Wikipedia talk:Arbitration Committee Elections December 2008/Archive 2

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

Review of past processes

Moved from Project page; Since it's discussion, I've placed it chronologically on this page for later archival. UltraExactZZ ^Claims ~ Evidence 12:56, 27 October 2008 (UTC)

Please offer feedback concisely below:

1. I think last year's format, particularly in the questioning/answering sector, was good, but largely inferior to the layout of the recent Wikimedia Board Election at Meta. Would like to see a questioning format wherein a question is offered to all candidates and each posts their response in a table, as done here. -- Anonymous Dissident^Talk 13:02, 9 August 2008 (UTC)

   Ideally we would have the option to ask questions to all candidates, and specific questions to individual candidates, preferably all in the one location. —Giggy 05:11, 29 August 2008 (UTC)

   What Giggy and AD said. The Meta way was far better. You can see everything in one place. rootology (T) 05:14, 29 August 2008 (UTC)

   Agree that format might be worth considering, as it had clear advantages organizationally and might have been less work for questioners as well? ++Lar: t/c 22:46, 30 August 2008 (UTC)

   Compared to what I saw last year trying to navigate through each candidate page, the boardvote method is far easier to compare and read. MBisanz ^talk 22:50, 30 August 2008 (UTC)

   Indeed, the boardvote method is much better. Wizardman 23:10, 30 August 2008 (UTC)

   That'd be a really simple format to set up, as well. absolutely agree. UltraExactZZ ^Claims ~ Evidence 00:10, 31 August 2008 (UTC)

   Whatever the format is, can we try to keep the page size below 300kb for the benefit of those on slower connections or mobile devices. (Yes, I read Wikipedia on my Blackberry, but I can stop any time.) Jehochman ^Talk 04:20, 15 September 2008 (UTC)

   I doubt many pages will be under 300kb, given that we get anywhere from 30-60 people running in arb elections, I suspect there will be epic page sizes. MBisanz ^talk 12:01, 15 September 2008 (UTC)

Pages and header created

I've created a bunch of pages for the elections, including /Candidate_statements and /Menu; /Voting is forthcoming.

Queries

• Are we retaining last year's minimum requirements to run for the Committee (1000 votes as of November 1, 2008)?
• Are we retaining last year's maximum word count for candidate statements (400 words)? Too short, perhaps, or just right?

All are invited to work on, expand, and mercilessly change my work. :-) Category:Wikipedia Arbitration Committee Elections 2008 is available for tracking of ACE2008-related pages.

Anthøny ✉ 16:37, 15 September 2008 (UTC)

I assume you mean 1000 edits? That seems like a reasonable number although I'd prefer something close to 3-4K. 400 words seems like a good number. MBisanz ^talk 16:49, 15 September 2008 (UTC)

Agree with keeping the statements that short or shorter. Der Wohltemperierte Fuchs ^(talk) 04:21, 16 September 2008 (UTC)

The idea is that anyone with 1000 edits knows enough to realistically assess their chances - and to assess them quite low in most cases if 1000 edits is all they have. Haukur (talk) 12:43, 16 September 2008 (UTC)

User:Manning Bartlett actually did quite well last year with ~1200 edits (60% or so, I think), so it does happen that 1000 edit people do well. Wizardman 13:42, 16 September 2008 (UTC)

Well, he actually had 2,239, but fair enough point, 1,000 hopefully idiot-proofs it. MBisanz ^talk 13:57, 16 September 2008 (UTC)

A sometimes overlooked bonus that comes with having few edits is that you have less of a record to scrutinize—that can be helpful to a candidate, because having a limited history makes it easier to appear perfect, but it can be deceptive to voters. I'd support a requirement of no less than 3,000 total mainspace edits, and at least 1,000 mainspace edits in 2008. Everyking (talk) 06:04, 17 September 2008 (UTC)

Installing a minimum requirement of 1000 mainspace edits would not be a fair move in my view, and indeed seems to attempt the elimination of a certain type of candidate (one which is very much "project space-based"—a very common type of Wikipedian) from the election. Whilst the minimum raw edit requirement is understandable, the minimum article space requirement proposal seems to be a move which is more the purview of the voters (it is they should decide which type of Arbitrator they wish to sit this year) rather than the Election maintenance folks. (Of course, it does stand to reason that an editor with several thousand edits, but under 1000 mainspace edits, would be disadvantaged due to his or her lack of involvement with the core of the encyclopedia; the key difference in principle there, however, is that such a candidate is disadvantaged—indeed, to the point of being unsuccessful in his or her candidacy—as a result of the Voters' choices, rather than of the editors generating the minimum candidate requirements.) Anthøny ✉ 06:36, 17 September 2008 (UTC)

Well, we could just say no requirements at all, and let the voters decide it all, right? But I think it's better to have a more tightly organized process in which only serious candidates with substantial encyclopedia contributions get to run. That way, we'll have a shorter list of candidates and a more focused process. Furthermore, there is the problem I mentioned above, which is that candidates with fewer contributions can sometimes seem deceptively "perfect". As long as the candidates have a reasonably substantial number of mainspace edits, we will have enough history to properly judge their candidacies. Everyking (talk) 07:06, 17 September 2008 (UTC)

We tried the no limit aproach we got a bunch of no hoper candidates who didn't realise what they are getting into. 1000 is enough to make sure that everyone has some idea what they are getting into without restricting anyone who has the faintest chance of getting elected.Geni 00:32, 20 September 2008 (UTC)

I'm strongly opposed to the 1000 mainspace edits in a year requirement. Someone may have perhaps a few hundred edits at most in a year but end up contributing multiple featured articles in difficult subjects, while someone with 10000 mainspace edits in a year may have spent all their time mistakenly correcting 'spelling errors' like espresso to expresso. However, neither ability to make featured articles, or ability to trot up a large number of individual edit actions, are really relevant qualifications to acting as a member of an arbitration panel.

To use one of my famous convoluted analogies. A bakery doesn't hire it's accountant based on how many loaves of bread he baked in the last year, they hire one on their ability to be an accountant.

Why not just make a statement that demonstrated experience as a Wikipedia editor is a requirement for the post, and candidates are required to give a short, 250-500 word, summary of their significant contributions to Wikipedia either in content or in policy. As well as a requirement to give a short, 250-500 word answer to the question "What is Wikipedia?". This turns a 'edit count hurdle' into a check that the candidate is willing to answer a simple essay question demonstrating significant experience, and their understanding of the purpose of Wikipedia.

I think we can agree those are relevant basic tests for a candidate. It might also stop us returning candidates to the office who can't draft a proposal on their own. --Barberio (talk) 19:58, 20 September 2008 (UTC)

That is subjective. We are looking for objective criteria.1000 edits across all namespaces appears to have worked in the past and there is no pressing need to change it.Geni 20:47, 20 September 2008 (UTC)

No, ability to complete two essay questions is objective, their contents are subjective. We don't need to assess the contents beyond filtering out obvious gibberish, the mere requirement to provide an essay question answer will weed out people who aren't able to produce one.

The assessment would be

• Is the answer in the approximate range of 250-500 words?
• Is the answer intelligible English language?
• Does the answer try to address the question?

All of which are much more objective assessments. And as surprising as it may seem, this is a relatively high bar on entry and will filter out people just doing it for the Lulz.

While edit-count is an objective measure, it's not a meaningful objective measure in selecting candidates. Shoe size is an objective measure too. --Barberio (talk) 21:15, 20 September 2008 (UTC)

Objective? Please provide a mathatical description of the method of judgeing the second and third requirement.

The edit count requirement isn't meant to select candidates. Not even meant to stop people running for the lulz. Selecting candidates is a matter for the voters.Geni 23:23, 20 September 2008 (UTC)

Er... Voters would have a hard time selecting candidates excluded from the process because of an edit count.

And if you really think that "Is this intelligible english" is a more subjective question than objective, I wonder how you cope identifying the cnksibtcjbb c hdugvbdh efhh hjd wugaba? --Barberio (talk) 12:39, 21 September 2008 (UTC)

We know that someone with less than 1000 edits has no real chance of being elected we know there is a correlation between low edit count and not realiseing what you are getting into. Thus a 1000 edit requirement deals with the problem of people not knowing what they are getting into without impacting the outcome of the vote. Is "Colourless green ideas sleep furiously" intelligible english?Geni 18:22, 21 September 2008 (UTC)

Check the requirement being proposed by Everyking again, it's 1000 edits *during 2008*, if we open nomination in November, that's a 100 mainspace edits per month. There are many worthwhile editors of the wiki that don't have that sustained edit count, and it'd rule out anyone who had taken a short sabbatical this year.

And no, we *don't* categorically 'know' there is any such correlation between low edit count, and competency to be part of the ArbCom.

And yes, "Colourless green ideas sleep furiously" is intelligible English, just meaninglessly so. But it wouldn't try to address the question, or be between 250-500 words. --Barberio (talk) 20:44, 21 September 2008 (UTC)

I know what everyking is proposeing but I'm argueing for the status quo. I never said "competency to be part of the ArbCom" I said not knowing what you are getting into. Since you appear to have defined intelligible English as uses real words a markov chain genorator could pass your test.Geni 21:13, 21 September 2008 (UTC)

I assumed you were arguing against my oposition to the 1000 per year limit, and proposal of an alternative instead of it?

And No, a Markov Chain Generator could not be nominated, because we could make a very objective observation that Markov chain generated gibberish was not addressing the question. If in the future, we create an AI that could answer the questions, I welcome their candidacies! --Barberio (talk) 21:34, 21 September 2008 (UTC)

Nope not gibberish you accepted colourless green ideas sleep furiously is intelligible English which means you accept any gramiticaly correct sentance containing real english words as intelligible English. If you wish to use "addressing the question" as a criteria please describe that in a robust manner.Geni 02:47, 22 September 2008 (UTC)

Actualy, markov chain generators do not reliably generate intelligible English at all, and are prone to various problems in generating sentences. And I don't know of any program that can reliably generate a large amount of intelligible english.

As for a 'robust description' of "Addressing the question", the answer is "If consensus agrees that the text was not attempting to address the question". If there are some people who think it was addressing the question, then it passes.

Remember, this is explicitly not a test of competence to stand, it's a very very very low bar on entry, to filter out people 'doing it for the lulz'. I suggest the essay question requirement because that can then be used by voters as part of their decision making too. --Barberio (talk) 16:41, 22 September 2008 (UTC)

Does User:Bishzilla write intelligible English? Haukur (talk) 16:52, 22 September 2008 (UTC)

Bishzilla writes very intelligible english for a monster. Also passes the hypothetical 1000 mainspace edits in the past year, so would be eligable to stand under that rule too! --Barberio (talk) 18:46, 22 September 2008 (UTC)

Leaving aside Barberio's hypothetical case of an editor who manages to produce FAs in only a handful of edits, in practice we can assume that anyone with less than 1,000 mainspace edits in 2008 isn't someone who is substantially active in the mainspace, to the extent necessary to understand matters with the kind of depth we need on the ArbCom. I feel that one of the major problems with the ArbCom is that arbitrators are, for the most part, not just out of touch with the community (as I argued in the section above), they're out of touch with the editing process and the practical needs of the encyclopedia as well. Even if they were highly active at one point, Wikipedia is constantly changing, and a high level of activity in past years isn't sufficient for an arbitrator. There may very well be some people who have exceeded 1,000 mainspace edits in 2008 but still aren't really all that active, due to the nature of their edits, but this measure will suffice to bar most insufficiently active candidates.

I don't agree with the baker/accountant analogy, either. Arbitrators need practical understanding of the problems of editing in order to arrive at intelligent solutions during arbitration cases, and the only way to get (and keep) that understanding is to consistently edit encyclopedia articles. In the case of bakers and accountants, they have clearly distinct tasks, and skills developed in one trade will not help in the other. With arbitration, the process of considering complex cases and reaching appropriate decisions requires substantial experience in the editing process—editing serves as ongoing education for arbitrators, and to me requiring 1,000 mainspace edits in 2008 for candidates is equivalent to expecting an accountant to have a high school diploma. Everyking (talk) 06:47, 22 September 2008 (UTC)

What you are trying to do here is to set candidacy criteria that will rectify perceived problems with the past and current ArbCom. It is not your place (or mine, or any other individual editor) to determine who is or is not a suitable candidate. The vote will select those candidates that the community as a whole feels are best suited to further the ArbCom's mandate; if consensus agrees with you that people without 1000 mainspace edits are unsuitable, then those candidates will not be elected. The entry criteria are intended solely to weed out the 'Lulz' candidates: people who nominate themselves purely for the fun of it, with no intention of discharging the office properly if they were elected. To do that we need to set the bar no higher than somewhere below which we can all agree that no one stands any hope of being elected. Your proposal simply does not meet that requirement, while the 'status quo' of 1000 edits total probably does. We're all open to suggestions as to how to improve the functioning of the ArbCom, but preselecting candidates simply isn't a workable one. (also)Happy‑melon 11:05, 22 September 2008 (UTC)

I concur with Happy-Melon, candidacy requirements must be minimal, and not intended to shape the outcome of the election, but only exclude spam candidacies. --Barberio (talk) 18:46, 22 September 2008 (UTC)

Update

Per [1], I've added another seat up for election. MBisanz ^talk 15:25, 23 September 2008 (UTC)

Timeline - Calling the question

We had (briefly) discussed an altered timeline, which ran something like:

Saturday 1 November 2008 - T-30 days - Editors must be registered and have 150 mainspace edits by 23:59 UTC on this date to vote.

Monday 3 November 2008 - T-28 days - Invitation for questions ("Think about what you'd ask/add it to this list if you want to ask everyone")

Monday 10 November 2008 - T-21 days - Nominations open (14 days)

Monday 17 November 2008 - T-14 days - Question pages opened/Questions transcluded/etc. - Question Phase Begins

Monday 24 November 2008 - T-07 days - Nominations close

Monday 24 November 2008 - T-07 days - Voting pages created/Discussion pages created/Quickvote Created

Monday 1 December 2008 - T-00 days - Voting Begins

Monday 15 December 2008 - T+14 days - Voting Closes, Vote Pages protected for 3 weeks, Votes Reviewed for socks

Friday 19 December 2008 - T+18 days - Final Vote totals/percentages/statistics confirmed

Monday 29 December 2008 - T+28 days - Jimbo Wales Certifies Election/Announces Winners/etc.

Thursday 1 January 2009 - T+31 days - New Arbitrators take office

This would incorporate the 21 day proposal from Wizardman, where there are two weeks for nominations and two weeks for questions, but they overlap - There are 7 days between the close of nominations and the opening of the election during which candidates can answer questions, even if they joined up late in the process. This also keeps everything on Mondays, for simplicity's sake. And, as before, we avoid major events on any of the holidays during this period. Jimbo announced winners on 26 December last year, which was a Wednesday; since it's a friday this year, I moved it to the following Monday - though, obviously, winners will be announced when they're announced, which is fine.

I think it's important to ensure that there is consensus on this item, particularly the date for suffrage (150 mainspace edits by 1 Nov), so I'm sort of calling the question. Thoughts? UltraExactZZ ^Claims ~ Evidence 17:55, 23 September 2008 (UTC)

I see nothing wrong with any of those dates. — Coren ^(talk) 18:30, 23 September 2008 (UTC)

Looks good to me, I support the 21 day proposal. Der Wohltemperierte Fuchs ^(talk) 18:48, 23 September 2008 (UTC)

Where do we put in the ratification vote for the policy changes? --Barberio (talk) 19:28, 23 September 2008 (UTC)

Barberio, Are you referring to a vote on whether to adjust the election procedures, or a vote on a proposal to adjust the Arbitration policy (such a vote being held parallel to this year's Election, as has been suggested in the past)? Anthøny ✉ 20:19, 23 September 2008 (UTC)

See above section Wikipedia talk:Arbitration Committee Elections December 2008#Policy changes from the RfC. These pretty much have community support already from the RfC, but some decisions between alternates need to be made, plus some may not get full support again after everyone's had a chance to settle down from the big discussion and problems ArbCom have had this year. The ratification on a couple are all but a formality, but it's best to follow precedent that changes to the arbitration policy need community voted on ratification. Some of the changes do change how the election process will work in the future. --Barberio (talk) 01:38, 24 September 2008 (UTC)

A link to the ratification vote page should probably be put in the quickvote page when it's created, for instance. (Incidently, I've not seen anyone else edit the vote page for it yet? Surely I can't have got it perfectly right on my own?) --Barberio (talk) 02:10, 24 September 2008 (UTC)

Let me tinker a bit. I'm thinking that we should have a "candidate" page for each of the policy subheadings, and then each such candidate page would lay out the proposals under that category. Each individual proposal would have its own voting subpage. You know, as I'm reading this, I can see questions coming up - it might be worthwhile to appoint an advocate for each proposal or set of proposals; they could serve as the "candidate". The alternative would be to redirect the question page to the Voting talk page for each proposal, since there will presumably be some discussion. UltraExactZZ ^Claims ~ Evidence 12:51, 24 September 2008 (UTC)

Don't forget there will need to be sufficient time between announcement and commencement to allow for identification to the Foundation. --bainer (talk) 01:29, 24 September 2008 (UTC)

Last year, winners were announced on the 26th, and I know of no problems with getting the proper identifications in place before the 1st. It might be simpler to remove that date, since it is not binding on Jimbo at all - and we would trust that he would approve the winners with sufficient time remaining in the year to identify themselves. The other option, if there is some delay, would be to seat the arbitrators only when they are confirmed to have properly identified themselves. Since new arbs don't hop in on old cases, the only issue would be if we got a new case in that first week of the year. UltraExactZZ ^Claims ~ Evidence 12:21, 24 September 2008 (UTC)

another bang of the drum :-)

I hope nobody minds me returning to a thread which has lay fallow further up the page for a while - my feeling is that this is of increasing importance, both as time passes, and as more arb.s resign... I believe we need to have a solid answer to the 'who is being elected' and 'how will the election winner be judged' questions.

I'd assume (and propose, and support) the straight forward election of 7 people to the vacant seats, with seats 6 and 7 filling the recently resigned posts, and hence being up for re-election sooner than their peers. By 'straight forward' I mean 'highest percentage of support' - and I really hope that this is uncontroversial enough to get broad approval well ahead of time :-) Privatemusings (talk) 03:11, 24 September 2008 (UTC)and feel free to edit the below mercilessly to aid clarity!

In fact, to that end, here's an 'ol wiki straw poll;

I support this system

• Privatemusings (talk) 03:11, 24 September 2008 (UTC)
• Strongly support. Everyking (talk) 07:34, 24 September 2008 (UTC)
• Of course. No idea why people value Jimbo's opinion over the opinion of the community, or indeed, their own opinion. Bizarre. naerii 12:42, 24 September 2008 (UTC)
• Support, trust the community. Wikipedia is not one man's plaything. DuncanHill (talk) 22:13, 25 September 2008 (UTC)

I have a question

go ahead :-)

I have further input

no worries... :-)

I oppose this proposal

• We leave it to Jimbo. MBisanz ^talk 03:12, 24 September 2008 (UTC)

What's this "we" crap. Please speak for yourself. --Duk 16:29, 25 September 2008 (UTC)

Nice to see you again Duk; my we have a way of bumping into each other alot; my statement remains that I oppose this proposal since we, the Wikipedia community, entrust the role of community founder to Jimbo Wales and therefore should leave such matters to him. MBisanz ^talk 16:37, 25 September 2008 (UTC)

• Nothing broken, nothing to fix. — Coren ^(talk) 03:30, 24 September 2008 (UTC)
• I trust Jimbo. Utan Vax (talk) 22:46, 24 September 2008 (UTC)

so do I :-) - in fact I'm not really certain that signatories to this section are necessarily even in tension with the above?! - this doesn't seem to me to be 'opposition' to the idea that the 'highest percentage of support' system is a bad one. I think signatories here are saying 'we're happy for Jimbo to make the call' - I'd pretty much agree, except that I'm asking for his input as to how he's gonna do that before votes are cast, which seems to me to be a good idea! cheers, Privatemusings (talk) 07:08, 25 September 2008 (UTC)

• I'd actually rather see "approval voting" which counts only supports, not neutrals or opposes. But I'm not sure this proposal will actually decide anything one way or the other. ++Lar: t/c 13:56, 25 September 2008 (UTC)

I (think) I agree actually, Lar (both about the approval votes, and the fact that this discussion may not be of particular influence) - but I still think it's worth having, and I think it would be a good thing to confirm this stuff ahead of time :-) Privatemusings (talk) 21:50, 25 September 2008 (UTC)

• Jimbo is generally going by the percentages, but I like the idea of somebody checking to make sure there is no gaming. Numbers sometimes don't tell the whole story. Jehochman ^Talk 14:30, 25 September 2008 (UTC)
• Oppose a removal of the layer to the system that Jimbo provides, until such times as the proposal can fill the void with a new check/balance. We really do need something to make sure we're not doing something stupid (as per Jonathan, above: numbers sometimes don't tell the whole story), and unless there's a better idea on the table, Jimbo more than suffices. (I suspect there will be very little playing around with the results this year by Jimbo, anyway.) Anthøny ✉ 20:05, 25 September 2008 (UTC)
• Oppose, because whether God-King or constitutional monarch we need a check on the "tyranny of the majority". The very last thing we need is to turn into one of those web communities that periodically has palace revolutions and changes its mission. Jimbo provides continuity and clarity of vision: this project is about the encyclopaedia, and the community is a secondary and subservient entity. As we have ever more articles and the community spends ever more of its time on internal politics, format wars and deciding exactly how many articles we need on each episode of The Simpsons, we need Jimbo more not less, to remind us that we should not seek to become more like those who imitate us only poorly. Guy (Help!) 21:25, 25 September 2008 (UTC)

what do you think of the suggestion below that the board could / should play such a role, Guy? :-) Privatemusings (talk) 21:48, 25 September 2008 (UTC)

Answering this seriously, I note that Jimbo was the author and/or a major proponent of the founding principles that still guide Wikipedia today. Not that you are not qualified to function as our godking and final authority, Privatemusings; :) but I think Jimbo is the man best qualified for the job. Sacking him now would simply be a lynching for lynching's sake. Anthøny ✉ 22:50, 30 September 2008 (UTC)

Oh, but I'm a big fan of the constitutional monarchy! - and mentioned this a bit further up the page! (besides, I'm not sure the tiara would match my thongs!) - I like the idea of having Jimbo formally appoint the arbs following an election, but maintain that it's probably 'best practice' for the system by which the appointments will be made to be clarified :-) - Interestingly, I'm getting more and more feedback that the system is in fact perfectly clear, and that Jimbo always picks the 'highest percentage of support', in which case my banging on could be resolved fairly simply... hopefully we can get confirmation soonish :-) Privatemusings (talk) 23:07, 30 September 2008 (UTC)

(edit conflict) Come on, Anthony! The existence of a king is sufficient reason to overthrow the king. I mean, you don't think people would dump tea into the harbor just because they disagreed with a minor proviso in the tax code. Come on man-- you're Scottish. You paint your face blue, I'll dress up like an native american and throw away some tea... If we can get somebody from the France to storm the Bastille and talk somebody from India into going on a hunger strike, and we got us a anti-monarchist shin-dig. :)

That said, this definitely isn't the time or place for that argument. :) --Alecmconroy (talk) 23:16, 30 September 2008 (UTC)

Not the time or place to 'decide' or 'bang the drum' for this

Quit it. Seriously. Next 'proposal' to change the number of people elected or function of the committee, I just remove from the page out of hand. We've had a recent lengthy RfC on various proposals to change arbitration policy, we worked out a set of changes that will be up for ratification vote. These kinds of changes are *MOST STRENUOUSLY NOT* to be decided by a group of self selected people running the election. --Barberio (talk) 04:05, 24 September 2008 (UTC)

(unhidden - really just to make the table of contents links at the top work, and the link I've dropped in to Jimbo's page - no biggie I hope)

my post may not have been clear enough, but I really do want to avoid conflating some of the more radical 'proposals' with what I hope is a simple request to answer what I see as a pretty basic question (sure, I asked two above, but let's kick off with just the one!); how will the election winners be judged? - I just see it as a bit... um.. unusual? to run an election process without confirming ahead of time how the victors will be determined! Re : MB and Coren - sincere apologies if the answer to my question is really really obvious (it wouldn't be the first time I've missed something like that!) - but could you possibly just indicate your understanding of how this will work? (or maybe asking for your understanding of how Jimbo judges the matter is a more appropriate question?) - thanks! best, Privatemusings (talk) 07:26, 24 September 2008 (UTC)

Agree that this page is perfectly appropriate for proposals and this should not be hidden. Everyking (talk) 07:34, 24 September 2008 (UTC)

I sort of agree, kind of. This page is perfectly appropriate for proposals on how we generate a list of editors, ranked by the percentage of editors who support them as taken from the total number of editors who support and oppose them. How that list is turned by Jimbo into new arbitrators is beyond the purview of this page, I think. If the community's will is unclear from the RfC, then I'm not sure what we can do here - though I seem to recall someone analyzing the last few elections to see how far Jimbo strayed from the results of the voting. UltraExactZZ ^Claims ~ Evidence 12:31, 24 September 2008 (UTC)

Any Change to Arbitration Policy itself has to be made in a very public fashion, with wide spread community consultation, as happened during the ArbCom RFC. So no, this page is not an appropriate place, because currently it's just being used by the people interested in setting up the election method.

The RfC was pretty widely advertised, and when it came down to it, people just weren't willing to support the complete removal of the Foundation from the process, and ultimately it's them who decide to accept arbitrator, Jimbo acts on their behalf by tradition.

If you're going to propose removal of Jimbo/Wikimedia from the election process, it's going to get tricky anyway as we don't have that kind of authority in the first place. And you'd probably have to go about getting yourself or like minded people elected to the Wikimedia Board. --Barberio (talk) 15:47, 24 September 2008 (UTC)

I'm under the impression that the Foundation aren't involved at all, Barb - but it's certainly worth checking :-) Privatemusings (talk) 05:45, 25 September 2008 (UTC)

Arguments against direct election?

Are there any arguments in favor of allowing Jimbo veto power over the community decisions in this election? The two votes presently in favor of letting Jimbo keep his prerogative seem to favor it simply cause it's "the way it's always been done". I'd like to know if anyone has a reasoned opinion about why it's better to let Jimbo make the final decision. Everyking (talk) 12:33, 24 September 2008 (UTC)

I like the last resort panic button. While it's vanishingly unlikely to happen, having someone who is generally considered sane get the last tweak may prevent a lot of problems. Remember that our "election" process is fraught with fraud and deception (because of the anonymity, and lack of accountability); while the fraud tends to go every which way and cancel out on average, it nonetheless exists. — Coren ^(talk) 12:50, 24 September 2008 (UTC)

Given the countless number of uncertainties, proven best perhaps by the PoetGuy saga that fooled all of us, I like the idea of Jimbo as a pressure valve. MBisanz ^talk 13:42, 24 September 2008 (UTC)

As mentioned above, it's not even something the Wikipedia Community has authority on it's own to change. For a variety of reasons, including some legal ones, the Wikimedia Foundation *must* have final veto over selection and appointment of Arbitrators. And it's really a board level decision to change how they do that. --Barberio (talk) 15:49, 24 September 2008 (UTC)

What legal reasons? I'm not aware of any. The foundation has ultimate veto over anything anyway, by virtue of they own the site and can take Office actions. But, I'm not aware of any law requirements requiring them to have say over selection of the arbitration committee. ⇒SWATJester ^{Son of the Defender} 08:00, 25 September 2008 (UTC)

(per above) - my understanding is that the Foundation isn't involved at all in the appointment of Arbitrators... this is certainly worth checking etc. - might it be possible to link to some further info. on how this works? (I'll ask around too...) cheers, Privatemusings (talk) 05:47, 25 September 2008 (UTC)

There are at least a few strong reasons why this serves the project well.

1. It isn't abused. A negative, but an important one. In the last 3 elections, the community's view was that which was carried out. (Minor exception - January 2006 to reappoint two existing arbitrators as positions 9 and 10, or some such.) So the odds are good that unless there is a serious problem, it will be that way in future.
2. The community is only part of the recipe. In brief, the mandate of Wikipedia is a community to write an open encyclopedia. If the community goes off the rails, now or future, or endorses matters that do not support that aim, then Jimbo (or some outside overriding mandate) can and should step in. This isn't an experimental democracy or a game of life, it's a community given that control with the good faith understanding it will use it well. A dead man's switch exists in case that ever were to change. The community may at some hypothetical time fragment into factions that do not all have the best interest of the project at heart, or stonewll each other, or some other matter may arise in the next 20 years.

It may not please some who are communal power advocates. But the position now is what it was then - the right to leave, and the right to fork. While we all stay, the project has directions and one person to ensure we don't forget it in future is a sensible final precaution given volatile human nature. FT2 ^{(Talk | email)} 08:30, 25 September 2008 (UTC)

While Jimbo has always made good decisions in previous elections, I'm not entirely comfortable with giving that power to one man. Wouldn't it be better to give the whole board the power? It doesn't need to involve lots of bureaucracy - just declare that the winners of the election become arbs and the board has the power to object within a month. In the vast majority of cases, the board will do nothing, on occasion a member of the board may have concerns and can call a board meeting to discuss them. The foundation already has some involvement in the process, since arbs are almost always checkusers and therefore have to identify themselves to the foundation. Giving the foundation (via the board) the power to stop someone because an arb for other reasons is a natural extension of this. If the board believe this would be inpractical, they could delegate the power to a committee of board members, or even just to an individual member (Jimbo, say), but acting as a representative of the board and accountable to them, rather than acting on their own volition. --Tango (talk) 14:31, 25 September 2008 (UTC)

This seems reasonable to me. Aside from the issue of giving the power to one person, I feel there is an issue of impartiality—Jimbo actively takes positions in enwiki affairs, he exercises powers occasionally, and it is generally understood that there is a segment of the community favored by Jimbo and a segment that is not. I would feel much more comfortable if we instead reserved veto power for the foundation, and appointments were deemed automatic by community vote unless the foundation said something. I don't think the foundation would alter the decisions except in truly exceptional cases, and I don't think there be would much risk that they would play politics with the process. Everyking (talk) 15:09, 25 September 2008 (UTC)

We (community) actually never did give that to one man in the first place. That's a fairly major misconception. Rather, historically, we (this community, me, you) were given the rights we have on enwiki, and over time a few more. But formally appointing and removing arbitrators and oversight of the committee as a whole, was not one of those. The Foundation aren't tasked with overseeing individual communities in detail either. On this wiki, that's Jimbo's final say-so. FT2 ^{(Talk | email)} 17:54, 25 September 2008 (UTC)

History isn't relevant - what the community says, goes. If the community decide they don't like Jimbo's ArbCom we can just make a new one and ignore any rulings by his. There is no actual need for an officially sanctioned ArbCom apart from CU, any group that admins are willing for enforce the rulings of has the same power as ArbCom. I doubt we'll ever get to the position of having to do that, since the foundation in general and Jimbo in particular aren't stupid and won't try and fight the community if there's a clear consensus, but if it comes down to it, the community would win. --Tango (talk) 18:14, 25 September 2008 (UTC)

It depends on what you mean by 'win'. If a user disobeys the arbcom, they will be sanctioned (probably blocked) by an admin, or in extremis by an ArbCom member. If an admin reverses that block, they can be desysopped by the stewards. If the stewards refuse to obey the arbcom, a developer (employed directly by the foundation) can write the rights changes directly into the database tables. No one, no matter how large a group or however well-placed, can exist on wikimedia wikis in defiance of the Foundation and on en.wiki that authority is devolved to the ArbCom. As FT2 says, that authority was devolved from above, not instituted from below. The only thing the ArbCom can't do is force you to stay - the right to leave and the right to fork are the only absolutely inviolate rules that no one has the power to change. Of course, the situation where the ArbCom and Foundation reign supreme over an empty wiki hardly counts as a 'victory', for anyone involved. But the belief that the ArbCom are somehow empowered by the community, and hence beholden to them, is simply not the case. The community exists in symbiosis with the Foundation: they provide the technical resources and administrative framework, and we provide the free content that their charter requires. The only sense in which one is dependent on the other is that without one, the other could not survive. And that cuts both ways. Happy‑melon 14:44, 26 September 2008 (UTC)

If there were a consensus among admins to ignore an ArbCom ruling, the stewards would end up desysopping the lot of them and then the site would dissolve into chaos. At which point all the admins would be resysopped by a desperate foundation that can't cope with all the complaints from people unhappy with their article saying they're gay. That's a win for the community. It will never come to that, though, it would require immense stupidity from the foundation and I'm confident they aren't immensely stupid. Where the power comes from in a de jure sense is pretty irrelevant, the community has the de facto power. --Tango (talk) 15:58, 26 September 2008 (UTC)

Desysopping the entire admin body would simply destroy the Community, and not something that would ever happen - even if the Foundation adopted a course of action so wildly at odds with the Community to drive 99% of en.wiki admins to rebellion, that would still leave 16 of them 'loyal' to the Foundation. They wouldn't throw the baby out with the bathwater - they'd know how badly they needed those people who were prepared to go along with them. But the point is the symbiosis: the Foundation and the Community have to work together even though our interests are sometimes misaligned. The ArbCom and Jimbo are the main intermediaries who have to negotiate the compromise between the two, as well as dealing with all sorts of other crap from both ends. The whole situation is only really kept stable by a healthy mutual respect; saying that anyone has 'power' over anyone else is disingenuous. Happy‑melon 17:44, 26 September 2008 (UTC)

Arbitrators have to identify to the Foundation whether or not they are checkusers as access to the mailing list means they often have access to CU data. Sam Korn ^(smoddy) 16:09, 25 September 2008 (UTC)

Policy Change Referendum Voting Thing

OK, I've attempted to create what would be the "Candidate Statement" page for the policy proposals. The draft may be found at Wikipedia:Arbitration Committee Elections December 2008/Policy Changes/Election. It includes the first three proposals, and groups all of the myriad options under proposal two (term of office, tranches, etc) into one item. Each proposal would, in theory, have a discussion page (which takes the form of the Question page for candidates). Eventually, each would also have a voting page, where the supports and opposes would go. The amboxes make everything look hinky, so please correct my formatting - but, ideally, each of the four sets of proposals would get a page like this, which would be transcluded onto the Policy Changes page. I included space for the link to discussion on the proposal, including the support from the RfC to move it forward, as there is no set of userlink-style resources to evaluate these proposals. Please review. UltraExactZZ ^Claims ~ Evidence 14:18, 26 September 2008 (UTC)

I think this is needlessly complicating matters? Separate pages for discussion on the options, that sounds like a good idea, but separate pages for each item to vote on, makes it more effort, and would turn people off voting. It seems a lot easier to have one page for conducing the vote on the policy changes. --Barberio (talk) 19:37, 26 September 2008 (UTC)

I've put in some redlinks to pages and talk pages for the major groups of proposals onto the page at Wikipedia:Arbitration Committee Elections December 2008/Policy Changes. This will keep the discussion and advocacy of the individual policies off in their own pages, while putting the actual voting in one page. --Barberio (talk) 19:59, 26 September 2008 (UTC)

Edits and a rename made to the page for elections. Most of the election changes came from the RFC, and I've linked to the section there rather than a diff, so discussion that took place can be viewed. Since all the other sections only have proposals from the RfC, this is what I'll do for those ones as well.

If you're going to create the other pages, Wikipedia:Requests for comment/Arbitration Committee/Summary lists where those policies came from, but you should link to the same sections on [[Wikipedia:Requests for comment not the summary page to show the discussion that took place.

Further discussion of the proposals can now take place on the talk pages of the subarticles, which are also red-linked at the moment on the voting page. --Barberio (talk) 20:31, 26 September 2008 (UTC)

My understanding is that these proposals would come to a vote in the same manner as Arbitration Committee candidates, and each candidate gets a voting page, for a variety of reasons. In my mind, either these are serious proposals that deserve the full dog-and-pony show of an arbcom election (The most prominent community input we get, I believe), or they're not and a straw poll would suffice. We can't do it both ways. UltraExactZZ ^Claims ~ Evidence 20:57, 26 September 2008 (UTC)

I think that the voting should be consolidated on one page until such a time as it needs to be split for size purposes. Discussion can be separated, but as Barberio says, fragmenting the voting too much will probably put some people off, which is not what we want. Happy‑melon 21:24, 26 September 2008 (UTC)

There's value to that, but it would be difficult to justify moving votes from one page to another during the voting period itself. Plus, there are structural reasons why I prefer to have each vote on a separate page. It helps to check for duplicate votes, for one. It also would help to keep the voting clear - there's only one place to support or oppose on each page, making confusion impossible. Third, if someone votes without being able (i.e. they lack 150 mainspace edits on 1 November), it's easy to check their Wikipedia-space edits for other votes on other pages; this would be trickier (though not impossible) on a single page. We could transclude the voting pages onto a single page, as we do with the candidate statement pages; this way, everything would be in one place (as with RfA, for example), but the votes themselves would go to subpages. UltraExactZZ ^Claims ~ Evidence 03:00, 27 September 2008 (UTC)

If, by transclusion, you can make it so someone can put their own votes down by only having to read through one page, then yes.

If they have to read through multiple pages in order to make a vote, then no. --Barberio (talk) 15:34, 27 September 2008 (UTC)

OK, here's what I would suggest, then. Each proposal has a subpage entitled "Vote", just like the candidates. That page transcludes the proposal itself (in lieu of the candidate statement), and then has headings for "Support" and "Oppose". Editors then add their names where they wish to vote. Each of these pages is then transcluded to a main Voting page, where all proposals (or each of the four groupings, or whatever) are listed. It would be similar to RfA or AfD; all of the active RfAs are on one page (WP:RFA), but editing one to comment brings you to the RfA itself, which is just one candidate. We could <noinclude> header instructions and links, so that someone could read the full list, vote on one item, and then click to go back to the full list. It sounds much more complicated than it is. We had over 8000 distinct votes last election, so we need subpages; but this would keep the proposals together. UltraExactZZ ^Claims ~ Evidence 18:31, 27 September 2008 (UTC)

I could live with that, as long as we include some clear instructions. Happy‑melon 10:03, 28 September 2008 (UTC)

Also, is there a reason we couldn't open voting for the ratification on november 1st? The proposals aren't likely to change after then, and people can always come back and change their vote until we close voting 15 December. I'll go ahead and change the 'policy changes' page to say that voting is open from November 1st and closes 15th December? --Barberio (talk) 20:53, 26 September 2008 (UTC)

I think that would cause confusion - some voting starts on one date, some on another? Again, if these proposals are part of the Arbcom election process, then they need to be fully part of it - which means two weeks of heavily-advertised voting in December, alongside the candidates for arbcom. UltraExactZZ ^Claims ~ Evidence 20:59, 26 September 2008 (UTC)

I agree with Ultraexactzz - full integration with the main ArbCom elections will give this poll the greatest possible authority. Happy‑melon 21:24, 26 September 2008 (UTC)

Fair enough, but in that case should we remove, or lower, the 40 support requirement? This is something I carried over from previous ratifications which had much longer time periods. --Barberio (talk) 21:30, 26 September 2008 (UTC)

You'd get 40 votes, no doubt. Hell, the vast majority of RFAs get that, and this would be advertised. If you got any less than that, it wouldn't be much of a consensus. Just like an RFC that most of the community didn't take part in isn't consensus. Sam Korn ^(smoddy) 22:02, 26 September 2008 (UTC)

To reply to your final comment. The RFC was advertised pretty much everywhere suitable on the wiki, including the Watchfor notices. Anyone who wanted to get involved in the debate almost certainly knew about it, and got involved. And I think it was probably the RfC with the highest rate of community involvement that we've seen. If I thought that on that alone I could say 'The community fully supports these changes', I would have, instead we've brought all the proposals that were accepted as having significant support up for a ratification vote. I'm not sure what you see wrong with this process? --Barberio (talk) 22:13, 26 September 2008 (UTC)

And I think in future, yes, the requirement for putting arbitration policy changes up for ratification is having had an RfC showing great support that these changes should be made. --Barberio (talk) 22:15, 26 September 2008 (UTC)

Well, I didn't take part in the RFC because I was too busy to take part helpfully. It wasn't a matter of it being unseen. I fail to see how the proposal "we should have two-year terms with biannual elections" can only be voted on as one piece when it is clearly two issues. Incidentally, you know you are innovating in policy here? Not that there's anything at all wrong with that, but this is a new way of changing policy. Sam Korn ^(smoddy) 22:21, 26 September 2008 (UTC)

Yes, it's a new way of changing the arbitration policy. And yes, it's my innovation. The reason for that is because we have a lot of people calling for changes to arbitration policy including arbitration committee members themselves, and *someone* had to come up with a way to change arbitration policy that demonstrates legitimate community support, and no one else has done it yet. --Barberio (talk) 22:38, 26 September 2008 (UTC)

"Not that there's anything at all wrong with that" Sam Korn ^(smoddy) 23:13, 26 September 2008 (UTC)

Another option for voting?

Forgive me if this has been proposed before, but I would like to propose another system. Instead of simply going off of percentage, why don't we subtract oppose votes from support votes? Then take the percentages and add x percent of votes to each total. Thoughts? Geoff Plourde (talk) 15:42, 29 September 2008 (UTC)

Too complicated - the adage that every equation halves the sale of a book applies equally to wiki. More pertinently, what advantage would it offer over the current system? Can you give examples of past votes where your proposal would have been superior? Happy‑melon 16:32, 29 September 2008 (UTC)

Interesting proposal, but one I don't think in practice would attract any improvements in the system, if adopted.
Anthøny ✉ 13:29, 4 October 2008 (UTC)

Note from FT2

Any user who is considering running for the Arbitration Committee, and is likely to be a plausible candidate for the community, and who wants to discuss Arbcom or any related matters in private to be better prepared (for example, to gain an understanding of some aspects, or check some assumptions and facts before or after deciding), is welcome to contact me for the purpose.

FT2 ^{(Talk | email)} 18:40, 11 October 2008 (UTC)

(This is cross-posted from User talk:MBisanz/ACE2008. -- how do you turn this on 23:21, 11 October 2008 (UTC))

Questions: which model to follow?

Wikipedia:Arbitration Committee Elections December 2008/Candidate statements/Questions/For all.
I have developed a possible model to follow, with regards to a "questions portal page", as suggested on WP:ACE2008.

The intention of that page is for questions to be asked once, to which all candidates can respond on the one location. Compare, m:Board_elections/2008/Candidates/Questions/1. This "one questions location for all candidates" page is intended to run parallel to the standard individual-questions-page-for-all-candidates system we've used in every ArbComm election to date.

Thoughts? Anthøny (talk) 10:32, 15 October 2008 (UTC)

Parallel? Why have both? I certainly think one system would be better. The latter, preferably. Though this may be a problem if someone wants to ask a question that doesn't have anything to do with the other candidates. -- how do you turn this on 11:29, 15 October 2008 (UTC)

One page where a question is asked and all candidates answer, is highly attractive, and I have thought hard on this one, and realized that I would have a very serious concern about an unintended side effect. I find myself very concerned that placing all answers to a question side by side, will have an unintentional and subtle "chilling" effect on the breadth and individuality of answers received.

In the 2007 elections, each person faced their own questions. Their style of reply (whether detail or general, packed with diffs or containing few, long or short, out on a limb or mainstream) was theirs alone. Initially I'm sure many candidates checked what others were saying, but you can't do that so easily all the time, and you're writing your page, in your style. The reader read 30 or 40 answers, and got a good sense of that users style, attitudes, the themes they nominated upon. All of that would be in the new format. But the much closer "side by side" method will disturbingly encourage much more "following the pack". I wrote reams, it was my style - if I was side by side with 20 users writing 10 - 15 lines, for question after question, would I not have been tempted to change my style more? I think so, it's human nature for a lot of people; even those who are very independent might have slightly modified their style, perhaps without realizing it, to some style closer to the "page average". Or perhaps read more commonly others views and made it a "head to head" where one tries to appeal to all the users ones opponents do, plus a few more, rather than simply state one's own view.

It's a concern for me, as I see in it the seeds of moving from "what a user is like" to "political" or "cause less (or no) offense" style answers.

What might be nicer is maybe a list of central questions, which each candidate answers on their own page. The Q&A pages for candidates would then have two sections - questions being asked to all candidates, and questions just for that user:

(user name)'s question page

== Questions specific to (user name) ==

=== 1. (question) ==
=== 2. (question) ==
=== 3. (question) ==

FT2 ^{(Talk | email)} 16:19, 15 October 2008 (UTC)

I like this idea, and understand what you're saying. For example, to use you as an example, in the 2007 elections, your answers were probably the longest out of everyone's. I'm not sure how well such answers could go together. I'm not saying this is the case at all, but it could give different users advantages and disadvantages unintentionally. As you say, every candidate is unique. I do agree a "set" amount of questions could be answered on one page (for example, answering what user rights/privileges they have). But I also agree, that some of the more "tough" questions asked of candidates would probably be better separately. -- how do you turn this on 16:25, 15 October 2008 (UTC)

I only just saw this when I saw this edit on my watchlist. Hopefully I won't be seen as hopelessly compromised.

My concern with mandating that all questions be asked of all users is that there won't be the opportunity to scrutinise individual users as well as should be the case. I understand that the questioning process can be abused. I am concerned, however, that adopting this system will have unhappy consequences.

Sam Korn ^(smoddy) 16:26, 15 October 2008 (UTC)

If you honestly think that your question is important enough to be asked of all candidates you can go to the effort of adding to every candidate page.Geni 18:39, 16 October 2008 (UTC)

There's something to be said for that, granted, but why make it harder? This will also let editors thinking about the process - and the committee's composition - before seeing and evaluating nominees, which has its own appeal. I'll add that I'm curious to see how different candidates answer the same question; having some of those same questions provided by template would ensure that the questions are identical, emphasizing those differences. UltraExactZZ ^Claims ~ Evidence 13:27, 27 October 2008 (UTC)

Identification

There is a good chance that any person wishing to sit for Arbcom may need to identify to WMF. Of course the majority of Arbitrators do, but at present it is not a requirement that a user will do so. I think this is unavoidable, and worth raising prior to nominations. I've raised it on arbcom-l also.

Access to non-public data is governed by the Foundation's Access to nonpublic data policy. It contains the following provisions:

The intent of this policy is to ensure that volunteers who have access to nonpublic data covered by the Wikimedia Foundation privacy policy are personally and legally accountable.

1. Only persons whose identity is known to the Wikimedia Foundation shall be permitted to have access to any nonpublic data or other nonpublic information produced, collected, or otherwise held by the Wikimedia Foundation [...which is covered by...] the Wikimedia Privacy Policy.

2. Any volunteer who is chosen by any community process to be granted access rights to restricted data shall not be granted that access until that volunteer has satisfactorily identified himself or herself to the Foundation...

It is already agreed (WT:ACE2007#Ruling on age limit) that users standing must meet the age requirement for non-public data. Rationale for the belief that identification is unavoidable is in that thread, and also as follows:

1. A user appointed to the Committee, even if they do not request Checkuser or Oversight, will join the arbcom mailing lists. This automatically gives "access to... nonpublic data or other nonpublic information produced, collected, or otherwise held by the Wikimedia Foundation" even if they do nothing else. (See also [2].)

2. A user who avoided arbcom mailing lists, and is not party to dialog or emails containing non-public or personal data about users, would effectively be so "out of the loop" (based on experience of arbcom activities) that they could not substantively do the role. Even if not every case involves such data, it would only take one or two significant points to involve such matters, and the Arbitrator would probably be unable to fully review the whole case as his/her colleagues could.

   (I haven't formally counted what proportion of activity and threads include at least one email with such discussion but I'd hazard a guess roughly as follows: 30 - 50%, rising sharply for the more significant and high profile cases, and for appeals. Enough.)

3. The presence of an Arbitrator who was unable to be sent emails that touched on, or contained quotes of, non-public information, would be highly disruptive and difficult for other arbitrators, who must discuss these with their colleagues on a daily basis, usually via their mailing lists.

FT2 ^{(Talk | email)} 16:31, 16 October 2008 (UTC)

• I have added "be willing to identify to the WMF if appointed" as a criterion.[3] Anthøny (talk) 16:44, 16 October 2008 (UTC)

I really think it's a bit extreme to require all candidates to provide identification, but I think it's more or less accepted that it may be necessary to provide ID to the WMF after being elected, but before being appointed. If nothing else, processing and verifying the ID credentials of 50+ hopefuls would probably take up a whole bunch of time for WMF volunteers that could be more productively spent on something else. Lankiveil ^{(speak to me)} 04:58, 18 October 2008 (UTC).

I asked Cary and he does not care about doing it before, during, or after the election. MBisanz ^talk 05:00, 18 October 2008 (UTC)

Yeah, but why waste his time doing ID for the obviously nonviable candidates that are going to pop up? I've no idea how much work is involved, but it seems silly to do all that work (and not just for Cary, but for the candidate who has to find and send the appropriate authorisation), not to mention the potential for privacy violations, for candidates who stand no chance at all of being elected. Lankiveil ^{(speak to me)} 05:16, 18 October 2008 (UTC).

My thought would be that ineligible candidate User:X runs, wins, refuses to identify, is not seated on arbcom, spends the next year reminding everyone how he was denied his seat because the Foundation is ageist/etc. To identify, all one must do is email/fax a copy of their driver's license/passport to the Foundation, so there is not that much paperwork involved. MBisanz ^talk 05:20, 18 October 2008 (UTC)

Sorry, when I say "nonviable" I'm talking about people like User:X who may just barely meet whatever entry criteria we set, but who have a profile or history that will more or less prevent them ever being elected. Say, a 21 year old with 500 mostly-automated edits, and a history of lighting more fires than putting them out. This person, although technically eligible, will never ever be elected, so what's the point in wasting the time validating their identity and age?

Now, I'm fine with requiring a user to disclose their identity to WMF in order to be elected, but it seems like it would be a lot of fiddly mucking about to do "a little bit" of paperwork for fifty people, when we can possibly get away with doing it for under ten. That's why I think we should require identification as FT2 has suggested, but we stipulate that a person is not required to do so until after they have been successfully elected (but probably prior to appointment by Jimbo).

If anyone refuses to identify themselves for any reason, we can then always point at this notice and say "Well, you knew it was a condition of being elected". Lankiveil ^{(speak to me)} 05:42, 18 October 2008 (UTC).

Okey, works for me , but I'm reserving the right to call shenanigans if someone tries to lawyer the identification process after the election. MBisanz ^talk 08:22, 18 October 2008 (UTC)

There is the point, however, that since identification is a si ne qua non requirement of being actually appointed, someone who does not or cannot do so at the time of nomination is a huge waste of everybody's time in the whole voting process. I think the prerequisite identification would be a valuable filter. — Coren ^(talk) 19:15, 18 October 2008 (UTC)

Proposal: Identification as a prerequisite

[Forked into separate section for navigation.] Quite so. Additionally: the only argument against pre-identification is that it's a waste of time to have every candidate identify. Cary has already stated that he's fine with folks identifying before; from that point of view, that rationalisation is somewhat moot. I move to make pre-identification a prerequisite for running this year right here and now. Anthøny (talk) 19:27, 18 October 2008 (UTC)

I have created {{ACEidentity}} for the purposes of noting which candidates have identified, if this goes ahead.

Identification:     Candidate has identified.

Identification:     Candidate has not identified.

Identification:     Identification is in progress.

Example, Wikipedia:Arbitration Committee Elections December 2008/Candidate statements/Example. Anthøny (talk) 19:59, 18 October 2008 (UTC)

Given all the concerns about the privacy of identities, I would think that we would want as few disclosures as possible-- i.e. identifications AFTER the election as a final step before actual appointment. Most of the candidates aren't going to be appointed to arbcom-- why make them all identify? Just make sure they all know they'll have to before they could serve. --Alecmconroy (talk) 20:04, 18 October 2008 (UTC)

Hold on - where is Cary's post on this subject? Where is the Foundation's description of what information will be collected, how it will be kept secure, how long it will be kept, and how it will be destroyed if no longer required? This sort of stuff isn't covered in the Foundation's privacy policy. Unlike candidates for the WMF Board, arbitrators have no fiduciary responsibility mandated by law. Let's not kid ourselves, privacy is a very big deal on this project, and this is moving the goalposts a very long way. Note that I have changed the title to include the word "Proposal". Risker (talk) 20:06, 18 October 2008 (UTC)

MBisanz privately contacted Bastique. If Cary could be asked to pop by here and confirm that he is indeed willing to accept identifications from candidates pre-Election, it would be all the better. Anthøny (talk) 20:56, 18 October 2008 (UTC)

Nevertheless, if you nominate yourself for the ArbCom you must be willing to identify yourself and trust that the Foundation isn't going to be misusing that identification. Otherwise, you obviously have no intention of actually serving if elected (given that those would be requirements). Yes, it means that you need to make an affirmative step when you nominate yourself— I would expect this to weed out frivolous candidacies if nothing else.

If you don't trust the Foundation with your identity before the election, something which is a legitimate concern, then what would make you suddenly trust it after the election? — Coren ^(talk) 20:30, 18 October 2008 (UTC)

Actually, with Arbitrators gaining Oversight rights, and most having CheckUser right, identification is required for Arbitrators. Since data is apparently discussed on the list, it would be nonsensical for someone to run who wasn't willing to identify, or who was under 18. -- How do you turn this on (talk) 20:32, 18 October 2008 (UTC)

It wouldn't be the end of the world, I don't know that any candidate would have a huge problem with it, but it's just good practice to not collect information you don't need. The easiest way to guarantee that something doesn't get stolen is to make sure you never possess it in the first place.

As always happens, some candidates are going to find out that, although they weren't previously aware of it, they are actually extremely unpopular, or at least untrusted. Those with hurt feelings will probably will regret having revealed their identities, and if those people are outed later, they may allege a leak from within the foundation.

In short, we could do five times the amount of work, and all we'd get for our trouble would be potential problems. --Alecmconroy (talk) 20:39, 18 October 2008 (UTC)

Does it actually matter if the identification is received before or after? As long as it's received some time between nominating, and say, a week after the new Arbitrators are announced, it shouldn't be an issue, should it? As someone said, it's a waste of time dealing with IDs for people who run but aren't successful. -- How do you turn this on (talk) 21:52, 18 October 2008 (UTC)

And it's a waste of time dealing with election for people who cannot or will not identify themselves to the foundation. The only question are, in practice, (a) How likely is it that someone who gets pick then does not identify themself, and (b) how many candidates will rethink their participation if they are presented with the requirement to identify beforehand. For the record, I think (b) is a positive, not a caveat. — Coren ^(talk) 21:57, 18 October 2008 (UTC)

Identification simply says that person is over 18. No other personal information is ever revealed. I personally don't see what the problem with identifying beforehand is, but some people seem to. -- How do you turn this on (talk) 22:00, 18 October 2008 (UTC)

Identification includes providing evidence of identity, not just age. Cary is quite rightly not concerned about Arbcom elections, which is why he doesn't specifically comment. His requirement will be Foundation's policy compliance. Foundation policies include a requirement that any user with access to WMF non-public data (so defined) must identify, or not be given that access.

My observation is that access to non-public information is integral to Arbitration Committee work. The norms and process for that identification would be the same as for any user, who requests access to a project role, that involves access to data covered by the WMF's non-public data policy.

Basis of observation, and background to this: Arbitration Committee ruling on age limit, WT:ACE2007#Ruling on age limit and WT:ACE2008#Identification. FT2 ^{(Talk | email)} 22:20, 18 October 2008 (UTC)

Sorry; I might not have been clear. I meant that some people may have been wary of identifying because of fears of personal data being "leaked", or their identity being public knowledge. The only thing that is public knowledge is the fact the person is over the age of 18. It is, of course, essential that ArbCom members identify themselves, due to the nature of their work. The point of this discussion is determining when to do so: before voting, or after once the new Arbitrators are announced. -- How do you turn this on (talk) 22:26, 18 October 2008 (UTC)

I like Anthony's template, given that identification for me was as simple as taking a picture of my driver's license with my digital camera and emailing it, I'm not seeing the time strain either way. Given the deceptions we've had over the last year with PoetGuy and Archtransit (as recently as yesterday at commons with Archtransit), it is a nice, tiny, extra safeguard to have everyone identified before/during the election. Given that the pre-election stage is nearly a month long, that should be plenty of time. Risker, the new privacy policy was released this week, page 6 of wmf:Image:Privacy Policy 10.14.08.odt - NeoOffice Writer.pdf describes the circumstances under which the Foundation (Cary) could disclose the information. Basically it is the same level of protection required by CheckUsers, but since Cary's job is on the line (as opposed to the volunteer checkusers), it is a pretty good protection. Now if you review the wmf:Access_to_nonpublic_data_policy document, it clearly says

Only persons whose identity is known to the Wikimedia Foundation shall be permitted to have access to any nonpublic data or other nonpublic information produced, collected, or otherwise held by the Wikimedia Foundation, where that data or other information is restricted from public disclosure by the Wikimedia Privacy Policy.

It goes on to say that the type of identifying data required is

...which may include proof that such user is at least 18 and explicitly over the age at which they are capable to act without the consent of their parent in the jurisdiction in which they reside.

So I'm not seeing how requiring pre-identification from someone who actually intends to serve an an arbitrator is that much different than what we've always done. MBisanz ^talk 22:23, 18 October 2008 (UTC)

I fully agree that identification is required under the Foundation's policy. It would be quite impossible to do arbitration work without access to private data. There doesn't seem to be any way round this in my opinion, and that's fine by me.

I'm not convinced, however, that identification is useful before we have chosen our arbitrators. It seems like a lot of unnecessary work both for the candidates and for Cary. I think the situation that would make it worth it is a pretty remote possibility.

Sam Korn ^(smoddy) 22:29, 18 October 2008 (UTC)

A user who was appointed then played round with identifying might well find the response was withdrawal of invitation, rather than playing along. Arbitrators are expected to be above that, and to have considered and checked matters if they had concerns. FT2 ^{(Talk | email)} 22:32, 18 October 2008 (UTC)

I'm assuming any user who got the level of support needed for Arbitrator-ship would be the kind who wouldn't do that; I expect they'd be fully prepared, and willing to work with the system, not against it. -- How do you turn this on (talk) 22:34, 18 October 2008 (UTC)

Sam, I don't mind either way, before or after the election. But given traveling and holiday schedules, it might be a nice option that people can identify before if they want and that Athony's template can keep track of it all. Not required before, just a "By Jan 1 (?) you should have identified" sort of thing. MBisanz ^talk 22:51, 18 October 2008 (UTC)

I think that identification beforehand is a good thing (disclaimer: I made no secret of my intent to run, and I obviously have no problem with this), because:

• It ensures that anyone who runs qualifies for the position should they be selected;
• it requires a positive act, possibly weeding out candidacies "for the heck of it" which distract the voters from evaluating the other candidates (this has been observed in previous elections, but is admittedly not a major problem); and
• it significantly reduces the probability of attempts to game the system.

In all fairness, there are drawbacks as well:

• It identifies people who have not been selected to serve to the Foundation who otherwise wouldn't have had to;
• it may impose a dissuasive barrier to some less confident candidates; and
• there is the legitimate (albeit not necessarily justified) fear that this information, which is very private, is then lost or misused.

I think that the workload to identify the candidates is a red herring; not only has Carey already confirmed that it is not a problem in itself. So the end result is that the primary factor is that some people will have their identities known to the Foundation who otherwise would not; but if they were sincerely being on the ArbCom, then they were necessarily prepared to have that information known. Given the fact that the Foundation takes privacy very seriously (as has been amply and regularly demonstrated in the past) then I can't think of a reason why we'd not streamline the election process by starting the identification at nomination time. — Coren ^(talk) 23:51, 18 October 2008 (UTC)

Indeed. All in all, I think requiring candidates to identify pre-Election is a net-positive. Adjusting the election procedures to ask candidates to quickly identify is a good move. Anthøny (talk) 00:04, 19 October 2008 (UTC)

It is all very lovely that MBisanz and Cary had a private chat. Cary needs to come here and say the same thing, and also tell people exactly and precisely what the Foundation's policy is with respect to requiring personal information, and what they do with it. Not one person commmenting in this thread, including present and former arbitrators, is qualified to speak on behalf of the Wikimedia Foundation on this point. As self-identification is not explicitly discussed in the Foundation's privacy policy, this is not a minor point. There is zero net positive in pre-election identification. I will point out that in our recent Checkuser appointments, none of those who self-nominated were requested to self-identify as part of their self-nomination; this is clear based on the rights logs, and at least one checkuser confirming that he was not required to provide personal identification until after the offer of appointment.

I work with huge amounts of personal private information on a daily basis, and such a cavalier attitude toward the collection, use, retention and destruction of such information in a project where personal privacy has been a major issue just leaves me shaking my head. Risker (talk) 00:23, 19 October 2008 (UTC)

Update: I've asked Cary Bass to comment on this issue, both specifically and in a general sense, because of course there is no doubt that successful candidates would have to identify themselves to the Foundation. My request to him is here on his user page. We have to be reasonable and understand that this question relates to his work as the Foundation's Volunteer Coordinator, and because it's a "work" question we should not expect him to answer it outside of his usual business hours. I hope for his sake that doesn't include Saturday evening. Risker (talk) 00:56, 19 October 2008 (UTC)

I'm not sure why any conversation I had on IRC with someone who asked me some questions is even mentioned in this discussion, and I'd certainly like it if people stop referencing any conversation I might have had with them. It's kind of creepy to learn of your own discussions on IRC days later on the wiki.

There are posts ad nauseum either on mailing lists or on meta about the retention of data as well as the method of submission and what happens to the data after identification. Members of Arbcom and others can certainly repeat what's already been said by the Foundation about the issue. Asking me to continue to repeat the same information again and again takes my time away from other valuable resources. As I'm certain there are literally hundreds of people who watch this page, someone can find the information you're looking for, without demanding my time to make the same "official statement" over and over.

I would like to add that the insinuation that any staff member of the Foundation that is delegated to perform the task of identification might leak anyone's personal information is baseless and absurd. It is baseless because there's no evidence, in all the identifications that have been made (See m:Identification noticeboard for a list), that anyone's personal information has been leaked. It is absurd to even consider that I might be "personally motivated" to leak someone's information; in my capacity at the foundation I have handled private information of a wide variety of individuals in a wide variety of matters, many of whom did not have ideal relationships with myself or the foundation. Even if I had personal feelings about any of the candidates, it certainly would not affect the way I handle their information. Bastique ^demandez 03:04, 19 October 2008 (UTC)

I'm surprised anyone is even suggesting the Foundation is leaking anything. How ridiculous. -- How do you turn this on (talk) 03:09, 19 October 2008 (UTC)

Speaking for myself, I don't myself mean to suggest they would leak, and (hopefully) that's not what any here is saying. It's just good data handling to only collect what you need. For one thing, right off the bat, even if we trust the humans involved, computers get compromised all the time.

For two, "Don't collect what you don't need" shields the foundation from false allegations of leaking-- it is impossible to leak what you never had. This is why police officers on a traffic stop will insist that you hand them only your identification, rather than you entire wallet-- if they were never in possession of your wallet, it is impossible for them to have mishandled it.

I'm just saying-- I estimate that the probability that we'll have to deal with one of top-8 most trusted people on the project being unwilling to identify if actually elected is basically zero. On the other hand, the probability that a disgruntled losing candidate might try to stir up trouble by alleging mishandling of information is distinctly non-zero. --Alecmconroy (talk) 05:13, 19 October 2008 (UTC)

Thanks for your reply, Cary. I'm sorry if you felt annoyed by the requests to make an 'official statement' of any kind; I suppose the justification was that this really is a first-of-its-kind situation (an individual project demanding candidates pre-identify; until now, only WMF-wide elections [steward, Board of Trustees] seem to have required that).
I suppose a useful compromise would be to simply make it all the more clear to candidates that, if elected, they will have to identify before taking their seats. I'm going to create an editbox with a pre-loaded notice that gives a few disclaimers to candidates whilst they're creating their statement. (I'll link here in a moment.)
Anthøny (talk) 11:11, 19 October 2008 (UTC)

No. Abcom is in theory at least chosen by the community. If the community choses someone who isn't prepared to identify that is it's choice.Geni 13:16, 19 October 2008 (UTC)

• Actually, the community chooses a candidate on the assumption that s/he is willing to identify. If the candidate later reveals he isn't willing to identify, then he's mislead the community, and in my view, the appointment is void. Anthøny (talk) 13:47, 19 October 2008 (UTC)

Not the case I was thinking of. The community may want to chose someone who refuses to identify.Geni 13:50, 19 October 2008 (UTC)

And the point would be what, exactly? The community would chose someone who would then not be on the ArbCom? Please remember that the requirement to identify is a Foundation issue; the community has no say in the matter. What you propose, at best, is to have a Pope at Avignon. — Coren ^(talk) 14:21, 19 October 2008 (UTC)

The person could be on arbcom if a rather ineffective memeber. The point would be that the rules for running the election are not meant to make it easy for the community to pick the candidate it likes. Not for the people seting the rules to pic the candidate type they like. If someone who refuses to identify manages to win (I suspect it would be imposible but they should be free to try) that would suggest the community accepts that it is legitate not to trust the foundation's data handleing procedures. If that were to happen (I really really doubt it would) perhaps the foundation would feel the need to review those policies. The ability for people with non standard views to run for arbcom is an important check and balance in en's relationship both with itself and with the foundation.Geni 15:14, 19 October 2008 (UTC)

Actually no, the person could not be on ArbCom at all; there is no community leeway here: members of ArbCom are and must be identified, and someone who is not identified cannot be a member of ArbCom. — Coren ^(talk) 17:31, 19 October 2008 (UTC)

Not so. They couldn't be on the arbcom mailing list. They would not be allowed acess to priverlaged information but that does not de jure prevent them from being part of arbcom. Heh I understand some wikis have arbcom memebers who are non admins.Geni 22:08, 19 October 2008 (UTC)

That would be correct, if it weren't wrong.  :-) Last year, there was an official clarification to exclude minors from ArbCom because they cannot identify, and that identification was sine qua non to being a committee member. — Coren ^(talk) 22:41, 19 October 2008 (UTC)

Nope no such board resolution exists. The foundation may request that the community limits itself to choseing people prepared to identify but it cannot force the community to unless it is prepared to take over the functions of arbcom. It can block arbcom-l acess or simular. I suspect such a request would have significat weight but that is a descision for those voteing not those sorting out the running.Geni 23:04, 19 October 2008 (UTC)

Well, you have sort of made my point for me, Coren. It is a Foundation issue, not something to be decided by individual wikis, and any questions about it need to be addressed to Foundation personnel; nobody in this thread with the exception of Bastique is in a position to answer them authoritatively. This appears to be the first time where an individual wiki has proposed to require pre-identification as part of the self-nomination process; the Foundation's policy on access to nonpublic data says "Any volunteer who is chosen by any community process to be granted access rights to restricted data shall not be granted that access until that volunteer has satisfactorily identified himself or herself to the Foundation".(emphasis mine) I'm not quite getting why en.wp would want to require, for this one particular role, an identification policy that goes far beyond that of the Foundation itself. Please note that candidacy for an Arbcom is not comparable to candidacy for the Foundation Board (the only other place where pre-identification is required). Foundation Board members are required by law to be publicly identifiable. Arbcom members need only to self-identify to the Foundation (not the whole world), and only in response to an internal policy. Risker (talk) 15:15, 19 October 2008 (UTC)

Actually, you got this pretty much backwards. Board members are, indeed, required to be identifiable - not candidates. The reason why pre-identification was required was to ensure that all candidates, if elected, could actually serve. Handling of identification of candidates that were not selected is exactly analogous to what would happen here. The data handling policies of the Foundation are nothing but a red herring in this case: they remain exactly the same whether identification is made at the time of nomination or after the vote. There is no legitimate reason to request or expect that running for ArbCom can be made pseudonymously given that the position absolutely requires that identification be made. Anyone who wishes to run without identifying is, in effect, stating that they refuse to serve (given that they refuse to fulfill the only absolute qualification). If one wants to remain pseudonymous, one needs to not request positions of access to private information. — Coren ^(talk) 17:29, 19 October 2008 (UTC)

(The though occured to me that, as well, someone could run and refuse to identify themselves if they fully expected that they could not be elected and that identification would be moot— this is even worse for two reasons: if they indeed have not been elected then they have wasted a great deal of community effort; if they do get elected against their expectation then they would either have to identify themselves in the end, or would have not only wasted community effort but displaced another, viable, candidate). — Coren ^(talk) 17:37, 19 October 2008 (UTC)

I think the appropriateness of requiring advance identification depends in part on how many candidates there are. If there are 10 candidates for 6 or 7 positions, it would be perfectly reasonable to ask everyone to identify. If there are 50 candidates, it would probably be excessive to ask the Office to verify more than 40 extra people, even though Cary has indicated that he is willing to do so.

If we don't go with advance identification, then it will be simple enough for someone to post a question to all the candidates: "Are you willing to identify yourself to the Foundation Office immediately upon being elected?" Anyone who doesn't answer that question with a yes isn't going to be elected anyway. Newyorkbrad (talk) 17:33, 19 October 2008 (UTC)

The corollary to that question is whether requesting advance identification will affect that number downwards. I would expect that a number of "what the heck" candidates for ArbCom will reconsider if nominating oneself involves more than simply writing and transcluding a page. Sorta of a "you must be this serious to run" sign on the door... whether that is a good thing or a bad thing on its face is open to debate (I think that's a good thing). — Coren ^(talk) 17:40, 19 October 2008 (UTC)

I just wish I worked at the office, because I want to see Bishzilla's birth certificate. :) Newyorkbrad (talk) 17:58, 19 October 2008 (UTC)

What office? Bishzilla just ate it. Jehochman ^Talk 03:09, 20 October 2008 (UTC)

FWIW, Jimbo has restated that he will not appoint someone who has not identified to him or the Foundation, but that he sees "no reason to bother" identifying beforehand. Given that there are privacy implications to requiring the identification of all candidates, and that the change in procedure is not clearly undisputed, I'm changing my position to "Meh, it would have been a good idea but I'm not deadset about it."  :-) — Coren ^(talk) 03:07, 20 October 2008 (UTC)

It is my belief that an individual project can make more stringent requirements for things like access to data (and many other things) than the foundation, but it cannot make less stringent requirements, the WMF requirements are a floor. So a project could (if such consensus were demonstrable) require that candidates have the first name Brad, or be ineligible to hold office. There are many projects already that impose more stringent requirements than mandated on things such as how CU checks are carried out, or who can be a CU. This strikes me as no different. I think a commitment that one WILL identify to the foundation if elected is an absolutely reasonable one to ask of candidates. I think candidates should be allowed to identify in advance if they so wish, and if Cary's willing to process them, great. (Note that to run for steward you are supposed to self identify in advance, not after. Note also that I am not necessarily saying that a committee that had only Brads on it would be a good thing... Or am I?) ++Lar: t/c 04:46, 20 October 2008 (UTC)