: I'm ok with wayne's approach to co-chairs
... and letting them figure out whether one drops out or not

Ian: you could simplify the algorithm
... so it only takes one person to trigger an election

<Suresh> If we could collapse to one algorithm that would be ideal

Ian: there's another piece to prevent another election from happening for 2 months
... to give the newly elected chair some time to prove they're ok
... maybe we don't need the 5 participants bit with the 60 days

Suresh: having 2 aglorithms seems too much
... rather than "simple" and "more complicated"

Ian: i'm only proposing "no algorithm" if there's a problem you have to use "one algorithm"
... the notes are just notes
... to see how much you can get out of the process in as little text
... I don't know which you're trying to say

Suresh: just the latter
... on "5", i think a percentage instead of an absolute number
... groups could be very big

Ian: my sense on that one is
... average group size is 25
... 5 is about 20% of that
... for big groups, you'd never be able to trigger an election
... we chose 5 to express support for a group
... in a big group, you'd basically not have a chance to trigger an election
... do we want that?

Suresh: if there's a chair, and a motion to change the chair
... the people requesting for changing chairs should be significant
... if 20-30% are upset, it makes a case
... but if 5 in a group of 200

Ian: we don't have to resolve it here

<Ian> scribenick: Ian

Josh_Soref: In a big group, only a small number of active participants
... hard to get percentage among inactive participants
... in theory you want more of a quorum.
... you want to force people to nay/yea and ignore those who don't vote.

Milan: +1 to One vote per organization.

<koalie> scribenick: koalie

Ian: if we go %age model, maybe we don't need the 60 days
... Summarizing for Wayne who just joined -- there is support on the call for 1-2-3 phasing, can 5 person trigger an election, or a percentage
... Josh_Soref's amendment is that only people who are active may weigh in
... we don't have to resolve this on the call today
... I want to see whether there's support for current proposal
... either 1) small number of people, or 2) percentage, or 3) percentage of active people
... any objections to any of those

Milan: I don't like 3)

Ian: the idea is you put out a survey, if someone doesn't bother to answer, you ignore them
... the percentage determination is based on those who answered the survey
... this is a "should be a vote".

Wayne: I propose to keep it simple, and go for "5 people"

Ian: OK. We can change that number.
... which is good for groups which have > 5 participants

<Ian> W3C Process appeal is 5%

Wayne: W3C process says 5%

Ian: Yes
... People can continue discussion by list
... thanks for discussion that already happened on the list
... I invite people to annotate the wiki, edit, add comments
... we'll continue to massage that
... summary:
... 2 questions came up
... 1) all groups should use the same algo
... 1-2-3- steps or whether 3-phase approach is preferable
... Second part of policy item is about copyright topic
... initially 1 IPR model
... with two agreements.
... both have copyright commitments and patents commitments
... for groups that don't need patents commitments, we initially said "too bad", but now, we have enough groups in that case that we need to reconsider
... we need lighter-weight commitments.
... one of the risks I heard is that this is going to turn into google groups or such, but I'm not worried
... it's an implementation issue that needs a checkbox
... next question: what is a group needs patents commitments?
... we can close that group and start a new
... so I think all this can be addressed; it implies some changes to the policy, and implementation time
... second topic is changing to CC-BY

wayne_carr: +1 to need for copyright-only gorups

<Josh_Soref> Josh_Soref: +1 for wayne_carr

[Ian gives background]

<Ian> proposal 2 - move copyright to cc-by

<Ian> wayne_carr: sounds fine

<Zakim> Josh_Soref, you wanted to say i'm not opposed

Josh_Soref: not opposed

Ian: next copyright topic has to do with people who need terms that are different from W3C's terms
... because our copyright is permissive, people can be definition, license under alternative copyright terms --that isn't in dispute.
... it should also be the case that you can't force people to grant patent commitments
... question comes up: whether we can to create where we require people to sign other terms --this isn't my current expectation, we'll update the FAQ
... any reactions to that?



Ian: one of the big topics at the breakout session had to do about how we talk about this work
... specifically how to distinguish from WG products
... current mechanisms: CG work is not listed under /TR/ where we list WG products
... we use a different style
... policy says they can't look like a WG product

<Ian> "Community Submissions must not use a style that will cause them to be confused with W3C Technical Reports. W3C may publish additional policies to govern publication of Community Submissions."

Ian: third, we require boilerplate text that this isn't on the standards track.
... Since the launch, people have pointed out a couple of other things
... 1) no use of the word final
... we can certainly change that.
... 2) use of the word "specification"
... we talk about "reports"
... we may not be consistent throughout the documentation and we can fix that
... "report" type of specification vs. type of specification
... what we mean: primer, use-case, etc.
... I don't mind avoiding "specification" as a class of things,
... but I am not sure I want to forbid people to use the word specification

Wayne: I thought they needed to use "community specs"

Ian: two contexts
... one is the title of a document
... I'd push back "SVG specification" but not "fooML specification"
... and then @@, we call them "Community Groups Reports"

Wayne: overall list called "report" is ok

<Suresh> +1 to calling final deliverable a "Community Report"

Ian: you'd push back on using "specification" and replace with "community report"

<Ian> wayne_carr: sounds good to have reports on index, and in title MAYBE add "community" before "specification" if specification is in the title.

Wayne: Yes, that's what I heard people concerned about

<Ian> Suresh: I think calling the class 'Community Report' would work

Suresh: people tend to read specifications with expectations (should, must, etc.)

Ian: CGs are allowed to do that, they can do the sort of specification you're talking about.
... the only question was whether the specification title should say "fooML" or "the fooML specification"
... I heard people shy away from the word specification anywhere.

Milan: I'm uncomfortable with "community specification" just using that term as a title

Ian: FooML will be called a Community Report

Proposed: class is "Community Group Report"

<Milan> +1

Josh_Soref: where is class?

<Ian> From the process:

Ian: the index

<Ian> "The label "Community Group Report" refers to any document produced by a Group. "

Ian: second question about the document's title
... I don't mind "specification" in the title if that what the document does

<Suresh> +1

<Ian> Milan: Suggest that if "specification" is used in title, prefix with "Non-Standard"

<Josh_Soref> +1

koalie: +1

<wayne_carr> it wasn't me saying I don't like specification

<wayne_carr> I think community specification in a title is fine

Suresh: @@

<Suresh> Wouldn't the final deliverable be called a ''XX CG final report" instead of a specification?

Ian: Status aren't required in CGs
... we could have a sub-title that says "draft" or "stable", or other vocabulary

Ian: I'm reluctant to put "draft" in the title as it's metadata
... you could have a watermark that says "not a standard"
... I'm throwing out ideas I've heard

Suresh: need to indicate it's not a standard

Ian: the requirement already exists, I've heard this is not enough.

Josh_Soref: random parsers could easily avoid a watermark
... any form of metadata can easily not ne noticed.

Ian: I wouldn't say there should be one exclusive way
... I want to enhance the current requirement

<Ian> Josh_Soref: I think we need even more nearer the title

Josh_Soref: reporters and managers are likely to see the word specification
... wikipedia's definition of specification is that it's a technical standard

Ian: we shouldn't struggle with the word "specification" which is accurate in some cases, we need to enhance the distinction

Josh_Soref: a report is complete, you got that when it's done

Ian: can call incomplete report a "draft report"

<Suresh> Actually, I don't see current W3C specs using the word "specification" in the title e.g. HTML5, Widget Interface, so the question is how important is this?

<Suresh> Atleast not many of them currently

Ian: proposal "fooML specification, a draft community report"

Suresh: current w3c specs don't use "specification" in the title, it uses the technology

<Ian> http://www.w3.org/TR/tr-date-all

Ian: I agree a proportion of them do

wayne_carr: where it is a specification it's important to say
... and you really end up with one

Josh_Soref: CSS used to have documents with 'specification' in title, they have moved away from doing so

wayne_carr: whether it's in the title or not is irrelevant


Ian: I'm inviting people to suggest alternatives on the list

Milan: boilerplate is not sufficient

Ian: Yes, agreed.

<Suresh> Need to drop off ...sorry!

[Suresh leaves the call]


Ian: documentation (we have a fair amount, we're waiting for new design to fully integrate)

<Ian> http://www.w3.org/wiki/ModernGuide

Ian: and we're also separately working on overhauling documentation for Chairs
... A lot is going to happen in terms of good practices
... Now, on charter habits
... I heard let's have a draft charter template
... we already have a charter generator that is available to WGs
... a static document might be enough to get started.
... anything people want to call out?
... I'd welcome if someone wanted to take a stab at drafting that document.
... coremob cg have already robust material to draw on
... any volunteer?

[no volunteer]

[Ian and Coralie will work on that doc and socialize it]

[Sorin Stefan, who needs to leave the call, is the one who made the CGs design and is working on a new version that is great]

Ian: Milan wanted to cover how we raise awareness in a group about the group's operating procedures

Milan: as long as a group wasn't posing any problem, fine, but when there are and there are no laws, some are needed

Ian: Current process says groups can have operating procedures
... Implementation of your comment: when a group has such procedures, give us the URL and it appears near the "join" button.
... if group has none, we show a warning
... that this group had no additional procedures other than those in the policy

Milan: It has to be a strong warning
... I got badly burned

Ian: do you want to suggest language on the list for such a warning?

Milan: Sure

wayne_carr: Distinguish "Not yet provide" from some other thing
... in other words, groups should want to document their procedures and
... cause the "warning" to go away

<koalie> ACTION: Milan to suggest wording on the list for a warning that a group doesn't have additional procedures beyond the policy's [recorded in http://www.w3.org/2012/11/29-council-minutes.html#action01]

<trackbot> Created ACTION-67 - Suggest wording on the list for a warning that a group doesn't have additional procedures beyond the policy's [on Milan Young - due 2012-12-06].

Ian: I want to avoid requirement about decision-making
... I heard Milan say he's ok with disclosure as starting point

wayne_carr: If someone said "We will listen to you but do whatever we want" then I think it should be shut down

Milan: I agree it should not be a requirement that we have "fairness" in the democratic sense. They do need to disclose.

Ian: I will go back to this

wayne_carr: In the real situation where there are a small number of companies that dominate a market and they decide everything,

<Ian> the question is...does W3C want to enable it or not.

Ian: we recommend consensus but do not require it. Not opposed to some minimal decision process

<Ian> from coremob charter:

<Ian> "It is the Chair's responsibility to ensure that the decision process is fair, respects the consensus of the CG, and does not unreasonably favor or discriminate against any group participant or their employer."

<Ian> http://www.w3.org/community/coremob/charter/

IJ: I would support good practice like that in charter template

<wayne_carr> consensus may not be the right word, since implies near unaniminity, maybe will

<Ian> actions?

[Josh_Soref leaves]

<Ian> ACTION: Ian to work with Coralie on a draft charter template [recorded in http://www.w3.org/2012/11/29-council-minutes.html#action02]

<trackbot> Created ACTION-68 - Work with Coralie on a draft charter template [on Ian Jacobs - due 2012-12-06].

<Ian> ACTION: Ian to review simple algorithm for chair selection and annotate with some options that we discussede [recorded in http://www.w3.org/2012/11/29-council-minutes.html#action03]

<trackbot> Created ACTION-69 - Review simple algorithm for chair selection and annotate with some options that we discussede [on Ian Jacobs - due 2012-12-06].

<Ian> ACTION: Ian to summarize suggestions/considerations regarding document class, title, and status info [recorded in http://www.w3.org/2012/11/29-council-minutes.html#action04]

<trackbot> Created ACTION-70 - Summarize suggestions/considerations regarding document class, title, and status info [on Ian Jacobs - due 2012-12-06].

Ian: next meeting?

<Ian> Proposed: review deliverables by email.

Ian: for the time being we can review deliverable by mail


Summary of Action Items

[NEW] ACTION: Ian to review simple algorithm for chair selection and annotate with some options that we discussede [recorded in http://www.w3.org/2012/11/29-council-minutes.html#action03]
[NEW] ACTION: Ian to summarize suggestions/considerations regarding document class, title, and status info [recorded in http://www.w3.org/2012/11/29-council-minutes.html#action04]
[NEW] ACTION: Ian to work with Coralie on a draft charter template [recorded in http://www.w3.org/2012/11/29-council-minutes.html#action02]
[NEW] ACTION: Milan to suggest wording on the list for a warning that a group doesn't have additional procedures beyond the policy's [recorded in http://www.w3.org/2012/11/29-council-minutes.html#action01]
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2012-11-29 20:58:36 $