Wikipedia talk:Arbitration Committee Elections December 2011/Archive 3

Examine the ACE voter log!

Or, "wabbit season is now open"

In addition to the real time vote log via the securepoll interface, we have a bot populated voter log right here, which has the username of every user who has voted. We need help inspecting the log for any irregularities such as: someone voting on multiple accounts, a banned user voting on any account, victim of an account or browser hijacking registering a vote that the account owner didn't intend. Take a look at 2009's voter log for examples of what we're looking for. Thank you for all the help!

--Tznkai (talk) 21:50, 5 December 2011 (UTC)

Voter numbers and election publicity

Is anyone keeping track of whether the number of votes cast at this stage of the elections is more or less than at the same (or similar) stage last year or the year before (those were both SecurePoll elections, I think)? I'm asking because I made a comment here that indirectly asks that question. Also, is there a list of where and when notices have been left about this election? Such a 'publicity list' should be made each time, as trying to reconstruct such a list from digging through page histories is difficult. Carcharoth (talk) 14:13, 6 December 2011 (UTC)

With four days to go in a 14 day voting period, there have been 578 votes cast in this year's election.

With four days remaining in the 10-day voting period of 2010, there were 574 votes cast; final turnout 854.

With four days to go in the 14 day voting period in 2009, there were 770 votes cast; final turnout 994.

Imprecise comparisons, but perhaps useful.

See the coordinators' talkpage for discussions of notices; no publicity list has yet been compiled to my knowledge. Skomorokh 14:23, 6 December 2011 (UTC)

Thanks for the stats. Looking through the contributions of the co-ordinators should enable some list to be made at some point for this year. It was when I considered comparing that to where and when and how often there was publicity in previous years, that I realised this is only possible if a similar list of notices and their details was made in previous years. If the numbers end up being lower this year, that downward trend might be something to raise on the feedback page. What I was wondering is whether the previous elections had any last-minute publicity push to make sure no-one who would have wanted to vote was unaware of the elections, and whether that is happening this year or not? I have a vague memory of "last few days left to vote" notices being left last year, but might be mis-remembering that. Carcharoth (talk) 14:47, 6 December 2011 (UTC)

In 2010, 321 people had voted after 47 hours. I don't know how this compares to his year though. The figure comes from Wikipedia talk:Arbitration Committee Elections December 2010/Archive 3#Encouraging others to vote. That thread also has complains about lack of publicity. Thryduulf (talk) 17:22, 6 December 2011 (UTC)

In the end raw votes cast were 734 in contrast to 854 in 2010. That's 120 less and amounts to a 14 % decline, the same as between 2009 and 2010. --Tikiwont (talk) 21:26, 11 December 2011 (UTC)

Well, that pretty much rules out the voting period length as a major cause of declining participation. Jclemens (talk) 21:48, 11 December 2011 (UTC)

Would there be any privacy or other issues in comparing the voter logs for the past 2 years to make a list of those who voted in 2010 but not this year and determining which of them are still active and asking them for their reasons for not voting this year? Thryduulf (talk) 23:49, 11 December 2011 (UTC)

Curiosity caused me to scrape the data from Special:SecurePoll and manually extract the unique voters, with these results:

Voted	ACE 2010	ACE 2011
One year only	470	350
Both years	384	384
Total	854	734

Johnuniq (talk) 07:44, 13 December 2011 (UTC)

Although some of those 470 people who voted last year but not this year will no longer be active, I think it unlikely that all of them are. This suggests that we cannot rely on people simply remembering that it's a year since they last voted for arbcom to get them to vote this year. I'd even say it's rather worrying that 55% of last years voters didn't return this year. If my maths is correct (and it might not be) then if we lose 55% of this years total vote (330 voters) and attract 14% fewer voters in total (631 total voters) that would mean just 301 new electors next time. a very poor hit rate. If I'm reading the stats correctly, there were 85,875 new Wikipedians making 10 or more edits between October 2010 and September 2011. Obviously not all of them will have made it to the eligibility criteria for this years election, but if we assume 30% did (I haven't been able to find any stats on this, so it is an arbitrary figure) then that's a potential new electorate of over 25,000 (which seems very high, so I've probably made an error somewhere). Thryduulf (talk) 23:42, 14 December 2011 (UTC)

As I see it, there were 470 who didn't come back this year, but there were 350 who came new. That's a difference of only 120, and no reason to extrapolate that the trend will necessarily continue at the same rate (or continue at all) in the future. --Tryptofish (talk) 23:48, 14 December 2011 (UTC)

I agree Tryptofish, though comparison with other years is needed to see whether the figure of 120 is large or normal. I also think it would be legitimate to ask if the election co-ordinators would be prepared to contact still-active users who didn't vote this year and ask if they are willing to leave feedback on the feedback page as to why they didn't vote this year. The analysis could also be extended back a few years to get a fuller picture. There is a full record for every election of who voted. Carcharoth (talk) 23:58, 14 December 2011 (UTC)

Releasing results?

When will the preliminary results be released?

Last year, it seems that 2 of the 3 scrutineers had already certified results by now .... Should we stay up or reward our public with beauty sleep?

 Kiefer.Wolfowitz 00:24, 11 December 2011 (UTC)

The tally of the votes will be released once the scrutineers are satisfied everything is above board; that will take up to a week. Judging by previous years, no amount of comments, questions, requests for updates and so forth is likely to have any impact whatsoever on that timeline, so for all (both?) you election junkies out there, ample beauty sleep and/or article writing is encouraged. Skomorokh 00:33, 11 December 2011 (UTC)

Thanks, Skomorokh!

Last year, voting ended on the 5th, which made it easier to post results on the 10th or 11th.

Cheers,  Kiefer.Wolfowitz 00:38, 11 December 2011 (UTC)

You're most welcome. If you want to suggest an earlier scheduling (or any other changes for future elections), now is the perfect time to kick off a discussion on the feedback page. Skomorokh 00:46, 11 December 2011 (UTC)

Yeah, the scrutineers know that they can take as long as they judge necessary to ensure the election is honest and proper. This might be a day or it might be five: we just don't tell, and nor should we be able to. Tony (talk) 09:12, 11 December 2011 (UTC)

Thanks to the candidates

I forgot to vote. So I thought I would try to make up for it by thanking all the candidates. --FormerIP (talk) 00:44, 11 December 2011 (UTC)

ACE2011 denied banner advertising on the basis of a local interpretation of the legitimacy of the RfC?

Risker has raised an interesting point of which I was quite unaware:

"This year's RFC was considered to be so poorly participated in that the people who control the MediaWiki interface felt quite justified in refusing to allow a site banner advising of the election, for example; when those site-wide banners ran to all users, even for a few days, there was considerably more participation."

I believe it's of great concern that a group of people in another place made their own interpretation of the level of participation in the RfC before ACE2011, and dismissed the legitimacy of the RfC. Does this mean that they used their privileged position, exclusive of the community, to make an important decision to the disadvantage of the electoral process? Is the denial of site-wide banner advertising part of the reason the voter numbers are down this year? It seems extraordinary that the community was given one set of RfC closures by trusted admins in a very public place where we could all see the outcomes, yet behind the scenes other people used their power to negatively affect the community's awareness of a critical event for en.WP.

I'd be very pleased to be informed that I'm wrong. Tony (talk) 09:47, 11 December 2011 (UTC)

It's well known that I'm not a fan of our overlords, who are neither all knowing nor particularly benevolent. The problem is that they still are pretty all powerful, and we're pretty stuck with them. Unless everyone decides to leave at the same time and start a new, WMF free fork, we're going to have to live with them. It's not a major problem now, but every time the WMF decides to overrule the community, more and more people lose faith in the WMF. I lost faith in them a long time ago, I wonder how many more stupid mistakes they're going to make before more people become as bitter as I am. Sven Manguard Wha? 10:56, 11 December 2011 (UTC)

This has absolutely nothing to do with off-wiki machinations or the WMF; it was a decision made in public by local administrators. The relevant discussion is here. Skomorokh 13:35, 11 December 2011 (UTC)

Apologies. It would appear that I was pegging this particular violation on the wrong behind the scenes group. (My contempt for the WMF is not diminished though). Sven Manguard Wha? 14:39, 11 December 2011 (UTC)

As Skomorokh said, it was not "behind the scenes", but an open discussion held in public. All the above really seems to demonstrate is that if someone (you) are (self-admittedly) bitter about something, you will jump to the wrong conclusions. Though in mitigation, the wording "the people who control the MediaWiki interface" can be interpreted several ways. You seem to have assumed that refers to the WMF, when that part of the interface is actually something that can be modified by any admin (I think). Though people are rightly wary of making changes to a notice that is displayed to all users. Wheel-warring over that would likely and ironically have led to some sort of arbitration request. It is a rather esoteric location anyway. Look at the page history and try and work out how to make changes there. Compare it to the page history at the watchlist notice. Technical stuff like this tends to, by its nature, discourage random admins from doing stuff like this, similar to how only a few people know how to edit and use spam blacklists and whitelists. Trouble is, that can lead to only a few people actually doing anything in a particular area and effectively becoming both a single point of (potential) failure (similar to when a widely used bot goes down without a replacement) and a fiefdom. Carcharoth (talk) 03:28, 12 December 2011 (UTC)

I've started discussions in order to attract more eyes to the Sitenotice issue during the election: here, here, here, and here, but only a few people decided to participate in those discussions. There should've been more people complaining during the election rather than waiting until the election ended. The other problem is that the !voters who has their !votes overturned at MediaWiki_talk:Sitenotice#ACE2011_Edit_Request weren't aware of the discussion. I didn't know about it, and judging from Sven Manguard's apology above, he or she might have not known about that discussion as well. There was a serious lack in communication. A note should've been added to the RfC or its talk page. --Michaeldsuarez (talk) 20:54, 11 December 2011 (UTC)

I wasn't aware of it, and I'm an election coordinator. Tony (talk) 00:28, 12 December 2011 (UTC)

This issue was raised in several places including on the co-ordinators' page here. And I raised it indirectly here (by the link to my comment at the sitenotice talk page). At least one of the co-ordinators replied to me there, so I assumed all of them knew of this. How could you have not been aware of this? This is definitely one of the things that needs to be codified, though this time a note can perhaps be left at that talk page so that those watching that page can then argue against it if they wish to do so. Carcharoth (talk) 03:10, 12 December 2011 (UTC)

An interesting thing, since one of the concerns is the banner appearing for ineligble users particularly anons, couldn't the fancy new code used for the research banners be used next year to migate this? Nil Einne (talk) 13:05, 20 December 2011 (UTC)

Chart on participation in 2010 and 2011

There has been low-intensity edit-warring about a chart, that seemed informative and relevant to this page. (Perhaps I am missing something.)

Will anybody explain the deletions and reinsertions, please?

Thanks!  Kiefer.Wolfowitz 12:59, 14 December 2011 (UTC)

Wow. I would have thought something called "secure poll" would be a bit more secure than that. --FormerIP (talk) 13:09, 14 December 2011 (UTC)

What, exactly, do you think is insecure about it? The names of the users who have voted is, and always has been, public knowledge; that is a vital element of transparency. The way those users voted is, of course, secret, and always will be. Happy‑melon 13:22, 14 December 2011 (UTC)

It's not a normal element of transparency in elections, though, is it? Why do I need to know who voted? --FormerIP (talk) 13:42, 14 December 2011 (UTC)

Perhaps it is unusual to you, but certainly it is public information in just about every democracy *who* voted. What isn't public is *how* they voted. The list of voters for SecurePoll is public information, always has been, and it is perfectly acceptable for it to be used for statistical purposes. I am fairly certain that it has been used for statistical purposes, including comparisons to back when the votes themselves were public, over several elections. I seem to recall a Signpost article on this subject a few years ago. Risker (talk) 13:54, 14 December 2011 (UTC)

No, I think it most democracies it would be illegal to disclose lists of people who voted. You normally vote in a secret ballot. I think its decidedly odd that we break from that normal expectation without telling users up front. --FormerIP (talk) 14:08, 14 December 2011 (UTC)

No it's not illegal. Quite frankly no one discloses who voted because no one really cares, except when you sometimes hear that someone running for office hasn't voted since Jimmy Carter was in the White House or something like that. Hot Stop talk-contribs 14:36, 14 December 2011 (UTC)

(ec) In modern democracies the ballot is secret, but the lists of voters (both of individuals who are eligible to vote, and of individuals who actually did vote) is not. The integrity of the voting process breaks down if the public cannot verify that the voter's list hasn't been padded with ineligible (or fictitious, or deceased) voters' names, or if there's no way to detect ballots that have been stuffed under the names of voters who didn't actually vote. Depending on your jurisdiction there may be paperwork to fill out, or you may have to be affiliated with a political party or candidate, or you may just be able to walk in off the street and ask at clerk's desk—but there will be a mechanism for just about anyone to see the voter's list. While many people may be unaware of this aspect of their democracies, it's certainly not an unusual practice. TenOfAllTrades(talk) 14:37, 14 December 2011 (UTC)

I still don't think that's correct, Ten. In the US (which isn't in any case "most democracies", there is are lists of voter registrations. In the UK, for example (and I know this for certain fact), you have the right not to have your registration info made available and records of who actually voted are anonymised and then destroyed. --FormerIP (talk) 14:49, 14 December 2011 (UTC)

In the UK, the full electoral register (not the edited version available for commercial sale, which you can opt out of) is a public document that is available for viewing by members of the public. They won't give you a copy except under certain very limited circumstances, but anyone can walk in off the street and have a look. See, for example, Electoral roll#United Kingdom, or here, here, or here. TenOfAllTrades(talk) 15:28, 14 December 2011 (UTC)

TenOfAllTrades is correct. After each election in the UK, the official list of people who actually cast their vote is made available to the political parties (there is a fee to be paid). It's known as the 'marked register'. (The parties may also have their own record because in closely fought elections, they have tellers outside polling stations to take the numbers from people who have voted). Sam Blacketer (talk) 16:43, 14 December 2011 (UTC)

OK Sam, you're right about the existence of the marked register, but it isn't a public document. --FormerIP (talk) 17:13, 14 December 2011 (UTC)

It is widely available. See page 7 of this PDF from the electoral commission. Thryduulf (talk) 17:37, 14 December 2011 (UTC)

Wikipedia:Arbitration_Committee_Elections_December_2011/Log – It appears to be a means to prevent voter fraud. Some votes are being struck out (eg. [1]). Without public scrutiny, unscrupulous users could potentially strike out innocent votes in order to influence the election's outcome without anyone knowing. --Michaeldsuarez (talk) 14:43, 14 December 2011 (UTC)

I can certainly confirm that this is the case in Canada, at least. It's one of the favourite sports of journalists when an election has been close in a riding.  :-) — Coren ^(talk) 14:44, 14 December 2011 (UTC)

Knowing who voted in a normal election is as simple as sitting outside the polling station watching people come and go; teams of monitors affiliated to political parties do so all the time. Happy‑melon 15:28, 14 December 2011 (UTC)

But you don't have to give them your name. I'm blown to bits by the fact that everyone is defending this, though. Should users not have a right to cast a vote without being publicly logged and monitored? --FormerIP (talk) 17:13, 14 December 2011 (UTC)

You don't have to give them your name? Voter fraud? Hot Stop talk-contribs 17:16, 14 December 2011 (UTC)

(edit conflict)You don't have to give the tellers your name, you do have to give your name and/or show your polling card (which has your name and voter number on it) to the officials manning the polling booth. Those officials mark you off on their register to say that you have cast a ballot. After the election this register ("the marked electoral register") can be examined, currently, by anyone who wants to. Anyway, regardless of what happens in real world elections, the voter log for Wikipedia elections is explicitly public, has always been so, and is thus available for anyone to examine. Thryduulf (talk)

(edit conflict)I can not speak for other jurisdictions in the United States, but in the State of New York, the official procedure for voting instructs the election workers to call out the name of each voter when they sign the voter registry, so that any election observers present may record them, compare them against any lists they have, and if they have cause, challenge the eligibility of the voter to vote. In actual practice, most polling stations either don't have observers present, or the observers don't care to have the names announced, but who actually votes is certainly public information if anyone cares to find it out. Monty₈₄₅ 17:42, 14 December 2011 (UTC)

As a side note, still in 2008 the votes themselves were also public, rightly changed. --Tikiwont (talk) 17:27, 14 December 2011 (UTC)

Monty, what you're describing in NY doesn't sound to me like publicly available lists, it just sounds like list available for the purposes of administration. But whatever.

What would be wrong with putting a message on the voting page just telling people that there will be a public log of usernames? --FormerIP (talk) 17:58, 14 December 2011 (UTC)

Makes sense. Hot Stop talk-contribs 18:07, 14 December 2011 (UTC)

I see no reason why that couldn't be included, but I'm honestly surprised you did not expect this list to be a normal part of election transparency. I can think of no way except making it available to make sure that everybody who voted was counted and nobody who didn't was. Otherwise, we end up with scenarios like "734 people voted, here are the counts. Trust us." While I have no hesitation to, in fact, trust the current scrutineers and election officials, being unable to check that information would be very harmful to transparency. — Coren ^(talk) 18:51, 14 December 2011 (UTC)

I don't see that it does much for election transparency. If I were to allege the election was rigged, I don't see how being able to point to a list of voters, but not how they voted, would satisfy me.

I don't think most users would be expecting it. Not that we have to model our practices on any particular RL election, I really don't think that it is normal for this type of list to be public. Maybe in some jurisdictions, but not normal and I can't see any good reason for it. It's also not expected because it's an online form, and we normally expect data we provide using online forms to be confidential to the public (at least if the website is reputable).--FormerIP (talk) 19:03, 14 December 2011 (UTC)

I believe that in the United States, records showing who voted are generally public records. I know that in New Jersey, they are public records. The information is not necessarily provided in as convenient a form as the SecurePoll voter list, but it is publicly available. So it seems perfectly "normal" to me. Of course, as others have said, how "regular" elections are conducted is of little relevance here. The SecurePoll voter list for this election has been posted and remains visible. I see no justification for removing the table that was based on that publicly available information. (I am not sure what lessons, if any, can be drawn from the table, other than that participation went down from one year to the next, but that has nothing to do with whether the table should have been removed.) Neutron (talk) 19:10, 14 December 2011 (UTC)

@FormerIP: If the voting log were private and its oversight were conducted by a group of "trusted" scrutineers, then who watches the scrutineers? I believe that the solution to that problem would be to have everyone (ie. the public) become the scrutineers' scrutineers. --Michaeldsuarez (talk) 21:09, 14 December 2011 (UTC)

OK, you've forced me into it. How elections are run in the United States is not a good guide to how Wikipedia or anything else should run things.

Editors who live in places where elections are run differently will not be expecting their names to be posted on a public list because they voted. This is especially the case since the voting is done my means of an online form, which people normally expect to produce secure data. I've given up caring about whether its a good idea or a bad idea. It just seems like a fairly pointless idea and the notion that it ensures transparency is just cobblers.

But what would be wrong with simply informing editors at the point where they vote? --FormerIP (talk) 21:23, 14 December 2011 (UTC)

Didn't Special:SecurePoll/vote/240 contain a link to Special:SecurePoll/list/240? Does anyone have a copy of what Special:SecurePoll/vote/240 used to said? --Michaeldsuarez (talk) 21:40, 14 December 2011 (UTC)

For what it's worth, right at the top of this page is a link to the "Voter log", which is the list of everyone who voted, and I know for certain that the same link was on the election page during the election process, because I clicked on it after I voted, to make sure that my vote had been entered properly. And the reason that I did that was because I read something in the instructions to voters saying that one could do that. It seems to me that any voter who read what I read would have understood that a viewable list of all voters was being generated. --Tryptofish (talk) 21:56, 14 December 2011 (UTC)

I did the same thing, Tryptofish. I am not sure where I saw the message that told me I could (and I knew I could from last year's election), but it must have been somewhere visible. Maybe one of the election administrators could confirm exactly where this message was placed. Neutron (talk) 22:18, 14 December 2011 (UTC)

You can see the contents of the various SecurePoll messages at Special:SecurePoll/translate/240/en; the description of, and link to, the public log is absolutely clear: "This record will not reveal to anyone how you voted, just that you have voted.". Happy‑melon 22:27, 14 December 2011 (UTC)

Yes, that's what I remembered. It said very clearly that you could see a list of everyone who voted, but it would not show how anyone had voted. I remember it as being quite conspicuous and clear on the voting page. --Tryptofish (talk) 22:31, 14 December 2011 (UTC)

Exactly. If FormerIP really wants to take this further, I suggest the feedback page might be a better place. What is clear is that during this election and the two previous ones, a vast and silent majority of hundreds of voters have been aware that a voter log was being generated and have raised no objections. I too would like to know why Ebe123 and Sven Manguard carried out their reverts - there seems to have been little to no justification at all for those reverts. Carcharoth (talk) 22:35, 14 December 2011 (UTC)

All of the above is a mildly interesting distraction to the question asked at the top of this section: why was the inclusion of the table, based on publicly-available data, reverted by Ebe123 and Sven without explanation ? (I don't count an edit summary "Wth?" as particularly enlightening). Alexandr Dmitri (talk) 18:00, 14 December 2011 (UTC)

Thanks! My goal was only to stop the edit-warring by encouraging a discussion. Nonetheless, I would be interested in explanations. (I admit to some jealousy that another editor had a contribution removed by Sven! I thought we had a special relationship!)  Kiefer.Wolfowitz 21:41, 14 December 2011 (UTC)

Block each of them for 15 minutes to record what they were up to on their "report cards." Cla68 (talk) 22:39, 14 December 2011 (UTC)

Can we try and keep discussion serious here? What they did wasn't helpful, but neither was your comment. Carcharoth (talk) 22:45, 14 December 2011 (UTC)

I was serious. People edit war too often in project space because they aren't held accountable for doing so. The block log is the closest thing to a report card that editor accounts have, so note what they did in their block log with a short block. Maybe then, editors will try to talk things out before resorting to revert warring on pages like this one. Cla68 (talk) 22:57, 14 December 2011 (UTC)

There is, though, the possibility that it was a misunderstanding (someone thinking it was non-public data being disclosed). You would normally ask for an explanation before going any further there. Compare that to the snarky comment made above by Kiefer.Wolfowitz about Sven Manguard. That is something I'd be comfortable removing outright as there is no chance of that being a misunderstanding (it was clearly designed to bait Sven rather than contribute to anything productive here). I only haven't removed it because I wanted Kiefer to do the right thing and remove it himself, and because an argument about that comment would only distract further from what should be being discussed here. About 'report cards', warnings and notes left on user talk pages serve the same function, which is why those pages are usually not deleted on editor request, and selective archiving reflects poorly on the user that engages in that practice. Carcharoth (talk) 23:31, 14 December 2011 (UTC)

(after edit conflict) Kiefer, there is no need to bring any past history between you and Sven Manguard into this. Could you please remove that comment where you refer to him. Carcharoth (talk) 22:42, 14 December 2011 (UTC)

Apologies are in order on my part. I'm not entirely sure what happened with my "wth" edit, but something clearly went wrong. I thought that the chart was being removed (which sparked the "wth"), and that I was undoing the removal, not re-removing the chart. I actually liked the thing and am all for sticking it back in. It's not the first time I've screwed up an edit, and it probably won't be the last. Oh, and thanks for telling me about this conversation everyone. Sven Manguard Wha? 06:29, 15 December 2011 (UTC)

Thanks for the explanation and the apology. Now we need Ebe123 and Bishonen to explain why they used the rollback tool: [2], [3]. --Michaeldsuarez (talk) 12:56, 15 December 2011 (UTC)

Or we could just wait for the election results, at which point no-one will care less about this minor flap. Carcharoth (talk) 14:20, 15 December 2011 (UTC)

[4], [5] – It seems as if no one really knew what was happening. --Michaeldsuarez (talk) 13:24, 19 December 2011 (UTC)

I'm very sorry about the rollbacking. I don't recall doing it to this page but nonetheless, sorry. If I would of noticed, I would of undo-ed my edit right away. ~~Ebe123~~ → report on my contribs. 12:24, 19 December 2011 (UTC)

In the Australian ballot, the origin of the secret ballot, as currently practiced voting is compulsory and the Australian Electoral Commission is generally circumspect about its prosecutions of people who violate the law (without justifiable excuse). I assume everyone knows I voted, because I come from a culture where voting is an obligation of members of society, but the publication of administrative data is highly unusual to me and offensive. Fifelfoo (talk) 09:10, 15 December 2011 (UTC)

One point about publishing who has voted is that if everyone except one person reveals how they voted (and are truthful about it), that will reveal the votes of the one person not saying anything about their votes. It is, of course, everyone's right to say nothing about how they voted. This is less relevant to votes involving several hundred people, even less relevant to local and national elections involving thousands to millions, but it does become relevant for committee votes involving only a few tens of people, which ironically is the size of ArbCom. Though of course committee votes are very different from electoral votes. It is interesting that there are more concerns about publication of a voter log that might have been assumed. Carcharoth (talk) 14:20, 15 December 2011 (UTC)

You can't prove that everyone was telling the truth, though (essential in order to avoid ballot-selling); even if the votes stated by the N-1 other candidates add up to the correct totals minus one, all you can know with any confidence is that an even number of people have lied... :D Happy‑melon 16:41, 15 December 2011 (UTC)

Good point. BTW, might the election admins and/or co-ordinators keep us updated with progress? Something along the lines of confirmation that the scrutineers have confirmed that they are doing stuff and that no unexpected glitches have been encountered before the white smoke goes up to signal the results. Carcharoth (talk) 01:30, 16 December 2011 (UTC)

As can be seen from monitoring the on-wiki voting log, the scrutineers have been attentively investigating suspicious votes, and striking those suspected of ineligibility. Several voters with discrepancies in their votes (i.e. evident cross-site request forgeries) have not responded to initial inquiries from the scrutineers and have been provisionally stricken. The scrutineers intend on giving them the full time available to respond, given that a small number of votes can mean the difference between successful candidates and also-rans. In other words, all is proceeding according to plan and schedule and rest assured, the watchers of this page would be the first to hear if they were not. Skomorokh 01:37, 16 December 2011 (UTC)

For someone whose last programming assignment was a cross-site request forgery attack, that sounds concerning. Is this a weakness in the current software? --Rschen7754 01:41, 16 December 2011 (UTC)

As someone with markedly less programming nous, I couldn't responsibly comment as to whether the software is particularly vulnerable to such attacks. You might want to take that up with the developers. I can tell you that it does flag potential CSRFs to election administrators (incl. scrutineers), for what its worth. Skomorokh 02:04, 16 December 2011 (UTC)

Well, sort of, Skomorokh. I believe the schedule was laid out with the intention of results being posted by the end of the day on 15 December (if not sooner), but given that I have myself raised a bit of a concern about a technical glitch given the high percentage of CSRFs, I won't complain if the scrutineers need a bit of extra time to consult with the sysadmin liaison. Rschen7754, I've had quite a bit of experience with the SecurePoll extension, having acted as a scrutineer myself in its last use (the image filter plebiscite). I'm wondering if some of the software changes have had an impact on the manner in which cookies are being activated, because the percentage of CSRFs is significantly higher in this vote. One thing we noted was that certain browser settings can cause a "false positive", but it may not be entirely possible to sort that out. It's well outside my scope and I acknowledge a conflict of interest here, but I am concerned that striking votes for what is intended to be a "heads up" as opposed to an alert of likely duplication may have an effect on the outcome. Risker (talk) 01:47, 16 December 2011 (UTC)

I'm not sure where you got that idea of a schedule from; the official timetable rightfully does not set a strict deadline. The informal instructions to scrutineers sets a target date of the 17th. I'm not sure it's altogether helpful to discuss putative CSRF percentages given the necessarily tiny sample size, restricted access to such information and the privacy of voters. Striking votes or not is at the absolute discretion of the scrutineers, and could scarcely be otherwise, as the community has not laid down any guidelines save the voter eligibility criteria. We've asked these volunteers to help with our imperfectly planned elections on imperfect software, and to my mind should let them get on with the job and concentrate on setting our own house in order for future elections. Skomorokh 01:56, 16 December 2011 (UTC)

Entirely correct on all counts. I'd not noticed the instructions to them. And yes, from the scrutineer perspective, SecurePoll is a pain in the neck. Risker (talk) 02:00, 16 December 2011 (UTC)

Indeed. The enwiki community is very welcome to investigate the frequency of CSRF-flagged votes in ACE elections (as I'm still listed as election admin for all of the past three elctions I'd be able, and very willing, to extract the necessary data); but not while an election is ongoing. It would be even better if the community were to reach its own decision on them, rather than leaving it to the scrutineers. But for the time being, we have devolved full authority to the scrutineers to decide how these things should be handled. Happy‑melon 12:55, 16 December 2011 (UTC)

As someone who has never heard of a "cross-site request forgery" before, can someone explain (in less technical terms) what is wrong with these votes? Is there evidence that someone other than the account-holder was using the account to vote? Or is it something else? And (again, without being too technical) how strong is the evidence? Neutron (talk) 19:26, 16 December 2011 (UTC)

Basically, it means that the vote could have been faked, not casted by the user themselves. We actually contacted those users more than 10 days ago to confirm if they actually casted the vote themselves. But till now they have not responded. To ensure a secure election, if we know that there is a possibility of fake vote, it is irresponsible to leave them there. We still chose to struck those 5 votes after some discussion (with election admins).Ben.MQ (talk) 01:28, 19 December 2011 (UTC)

I would add that even if you don't publish the list of people who voted, many people may feel free to publicly declare that they voted, without expecting it will lead to who they voted for revealed. Yet if T (= total number of people who voted) - 1 people declare how they voted, some may assume this reveals how the person who publicly declared that they voted, voted. Of course there's even a greater risk from lying since in this case your pool of liars is all possible voters rather then all voters bar one, but I think the point remains when you have 700+ voters, such risks are too insignificant to be worth considering even if more accute then with 1 million voters. (In NZ results from polling places with less then 10 voters are only included in a general tally to avoid the possibily how someone voted may be revealed, 10 is obviously a far bigger risk.) Nil Einne (talk) 13:30, 20 December 2011 (UTC)

Timing

One aspect which surprised me was, the way voter names were added to the page meant it was possible to see when a person voted. I wasn't expecting that information to be public. I'm not especially bothered (with regards to my own voting), but I'm not convinced it's a great idea to show the list during the election. An important part of closed ballot voting is, that nobody but the voter has any right to know what their decisions were based upon; certain aspects of voting methodology can be easily inferred from the timing aspect. For example, someone choosing to vote within the first hours has clearly not bothered worrying about questions that have not yet been responded to.

So for future elections, I would suggest that the list of users who voted at least not be public until the voting is concluded.  Chzz  ►  15:28, 15 December 2011 (UTC)

I'd say that again, this is not at all different from an ordinary election, where anyone can see what time a person voted simply by sitting outside the polling station and waiting for them to turn up. But more importantly, a) does the timing of a vote really give information about a voting methodology, and b) what possible use is that information to anyone even if it does? In the case you mention, you can indeed infer that the voter is sufficiently satisfied with a suffienct number of candidate's answers to a sufficiently wide range of questions that they are happy to vote based on the information they currently have, and do not want to wait for either answers to further questions currently posed-but-unanswered, or for hypothetical questions which may or may not be asked in the future; where all those "sufficiently"s are personal preferences about which an observer knows nothing. Or they may not be so satisfied, and intend to come back and revise their vote later.

So basically, the one piece of information given to an observer allows them to say absolutely nothing with any confidence about the voter's motivations. Case in point: I voted just a few minutes after the start of voting this year. How much do you think you can say about the "methodology" behind that vote? Happy‑melon 16:37, 15 December 2011 (UTC)

My core problem with it is the "surprise factor". When I participated in that ballot - which assures one of confidentiality - I was alarmed to notice, a few minutes later, that my participation has just been announced to the world on a highly-visible page. As I said, it didn't bother me - but it surprised me. I could think up various reasons why people might object to it; to take one example, how about if a group of users that were known to work together within some project area, and happened to all cast their votes at the same time, people might be suspicious of sock/meat puppetry.

Let's turn the question around - what is the benefit of listing voter usernames during the election?  Chzz  ►  12:22, 16 December 2011 (UTC)

First of all, we can count the number of voter names 'the powers that be' claim have voted, and verify that there are as many as there are votes; that proves that there has been no ballot stuffing. Then we can check the list for voters who are not elegible or do not exist; and satisfy ourselves that the election has not been subject to ballot padding. If we are particularly cautious or concerned, we could ask some or all of the voters whether they did in fact vote, and thereby detect any ballot theft. Happy‑melon 12:35, 16 December 2011 (UTC)

Agreed, additionally people who don't vote have less cause to complain about their decisions.--Wehwalt (talk) 12:37, 16 December 2011 (UTC)

It also makes it trivially easy to check whether newly discovered sockpuppets voted more than once. Thryduulf (talk) 14:30, 16 December 2011 (UTC)

I can understand those as reasons for listing the names of voters - and that isn't my concern; I don't see why any are valid reasons for listing them at the time the votes actually occur.  Chzz  ►  10:49, 17 December 2011 (UTC)

Well, it lets you doublecheck that your vote registered.--Wehwalt (talk) 12:59, 17 December 2011 (UTC)

Real life elections do sometimes involve a checklist of party faithfuls at the door of the Polling station. Seems a little unlikely here, with 17 independents and no parties. But I'm glad I received no last minute prompting from any of the candidates! Maybe that's aginst the rules. Who would want my vote, anyway. I'm not sure why the list of voters had to be time-stamped. Personally, I would have found it more interesting to see a cumulative indication of vote turn-out precentage. Martinevans123 (talk) 14:21, 17 December 2011 (UTC)

The time-stamping is useful, because you can change your votes any time up until the close, and without the time stamp you can't be sure your new vote has registered -- Boing! said Zebedee (talk) 14:53, 17 December 2011 (UTC)

Yes, that's true. Although some kind of automated personal reply message, to one's talk page, might work equally well? Martinevans123 (talk) 14:59, 17 December 2011 (UTC)

I don't see how that's going to help with Chzz's concerns tho, not that I agree with them. Nil Einne (talk) 13:33, 20 December 2011 (UTC)

I'm not sure why this is an issue at all. A secret ballot guarantees that the choices made by a voter are known only to that voter (unless he or she chooses to make them public). Everything else is usually on the public record and for good reasons. The more that is publicly known about the process, including who voted and when, the more the likelihood that the process is 'free and fair'. --regentspark (comment) 14:46, 17 December 2011 (UTC)

Update on the announcement of results

The election admins have been in contact with the scrutineers and can tell you that they are now entering the final checks phase. The release of results should follow before too long, once the checks are done and the scrutineers are all available to announce and sign-off on them. In the meantime, we would like to re-stress the importance of having your reflections and suggestions for improvement on this year's feedback page.

For the election administrators, Skomorokh 23:47, 17 December 2011 (UTC)

Yep, I've just a few checks to do. --Vituzzu (talk) 10:05, 18 December 2011 (UTC)

Its now been over a week since voting closed --Guerillero | My Talk 20:04, 18 December 2011 (UTC)

.. election, was there an election? Martinevans123 (talk) 21:03, 18 December 2011 (UTC)

Skomorokh, with the greatest of respect, this isn't really an update... The Cavalry (Message me) 20:28, 18 December 2011 (UTC)

Results are out now, there was some trouble with CSRF (as usual) and timezones...--Vituzzu (talk) 21:52, 18 December 2011 (UTC)

All that and no link to where the results are? You guys can be pretty frustrating sometimes.. all right, I'll just go by which of the candidates have been congratulated on their pages by Hersfold, then.[6] That works too. Bishonen | talk 22:51, 18 December 2011 (UTC).

Just click "Project page" at this top of this page to go to the page with the results. –MuZemike 22:53, 18 December 2011 (UTC)

Wikipedia:Arbitration_Committee_Elections_December_2011#Results. --Michaeldsuarez (talk) 22:54, 18 December 2011 (UTC)

Some voting trivia

There were 729 votes tallied for the results. If percentage was calculated as Support/(Support + No Votes + Oppose) the ranking would remand the same except for flipping seats six (SilkTork) and seven (AGK). In order to change the results between seats seven (AGK) and eight (Jclemens), 51 votes would have to be struck, each one of those voting support for AGK and oppose for Jclemens. To change the results between eight (Jclemens) and nine (Worm That Turned), 8 votes would have to be struck, each one of those voting support for Jclemens and oppose for Worm That Turned.--Tznkai (talk) 22:10, 18 December 2011 (UTC)

But what was overall percentage voter turnout? Martinevans123 (talk) 22:25, 18 December 2011 (UTC)

Against what? Potential voters? "Active" users? People who have commented in any Arbitration thread or case or request?--Tznkai (talk) 23:08, 18 December 2011 (UTC)

All eligible voters. Thanks. Martinevans123 (talk) 23:09, 18 December 2011 (UTC)

While you wait for a definitive answer, I can offer you the number of active users: 120,065. The turnout as a percentage of active users was 0.60717111564569% (and you thought that voters in your local elections were apathetic). --RexxS (talk) 00:16, 19 December 2011 (UTC)

Depends how we're defining "active". The vast majority of Wikipedia's editors participate on the project only casually, meaning a few edits per year. They tend not to run into situations requiring dispute resolution, and probably care little about the politics in the royal court. Most of them probably don't even know what ArbCom is. A better number to measure activity is the "more than 100 edits/month" metric, which shrinks the community size down to a core of about 3500.[7] But even among those, again, most probably care little for the Conflict du Jour, they're just working off in their corners of the project. Which is kind of nice, actually, to think that there are more editors working on articles, than caring about the intrigues of wiki-politics. --Elonka 03:15, 19 December 2011 (UTC)

Thanks, Rex. A very useful little equation, even with the spurious over-accuracy. Isn't that what most people would call "half of one percent"? Even by-elections in Northern English towns, on a wet day in December, do a hundred times better than that. But, I guess, to add any voter eligibility condition of e.g. "100 edits in the past month", would unfairly disenfranchise just too many hewers of wood and drawers of water? Even if they could care less. Democracy means you can vote if you want to - even if you don't actually exist? Martinevans123 (talk) 21:04, 19 December 2011 (UTC)

Why on earth would a solution to the (alleged) 'problem' of low turnout, be to reduce the number of people elegible to vote?! Happy‑melon 21:27, 19 December 2011 (UTC)

Perhaps because the vote of the editor who has made 150,000 edits is worth only as much as that of the editor who has made 150. But that's the beauty of democracy. Problem? what problem? As long as there are more voters than candidates, that's all that matters, isn't it? I guess most editors don't care about ArbCom because, fortunately, they never need it. Perhaps we ought to be grateful that voting is not simply restricted to administrators? But that's a meritocracy isn't it, like in North Korea. Martinevans123 (talk) 21:40, 19 December 2011 (UTC)

If you're making a point, it's not coming across, but I suspect you're not, or at least not strongly. If you do want to argue that the ArbCom Elections should be non-democratic, and that the votes of admins, the 55 users with more than 150,000 edits, or whatever other 'elite' group should carry more weight, then you're welcome to, but I doubt you'll find much likeminded sentiment. Happy‑melon 22:00, 19 December 2011 (UTC)

The point I'm trying to make is this: although low "turnout" makes this election look "unrepresentatitve", that's a red herring because most of the electorate don't care or don't even exist in any meaningful way. But there's no better or "more democratic" way of doing things. Thanks. Martinevans123 (talk) 22:16, 19 December 2011 (UTC)

These elections are not "unrepresentative" because any editor (assuming 150 edits and no active ban at the time of voting) may vote. The fact that most do not choose to do is unfortunate, but doesn't affect the representative nature of the election. The non-voters choose to allow those who do vote to make the decision for them. Neutron (talk) 23:30, 19 December 2011 (UTC)

I fully agree with you. Martinevans123 (talk) 17:38, 20 December 2011 (UTC)

Sadly many or most of them are just doing f***ing geo-locations, project assessments, & other cruft or semi-cruft, or vandal patrolling. But I entirely agree that, with a few exceptions who work offline & produce finished articles with one edit like conjurors, the 100 edits per month figures are the most meaningful level we have for assessing effective actives. Johnbod (talk) 04:25, 19 December 2011 (UTC)

That isn't sad. This is a volunteer project, and we should welcome any help we get.--Tznkai (talk) 04:38, 19 December 2011 (UTC)

Concur strongly with Tznkai. It is wrongheaded to think that only those editors who write full articles are the "valuable" ones of the project. Wikipedia is a living example of how thousands of editors each making small edits, can still produce an amazing resource. We should celebrate and encourage all of our constructive editors, not just those who produce "finished articles". --Elonka 15:33, 19 December 2011 (UTC)

Congratulations to the candidates

And thanks to those of you who freed me from servitude.  :-) This was the last time I ran, regardless of the results. I offer my most sincere congratulations and thanks to the new (and old) members of the committee; it's a thankless job that needs to be done, but it's a pain to do and you all deserve commendations for volunteering to do it.

Good luck. — Coren ^(talk) 22:54, 18 December 2011 (UTC)

I'd like to take a moment to thank everybody that has been on the Arbitration Committee. It's a dirty, nasty, thankless job. Regardless of whether I agree or disagreee with certain positions, I have a great deal of respect for anybody that's willing to serve our community in this fashion. Taking on this respobsibility is an enourmous task and the hostility thrown at ArbCom is intense and never ending. Regarless of whether the hostility is justified or not, these are all volunteers that offer to do the most difficult work under the most difficullt situations, and they should be thanked for their efforts. They don't do this for fame or wealth, they do this to help the community and their efforts should be commeded. These are real people with real feelings and emotions and there are real world risks taken by these folks as their decisions can have a backlash to their personal lives. Serving on ArbCom is neither easy nor fun, but it's a task that needs to be done and I'm thankful for those willing to take on this task. Thank you to all that have served on ArbCom. I appreciate your efforts. - Hydroxonium (T•C•V) 00:44, 19 December 2011 (UTC)

As an added observation, I think it's gratifying that everyone who was seated had greater than 60% support. I remember some comments around the beginning of the election that we would have trouble finding even seven people who could get over 50%. It turns out that wasn't true. --Tryptofish (talk) 16:46, 19 December 2011 (UTC)

In 2009, the top eight candidates had percentages over 60%. In 2010, the top nine candidates had percentages over 60%. --Michaeldsuarez (talk) 00:23, 20 December 2011 (UTC)

But I believe there was one person elected in 2009 with just under 60 percent, and in 2010 there were three people elected with less than 60 percent. What I find interesting is that this year there were people who advocated for a "zero" threshhold because a 50-percent requirement would result in seats going vacant -- which did not happen -- and there were other people concerned that 50 percent was too low. I think the community got it just right, in the middle, at 50 percent, and the voters took care of the rest, and satisfied everybody. (Well, probably not everybody, but close enough.) Neutron (talk) 01:41, 20 December 2011 (UTC)

No spoonvote

It seems a bit odd to log "no votes". I get it but... if someone hasn't voted, why show it? They've not said that they wanted to give any opinion whatsoever, so it seems strange to record their "non choice" in a tally of results?  Chzz  ►  08:58, 19 December 2011 (UTC)

Seems six one way, half dozen the other. You could WP:SOFIXIT remove the column, but it might get reverted. Nobody Ent (Gerardw) 11:42, 19 December 2011 (UTC)

Considering that the table shows the "official" results as certified by the scrutineers, I don't think it should be touched without their permission. And I really don't think it is worth being concerned about at this point. The footnotes explain what the numbers mean, so there should not be any misunderstanding. Neutron (talk) 14:57, 19 December 2011 (UTC)

Mysterious colors in table

Per [8]. My concern is not the mechanics of seating arbitrators but lack of clarity of information -- if the table is to have light green and light blue colors there should be some explanation of what they mean. Nobody Ent (Gerardw) 12:03, 19 December 2011 (UTC)

Yes, but with a little care. You never know...something could happen before they're actually appointed. So it'll have to be something like, "The candidates marked in THIS COLOUR obtained enough votes to be appointed for 2 years; Jimbo will actually announce their appointment...blah". Something like that. We shouldn't say (I don't think) that "These are the new arbs". Gotta word it right. I noticed this edit. Actually, I'll shout out to Happy Melon.  Chzz  ►  12:09, 19 December 2011 (UTC)

No. Colour alone is never sufficient to convey information. You either give the information to everybody or nobody. Let's not discriminate against those visitors who are visually impaired. WP:COLOR contains the relevant advice, derived from WCAG, the world-wide standard. --RexxS (talk) 12:15, 19 December 2011 (UTC)

Without an explanation, the colour is indeed just decoration. The point of my revert was not that the meaning should be conveyed purely through colour, but rather that the colour should not carry (that) meaning at all. Happy‑melon 13:15, 19 December 2011 (UTC)

It's not reasonable to simply have a ranked list and have readers scratching their heads as to what it means -- I've made a good faith stab at annotating the results column as unofficial and added a note; if I've missed the mark a bit revising the title and note are preferable to simply reverting. Nobody Ent (Gerardw) 12:23, 19 December 2011 (UTC)

They're more "provisional" than "unofficial", but the sentiment is right on. The way these elections work is that all our balloting and SecurePoll-ing and questioning goes to produce precisely that: a ranked list of candidates according to S/(S+O). It is then up to Jimbo to decide what that means; and he has done some pretty wierd things in the past with the nominations, so as I said in my edit summary it is definitely not over until the fat lady sings. There are many people who think that it shouldn't work this way, that at this point we should be able to say that those seven green-highlighted candidates are appointed to two-year terms and that Jclemens is appointed to the one-year term, and maybe that's the way it will work in 2012. But for now we have this extra stage, and there's nothing lost in waiting for it to pan out properly. Happy‑melon 13:15, 19 December 2011 (UTC)

It's not at all up to Jimbo to decide what it means. The community runs the election, and if there's an issue, an RfC is required. Jimbo does the ceremonial bit. Tony (talk) 02:35, 20 December 2011 (UTC)

The stewards

Has anyone thanked them? I suppose I'll go to their pages and give them flowers. They're not at all obliged to help us. Tony (talk) 02:32, 20 December 2011 (UTC)

Appointments

For whom it may concern: User_talk:Jimbo_Wales#Arbcom_Appointments. Skomorokh 21:34, 20 December 2011 (UTC)

So there won't be any "surprises." That's good. But it does leave an open question as to whether there could, theoretically, be "surprises" in some election in the future. Neutron (talk) 23:22, 20 December 2011 (UTC)

And here are the final appointments: ArbCom Appointments 2011. Trijnstel (talk) 11:48, 22 December 2011 (UTC)