Learning from last Co-Chair election

ChristopherA commented 4 years ago

This last election of co-chairs, under our updated 2020 charter, could be clarified and improved.

Some thoughts:

We didn't in the 2020 charter define membership and precisely who can vote, and W3C doesn't quite define it either. Company, individual, active individual, etc? This email thread captures it fairly well https://lists.w3.org/Archives/Public/public-credentials/2020Jun/0070.html
It wasn't clear to me that the charter says that the only people can vote are those who could vote as of a particular date, which is what the co-chairs chose for this last election to prevent last-minute ballot stuffing. The co-chairs are allowed to choose this under the 2020 charter, but we may want to make this more official in the next charter.
The chairs were challenged by finding technical tool support for voting, especially given membership and date limitations. The w3c collect and volunteer tally was a lot of work for what should be a simple election, but the co-chairs could find no tool to automate this.
Currently if there is more than one nomination for a chair seat, the current chairs decide what form of vote we the community will use. I think it worked reasonably well for this last election, but I see two possible challenges with this in the future:
- When one of the existing chairs is also a nominee, it is possible for there to be a perception of choice of voting system to favor the existing chair. In this last election Joe recused himself, but if Kim and I had not agreed on election style, we could have had an impass.
- Kim and I chose simple "Joe or Heather" choice in the last election, but it probably would have been better to instead have it be a greatest approval. Approval voting also has even better properties when there are more than two candidates. We may want to formally add approval voting style to the charter.

Add your thoughts about other improvements to elections, but as changes to charter, or recommendations to the co-chairs for next election (who again, under current 2020 charter, can decide for each election).

-- Christopher Allen

kimdhamilton commented 4 years ago

To add: I think some people are confused about whether they are members of the CCG for these reasons:

CCG member vs mailing list recipient
DID WG member vs CCG member

ChristopherA commented 4 years ago

Heard from an anonymous CCG member:

The criteria for voting membership should be tighter solely the CCG membership, along the lines of attending at least 3 calls over the last 3 months (or maybe also 9 over the past 9 months) to remain in good standing. I worry about ballot stuffing by those who do not fully participate.

One thing I like about this idea is that it is in our control, unlike the knowing the details of the CCG membership which is under W3C control and can't be easily changed. We already have tools that filter the meeting transcripts for attendees using the +present tag. With that data we can know who has attended regularly.

We can even automate the "scribe" list, moving people to bottom when they scribe, but also remove them from the scribe list if they have not attended for a long time.

Another advantage of this is that we can do something more to encourage scribing, such as having scribed at least once a year to be in good standing to vote, or offer other incentives.

-- Christopher Allen

kimdhamilton commented 4 years ago

from dupe: https://docs.google.com/document/d/10vsEEzy901C7XOpqziqOUaKL_uo41_hV0lJkOYEuLI8/edit

kimdhamilton commented 4 years ago

From another tracking issue:

Two items came up in the vote process.

What is the cutoff for joining? Current position: when the election date is announced
What is a "member". Company level or individual? Current position: individual

Worth discussing and proposing to the group.

wyc commented 3 years ago

We have some proposed updates to the charter here by @vmwvmw and we're ready for some public comments & feedback. If you are interested in this process, please review and leave comments here:

https://docs.google.com/document/d/1YJ2kPg_zzR8oINq9MDf6KcAma21OJtmwAENjzzHe-iA/edit

[x] Send this to the list to get it reviewed
[x] Discuss on our next meeting Jan 5.

ChristopherA commented 3 years ago

On Mon, Dec 21, 2020 at 11:28 AM wyc notifications@github.com wrote:

We have some proposed updates to the charter here by @vwvwvw https://github.com/vwvwvw and we're ready for some public comments & feedback. If you are interested in this process, please review and leave comments here:

I have left two comments:

a +1 for change to approval voting.
on requirements for activity in CCG for nomination: I would prefer over 6 months AND include other measures of CCG activity. Two reasons: the candidate may not have been able to commit to weekly meetings due to conflicts or is active in other CCG meetings or work items; also not uncommon for someone to need to concentrate on a project and then free up to be then commit to CCG.

— Christopher Allen

vsnt commented 3 years ago

We will discuss at 1/6/2021: 10am PT / 1pm ET meeting. Details: https://lists.w3.org/Archives/Public/public-credentials/2020Dec/0208.html

lrosenthol commented 3 years ago

Participation in the CCG comes in many forms - not just attendance at meetings.

For example, I consider myself an active member of the CCG through my involvement in the mailing list and github - but I am unable to attend most meetings due to conflicts. Should I be unable to vote because of that?

wyc commented 3 years ago

Great point @lrosenthol. To address this point, we've added the following language:

A voting member of the Community Group is an individual member of the Community Group who has attended at least 25% of the Community Group public teleconference calls over the 6 months prior to the date of the announcement of the election, which will also be the record date on which voting membership is determined. Attendance is counted through the list of attendees present on meeting transcripts through the “present+” function on the public teleconference calls. A single meeting tally may be substituted by each instance of the following evidence of meaningful CCG contributions: (a) per each role served as owner for active work items or (b) three messages of substance, as determined by the chairs, within 30 days on the CCG mailing list, not to be recounted.

Does this work for you?

ChristopherA commented 3 years ago

This reads like it would require attendance of the least 7 of the last 26 meetings (52 weeks / 2 / 4) OR 21 (7 * 3) substantive posts or issues over 6 months.

Still seems a lot. Even @lrosenthol might be challenged depending on what qualifies as substantial. Do replied to posts or issues count? Would this as a reply count?

Related, I suggest you do the math and show the totals (eg 7 meetings or 21 posts or issues over 6 months , or a mix of) rather than the more obscure 25% and 3 substantial posts counts as one meeting.

jandrieu commented 3 years ago

Unfortunately, I think this approach in general is far too strict.

The community group exists to offer an open forum for people to participate in developing these standards. Restricting participation in governance, IMO, undermines that function.

I do agree we need an explicit boundary on what constitutes membership--and when that determination fixes the voting constituency, with better communication about that boundary and notification through various channels so people aren't surprised they actually aren't members.

However, I believe the right boundary is signing the collaborators agreement (via joining the group on W3C).

Imposing any more restrictions reinforces the "clique" and "cabal" factors that limit the ability and belief that new participants are welcome and recognized as peers.

-j

On Thu, Jan 21, 2021, at 11:31 AM, wyc wrote:

Great point @lrosenthol https://github.com/lrosenthol. To address this point, we've added the following language:

A voting member of the Community Group is an individual member of the Community Group who has attended at least 25% of the Community Group public teleconference calls over the 6 months prior to the date of the announcement of the election, which will also be the record date on which voting membership is determined. Attendance is counted through the list of attendees present on meeting transcripts through the “present+” function on the public teleconference calls. A single meeting tally may be substituted by each instance of the following evidence of meaningful CCG contributions: (a) per each role served as owner for active work items or (b) three messages of substance, as determined by the chairs, within 30 days on the CCG mailing list, not to be recounted.

Does this work for you?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/w3c-ccg/community/issues/136#issuecomment-764884849, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABNTTZYCDV2JLXIL356CP3S3B6K7ANCNFSM4N3V6F3Q.

-- Joe Andrieu, PMP joe@andrieu.net +1(805)705-8651 http://blog.joeandrieu.com

wyc commented 3 years ago

@jandrieu you make good points about inclusivity, and certainly I support inclusivity. However, I'm curious how your proposed approach would address an influx of users prior to an election or vote who have joined the community group with the sole intent to "pack the vote", without any other contributions.

Imagine if a community member was unhappy and requested their 100,000 Twitter followers to join the CCG, sign the collaborators agreement, and then vote a certain way in an upcoming election. If we impose a time delay, these followers may simply lie in waiting to vote as instructed, without contributing to the CCG as many other members do while enjoying the same governance privileges. Do you think we should allow this behavior or am I missing something?

kimdhamilton commented 3 years ago

@wyc I fear the former US president has you thinking about extreme abuses of process that we'd have to guard against. ;)

In general the stricter proposed measures seem more likely to alienate "real" members (e.g. @lrosenthol) than ward off real threats.

I'll assume this obvious straw man is just for discussion purposes; but to state the obvious, yes, let's everyone here is reasonable and wants to avoid the 100K twitter follower scenario case. Moving forward, how can we balance those concerns with the concerns like Leonard and @jandrieu mention?

wyc commented 3 years ago

@kimdhamilton I cannot deny that I am likely biased by recent events--thanks for pointing that out.

I am further agreed that we should err on the side of not alienating membership that have made contributions. Perhaps 6 of any of the following actions over the past 12 months:

Coherent mailing list posts over 140 characters
Weekly meeting attendances
Github commits against CCG repos

@kimdhamilton do you think this a good compromise?

vsnt commented 3 years ago

Heard from an anonymous CCG member:

The criteria for voting membership should be tighter solely the CCG membership, along the lines of attending at least 3 calls over the last 3 months (or maybe also 9 over the past 9 months) to remain in good standing. I worry about ballot stuffing by those who do not fully participate.

One thing I like about this idea is that it is in our control, unlike the knowing the details of the CCG membership which is under W3C control and can't be easily changed. We already have tools that filter the meeting transcripts for attendees using the +present tag. With that data we can know who has attended regularly.

We can even automate the "scribe" list, moving people to bottom when they scribe, but also remove them from the scribe list if they have not attended for a long time.

Another advantage of this is that we can do something more to encourage scribing, such as having scribed at least once a year to be in good standing to vote, or offer other incentives.

-- Christopher Allen

@ChristopherA @jandrieu I'm confused on the push back for tighter requirements, since @wyc specifically used the feedback in this thread to guide the changes. Take a look at the first two comments by @ChristopherA. Are you saying you do not want tighter restrictions which would go against the anonymous concerns that @ChristopherA commented above? If that is the case, how would you address those concerns or would you ignore them?

The charter amendment specifically addressed these concerns, but now we're hearing you don't want restrictions and want it open like it was before.

@kimdhamilton the ballot stuffing scenario was a legacy concern from the last election and is mentioned in the first comment on this issue as well as the one I quoted here. If that is no longer an issue, we can remove addressing it as a requirement.

kimdhamilton commented 3 years ago

@wyc -- I like the direction, but some of them seem hard to measure and may cause unnecessary complications in their specificity, e.g. "Coherent mailing list posts over 140 characters" -- why 140 characters? who gets to decide "coherent"?

Curious what @ChristopherA and @jandrieu think based on their experience. In general, we tended to optimize for the most likely scenarios, thinking that if we had to worry about some extreme case it would be dealt with in other ways (e.g. appeal processes). So establishing metrics like this makes me wonder if we're going in the right direction.

wyc commented 3 years ago

Thanks for the feedback @kimdhamilton. I'm open to dropping the possibly subjective criteria, such as character count or "coherent"-ness. An appeals process sounds like a fine way to proceed without requiring us to define all exceptions upfront--but who then would determine the outcome of the appeals? I'm reluctant to bring W3M into this due to their already heavy workload, but it could be an option for exceptional circumstances. Perhaps we could even ask W3C Advisory Committee members if, e.g., a mailing list post is "on topic" or "off topic" participation with respect to the CCG charter. Looking forward to more feedback from Chris & Joe as well.

jandrieu commented 3 years ago

You set membership qualification at the point the chairs select the election date. That is probably the best you can do, but it is pretty good. It even helps ameliorate the advantage a sitting chair might have from insider knowledge about an upcoming election. Not completely, but it's better than at the announcement date, which could be gamed more easily. Sent from my Galaxy -------- Original message --------From: wyc notifications@github.com Date: 1/22/21 12:35 PM (GMT-08:00) To: w3c-ccg/community community@noreply.github.com Cc: Joe Andrieu joe@andrieu.net, Mention mention@noreply.github.com Subject: Re: [w3c-ccg/community] Learning from last Co-Chair election (#136) @jandrieu you make good points about inclusivity, and certainly I support inclusivity. However, I'm curious how your proposed approach would address an influx of users prior to an election or vote who have joined the community group with the sole intent to "pack the vote", without any other contributions. Imagine if a community member was unhappy and requested their 100,000 Twitter followers to join the CCG, sign the collaborators agreement, and then vote a certain way in an upcoming election. If we impose a time delay, these followers may simply lie in waiting to vote as instructed, without contributing to the CCG as many other members do while enjoying the same governance privileges. Do you think we should allow this behavior or am I missing something?

—You are receiving this because you were mentioned.Reply to this email directly, view it on GitHub, or unsubscribe.

bumblefudge commented 3 years ago

As an infrequent, frequently terse and more frequently incoherent contributor, I thank @kimdhamilton for sticking up for my addled brethren 😆

jandrieu commented 3 years ago

Heather wrote:

@ChristopherA https://github.com/ChristopherA @jandrieu https://github.com/jandrieu I'm confused on the push back for tighter requirements, since @wyc https://github.com/wyc specifically used the feedback in this thread to guide the changes. Take a look at the first two comments by @ChristopherA https://github.com/ChristopherA. Are you saying you do not want tighter restrictions which would go against the anonymous concerns that @ChristopherA https://github.com/ChristopherA commented above? If that is the case, how would you address those concerns or would you ignore them?

As I have said already, you address it by setting the date of eligibility on the date that the chairs decided to announce the election. Not on the date of the announcement, but when the chairs, acting in concert, formally agree to announce.

The charter amendment specifically addressed these concerns, but now we're hearing you don't want restrictions and want it open like it was before.

That is correct. Anyone who has signed the W3C collaborators agreement for the group is a legitimate member. That should be the only restriction. Full stop.

@kimdhamilton https://github.com/kimdhamilton the ballot stuffing scenario was a legacy concern from the last election and is mentioned in the first comment on this issue as well as the one I quoted here. If that is no longer an issue, we can remove addressing it as a requirement.

Ballot stuffing remains an issue. That's why you must have a clear, unambiguous date by which eligibility is set, and that such a date must be as free from meddling as possible. This was an unfortunate oversight in the drafting of the current charter.

However, adding additional hurdles is, IMO, an inappropriate barrier to participation, as the body politic that cares about this work is much greater than the working population who can actively attend meetings and drive the work.

It is important to note that in a functioning democracy, voter turnout is low, as is attendance at town hall meetings and other forums for gathering input. When things are working well, even those who care deeply about the work can happily observe the work without having to become a key cog in every discussion. THAT is the ideal. It's when problems arise that constituents show up and voice concerns. What we need is the opportunity for those who care about this work to choose the leadership they feel will best be able to advance it, without false hurdles that restrict their ability to do so.

-j

On Fri, Jan 22, 2021, at 3:14 PM, vsnt wrote:

Heard from an anonymous CCG member:

The criteria for voting membership should be tighter solely the CCG membership, along the lines of attending at least 3 calls over the last 3 months (or maybe also 9 over the past 9 months) to remain in good standing. I worry about ballot stuffing by those who do not fully participate.

One thing I like about this idea is that it is in our control, unlike the knowing the details of the CCG membership which is under W3C control and can't be easily changed. We already have tools that filter the meeting transcripts for attendees using the +present tag. With that data we can know who has attended regularly.

We can even automate the "scribe" list, moving people to bottom when they scribe, but also remove them from the scribe list if they have not attended for a long time.

Another advantage of this is that we can do something more to encourage scribing, such as having scribed at least once a year to be in good standing to vote, or offer other incentives.

-- Christopher Allen

@ChristopherA https://github.com/ChristopherA @jandrieu https://github.com/jandrieu I'm confused on the push back for tighter requirements, since @wyc https://github.com/wyc specifically used the feedback in this thread to guide the changes. Take a look at the first two comments by @ChristopherA https://github.com/ChristopherA. Are you saying you do not want tighter restrictions which would go against the anonymous concerns that @ChristopherA https://github.com/ChristopherA commented above? If that is the case, how would you address those concerns or would you ignore them?

The charter amendment specifically addressed these concerns, but now we're hearing you don't want restrictions and want it open like it was before.

@kimdhamilton https://github.com/kimdhamilton the ballot stuffing scenario was a legacy concern from the last election and is mentioned in the first comment on this issue as well as the one I quoted here. If that is no longer an issue, we can remove addressing it as a requirement.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/w3c-ccg/community/issues/136#issuecomment-765737698, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABNTT42CS5LKCNF6QD75S3S3IBFNANCNFSM4N3V6F3Q.

-- Joe Andrieu, PMP joe@andrieu.net +1(805)705-8651 http://blog.joeandrieu.com

kimdhamilton commented 3 years ago

That's why you must have a clear, unambiguous date by which eligibility is set, and that such a date must be as free from meddling as possible. This was an unfortunate oversight in the drafting of the current charter.

+1. And thanks for the reminder of the concerns we were trying address at the time. Last election we thought of the concern too late, meaning we had to pick a date in the past, and some votes went uncounted as a result.

vsnt commented 3 years ago

Per the 1/27 conversation, the community decided against considering anonymous comments as principle objections. @wyc will revise the charter amendments to address the objections in this thread.

wyc commented 3 years ago

Hi @jandrieu @kimdhamilton @ChristopherA @lrosenthol

I've simplified the voting eligibility criteria to this:

Any member of the Community Group who has also accepted the W3C Community Contributor License Agreement (CLA) as part of the enrollment process is eligible to vote on elections, charter amendments, and other community matters so long as they have completed enrollment by the date that the chairs announce those events to the public mailing list.

Could you please indicate a confirmatory response if this addresses your concern? This would help us move along.

ChristopherA commented 3 years ago

Could you please indicate a confirmatory response if this addresses your concern? This would help us move along.

+1

kimdhamilton commented 3 years ago

+1

jandrieu commented 3 years ago

+1

wyc commented 3 years ago

The 2 week period has passed without strong objections and we have addressed the concerns above and also across the mailing list. I will publish this to the CCG website shortly and that should resolve the issue.

wyc commented 3 years ago

Resolved

w3c-ccg / community

Learning from last Co-Chair election #136