Jump to content

Talk:Movement Charter/Drafting Committee/Elections

Add topic
From Meta, a Wikimedia project coordination wiki
Latest comment: 2 years ago by Nealmcb in topic Dump of votes

The following Wikimedia Foundation staff monitor this page:

In order to notify them, please link their username when posting a message.
This note was updated on 02/2022

Elections compass

[edit]

@Yair rand: Could you elaborate what do you mean by "completely broken"? I wouldn't want to jump ahead and revert something not working, but I'm not sure what the issue with the compass is --Abbad (WMF) (talk) 10:43, 12 October 2021 (UTC).Reply

@Abbad (WMF): The positions and statements are scrambled, at least 25 of them are misplaced from other candidates, misattributing positions they never held. (I've been told that fixing this is being prioritized.) --Yair rand (talk) 10:49, 12 October 2021 (UTC)Reply
@Abbad (WMF) I don't see a link to the elections compass on the landing page. Ad Huikeshoven (talk) 11:13, 12 October 2021 (UTC)Reply
@Ad Huikeshoven: I removed it (temporarily, until fixed), because of the issue mentioned above. --Yair rand (talk) 11:17, 12 October 2021 (UTC)Reply
@Yair rand: I find it highly controversial that you as a candidate decide to edit what information voters see or don't see on the official pages. Just for the record and probably stating the obvious.
However, it would be helpful to know if user:Abbad (WMF) was able to replicate your problem (whatever it may be - unclear) and if yes - whether it has been addressed (given the exposure the compass has received). Effeietsanders (talk) 23:44, 17 October 2021 (UTC)Reply
@Effeietsanders: Very shortly after I made the above comment (and associated edit), the compass was taken down on the toolforge end while it was being fixed. Afterwards, the compass link was then re-added to the election page pretty promptly.
The original source of the problem was that the (over 1000) candidate inputs were being manually copy-pasted into the system, resulting in numerous mistakes, such as position statements from one candidate being placed as that of another candidate. (I'm aware that my direct involvement in editing the election page to remove the link was less than ideal. I further admit that I was additionally biased by the fact that statements I never made (in particular, of positions I strongly disagree with) were being falsely attributed to me, which made me extremely uneasy.) --Yair rand (talk) 00:07, 18 October 2021 (UTC)Reply
@Effeietsanders: Thank you for the valid concern. I have not replied here because Cornelius (who is leading on the compass) followed up himself on a another talk page about these which were, later, resolved. For the sake of clarity and summarizing that discussion, there were two separate issues that were addressed in the above discussion and led to the temporary removal of the compass:
  1. The copy/paste issues that Yair mentions, which led to some errors in attributing candidate statements, although the compass was still working.
  2. A reset procedure that Cornelius was working on, which caused the compass to be offline for about a day.
And yes, ideally candidates wouldn't directly edit, but since there were issues being discussed I believe it made sense to temporarily remove the compass link. Now that both issues have been resolved and the compass link is back, things should be fine! --Abbad (WMF) (talk) 07:33, 18 October 2021 (UTC).Reply

Batchinput

[edit]

Hello is there a possibility of an alternative input mode for the election. I think it is difficult to find a person in the long list. I thinked about a possibility of a batch input through a text field where all Candidates that someone wants elect are inserted. One per row and so the one in the first line has the highest priority and so on. After I dont have experience what needs to be there that it is secure I dont know if it is possible but it is something that I think could be helpful.--Hogü-456 (talk) 17:40, 12 October 2021 (UTC)Reply

Thank you for the question! I don't know whether that would be feasible. I can check. Something that could facilitate filling in the fields would be using the letter of the candidate you are searching in the dropdown box, as it will take you to the candidates with that letter. Still it will take time with so many fields --KVaidla (WMF) (talk) 15:55, 13 October 2021 (UTC)Reply

Copy&paste mistake

[edit]

This is a copy&paste mistake from the Board's elections: "On the voting page is a sequence of dropdown boxes. Starting from the top of the page, select the candidates in order of preference, from "Preference 1" (most preferred) to "Preference 19" (least preferred)." Preference 19 does not make sense here and should be changed to Preference 70. 4nn1l2 (talk) 17:44, 12 October 2021 (UTC)Reply

It was left in intentionally, as there is the highest value of doing preferred voting for the top candidates (as suggested on this landing page). However, we have received the notifications regarding this from a number of users, so it seems to be distracting and we have changed the number to 70. --KVaidla (WMF) (talk) 15:58, 13 October 2021 (UTC)Reply

Heads up for alternates

[edit]

You are voting to elect 7 Movement Charter Drafting Committee members and 2 alternates

The current implementation of the Meek voting system may not give you 2 alternates. I assume that your alternates are the "best losers", but SecurePoll does not rank unelected candidates at the moment. While I could easily rank all the candidates in the Foundation's latest elections, this is not always the case. This needs the attention of User:NKohli (WMF).

Please see phab:T288185 for more details. 4nn1l2 (talk) 17:51, 12 October 2021 (UTC)Reply

Thank you for surfacing this, 4nn1l2!
Indeed, how the Single Transferable Vote is set up in the SecurePoll it is not designed for also electing alternates. We have another restriction in these elections, which complicates things even more, as there is a cap of maximum 2 people from the same main wiki project to be elected to the committee. ::For caps: it is possible to run the elections as is and remove the candidates who were elected last and transfer them to the last who were eliminated OR it is possible to eliminate the same main wiki project candidates as soon as two have been elected, to allow transfer of votes.
'For alternates: it is again possible to run the process as is and have the last two who were eliminated as alternates OR it could be that the process is run for 9 seats and then the last 2 who are elected will be alternates OR it is possible tor un the process as is now, keep the votes, and if alternate is needed, re-run the process with the person who has stepped down being removed.
It does indeed make sense to discuss it in further detail with technical experts to see what are the most feasible options. I will return to this thread once we have a defined, technically feasible approach. --KVaidla (WMF) (talk) 16:15, 13 October 2021 (UTC)Reply
Regarding alternates, the only solution is to fix SecurePoll to rank all candidates. Currently, SecurePoll breaks counting votes as soon as k (7 in the current elections) seats have been filled, but to determine the "best losers", the software should continue voting. In the Foundation's latest elections, the last seat filled in round 18 and all candidates got ranked automatically. But here we have 70 candidates and you want to elect only 7 of them, so it is very possible that the last seat gets filled in, say, round 50, and then SecurePoll stops counting. But you want 2 alternates too and normally they are the best candidates among the "losers", and these best losers are only determined in the last round (round 69). A real PROBLEM.
One might consider changing the elections settings and select 9 seats rather than 7 seats. But this is not at all a solution because it may change the elected candidates. Please see Talk:Wikimedia Foundation elections/2021/Votes for example. If you set k=4 and select one alternate (the "best loser"), you will end up with Rosie, Victoria, Dariusz, and Lorenzo as elected and Eliane as alternate. But if you set k=5, you will have Rosie, Dariusz, Victoria, and Eliane as elected, and Lorenzo as alternate.
Re-counting the votes is also not a solution. This again may change the body. Suppose you have A, B, C, D, E, F, G as elected members of the body. A leaves the project and their seat becomes vacant. You may think that if you re-count the votes, you will have B-G as members again and X will be added instead of A. But this may not be the case. You may end up with a completely different body composed of, say, B, M, N, D, I, J, and X! You can't change the members of the body just because one person has left the project. 4nn1l2 (talk) 05:57, 14 October 2021 (UTC)Reply
Thank you for providing this insight regarding the mathematical implications running the Single Transferable Vote. As I said, we are already looking into it and we will certainly plug this analysis into the conversation. We will see what are the options regarding SecurePoll fix and try to find the best solution. Your contributions are really appreciated. --KVaidla (WMF) (talk) 13:24, 14 October 2021 (UTC)Reply

@KVaidla (WMF): Thanks for all your efforts. Winners have been announced recently, but not alternates. I would like to know if they have been determined, and if yes, how. It would be great if you publish the full results (including round table) like Wikimedia_Foundation_elections/2021/Results#Rounds_table. I would like to know how votes have been transferred, especially for the MENA candidates. Thanks 4nn1l2 (talk) 17:45, 1 November 2021 (UTC)Reply

@4nn1l2: Thank you for the question! Results table has been published here. Regarding the alternates the suggestion was that instead of nominating the "best losers" or last eliminated, we would re-run the tally with elected members marked as eliminated. We have not yet run this script. --KVaidla (WMF) (talk) 18:02, 1 November 2021 (UTC)Reply

Are Candidates allowed to vote

[edit]

May sound silly but just wanna confirm Uncle Bash007 (talk) 12:15, 13 October 2021 (UTC)Reply

@Uncle Bash007: "Are Candidates allowed to vote?" Yes. For more info, you read Movement_Charter/Drafting_Committee/Set_Up_Process#Election_process. ✍️ Dušan Kreheľ (talk) 12:47, 13 October 2021 (UTC)Reply
Not silly question at all. I am adding the response with reference:
The eligibility criteria are following:
  • not be blocked in more than one project;
  • and not be a bot;
  • and have made at least 300 edits before 12 September 2021 across Wikimedia wikis;
  • and have made at least 20 edits between 12 March 2021 and 12 September 2021.
Candidates who meet these criteria are eligible to vote. --KVaidla (WMF) (talk) 16:00, 13 October 2021 (UTC)Reply
Ok thanks for the info. really appreciate — The preceding unsigned comment was added by Uncle Bash007 (talk)

Tools for examining candidates

[edit]

These two tools may be useful for folks trying to evaluate the 70+ candidates.

-- Fuzheado (talk) 01:16, 14 October 2021 (UTC)Reply

The table would be A LOT more useful (ie actually useful) if the top or total number of edits on projects were included! On the candidates I've looked at so far, the total ranges between over 500,000 and 10 (yes, ten). Johnbod (talk) 17:59, 14 October 2021 (UTC)Reply
@Johnbod There's a table with total edit counts at User:Guerillero/MCDC Voter Guide, which I've found useful. the wub "?!" 21:37, 23 October 2021 (UTC)Reply
Yes, thanks - found that now & it is very helpful - actually they range from over 1.5 million to 10! Johnbod (talk) 03:40, 24 October 2021 (UTC)Reply

✍️ Dušan Kreheľ (talk) 08:20, 24 October 2021 (UTC)Reply

Changing the vote

[edit]

It said somewhere that once you vote you can change your vote later. How can I do that? --Gereon K. (talk) 15:46, 14 October 2021 (UTC)Reply

@Gereon K.: I think you can simply go to the voting link again and recast it. Did you try that? --Abbad (WMF) (talk) 16:10, 14 October 2021 (UTC).Reply
Hi Abbad (WMF). I tried that, but just saw a blank voting page with none of my old votes. --Gereon K. (talk) 16:14, 14 October 2021 (UTC)Reply
@Gereon K.: While I'm not expert on the SecurePoll, I believe this is how you'll have to recast it, but I'll double check for you --Abbad (WMF) (talk) 07:46, 18 October 2021 (UTC).Reply

Confusing ending time (AoE)

[edit]

The voting starts in UTC time and ends in AoE, and the latter is now mentioned in many Central banners (with no link, just the acronym). I don't think that AoE is well known, or at least I had never heard of it before, and the Wikipedia article exists only in four languages. And I still don't understand when the voting actually ends where I live. I think it should be changed to UTC and in the future only UTC should be used in the banners (and in elections in general). Even though people might not know what time UTC is in their own timezone, at least they have heard of it. kyykaarme (talk) 20:54, 14 October 2021 (UTC)Reply

Agreed. AoE is nonsense.
By the way, voting ends on 25 October 2021, 11:59 UTC. 4nn1l2 (talk) 20:58, 14 October 2021 (UTC)Reply
Folks, AoE is not hard. All it means is something ends on a given date. Doesn't matter where you are on earth. That's it. Don't worry about timezones. It's that simple. - Fuzheado (talk) 00:43, 15 October 2021 (UTC)Reply
Just visit the English Wikipedia and see the text above your watchlist: Voting has begun for the Wikimedia Movement Charter Drafting Committee election! Voting ends at 23:59, 24 October 2021 (UTC). (permalink) The fact that the biggest project on wiki-verse is wrong about when the voting ends, refutes your last sentence that it's simple. It's total nonsense. 4nn1l2 (talk) 01:34, 15 October 2021 (UTC)Reply
Now the watchlist notice says Voting ends at 11:59, 25 October 2021 (UTC), while the Central banner says Vote until October 25!. kyykaarme (talk) 17:56, 15 October 2021 (UTC)Reply

What a farce! Look at votewiki:SecurePoll:1208. According to the SecurePoll settings, voting ends at "2021-10-25T00:00:00Z". Was it really necessary to use such as ridiculous timezone and confuse many many users? 4nn1l2 (talk) 18:52, 15 October 2021 (UTC)Reply

@4nn1l2: Hm, If You look here, all is ok: https://vote.wikimedia.org/w/index.php?title=Special:SecurePoll/entry&sort=el_end_date&limit=50&asc=&desc=1 ✍️ Dušan Kreheľ (talk) 19:18, 15 October 2021 (UTC) & ✍️ Dušan Kreheľ (talk) 21:15, 15 October 2021 (UTC)Reply
Very very confusing! The start time is "10:00, 12 October 2021" [where the heck 10 comes from?] according to [1], but "2021-10-12T00:00:00Z" according to [2]. By the way Z indicates UTC as mentioned here. Probably someone has manipulated (i.e., fixed) the start and end dates using command-line interface. Still it demonstrates how confusing the AoE concept is. Granted, AoE is simple, but not at all in common use. Most of Wikipedians are regular folks, not IEEE engineers. 4nn1l2 (talk) 19:33, 15 October 2021 (UTC)Reply
Why the 10th hours, look this. I like the AoT idea. ✍️ Dušan Kreheľ (talk) 21:15, 15 October 2021 (UTC)Reply
That doesn't explain anything. 12 would be understandable, but 10 not. I urge the Wikimedia employees to drop the idea of using this ridiculous timezone in their future elections. It only wastes volunteers' time and makes many people confused as seen here. Nobody looks smarter if they use obscure concepts such as AoE. 4nn1l2 (talk) 21:40, 15 October 2021 (UTC)Reply

Given the table below (from the Anywhere on Earth article on enwiki), now I understand where 10 comes from.

The day 2021-12-24
The time The country or the time zone
The first second of day 2021-12-23 10:00:00 UTC UTC+14
Pacific/Apia, Pacific/Kiritimati
The last second of day 2021-12-25 11:59:59 UTC UTC−12

But it raises another question. Why has the start time been expressed in UTC but the end time in AoE? Why couldn't you stick to one of them? At first, I thought that AoE is just for end dates, but now I see that it can also be used for start dates.

This table also shows AoE is more complicated than one might think. It's not always add 12 more hours, as one may first think [Anywhere on Earth (AoE), a calendar designation for UTC−12:00]. Sometimes you need to add 12 hours to the end dates and sometimes you need to subtract 14 hours from the start date to obtain the actual times! Totally ridiculous. 4nn1l2 (talk) 23:31, 15 October 2021 (UTC)Reply

My main/only concern is the Central banner and how it confuses users. For example, the translation of the Central banner for the Portuguese language says "Vote até 24 de Outubro, 23:59 (AoE)!". I went to the Portuguese Wikipedia, searched for "AoE", and through a redirect I ended up in the article for pt:Age of Empires. Volunteer users do not know what AoE means, and it should not be used in the central banner. The banner is supposed to inform users about something instead of confusing them with an obscure (and English!) acronym that is used almost Nowhere. on. Earth. I guess the AoE translations are not reviewed so they haven't been pushed to the projects yet, and instead the live banners mention the 25th of October which was in the banner originally (and still is in the English banner). I edited the banner in my language to include the ending time in UTC. Here are all the translations [3]. -kyykaarme (talk) 13:18, 16 October 2021 (UTC)Reply

Password for voting

[edit]

What am I missing? My sign-in name and password are valid on all the sites. For instance, I can edit this page under my own user name. But when I try to vote on this, it wants me to enter my user username and password, and then tells me my password is invalid. Only there, and no place else. What is going on with that? Maile66 (talk) 21:33, 14 October 2021 (UTC)Reply

@Maile66: To be honest, I'm not sure what you're describing (I've tried entering the poll with two accounts and it doesn't ask me to sign in). I'm wondering if you've tried to use the "login" button in the SecurePoll website, in which case yes: Your normal login credentials won't work, but you do not need to login. Just go to Special:SecurePoll/vote and you'll be able to cast your vote. If you go the the main page, it says the following:

This wiki is maintained only to host elections that use SecurePoll. There should be no need to edit it, and you do not need to be logged in to vote. Only a limited number of accounts exist on this wiki.

If you still have an issue, though, please ping me here or privately and I'll get you in touch with someone who might help --Abbad (WMF) (talk) 07:44, 18 October 2021 (UTC).Reply

Voting options

[edit]

Why isn't there an option to cast a negative vote? To express you don't want a certain candidate to be on the committee! --Sb008 (talk) 10:59, 18 October 2021 (UTC)Reply

@Sb008: This elections follows the model of the most recent Board of Trustees elections in using the single transferable vote method. As you've probably seen, this means that you express your preference to candidates by ranking them. If you don't want a candidate, you'll typically have to rank them last (which, I get, is a decent challenge with the amount of candidates here). Because of the voting method, however, not ranking a candidate at all is perhaps your best option: Since this is a "single vote", it's the most guaranteed way for your vote never to go to them --Abbad (WMF) (talk) 13:47, 18 October 2021 (UTC).Reply
@Abbad (WMF): You suggest 2 options
  1. Rank all and put the undesired person last. -- This doesn't express my desired result, since I have to rank a few dozen people on which I've no opinion and would, if possible, rank all equally. Basically the neutral vote.
  2. Not ranking the undesired person at all. -- This also doesn't express my desired result. It would put the undesired person at the same level as all other peope I don't rank. And they are, for me, not on the same level, where the others are concerned, I've a neutral opinion, whereas on the person in question I've a negative opinion.
Suppose we have a single fictive person and in an election you can vote suitable, unsuitable or neutral. Suppse the voting result would be suitable 20%, neutral 15% and unsuitable 65%. Clearly the fictive person wouldn't be voted in. Now we turn to the system used here. Basically we can only vote suitable (with a differentation option between all those suitable) and neutral. Suppose all people vote the same, the result regarding this fictive person would be suitable 20%, neutral 80% (since you can't vote on unsuitable and only rank so low or not at all it effectively becomes neutral). A strong "no" is turned into a negligible "yes". With a 20% support, a person could very well be voted in since all opposition is eliminated from the voting process. A voting process in which all opposition is eliminated is not a fair process. Suppose this would be survey about hotels. If you elimiate the 65% of bad reviews for a specific hotel, the hotel could very well end up in the top 15. If on the internet all negative reviews for hotels would be deleted, we would consider it fraudulent. Not hard to figure out how I rate this election process. Kinda funny, a committee which, among others, is supposed to focus on democratic aspects, is elected in an, undemocratic would be to strong, so let's call it a flawed election process. Perhaps a 2 step election would be better. In step 1 the community could decide on whether someone is considered suitable and step 2 would be this election with the difference that people considered unsuitable in step 1, will not participate in step 2. --Sb008 (talk) 18:38, 18 October 2021 (UTC)Reply
I voted and ranked all candidates. Just filling out the form took about 15 minutes. I had prepared a list with all the candidates in my preferred order (which took DAYS). I cared a great deal about people I wanted to support and also about people that I absolutely do not want to see elected. What would have worked better for me would have been a list of all names, scrambled, in random order, with a UI where I can drag them into place (this would make the experience significantly better on a mobile phone) and then all I had to do is move my most preferred candidates up, and my least preferred candidates down. That way we'd accomplish two things: voting is easier, and everyone ranks all the candidates. Win-win. Vexations (talk) 18:58, 18 October 2021 (UTC)Reply
@Vexations: I'm sincerely sorry about the problems that you're facing. We have discussed multiple times, before the elections, possible modifications to make the interface a little more use-friendly. Unfortunately, though, the current SecurePoll system allows for very little changes of this kind. One of the obvious lessons of this process is the need for UX improvements for future elections --Abbad (WMF) (talk) 15:36, 22 October 2021 (UTC).Reply
No need to apologize, but if you want to quote a random user as advocating for a VERY significant investment in the free and open source software tools that facilitate collaboration, even at the cost of scaling back many of the foundation's initiatives, then feel free to quote me: "Before we try to effect change, we need to build the tools that enable it. Much of what we use is not fit for purpose. Spend the money (on grants for wildly innovative software for example). We need radically better tools for massively multi-user online consensus-building". And no, I don't think we should rely on corporations to build a commercial tool like Facebook's Horizon that we could use instead. Vexations (talk) 16:03, 22 October 2021 (UTC)Reply
@Abbad (WMF): I absolutely agree with Sb008. The fact that we cannot cast a negative vote is a huge problem with the current voting system, and the fact that it was used before is not a justification to keep using it. --Ita140188 (talk) 04:31, 19 October 2021 (UTC)Reply
@Sb008 and Ita140188: While I haven't personally supported or made a decision on this voting mechanism, I'm not sure if it's accurate to describe the lack of votes "against" a candidate as a flaw in "democratic aspects". AFAIK, in many or most political and democratic elections you are given the option only to vote for candidates, not against them. That is not to say that your concern are not valid: This is, indeed, different from the way Wikimedia elections are traditionally held. After the elections are complete, we will be evaluating how the process went and what are some lessons for the future, and this could well be one of them depending on the useful feedback that we are getting from you (and hopefully others) --Abbad (WMF) (talk) 15:33, 22 October 2021 (UTC).Reply
@Abbad (WMF): You mention other political and democratic elections.
These elections in general differ in 1 or more of the following aspects.
  1. They are elections between a select few number of people, who we mostly know, for a single position. E.g. presidential elections.
  2. If the elections are about more than 1 position, they are in general party based. E.g. parlement elections. Although you vote for an induvidual, the election is party driven. Not likely you'll vote for a member of a party which ideas you don't like. In general you choose the party first, and next you vote for the candidate you consider best of that party.
  3. You know (more or less) the individual candidates or the views of the parties. What they stand for.
This election completely differs. Most candidates you've never heard of. E.g. a candidate who operates on the Japanese Wikipedia only, I don't know cause I never operate there. Depending on your origin and the origin of the candidates, in the worst case you don't know even 1 of them and if you lucky you know a handful of them. So most of us have to vote on people of who you know none or just a few. So what do we have to base our vote on? You'll probably say, they all present themselves and express their views. My response is, actions speak louder than words. It's not uncommon that candidates speak of what voters like to hear. It's also not uncommon that of all promises made, few become true. And with all respect, I don't have time to check 50+ candidates for their actions in the last couple of years. People I don't know at all, I don't vote for, not a positive nor a negative vote. Basically I remain neutral. The few I do know, I would like to vote on. Those of which I've a positive view, I like to "reward" with a positive vote. But those of which I've a bad view and think, I don't care who but not him/her, I like to express this too, by a negative vote. I cannot do this, I can only not vote for this person. And that's not right, cause it's effectively a neutral vote.
In the current voting system it's possible a certain government will make sure they get a pawn on the committee. It's not hard for this givernment to organise 500, or 5000 or whatever amount of votes is needed to get their pawn elected. Since negative votes are not allowed, it's almost impossible to keep this pawn out of the commitee. So yes, the voting system is flawed and wide open for corruption.
Maybe you didn't support or made a descision in this voting process. But those who did remain silent. One of the first things each politician learns, if you get caught on something flawed, a lie or whatever, stay quiet and the sooner it will pass by without any changes. --Sb008 (talk) 00:44, 25 October 2021 (UTC)Reply

Votes for project-constrained candidates redistributed?

[edit]

@Abbad (WMF): I have an urgent question about the rules for the STV distribution that was not covered by the rules.

The rules of the election state that there is a constraint of no more than 2 elected members per wiki project (Movement_Charter/Drafting_Committee/Set_Up_Process#Election_process).

Will votes for candidates who are in the top 7 but disqualified under this rule be redistributed as if the candidate was eliminated through STV?

If not, this will be fundamentally unfair as a substantial amount of STV votes will be "wasted" on the candidate when they could have influenced the rest of the election.

Without accounting for this constraint, the election process will be quite unfair and will discourage people (including myself) from ranking their true preferences. Best, KevinL (aka L235 · t) 15:48, 21 October 2021 (UTC)Reply

@L235: Thank you for raising this and I get the urgency. I've forwarded your question to my colleagues yesterday, since it's outside my technical expertise. Just wanted to assure you that I'll respond (or they will) once I have a confirmed answer --Abbad (WMF) (talk) 14:56, 22 October 2021 (UTC).Reply
Hi Abbad (WMF). I understand that teams are resource-constrained but it's deeply unfortunate that this question hasn't been resolved before the close of voting in a few hours. I hope that when the votes are tallied, the following point be made clear:
  • If two candidates from the same wiki project have been elected after a particular round of STV, all other candidates from the same wiki be immediately eliminated and their votes transferred before any other candidates are eliminated or elected.
That appears to me to be the only way to retain fairness for those voters who ranked more than 2 candidates in the top 7. Best, KevinL (aka L235 · t) 05:31, 25 October 2021 (UTC)Reply
@Abbad (WMF): Hello! I'm somewhat surprised this wasn't picked up during the planning phases for this election - vote redistribution is the core tenet of STV effectively, and to 'tinker' with its mechanics (i.e. in this case only allowing 2 members per project) can lead to all sorts of unintended consequences if not properly thought through. I agree that KevinL's solution (eliminating constrained candidates as standard and redistributing) is the only fair one. Either way, I'd hope to see it made clear in the final results which method was used in order that the results can be properly interpreted. I realise this isn't your direct area of expertise, and that you aren't responsible for the voting system, but would ask that you pass my comments along to your colleagues. Thanks! firefly ( t · c ) 06:19, 25 October 2021 (UTC)Reply
@Abbad (WMF): This is probably obvious, but: If there are any remaining ambiguities about how the results will be calculated, it is essential that they be figured out before opening any of the votes. If necessary, maybe consider delaying unsealing the votes. --Yair rand (talk) 08:03, 25 October 2021 (UTC)Reply
Let me echo that I (really) think Kevin's solution should be implemented but in any regard this needs to be decided and communicated clearly before "opening" the votes per Firefly and Yair. Best, Barkeep49 (talk) 18:33, 25 October 2021 (UTC)Reply
I didn't notice this piece of the election rules before now. I understand the sentiment -- wanting linguistic & project diversity in the pool -- but I don't think it makes sense in practice. Most of these candidates have a long and impressive resume in the Wikimedia world, working across multiple projects and sometimes across multiple languages. In addition, many of them (as is natural for this kind of meta-governance work!) have focused their efforts on 'real-world' outreach and affiliate work, including governing affiliates and the Foundation -- which I can tell you from experience doesn't leave a lot of time for hands-on project editing. I'm not a candidate, but thinking about myself, I would have a hard time telling you if I consider my "home project" to be Wikipedia -- or perhaps Wikimania, or the WMF itself, where I've put in just as many hours of work. I would encourage the committee to drop this requirement, especially if you haven't figured out how it works with STV. -- phoebe | talk 14:51, 25 October 2021 (UTC)Reply
I absolutely oppose changing the rules now. This restriction is superb and makes the committee more diverse. I encourage the WMF to uphold this requirement. Thanks 4nn1l2 (talk) 08:54, 26 October 2021 (UTC)Reply
I guess the constraint is out of question. It does not make sense to change the rules at this point. But how will the constraint be applied to the count process is still undefined? MarioGom (talk) 14:02, 26 October 2021 (UTC)Reply
Response - method to be applied in tallying the votes

Thank you so much for raising this important question, KevinL. Thank you for your contributions to the discussion Firefly, Yair rand, Barkeep49, phoebe, 4nn1l2 and MarioGom.

I am truly sorry that we did not get back in time prior to closing of the vote. We do understand that how the STV is being implemented has a potential effect on how people vote. We were still weighing technically feasible solutions and fairness of the approaches and this is where we have landed:

  • We believe that implementing the constraint is necessary to keep the integrity of the process.
  • We will run the STV script as is until the 7 people are elected.
  • In addition to 2 people from the same project cap, we will also name 2 alternates as the result of the elections. This will help us to avoid rerunning complex processes if any of the elected members should step down.
  • This means that if a third candidate from the wiki, where already 2 members have met the quota and have been elected, will also meet quota, they will be moved to the alternate seat and be replaced by the person who was eliminated last.
  • There is no transfer of votes in this approach from the candidate that gets moved to the alternate seat from being elected. We have estimated that occurence of such situation in early rounds is unlikely and in later rounds it will have lesser negative effect.
  • We will tally the votes based with this methodology and will document lessons learned for future instances.

I am available for any further questions or specifications regarding this matter. --KVaidla (WMF) (talk) 19:54, 26 October 2021 (UTC)Reply

@KVaidla (WMF) can I understand more about how We have estimated that occurence of such situation in early rounds is unlikely and in later rounds it will have lesser negative effect? Thanks, Barkeep49 (talk) 19:58, 26 October 2021 (UTC)Reply
There's lots here that is surprising and upsetting to me, but regarding technically feasible solutions, this seems like a spurious reason. If need be, we could do this right in Excel. It certainly isn't ideal, but giving certain people's votes significantly less weight for reasons that were not made clear in the election rules is far less ideal. As for We have estimated that occurence of such situation in early rounds is unlikely and in later rounds it will have lesser negative effect., I don't see how you could know this without looking at the votes and I also don't think it's true; I know how I voted, and I'd be surprised if this issue doesn't negatively affect my vote. Best, KevinL (aka L235 · t) 21:08, 26 October 2021 (UTC)Reply
@KVaidla (WMF) - with regard to There is no transfer of votes in this approach from the candidate that gets moved to the alternate seat from being elected. Why not? It’s eminently possible to do the transferral of votes even in this scenario. firefly ( t · c ) 21:11, 26 October 2021 (UTC)Reply
@KVaidla (WMF): can I ask how this was not determined well prior to vote close? - assuming I'm reading the Telegram log right (and please correct if not), this was raised on the 12th October, and it looks like the team was already aware of the options at the time. You also note technically feasible solutions - multiple runs of the results depending on what happens would seem to avoid all of these issues. If needs be, the anonymised details can be transferred out Securepoll and into Excel. As the data log is being made publicly available, anyone will be able to run their own versions of these checks to ensure that no error has occurred between SP and Excel or in Excel. Nosebagbear (talk) 21:31, 26 October 2021 (UTC)Reply
As a follow-up note, and will document lessons learned for future instances. doesn't hold up. There will be future (s)elections, but there won't be any future MCDC (s)elections in the foreseeable future. It doesn't have a fixable future process, so we need to avoid the issues now Nosebagbear (talk) 21:53, 26 October 2021 (UTC)Reply
Re technical feasibility: If it would help, I am confident that there are plenty of technically-inclined volunteers who would be glad to help with anything needed on this, even in the limited time remaining. This doesn't need to be restricted to pre-existing software. --Yair rand (talk) 22:05, 26 October 2021 (UTC)Reply
Response to the points raised
Mistake in assessment - Thank you for surfacing the points regarding the assessment! I have revisited it and made a significant omission that led to misinformed decision. Namely, I was concentrating too much on the elimination rounds, so at some point I managed to overlook the potential of meeting the quota in earlier rounds. The project with most candidates is English Wikipedia with 10, which means that on grounds of elimination the cap would be triggered in final 15 rounds. However, the same project has most voters and it is possible that with uneven distribution, candidates from the same project could meet quota in rather early rounds. This changes significantly the potential impact of votes being transferred or not transferred. I have revisited the assessment and have to admit my profound mistake.
Technical feasibility - SecurePoll with its current setup has limitations for tallying the votes. Unfortunately, we are not currently in the position to change that. As a result, an approach was suggested how to use existing tallying setup to come to the solution for project cap and alternates. However, as the assessment of the importance of transferring the votes is corrected, this is no longer feasible, because of potential significant impact to the final results. If the cap needs to be implemented, we will set up a technical solution beyond SecurePoll.
Transfer of votes - It does make indeed sense to implement STV to its full potential and make use of the transfer of votes.
Documenting lessons learned - I agree that it will be unhelpful for current drafting committee. Yet, it is of high value for future elections that can have a similar scope.
Thank you for your kind patience and attention to the matter! --KVaidla (WMF) (talk) 16:53, 27 October 2021 (UTC)Reply
Elaborated methodology

Thank you everyone involved for the swift reaction to the post and raising your concerns as well as sharing your insights. I have revisited the previous assessment and have identified an essential miscalculation (I will elaborate further responding to the thread above) as well as understood how the resolution was too much based on the technical aspects, rather than the principles of STV. I have had a conversation with the team members involved and there is an agreement that the methodology that has been proposed in this thread (and previously on other threads) is actually the most fair one that needs to be used.

  • We will implement the constraint of maximum 2 people per project.
  • If 2 people from the same project will be elected, then the other people with the same project will be eliminated and votes redistributed.
  • The process will be run until 7 candidates meet the quota and are elected.
  • Added per specification of 4nn1l2 As to STV methodology, Meek implementation with Droop quota is used, as during Board of Trustees elections.
  • Added per specification of 4nn1l2 1) Initially SecurePoll script will be used to tally the votes. 2) We are looking to have a script outside of SecurePoll if the implementation of the gap is needed. 3) If we do not have the script in timely manner, Excel will be the last resort.
  • Added per specification of Tgr Implementing the cap not based on meeting the quota, but elimination round:
    • Calculate the number of the occurrence of the same main wiki in the pool for each project with more than 2 candidates (n).
    • As 2 candidates can make it to the pool calculate 5 + n to define the number of remaining candidates when the elimination rule needs to be implemented.
    • When the defined number of candidates is met (i.e 5 + n), eliminate the candidate from the project that has least votes at that point.
    • If more than 1 wiki meet the criteria, more than 1 candidate will be eliminated for all the projects meeting the criteria.

--KVaidla (WMF) (talk) 12:36, 29 October 2021 (UTC)Reply

Thank you KevinL for suggesting this approach on this thread!

There is still an open question regarding choosing the alternates. I see three options:

  • Still go with 2 last eliminated people in the STV process (that would not include the people eliminated with the cap restriction).
  • Rerun the STV with elected candidates removed (that might have significantly different results compared to the original tally, yet would include people eliminated per cap.

What are your thoughts?

Thank you for your kind attention and constructive contribution to ensure that the tallying is done in the right way! --KVaidla (WMF) (talk) 16:20, 27 October 2021 (UTC)Reply

For the alternates, re-running and going with the two highest candidates who are not already sitting on the MCDC (which may indeed mean 1st and 2nd place!) go on. That reduces alternates being hit by the project cap when they may remain popular Nosebagbear (talk) 16:27, 27 October 2021 (UTC)Reply
As we would like to move towards decryption of votes, yet need to have the process confirmed, I am pinging people who have engaged with this thread to know that the proposed solution is fair. @L235: @Barkeep49: @Nosebagbear: @Yair rand: @Firefly:
STV is a family of voting systems. Which method are you going to use to count the votes? en:Meek STV as in the SecurePoll? If not, which one exactly? Different methods may result in different winners.
How are you going to count the votes? By hand? By "Excel" as suggested above? By SecurePoll? Currently, SecurePoll cannot perform this step: "If 2 people from the same project will be elected, then the other people with the same project will be eliminated and votes redistributed". 4nn1l2 (talk) 17:23, 27 October 2021 (UTC)Reply
"If 2 people from the same project will be elected, then the other people with the same project will be eliminated and votes redistributed."
If You want maximal fair use, then votes have been forfeited. The limit of 2 people was known in advance. If I don't follow, then I can manipulate. I will give as many candidates as possible to the beginning so that there are as many votes as possible later when transferring votes for my local wiki. ✍️ Dušan Kreheľ (talk) 18:59, 27 October 2021 (UTC)Reply
To do this however necessitates that you have excellent knowledge about how others are voting, or you can end up doing yourself a mischief. Of the various strategic voting mechanisms STV offers, I can't see it being very likely Nosebagbear (talk) 19:52, 27 October 2021 (UTC)Reply
What exactly is meant by "votes redistributed."? Is it redistributed for all?
Doesn't it do more easier to generate a complete list than in a standard vote and then to strike 3+ rank candidates and then final output list. You don't must change a algorithm. ✍️ Dušan Kreheľ (talk) 14:09, 28 October 2021 (UTC)Reply
Re main methodology: looks good. Re alternates system: I think either of these could probably work. The former preserves proportionality relative to the other elected candidates more, but I'm not even sure if that's something we're going for for alternate spots? The alternates won't be on the committee at the same time as all seven original members. (If I were designing the system from scratch, and also didn't have to worry about too many people placed on the "alternate" list, we could probably have a system to find a (proportional-compatible) backup for each given elected member potentially needing to be replaced. Obviously, that's not an option here.) Of the two options given, I'd lean very weakly towards the latter, given that the alternates' potential relevance is in a situation where one candidate is already removed from the group.
(On the cap hitting the alternates: That could be positive or negative, depending on whether the addition of the particular alternate would put the membership over the cap, which also may depend on who's being replaced. I can't tell how likely the scenarios might be.) --Yair rand (talk) 22:43, 27 October 2021 (UTC)Reply
@KVaidla (WMF) note that it's possible for the two-per-project quota to kick in at a point where it would result too few candidates left. As an extreme example, imagine candidates with few votes are eliminated until only 10 enwiki, 5 dewiki and 5 frwiki candidates are left. The quota does not prevent that situation from happening but it does prevent getting from there to seven winners.
I think you'd need an additional rule that would look something like this: whenever a candidate would be eliminated,
  • take the pool of remaining candidates plus the candidates who have already met the quota
  • count how often each wiki occurs in that pool
  • reduce the count of each wiki to 2 if it is larger
  • if the sum of the resulting numbers is less than 7, instead of the to-be-eliminated candidate, eliminate one of the candidates belonging to the wikis with more than 2 candidates (presumably the one with the least votes).
(I agree with @Phoebe that this was an unfortunate process choice. Oh well.) Tgr (talk) 04:05, 28 October 2021 (UTC)Reply
Re: alternates, if you use the last eliminated candidates, some votes will count twice, which seems to go counter to basic expectations of electoral fairness (granted, in national elections and such, it's not uncommon for votes cast in different districts to have different weights). So IMO, either re-run the process with the 7 winners removed and 2 open seats, or run the process initially with 9 open seats, then run it again just for those 9 candidates, for 7 seats. (The first doesn't apply the cap to the alternates, the second does.) Tgr (talk) 04:14, 28 October 2021 (UTC)Reply
To ad my 2 pence worth:
I endorse the proposal that the cap is effected by marking all remaining candidates from that project as eliminated/withdrawn. This is the fairest way to handle the election in my view and means that votes cast for the 'capped-off' candidates can be transferred to next choices. (I suggested this in a Telegram group, happy to share my thinking here for transparency)
Regarding the 'runners up'. There are many ways of doing this. The most consistent with the election process is to re-run the election with each elected person marked as 'withdrawn' and then identify the first person who is then elected, who is not already on the MCDC. This would create an 'alternate' for each member of the Committee (though it's plausible that all members could have the same alternate). It also means that potentially someone who is removed because of the cap, could nonetheless be a runner-up, e.g. if P and Q from En.wp are elected but Q then resigns, R also from en.wp is a valid alternate. Chris Keating (The Land) (talk) 14:38, 28 October 2021 (UTC)Reply

Broken voting page (pl)

[edit]

Could somebody please fix a link on the voting page. Thanks :-).

Voting page (link has broken syntax)

Nux (talk) 15:50, 23 October 2021 (UTC)Reply

Use comboboxes instead of dropdowns next time

[edit]

Please use comboxes instead of drop downs next time. Long selects (dropdowns) are very hard to use even when properly implemented. And when the list is not sorted alphabetically the experience is just... Well, not good. It is very hard to find someone you want to vote on. What makes this even worse is that the compass shows nicknames and nicknames are shown in parentheses. It is worse because you cannot even use keyboard to jump to specific nickname by letter.

I can see that oo-ui combobox is broken (or appears to be broken in FF), but maybe you could use jQuery UI instead. See: jqueryui.com/#combobox. It works rather well. I have a bit more cleaned-up version here: github.com//flyspray//javascript/jquery-ui/extra. It's a drop-in solution to convert plain `<select>` into a combobox. You would have to figure out a best way to remove items in votings I guess, but simply removing <option> from <select> works fine. You could also change `_source` method to e.g. not match disabled options. Nux (talk) 16:24, 23 October 2021 (UTC)Reply

I think it shouldn't be a list of dropboxes or comboboxes at all. This is bad UX for ranking. It doesn't allow easy reordering of the ballot after selection. The whole UX for ranked vote needs to be redesigned from scratch. MarioGom (talk) 06:57, 25 October 2021 (UTC)Reply
I agree that having an ability to re-order would be nice. A sortable list more or less like in https://jqueryui.com/sortable/#connect-lists would work probably (one column would be for chosen candidates and the other for potential ones). But that would be harder to implement as there would be accessibility problems with that (no keyboard interface). So in that comboboxes give better experiences to some users. Nux (talk) 09:02, 25 October 2021 (UTC)Reply
Thank you for the feedback Nux and MarioGom! It has been well noted. --KVaidla (WMF) (talk) 19:34, 26 October 2021 (UTC)Reply

Movement Charter/In the cube

[edit]

For people, what this theme opening first time: Movement Charter/In the cube. ✍️ Dušan Kreheľ (talk) 15:23, 24 October 2021 (UTC)Reply

Tardy Selectors

[edit]

I put this on Qgil-WMF's talk page a few days ago but since things are time sensitive I thought I would cross post here: I'm wondering if there is anything being done to ensure all selector rankings are in before wiki voting results are revealed to ensure that all groups are acting simultaneously and to keep the process fair for all. Thanks and best, Barkeep49 (talk) 18:34, 25 October 2021 (UTC)Reply

I agree that it is important to have selector rankings before election results are unsealed and tallied. KevinL (aka L235 · t) 22:25, 25 October 2021 (UTC)Reply
Dear Barkeep49, thank you for this question and KevinL for the supporting point. I am providing a wider context of the situation and then get to the current solution we have for the issue raised. I hope this is helpful.
Indeed the process was designed to have equivalence between the election process (mainly designed to bring in the voices of the project contributors) and affiliate selection process (mainly designed to bring in the voices of the movement organizations). One key aspect in ensuring the equivalence has been the set up to run the processes in parallel. For this reason the group of affiliate selectors has been convened in parallel to the start of the elections.
There has been a complicated design element related to running parallel processes. Namely, consolidation of 2 separate lists of elected and selected candidates. Parallel processes might result in overlap of candidates and there needs to be solution mitigating this. The approach taken is that the election tally will be run with 7 candidates and the consolidation happens with the selection process with the selectors creating a list of preferences of 13 candidates. This would ensure that even if there is a full overlap in 2 lists, we would still get the needed 6 selected members from the selector list.
However, even prior to convening the group of selectors, there have been discussions related to the selection process and the opportunity to bridge diversity and expertise gaps in the committee with the selection. The suggestion has entailed a proposal to run the processes in sequence, so the selection can be done after the fact of having the election results. This suggestion, however, is in clear conflict with the design based on the equivalence principle and, as a result, is not implementable. This means that the point you have raised is valid and acknowledged.
At the same time, we have had conversations with the selectors and it has become clear that there might be a way how to tap into the potential of the selection process, while keeping the integrity of the equivalence design. The solution we have is that 1) as the first step, the group of selectors is creating a shortlist of 13 candidates to express their preference related to the composition of the drafting committee. This will be done prior to tallying the votes; 2) after votes have been tallied, the selectors will have access to that information, to make their final selection of 6 members to be appointed to the drafting committee. They can only select the candidates that have been shortlisted prior to the tally.
We plan to publicly announce the results of the election and selection process together next Monday, i.e. on November 1, 2021. --KVaidla (WMF) (talk) 19:31, 26 October 2021 (UTC)Reply
@KVaidla (WMF) if this had always been the plan I would have been raising a stink from the start. To the extent that issues around the election were not fully considered it seems like that is the case because of deadlines set by the foundation, perhaps at the board level but by the foundation never-the-less, and not the broader volunteer community (affiliate and wiki editor alike). I could live with a 7/6/2 split compromise assuming that the wiki voted and affiliate selected at the same time and so I said nothing but even this seems to give the affiliates a disproportionate ability to select members relative to the number of people who are active editors vs affiliate members. However it seemed like a reasonable process and one I could live with. And now, with voting over for me, my ability to directly impact gone, I'm being told that the election process is changing. So I'm left with questions:
  1. Why have the selectors gotten to have this conversation with you and not the community at large? I see nothing in your contributions that suggests this conversation happened on some place on meta that I simply wasn't following.
    1a. If it was happening on some off-wiki but public forum (i.e. a wikimedia mailing list) can I be told where so I may at least read the discussion?
  2. Why did it take me wondering about this to discover that a substantial change - and regardless of whether this meets the technical provisions laid out it is a substantial change - is being considered (or perhaps it's already being implemented)?
  3. Is this the right place to let others know to express any opinions they may have about the topic? If not where is the right place?
Thank you for your time, Barkeep49 (talk) 19:51, 26 October 2021 (UTC)Reply
(Edit conflict.) This seems like a drastic change to the system, giving more influence to the affiliate selectors. --Yair rand (talk) 20:17, 26 October 2021 (UTC)Reply
Apologies but one more question KVaidla (WMF): are there any other changes to the election process being considered at this time? Best, Barkeep49 (talk) 21:15, 26 October 2021 (UTC)Reply
Thank you for posting these questions, Barkeep49; I too would deeply appreciate an answer to them. Best, KevinL (aka L235 · t) 21:25, 26 October 2021 (UTC)Reply
Thirded on “would like to see responses”. Thanks Barkeep. firefly ( t · c ) 21:40, 26 October 2021 (UTC)Reply
  • @KVaidla (WMF) and L235: this is a huge shock to me, and I'm a candidate (and have thus been paying pretty close attention). What is the selection method doing changing now?! Your phrasing suggests that drawing members from a shortlist of 13 has no negatives re the equivalency principle, but when we were writing the proposal together, we were anticipating between 20-25 candidates. 13 would have been over half of those, and we never would have considered that functionally the same process. This is still a MASSIVE change. If you were considering it, then it should have been given top of meta's front page and on the various MCDC talk pages, pulsed to everyone who had voted, highlighted on every channel that occurred, and at least dropped on as many local noticeboards as two hours' frenetic posting would allow. I have to 2nd Barkeep's "why is this a discussion being had with selectors and not others". Mid-election changes like this shouldn't happen - we built in the MCDC having the ability to select additional members for after the fact filling out of expertise and diversity matrices - we didn't consider an equivalent after the fact change to be within the affiliates' selectoral remit, and we wouldn't know that those who engaged (or chose not to engage) with the merged method wouldn't have done so differently had this been included, so it's not necessarily something that even a broad scope discussion could handle now. Please do not go ahead with this. Nosebagbear (talk) 20:15, 26 October 2021 (UTC)Reply
I am astounded to read that this change - seemingly to upweight the opinions and selections of the affiliate selectors, in effect if not in intention - is being considered at this point after the community voting period has ended, and without any kind of formal communication. I concur with my colleagues above that this must not happen. firefly ( t · c ) 21:01, 26 October 2021 (UTC)Reply

Response to questions raised by Barkeep49

[edit]

Thank you for raising the questions and concerns to ensure that the integrity of the process is not broken and the parity is ensured across the stakeholders! This is a really important aspect to ensure the success of the Movement Charter process moving forward and I appreciate your care.

1) Our commitment with the Movement Strategy and Governance Team facilitating the process is to provide support to all different contributors to the process. We have put a significant effort into supporting the elections by a) working with the candidates to surface additional information regarding their expertise, 2) put together the election compass to make it easier to browse the 70 candidates and disseminating similar tools developed by community volunteers, 3) we have disseminated the information across the languages to support wider turnout across the projects, 4) we are also invested to be responsive to all the queries coming from the community members either, whether on meta, local wikis, mailing lists, social media channels or when receiving private queries.
In a similar way, we have tried to support the selection process, being in close contact with regional affiliates to support the selection of selectors and now working with the selectors to advance the process to the conclusion. The reason why selectors get to have this conversation with us is that they are appointed "super-delegates" of their regions with a mandate for certain tasks, needing to implement their process. The reaction we received from them when presenting the proposal was that there is a built in tension or polarity between the ask of "The selection process will run in parallel to the election process and will focus on adding diverse expert profiles to the Drafting Committee based on the Diversity and Expertise matrices" and the need to do this with 6 seats without knowing the results of the elections. We tried to find solutions to this in our conversations, yet in the end where we landed was that in order to work with diversity and expertise, some disclosure of the election information would be of high value.
We did not publish information regarding these conversations earlier, because we hoped that there is a way to resolving it and enabling selectors to do their work without any iterations to the process. The deeper we went with the discussions, however, it became obvious that it is not the case. We were planning to post information regarding the planned change on meta and also on the Movement Strategy Telegram channel were many of the reflections happen (as we had done previously with the change to WMF appointment), but this urgent thread beat us to it.
1a) There have been prior discussions regarding this on semi-public forums. Many Wikimedia conversations have moved to Telegram channels and there have been couple of threads regarding this matter on Movement Strategy Telegram channel and also on WikiConference North America channel. In addition, some prior discussions have happened on the calls like [Wikimedia Affiliates Network call in July] and also at Wikimania 2021. Unfortunately, even though the sessions were public, the documentation from these meetings is lacking the information regarding the points raised about the selection process.
2) The process is not implemented. We have not provided any information to the selectors on the election results. We were late with the consultation regarding this matter, because we were trying hard to find other solutions that would not entail changes to the process. We did not manage it prior to closing the elections, when this urgent thread beat us to it.
Ideally, these matters should have been fully resolved prior to either of the process started. While there were some concerns raised here and there prior to the process regarding the shortcomings in the selector process, they were not presented consistently, with a sufficient rationale to actually make changes to the process. It was only when the selectors were nominated and needed to implement the process when the issue escalated due to integral conflict in demands to, mandate of and the process available to the group of selectors.
3) This is a good place to have these conversations. We are trying our best to keep it connected to discussions happening elsewhere, mainly Movement Strategy Telegram channel.

I hope my responses provide some clarity on the context and support your thoughtful engagement in resolving this matter! --KVaidla (WMF) (talk) 12:25, 28 October 2021 (UTC)Reply

I'll chime in here as the affiliate selector for North America. The selectors have been discussing the full composition of the MCDC; what mix of skills, diversity, and experience do we want to have on this group in order to represent not just the movement we have, but the movement envisioned by the Strategy 2030. When you start identifying all the experiences and skills we want to bring to the table, you quickly realize how difficult that is. Even if we wanted to make this strictly regional, to ensure we have a diversity of geographic perspectives, there are 8 geographic regions represented by selectors. Even if we only nominated one person from each region, we'd still have to cut out two regions; which two do we cut? The expertise matrix also has 15 different things we want members to have, irrespective of the diversity matrix. We can pick our seats with complementary skills if we know which skills are already there from the community election.
Yes, the process allows for the MCDC to add people if skillsets are missing; however, I think 15 people is already a lot (we selectors are having scheduling woes trying to meet and there are only 9 of us); none of the Working Groups had even that many people on them. And the process has already been delayed repeatedly; waiting until the group adds more on seems like an unnecessary delay.
I have sympathy to the objection that it's another change in the middle of the process; however, so have been the other ones that have gone through. It's been messy. None of us expected to have 70 candidates! I was there when we were speculating if we'd even have 13! If we'd known this is what we were going to have in terms of a candidate pool, I think we'd have done a lot of things differently, and in the "improving as we go" spirit of our movement, I personally am ok with making changes like this that (in my mind) improve the end goal of this effort: Having a diverse set of people with the necessary skills to collaboratively craft a Charter for our Movement. I can respect if you disagree with this from a process perspective.
Finally, I'll address the question of "does this give Affiliates more power" -- please also keep in mind that Affiliates have a huge stake in the outcome of the Movement Charter. Defining affiliate roles and responsibilities, the Foundation's role going forward, how much we decentralize existing power structures, governance rules, funds dissemination -- these all have only minor impacts on individual content contributors but *huge* effects on Affiliates. I think individual content contributors are the heartbeat of our movement, and as some of you may remember, I pushed really hard in the discussions earlier this year for the community election because I believe so much in the importance of engaging people who aren't in the Meta weeds in this critical moment in our movement's future. But I also think Affiliates are important, and the voices of those representing the structured entities in our movement who are perhaps the ones most affected by the content of the charter are needed on the Drafting Committee. The proposal KVaidla (WMF) outlined above -- where we create a list of 13, then see the community election results, then pick our 6 from our list of 13 to fill in the gaps from the community election, in my mind does not give undue power to Affiliates. This may even be a totally moot point if the 7 community elected people are already on our list of 13; then our other 6 just get added by default. Liannadavis (talk) 23:02, 26 October 2021 (UTC)Reply
If it was seen as vital to have all 8 geographic regions represented that could have been in the structure and done in any number of ways. Similarly with the expertise matrix. But that wasn't seen as vital enough for the process to have landed there. It did land on a process that was largely accepted by everyone and did have some measures to ensure diversity including a limit of 2 people per project (reasonable) and the ability to bring in more people if needed (also reasonable). It also setup a process where selectors and wiki voters were to be given equal amounts of information when making their choices. I certainly would have liked more information when I made my choice than I had. But the way the process landed is fair to all. Changing it now and seemingly with a discussion only because I happened to wonder about it is not fair and it is not in the spirit of what this global charter will hopefully do. Finally, I'll say that you didn't actually address "does this give Affiliates more power", you address why the Affiliates deserve power. I agree that Affiliates have an important role to play in the movement - and have repeatedly tried to make sure the UCoC enforcement draft is something that works equally well for them as it does for wikis. So yes the Affiliates deserve power, but that doesn't mean they deserve more power than what was laid out for them in the process at the start and it definitely should mean that they shouldn't be able to get more power by appealing to common goals like diversity at the 11th hour. Best, Barkeep49 (talk) 23:19, 26 October 2021 (UTC)Reply
(Edit conflict.) Hi @Liannadavis:,
A few aspects where I can see valid disagreement arising here:
  1. While affiliates certainly are affected by the MC, there are vastly more individuals within the projects than active in the affiliates, the effects are likely even greater (given that effects on affiliates are effects on projects) and there was already major disagreement about affiliates receiving 6 seats. Still, your reasoning does indeed specify why affiliates deserve some say. But that's not the dispute, the dispute is "do they deserve additional electoral say than that provided in the process", and that is not covered adequately.
  2. While you note "improving along the way", two main issues arise there: proposal structure and voting disruption. Firstly we have that this set-up was designed to balance things, and so "improving as we go" only works if it doesn't impart any negatives that can't be fixed. One side of the equation is being changed at a stage when others can't be. The people who agreed to the merged proposal may well not have had they known these changes would go in.
  3. The second of these aspects is that the voting structure is closed. We've known we were going to have 70 candidates for weeks and weeks. If someone viewed this as an issue to be improved it should have been raised on meta where there was sufficient time to get widespread consideration, voters could make their choices in an informed fashion, election compass queries could have been crafted with it in mind, and affiliates could have made their selector choices with it in mind. It's not like writing an article - the problems caused by a change can't be resolved, and the basis for that decision can't be verified as still standing. That the selectors raised it directly rather than in a public forum is also concerning, and in violation of how things should be done on wikimedia. Nosebagbear (talk) 23:28, 26 October 2021 (UTC)Reply
(Edit conflict.)I at least thought it was possible we wouldn't even *have* candidates from all geographic regions. I wasn't involved in every discussion about setting this up, but I think my point is the process we defined with our idea of how many people would be running simply isn't great with the reality we face now, for the things we want to accomplish. And re: the Affiliates power thing, I guess I don't see it as giving Affiliates more power; we still only get 6 seats to the 7 community ones. And isn't having those diversity and expertise skills not added by us, but instead backfilled by *additional* people being appointed beyond the 15, what will actually dilute the power of the 7 community elected seats? (edit conflicted with NBB and I have to run now but will check and reply later!) Liannadavis (talk) 23:33, 26 October 2021 (UTC)Reply
(Edit conflict.) (Re "None of us expected to have 70 candidates": One of us *cough* suggested having a committee of 56, with 36 directly community-elected.)
Yes, the affiliates think the affiliates are important in this, while others may disagree. And we did end up giving the affiliates "bonus" seats (while allowing many affiliate members to also vote in the main election, and in fact doing more advertising to them than to the communities, many of whom had no idea what the election was even about) in a quantity almost on par with the entire community representation. While also giving any "extra" seats (that come about through overlap) to affiliates, even further. I doubt, if we had some way of getting widespread quick response to the question under discussion, that the communities would consent to giving even more influence to the affiliate selectors. --Yair rand (talk) 23:39, 26 October 2021 (UTC)Reply
To me it comes down to the value each of us individually place on having an MCDC with a mix of skills and experiences from diverse communities in our movement. Is it about individuals, or is it about creating a *group*? I think it's about creating a cohesive group that can do community consultations and outreach to as many parts of our movement as possible. Let's say there were two candidates from Antarctica who had similar profiles; in a parallel system, it's possible one is chosen by Affiliates and the other is elected by the community. Then two people who bring similar skills from the same community are on the MCDC, and another group in our movement lacks someone on the MCDC from their community who can do consultations with them following their community norms (and maybe even in their native language — it's easy for those of us, myself included, who speak English natively to forget the language barrier). I would like to avoid this hypothetical situation. Allowing Affiliates to select after the community votes enables us to see that the first candidate has already been elected, thereby having us add someone from that other group. Is this fair to the individual candidate who is then not selected since their fellow Antarctican was elected in the community process? Maybe not, but I value group diversity reflecting our global movement for something like this more. And I'll reiterate my earlier point that if we keep the process as-is, the existing rules state that the missing diversity can instead be filled by appointed seats after November 1. If we keep the process as-is, I think that outcome is *highly* likely, and that is what will really shift the balance on the MCDC away from community elected people. Liannadavis (talk) 01:12, 27 October 2021 (UTC)Reply
The MCDC doing a self-assesment and filling gaps it identifies sounds both like an excellent initial project for the committee to work on as it figures itself out and in keeping with the process as designed. Thus there is no need to, at the last minute, change the process in a way that just happens to give you more power. Best, Barkeep49 (talk) 01:34, 27 October 2021 (UTC)Reply
That means you would be ok if the group then appoints 3 more people who are mostly "affiliate" people since the composition is lacking representation of affiliates in 3 regions, shifting the balance to 9-7-2? And that's somehow not "more power" for affiliates, but us affiliate selectors selecting those same 3 people instead for 3 of our existing 6 slots because we know those people are missing from the community election, and maintaining the 7-6-2 balance, is? I'm not trying to get "more power" for affiliates here; I'm trying to maintain the 7-6-2 balance so community elected people have more slots. Liannadavis (talk) 02:07, 27 October 2021 (UTC)Reply
Yes I am trusting that the process as designed will produce good people and that those good people will make good decisions on how to proceed with this process. If I didn't trust the results of the process I would have been complaining earlier but I could and am willing to live with that. Best, Barkeep49 (talk) 08:26, 27 October 2021 (UTC)Reply
@KVaidla (WMF) I get it that the work the foundation has had to do with this election is immense. However the foundation chose the timeline for this whole thing. The idea that the selectors by virtue of their position get to have the conversation is fair enough. However, absent this thread it appears that the decision would have just been made by selectors with input by those who are privileged enough to know they should have been on telegram - a service not mentioned on any Movement Charter page I could find and definitely not mentioned on any Movement Charter Drafting Committee or MCDC election page. Even with this thread all it meant was that there are 24 hours for discussion. Barkeep49 (talk) 17:28, 28 October 2021 (UTC)Reply

This seems like a pretty straightforward issue to me. Selectors are tasked with ensuring the diversity of the committee alongside various axes. Diversity is one of the two "founding principles" of the entire strategy process, so this is kind of important. They can't do this job properly if they don't know the composition of the rest of the committee. So the process was set up in such a way that prevents them from doing their job. It's unfortunate, but mistakes happen. It's better to fix them late than to not fix them at all.

What I find really frustrating in this debate is that the arguments for keeping the election and selection process parallel just don't make any sense whatsoever. It's one thing to wish for more influence over the composition of the committee to go the editor communities; one can make reasonable arguments for that. But preventing selectors from doing their job won't increase the editor communities' influence by a bit! People who care about diversity will be worse off, and people who care about editor community representation won't be better of. It's the pessimal choice. If the selectors decided that they want to maximize the power of Nigerian women on the committee, or of affiliate board members, or whatever conspiracy theory we seem to be operating on, they could do that just as easily in the parallel process. The only thing it hinders is ensuring diversity, which, again, is the core principle of the entire strategy.

Sorry if this is rude, but I really hope the people running this process can make the correct choice between some people's confusion and the Strategic Direction. --Tgr (talk) 01:21, 27 October 2021 (UTC)Reply

I concur with Tgr - it is bewildering to see this sentiment of "communities vs affiliates" being touted here as if this actually applies to anyone outside of en.wikipedia. This step in the process shouldn't be about power, but for some reason the extreme voices got a foot in the door and ended up making this all about perceived power, even though that discussion comes later and is between all of us and the WMF. Braveheart (talk) 07:30, 27 October 2021 (UTC)Reply
I dislike being called an extreme voice. I dislike it because I have the confidence of not only my own project but the confidence of arbitrators from other projects when it comes to drafting UCOC enforcement. I dislike that two candidates for the MCDC, including one whose home wiki is not enwiki, have been called an extreme voice. I think it's hard to know what broader community thinking is because until today this change hadn't been known about so there had been zero ability for people to say what they think. The people objecting may be a minority voice or you may be. But none of us are extreme voices. Best, Barkeep49 (talk) 08:32, 27 October 2021 (UTC)Reply

To my mind the affiliate selectors will be able to do a more effective job if they know the results of the community election first. The purpose of having affiliate selectors is not just to give affiliates a voice, it's to help balance the committee by paying attention to parts of the skills and diversity matrix that haven't already been filled by community election. If the community election returns 6 men and 1 woman, the affiliates can be expected to mainly appoint women. If the community election has no-one from Africa or South Asia, the affiliates can be expected to ensure those regions have at least one person represented. This is much easier to do if the affiliate selectors know what gaps there are to fill. Also, I really don't understand how approaching it this way affects 'the power balance' or whatever it is people are concerned about. If the affiliate selectors were really concerned about having more power on the committee, they wouldn't be asking for this change, they would just be giving a list of the 13 candidates who they think would most reliably promote the interests of affiliates. Chris Keating (The Land) (talk) 08:08, 27 October 2021 (UTC)Reply

Not that I believe the current affiliate selectors would do such a thing, but it would be entirely possible for them to look at the candidates, decide they were "anti-affiliate" or "movement skeptic" or "anti-WMF" in some way, and balance the selection in that regard. I don't believe that's likely to happen, I trust the affiliate selectors will look for balance in the way LiAnna and Chris has suggested. But it does come down to trust. Had this happened a little over seven years ago, I would have had a lot less trust in "movement" people (at the time I assumed they were mostly disgruntled former Wikipedia editors who had created smaller ponds for themselves to wield power and play off relationships with WMF). Now I have a much better sense of the movement and WMF, and which gives me a much better sense of the complexity and diversity of perspectives in the movement. But I fully understand that distrust isn't some extreme position. Guettarda (talk) 14:17, 27 October 2021 (UTC)Reply
Given that the selectors were chosen by the affiliates, I don't think it's particularly paranoid to assume that they might (consciously or unconsciously) prefer candidates with an affiliate background. But if that were the case, they would just select six such candidates, whether they have to do it now or next week. (They wouldn't have to try very hard - it's not like there is an abundance of "movement skeptic" people on the list of candidates.) I still don't see how this makes sense as an argument against letting selectors know the election outcome. As Chris said, if the selectors were predisposed to prefer a certain group, they could select candidates based on that in the current process just fine. It's only problematic for selectors if they truly care about diversity, as opposed to promoting some specific POV or minority group. Tgr (talk) 16:40, 27 October 2021 (UTC)Reply
@Tgr: There are several axes that selectors might select for, and they have relatively few degrees of freedom, given the available candidates and certain qualifications that they need to look for. (They can't just pick six clones that have all the opinions and mindsets that the selectors like and are also fully qualified, because no such options ran. All the more, given the need for diversity as well.) Knowing the election results makes it much, much easier to specifically modify the committee to the particular outcomes they may prefer. --Yair rand (talk) 00:39, 28 October 2021 (UTC)Reply
@Yair rand if the committee preferred some particular outcomes or opinions or mindsets, they would just select the 6(+7) people best matching those, and the result of the election would be completely irrelevant for them. The only case in which knowing the election results matter is when your scoring function for a given candidate depends on the presence/absence of other candidates in the committee (ie. some kind of diversity or conformity metric). It's just basic math. Tgr (talk) 04:22, 28 October 2021 (UTC)Reply
I agree with Tgr and The Land. This small change does not change the balance beteeern the two elector groups, but it ensure that the endresult will not be cotradictory to the aim of the composition of the group.Yger (talk) 14:45, 28 October 2021 (UTC)Reply
  • How is this discussion being spread? I noted a few options above, but as far as I can tell, it's basically just telegram and those who were already on this page. As an effect on the (s)election method, it should logically be raised to the 1040ish voters, in some neutral form. Strategy discussion with insufficient awareness of applicable parties is almost our catchphrase by now. In some cases, fairly immense efforts have been made (with still limited success, but there's a "what can you do" sense there), but this one would not appear to fall into that camp Nosebagbear (talk) 16:17, 28 October 2021 (UTC)Reply

Updated proposal for resolving the issue

[edit]

Summary of the issue: There is an intrinsic tension in "the selection process will run in parallel to the election process and will focus on adding diverse expert profiles to the Drafting Committee based on the Diversity and Expertise matrices", i.e. ensuring diversity and expertise of the group, only having 6 slots to do so, and not knowing the results of the elections. This issue is preventing the selectors to do their work well and finalize the results. Ideally, the issue should have been resolved prior to either processes starting, yet there is still a possibility to do this right.

Key considerations: 1) Integrity of the process needs to remain intact and general equivalence between the processes preserved, 2) selectors need to be enabled to fulfill their task and the mandate, 3) equity does not happen by itself, but needs proactivity - this is the point where action needs to be taken to avoid gaps in the process that will grow over time.

Proposal:

  • Selectors put together a shortlist of 13 people with 6 first choices. (in progress)
  • Provide selectors with anonymized data from the elections prior to final decision, focussing on the key areas of expected gaps:
    • Regions not covered by the election outcome
    • Areas of expertise not covered by the election outcome
    • Previous experience in movement committee roles not covered by the election outcome
  • Selectors will make the final decision of choosing 6 members with an access to this anonymized and quantified data, but not full election results.

Question: Would this proposal work in your perspective? Does it keep integrity of the process? Does it provide sufficient support for the selectors.

We would need to make a final call regarding this in less than 24 hours, as we have planned for a deliberations meeting with the selectors tomorrow, so looking forward to your swift feedback! Thank you for sharing your insights and considerations. --KVaidla (WMF) (talk) 16:56, 28 October 2021 (UTC)Reply

Pinging everyone in the thread to ensure that they have noted this proposal and can contribute their thoughts, as well as because there are time restrictions for making a final call on this: @Nosebagbear: @Yger: @Tgr: @Yair rand: @Guettarda: @The Land: @Barkeep49: @Braveheart: @Liannadavis: @Nosebagbear: @Firefly: @L235: --KVaidla (WMF) (talk) 17:11, 28 October 2021 (UTC)Reply
As this is what was suggested above the comments made above reflect my feelings (and likely those of the other participants). Best, Barkeep49 (talk) 17:14, 28 October 2021 (UTC)Reply
Since I wrote a lot above let me emphasize one point: I would much rather have the committee itself fill in gaps with diversity and expertise, gaps that might not be immediately apparent from the information selectors will have, and in accordance with the process laid out at the start than to have the selectors do it. Best, Barkeep49 (talk) 18:51, 28 October 2021 (UTC)Reply
I concur with Barkeep’s views and analysis entirely. firefly ( t · c ) 20:05, 28 October 2021 (UTC)Reply
To clarify, we're talking about the 8 regions, the items on the matrices, and experience in a specific set of committees (could these be listed?), listing specifically which of each are completely absent from all nine members (7 elected and 2 appointed), and not getting any more specific than that? --Yair rand (talk) 17:31, 28 October 2021 (UTC)Reply
Yes, that is the idea. --KVaidla (WMF) (talk) 17:34, 28 October 2021 (UTC)Reply
OK för me, but I would also like the anonymized data to show men/women.Yger (talk) 17:22, 28 October 2021 (UTC)Reply
Anonymization makes sense (although it's unfortunate that the matrices are sometimes arbitrary; e.g. technology expertise is missing entirely, despite software development of some form making up about half of the total spending of the Wikimedia movement). The shortlist makes this proposal mostly pointless though :( Tgr (talk) 17:32, 28 October 2021 (UTC)Reply
I would definitely add gender to the list of characteristics that are being shared. I don't really believe that anything regarding candidates' self-assessed competencies is worth sharing - it would be pretty rubbish if the selectors were led to believe that a particular competency was covered on the MCDC because a one successful candidate suffering from en:unconscious incompetence had listed it on their profile, and this created the impression that all was fine. "Experience in movement roles" is also... difficult to assess (which roles are being listed?) Chris Keating (The Land) (talk) 18:09, 28 October 2021 (UTC)Reply
This is a somewhat concerning viewpoint - ensuring expertise balance was met is one of two goals, alongside diversity, and if you're waiving it, logically both should be being waived (not something we'd generally want, I imagine) - the text doesn't contain a prioritisation basis. Nosebagbear (talk) 18:16, 28 October 2021 (UTC)Reply
I'd love for balance of expertise to be considered, I just don't feel there is any meaning in using candidates' self-assessments as a basis of judging 'expertise'. There is, if one looks at the candidate statements vs the self-assessments an extremely wide range of experience and competence in people who have said yes to any particular kind of experience. And also clearly there are cases where individuals have very deep experience which is not listed among their top 3 competencies. Chris Keating (The Land) (talk) 18:26, 28 October 2021 (UTC)Reply
To be honest I stopped reading after "parity among stakeholders". This is what's wrong with the world. Demanding parity among unequal partners just benefits the status quo and in this particular case, the English-speaking, white male world view. Let's get this over with and focus on more important things. Braveheart (talk) 18:18, 28 October 2021 (UTC)Reply
What exactly is wrong with the process that was present before? I think it's pretty much antithetical to an outcome that's not only fair, but perceived to be fair, to hand off even partial election results to a group whose decisions will directly affect the final makeup of the MCDC, before they finalize their decision, and before the results are publicly released. Their job, just as the voting community's job, is to select candidates who are qualified, willing, and able to do the work required. Yes, everyone should be considering diversity of all types (qualifications, linguistic, regional, etc.). At the end of the day, though, we need a committee with 15 people with a proven track record of successful collaboration, understanding of the strategy, a history of working at the "global" level, ability to meet deadlines, and people who show up consistently. This is not a good place for capacity building; we need people who are *already* capable, and there are a LOT of candidates who meet that qualification. Leave the affiliates alone to make their selection in the way that they have always been intended to make their selection. Risker (talk) 23:08, 28 October 2021 (UTC)Reply
No lie, I was a little more than upset about finding out about this potential change. I listed 13 candidates I wanted to see on the committee, but I ranked them all using a criteria that heavily skewed in favour of representing diverse geographic interests. I would have completely changed my vote if I had known the result would be retroactively adjusted for those exact reasons. –MJLTalk 03:54, 29 October 2021 (UTC)Reply
I spent days on making my selection, carefully ranking the entire slate of candidates, and now you're telling me I voted wrong and the outcome must be adjusted to produce the desired results. Vexations (talk) 09:46, 29 October 2021 (UTC)Reply

Community election results leaked...

[edit]

Posting this here as it is widely discussed in some other places. The apparent results of the community election have been posted by the German Wikipedia Kurier. Based on that, the successful candidates appear to have been: Richard Knipel (Pharos), Anne Clin (Risker), Alice Wiegand (Lyzzy), Michał Buczyński (Aegis Maelstrom), Richard (Nosebagbear), Ciell (Ciell), Ravan J Al-Taie (Ravan). This sort of renders the discussion above moot (the affiliate selectors appear to have noticed that this has happened). Chris Keating (The Land) (talk) 09:56, 29 October 2021 (UTC)Reply

This is rather vexing on the discussion side, I don't believe it's technically leaking in the sense of someone spoke who shouldn't have yet, but not realising that the data dump was done automatically. I assumed it was manually posted by WMF staff after the results were confirmed, but apparently it's an auto-post built into securepoll Nosebagbear (talk) 10:48, 29 October 2021 (UTC)Reply
indeed, not leak in terms of someone maliciously making private data available, but more of a head-scratching "Securepoll does WHAT??" moment. Anyway, congratulations on your tentative, probably elected but not officially confirmed, status. Chris Keating (The Land) (talk) 11:10, 29 October 2021 (UTC)Reply
Thank you Chris Keating (The Land) for sharing the information discussed on MS Telegram channel also here and Nosebagbear for sharing your perspective.
Automatic data dump is indeed a feature of the SecurePoll. This is not automatically available, but will be generated as soon as decryption key for the votes is entered.
We could not hold back the decryption of the votes, because of the limitations of the built in SecurePoll STV script tally to account for the project cap (needing extra time for tally) and the fa.wp ArbCom elections waiting behind us to kick off their process (with some tasks to be performed on the SecurePoll prior to that.
The selectors have noticed this post (also being shared on the Movement Strategy Telegram channel). There are strong advocates among them to overlook the data that has been published, as it is not officially confirmed and proceed with the selection as planned. We will announce the election and selection results on Monday and I will follow-up on this thread to disclose the approach the selectors took in the end. --KVaidla (WMF) (talk) 11:17, 29 October 2021 (UTC) For some reason I had been logged out from wikiReply
Indeed, it's definitely not a "anyone's at fault" issue - in fact, for every other election ever run on Securepoll, it's a positive there as well.
We'll add it as additional motivation (the big one being the UCOC generating a U4C and various other arbcoms likely emerging) for, short-term: a 2nd securepoll/vote-wiki, and long-term, a major redo of Securepoll.
@KVaidla (WMF): I think that's really the best we can go with from this situation. When we finish, and you've had the chance for a week's rest with many beverages of your choice, I'd be interested in participating in running a lessons log. A non-basic election set-up is likely for the GC (and who knows about the hubs' methodology), so running through it seems like a worthwhile activity. Nosebagbear (talk) 12:31, 29 October 2021 (UTC)Reply
From what I understand three of these (risker, Pharos, Nosebagbear) have enwp as home wiki, and only two will be accepted from the same project? Yger (talk) 15:30, 29 October 2021 (UTC)Reply
Pharos's home wiki is Commons. Risker (talk) 15:32, 29 October 2021 (UTC)Reply
@KVaidla (WMF) - sorry I'm confused by I will follow-up on this thread to disclose the approach the selectors took in the end. and how that aligns with We would need to make a final call regarding this in less than 24 hours... What is the final call that has been made? Best, Barkeep49 (talk) 17:11, 29 October 2021 (UTC)Reply
Well, I can't see any reason to disbelieve this "leak". But what is done is done. Looks like between the elected and WMF-selected candidates, 6 of 8 geographic regions have some form of representation. I am sure that the affiliate selectors will have no difficulty coming up with suitable candidates from the Sub-Saharan Africa and ESEAP regions, and can then concentrate entirely on selecting the best candidates to round out the committee. And I suppose that knowing the actual names of successful candidates should make things easier for them (rather than some kind of expurgated version of "candidate profiles"). I do find it interesting that, for the first time ever, the community has elected a majority of women to any kind of position. Risker (talk) 16:54, 29 October 2021 (UTC)Reply

Althought knowing, that making the whole votewiki an Persian wiki could be interpreted as an poor attempt to confuse non-Persian readers in a time with translation tools widely aviable - I'm confused where the problem is. Per Movement Charter/Drafting Committee/Set Up Process#Selection_process the "The selection will be conducted between October 11 - 24, 2021" which already happened. Habitator terrae (talk) 19:15, 29 October 2021 (UTC)Reply

It's at least 6 years that Persian Wikipedia holds its elections exactly at this time (late Oct -- early Nov). It's the Wikimedia Foundation that is holding its obscure Movement Charter/Drafting Committee elections at a wrong time without even notifying us! We already postponed our elections for one week (phab:T292685#7412256). So if you are looking for someone to put the blame on, you should look at the WMF! 4nn1l2 (talk) 19:29, 29 October 2021 (UTC)Reply
So this wasn't an attempt to make the votewiki unreadable. No blame. Habitator terrae (talk) 20:04, 29 October 2021 (UTC)Reply
Now I understand the problem: The election-selection-process was changed after the start of the election. Habitator terrae (talk) 20:10, 29 October 2021 (UTC)Reply
  • To be honest, the wise thing to do would be to properly release the results of the election, explaining what happened, in order to stop all the nonsense. And at this point, neither the original nor the "revised" rules for the affiliate selectors make any sense. They should just be advised to make their decision based on the results of the election. So many of the rules have been changed at so many junctures during this election, that I think it is just better to accept that, and be forthright. There is no way to recover from this level of release of information, and it's just putting the affiliate selectors in a far worse position than they would be otherwise. Risker (talk) 20:16, 29 October 2021 (UTC)Reply
    I agree with this. Best, Barkeep49 (talk) 20:28, 29 October 2021 (UTC)Reply

The result is public: Movement Charter/Drafting Committee/Elections/Results/Announcement --Dušan Kreheľ (talk) 12:41, 1 November 2021 (UTC)Reply

Dump of votes

[edit]

How can one obtain the dump of votes? I need them to study the voting method (i.e., Meek STV) as you can see here: Talk:Movement Charter/Drafting Committee/Elections/Results#Rounds 39, 40, 41. In other words, how could the German Wikipedia leak the results? 4nn1l2 (talk) 09:47, 3 November 2021 (UTC)Reply

RAW result of vote is every-time to download on URL: https://vote.wikimedia.org/wiki/Special:Securepoll/dump/NUMBER_OF_VOTE. If You know a URL, so You can download and analyze.
The human like format of RAW MCDC vote: Movement Charter/Drafting Committee/Clean voter result. ✍️ Dušan Kreheľ (talk) 13:30, 3 November 2021 (UTC)Reply
More to study or analysis: Talk:Movement_Charter/Drafting_Committee/Candidates#The_final_MCDC. ✍️ Dušan Kreheľ (talk) 20:24, 4 November 2021 (UTC)Reply
@Abbad (WMF) Can we get a copy of the cast vote records (representing the actual ballots (votes)) in a standard ranked-choice representation, so we can plug the ballots into other software to check the results? The only form in which I find the actual votes here is an HTML table at https://meta.wikimedia.org/wiki/Movement_Charter/Drafting_Committee/Clean_voter_result, which is messy to work with. What tabulation software like flexsstv, OpenSTV, CIVS, STeVe or RCVRC expects are well-defined formats like BLT, preflib, ABIF, CSV etc. ★NealMcB★ (talk) 04:50, 9 February 2022 (UTC)Reply
@★NealMcB★ Thank you for the question. Releasing the ballots has been requested since immediately after the elections, and has been something that our team tried to acquire since then. I'm afraid to say that we have been unsuccessful in getting to this data ourselves. I cannot myself describe the technical challenge, but my colleague @Kaarel may be able to better explain them. Would it help if I we get you the same data from this page in a different format? Unfortuntately, it's all that we have so far --Abbad (WMF) (talk) 20:23, 9 February 2022 (UTC).Reply
Thank you, @Abbad (WMF)! Note that I was an election facilitator for the 2019 ASBS STV election (Affiliate-selected_Board_seats/2019) and led the effort to be transparent around the tabulation. I can figure out how to translate the formats myself, so my main notion is that from a transparency and election integrity standpoint, WMF should be publishing the results in a format that is more machine-readable for tabulation software, so that others less skilled can validate the results. In particular I know from my colleagues like Joe Kiniry that EVERY implementation of STV that he has dug in to (and that includes several in his professional work) has bugs that place it at odds with the relevant rules for what the software is supposed to compute. I dare say they are often odd boundary conditions, but I think it is important that it be easy to check the results, and I'm sorry I was too busy to dive into this months ago. The first question is probably what the best format would be, since there are many to choose from. ★NealMcB★ (talk) 21:55, 12 February 2022 (UTC)Reply
@Nealmcb Look Movement Charter/Drafting Committee/Submitted voting harks. (@Abbad (WMF) and KVaidla (WMF): Ping.) ✍️  Dušan Kreheľ (talk) 14:38, 14 February 2022 (UTC)Reply
@Dušan Kreheľ Thank you - that page is also helpful, but is still in wiki format, which we can't expect standard STV tabulation software to support. I can parse it and convert it, but I think we still need to provide the ballot data in a standard format like BLT, preflib etc. ★NealMcB★ (talk) 01:51, 16 February 2022 (UTC)Reply

Feedback

[edit]

You are welcome to share your feedback about the elections (and/or selection) process --Abbad (WMF) (talk) 13:52, 3 November 2021 (UTC).Reply