Wikipedia talk:Arbitration Committee Elections December 2017

2017 Arbitration Committee Elections

Status

• Voting in the 2017 Arbitration Committee elections is now closed. The results have been posted to Wikipedia:Arbitration Committee Elections December 2017.
• Feedback on the election may be left on the feedback page.

It is currently 14:53 (UTC), Sunday, 24 November 2024 ()

Strange...

Moved from Wikipedia talk:Arbitration Committee Elections December 2017/Candidates

For some reason User:Shock Brigade Harvester Boris's application seems to have gotten lost in the mail. Drmies (talk) 01:33, 16 November 2017 (UTC)

That doesn't appear to be the case. My, such unrequited love. wbm1058 (talk) 17:07, 20 November 2017 (UTC)

Social scientists

This is the place for general candidate discussion, yes? Can we get some sociologists and social scientist noms, or have any editors with such background expressed an interest elsewhere? I'm most interested in candidates who have knowledge of and experience evaluating social systems. czar 00:06, 17 November 2017 (UTC)

Nominations

On 21 November 2017,‎ @QEDK: changed the order of the standing candidates with the edit summary "just helps preserve the chronology for people making voter guides or so". What is the basis for the revised order? Cwmhiraeth (talk) 13:49, 22 November 2017 (UTC)

Just speculating here but the intent seems clear: Since the candidate list is randomized, people creating voter guides will probably either go alphabetically or use the order in which the statements were added. However, the latter only works when people add their candidacies at the bottom and not, as has happened, partly at the top and partly at the bottom. QEDK seems to just have restored said order. I see no harm. Regards SoWhy 14:05, 22 November 2017 (UTC)

It's helpful for anyone who has already asked some candidates questions or written a voter guide, as it helps them work out who they need to go back and add. Voters may also be interested. Someone who declares early opens themselves up to more scrutiny, particularly by question. Voters might also subjectively choose to interpret later declarations in terms of candidates' state of mind. Or they might not. In any case, it's not a bad thing to do. --Dweller (talk) Become old fashioned! 14:23, 22 November 2017 (UTC)

SoWhy and Dweller put my intentions exactly. Also, I realized it's easier if anyone wants to copy over the links to the Questions page, since that's done chronologically. --QEDK (愛 • 海) 14:39, 22 November 2017 (UTC)

I started this thread because I couldn't understand why QEDK would want to rearrange the candidates into an apparently random order, but I now see that, every time a new candidate entered the fray, or in fact whenever anyone edited the page, the software rearranged the order of the candidates. This is unhelpful to anyone wanting to make a voter's guide. It would be better in my view if a chronological order was maintained throughout the nomination process, and if a random order was wanted before voting takes place, the randomisation should be done after the nomination period has closed. Cwmhiraeth (talk) 09:51, 23 November 2017 (UTC)

The module randomizes on each purge/load by client, not just each edit or with new entries. It's done primarity to prevent a tl;dr sort of bias, as candidates at the top would get undue importance simply because they occurred higher in the list. Also, we cannot exactly have the solution you have in mind, because for the period it's up, they would get the priority, leading candidates to apply as fast as possible to gather most audience. --QEDK (愛 ☃️ 海) 17:46, 24 November 2017 (UTC)

Overvotes

I clicked the submit button once but the log has recorded that I voted twice. Why is that? Finger too long on the submit button due to elderly stubby fingers unable to work modern technology? Or what? How can I now be sure that my vote is recorded accurately (I opposed three and supported four)? Should I re-submit to be sure? DrKay (talk) 18:00, 27 November 2017 (UTC)

@DrKay: You can be sure your vote counted if you have one non-struck vote in the log. You can always resubmit if you want, but you should be fine. ~ Rob₁₃^Talk 18:10, 27 November 2017 (UTC)

Mass message?

Is there a reason a mass message hasn’t gone out yet? ~ Rob₁₃^Talk 18:46, 28 November 2017 (UTC)

I am against it, as pointed out before (forgot where). IF a message, please avoid "now open" in the header, which will become wrong right after the election. I keep removing the ugly and dated thing from the pages of users who don't take care of their talk anymore. IF going round, can a bot perhaps remove 2016 and 2015 messages that are still on talk pages? --Gerda Arendt (talk) 18:51, 28 November 2017 (UTC)

Looks like technical difficulties over compiling the list without errors, and then to trim it down to only users that have edited in the past year. Alex Shih (talk) 18:55, 28 November 2017 (UTC)

• @Yunshui: - I do think this is a critical issue for the election. As of a few minutes ago, only 261 votes had been cast, compared to well over 1,000 in the same period in 2016. You can however take the time to do this right over the next few days. In 2016 the total number of voters was only 1,950 , so the mass message would likely do most of its work in just a few days. So proceeding will all deliberate speed is probably the best way to go, rather than trying to do it all in 1 day. I understand that there are technical difficulties. anything you can do would be appreciated. Smallbones_(smalltalk) 19:16, 28 November 2017 (UTC)
• I agree with Smallbones. This is a critical issue. The whole point of the bot is making it so users who aren't familiar with WikiPolitics know that they have a voice in this election, and it should go out as soon as possible to give them time to vote and review the candidates. TonyBallioni (talk) 19:26, 28 November 2017 (UTC)

And that's what you want, the votes of users who are not familiar with the politics?? Brexit ... German election: that's what you get. - If you want to inform all who log in (and a message won't reach those who don't anyway): why not something like telling us about Asia month, something that comes up on top and blinks on top and then is mercifully over, instead of talk pages clogged with soon-to-be-stale messages. --Gerda Arendt (talk) 20:07, 28 November 2017 (UTC)

Step right up, place your bets as to who gets to be Arbcom 2017 .... oh dear, it's gone in "0", all powers return to Jimbo, bad luck.....

Don't look at me Gerda, I went straight down to the polling station at 8am and voted "remain"! Anyway, I remain unconvinced this is a worthwhile exercise, except perhaps for notifying people who couldn't vote last year, but now can. For everyone else, they already know, they're indifferent, or they'll get upset / annoyed / angry about "spam". As for the lack of votes, perhaps that's due to general indifference that ArbCom can do anything, cases take forever, emails are never answered etc. (I realise there are explanations for these things, but that's not what the electorate sees Ritchie333 ^{(talk) (cont)} 20:16, 28 November 2017 (UTC)

Ritchie, thank you. What I see is that we have 12 decent candidates, so we could though a dice without changing the world much. Thank goodness, arbcom became mercifully unimportant. --Gerda Arendt (talk) 20:38, 28 November 2017 (UTC)

• The election coordinators have responded at Wikipedia talk:Arbitration Committee Elections December 2017/Coordination#Mass message. Mkdw ^talk 20:27, 28 November 2017 (UTC)

Wow, some people seem super desperate to get more coverage here. Mind you, at about 30 votes a day right now, this is really becoming something of an omnishambles. I'm sure it'll give Jimbo enough ammo to spike the whole thing. But what then? The Rambling Man (talk) 22:22, 1 December 2017 (UTC)

We're working on it. :) With a voter pool the size of Gloucester and with no simple way to filter out those who haven't edited in years, it's proving more difficult than one might expect to alert voters to this election without posting literally thousands of messages nobody will read. Joe Sutherland (WMF) (talk) 22:40, 1 December 2017 (UTC)

Well, as I've already said (elsewhere), most people will have actually considered how they would alert the eligible voters before being three or four days into the election. WMF clearly lacks the technical ability to communicate with the subset of editors eligible to vote, that's a serious failing on WMF's behalf. Fixing it (or trying to) half-way through the actual voting period is a complete joke and should render the entire process null and void. The Rambling Man (talk) 22:44, 1 December 2017 (UTC)

Sure. We're going to continue working on a solution. Joe Sutherland (WMF) (talk) 22:58, 1 December 2017 (UTC)

I'm not sure that actually reflected an answer to the previous statement. Continue to work on a solution, "sure", but you should have had a solution before the election criteria were even declared. Right now you're running a sham of an election with barely a couple of dozen people voting per day. And changing the methodology mid-way through voting should also be subject to serious scrutiny. We are where we are and somehow trying to modify the way things are working mid-vote is actually tantamount to vote-rigging. So "Sure. We're going to continue working on a solution" is all very well for 2018, but right now, you can't change the baseline without nullifying the results. The Rambling Man (talk) 23:08, 1 December 2017 (UTC)

I'm not sure I follow you on "changing the methodology". This is get-out-the-vote, not a changing of how the vote is taking place. ~ Rob₁₃^Talk 23:12, 1 December 2017 (UTC)

No, this is the "change how we advertise the vote because some people are getting edgy about not getting enough votes". WMF are failing wholesale on this, you must never advertise a set of voting criteria that you cannot differentiate from the rest of Wikipedia, and WMF have done just that. It's a complete joke and in most circumstances would render the entire process void. Even so, in such circumstances, the WMF should be honest and open about the entire situation. The Rambling Man (talk) 23:20, 1 December 2017 (UTC)

I'm not sure how the WMF's opinion is relevant to a decision made by community consensus. I'm also not sure why multiple candidates in the election are arguing publicly about this during the vote. power~enwiki (π, ν) 23:38, 1 December 2017 (UTC)

To your first point, I agree, we should all abide by the results, regardless of WMF's incompetence of arranging an election over which they can't suitably govern. To your second point, I agree, why are some candidates making so much fuss over this? It is what it is. If the urge to become an Arb is so strong, wait until WP:ACE2018, by which time hopefully WMF will have got their house in order and be actually technically capable of hosting an election with such intricate eligibility requirements. The Rambling Man (talk) 23:41, 1 December 2017 (UTC)

The eligibility requirements for voting are actually quite simple - the problem seems to be the requirements set by the community this year for restricting the mass message to a smaller subset of eligible voters rather than to everyone who's eligible. Boing! said Zebedee (talk) 23:54, 1 December 2017 (UTC)

Hmm...I, after going through the entire RFC discussion find increasingly swayed by Tony's and Rob's arguments.There was a near-unanimous consensus in the pre-ACE RFC, which was participated by numerous respected editors, whose spirit was that some sort of mass-message shall go out .The exact requirement, generated by Floq was hardly mentioned except in a few !votes and seems to be the result of Floq's own choice out of some criterions that were loosely thrown about.Greater participation was the major argument in most of the !votes.Also, I don't find much problems in any delay in the process.So, in a sense, not-sending any mass-message at all do seem to be more against the consensus than to send the message to all eligible editors.But, I am fairly opposed to deviate from the timelines and extend the voting-periods etc.Winged Blades ^Godric 16:57, 2 December 2017 (UTC)

No, it was not "Floq's own choice out of some criterions [sic] that were loosely thrown about", it was using the exact same criteria that were decided on the year before; we do things the same way each year unless there is consensus to change it. If you don't know what you're talking about, it's usually better to ask, rather than just casually assume I supervoted. --Floquenbeam (talk) 20:38, 2 December 2017 (UTC)

But, the point that remains is that since the election commissioners were chosen so early, how were the problems not resolved with help from competent parties (and WMF), already? More specifically, why the haste that is prominently visible across different t/p(s) not visible prior to a week back?Winged Blades ^Godric 16:39, 2 December 2017 (UTC)

Whatever happens to the list, we badly need smth like Wikipedia:Arbitration Committee Elections/Memo for Electoral Commission. I hope current and former commissioners will find time after elections to add best practices on this page. I may start the page but I only encountered a limited number of things while I was still on the commission.--Ymblanter (talk) 17:19, 2 December 2017 (UTC)

An excellent idea; I was going to suggest something similar. --Floquenbeam (talk) 20:38, 2 December 2017 (UTC)

If I can temporarily take my candidate hat off, and replace it with my "person who analyzed the voter data for the past two years" hat: in 2015 the original plan was to message those who had edited within the previous three months. The message was then accidentally sent to all eligible voters with no filter for past edit times. There was a great deal of concern at the time over what would happen if large numbers of long-dormant editors voted. In the final analysis, about 10% of the voters had not edited within the three months preceding the election. That gives 10% as a generous upper bound on the actual effect of dispensing with this requirement. (I suppose I could check the one-year numbers if there's interest, but I'm going to make an edumacated guess at 3%.) Opabinia regalis (talk) 04:31, 3 December 2017 (UTC)

For complete transparency to the voting public, the WMF should, after the election concludes, release a graph of cumulative votes, to establish the impact that a mass message makes. The "mass message" should have been sent out three weeks ago, yet here we are, mid-vote, and it's interesting to see the rush to get out a spam festival. The impact of these kinds of messages needs to be assessed quantitatively, particularly mid-way through the election to which they relate. The Rambling Man (talk) 21:36, 3 December 2017 (UTC)

Anyone could create such a graph using the data at https://vote.wikimedia.org/wiki/Special:SecurePoll/list/673, which shows the votes and the day/time they were cast. So far: — Diannaa 🍁 (talk) 22:27, 3 December 2017 (UTC)

Date	Votes cast
Nov 27	222
Nov 28	84
Nov 29	58
Nov 30	30
Dec 1	37
Dec 2	45
Dec 3	433
Dec 4

Thanks for the mass mailing! I wish you had all let me known you wanted candidates earlier (or at least within my wikipedia projects, ie WP:feminism). There are some great candidates out there, that might have been missed! Fred (talk) 22:25, 3 December 2017 (UTC)

Next time you send a mass mailing, could you please consider excluding bots (e.g. User talk:BattyBot#ArbCom 2017 election voter message. Thanks! GoingBatty (talk) 22:54, 3 December 2017 (UTC)

There were a lot of technical difficulties that made it hard to get the list down to the ideal set, discussed at Wikipedia talk:Arbitration Committee Elections December 2017/Coordination. Boing! said Zebedee (talk) 23:34, 3 December 2017 (UTC)

Stupid question time! Yay!

Forgive me for being so obtuse but I can't seem to find this info...how many people are being elected to ArbCom? In other words, how many slots are available? Thanks, Shearonink (talk) 22:42, 2 December 2017 (UTC)

8 vacant seats. See #Vacant seats section on this talk page's project page for more info. --Floquenbeam (talk) 22:46, 2 December 2017 (UTC)

Thanks Floq. I wish that information were clearly stated somewhere on the voting page, I couldn't find it.

If anyone else comes here looking for the same info it's

• EIGHT/8** vacant slots. (Wanted to make it clear in case anyone else comes asking the same question Floquenbeam). Shearonink (talk) 21:30, 3 December 2017 (UTC)

Thoughts from a first-time voter

I'm not sure if this is even the place for this, but the whole process for this seems a little odd and unrefined. I was hoping somewhere there would be something resembling a guide for the voting process that would lay out what is considered necessary to make an informed vote, but all of the 'guides' listed are just breakdowns of how some people voted and why for the most part. I had to do a lot of reading on the candidates that took a sizable amount of time before I felt like I was in any position to make an informed vote and there didn't seem to be much help along the way. The graphic in the vacant seats section is large and not particularly helpful. The guides section just links to the user guides which aren't really guides. Any information someone new on the scene would care to learn has to be gleaned from the statements and questions which contain lots of references and abbreviations that make knowing who your voting for somewhat inaccessible. I enjoy participating in the community when I get the chance and was interested at the notification about the voting, but the somewhat cryptic nature of the discussion of topics and lack of a definition for how to be a good voter made this a more difficult and time-consuming ordeal than I feel it needs to be. I just feel if it's not supposed to be accessible for people not active in the arbitration committee discourse then I shouldn't have been notified, and if it is meant to be accessible to everyone then some more standardized questions or other means of evaluations of the candidates would help give a more consistent perspective on the candidates. I felt like I mostly had to just discount some of the candidates because I couldn't learn enough about them in a reasonable amount of time and that there should be something resembling standard, broad issues for the candidates to come down on in discussion or questions that all of the candidates can respond to, instead of questions only from people already informed enough to ask. Penitence (talk) 06:40, 3 December 2017 (UTC)

I agree completely, you have probably next-to-no-chance to assimilate the information you need to make a considered vote, especially as we're now seeing WMF push out a "please, we don't have any votes" message mid-way through the election rather than three weeks ahead of it. I'm sorry to read about your experience, hopefully we could make it better in the future, but that depends on transparent voting mechanisms, technical competence and project management skills, none of which appear to have been demonstrated in this "election". The Rambling Man (talk) 21:39, 3 December 2017 (UTC)

Just one point: In former years, all candidates had to answer the same questions plus individual ones. That was a load on the candidates. I asked all candidates the same question. --Gerda Arendt (talk) 22:05, 3 December 2017 (UTC)

In the pre-web epoch, judging politicians' objectively documented past behaviour in decision-making and facilitation, rather than making guesses based on their promises, their political party affiliations, their charisma and their propaganda/advertising required much more work than most individual voters had the time to invest in. Today, for serious voters with some minimum verbal and web skills, the task of making evidence-based voting decisions is a huge amount easier. Getting to Wikipedia ArbCom, while it's good to see what the candidates claim they intend to do (or have done), we voters have an extraordinarily rich and efficiently consultable public database of the candidates' past behaviour. Below each candidate's statement I see a list of links: (talk · contribs · count · logs · target logs · block log · lu · rfas · rfb · arb · rfc · lta · spi · checkuser · socks · rights · blocks · protects · deletions · moves). Is X experienced, competent, neutral and effective in conflict resolution? Was X patient in handling disputes? Did X calm down huffed tempers? Did X let him/herself be manipulated by a certain Wikipedian? Did X help mediate by helping different people see each others' arguments? Has X him/herself shown serious commitment to developing quality content on Wikipedia and/or other Wikimedia projects? Editing a text file locally and making your own private analysis of candidate X, Y or Z by copying/pasting bits you find relevant is incredibly straightforward, though it does require some work. Keep your analysis private - just for making your vote - and nobody will be able to claim that you judged X unfairly. In any case, I do see rather lots of questions on the Questions page... But again, what the candidates have done on Wikipedia will tend to provide a good predictor of what they will do if elected. Boud (talk) 23:35, 3 December 2017 (UTC)

Thank you all for the questions and especially to Boud for the thoughtful reply. GeeBee60 (talk) 01:44, 4 December 2017 (UTC)

Right, Boud, I went through the candidates I couldn't eliminate quickly's user pages and histories etc. it just took a long time and it's not necessarily accessible to just ask someone to go through archives with thousands of entries and read all of them. If you're not steeped enough in the culture of Wikipedia user interactions it may not be very meaningful. I would have been able to discount a couple candidates earlier, perhaps, if they did go into their intentions, goals, ideals, etc. and then not only could I potentially have to be going through less logs, I might also have specific things I'm looking for to see if they are consistent with what they advertise themselves as. I would argue to the average user those mediums are not efficiently accessible, either. I refer back to my issue that if they want votes from everyone they've pushed out that message to, then whatever threshold whether or not you are invited to vote should be the baseline for how experienced you can expect that user to be. It personally took me several hours to go through all the info until I felt like I could cast a comfortably informed vote, and I certainly did not feel like I was welcomed with guidelines appropriate for someone of my experience as compared to someone with tens of thousands of contributions. If there are people that can easily access that qualifying information about the candidates via those tools and quickly identify what sort of things to look for, it would be nice if one such user could create a comprehensive guide on what to look for, how to look for it, and why that information is meaningful and relevant to candidacy for ArbCom in terms of predefined rules for Wikipedia in general and any ArbCom policies. Penitence (talk) 04:31, 6 December 2017 (UTC)

All of that is true, but I think it is important to note how significant even having the bot message go out is: ArbCom is pretty much the most inside baseball thing there is on Wikipedia (the only other thing that comes close is [[WP:RFA|requests for adminship). Even having the bot go out so people who aren't obsessed with the project space politics would be aware that their voice maters was a big deal (see this discussion for how technically difficult it was this year.) I think we should do better in making this more accessible, but also think that we should take credit that we had volunteers work so hard to try to recruit more voters. I think thats a major positive, and while work can still be done, we should always look at the good things here :) TonyBallioni (talk) 04:39, 6 December 2017 (UTC)

Alternative system proposal

Hi,

Maybe it might seem like obvious for others, but is there a page explaining why (support)/(support+oppose) was decided? And how different is a neutral vote compared to a vote being rejected in a real election? tks Yaḥyā ‎ (talk) 03:00, 4 December 2017 (UTC)

In the case of a real election, votes are rejected when it is considered non-compliant (under strict guidelines), not because they were neutral. Here in this case, users have really the choice to vote neutral. I am not aware of the electronic ballot system or if there is a neutral choice. Some could see as arbitrary to promote 80% support over 100% neutral.

Besides democracy isn't the only political system worldwide... some will call this cultural bias.

Given that Wikipedia system should be representing the World, the only sustainable way is to form a wholocracy!

• 1) The most supports --- Promoted... isn't allowed to see (the identity of members involved in cases)... because reputation (image) got them in power.

remains innocent, because they couldn't see who was involved.

• 2) The most opposes --- Promoted... isn't allowed to speak (give his own opinions)... because most don't want those opinions.

remains an idiot, because they couldn't say who did wrong.

• 3) The most neutrals --- Promoted... isn't allowed to hear (others opinions)... because those are the ones it's more likely have no opinions.

remains ignorant, because they couldn't hear their (those involved in the cases) defense.

Under Western juridical system, the three would get away and stay in power for all eternity.

• Three_wise_monkeys are only wise when they're together. :b Yaḥyā ‎ (talk) 17:44, 4 December 2017 (UTC)

Any answers?: More evidence that it’s cultural bias, see the first argument of Avicenna (Muslim philosopher) [1]: Appearance says little about substance. Under Islamic culture (reason why democracy has problems in those countries) Zahir (appearance, determine who is popular) can not have authority over Batin (substance) [2]. If there is a place where every cultures should be fairly represented, by the exchange of knowledge without cultural bias, it’s here. What I have proposed is the only system which was consistently proposed irregardless of cultures. Yaḥyā ‎ (talk) 21:13, 5 December 2017 (UTC)

Just a comment, I don't know if you'll find it helpful.

When I first saw that candidates are ordered by support/(support+oppose), I thought "that's weird! It would make more sense to use support/oppose". Then I realised that the two methods must always yield the same ordering. And both have the snag that all candidates with some supports and no opposes score the same, as do all candidates with no supports and some opposes. Maproom (talk) 21:55, 5 December 2017 (UTC)

support/(support+oppose) treats the sample as if support+oppose = total votes. support/oppose wouldn't be much different, because there is no place for neutrals in the formula. Compare that in sports or lets take chess. What would you think if Chess candidates weren't given their 1/2 in a competition for the draws? But this still will be besides the point, because even if neutrals were to be added... choosing solely on positive popularity is cultural bias. Yaḥyā ‎ (talk) 22:08, 5 December 2017 (UTC)

While I think this is an approach, it’s not the only. Why the most support to begin with (irregardless of the way they’re computed) should be promoted? That’s the relevant question, because it’s a cultural bias. The only non-arbitrary things such elections can tell us is who got most support, who got most oppose and who got most neutral. If I want a fair representation of seats, I’ll take portions of each. The only thing most support (again irregardless of the way they’re computed) tells us is that they’re positively the most popular. What about negative populars or neutral populars…? See in a card deck, there are four kings (of hearts, of tiles, of clovers and of pikes)… in what I proposed above, Shizaru [3] the Clovers kings (those of professions) are missing… that’s because those have hands tied (like Shizaru who might end up like a NAZI doctor if hands weren't tied), they should only be consultants. Card decks have a long history, irregardless of political systems, less arbitrary than the current system.  :) Yaḥyā ‎ (talk) 19:17, 5 December 2017 (UTC)

Proving positive popularity argument is arbitrary

• Be three set of candidates

• Present
  • C1: Got more support.
  • C2: Got more oppose.
  • C3: Got more neutral.

• In a hundred year
  • C1': Got more support.
  • C2': Got more oppose.
  • C3': Got more neutral.

• Since opinions about members rely on social norms and subjected to cognitive bias, they change in time.
  • Which means that the exact same candidate might be promoted now, but not in a hundred year.

• Only thing which remains stable:
  • Each being the extremes, we could say that:
    • C1 and C1' have set of behaviors which are considered as positive in their given era.
    • C2 and C2' have set of behaviors which are considered as negative in their given era.
    • C3 and C3' have set of behaviors which are considered as neutral in their given era.
      • Ct : All human behaviors (irregardless of if they're considered as positive or not)
        • C1 + C2 + C3 = Ct
        • C1' + C2' + C3' = Ct
          • C1 + C2 + C3 = C1' + C2' + C3'

• In conclusion: All set of human behaviors should be included in those promoted, because that's the only thing which is constant in time. So wiki positive popularity is an arbitrary way to decide who gets promoted. Yaḥyā ‎ (talk) 16:55, 6 December 2017 (UTC)

• Could you please take this somewhere else? --Floquenbeam (talk) 17:01, 6 December 2017 (UTC)
  • Hi, do you have somewhere to propose? tks. Yaḥyā ‎ (talk) 17:12, 6 December 2017 (UTC)

A subpage of your user space. --Floquenbeam (talk) 17:14, 6 December 2017 (UTC)

Do you mean that there is no formal place to discuss the way admins, arbs etc... are chosen? If not, do you have an idea how such a formal place could be created? tks Yaḥyā ‎ (talk) 17:18, 6 December 2017 (UTC)

The way arbitrators are chosen is decided every year in an RFC in September. However, based on what you're saying here, the most appropriate place to post it would actually be user talk:Jimbo Wales. --Floquenbeam (talk) 17:31, 6 December 2017 (UTC)

Oh, I didn't know about the RFC sorry... with all those rules it's hard to know the wheres, hows and whens... I had previously renamed the section as suggested and moved a reply back here, to comply with others wish. I won't oppose if you move this elsewhere in my userpage if you find it more appropriate. tks for the reply. Yaḥyā ‎ (talk) 17:39, 6 December 2017 (UTC)

Disqualification

@DoRD: can you please modify your notice about A Den Jentyl Ettien Avel Dysklyver being disqualified to include a brief explanation of the reason? Yes, eventually users can find Wikipedia:Administrators' noticeboard/Archive294#Mentoring and removal of permissions needed, but even then it takes a while to find/understand the exact reason(s). - dcljr (talk) 03:49, 4 December 2017 (UTC)

Actually, let me rephrase that: Could you add a quick explanation that the user was disqualified after getting indefinitely blocked (linking that phrase to the AN archive page, as done here). Although the exact reasons for the block itself are not really relevant for explaining the disqualification, the link would be nice for users who want to know the details. - dcljr (talk) 04:14, 4 December 2017 (UTC)

That's a sensible suggestion. Done. ​—DoRD (talk)​ 11:15, 4 December 2017 (UTC)

algorithmic voting

I am too lazy to go through all these candidates' edits, etc. Surely by now there ought to be some algorithms written (Watson, for example?) that can do this and give a summary of what each candidate is like. I would be happy to vote based on an algorithmic assessment of each candidate. Has this been looked into yet please? --Rebroad (talk) 04:53, 4 December 2017 (UTC)

"Neutral vote does not affect the outcome" false claim (specific counterexample)

Thanks for working to improve the instructions.
Counterexample, showing a "neutral" vote affects the outcome:
Candidates A and B start with equal probability of winning.
Voter #1 supports both A and B (a non-decisive ballot).
Voter #2 opposes both A and B (a non-decisive ballot).
Voter #3 has a neutral vote for A or supports A, and opposes B.
A wins.
CONTRADICTION: We cannot say a supporting vote can have an impact on the outcome of the election but a neutral vote cannot, when they produce the same election outcome. (Note also Voter #3's opposition to B cannot affect the outcome independently of the vote for A, because the vote for A must be either neutral or support. Therefore, we cannot say that the opposition to B affects the outcome but the neutral vote does not.)
Filingpro (talk) 16:48, 5 December 2017 (UTC)

Thank you Yaḥyā ‎ for the comment and I agree the current voting method is not multi-winner proportional. Because this section concerns fixing the instructions of the existing system, can you please consider moving your comment to a new section about an alternative system proposal, so others may respond? Thank you Filingpro (talk) 22:48, 5 December 2017 (UTC)

Moved, misunderstood your intention: But I disagree with you in a technical sense, does not affect as in support/(support+oppose)... obviously not supporting someone would affect the results... not because neutral was voted. — Preceding unsigned comment added by Yahya Talatin (talk • contribs) 23:07, 5 December 2017 (UTC)

Thanks. The counterexample shows that for Voter #3 voting either "neutral" or "support" for A has the same results, not that "not supporting someone would affect the results." Do you see? Filingpro (talk) 23:45, 5 December 2017 (UTC)

I am perfectly seeing where you are heading. I don't know his background or yours, but I'd assume his is science related. They will check directly the formula,... every relationships between neutrals, supports and opposes are treated as confounding factors, correlations etc... but those are considered as indirect. (while some papers might disagree, but those usually are considered by science POV as fringe) Means that if in the formulas that variable is not directly there... it's usually indirect (I'm not involving parametric equations here). That's one of the reasons why Clovers kings should act only as consultants (see my comment here [4]), because their languages might not represent the majority. I proposed directly to be added... will see what others say. — Preceding unsigned comment added by Yahya Talatin (talk • contribs) 00:04, 6 December 2017 (UTC)

Please see "(re mathematical approach)" reply to your post below. Thanks Filingpro (talk) 06:56, 6 December 2017 (UTC)

The instructions are intended to communicate that voting neutral does not affect the outcome compared to not voting. Of course, withholding support does affect the outcome (which is what your counter-example demonstrates). ~ Rob₁₃^Talk 22:54, 5 December 2017 (UTC)

Can you explain what you mean by "voting neutral does not affect the outcome compared to not voting"? If we wanted to make a true statement, we could say, "voting neutral for every candidate has the same effect on the outcome as not submitting a ballot." But that is a categorically different statement than the false claim made in the instructions. Are you agreeing to the falseness of the statement?Filingpro (talk) 23:18, 5 December 2017 (UTC)

What do you mean by "withholding support"? The example shows specifically that Voter #3's neutral vote for A affects the outcome (the same way a supporting vote affects the outcome! Filingpro (talk)). Filingpro (talk) 23:18, 5 December 2017 (UTC)

He’s taking it from a mathematical point of view. The formula! One got an oppose the other didn’t… (without involving neutral), Maybe Neutral vote does not directly affect the outcome would be better. — Preceding unsigned comment added by Yahya Talatin (talk • contribs) 23:38, 5 December 2017 (UTC)

Yes! So a correct mathematical statement we could make is: "A neutral vote for a candidate does not affect that candidate's absolute score." (a categorically different statement!) HOWEVER, this would be irrelevant and highly misleading to the voter because candidates are elected at the exclusion of others based on their relative scores. Imagine an approval voting system where voters are instructed to mark for each candidate "Yes" or "No", and the FORMULA used to score candidates is -1 x "No"s (+ 0 x "Yes"s). By the proposed false "mathematical logic", we could claim "A Yes vote does not affect the outcome in any way." Or, "A Yes vote does not directly affect the outcome." Filingpro (talk) 00:25, 6 December 2017 (UTC)

I understand what you mean. But yet, mathematically the initial statement could be considered valid, irregardless of what it means for most. Because what it means is not necessarily what is written (see my userpage :) to see why). That's why users like you should be deciding the wording, and clover kings should only be consulted to see if it is compatible with all those standardized constructs their professions came up with. Because you see meanings they see constructs.

(re mathematical approach)

Briefly thanks for the vote of confidence that someone like me "should be deciding the wording," and I take as a friendly gesture you did not regard me as having a science background. Incidentally I do consider myself a scientist (and a humanist).

Yes I agree the initial statement is valid but only when using an incorrect mathematical model (see below), and so it is in fact unambiguously false. Firstly, if it were true, it would not be contradicted by the counterexample above (I believe has not been rebutted). In my view, a mathematician would be concerned with the contradiction. I believe the confusion arises when we wrongly assume a neutral vote cannot affect the outcome of the election merely because a neutral vote does not affect the absolute score of a single candidate, but this does not logically follow, and the reason may not be intuitive. As a start, if this logic were true, please consider the absurd results we see in the "Approval" voting example I provided above, i.e. we could claim "Approving a candidate does not affect the outcome of the election!". Again, in my view, a mathematician would be concerned with this contradiction. In my experience, the reason a "mathematical" argument can appear logical and be false, is because the incorrect mathematical model of the subject-matter is used. For example, to justify the current instructions as written, we must assume that the voter is voting for each candidate in an independent election, when in fact, candidates are elected at the exclusion of others according to their relative scores, axiomatically. Simply put: The reason a "neutral" vote affects the outcome is because a "neutral" vote causes a distinctly lower (or equal) score than "support" and causes a distinctly higher (or equal) score than "oppose". Because candidates are elected based on relative scores, a "neutral" vote necessarily can affect the outcome of the election. Filingpro (talk) 06:56, 6 December 2017 (UTC)

It's the oppose of the other candidate that affected the outcome, not the neutral. ~ Rob₁₃^Talk 02:27, 6 December 2017 (UTC)

Challenge Q for Rob₁₃: Then why does Voter #2's oppose not affect the outcome? Filingpro (talk) 03:37, 7 December 2017 (UTC)

They are not the median voter in your hypothetical two-candidate, one-seat example. ~ Rob₁₃^Talk 03:40, 7 December 2017 (UTC)

Why is Voter #3 the median voter and Voter #2 is not? Filingpro (talk) 03:54, 7 December 2017 (UTC)

Would Voter #3 be the median for any vote on A? Filingpro (talk) 04:02, 7 December 2017 (UTC)

As voter 1 and 2 are casting indifferent votes (votes for both A and B), voter 3 is effectively the only voter, and so also the median voter. ~ Rob₁₃^Talk 04:06, 7 December 2017 (UTC)

You said Voter 3 opposition to B affected the outcome, not the neutral vote for A, but now you say a differentiating vote for and A and B is necessary condition to affect the outcome - i.e you proved your first claim false. Do you see? Filingpro (talk) 04:17, 7 December 2017 (UTC)

Consider an alternative situation where voter 3 votes oppose for B and doesn't vote at all for A. The outcome is the same. That is what is meant by a neutral not affecting the result. ~ Rob₁₃^Talk 04:26, 7 December 2017 (UTC)

Challenge Q for Rob₁₃ Re: “alternative situation” What about the situation we were just discussing which led to a contradiction? A contradiction in one case proves the statement false “A neutral vote does not affect the outcome.” We cannot ignore a failure case by switching to a different case to make it go away. Filingpro (talk) 04:57, 7 December 2017 (UTC)

You can't oppose B and "not vote at all for A" (OR, how does a voter do that?) Filingpro (talk) 04:46, 7 December 2017 (UTC)

That's my point. A neutral is not voting. You're correct that voting neutral instead of support or oppose changes the result, but that's not because of the neutral; it's because of the lack of a support or oppose. It's equivalent to not voting. (The above scenario is not a contradiction, because it falsely states the neutral caused the result to change when it was the oppose that caused the result to change. If that voter had submitted a fully neutral ballot, results would stay the same. This is not true of support/oppose ballots in all cases, where one could push a candidate over the 50% line.) In any case, we should continue this on my talk if you'd like; this is more an academic discussion than anything else. ~ Rob₁₃^Talk 13:35, 7 December 2017 (UTC)

How can we say "a neutral is not voting" when "neutral" is one of three mutually exclusive choices on the ballot (each affecting a candidate's score differently!)? Filingpro (talk) 23:20, 7 December 2017 (UTC)

Hi Rob, could you review his proposal at the end of this section? Yaḥyā ‎ (talk) 02:24, 8 December 2017 (UTC)

You say "neutral instead of support or oppose changes the result" contradicts the instructions. (I understand your argument but failed to show you the problem). Yes academic but a false claim on wiki not good, and may disenfranchise voters. Will take to talk page per suggestion. (Apologies for addressing you specifically; I'm attempting to reach someone who is admin.) Filingpro (talk) 23:20, 7 December 2017 (UTC)

Clarification: Rob13 said my counterexample "falsely states the neutral caused the result to change when it was the oppose." To clarify, I did not say explicitly that the neutral caused the result to change and the oppose did not. I am saying it is false to claim the oppose changes the outcome and the neutral does not, because the neutral choice is a necessary condition, so that claim is a logical impossibility (stated in my opening post). Filingpro (talk) 17:08, 10 December 2017 (UTC)

Thanks Rob₁₃, I had also considered that, but we cannot assume Voter #3's opposition to B causes A to win independently of the vote for A. Do you see this is a false claim? (Notice that the vote for A cannot be any of three ratings: support, neutral, oppose.) Please see also "(re mathematical approach)" above for a more detailed discussion. Filingpro (talk) 07:05, 6 December 2017 (UTC)

Hi Filingpro, I'm answering here, because it's becoming harder to follow the line. Every discipline check things by starting from their object of interest (the model). Here it’s the formula and any following interpretations has to be within the specified parameters (here set by the formula)… you are opposing their deductive (object oriented, quantitative) language to yours which is inductive (subject oriented, qualitative). Deductive inevitably has to rely on some form of conventions and work from there.

So, deductive says:

• Formula says no neutral involved.
  • Interpretation has to follow formula.
    • Answer: One got oppose (without involving neutral).

That does not exclude some form of other relationship between variables (or systematic bias), but this is mostly treated independently or the formula is falsified (but that’s another issue). I am not agreeing nor opposing either ways of seeing things (both are important). I shouldn’t have assumed your background from your post. I apologize if that might have offended you.

What about you propose a wording and see if it's compatible with them? Yaḥyā ‎ (talk) 14:35, 6 December 2017 (UTC)

1. No offense taken.
   
2. The formula to determine the absolute score of a candidate is not a mathematical model of the election. The election is a function that receives as input ballots with 3^C possible unique ballot markings, where C is the number of candidates, and outputs an ordered list of winners.
   
3. The neutral vote (for A) is a necessary condition for the oppose vote (for B) to change the outcome of the election. Therefore we cannot say the neutral vote is not "involved" because it is a necessary condition in affecting the outcome of the election.

Filingpro (talk) 16:44, 6 December 2017 (UTC)

But the wording is merely based on the formula. I am not saying that I agree with it. See I can not answer your reply unless you propose a wording... because if you come up with a better wording, then this discussion is pointless. :)

Instructions

• Use the radio buttons. For each candidate indicate whether you support, oppose, or are neutral to that candidate. See below for how ballots are counted.
• ...
• ...
• ...
• Submit your ballot. Click the "Submit ballot" button to cast your vote.
• ...

How ballots are counted
Candidates are scored by the formula support_total / (support_total + oppose_total). [# open seats] candidates with highest scores are elected.

I would WARN: composing new wording may be ineffectual if the wiki board does not understand the mathematical error with the existing instructions! Filingpro (talk) 22:30, 6 December 2017 (UTC)

Alternate instructions (see also above and consider combining language):
For each candidate indicate whether you support, oppose, or are neutral to that candidate. Candidates with the highest percentage of support are elected. Neutral votes are not included when determining a candidate’s percentage of support.
Filingpro (talk) 07:22, 7 December 2017 (UTC)

Finally you came up with one. :) — Preceding unsigned comment added by Yahya Talatin (talk • contribs) 16:37, 7 December 2017 (UTC)

Rob13 closed talk so my final comments here for future elections (I did not intend to change ballot instruction mid-election).

1. "Not voting": The problem with saying voting neutral for a specific candidate is "not voting" while simultaneously supporting another candidate is voting, is that there are not two separate elections, so there is no basis to the claim. There is one election. Voting is casting a ballot. Not voting is not casting a ballot. I see no need for philosophical discussions about whether "not voting" changes the outcome.
   
2. Voter's perspective: From the voter's perspective, the ballot has 3^C possible markings (the input) each of which can change the outcome (the output). The instructions are a statement to the voter about how the choice of a neutral vote affects the outcome, which it does. A voter can support a preferred candidate and cause that candidate to lose by choosing to vote neutral for a lesser preferred candidate, rather than oppose.
3. Participation failures are red herring: The failure of ordinal participation criterion by voting all-support or all-oppose is indeed a negative result, but unrelated. (Otherwise none of the columns in range voting 1-10 would "affect the outcome" because they satisfy participation).
4. Remedy: We can inform the voter by making a true statement, e.g. "Candidates with highest % support are elected. A supporting vote increases a candidate's % support, while opposing decreases. A neutral vote neither increases or decreases a candidate's % support." (It can be debated how to word this.) Filingpro (talk) 00:42, 12 December 2017 (UTC)

Candidates deleting questions

@Ritchie333:, @Yunshui:, @DoRD:: I am just wondering if there is any protocol when candidates running for ARBCOM election delete questions without justification, as User:Sir Joseph has here. He has been challenged on this decision both on his own talk page and here, both by myself and by other editors, and yet has not taken action or responded to these claims. We are now in the run-in to the election, and I feel that it is unfair for a prospective candidate to bury a legitimate answer, and am growing increasingly tempted to revert the action as per WP:BOLD. However, this could constitute edit warring and a lack of civility on my behalf, which I am wary of. Would I be entitled to do so, or is there any official measures which you, as co-ordinators can deploy to resolve this with the necessary haste? I only ask, as all other avenues to resolve the problem have been exhausted as I am being blind-sided at this crucial juncture of the election. Regards, Stormy clouds (talk) 19:11, 5 December 2017 (UTC)

@Ritchie333:, @Yunshui:, @DoRD:: I have decided to reinstate the questions in accordance with WP:BOLD on the page listed. If any of you disagree with this position feel free to pull it again. Stormy clouds (talk) 21:33, 5 December 2017 (UTC)

Your comment on your election guide, IMO, crosses the line. I suggest you strike out or delete the comment. Sir Joseph ^(talk) 21:09, 7 December 2017 (UTC)

@Sir Joseph: - Which one specifically? Cite it and your wish is my command. Stormy clouds (talk) 21:31, 7 December 2017 (UTC)

Removed the likely suspect. If you have a problem with the new wording, say so and I will remedy it. I don't care about your faith - my issue lies with you utilising it as a baton to attack those who critique you. Stormy clouds (talk) 21:39, 7 December 2017 (UTC)

I don't shoot down criticism. Look at many user guides criticizing me. I do have a problem with comments that are antisemitic and with people attacking me for raising concerns, as you have now done. The initial comment was antisemitic and as I wrote then, the civil thing to do would have been to apologize and edit. It is also a bad thing to criticize people bringing good faith complaints, as I'm sure you read that academic study linked on my talk page. Sir Joseph ^(talk) 23:46, 7 December 2017 (UTC)

Switching the Support and Oppose around

Hi, For me the current vote layout is Oppose | Neutral | Support,
Shouldn't it be Support | Neutral | Oppose ?,
I nearly ended up Supporting those I oppose and Opposing those I support ....,
Not really a big deal but I believe the prev layouts was S|N|O so thought it might be worth a mention,
Thanks, –Davey2010^Talk 22:28, 6 December 2017 (UTC)

Yeah so did I..usually yeses/okays/supports are in the left and nos/cancels/opposes are in the right. Galobtter (pingó mió) 18:41, 10 December 2017 (UTC)

Here is a new voter guide for those who are interested.

I know it is a bit late in the game, but I wrote a voter guide for the election. Relatively inexperienced in this regard, but welcome all feedback or questioning regarding the guide.

Pinging the other editors who made other guides to prompt discussion and debate here. Stormy clouds (talk) 20:59, 7 December 2017 (UTC)

How long before we get the results?

Does anybody know how long it's likely to be before we get the results? (And please don't just say 'same as last year' - if I could remember how long that meant I wouldn't be asking the question). Tlhslobus (talk) 10:20, 11 December 2017 (UTC)

Likely couple of weeks.--Ymblanter (talk) 10:50, 11 December 2017 (UTC)

The names of the pages are the same each year, only the year date changes, so it's easy to find past years, see Wikipedia talk:Arbitration Committee Elections December 2016 where the same question was asked and answered. SilkTork (talk) 11:23, 11 December 2017 (UTC)

Thanks, but in the end I found it easier to look up the edit history of the results pages.

Last year results were announced at 22.41 on 15 December in this edit.

The timeline section says voting closed at 23:59, on Sunday 4 December, UTC)

If it takes exactly that long this year we will get the results at 22.41 on 21st December; if it takes much longer maybe scrutineers will have to take Christmas and New Year breaks as well.

But on the other hand 2015 results were posted at 19.10 on 9 December 2015, less than 3 days after voting closed at 23:59, on Sunday 6 December. If it takes exactly that long this year we'll get them at 19.10 tomorrow (Wednesday 13 December 2017).

Maybe we need a few bookies so we can bet on when they'll be out :) Tlhslobus (talk) 14:21, 12 December 2017 (UTC)

Meanwhile I guess I'll just have to put the results page on my watchlist. Tlhslobus (talk) 14:32, 12 December 2017 (UTC)

It obviously depends, among other factors, on the number of voters.--Ymblanter (talk) 14:43, 12 December 2017 (UTC)

Maybe, but I doubt if there were 4 times as many voters in 2016 as in 2015. I think there was some new problem in 2016. Hopefully that's now fixed. Tlhslobus (talk) 15:41, 12 December 2017 (UTC)

But if the voters are increasing fourfold every year we won't get the results until late January.Tlhslobus (talk) 15:44, 12 December 2017 (UTC)

I hereby propose the cabal institutes exit polling to allow us modern election style coverage the minute the polls close. Regards SoWhy 15:52, 12 December 2017 (UTC)

• For the younger pups around here, DYK that in the good old days, ArbCom voting was public (like RFA)? Results were known pretty much the instant the voting was over, except for the "will Jimbo ignore the voters!?!" rumors that swirled until he validated the results. --Floquenbeam (talk) 16:02, 12 December 2017 (UTC)

Funny, I would have considered you a younger pup as well. Then I considered that 2008 was nine years ago and felt really really old. Regards SoWhy 16:38, 12 December 2017 (UTC)

First, that's the nicest thing anyone has said to me this week. Second, I was around a while before 2008 too. --Floquenbeam (talk) 16:40, 12 December 2017 (UTC)

You've both been here longer than me (ha ha), but I remember that in one of the first elections after Secure Poll was started, they briefly published the results upside-down (ie, the candidates with the worst results coming in first, etc), with predictable brief drama. Sometimes, waiting ain't so bad. --Tryptofish (talk) 16:43, 12 December 2017 (UTC)

I have a feeling that would cause a riot this time around. It would be somewhat interesting to discuss whether scrutineering is necessary for a non-close result next year. For instance, 2016 had a difference of eight percentage points between making the Committee and not. At the end of the day, after ten days of scrutineering, the scrutineers struck 8 votes. Perhaps it makes sense to skip the step if and only if there is a gap larger than 5%? Food for thought ten months from now. ~ Rob₁₃^Talk 17:27, 12 December 2017 (UTC)

That doesn't strike me as a good idea. I would rather have legitimate results than faster results, regardless of how likely it is that there will be a difference between the two. Mz7 (talk) 07:18, 14 December 2017 (UTC)

I am with you Rob. In every real life election that I have encountered, the results have almost always been made available within 24 hours after the polls have closed. At the end of the day, nobody is going to remember (or care) if somebody got elected with 61% or 60% of the vote. The important fact is that one got elected. -- Dolotta (talk) 13:52, 19 December 2017 (UTC)

Well in this case 3% would've been enough to change the results..there was a small chance (but not non-existent) that the results could change. Galobtter (pingó mió) 13:59, 19 December 2017 (UTC)

The scrutineers struck a single vote, as far as I can tell, not counting the struck vote resulting from testing the election interface before the election opened. ~ Rob₁₃^Talk 14:15, 19 December 2017 (UTC)

I'm looking at a more interesting result right now.... Ritchie333 ^{(talk) (cont)} 19:01, 12 December 2017 (UTC)

Yeah... there's an unbiased piece of writing. Fyunck(click) (talk) 06:46, 13 December 2017 (UTC)

As for the election returns there, whew, I feel better now! --Tryptofish (talk) 14:45, 13 December 2017 (UTC)

When 3 people are beating you with sticks, it feels slightly better when one stops, but it still sucks to get beaten by 2 people with sticks. --Floquenbeam (talk) 15:48, 13 December 2017 (UTC)

Darn those sticks. Dave Dial (talk) 23:05, 15 December 2017 (UTC)

Publish preliminary and/or interim results?

Does anybody think it would be a good idea if preliminary unscrutineered results were published as soon as they were available, in the interests of both transparency and of not keeping us all unnecessarily on tenterhooks (I'd like to know how some of my favorite and least favorite candidates are likely to do), with it being clearly stated that they were provisional results and that we would still have to wait for the official scrutineered results? Tlhslobus (talk) 14:01, 16 December 2017 (UTC)

Periodically publishing Interim scrutineered results (perhaps at the end of each day) might also help.Tlhslobus (talk) 14:02, 16 December 2017 (UTC)

I'm afraid I think this would be a very bad idea. It would merely cause all sorts of opportunity for drama, with people speculating or insisting what elements of the results would change after scrutineering, and then upset with whatever changes (if any) did take place as a result of scrutineering. And the provisional winners and losers would be in a very strange limbo, sorta (non)elected but not clear. There really is no urgency, whether it takes hours or days or weeks, just wait for final results! Martinp (talk) 16:08, 16 December 2017 (UTC)

Glad Martinp types faster, because now all I have to type is "Exactly what Martinp said". --Floquenbeam (talk) 16:13, 16 December 2017 (UTC)

Gosh, I guess it's good to learn that Wikipedia is, and will forever remain, an organisation so self-evidently perfect in every respect that it has no need of transparency. Such a pity that the rest of humankind isn't as pure and incorruptible as us... Tlhslobus (talk) 16:32, 16 December 2017 (UTC)

And re limbo, all the candidates are in limbo now, instead of only those few who are borderline winners or borderline losers. And the same goes for any voters who are hoping for or dreading some particular result. Tlhslobus (talk) 16:40, 16 December 2017 (UTC)

I can't escape the suspicion that "transparency" is used here to actually mean "impatient curiosity". There were no such objections raised to this form of voting during the election planning process. I can recall only one significant meatspace election where the counting of individual ballots and subsequent inspection of them took place under public scrutiny (the recounts in Bush v. Gore in Florida, of course). Those "vote counts" you see on election night on TV or the 'net aren't even actual vote counts but estimates/projections based on limited reporting results. The ability of the scrutineers to change the results outside of your personal inspection is, at best, extremely limited and subject to mutual review and disclosure. If you have that little faith in your fellow editors, then become a scrutineer yourself next time around. Eggishorn (talk) (contrib) 17:03, 16 December 2017 (UTC)

That's actually not possible, since scrutineers may not have enwiki as their home wiki to prevent any appearance of bias. ~ Rob₁₃^Talk 17:07, 16 December 2017 (UTC)

Thank you for the correction, but I think the essence of the point remains. Eggishorn (talk) (contrib) 18:42, 16 December 2017 (UTC)

• What? Can't possible see how that would help anything. Just cause unneeded drama. Dave Dial (talk) 17:38, 16 December 2017 (UTC)
• No need for this. TonyBallioni (talk) 18:44, 16 December 2017 (UTC)
• Incoming Arbcom members don't take up their positions till Jan 1, so it doesn't really matter on what day in December the results are issued. -- Euryalus (talk) 18:56, 16 December 2017 (UTC)
  • I'd say it probably matters that it's done before Christmas, because there's some back-end stuff that needs doing to add new arbitrators to mailing lists and get them caught up on what's going on behind-the-scenes. Having said that, the scrutineering will almost certainly be done by then. (We could give some thought to holding the election a week earlier next year, which I think would be a positive change.) ~ Rob₁₃^Talk 19:30, 16 December 2017 (UTC)
    • @BU Rob13: I think it was delayed a week by the WMF being committed to something else. Doug Weller talk 19:35, 16 December 2017 (UTC)
      • Ah, that explains it. ~ Rob₁₃^Talk 19:38, 16 December 2017 (UTC)

        Yes, they were running a similar secure poll on another project, and we had to shift elections by a week.--Ymblanter (talk) 19:41, 16 December 2017 (UTC)

• As far as I understand it, the actual vote tally isn't available on a preliminary basis because it is encrypted. Only after the scrutineers have completed their review, and the WMF decrypts it, is the tally available for dissemination. ​—DoRD (talk)​ 20:38, 16 December 2017 (UTC)
• It's impossible per DoRD above, and even if it were possible there's just no possible good reasons to do it. If the votes are encrypted as they're supposed to be, then once the tally is ran, then the election is over and no more changes are possible. You don't want election admins (scrutineers) to run (temproary) results during the election and possibly affect it based on whether they like how it's going. You also definitely don't want a decision on whether to strike any particlar vote to be based on the result. Just imagine the scenario - run the result, wait my favourite candidate just missed out by a few votes, let's strike these votes and see if we can get the result we want. -- KTC (talk) 21:19, 16 December 2017 (UTC)
• Hey all. Just want to chime in from Foundation-side. As others have correctly pointed out, the votes are encrypted, and will not be decrypted until the stewards are finished scrutinising them. For the reasons KTC points out above, decrypting them early is not a good idea. I'm sorry there is a delay in getting the votes tallied, but unfortunately this delay will not be improved by publishing interim results. Joe Sutherland (WMF) (talk) 02:48, 17 December 2017 (UTC)

Thanks, DoRD, KTC, and Joe Sutherland (WMF), you have convinced me. Tlhslobus (talk) 08:50, 17 December 2017 (UTC)

• I would favour immediate results without the need for scrutineers, or even a return to open voting. This secret voting and intense scrutineering heightens the view that this election is more important than RfA, and that ArbCom is an elevated status. The Committee acts as a group, so lessening the impact of any rogue individual. The most evil the Committee can do is ban someone from the site without the ability of that decision being overturned, except by another Committee. If the Committee decided to ban someone that the community widely felt should not be banned (not the usual wailing from the friends of the banned individual, but a community wide consensus that the Committee had done a bad thing), then protests could result in either the ban being overturned or the Committee being overturned or both. The Committee operates entirely by the consent of the community. We are not electing rule makers or breakers here. We are electing from the community some individuals to make awkward decisions regarding conduct of some users. That is it. Unless we plan to introduce scrutineering to 'Crat and Admin elections (both of which carry more potential for harm in the wrong hands), then we should do away with it here. Not because of the extra time it takes, but because it makes this election seem more important than 'Crat and Admin elections. SilkTork (talk) 08:41, 17 December 2017 (UTC)

the admins may have the potential for direct damage, but Arb Com has the responsibility for supervising the admins, and thus the greater importance on absolute neutrality in the election. Considering the utter disgrace some RfA "discussions" have been, there is also a place for a more objective process. WP goes generally by consensus, which is a great idea that has served us well, but does have some limitations, particularly in connection with log-rolling; the plain election of arbs (and the policy by which they make their findings by a plain vote rather than rough consensus) is aa necessary corrective. DGG ( talk ) 09:56, 17 December 2017 (UTC)

ArbCom does not have "responsibility for supervising the admins", and there would be a howl of protest if that were so. The community holds that responsibility. We call out admins we feel are behaving inappropriately - we can do this individually with a comment to the admin, or at a group discussion venue; if the community feels the situation is serious enough (such as where the admin is denying inappropriate behaviour), we call in the Committee to examine evidence impartially and come to a conclusion. ArbCom are not Judge Dredd who can arrest, try, and execute, except in emergency situations where an admin is going wild, and even there all ArbCom can do is report the admin to a 'Crat or Steward for desysopping. We like to have roles separated. SilkTork (talk) 10:31, 17 December 2017 (UTC)

• I must say I would favour the announcement of preliminary results after the election, but certainly not during voting. I don't find the arguments against above very convincing at all (no doubt the technical problems are sortable). This is after all what the mature Anglophone democracies all (afaik) do in political elections. It seems the scrutiny process typically changes only a handful of votes, and would only affect 2-3? candidates. Do we know if any past results were actually changed by the scrutiny process? Apart from satisfying the idle or not so idle curiosity of the wider community, it seems tough on the candidates. The winners have only the holiday season to prepare themselves before taking up the workload on 1 Jan, which doesn't seem sensible. KTC's main argument above shows little trust in the integrity of the scrutineers. Johnbod (talk) 14:36, 17 December 2017 (UTC)
  • @Johnbod: The biggest problem identified above is that, if the results were decrypted and made available to the community in a preliminary sense, the exact manner in which each individual voted would be seen by the scrutineers while they were scrutinizing. This leads to a potential for bias (or at least the appearance, since the scrutineers are trusted stewards). That can't be gotten around; the results are either encrypted or not. ~ Rob₁₃^Talk 14:57, 17 December 2017 (UTC)
    • Without knowing anything about the process, this doesn't immediately make sense. Why do they need more than the user ID? I don't see a special risk in them seeing "the exact manner in which each individual voted", either in terms of actual or perceived bias being deployed. Johnbod (talk) 15:03, 17 December 2017 (UTC)
  • "taking up the workload on 1 Jan". Committee members decide themselves how much they take on. Existing cases continue to be worked by the existing Committee members. If there's a case request live, then the new members have the choice of making a comment or not. No new members would be expected to fling themselves into an ongoing discussion. The amount of work put in is entirely down to individuals. While I was on the Committee there were some real workhorses who got involved with everything, and there were others who wafted in and out now and again to pass a comment. So the time of year is not as relevant as the character and personalty of the Committee member in determining work load. SilkTork (talk) 15:11, 17 December 2017 (UTC)
  • It doesn't matter whether one trust the scrutineers or not. You don't design an election process where that kind of abuse is easily possible, precisely in case your trust turns out to be misplaced. Why would one invite even perceived or potential if not actual bias, all for the sake of satisfying idle curiosity. And technically, it's simply not possible for this election. Editors are of course free to propose doing things differently for next year as SilkTork is above. -- KTC (talk) 18:56, 17 December 2017 (UTC)
• Publishing unverified and quite possibly wrong results would be a transparency problem not solution; for all anyone knows some answer or dispute toward of the end of voting strongly skewed some last-minute voting; if the scrutineers reported the first few days results it might signal a land-slide victory for someone who later turned out to not be elected at all, and this "success" by the candidate might be widely reported in the interim.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  15:24, 17 December 2017 (UTC)
• No, here be dragons. Just yet another excuse to whip up drama, scandal, retribution, hate-speak, and all the things we should be avoiding as we build Wikipedia's encyclopedic content. Who actually cares what the "preliminary results" are? If you want an exit poll, just make one up based on the voter guides, but waiting for the final result shouldn't tax too many people around here, certainly those who have followed recent Arbcom cases which take a month of Sundays to reach even a starting point, let alone a conclusion. Patience all. Note: if the manner in which the elections are conducted is in question, that's a different topic altogether.... The Rambling Man (talk) 20:39, 17 December 2017 (UTC)
• I disagree with publishing anything that is not entirely accurate. However, how a supposedly modern, forward looking organisation can take so long to tally and verify no more than a couple of thousand votes is completely beyond comprehension. Third world countries can count tens of millions of election votes in a couple of days! Leaky Caldron 09:17, 18 December 2017 (UTC)
• Publishing unverified results would not be good. If people are interested in polling, there should be a clear separation. Independently-run optional exit polling would be an alternative. Mkdw ^talk 19:31, 18 December 2017 (UTC)
• Well we have managed to get the worst of both worlds! A long wait, and a slightly inaccurate first announcement (see below). Yet the world does not seem to have come to an end, as suggested by several above. We should take note of this for next year. Johnbod (talk) 02:55, 19 December 2017 (UTC)

How did we get from, "...all sorts of opportunity for drama..." to, "...the world ...come to an end..." in one jump? That's a pretty large leap in logic. Eggishorn (talk) (contrib) 03:02, 19 December 2017 (UTC)

Literal guy, huh? Don't forget the "scandal, retribution, hate-speak, and all the things we should be avoiding as we build Wikipedia's encyclopedic content...". And dragons. It seems from the section above the scrutiny process changed the results by one vote. Johnbod (talk) 15:20, 19 December 2017 (UTC)

• I also disagree with this. No one here has really provided any compelling reason for exactly why it is necessary to publish preliminary results. We get them eventually, and as far as transparency goes, not even the scrutineers know how each person voted until they are completely finished with their process, so as others have pointed out, we know the exact same information they do as far as the overall results go. I suspect simple patience is the solution – just as we've done so for the past 8 years. Mz7 (talk) 17:59, 19 December 2017 (UTC)

Copyediting for next year

The bit "Candidates may continue to answer questions until the end of the voting period (23:59 UTC, 10 December)." should just be deleted. It's entirely irrelevant if a candidate (or someone asking them something) posts to the questions page after the closure of the voting period. In my case, I had a socking anon post on my candidate discussion page after the close of voting, with a bunch of weird accusations, so of course I responded to them, and pointed to the WP:SPI case. This could just as easily have happened at the questions page instead of or in addition to the discussion page. If this language is to be retained, then those clerking the process would have to be really on top of policing the questions pages to revert anything added to them by anyone, and that's pointless maintenance that probably would not actually get done by anyone.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  15:21, 17 December 2017 (UTC)

I think it's meant to communicate that candidates aren't expected to just stop answering questions because voting has started. Of course, any editor can ask another editor a question anywhere on the project at any time. The questioned editor can answer it or not. ~ Rob₁₃^Talk 19:17, 17 December 2017 (UTC)

Perhaps reword, then, and leave off the closing date: Candidates may continue to answer questions after the voting period has begun.—Odysseus1479 04:16, 19 December 2017 (UTC)

Sensible. ~ Rob₁₃^Talk 10:10, 19 December 2017 (UTC)

Results

The results have been posted, and are just awaiting certification by the scrutineers. Pinging RadiX, Matiia and shanmugamp7 for their signatures. Yunshui ^雲_水 10:42, 18 December 2017 (UTC)

The currently-posted results say that Alex Shih's vote totals were 565/997/396, which only adds up to 1958. The vote totals for other candidates add to 1991. Can somebody check this? power~enwiki (π, ν) 21:45, 18 December 2017 (UTC)

That's... a good catch, and kind of disquieting. Why are scrutineers signing off on incorrect data? I note that if the missing 33 votes (1991-1958) were all supports, then the 60.16% figure would be correct (right now it is wrong). --Floquenbeam (talk) 21:52, 18 December 2017 (UTC)

@Yunshui, DoRD, Ritchie333, RadiX, Matiia, and Shanmugamp7: --Floquenbeam (talk) 21:54, 18 December 2017 (UTC)

DoRD has fixed it as expected. The "Net" vote total also needs to be updated. power~enwiki (π, ν) 22:06, 18 December 2017 (UTC)

(ec) I assume that it was a typo or transcription error. The correct number is indeed 598. I have also updated the net number. ​—DoRD (talk)​ 22:08, 18 December 2017 (UTC)

To be honest, I saw this morning that the candidates were posted in the correct order per the official list from the WMF, but did not look at every single number. All of the numbers now match the official results. I agree that the scrutineers should have verified them before signing off, though. ​—DoRD (talk)​ 22:20, 18 December 2017 (UTC)

In fairness, RadiX hasn't signed off on them.Tlhslobus (talk) 23:42, 18 December 2017 (UTC)

Done. Checked. RadiX∞ 00:15, 19 December 2017 (UTC)

Thanks for fixing it. I did check the numbers and made a correction but somehow overlooked this transcription error while comparing the numbers. --Shanmugamp7 ^(talk) 02:19, 19 December 2017 (UTC)

Apologies for any transcription errors; that's entirely on me and my fat-fingered manual transcription of the results. Yunshui ^雲_水 08:53, 19 December 2017 (UTC)