I went to update the Quick guide last night, and wanted to be sure that the wording there, and here, reflect how the election is supposed to work, since at the moment they don't quite agree. What this main elections page says now is: The minimum percentage of support that is required is 60% for a two-year term, and 50% for a one-year term. Candidates who receive 60% support will be appointed to serve a full two-year term. If applicable, candidates who receive ≥50% support but fall below the 60% threshold will be appointed to serve a one-year term.

My question is what the actual minimum vote needs to be. Since only support and oppose votes count, if there are 1000 support and oppose votes, is 500 support and 500 oppose sufficient to qualify for a seat, or does it need to be an actual majority of these with at least one more support than oppose votes. (In this case, 501 support and 499 oppose, but 500 to 499 or 501 to 500 would also hit the minimum.) I ask because both the first sentence ("minimum percentage of support that is required is ... 50% for a one-year term") and the second ("who receive ≥50% support") are saying that if 1000 eligible support and oppose votes are cast, 500 support, or exactly 50%, is the minimum required.

The Quick guide created in 2018, which may be wrong on this point, has had the following statement since its inception: Seats will remain vacant if there is an insufficient number of candidates with more supports than opposes; to date, this has not occurred. This wording does not allow a 500 support/500 oppose scenario. Is this correct, or should it be changed to "with at least as many supports as opposes" or an equivalent phrase?

As for the 60%, this seems straightforward—using the 1000 example once more, 600 support and 400 oppose (60% on the nose) would be sufficient. (Let me know if it isn't.) Thanks for your patience in wading through this. BlueMoonset (talk) 16:27, 5 October 2019 (UTC)

@BlueMoonset: Swarm, in closing the 2018 ACERFC, wrote There's a clear consensus to increase the minimum threshold of support for appointment to two-year terms to 60%. If applicable, any candidates who receive ≥50% but <60% support will be appointed to one-year terms. The operative term seems to be "≥50%"; if there's exactly 50%, that counts. That's just my read, though. Best, Kevin (aka L235 · t · c) 16:44, 5 October 2019 (UTC)

I've gone back a number of years before that change, and before the 60% level was added, it had simply said The minimum support percentage is 50%. So I think you're right, L235—it had been minimum 50% before the change, and the new 60% level required for a full two-year term did not change the 50% minimum eligibility. I'll go change the Quick guide's wording. BlueMoonset (talk) 17:39, 5 October 2019 (UTC)

You guys have got it right, good catch. ~Swarm~ ^{sting} 15:54, 6 October 2019 (UTC)

Out of interest, why do we have such a long "selection period" for the election commissioners? I understand there's a week between the end of that and the start of nominations, because they have to sign, and have processed, the NDA (it just happened that was already done this year). However, the selection is functionally instantaneous (and even if it wasn't wouldn't take more than a day), and NDA processing usually only takes a few days.

Out of interest for next year, is there a reason for this that still stands (or could stand) that I've missed or is it just a leftover from prior years? Cheers to anyone who can answer Nosebagbear (talk) 19:05, 22 October 2019 (UTC)

As I understand it, the theory is that all the discussion and voting on candidates for the commission happens during this period, with the nomination period being literally just for nominations. In practice evaluation and voting happens as soon as candidates put their name forward. This is certainly something that can be discussed next year, and indeed xaosflux has already made a note of it - see Wikipedia talk:Requests for comment/Arbitration Committee Elections December 2020#Topics to review for next year. Thryduulf (talk) 23:36, 23 October 2019 (UTC)

@Nosebagbear: As far as the 2 week period, some editors may only get on once a week - so with current timing this would give them a chance to !vote for someone that self-nominated right at the end of the current period. We should look over all the timing aspects of this next year for sure. — xaosflux ^Talk 23:43, 23 October 2019 (UTC)

Based on the proposed motion at Wikipedia:Arbitration/Requests/Motions#Temporary checkuser permission for election scrutineers it appears that Base, Shanmugamp7, and Einsbor, stewards whose home wikis are the Ukrainian, Tamil and Polish Wikipedias respectively will be the scrutineers for this year's election. Thryduulf (talk) 23:40, 23 October 2019 (UTC)

Yes. I requested the volunteers on m:SN after agreement from the other commissioners and sent an email to ArbCom requesting they be granted CU. TonyBallioni (talk) 02:22, 24 October 2019 (UTC)

I think what happens is that the seats are filled from highest % vote getter to lowest, until the number of seats run out, correct? So, just want to clarify, if there are, for example, 8 qualified persons who meet the above 60% they get two year seats by order of magnitude of thier percent, and if anyone else gets above 60% but below the others who took the 8 seats that persons gets a one year seat, or if all eleven get above 60%, the three-lowest will get one-year-seats.

There are 15 seats, four are already filled leaving 11 to be filled in the election. The current wording "eight seats may be filled for either a two-year term (January 1, 2020 – December 31, 2021) or a one-year term (January 1, 2020 – December 31, 2020), and three seats may be filled for a one-year term only. . .Candidates who receive 60% support will be appointed to serve a full two-year term. If applicable, candidates who receive ≥50% support but fall below the 60% threshold will be appointed to serve a one-year term." Seems to exclude above 60%-ters from getting any of the 3 one year seats, right? Should this be clarified? And whats with "eight seats may be filled for either a two-year term (January 1, 2020 – December 31, 2021) or a one-year term (January 1, 2020 – December 31, 2020)"? Are not all of these 8 seats two years and then there are 3 one year? Alanscottwalker (talk) 12:16, 24 October 2019 (UTC)

@Alanscottwalker: on the first point you're correct - if more than 8 get more than 60%, it just ranks highest to lowest, so yes it should be rephrased. The second part ""eight seats may be filled for either a two-year term (January 1, 2020 – December 31, 2021) or a one-year term (January 1, 2020 – December 31, 2020)"?" is correct, because if (say) only 6 people get 60+%, and 5 people got between 50-60%, then there would be 6 2 year terms and 5 1 year terms. Nosebagbear (talk) 12:48, 24 October 2019 (UTC)

Hmm? We have two tranches, about half are to be filled every two years, so it makes more sense to be clear that eight seats need to be set aside as two year seats (requireing 60% or above) and 3 seats need to be set aside as one year seats (requiring 50% or above). Alanscottwalker (talk) 12:56, 24 October 2019 (UTC)

If there are fewer than eight candidates who receive >= 60% support, but more than eight who receive >= 50% support, then the balance of the eight seats will be filled, but only for one year. As discussed last year at Wikipedia talk:Arbitration Committee Elections December 2018 § Election after expiry of one-year term, the RfC didn't really cover what would happen in the subsequent election; the participants in that discussion assumed that to maintain the tranches, the next year's election would refill those seats for another year, which is a reasonable assumption. isaacl (talk) 05:28, 25 October 2019 (UTC)

That opens the possibility that you have 11 one year seats in perpetuity. Alanscottwalker (talk) 14:03, 25 October 2019 (UTC)

Suggested amendments: eight seats may be filled for a two-year term (January 1, 2020 – December 31, 2021), and three seats may be filled for a one-year term (January 1, 2020 – December 31, 2020). . .Candidates who receive ≥60% support will be appointed to serve two-year seats until filled, and then appointed to one-year seats until filled. If one-year seats still remain open, Candidates who receive ≥50% support will be appointed to serve a one-year term until those three seats are filled.

With the current last line that says that some seats in either tranch may go unfilled, this probably takes care of it. Alanscottwalker (talk) 13:53, 24 October 2019 (UTC)

Looks good, I like it Nosebagbear (talk) 14:00, 24 October 2019 (UTC)

Alanscottwalker, I think this wording as proposed is fine and we should use this.—CYBERPOWER (Trick or Treat) 14:40, 26 October 2019 (UTC)

@Cyberpower678:, I do not feel that the consensus at Wikipedia:Requests for comment/Arbitration Committee Elections December 2018 § Percentage support needed for appointment was for the seats in the expiring tranche to be designated as two-year terms only, and thus anyone receiving between 50% and 60% support would not be able to fill those seats. If that were the case, then the RfC would have specified that candidates receiving between 50% and 60% support could only fill mid-term replacement seats. The key phrase from the proposer is This would mean that only candidates with 60% support can be appointed to two year terms, with any remaining vacancies to be filled with candidates who have 50%+ support. isaacl (talk) 17:04, 26 October 2019 (UTC)

Isaacl, It does leave the interesting question of what happens to vacancies for 2 year terms when no one gets above 60%. Then everyone gets one year, and the alpha and beta tranches are effectively dissolved. This is just my personal opinion. But yes, you do make a good point, I neglected to see it from your angle.—CYBERPOWER (Trick or Treat) 18:45, 26 October 2019 (UTC)

It is true that we'd lose the effect of keeping some kind of consistency across elections if no one receives 60%, but it doesn't necessarily invalidate the concept of tranche, assuming we as a community wanted to keep it after an election like that. If we ended up with a situation where no one receives 60% of the vote, it is true that everyone elected would get one year only and all fifteen seats would come up for election in 2020. However, only the seven seats of Tranche Alpha would be available for two-year terms, and the eight seats of Tranche Beta would be for one-year terms regardless of support level. With strong enough candidates capable of gathering sufficient support, we could get back to the tranche concept pretty quickly. Obviously the tranche system only works if we consistently nominate strong candidates whom the community will broadly support, so it's really on us to make sure that enough of these candidates actually do nominate themselves. CThomas³ (talk) 22:33, 26 October 2019 (UTC)

My apologies; I did not realize this was under discussion before I modified the elections page to further describe how the seats are assigned. How would everyone like to proceed? isaacl (talk) 05:07, 25 October 2019 (UTC)

The election commission needs to weigh in. We either are going to maintain two tranches or we are going to allow potentially the election of many one-year-seats above the three that were discussed in the 2019 RfC. My wording was aiming to maintain the tranches which aim to have half the committee selected every two years, which means you fill the two year seats with above 60 percenters and if no one is left, you leave them unfilled. And you fill three one-year seats with the above 50%ters, which include above 60%ters who did not get one of the 8 seats. I think your wording opens the possibility of 11 one-year seats, and never getting back to the half the comittee elected every two years.I guess we could go either way, although I think my way was more in tune with the past, but the commission needs to nail this down. -- Alanscottwalker (talk) 13:46, 25 October 2019 (UTC)

I don't think the concept of two tranches has changed at all. The eight seats that could be filled with either 2-year or 1-year terms are the eight from Tranche Beta. The three seats that can only be filled with one-year terms are those that are replacing those that resigned/retired from Tranche Alpha. Perhaps we should be explicit about that on the election page.

Here is my interpretation of how this will/should happen: If at least eleven candidates receive 50% or more of the vote, we will fill all eleven seats; otherwise we will fill only as many seats as we have candidates with at least 50% support. All candidates will be rank ordered from most support to least. The top eight, assuming all are at least 50%, will be seated in Tranche Beta, with those receiving 60% or more given two-year terms and those receiving less than 60% receiving one-year terms. The next three, again assuming all are at least 50%, will be seated in Tranche Alpha for a one-year term. If there are eight or fewer candidates with at least 50% support, they will all be seated in Tranche Beta and none of the Tranche Alpha seats will be filled this time around. CThomas³ (talk) 16:35, 25 October 2019 (UTC)

Right now there are eight seats in tranche beta up for renewal. For simplicity, let's ignore vacancies in tranche alpha. If five persons reach 60% support and three more reach 50%, then the discussion last year felt that the most reasonable approach was to appoint three persons to a half-term in tranche beta, and fill those seats in the following year with a one-year term, so the next year would have the eight seats of tranche alpha available for a two-year term and three for a one-year term. So every year there will be one tranche up for renewal with a two-year term available, for candidates who reach at least 60% support. The 2018 RfC question at Wikipedia:Requests for comment/Arbitration Committee Elections December 2018 § Percentage support needed for appointment did not specify that the number of candidates elected to a one-year term based on attaining only 50% support would be limited to the number of vacancies in the tranche not up for renewal. isaacl (talk) 16:43, 25 October 2019 (UTC)

There is no need in lengthy posts to go over this again. We have two choices, either all eleven can be filled with one year terms (and we somehow finesse the "8 2-year terms" language voted for in the 2019 RfC) or only three can be filled with one year terms. If the former, that means we may get a super-majority of the ctte to be one year terms, and again next year a high turning-over election, and high turnover elections annually for the foreseeable future. (possibly never getting back to half the committee being experienced together) -- Alanscottwalker (talk) 17:25, 25 October 2019 (UTC)

This depends on their being candidates who receive >= 60% support. If the seats for the expiring tranche are left empty when there are no candidates who reach 60%, then either the entire committee will empty out, or, if the vacant seats are put up again for a one-year term in the following year, then there will also be a high turnover of one-year terms. The only difference is whether or not the seats will stay vacant in the meantime. isaacl (talk) 17:47, 25 October 2019 (UTC)

Not just vacant, the group working together is either highly inexperienced working together and doing the work (and with lower community support), or has a smaller group to work with and on-board or acclimate. Alanscottwalker (talk) 18:07, 25 October 2019 (UTC)

Sure, both scenarios will have groups that may be inexperienced with working together, and the scenario that perpetually leaves half the seats empty will have a smaller group that garnered greater community support, though in all cases it will be above 50%. isaacl (talk) 18:16, 25 October 2019 (UTC)

And the tranch system was designed with the idea that roughly half the committee will not stand in election every year for continuity and experience. (One year terms initially were to complete an uncompleted two year term, and the 2019 RFC only spoke of "8 2-year terms" and the rest one-year terms. But now, I take from what you wrote that one year terms are potentially all terms, regardless of tranch. Last year, we eliminated 3 2 year-terms and this year we are replacing them with optional two or one year terms, and doing the same for all tranch seats). So, I take it, we have designed by ad hoc accretion, allowing the super-majority or entire committee standing annually, there may or may not be any 2 year terms so that experience/continuity remains half or some number on the committee, depending on only percent. -- Alanscottwalker (talk) 13:34, 26 October 2019 (UTC)

The issue is that based on past voting history, participants in the 2018 RfC assumed that most of the seats in the expiring tranche would continue to be filled by candidates with >= 60% support, thus maintaining continuity. This may be an assumption worth revisiting. isaacl (talk) 17:01, 26 October 2019 (UTC)

Would any of the non-reserve members of the electoral commission like to discuss this topic? isaacl (talk) 19:18, 12 November 2019 (UTC)

I haven't commented only because I thought this was resolved three weeks ago. The minimum threshold percentages are 60% for two-year terms and 50% for one-year terms. That same RFC is quoted to say If applicable, any candidates who receive ≥50% but <60% support will be appointed to one-year terms. There are 8 open spots on Tranche Beta, so one of the following (let's call them "TB") will happen:

1. At least 8 nominees receive ≥50% support. Ranked in order of percentage, those receiving ≥60% will receive two-year terms and the rest (if applicable) will receive one-year terms.
2. Less than 8 nominees receive ≥50% support, in which case those receiving ≥60% will receive two-year terms and the rest will receive one-year terms, with the remaining spots left vacant.

There are also 3 spots open on Tranche Alpha, so one of the following will happen.

1. TB1 happens, which case:
   1. If there are ≥3 nominees remaining receiving ≥50% support the top three will receive one-year terms (regardless of overall percentage)
   2. If there are <3 nominees remaining receiving ≥50% support they will receive one-year terms (regardless of overall percentage) and the remaining spot(s) will remain vacant
2. TB2 happens; the three spots on Alpha remain vacant.

Based on past election results (mentioned in a similar discussion last year) it is extremely unlikely that the majority of those elected will not be elected by ≥60% support. Primefac (talk) 20:16, 12 November 2019 (UTC)

The page currently says "If any of the seven arbitrators with unexpired terms resign or otherwise leave the committee before the start of voting, the seat they vacate will be filled for a one-year term in this election." But that does not seem right, only four of the current arbitrators remain arbitrators after January 1, if they resign their seat has to be filled but the 3 whose terms are expiring are already being replaced by new two year terms. Alanscottwalker (talk) 13:43, 24 October 2019 (UTC)

Suggestion: If any of the four arbitrators, who are in midterm, resign or otherwise leave the committee before the start of voting, the seat they vacate will open another one-year-term seat for this election." Alanscottwalker (talk) 14:09, 24 October 2019 (UTC)

• I'd suggest removing the numerical portion all together and change it to "If any of the arbitrators who are in midterm". But yes, there are only 4. Worm^TT(talk) 17:10, 26 October 2019 (UTC)

Thanks, I made the change, but I kept the referral to "four current arbitrators" because that helpfully, I think, repeats the phrase from the first sentence in the Vacant Seats section, but if you want me to change it more take a look and let me know. Alanscottwalker (talk) 00:08, 27 October 2019 (UTC)

Is it a bit of a worry that there are only three candidates so far? Is there usually a flood of nominations towards the end of the period? It will be a mighty small ArbCom on current numbers, even assuming all three got the 50%. Would it even be viable? Peacemaker67 (click to talk to me) 06:09, 6 November 2019 (UTC)

Peacemaker67, It's not terribly uncommon. SQL^{Query me!} 09:24, 6 November 2019 (UTC)

Thanks for that, I was starting to get concerned, not being well-versed in the way it usually develops. Hopefully things will work out and we will have enough candidates to make a real choice. Peacemaker67 (click to talk to me) 09:28, 6 November 2019 (UTC)

@Peacemaker67:: This happens every year. The high-profile candidates only throw their hats in the ring at the last minute. Hawkeye7 (discuss) 08:08, 12 November 2019 (UTC)

Thanks Hawkeye. This is the first year I've monitored the nomination process and asked a question. An eye-opener. Peacemaker67 (click to talk to me) 08:25, 12 November 2019 (UTC)

I've considered off-and-on whether I'd be willing to run. If I were going to stand for an election, though, I'd rather stand in the steward elections. NinjaRobotPirate (talk) 13:54, 9 November 2019 (UTC)

NinjaRobotPirate, Go for it. You'd probably pass. Then I'll have my own personal steward in my pocket. ;-) —CYBERPOWER (Chat) 15:16, 9 November 2019 (UTC)

It seems possible that we could get a global CheckUser feature now that the WMF is working on the CU extension again. That would make cross-wiki sock puppetry a lot easier to deal with. If I were to go looking to heap more responsibilities on myself, I feel like that's where I'd be more useful. On the other hand, it might be useful to have a couple people familiar with the CU tool on Arbcom. But since the committee hasn't said anything, I guess it's not currently an issue. NinjaRobotPirate (talk) 01:12, 10 November 2019 (UTC)

I'm obviously misreading it, but ~2.3 seems to read that a voter can qualify by making 10 live edits, etc., from the night of Friday, etc., onwards; how is this possible?! ——SN54129 18:21, 12 November 2019 (UTC)

I assumed it means 10 live edits between Nov-1-2018 and Nov-1-2019.--Cube lurker (talk) 18:24, 12 November 2019 (UTC)

Yes, me too. Although it is a bit awkwardly worded.-- P-K3 (talk) 18:26, 12 November 2019 (UTC)

Ah! Thanks both both. Obviously that makes sense. Something like ...within one year prior to Friday 00:00, 1 November 2019 perhaps? ——SN54129 18:40, 12 November 2019 (UTC)

Changed to ...edits (in any namespace) in the one year prior to Friday... Primefac (talk) 19:10, 12 November 2019 (UTC)

Hey, thanks very much Primefac, nice one. Much better  :) ——SN54129 19:15, 12 November 2019 (UTC)

Is there a section for questions to ask for the 2020 RFC? I think we really need to discuss whether candidates are eligible to write their own guide, since a few this year seem to think that it is okay. --Rschen7754 19:45, 12 November 2019 (UTC)

@Rschen7754: Here you go! Lord Roem ~ (talk) 19:54, 12 November 2019 (UTC)

I'm withdrawing my candidacy. I feel it's only right to give a quick explanation as this is after the cutoff time. Simply put, I don't think I have a chance in the current field of candidates. I had to weigh whether it makes sense to have ACE2019 in the back of my mind for the next month when it looks increasingly likely that, even with 11 vacant seats, my odds are quite low. There are 9 current or former arbs running, each with a fairly high likelihood of taking a spot. That leaves 15 other candidates competing for 2 seats, or conservatively, 4-5 if some of the ex-arbs don't make it. The guides last year were very indicative of the final results, and those are fairly negative based on my activity levels over the past year.

The one thing that had me hesitating is a deep concern that the new Arbcom is going to largely look like past Arbcoms. The folks running are all incredibly talented, and there's many I respect very dearly and want to see return... but, at this critical juncture for the project, after a destabilizing year, fresh faces with new ideas are essential. There is a crisis of confidence in the Committee, in its slow-moving mechanics, its lack of transparency, and a structural inability to fix its perception as distant and difficult to engage with. I sincerely hope whomever is elected takes these concerns to heart. There is a better Committee out there, we just need to build it.

Thank you to those who lent their support, asked me questions, or took the time to read my answers.

Best, Lord Roem ~ (talk) 05:29, 13 November 2019 (UTC)

While I haven't decided who I'm voting for yet, you were certainly on my list of possibles as I think the Committee needs a mix of old and new faces and there is at least one former arbitrator who will definitely not be getting my vote. Thryduulf (talk) 09:06, 13 November 2019 (UTC)

I absolutely agree with your point that should we end up with current or former arbitrators taking the lion's share of the 11 available positions, the chances of improving things will be greatly reduced. Fish+Karate 10:07, 13 November 2019 (UTC)

Should the ballot state that eleven (11) are to be elected?Eschoryii (talk) 01:14, 19 November 2019 (UTC)

The ballot is just the ballot, and includes links to all of the relevant information (including the number of potential elected members). I would also suspect that there might be some undue bias introduced if the ballot itself were to state that, because a user can support as many candidates as they would like, and we shouldn't make them feel like they can only vote for 11. Primefac (talk) 02:26, 19 November 2019 (UTC)

Went theough all the hassle of voting, reading all the candidates offering, submitted my vote and it says you must log on to vote, which i was!!!You've had your chance I'm not going to do it again!!--Petebutt (talk) 02:05, 19 November 2019 (UTC)

Petebutt, I would assume you remembered (even vaguely) how you voted; there's no requirement to re-read all of the candidate statements, just the ticking of a few radio buttons. But hey, you do you. Primefac (talk) 02:28, 19 November 2019 (UTC)

I also got that when I first tried to vote. I suspect there might be a fairly short session timeout, so that one has to submit the votes quickly after opening the ballot. It might be good to add a note to that effect to the ballot, so people aren't caught off guard (I know it suggests to record one's votes, but it does so in a different context, and I don't think it talks about the potential for them getting eaten by the computer, although that is the general rule with computers...). —{{u|Goldenshimmer}} (they/their)｜Talk｜Contributions 06:30, 23 November 2019 (UTC)

See Wikipedia_talk:Requests_for_comment/Arbitration_Committee_Elections_December_2020#General_complaints_and_improvements for suggested improvements to next year's directions. — xaosflux ^Talk 00:45, 24 November 2019 (UTC)

Hello

How do I ask a question to all candidates? Thanks,

Regards, --Gryllida (talk) 02:59, 19 November 2019 (UTC)

Go to each candidate question page and ask your question. There is no way to ask the same question automatically to every candidate. Primefac (talk) 03:00, 19 November 2019 (UTC)

What's with this list? It says 0 people have voted. Is there a different link? ---Sluzzelin talk 06:04, 19 November 2019 (UTC)

The voting occurs on votewiki, not here. votewiki:Special:SecurePoll/list/759. — JJMC89 (T·C) 06:16, 19 November 2019 (UTC)

Thank you, JJMC89! ---Sluzzelin talk 06:18, 19 November 2019 (UTC)

If I go into the booth now, can I vote for those that I'm sure about; and go back a second time later, if and when I've made up my mind about those I'm not yet sure of? Thanks! Usedtobecool ^TALK ✨ 06:15, 19 November 2019 (UTC)

You have to make sure you submit your full preferences each time, since they'll strike your earlier vote(s) entirely. I don't believe it 'saves' your preference for a specific candidate, if that's what you mean. Lord Roem ~ (talk) 06:17, 19 November 2019 (UTC)

Lord Roem, hey, thanks! I guess I'll make up my list offline and go in once I know exactly how to vote on every candidate then. No point making incomplete votes that will be struck. Thanks again! Usedtobecool ^TALK ✨ 06:21, 19 November 2019 (UTC)

Hi, I received a notification for the voting on my User Talk page, but not for the preceding self-nomination period. Why? Regards, --oSeveno (User talk) 10:11, 19 November 2019 (UTC)

@OSeveno: the election RfC's have only supported sending out a single mass message to talk pages. The nomination period was advertised as a watchlist banner, in T:CENT, and in other various places. I don't expect this to change in next year's RfC, but you can certainly bring it up for discussion then if you would like. — xaosflux ^Talk 12:30, 19 November 2019 (UTC)

@OSeveno: In case you aren't aware, the list of things to discuss in next year's RFC is at Wikipedia talk:Requests for comment/Arbitration Committee Elections December 2020. The WP:ACE2020 shortcut currently targets that page. Thryduulf (talk) 13:09, 19 November 2019 (UTC)

I'm still opposed to IP (and other personal) information going to scrutineers. Wikipedia adopted this method 7 or so years ago. I haven't voted in Arbcom elections since. GoodDay (talk) 16:39, 20 November 2019 (UTC)

GoodDay, then start an RfC to stop this. Scrutineers are stewards and have the trust of the global community and WMF. —CYBERPOWER (Chat) 19:19, 20 November 2019 (UTC)

I may do so, in January. GoodDay (talk) 19:42, 20 November 2019 (UTC)

My understading was, they only really use the checkuser bit when votes seems suspect in order to strike them from the roll, and then pass on said infomation to our local CU team in order for them to action it via blocks or other means. On top of that, they are stewards who have all signed the Confidentiality agreement for nonpublic information, and would likely be in a ton of trouble with the WMF if they released any non-public info to the public. CodeLyoko^talk 16:22, 25 November 2019 (UTC)

CodeLyoko, scrutineers see all CU data that is collected at votewiki for each voter without having to run any checks. It is all there on votewiki for looking at by them. They are given local CU access because one action is rarely enough to come to a definitive conclusion, and giving them access to en.wiki CU data which usually will provide a clearer picture if suspicions exist. In cases where it’s a well-established user or the data is complex, they will usually request the assistance of local CUs. TonyBallioni (talk) 01:16, 26 November 2019 (UTC)

@TonyBallioni: given this has come up a couple of times, it might be useful to show an example (for user:Example and with fictional data, obviously) of what the scrutineers see. If that is possible and allowed. Thryduulf (talk) 09:56, 26 November 2019 (UTC)

@Thryduulf: as far as private data goes, it is: (1) Your IP Address, (2) Your User-agent. See File:SecurePollSample-2019-11-26.PNG. — xaosflux ^Talk 10:31, 26 November 2019 (UTC)

That is what is on every vote without using the actual CheckUser tool, which is the normal process. — xaosflux ^Talk 10:31, 26 November 2019 (UTC)

Thank you, that's helpful. Thryduulf (talk) 11:54, 26 November 2019 (UTC)

FYI: Assuming nothing has changed since I served as one, the election coordinators also have access to the data on votewiki. —DoRD (talk)​ 03:03, 27 November 2019 (UTC)

DoRD, Since last year, they do not. SQL^{Query me!} 03:12, 27 November 2019 (UTC)

Correct, neither the coordinators nor the commissioners are meant to have access to this once the election starts. In certain situations (such as in pre-election testing, or if there is a delay or errors in accoutn processing) pre-election access may temporarily overlap the start of the election, which is why these users must execute the NDA and get approved by arbcom under the checkuser criteria. — xaosflux ^Talk 03:28, 27 November 2019 (UTC)

Thanks for correcting me. —DoRD (talk)​ 03:59, 27 November 2019 (UTC)

In addition to that screenshot (xaosflux provided), the very same data showed up (IIRC — it's been a year since I last checked and I do not have an access to votewiki now) in voter list without even having to go to voter page, from the voters list. I sort of recall seeing IPs there, and I am sure I saw some sort of Yes/No data (I am not saying 'what kind of yes or no? per WP:BEANS) — regards, Revi 19:40, 3 December 2019 (UTC)

@-revi: yup - the "list view" for the scutineers looks like this roughly:

Time 	                Name 	        Domain 	                IP 	        XFF 	                        User agent 	           CSRF 	Dup
2019-11-26T10:20:26 	Xaosflux 	en.wikipedia.org 	a.b.c.d 	e.f.g.h, i.j.k.l, m.n.o.p	Mozilla/X.Y (qqqqq) zzz 	  	 

— xaosflux ^Talk 19:48, 3 December 2019 (UTC)

Scrutineering period (immediately following the voting period) → scrutineers, consisting of stewards whose main wikis are not the English Wikipedia, will check the votes (e.g. for duplicate, missing, and ineligible votes), and compile a tally of the results. Leaky caldron (talk) 18:44, 4 December 2019 (UTC)

I suspect that's an old potential issue, but I can easily see someone either forgetting to select (before "neutral" was the default) or the software removing votes from the final tally, meaning that the total votes cast might be different between candidates. Primefac (talk) 18:52, 4 December 2019 (UTC)

(edit conflict) @Leaky caldron: if it is completely "missing", nothing to do, else this would be reconciling that that tally of votes for candidates equals the number of voters to see if something went awry and didn't record an actual vote. — xaosflux ^Talk 18:54, 4 December 2019 (UTC)

It would be great if in future user talk page Election notices, there was an indication that the voting deadline was in the UTC time zone. I usually recognize this in regular editing and make adjustments for the differing time zones and the Election pages DO indicate the deadlines are UTC time zones but the user talk page messages didn't. It just says 23:59 on Monday, 2 December 2019.

The past two weeks have been a hectic ones for my family with packing up a house and moving and this one detail slipped past me. It would be a simple correction to just indicate in the user talk page message that the deadline was midnight...in the UK. I'm not sure where to put this suggestion so I'm leaving it here. Thanks. Liz ^{Read! Talk!} 15:49, 5 December 2019 (UTC)

@Liz: thanks for the feedback, I've updated the message template at Template:ACEMM to include the UTC TZ for next time. — xaosflux ^Talk 15:58, 5 December 2019 (UTC)

I come up with 1673 votes counted based on the numbers you've posted. Can someone official check that? 🌿 SashiRolls ^{t · c} 14:29, 14 December 2019 (UTC)

@SashiRolls: all of the votes are listed here - are you seeing a potential "off by one" situation? It looks like there is one unique vote status vote, 16:09, 20 November 2019 TimBuck2 : a vote that was both a double vote that was automatically discarded and also a stricken vote. — xaosflux ^Talk 14:39, 14 December 2019 (UTC)

It is amusing that the difference in votes counted and valid votes reported is exactly 22. I thought at first all the Arb Candidates votes were thrown away, but I'd noticed at least one didn't vote, so it can't be that... :P

Maybe, the missing 22 votes will be easier to see in my sandbox where I've added neutral and oppose together into !support?🌿 SashiRolls ^{t · c} 14:54, 14 December 2019 (UTC)

I see what you're saying... 1695 valid votes - 23 struck = 1672 which is off by one from 1673. 1783-111=1672. 🌿 SashiRolls ^{t · c} 14:58, 14 December 2019 (UTC)

@SashiRolls: One vote was both "thrown away" and also stricken. 1873 votes were made, 88 were thrown away, leaving 1783-88=1695 votes, of those 1695 votes 22 were further invalidated, 1695-22=1673. Does this resolve your concern? — xaosflux ^Talk 17:54, 14 December 2019 (UTC)

Yep, I'll fix the page so that it reflects this new info if it hasn't been done already. (In other words there were not 1695 valid votes as this page originally reported, but 1673.) Thanks for finding the source of the 22 extra valid votes. (I've gone ahead and fixed the numbers you accidentally inverted in your explanation above. I hope you don't mind. 🌿 SashiRolls ^{t · c} 18:48, 14 December 2019 (UTC)

@SashiRolls: seems fine - some of this may be confusing wording across the systems as well (there were votes, then some were automatically considered valid, then some were invalidated). — xaosflux ^Talk 19:19, 14 December 2019 (UTC)

• Much thanks to the scrutineers for working to bring this process to conclusion. A very good result. – Ammarpad (talk) 17:43, 14 December 2019 (UTC)

I want to thank those who ran to serve on ArbCom, both those who won and those who may win in the future. Volunteering in this capacity is not for the faint of heart, and as an active volunteer in service of open knowledge, I fully appreciate those who serve to help our community be better while often spending more time than anybody really appreciates. Thank you to those who ran and those who led this election process. --- FULBERT (talk) 16:22, 28 December 2019 (UTC)