Wikipedia talk:Arbitration Committee Elections December 2009

Notes from the 2008 elections.

These are issues that arose during the last vote, and are recorded here for reference. Refer to Wikipedia talk:Arbitration Committee Elections December 2008 for the discussion at the time.

Comments with votes used as 'ballot-box campaigning'

During the 2008 elections issues arose from people using comments with their oppose votes to make negative campaigning statements that could not be rebutted due to the restriction on threaded commentary on voting pages. --Barberio (talk) 20:20, 6 December 2008 (UTC)

Discussion of the issue resulted in an apparent strong majority of support that "The ballot box is not the place to campaign" and that discussion about the candidates should be kept to candidate talk pages, and general campaigning elsewhere. There was also a significant minority that felt that giving a "rationale" comment should be required for all votes. --Barberio (talk) 13:36, 12 December 2008 (UTC)

'Arbitrator concern' and canvassing issues

For me, so far, the following came up:

1. There needs to be a clear understanding how arbitrators may comment, or ask questions, of candidates. This is problematic since arbvitrators may have good knowledge of the role, or behaviors, and of issues, that (like any admin) they feel need raising; at the same time some users will react with "Arbitrator has this concern" as a sort of "kiss of death". Perhaps some admins willing to proxy for (well merited) arbitrator questions, so the issue can be asked without "OMG ARB" being an issue?
2. Likewise users need to know where concerns over canvassing should go - this may need to be off-wiki since it can often include copies of emails which can't usually be posted.

Will add more as needed. FT2 ^{(Talk | email)} 02:51, 9 December 2008 (UTC)

Secret ballot

Concerns were raised over if the election should be moved to a secret ballot. Jimbo Wales, presumably in his position as final judge of the election, made a comment in support of votes being taken in private as an option in addition to public voting. [1] --Barberio (talk) 13:23, 12 December 2008 (UTC)

Here is a discussion about secret ballots for Arbitration Committee elections. A majority of the users supported the Schulze method and secret ballots, like in the last Wikimedia Board of Trustees elections. A Horse called Man 21:38, 10 October 2009 (UTC)

The Audit Subcommittee elections that start later this month, and it will be support/oppose voting; we intend to use secret voting for this election unless some unanticipated difficulties arise in getting it configured quickly enough. Schulze method is not designed to select a large number of "winners" (indeed, it is designed to select a single winner), and is counterintuitive in an atmosphere where voting has always been done as support/oppose rather than required ranking of choices; there is no way to oppose a candidate. One step at a time; let's trial secret balloting first, and then have a full community discussion over a longer period about whether support/oppose or some other method is the preferred method of vote counting. Risker (talk) 21:46, 10 October 2009 (UTC)

So, the question becomes, then: Should we be preparing a standard arbcom election, in the style used for 2008 and prior years? I've updated some of the standard templates for 2009's dates, but don't want to dig too deep into that process if we're going to switch over to secret balloting or something just as different. UltraExactZZ ^Claims ~ Evidence 20:46, 12 October 2009 (UTC)

I thought it was pretty much settled that secret balloting was preferred. If so, someone can dig up the links and post them here. If not, the audit subcommittee elections end on Nov 7, with Arbcom voting beginning on Dec 1. Assuming secret ballot functions properly for the subcommittee election, that gives you 2 weeks to discuss whether to have a secret ballot for the arbcom election, and then one week to either configure the secret ballot software or to set up the traditional election pages. Thatcher 18:52, 26 October 2009 (UTC)

Thatcher, with all due respect, no, I don't think it was settled. The only consensus was in one small poorly advertised RfC. As I'm digging through, the only place that the RfC seems to have been advertised was in one small announcement at the Village Pump Policy page,[2] and then after the RfC closed there was one tiny announcement buried in a "Discussion report" of a Signpost article, saying that the proposal was "soundly endorsed".[3] But I don't believe that qualifies as adequate proof of community-wide consensus for such a major change. Better would be to make big announcements about it everywhere: WP:AN, WP:AE, Centralized Discussions. Then if there's still an obvious consensus, proceed. Otherwise, stick with the way we've done it before, and continue to engage in discussion about possible changes for next year's election. --Elonka 20:33, 26 October 2009 (UTC)

I didn't see any of what Elonka mentions (unless my memory stinks ... ummm, wellllll ...) SandyGeorgia (Talk) 20:44, 26 October 2009 (UTC)

Pointing out here what I said below, which is that the discussion was advertised at Centralised discussions. See here. I will agree that it can be easy to miss discussions if they are not well-publicised. But equally, it is easy to be busy during the period when such discussions take place. I've sometimes thought "how did I miss that?", and found out that I forgot to read the Signpost that week, or something. Carcharoth (talk) 08:14, 27 October 2009 (UTC)

I am most disturbed to see secret balloting in place, considering the disgusting canvassing in favor of one candidate last year who was elected to ArbCom on a marginal vote. How is the community to help expose this canvassing if ballots are secret, and how are we to have trust in the final outcome? I seem to remember a graph from last year that showed the upturn in one candidate's percentage, that coincided with the canvassing, but I can't locate that graph now. SandyGeorgia (Talk) 19:02, 26 October 2009 (UTC)

Sandy, I was one of those marginal candidates. If canvassing was going on with respect to other candidates, or on my behalf without my knowledge, I would have wanted to know about it. Can I ask that if you encounter such canvassing in future, that you do something about it? Maybe suggest a mechanism this year for people to report unacceptable levels of canvassing beyond things such as "I've voted for X, who are you voting for?" Simply saying that you were aware of "disgusting" canvassing, and not naming names or giving details, doesn't really help, other than cast a cloud over all the marginal candidates. The graph you are looking for is, I think, this one. Carcharoth (talk) 08:23, 27 October 2009 (UTC)

Carcharoth, I did not receive canvassing for you and wasn't aware of any; thanks for asking so I could clarify. But the canvassing that did occur was not only ethnic block voting. I really don't know how to reveal cavassing without divulging private e-mail, which is a barrier I don't want to cross. Perhaps others can suggest ways that canvassing can be reported, assuring that e-mails will be held in complete confidentiality, but since reporting canvassing would normally result in an investigation, I don't see how confidentiality can be assured. SandyGeorgia (Talk) 13:01, 27 October 2009 (UTC)

Canvassing is unfortunately inevitable. At least a secret ballot prevents the "oppose - molests small dogs" spoiler vote. Perhaps the only way to neutralise canvassing is to allow everyone to do it.--Scott Mac (Doc) 19:37, 26 October 2009 (UTC)

Of course canvassing is inevitable, but secret ballots make it harder to expose. SandyGeorgia (Talk) 19:38, 26 October 2009 (UTC)

Why expose it? It is only a problem if one tries to ban something when no ban can be effective. Better just to accept it.--Scott Mac (Doc) 19:52, 26 October 2009 (UTC)

I'm sorry, but your response makes no sense to me, so I can't reply. Do you really think canvassing was a factor, for example, with some of the top nominees? Or do you think it affected only the candidates who were appointed on slim margins? I get a pretty decent amount of e-mail, and try to be in touch with the community, so I know what my answer to that questin is. SandyGeorgia (Talk) 20:07, 26 October 2009 (UTC)

I have no idea. I strongly suspect that canvassing went on for ALL candidates. I was certainly approached by numerous people trying to solicit my vote for a whole variety of candidates. It is inevitable and unpreventable. I see absolutely nothing to gain by trying to stamp it out. People are going to canvas, on-wiki and off-wiki, using irc, IM, e-mail and doubtless other means. You may detect some of it, you will inevitably miss most of it. I can't see the point in caring.--Scott Mac (Doc) 21:11, 26 October 2009 (UTC)

←Scott, the point of caring is that a clandestine "please vote for me" contest doesn't necessarily get us arbs with the right skill-base. Candidates have ample space to describe their skill-base, and the election is widely advertised. It's a tough job being an arb, and elections and skill don't usually mix well; that is why canvassing should be open and public, not via email or IRC, where it's possible for lots of agendas to swirl around undetected ("I'll support Blogistan over the Twitter Republic if it comes up at ArbCom"—aargh). I don't see why different rules should apply to canvassing here than for RfCs. However, I'm not prepared to make a big thing of this at such a late stage. Tony (talk) 07:30, 27 October 2009 (UTC)

Tony, that might be true, but there is absolutely no way of preventing it. We might also use the same argument for saying people ought to read what the candidates say on their question pages, and not vote simply on the basis of how their wiki-friends vote. However, requiring people to read the questions is also unenforceable - just like trying to prevent canvassing is unenforcable . Anyway, most civilised countries agree that secret ballots overall tend to reduce the opportunities for electoral manipulation.--Scott Mac (Doc) 13:06, 27 October 2009 (UTC)

Sandy, with a secret ballot, there will be much less point in canvassing, because the canvassed can say, "Yes, sure, I'll vote for X, if you vote for Y," then ignore it, and vote whatever way they want. That's the beauty of it. All the usual double-dealing will be a waste of time. SlimVirgin ^{talk|contribs} 12:48, 27 October 2009 (UTC)

Quite.--Scott Mac (Doc) 13:06, 27 October 2009 (UTC)

This is my point also. Thatcher 14:25, 27 October 2009 (UTC)

(ec re to SV) Perhaps that's valid; I know I would be likely to lodge more Oppose votes on a secret ballot. In fact, with secret balloting, I suspect we're going to find much lower Support percentages, because people will be less afraid to oppose, causing Jimbo to dip even lower to fill ArbCom. SandyGeorgia (Talk) 13:08, 27 October 2009 (UTC)

Sandy, "dip even lower" is not the way I see it. The more I think about it, the more advantages a secret ballot has, including not publicising the votes afterwards. Tony (talk) 13:47, 27 October 2009 (UTC)

"Dip lower" in strict numerical terms, perhaps, but if the previous elections had inflated Support numbers, then the winning candidates could have the same level of real support. Thatcher

having a secret ballot has another advantage of just eliminating the "horse race" aspect of it, which can create some irritating behavior, and encourage last minute canvassing.--Tznkai (talk) 16:12, 29 October 2009 (UTC)

OK, I've come around on this, and see the benefits of a secret ballot, but have some followup questions. Who will be auditing (the Auditing Committee, I presume)? It is often "the community" who identifies socks, ineligible voters, duplicate votes, etc. ... will that task fall to them, or be another burden on an already overworked ArbCom. (I apologize if I've missed this; I can't keep up with the entire page :) Also, if this task falls to the Auditing Committee, will they pledge confidentiality if canvassing e-mails are forwarded to them? SandyGeorgia (Talk) 16:30, 29 October 2009 (UTC)

My understanding is that when an account votes, then the fact that they voted will NOT be secret, simply the content of the vote. This means that if people sock, the community can investigate that as before, without breach of the secret. If there is any other impropriety, that can be investigated by admins or arbcom, as before on the same grounds. The content of anyone's vote will not be disclosed for any purpose, including that of investigation. — Preceding unsigned comment added by Scott MacDonald (talk • contribs)

There will be some small group of scrutinizers (scrutineers?) who will have access to all the relevant data. They will either be drawn from en.wiki or the foundation level, I believe the decision is ultimately up to Jimbo, but the details are lacking.--Tznkai (talk) 18:11, 29 October 2009 (UTC)

That response concerns me, because the community is helpful in picking up invalid votes, and this seems to place a very high burden on a small number of editors. SandyGeorgia (Talk) 18:16, 29 October 2009 (UTC)

Woah, Hold on, I thought in that RFC all the talk was of a 3rd party voting system, not vote counting by some selected group of wikipedians?--Cube lurker (talk) 18:20, 29 October 2009 (UTC)

e/c I understand your concerns, but hopefully I can allay some of them. First of all, (if i understand it correctly) who is voting is immediately visible to everyone, so some problematic trends will become rapidly apparent. Second, we're about to test the secret balloting (barring an application of Muprhy's law) with the audit subcommittee elections.--Tznkai (talk) 18:26, 29 October 2009 (UTC)

Responding to Cube lurker, there has been a lot of talk and no decision that I am aware of. I personally advocate taking care of it in house, which seems tidier to me, for the same reason we rely on in house CU and OS. Others have made fairly compelling arguments that it should be outsourced to stewards (which are elected at the meta-level), other language CUs, or Foundation appointment (similar to this election committee). Tznkai (talk) 18:29, 29 October 2009 (UTC)

I think, particularly considering recent events still fresh in our minds, it's outrageous to expect us to trust the integrity of the process if it's done behind closed doors by a selected group of wikipedians.--Cube lurker (talk) 18:40, 29 October 2009 (UTC)

The list of all users that have voted is publicly-viewable, and everyone is welcome to participate in the auditing of that list, as with any election. The election admins have immediate access to checkuser-like data on all votes cast, which gives them additional resources in identifying sockpuppet votes and reduces the burden on CheckUsers (although their involvement may still be necessary), but gives them no other 'secret powers'. At no time does any person have the ability to determine how a particular vote was cast, and the election admins have no involvement with "counting votes"; that is done entirely by the software. Happy‑melon 07:36, 30 October 2009 (UTC)

I was under the impression there was some sort of human being involved in auditing the votes themselves, for those of us who are on occasion distrustful of software?--Tznkai (talk) 07:43, 30 October 2009 (UTC)

You mean in terms of counting the votes? If the WMF servers are incapable of correctly adding up a pile of ballots, we have greater things to worry about than a subcommittee election...

Election admins have the ability to strike votes that appear to be illegitimate; although neither they nor any other use have the ability to see how the votes they are auditing have been cast. The only information they have access to over and above ordinary users is checkuser-like data on the votes. Happy‑melon 07:49, 30 October 2009 (UTC)

If I understand what you've said, (only half way through my morning coffee), then I'm far less concerned then I was when I last posted.--Cube lurker (talk) 13:19, 30 October 2009 (UTC)

Without getting all conspiracy theory on you, machines screw up sometimes (more accurately, people programing, operating, and maintaining the machines screw up, but the point is the same). I am unaware of any significant voting system out in the real world that doesn't leave open the option of human beings chasing down the votes. I was under the impression that the board election, for example allowed a more manual tally by the appropriate staffers.--Tznkai (talk) 07:56, 30 October 2009 (UTC)

In any voting system there must be some way, some how, for some one to put the jigsaw pieces back together and check the figures in the presence of all the information. If you cant to get really technical, securepoll stores voting data in two tables: one table stores a vote id and details about the user that cast the vote, and the other stores the same vote id, and details about how the vote was cast. There is, quite deliberately, absolutely no method in the software to link between the two tables and get the complete picture on any vote; both tables are considered private data for the purposes of database dumps, toolserver replication, etc; you'll find them wherever you'd find checkuser data, and nowhere else. So in this election, the people who potentially have the complete picture are the sysadmins, who have shell access to the database. But remember that these people could quite easily be logging your form submissions and seeing how you vote anyway. Or they could just go and write a few extra votes into the database if they felt the urge. You have to trust someone.

If we get to the end of the vote, hit the "tally" button, and learn that every single voter apparently opposed every single candidate, we will get a sysadmin to look at the raw tables and check that that was indeed the case, but the chances of the votes being correctly recorded and yet tallied wrongly, are very low (in principle, of course, zero, but even assuming that there is an error somewhere in the code, it's much more likely to result in votes being recorded wrongly in the first place). Of course, machines screw up. We're not giving people the nuclear codes here. Happy‑melon 17:47, 30 October 2009 (UTC)

Requiring a petition, increasing numbers and reducing term

Is it at all on the table to completely revamp the system rather than just once again rearrange chairs on the Titanic? I would very much so like to see us abandon the "special class" of arbiters and go with something a little more community oriented, like having cases heard by a randomly selected pool of willing admins. That solves problems like burnout (there's 1 case/year max for each person) and arbiters being an aloof special class. If we are unwilling to consider drastic reorganization, there are at least some less drastic things we can do:

• Require a petition to stand for election. You must submit signatures from 50 qualified voters in order to run. Right now, there are enough candidates that reading all of the questions + answers for all of the candidates would potentially be a full-time job. So if we eliminate the ones who would never stand a chance of being elected, we at least reduce the scale of any problems we do have.
• Increase the number of arbiters. Burnout and inactivity are two serious problems arbcom has right now. Why not increase the number of arbiters to 20 and then assign ten of them randomly as each case comes along? By spreading the workload, you help resolve those issues.
• Rotate the membership more. Reduce the term to 18 months and have elections every six months. Again, this helps with burnout and it encourages more qualified people to step up (being an arbiter isn't going to consume the next 3 years of your life any more).

I'd really like to see us take more than token changes - there's a lot that needs to be reformed. --B (talk) 18:12, 10 December 2008 (UTC)

There have been a lot of very good ideas coming forward from the RfC; the one that I find most problematic from both a candidate and a franchised voter perspective is the every-six-months election. I can tell you honestly that answering the questions—over 250 of them for most candidates—consumed over 200 hours; reading the responses of 20 or 30 candidates over that many questions would take at least 50 hours. Requiring signatures of 50 qualified candidates is also an extremely high barrier, given the number of editors who actually vote (probably about 1000 total by the end of this election, of which about 40% will vote one way or the other on 5 or fewer candidates); in fact, if people are able to garner that many "votes" in advance, the election is something of a denouement. Keep in mind that three of the top-10 candidates have fewer than 250 votes total as I write this; a 50-vote petition is over 20% of their total votes, and would also mean that each election phase starts increasingly early as people work to collect signatures. To be blunt, Arbcom is important, but it isn't so important that it should preoccupy this much of the community's attention and resources. I don't really have a problem with limited-issue candidates, even if they are unlikely to succeed, as their contribution to the discourse is valuable in many cases. Risker (talk) 14:19, 12 December 2008 (UTC)

non-admin seat

Another reform that we should consider is having a non-admin seat on the committee. There are a lot of good Wikipedians who never become admins for a variety of reasons, and each year we have a number of candidates who are non-admins.

In 2007, Giano II was nearly elected, and this year The Fat Man Who Never Came Back was over the 50% threshold. I voted for the former, but not the latter because there were sufficiently diverse candidates this year, however a non-admin seat would have guaranteed a position for them both, and I think the project would be better off for it. John Vandenberg ^(chat) 00:53, 16 December 2008 (UTC)

I second the thought behind it (and I voted for both candidates). The committee should reflect WP in its plurality of editors with good judgment, good sense, and constructive dedication toward the encyclopedia, They come with and without the badge.

Are you suggesting

• a required minimum of 1 non-admin seat, or
• a fixed quota of exactly 1 non-admin seat (implying that all the other seats would be held by administrators)?

I might cautiously agree to the former, for one election year perhaps, but the latter is problematic. I don't think we should widen this construct of a trench between administrators and everybody else. One token non-admin seat seems a bit meager anyway. Do we want quotas or should all arbitrators be carrying the strongest community support during election, regardless of the presence of a badge?

Admin votes carry no greater weight than non-admin votes. Whichever way John's proposal goes, I'm optimistic that future elections will endow the committee with more than one non-administrator. The fact that this idea is coming from an arb-elect (and admin) makes me hopeful, and it might open some people's minds. I was quite surprised to see voters explaining their opposes with the knockout criterion not an admin this year. (The laconic reaction of at least one voter who supported per "not an admin", was just that and made me chuckle, of course). ---Sluzzelin talk 01:49, 25 December 2008 (UTC)

I don't mind how many non-admins seats are allocated, however if we are going to have non-admin seats, the number should be pre-determined before the election.

I would prefer that we have one non-admin seat allocated for 2010, so that there is fierce competition which will guarantee one very competent person is elected into that seat, which will reduce concerns, and should result in expansion in the number of non-admin seats in the future.

We can have the best of both worlds by allowing non-admins to run in a ballot for the non-admins seats, and also place them in the general ballot. In that way, the pre-determined non-admin seats will be filled with non-admins, however the community also has the option of appointing more non-admins into the other seats. John Vandenberg ^(chat) 06:35, 11 September 2009 (UTC)

The likelihood of a candidate being elected to ArbCom who could not pass RfA is very remote, and I'm not sure that I would want such a candidate on ArbCom. That means that the only people who have any chance of being elected to this seat would be those who choose not to run for adminship. How many of those are likely to choose to run for the much more perilous post of ArbCom? If the week of mudslinging at RfA is too great a hindrance for them, why would they want to sign up for three years of far greater abuse and nitpicking? I would be astonished if we had any candidates for such a seat who had any reasonable chance of winning it. Then there is the point that "non-admin" is not a permanent status: what happens if/when the successful candidate runs for RfA? Remember that, in a three year period, the likelihood of such an occurence is very high; I would say a near-certainty, especially if they acquit themselves as a competent Arbitrator. Should they be barred from RfA? Or should they be required to surrender the "non-admin" seat if they cease to be a non-admin? Or do we, by allowing the candidate to make the natural progression and hold both seats, accept the fact that adminship and ArbCom membership are both markers of the same qualities (trust and respect within the community) and hence that expecting a sensible candidate to have one without the other is not, in fact, realistic? IMO, creating a non-admin seat would be a purely political gesture which would both fail to achieve anything constructive, and which would actively move us further away from the WP:DEAL ethos which we should in fact be fighting to get back to. Happy‑melon 15:05, 23 October 2009 (UTC)

Thank you for your thoughtful response to this crazy idea ;-) You have raised a lot of good points.

The candidates who would be suitable for this non-admin seat are the type of person who is committed to not being an admin. Their role would be to provide a non-admin perspective. In my crazy mind, these candidates were expected to continue to not be admins throughout their term on the committee.

As you point out, folk who are non-admins by choice are not likely to want to commit large slabs of time to serving on Arbcom. But, maybe there are a few willing people out there... ? If so, hopefully they will find this discussion thread and raise their hand.

John Vandenberg ^(chat) 04:22, 24 October 2009 (UTC)

John, would it not be a major handicap to an arb if he could not read deleted edits?--Scott Mac (Doc) 15:00, 24 October 2009 (UTC)

I understand the general principle here, but I think that the real issue is getting non-admin voices respected on DR and other issues throughout the Wikipedia culture, not trying to shoe horn such respect by electing a non-admin. The technical problems however, are pretty easy to surmount, just ask the Roman Catholic Church. --Tznkai (talk) 08:11, 30 October 2009 (UTC)For those missing the reference, in theory the Pope could be any layman, who would then be ordained deacon, then priest, then elevated as a bishop

It would be necessary that arbitrators can see the deleted edits. This can be fixed at a technical level by raising a bug request similar to bugzilla:21044. Non-admin arbs could then be given checkuser, which would give them the ability to read deleted pages. --John Vandenberg ^(chat) 09:08, 30 October 2009 (UTC)

John pointed out in the first post of this thread that non-admins usually are not elected, even if it almost happened once. But this information leads imho to another conclusion than creating a special non-admin seat: That the community does not want non-admins on ArbCom or rather has, as happy-melon says above, seen RFA as a necessary pre-judging of the candidate's acceptance in the community. If the community was in favor of non-admin ArbCom members, they surely are free to elect them, are they not? Furthermore, the creation of one or more non-admin-only seats would back up the belief that some hold, i.e. that admins are different from "normal" users or that "normal" users and admins have different views on issues (they might have but not because they are admins or non-admins but rather because everyone has a different view). Such generalizations are not only incorrect, they are dangerously so by potentially increasing the rift between admins and non-admins.

Imagine a political election where a certain number of seats are reserved to, let's say, blondes. Nothing separates blondes from people with other hair color except said color and surely no one would argue that this leads to them having different views. To reserve a seat for a group that per definition has different views is accepted on the other hand (for regional minorities for example - at least here in Germany, for example the South Schleswig Voter Federation). But non-admins are not such a group and as such, it would be incorrect to assume that the status of not being an admin would lead to a ArbCom member having different views, just as the admin status does not in itself leads to having different views. Regards SoWhy 09:41, 30 October 2009 (UTC)

How to simplify and make better the AC elections - proposal

Every other Wikimedia project does fine, and functions fine with Jimmy Wales's "approval". I have yet to see evidence that his "approval" is required, nor even needed, for the AC elections. I have yet to see the value that is added by Wales "approving" members or "approving of" the community election. His two appointments to the AC, Essjay and Jayjg, did not work out well--Essjay lied to and deceived the community, and generated the single most embarrassing even that Wikipedia has yet to see in the mainstream press. Because of Essjay, our contributions and donations actually dropped (a month after the "Essjay Event", contributions did in fact drop). Jayjg, appointed to the AC by Wales, was later defrocked of his Checkuser and Oversight tools for behavior unbecoming. Again... what value does Wales's 'role' in the appointment process add?

What needs to happen next? How do start to ween ourselves off of Jimmy, to have our internal processes begin to stand alone?

AC elections process proposal

• Wales's previous role in the AC elections, as it existed through 2008, is discontinued.
• Selection to the AC is based upon, in 2009 and going forward, the following:

• A minimum 50% support in the elections.
• Willing to disclose their real identity to the WMF.
• The number of open seats in each tranche.
• The highest percentage number (support/oppose) of each candidate.
• The top winners by percentage are appointed to the longest-duration open tranche seat, down to the shortest duration open tranche seat.
• Wales has no approval role. If he feels a candidate is inappropriate for the AC, he may unilaterally veto them, and must provide private justification to the current AC.

• If Jimmy doesn't provide reasoning to the AC within 1 week of the election's close, the veto expires and does not stand.
• The AC, by a public up-and-down voted motion, may overrule any of Wales' vetoes. Departing Arbiters may not vote on these motions. The overrule motion must be ran in public, must occur within 1 week of Jimmy's veto, and the motion vote must last 1 week.

• The next highest candidate by percentage moves up. Wales may exercise the veto process up to three (3) times per election. Afterward, the will of the community stands.

Absurdly simple, and easy. Jimmy is still able to nix people, but the final decision really does here lie in the community--either by election, or by the overrule of Jimmy, via the people we elected.

Thoughts? rootology/equality 18:34, 26 May 2009 (UTC)

I'm not sure that that is intra vires the community. Stifle (talk) 13:08, 12 June 2009 (UTC)

Nothing isn't. rootology (C)(T) 13:22, 12 June 2009 (UTC)

• I think it is a bit anachronistic that we (the English Wikipedia, as opposed to all other language editions) still have a god-king here who is involved in our decision-making. We should be able to do without. In other words, I endorse your proposal with the exception that I don't think any single user should have a veto right (except, say, Cary Bass or somebody like him who would be the official verifier of identities). However, I would like to add that in the last elections, Jimbo Wales made excellent (and very fair) appointments based on slightly unclear results. The results were a looking a bit unclear because it was not clear to the voters how the winners would be chosen, so without a single final arbiter, we will need clear rules. Yours would probably work. Kusma (talk) 14:37, 12 June 2009 (UTC)

In the interests of checks and balances, I'd rather some other group of editors besides the AC fulfill the veto-monitoring role. Self-regulation is a very bad idea. I would prefer to see bureaucrats—by and large a trusted and trustworthy bunch—manage the vote counting and step in wherever judgements are needed to interpret The Will Of The People.  Skomorokh  14:47, 12 June 2009 (UTC)

While I find your proposal a bit too complex to gain traction, I like the idea of codifying how the seats are filled, particularly because it avoids the silliness and uncertainty that we have traditionally encountered after the close of voting. It should be clear, while people are voting, how many seats are up for election. People put a lot of thought into these votes, and have a right to know what their votes mean.

Jimbo has, in the past, added extra seats after the close of voting, so that he could appoint people who didn't quite make the cut without technically "vetoing" anyone. While he has of course always done so with good intentions, the Arbitrators he appointed to those seats have been of mixed quality. In Equality's proposal, the number of seats could not change after voting.

For those who are proposing to go farther and cut Jimbo out of the loop entirely, I disagree. I would prefer to maintain the Founder's right to veto people. Someone has to be able to do it, because if the community directly elected the Arbitrators, we would risk fragmenting Wikipedia into factions and political parties. If we form some "vetoing committee", that creates an extra layer of bureaucracy for no obvious gain. As long as it makes sense for one person to have this veto power, there is an obvious choice for who that person should be. rspεεr (talk) 04:17, 3 October 2009 (UTC)

Simplify the process without changing much: no more tranches

By the way, tranches are dumb. They are meant to maintain a cycle of seniority on the ArbCom and a relatively constant number of open seats, but only under the unrealistic assumption that most Arbitrators stay through a three-year term. We could simplify the election greatly without changing much -- we'd just stop diddling with term lengths and expansion seats -- if we replaced the "tranche" terminology with a single rule:

• At each election, the number of seats to be filled is the number that would result in a total of 18 Arbitrators. (This number of Arbitrators can be changed, in advance of the election, if the community forms a consensus to do so.)

Easy. This also goes very well with reducing the term length to two years, if we choose to do that. rspεεr (talk) 04:17, 3 October 2009 (UTC)

Numbers

Last election, 10 new members were appointed, taking the total in January '09 to 18. A quick look at the WP:ARBCOM page shows that we now have 14 arbcommers, of which 10 are active. What I have been thinking is that, if we want a committee of 18 members, we ought to appointed 20-25 to allow for the inevitable drop-outs throughout the year. -- EA Swyer ^Talk _{Contributions} 19:34, 6 September 2009 (UTC)

18 memebers is meant to be high enough that we can deal with the normal drop off rate.Genisock2 (talk) 19:39, 6 September 2009 (UTC)

Has the caseload increased? That's the question. We'd need some committee members, as we run up to the election, to say what size the committee should be to handle the work and from that we can extrapolate how many seats need to be filled. -- EA Swyer ^Talk _{Contributions} 00:34, 7 September 2009 (UTC)

I don't think there should ever be 25 Arbitrators, not even if we expect half of them to drop out. The ArbCom should be more like a Supreme Court than a Senate. rspεεr (talk) 04:28, 3 October 2009 (UTC)

Questions

I don't know who's going to be running, but I'm wondering:

• Will FloNight run for re-election?
• Will Coren ron for re-election?
• Will Stephen Bain run for re-election?
• Will the vacancies of Kirill Lokshin, FT2, Sam Blacketer or Jpgordon be filled following this election?

FloNight, Stephen Bain and Coren will have their terms expired at the end of this year. FloNight is longest serving current arbitrator, having served ArbCom since January 2007, two years, and nine months. Stephen Bain was appointed to a one year term in January 2008, but Jimbo Wales extended his/her term to a two year term following the resignation of Paul August. Coren has been part of the Committee since January 2009, and is appointed to a one year term. I wonder if they've decided if they'll be running for re-election. Also, all three are currently on Tranche Gamma. Right now, there are four vacancies, and should FloNight, Coren and Stephen Bain not run for re-election, there will be seven seats for newcomer arbitrators to fill, or if more seats are added to the Committee for this election, there would be more seats to fill. I am wondering what the intentions of the above three are, and whether those vacancies will be filled, but, while we can assume, we won't know until the end of the election. Only time can tell. Mythdon (talk • contribs) 20:48, 7 September 2009 (UTC)

Well, there has been discussion to make Arbitrators' terms uniform at two years (almost everyone agrees that three is too long for most, and that the rare case where someone could last longer then requires six years instead of four). Part of that discussion has informally raised the possibility of reorganizing seats in a way that would have my term extended to two years. Provided that gels, and that there is no strong community objection, I would accept — the very high amount of work that standing for election implies would pretty much knock me out of activity on the committee and is probably best avoided if not necessary.

Otherwise, yes, I'll probably run for a new two year term. I think I did a reasonably good job, I don't seem to have attracted undue amounts of hatred, and this committee has done a fairly good job collectively. — Coren ^(talk) 23:29, 28 October 2009 (UTC)

Holy crap Elections start in less than 3 weeks

I've updated some of the election pages and templates for 2009, matching closely what was done last year, including General Questions, Candidate Statement template, and so forth. It's almost too late to switch election systems to a secret ballot or some other method, and I haven't seen consensus to do so in any case. So, that said, I'm going to ping the arbcom for their take on the vacant seats for this election, what to do with the proposed non-admin seat (if any), and any additional procedural changes they need. If there's anything we're forgetting, please speak up! Thanks, UltraExactZZ ^Claims ~ Evidence 14:23, 21 October 2009 (UTC)

I think there's strong support for secret and preferential elections – can we ask the people behind the Board elections to help out with implementation?  Skomorokh, barbarian  14:27, 21 October 2009 (UTC)

This thread on Jimbo's talkpage is instructive – Jimbo intends "to keep the size of the ArbCom the same". This comment by arbitrator John Vandenberg regarding the consensus for election reform and the Board election committee is also worth reading in full.  Skomorokh, barbarian  14:36, 21 October 2009 (UTC)

Thanks for those links; that helps clarify things a great deal. I'll note the seats below, then. UltraExactZZ ^Claims ~ Evidence 19:08, 21 October 2009 (UTC)

We intend to trial SecureVote onwiki for the first time with the AUSC elections that start on October 30. After a lot of careful consideration, we have opted to go with straight support/oppose voting rather than preferential voting, because there is not currently a preferential voting system specifically designed to select multiple "winners", and the Schulze method is counterintuitive for many voters (the idea that not ranking a candidate is the equivalent of an oppose vote is a significant downside). As to the number of vacant seats, there will be a minimum of 7 and perhaps more; I encourage any of my colleagues who are contemplating resignation to make a decision and announce their intentions prior to the start of voting. Risker (talk) 14:42, 21 October 2009 (UTC)

I see you answered this question earlier on the page - and I replied to you! Sorry for the confusion. I'll note the current seats, expiring and vacant, in a following section, for clarity. Thanks, UltraExactZZ ^Claims ~ Evidence 19:08, 21 October 2009 (UTC)

Vacancies

Per {{ArbitrationCommitteeChartRecent}}, It appears that Tranche Gamma is expiring this year, which means the following arbitrators will be up for re-election, should they so desire:

• Coren (talk · contribs)
• FloNight (talk · contribs)
• Stephen Bain (talk · contribs)

Additionally, the seats vacated by the following are up for election:

• FT2 (talk · contribs) (Alpha Tranche, expiring 2010)
• Sam Blacketer (talk · contribs) (Alpha Tranche, expiring 2010)
• Casliber (talk · contribs) (Beta Tranche, expiring 2011)
• Kirill Lokshin (talk · contribs) (Gamma Tranche, expiring 2009)
• jpgordon (talk · contribs) (Gamma Tranche, expiring 2009)

I'll post this list on the ACE2009 page shortly, once I figure out the formatting. Unless Jimbo indicates otherwise, which seems unlikely, we should proceed with the understanding that all vacant seats will be filled. Thanks, UltraExactZZ ^Claims ~ Evidence 19:16, 21 October 2009 (UTC)

Are you sure jpgordon's seat is up for grabs? Per {{ArbitrationCommitteeChartRecent}} it does not seem to have been filled last year (despite Wikipedia:AC2008 indicating it was open), and adding five extra arbitrators to the current stock would give 18 rather than the 17 of January 1, 2008. If Jimbo wants the number the same as last year, does this not mean 17?  Skomorokh, barbarian  19:24, 21 October 2009 (UTC)

Deskana resigned following the close of last year's election, and no replacement was appointed. Thus we started 2009 with an empty seat. Technically, John Vandenberg was appointed to a 3 year term, but has stated he will step down at the end of two years, so he was put into the Alpha tranche instead of the Beta one, but I'm not quite sure by whom. This is why the tranches appear to be so imbalanced. Having said that, if people want to move to a two-year system (leaving the present 3 year appointments in place), this might not be a bad time to do so. Risker (talk) 19:46, 21 October 2009 (UTC)

You're right, Deskana's seat was specifically listed as planned to be filled in a by-election in Summer 2009 (oops). Since we had no such by-election, I feel comfortable listing that seat here as up for election. John Vandenburg should be under Tranche Beta (per Last Year's Results, with an already-filed retirement date of 31 December 2010 - which we would treat as if he said he was retiring then, rather than up front. Near as I can tell, the vacant 18th seat was actually in Tranche Gamma - Alpha is listed as having 7 (8, then take out John Vandenburg), Beta has 6 (5 plus John Vandenburg), and Gamma has 5. So if we fill FT2's currently-vacant seat with someone to be added to Gamma, then everything balances at 6 per tranche. (I'd update the chart, but it frightens me.) As for 2 year terms; It'd be simplest to split this year's winners between Alpha and Beta, with some getting a 1 year term and others getting two. Then, next year, we'd have 9 open seats to elect and we can proceed with a two-tranche system. There was discussion in the Arbcom RFC, lo these many months ago, about having staggered terms of 2 years, with elections every 6 months. Consensus, as I remember it, was that that would be too much hassle. But it's an idea as well. UltraExactZZ ^Claims ~ Evidence 20:18, 21 October 2009 (UTC)

Also - Not sure about jpgordon's seat. But if Jimbo wants 17 in total, it doesn't bother me, though it would make the tranches uneven - whicm might not matter. UltraExactZZ ^Claims ~ Evidence 20:20, 21 October 2009 (UTC)

I think Risker's Deskana explanation puts Jimbo's desired figure at 18. Your suggestion for the mechanics of a two-years system sounds appealing. I wonder what the arbitrators think of maximum two-year term lengths, and how much discussion would be needed to implement them?  Skomorokh, barbarian  20:25, 21 October 2009 (UTC)

As far as Implementation, either A) Jimbo could appoint to two year terms, as he wishes (though he's said he just wants to ceremonially approve the election, as opposed to using his discretion to change the results), B) The committee could issue some sort of ruling, though I'm not 100% sure that it is within their remit, or C) Candidates could individually agree to sit only for two years, reducing the term limit by consensus. The trick here is that they'd essentially be doing as John Vandenburg did - agreeing to serve only for two years of a three year term. There is not currently a process to stop one or more candidates from taking that third year, which would throw everything off. If there's consensus for 2 years, that becomes less of an issue. I do think we should know up front, though - quality candidates who only wish to serve for two years might not run if they'd be asked to serve three. UltraExactZZ ^Claims ~ Evidence 12:12, 22 October 2009 (UTC)

I won't venture to speak for my colleagues who are currently in the Beta tranche, except for John Vandenberg who made it clear he only intends to serve two years; however, I did state during my candidacy that would not be opposed to the term being reduced to two years. Do keep in mind that over 80% of arbitrators do not complete their terms, which also seems to favour reduced term length. One thing to be aware of is that there is a strong likelihood that each year more than half of the seats will be filled with new arbitrators; even if the early resignation rate is reduced, there is still the likelihood that one or more arbitrators will step down during a 2-year term as well. While I don't think that's necessarily bad, it will change the dynamics of the committee. And like you, Ultraexactzz, I wouldn't touch that chart with a ten-foot pole. Risker (talk) 20:59, 21 October 2009 (UTC)

If a reduction of term limits to 2 years is considered, we should think about deciding on this at the same time as the elections, although that would mean that we would only have approx. 5 weeks to get to a decision because imho it should be decided before voting begins so that all voters are informed on how long the people they vote for will serve. Regards SoWhy 21:06, 21 October 2009 (UTC)

I'd like to do it ahead of time; a good candidate is a good candidate, and I don't know that the length of the term would impact who votes for whom - but it could have an impact on who runs. UltraExactZZ ^Claims ~ Evidence 12:12, 22 October 2009 (UTC)

Calling the question

There appears to be some interest in 2 year terms for Arbcom, as evidenced here, last year's election discussion at WT:ACE2008, and the Arbcom RFC at Arbitration Committee change and reform (and in other sections; several views included a reduction in terms). As I note, term changes could be initiated by Jimbo (who I plan to ping regarding this discussion), by the committee, or by the community. Is 2 weeks and change enough time for a proper RFC calling the question? Would a motion or request for clarification to the committee be more appropriate? Or do we rely on candidates self-limiting to 2 years (or 1, if we stagger terms for a two-tranche system), and let the election results serve as a referendum? In other words, if candidates who explicitly pledge to sit for 2 years are the top vote-getters, would that be a clear indication of the community's desire for 2 year terms? Now that I think about it, that might be simplest, and it would inform Jimbo's decision in December. It would also force him to exercise discretion, which he as stated he doesn't want to do this time around (wanting his picks to be more ceremonial). UltraExactZZ ^Claims ~ Evidence 13:58, 23 October 2009 (UTC)

But since people don't only vote for those who make such pledges but mainly for those who they think are most able, the amount of votes for such candidates will not really be an helpful indicator, will it? Regards SoWhy 14:49, 23 October 2009 (UTC)

We wouldn't know until after - that's the rub. Last year, Jimbo read the voting as a mandate for change, and acted accordingly. I don't know how much of that came from the RFC that year, the votes themselves, or just the totals - and it's unclear what form a mandate for shorter terms would take here. I mention it as an alternative to a panicked RFC two weeks and change before nominations. UltraExactZZ ^Claims ~ Evidence 15:07, 23 October 2009 (UTC)

Why 2 weeks? Can't the RFC run until voting starts on December 1st and we simply inform all candidates that a change in term length is possible? Regards SoWhy 16:23, 23 October 2009 (UTC)

I have, by coincidence, just written to Jimbo about exactly this issue on his talk page:

Dear Jimbo

Last week I raised on your talk page the question of whether the current norm of three years for the terms of incoming arbs might be reduced to two years, suggesting that there would be consensus for such a change in the community, and asking whether you might sound out the current arbs on the matter. I know you've been busy and in transit recently, and the section has since been swallowed into the archives.

A colleague and I would be willing to hold an RfC before nominations are called (i.e., starting very soon) to gauge community attitudes to such a proposal, framed as advisory only. What are your thoughts on the matter, and would such an RfC be useful?

Tony

A quick count has revealed the astonishing fact that only six of the 57 arbs ever to have been appointed have served a full three-year term. A three-year term now would not be replaced until 2013, which seems like a long time to take our best and brightest out of circulation. There's the burn-out factor in such a long term, since being an arb is increasingly a challenging, complex and full-on job. There's the apparently quick learning curve of the current new crop of arbs. And there the option of a second term of four not six years.

I'm unsure why the tranches are needed, if elections can be treated as "top ups" of numbers to the desired size (18?) each January. The number of seats to be filled will almost certainly vary from year to year, which shouldn't matter.

I look forward to hearing your opinions on this. Tony (talk) 16:35, 23 October 2009 (UTC)

Finalization

As the elections process for the 2010 ArbCom is just three weeks away, it seems appropriate to request some definite information regarding this election. A confirmation of seats available, arbitrator vacancies, election method, term length and similar information from the current committee would be appreciated. —Finn Casey * * * 20:26, 22 October 2009 (UTC)

Funny, we were just discussing that in the thread above. We have 8 vacancies, as shown. There is consensus for a secret ballot, but - according to Risker - the committee wants to use the Audit Subcommittee Election next week as a dry run, which makes sense. Following that election, we would presumably discuss the matter with the community and come to a consensus on whether it would work for a large-scale election such as this. That's not going to happen before November 10, when candidates may self-declare, so I believe we will use the format from last year for this election - Support/Oppose voting, General Questions, etc. Term Length is an open question, as there may be support for a split two year term that folds the current gamma tranche into the Alpha and Beta - discussion is ongoing, above. UltraExactZZ ^Claims ~ Evidence 20:33, 22 October 2009 (UTC)

Thanks for the reply! I noticed the above discussion, but was not sure how widely publicized the discussion had become. I thought I would attempt to solicit remarks from other editors, and perhaps the current committee, with a view to forming a speedy consensus. As you note, time is running out to implement changes, so it is good to discuss the matter as soon as we can. The committee keeps very busy, and it is possible that matters could be inadvertently forgotten until too late (e.g. the overlooked summer 2009 election for Deskana's seat). Thus, this is just a friendly reminder and agreement with Ultraexactzz's attempts to get the ball rolling. I also posted on WP:ACN in order to perhaps draw additional attention to this important discussion. —Finn Casey * * * 20:47, 22 October 2009 (UTC)

By my reading, the "consensus" for the secret ballot was a fairly small one, and it's my guess that most members of the community who might be interested in the matter, have no idea it was even discussed. A better way to proceed might be to publicly announce that a secret ballot is planned for the election next year, and then make sure the matter gets a vigorous discussion, perhaps by asking all the candidates about it during the current year's election. --Elonka 20:50, 22 October 2009 (UTC)

• I did much of the mundane setting up of the various pages for last year's election. I think I'll take a back seat for this one. In most respects, however, the format for December 2008 seemed to work quite well (with the exception of "general questions", which, in retrospect, I would say were probably more bother than we needed). Additionally, I'd say: that Finn is quite right that we need to get a move on with setting up the pages; and that Elonka is quite right that a secret ballot for this year's voting would not be wise. AGK 22:13, 22 October 2009 (UTC)

I'm not sure we need a supermajority of the entire community to make this kind of change; it's been kicked around quite a bit, with pretty limited concerns expressed. If it works well with the AUSC election, and there are no major technical challenges to using it with ACE2009, then I see no reason to wait. The potential downsides are seemingly minor. Nathan ^T 22:36, 22 October 2009 (UTC)

The idea of a closed/secret ballot is an excellent one. Cirt (talk) 22:48, 22 October 2009 (UTC)

Springing a secret ballot on the community as a surprise would not be a good idea. For the record, I myself am strongly against the idea of a secret ballot, and was surprised when someone told me that it was even being considered for this year's elections. When I indicated surprise, they told me that there was consensus, and pointed me at the (already closed) RfC. But as I mentioned above, I don't think this RfC was as widely announced as it could have been. There was no Signpost article about it, and many editors who I would have expected to see at such an RfC, were completely absent. To try and push through a "consensus" about a secret ballot with such little community awareness about it, would not be wise, and might cause more chaos than it was intended to prevent. I'm reminded of the situation where there was a "consensus" to rename WP:V, but when the change went through, it caused a community uproar. So, a better option here is to proceed slowly and carefully. Make sure that the idea of a secret ballot is extremely well known, and give everyone time both to discuss their concerns, and also to get used to the idea. --Elonka 23:59, 22 October 2009 (UTC)

I like the idea of a Secret Ballot, if properly implemented with appropriate consensus, and I'll be watching the AUSC election with interest. I'm also a fan because it'd be less work and hassle during the election itself. As Risker notes, though, we really need a broader consensus before we implement. UltraExactZZ ^Claims ~ Evidence 14:03, 23 October 2009 (UTC)

I really don't agree at all with the assertion that this is being "sprung on the community". This RfC went on for a month and was pretty widely advertised. Going ahead with secret support/oppose voting, unless there are major problems with the AUSC election, makes perfect sense to me. Ultra, I think you misunderstand Risker. I read her comment as meaning that we would need a longer discussion on what the exact secret voting method should be (Schulze or support/oppose) after this years arcom election. — Jake Wartenberg 22:40, 23 October 2009 (UTC)

With all due respect, I disagree that the RfC was "widely advertised". Checking "What links here" shows a very small list.[4] There was one tiny mention, buried in a longer Signpost article, and one link from the Village Pump (policy) board. Plus, looking at the names of the participants who did go to the RfC, I'm seeing very few of the ArbCom regulars, which tells me that most of them had no idea that the RfC was even happening. To be "widely advertised", there should have been announcements on WP:AN, WP:AE, a full Signpost article, and a link on "Centralized Discussions". --Elonka 15:02, 26 October 2009 (UTC)

Agree that it could have been advertised more, but it was on the centralized discussion template. See here. You need to look through the history and (not always complete) archive to see what was listed there. Best practice, as I say at Wikipedia:Publicising discussions, is to make a note at the time (on the talk page) of where a discussion was publicised, rather than trying to work it out later. Carcharoth (talk) 07:16, 27 October 2009 (UTC)

Term lengths: On the matter of changing to two-year terms, it would be in everyone's interest to avoid the messiness of allowing individual candidates to choose for themselves, something that militates against an orderly process, clarity of purpose, and the appearance of commitment to the job. How would voters react to a declaration either way by a candidate? Hard to tell, which introduces an unnecessary imponderable into the process. There is apparently no significant opposition to two-year terms as a norm—not among current arbs, not by Jimbo, and not in the community. Jimbo has undertaken on his talk page to ask the arbs and take their advice. I advised him that a colleague and I were ready to go with an RfC on the matter, if that was necessary; no one, apparently, does think it's necessary. I believe we should work on the basis that it will be two years. Although it would be a lot of fuss, I am quite prepared to post a note at VP and Centralized Discussion at a moment's notice, asking for anyone who objects to pipe up and state their reasons on a dedicated page.

Seats vacant: Just checking: the election will "top up" the number of arbs to 18? This is as I understand Jimbo's response on his talk page when I put the question a while ago. 18 was the number after the last electoral "top up", including Deskana's seat. Eight "fallow" positions are listed above. There are 13 arbs currently listed. Is my arithmetic correct? How many seats, exactly, will be filled this time? [Later: yep, 13 minus 3 retirements plus 8 equals 18.]

Secret ballot. I thought it had been resolved that this election would be held as a secret ballot, as a matter of fair, clean process. I see no problem in implementing the secret ballot in the period between the close of Audit Subcommittee elections and the start of the ArbCom vote, provided there are no great cock-ups in the AS election. This is an important reform, and if I'd known its implementation for this election was suddenly going to be at issue, I'd have said something long before now. Tony (talk) 12:43, 26 October 2009 (UTC)

I put vacant seats back to "pending". Sorry for this guys, but the number of seats could still change, in the event that a sitting arb is transferred from one tranche to another (this HAS happened in previous years). This would change the number of seats being filled obviously. Arbcom are mighty busy right now, but I am hassling them to get a definitive answer. Manning (talk) 13:53, 26 October 2009 (UTC)

Jimbo has stated only a few weeks ago on his talk page that he will play a purely "ceremonial" role in this election. This was in answer to my request that we know before the election how many seats will be filled. Changing sitting members from one tranche to another is definitely not ceremonial. As far as I understand it, if a sitting member wants another term, s/he should stand for election. Tony (talk) 13:58, 26 October 2009 (UTC)

The situation is unclear. It could be argued that Arbcom have the right to transfer their members. Or not. It all depends on how you interpret Jimbo's transference of authority. This is why I'm trying to get it clarified ASAP so that the clerks can get the pages into order for the election. Manning (talk) 14:03, 26 October 2009 (UTC)

I think a formal RFC on the length issue would be good, if only to get an uninvolved party to close the discussion and document it as a best practices method for future changes to voting practices. I think that was done for the secret ballot (or maybe I'm forgetting something). MBisanz ^talk 14:08, 26 October 2009 (UTC)

Thanks, Manning. I've done some copy-editing of the associated pages. I made no substantive changes except for two. Two, I think, no one will object to: that candidate statements should not exceed 400 words, rather than that they should be less than 400 words (= 399 or less—seemed to be very fussy); and that where a statement does exceed the limit, it will be removed for reduction, not may be removed. It seems only fair to be strict about this, don't you think.

The third substantive change arises from an inconsistency, but will probably need to be OKed higher up. In at least one place, it says that Jimbo will "verify" the results. Yet on the "Voting process" page, it says that "Voting will run for two weeks (exactly fourteen days), and the results will be determined and announced by Jimbo Wales." I changed this to "verified", but please change it back if you think there's a problem. In the end, one term needs to be used in both places, I think. [5] Tony (talk) 14:32, 26 October 2009 (UTC)

Matt, an RfC would need to be launched very soon indeed. I suppose I'm willing to launch it, but can we sort out soon whether this is necessary? Can you ask the arbs? Tony (talk) 14:34, 26 October 2009 (UTC)

I can email them (which I've done), but Special:CompelUser isn't working this week :) MBisanz ^talk 14:40, 26 October 2009 (UTC)

Matt, Tony - I've managed to get in touch with a couple of arbs about these two issues (tranche changes, two-year terms) and a response from ArbCom on both should be forthcoming in the near future. (Bear in mind that some major arb work is underway at present). Nothing stops you from launching an RFC in the interim, but it *may* be moot. (Or it may not, hey what do I know?) Manning (talk) 14:55, 26 October 2009 (UTC)

Thanks, Manning. I'd rather wait if advice is forthcoming soon, but can launch it promptly if necessary. An RfC on term-lengths alone (not tranches) would be my preference—the simpler it is, the more likely there will be a clear-cut result and minimal risk of bloat.

Of course, consitutionally, Jimbo can simply say "ArbCom advised me that two years would be ideal, and I'm acting on their advice", since term-lengths are mentioned nowhere in policy. That is, in fact, what Jimbo suggested on his talk page. Tony (talk) 15:08, 26 October 2009 (UTC)

In the interests of transparency, I'm going to post my views here. I'm speaking here as a community member, with current experience of the arbitration committee, and as someone who thinks that ultimately the community should decide issues such as the size of the committee, and the term lengths, and the voting method.

• Term lengths - I think two years is a good term length. It is long enough to effect changes and to get up to speed in the first six months. Three years is too long, and two sets of two-year services (ideally with a break in-between) is better than two lots of three-year services. The problems arise with turnover and the need to keep the committee stable (having more than half its members change each year is not good). If you have elections to fill all vacant seats with 2-year terms, then the number of seats up for election depends on the number of resignations, and the number of departing arbs whose term has ended (though they can, of course, run for re-election). Ideally, you would have staggered terms (i.e. tranches), as this avoids situations where you have (say) eight seats up for election one year, six seats up for election the next year, and twelve the year after that (this is the current situation given the number of arbs on 1-, 2-, and 3-year terms, if all eight seats this year were appointed to 2-year terms). This means that to even things up, you need to have three people running for 1-year terms this year, and five for 2-year terms. Then you go back to 9 seats available each year for 2-year terms (i.e. effectively a 2-tranche system). If you stick with a 3-tranche system (and 3-year terms), you have a tranche of 6 already filled with those serving for one more year, and a tranche of 4 filled with those serving for two more years. Thus you would have elections for 8 seats, 6 being for three years, and 2 for two years. And then elections for 6 seats each year thereafter (plus resignations). The advantage of a three-tranche system is that you don't replace half the committee each year (that can also be a disadvantage), but the disadvantage is that if you have annual elections, you need three-year terms, which are too long for some. The advantage of a two-tranche system is that you only need 2-year terms, but the disadvantage is that you replace half the committee each year and if arbitrators resign the year they were elected, then the following election results in more than half the committee being replaced (i.e. less stability). It's not an easy thing to weigh in the balance.
• Size of committee - I think 18 is a workable size (though at the upper limit). 15 would also be workable (though if you have 2-year terms, you either need an even number of seats, or the theoretical number of seats available for election goes up and down by one each year). I think anything below 12 becomes unworkable when you have arbitrators either resigned, recused, or inactive/burnt out, as you can end up with as few as 5 arbitrators voting on an issue towards the end of the year, which is not good. So anything between 18 and 12 is workable, in my view, with a preference for a number between 15 and 18 to allow for resignations and inactivity/burnout.
• Voting method - I have reservations about secret ballots, but some of the arguments for this have swayed me. My primary concern is that the discussions and "endorsement" pages that will inevitably spring up will get out of hand. There should be vigorous discussions about the candidates, but it should be kept under control. I do think more discussion about the practicalities of a secret ballot are needed (the RfC endorsed the concept, but little was said about how it would work in practice). I would urge those following these discussions to follow the AUSC elections and help point out and correct any bugs or problems that arise.

So overall, 2-year term lengths, 18 on the committee, and secret ballot (for this year at least). With the caveat that three of the eights seats this year need to be for 1-year terms to avoid a pile-up of 12 seats for re-election in December 2011. And repeating here that this is my personal view, and that these issues are, ultimately, for the community to decide. Carcharoth (talk) 08:04, 27 October 2009 (UTC)

Only seven arbs at the start of this year were continuing (two of whom left soon after, and another after six months); nine were new. This seemed to work well and occurred under the three-year regime. Half of 18 is nine. I hope ArbCom will focus in the coming year on how to minimise the drop-out and burn-out rates. More delegation, streamlined hearings, structural reform, measures to protect arbs from abuse—these might all help. Given the uneven, unpredictable rates of dropping out during each year, the December elections for ArbCom have become a de-facto "top up" mechanism. Let's be realistic—it can only ever be that, and there's nothing intrinsically wrong with different numbers of vacant seats each December; as long as there are at least four or five continuing arbs on each turnaround, plus the supportive infrastructure of continuing clerks and functionaries, the system can be made to run more smoothly with two-year terms than it has done with three-year terms. Tony (talk) 08:59, 27 October 2009 (UTC)

I have created a shell of an RFC regarding term lengths and the number of seats: Wikipedia:Requests for comment/Arbitration Committee 2.
Before we get the RFC ball rolling, I think that we need to consider how to have an effective and quick RFC, due to time constraints, because I believe it is critical that this decision is made before the election commences. John Vandenberg ^(chat) 14:25, 27 October 2009 (UTC)

I have kicked off the RFC. John Vandenberg ^(chat) 09:25, 29 October 2009 (UTC)

Getting organised

OK, with elections just around the corner there is a need to get organised. I have some suggestions:

It would be good to organise a team of volunteers to keep election matters running smoothly. This team can coordinate on a user page somewhere (as these discussion pages will soon be full of election related discussion). Knowing what needs to be done and who else is available to help is a good thing. I'd nominate a subpage at User:Ultraexactzz as he/she seems to have already taken a leadership role in this year's election. (I haven't raised this with Ultraexactzz by the way, it's just my idea).

Also a few Arbcom clerks will be around to help. It should be noted that clerks have NO special responsibility/authority in elections, (they are run by the community and not by Arbcom). We can however be useful for getting in touch with Arbcom quickly if ever needed.

We also need to start to properly publicise the election.

Feel free to add comments or additional ideas. Manning (talk) 15:35, 26 October 2009 (UTC)

I am willing to give some time as an assistant. Tony (talk) 16:10, 26 October 2009 (UTC)

I am prepared to do the required donkey work. How about using a subpage of the Election page (i.e. here) for co-ordination? User subpages are harder to find and less enticing to contribute to for editors reticent to "trespass", and for posterity's sake it would be bad to have the lost to history as might happen if it were not attached to the election page.  Skomorokh, barbarian  16:30, 26 October 2009 (UTC)

 Done - Wikipedia:Arbitration Committee Elections December 2009/Coordination. I've added both your names to the list there as well. Manning (talk) 16:41, 26 October 2009 (UTC)

Agree, everything should be in one place, and we should keep in the WP space (mainly, so we can find it next year). I'll add myself momentarily. Publicizing the election usually takes care of itself; within the week we'll be asking the community for General Questions to ask all candidates, and the nomination process usually sparks a great deal of interest. Perhaps we should ping the Signpost as well? UltraExactZZ ^Claims ~ Evidence 17:11, 26 October 2009 (UTC)

 Done - Re the Signpost, I had actually already sent an email to phoebe before I went to bed last night. I also put some preliminary notices at the Admin noticeboard and at VPM. Manning (talk) 23:45, 26 October 2009 (UTC)

Questions Pages

Whatever sort of voting we do, we'll still be asking questions of the candidates. Last year, we accepted general questions for all candidates (here) from 3 November until 17 November. We then locked the page and transcluded these questions to each of the nominees at that time. Since nominations went from 10 November to 24 November, there was a week where new nominees got the questions right away - and there was a week where the candidates did not actually have general questions. Some candidates copied individual questions over, others prepared answers and waited for the 17th to post them.

One of the biggest headaches was making sure that candidates who answered some questions early ended up getting all of the questions. We couldn't just transclude, or risk losing the answers they had posted already. So here's my question - would there be an objection to moving Questions up a week? That means we'd close the General questions on 10 November, and all nominees would have all of the questions from day one. The downside is that this would mean we need to ask for questions Tomorrow, if we want to give two weeks for them as we did last year. I don't think this would be a problem, and the proper pages are already set up. Wikipedia:Arbitration Committee Elections December 2009/Candidate statements/Questions/General adds questions to Wikipedia:Arbitration Committee Elections December 2009/Candidate statements/Questions/General/List, which is later locked and transclude/substituted onto each candidate's question page, which uses Wikipedia:Arbitration Committee Elections December 2009/Questions header as a base. UltraExactZZ ^Claims ~ Evidence 19:40, 26 October 2009 (UTC)

tl;dr version - It'd be simpler to ask for questions for candidates starting tomorrow, if consensus exists to do so. Thoughts? UltraExactZZ ^Claims ~ Evidence 19:40, 26 October 2009 (UTC)

I agree that the proposed schedule would be an improvement. However, we urgently need acknowledgement from the current committee. —Finn Casey * * * 19:58, 26 October 2009 (UTC)

Jimbo's working on it, as is Risker and others. The issues they're looking at, though - the secret ballot thing and two year terms - wouldn't be affected by moving questions to tomorrow (or 3 hours from now, omw). nor do I think a watchlist notice is necessary; posts at AN, BN, ARBCOM (or ARBCOM/N), and VP - and CENT - would probably be sufficient to announce that questions are open for submissions. UltraExactZZ ^Claims ~ Evidence 20:57, 26 October 2009 (UTC)

I cant think of any reason to wait. In late October 2008 we had a number of "probable" candidates already answering some basic familiarisation questions developed by MBisanz. See the 27 October 2008 revision of his ACE2008 page, and the revisions for each candidate guide page as of the same day: CHL, Coren, Jehochman, Rlevse, Sam Korn, and Wizardman. John Vandenberg ^(chat) 10:34, 27 October 2009 (UTC)

 Done. I posted notices to the Village Pump, AN, BN, and WT:ARBCOM. I'll add it to CENT monemtarily. Thanks! UltraExactZZ ^Claims ~ Evidence 13:01, 27 October 2009 (UTC)

Canvassing

I'm concerned to read comments above such as "Canvassing is unfortunately inevitable" and "the disgusting canvassing in favor of one candidate". Canvassing, like socking, is impossible to stamp out completely, but it wouldn't hurt to remind people that this electoral process takes it seriously, would it? Something like this?

All candidates and voters are reminded that Wikipedia's WP:CANVASSING policy will be strictly applied in the election process.

It could go ... let me see ... underneath "Reminder to candidates regarding their statements"? Tony (talk) 04:45, 27 October 2009 (UTC)

Well for arbcom elections we do tolerate things we don't in other processes, such as userboxes indicating a person supports a particular candidate. Also, the scale of arbcom elections generally render canvassing impossible and just lead to long chases down foxholes. MBisanz ^talk 04:58, 27 October 2009 (UTC)

OK, will go with your advice. I was just concerned it might be a disruptive element, and that the election not be unnecessarily politicised. Tony (talk) 07:10, 27 October 2009 (UTC)

Be afraid! ;-)

There were a lot of useful ideas in that discussion; many should be revisited before we finalise the election rules for this upcoming election.

In that discussion almost everyone agreed there isn't much that can, or should, be done about canvassing. I suspect it will become more of an issue with a secret ballot because it will go unnoticed, but at the same time a secret ballot means that votes need not be affected by pressure from a group. People who are canvassed will be able to secretly break rank without fear of reprisals from those who are doing the canvassing.

Block voting will still go unnoticed where the bonds within the group are strong, however without the ability to say whatever they like publicly when they register their vote, they cant affect other voters unless they go to the talk page and express their concerns in a convincing manner.

John Vandenberg ^(chat) 10:13, 27 October 2009 (UTC)

John, you said "it will become more of an issue with a secret ballot because it will go unnoticed, but at the same time a secret ballot means that votes need not be affected by pressure from a group". I can't see how either assumption is correct. Correct me if I'm wrong: the idea of the secret ballot is that the votes are made public after the close of voting. Therefore, rewards and reprisals are possible for those who did or did not vote according to the wishes of a lobbyist (however, the entire vote is open to auditing afterwards, which is good).

Like sockpuppetry, which often cannot be proved, I see no reason why canvassing should be tolerated just because it can still be organised in secret. The least we can do is to post a strong reminder in at least one prominent place on the election pages. Heck, you have to be very careful even neutrally advising people that an RfC is being held on WP. As User:ChrisO said at your link above, "I'd think WP:CANVASS would apply in this as in all other community discussions.... I would say that [emailed lobbying for or against a candidate] undermines the integrity of ArbCom votes - they're not supposed to be proxies for fighting ethnic or cultural conflicts." Indeed. Tony (talk) 13:03, 27 October 2009 (UTC)

It isn't a secret ballot if the individual votes are published at the end. The WMF Board election votes are not released at the end; I see no reason why these election votes would be released at the end. I'll scream bloody murder if they are, because that was not what was discussed at Wikipedia:Requests for comment/ArbCom secret ballot.

A lovely note discouraging canvassing is nice, however there is no point in saying that canvassing isnt permitted unless we know what happens when someone does it. And they will do it. And they will swear black and blue that they didnt do it, despite the obviousness of it.

Are we going to deny people their voting right because they received a canvassing email? What about if they send an email?

John Vandenberg ^(chat) 13:18, 27 October 2009 (UTC)

Yeah, which is why the votes should not be public afterwards. Point taken, and see SV's comment above. Tony (talk) 13:43, 27 October 2009 (UTC)

SecurePoll ballots cannot be publicised. There is a public list of who has voted, and a list of what votes were recorded for each candidate, but there is no list, and no way to create a list, that combines both data. Fortunately. Happy‑melon 23:09, 28 October 2009 (UTC)

I vaguely recall, from voting in Foundation Board elections, that voters get a "receipt" of sorts where they can save their vote (in encrypted form) on their own computer. Am I misremembering that? I always thought that was designed so people could cross-reference their votes and make sure their votes were recorded correctly. If that is the case (and I'm likely talking through my hat here), then if every voter published their receipt, you could reconstruct the vote. But that is unlikely to happen, obviously. One question I did have is whether candidates can vote for themselves? This is normal in real elections, but traditionally hasn't been done in our open elections. Candidates who vote would appear on the list of those who voted, but it wouldn't be known who they voted for (you would have to believe what they said if they chose to disclose that). In the past, the issue of candidates voting for each other (and sitting arbitrators voting) has proved contentious, in that people who oppose each other in a public vote may end up having to work together. I suppose now the issue gets sublimated somewhat, if not completely resolved (the only way to avoid that completely is not to vote at all, and just be professional about who you end up working with, and to not be silly enough to ask your colleagues if they voted for you or not if they voted). Carcharoth (talk) 08:15, 29 October 2009 (UTC)

That is the encrypted-ballot feature of SecurePoll, which is designed to ensure that not even the server administrators can alter the votes cast; as you say, it is so that if the voters suspect that the host has committed fraud, the election can conceivably be reconstructed (although only with the agreement of all the voters). In these elections, the encryped-ballot system will not be used, as we have no reason not to trust the host (WMF) implicitly.

There is no restriction on candidates voting for themselves or each other, except that the fact that they have voted will be public knowledge. Happy‑melon 11:14, 29 October 2009 (UTC)

The ArbCom election is so widely attended, a single candidate may have 300-500 people commenting, that I am not all that concerned about canvassing. The canvassing problem is real on AFD debates with low attendance, and where recruiting 10 supporters might disrupt the discussion, but I don't see that having such an impact on the ArbCom elections. For very large scale polls like this, I think it is best to be liberal in accepting that people will have their opinions on whom the best candidates are, and that many will try to influence others to think the same way. Sjakkalle (Check!) 10:09, 28 October 2009 (UTC)

Most canvassing is negative, so 20 will make a big deal. YellowMonkey (bananabucket) 05:45, 30 October 2009 (UTC)

Special:SecurePoll monitors

Any idea who will be monitoring the votes at Special:SecurePoll to strike sockpuppet votes, double votes, etc? It has to be someone identified to the Foundation and ideally should be a disinterested party. I know I had mentioned the Meta crats and de.wiki arbs, but given that the de.wiki arbs have resigned en masse in the interim, I think the fr.wiki and nl.wiki arbcoms are the next most active. MBisanz ^talk 13:58, 27 October 2009 (UTC)

The extension has built-in mechanisms to prevent double voting, and to enforce eligibility restrictions. Not sure how the sockpuppetry issue is managed, or if its a significant problem - presumably if there is a relatively strict eligibility requirement (as there has been in the past) the risk of socks affecting the outcome will be pretty small. Nathan ^T 16:09, 27 October 2009 (UTC)

I'm nearly certain any bot operator would get at least two votes without anyone noticing (assuming they were ok being unethical), same goes for anyone with a public account or who has changed usernames by abandoning an old account. MBisanz ^talk 16:39, 27 October 2009 (UTC)

Specifying in the rules that accounts with the Bot flag are ineligible would solve one issue, and would be simple to check. Public accounts aren't an issue, so long as their votes do not overlap with the main account for that user (and so long as they have suffrage); that's something else that usually gets checked after the election. The other issue will be accounts blocked for sockpuppetry during the election - Sockmaster votes, then the sock gets caught double voting. Monitors would need to remove those votes from the count, as well. An aside - who is monitoring the AUSC election? UltraExactZZ ^Claims ~ Evidence 17:40, 27 October 2009 (UTC)

Well there aren't "formal" monitors since the last 3 elections were all on public pages and before that Jimbo/Arbcom appointed the monitors for the 2 previous securepoll elections. MBisanz ^talk 17:56, 27 October 2009 (UTC)

So, as a technical point, then - how does it work? Someone will have to go through the votes and verify suffrage, non-bot status, non-blocked status, non-sock status, etc. Who holds the bag? If we invite NL.Wiki's Arbcom to do it, for example, would one or more of us sit in with them as an english wiki liason? Can we set up our own secure poll, or does a dev need to do it for us? Have we made the request, to make sure it's feasible? I see a note at http://www.mediawiki.org/wiki/Extension:SecurePoll that the extension does not scale well with tens of thousands of users; we had 7200 votes two years ago, will we break securepoll? The board elections, by example, failed to breach 4000 votes. UltraExactZZ ^Claims ~ Evidence 20:10, 28 October 2009 (UTC)

If we do end up going with securepoll, I'm hoping Jimbo will appoint canvassers/monitors by fiat as a modification of his normal role, possibly drawing from AUSC and CU.--Tznkai (talk) 20:56, 28 October 2009 (UTC)

I would be very uncomfortable with English Wikipedia functionaries monitoring secret ballots. Surely other responsible wikimedians or external monitors would be preferable to (with all due respect) ArbCom-appointed members of the cabal-that-is-not?  Skomorokh, barbarian  21:05, 28 October 2009 (UTC)

Why? That means more identification, more spread of private data (in general) and decreased familiarity with whats going on. We regularly have AC clerks doing a lot the work. In the real world, election monitors are often partisans, elected officials with sufficient checks and balances, or they are from other jurisdictions in the area and thus still have an interest. I understand the general principle of disinterested parties, but there are perfectly good ways to do it without them. The problem with disinterested parties is they tend not to care enough to volunteer.--Tznkai (talk) 21:55, 28 October 2009 (UTC)

I think it is pretty important that whoever administers the elections not be involved with this project too heavily, and certainly not with anything to do with arbcom. These are secret elections; if functionaries, some of the most powerful people here, know who you voted for that almost defeats the point. — Jake Wartenberg 22:34, 28 October 2009 (UTC)

I agree with Jake. The whole point of a secret ballot is that people will feel free to support and oppose without fear of backlash. If there's a suspicion that those overseeing the election are interested parties who might use the information about voters' choices, it defeats the purpose, and we may as well not bother. SlimVirgin ^{talk|contribs} 11:22, 29 October 2009 (UTC)

To clarify a few points:

• The list of users who have voted is publicly viewable, although of course the way they voted is not indicated.
• No wiki user has the ability to see who voted for whom in a SecurePoll ballot. Votes are cast, and the list of who has voted is available. Election administrators can strike votes in this state, but they do not know which way the vote was cast. Once the striking process is complete, the software releases the final vote count: that is, the list of how many votes were cast each way. There is no ability to determine who cast each vote. "Election admin" is not synonymous with "has all data on the election".
• The SecurePoll for the AUSC elections is already configured and ready to go, you'll be seeing it in action in less than 26 hours time. Currently elections can only be configured by sysadmins; there is work in the pipeline to provide an on-wiki interface for creating polls, which will hopefully be finished before the ArbCom elections. If it is not, a sysadmin will configure the ACE2009 election manually as well.
• There is no danger of these elections placing undue strain on SecurePoll. The problems that were noticed during the licence update survey were caused by having large numbers (tens of thousands, as indicated) of votes cast, in a vote that used SecurePoll's encryption mechanism to provide triple-blind security to the poll (essentially, the collaboration of two of the three parties involved in the poll (WMF, the voter, and SPI who hosted the vote) was required to identify any vote); the large numbers of votes took too long to decrypt at the end of the poll, timing out the web interface that is normally used to tally results. As the encryption mechanism will not be used in these elections (redundant as the 'host' and 'scrutineers' are the same people), this is a non-issue. Also note that documentation at mw.org is often hopelessly out of date.

Happy‑melon 22:55, 28 October 2009 (UTC)

I've always believed the fear that people have has nothing to do with the "powerful users" but everything to do with what their friends will think of them. It certainly shapes editing patterns.--Tznkai (talk) 23:43, 28 October 2009 (UTC)

Secret ballot or not ? debate and endorsements

All seems confused, the election documentation subpages assume we go by public ballot, yet we had a RFC leaning towards a secret ballot, and the discussion above is not particularly decisive. Well, we need to sort this out.

In case we go by secret ballot; I think we should allow, actually encourage, discussion/debate on-wiki on the candidate. So in my opinion, we should deploy debate subpages - not just say go to the talk page, so with the system Wikipedia:ACE09/Candidate name/Debate or Wikipedia:ACE09/Debate/Candidate name.

Someone mentioned endorsements above, we could also have endorsement subpages, although they could arguably be merged in the debate subpages. I had also thought of dividing the election in two phases; debate/questions, then secret voting. Cenarium (talk) 02:07, 28 October 2009 (UTC)

I agree that the matter needs to be decided forthwith. I suspect a decision either way will surprise a significant portion of the community, and it is thus beneficial to make the final decision quickly. —Finn Casey * * * 02:56, 28 October 2009 (UTC)

I started preparing election pages on the basis of an open election, using last year's pages as a template, based on comments above that the AUSC election would be a test run for a Secret Ballot, and that - once complete - the experience from that election would be used to inform a community decision on whether that process is the way to go for ACE2010. I took this from Risker's analysis, above (here), though I may have read too much into that comment. There was an RFC on secret ballots, and the most supported option there was to use a secret ballot - though many comments also assumed a Schluze-style selection method, whereas the committee has recommended Support/Oppose voting (per Risker, here). There has been concern that the community did not have adequate input in the Secret Ballot RFC, or that it was inadequately publicized - which I can't speak to, as I was on wikibreak. We have this new RFC being put together, should a Yea/Nay Secret Ballot question be added to it? UltraExactZZ ^Claims ~ Evidence 19:45, 28 October 2009 (UTC)

I don't think so. I think we should go with secret ballots based on the previous discussions that concluded with support for the idea. I think the counting method is less crucial, and can be handled by a smaller discussion and consensus than an RfC. Since the RfC was pretty prominently advertised, and there was general approval for the idea, I'd rather not see implementation delayed for redundant approval. Nathan ^T 19:54, 28 October 2009 (UTC)

Candidate eligibility

Do we have a candidate eligibility description anywhere? Without explaining why I am asking such a dumb question (I have reasons, sadly), is there a page anywhere that explicitly says "Do not nominate yourself if you have less than 150 edits and/or will be under 18 on Jan 1, 2010"? Manning (talk) 03:11, 28 October 2009 (UTC)

I think there is not strict requirement, but no one with that few edits will be taken seriously. Steal from last year or if there isn't something, you could crib something from RFA, "Successful candidates are likely to be..." Thatcher 03:41, 28 October 2009 (UTC)

I would assume that if one is not eligible to vote, they would understand that they are not eligible to run. And per Thatcher, it appears to be more of a common sense matter. Newer users who are so bold as to nominate themselves for Committee candidacy are unlikely to be given any serious notice. That's not intended to put down the contributions of newer users - I myself just began getting actively involved in Wikipedia processes three months ago :) —Finn Casey * * * 05:06, 28 October 2009 (UTC)

The only formal requirements are that one identifies (or be willing to identify) to the foundation, and the age requirement (a legal issue, I think). Agree that no one with too few edits to vote would be taken seriously, and they can always be asked to withdraw. UltraExactZZ ^Claims ~ Evidence 05:42, 28 October 2009 (UTC)

One issue that may be relevant is that, using SecurePoll, the list of candidates cannot be modified once the election has begun: that is, candidates cannot be 'withdrawn' from the election once it has started (equally, new candidates cannot be added). A point for discussion, perhaps? Happy‑melon 23:01, 28 October 2009 (UTC)

My understanding wasn't that it couldn't be done, but that it needed a manual intervention to do. Perhaps I misunderstood? — Coren ^(talk) 23:35, 28 October 2009 (UTC)

Changing the list of candidates in any way, at any time, currently requires sysadmin intervention. Even when that ceases to be the case, it still won't be possible to change the list after the election has begun. Removing candidates would leave phantom, uncountable votes, which would break the current tallying system. I guess a system for 'locking' candidates could be implemented, where they would still be on the ballot list but greyed out so they couldn't be voted for, but even that might be problematic. Happy‑melon 11:03, 29 October 2009 (UTC)

One of the main purposes in withdrawing during the voting is to prevent the embarassment of a 6 support, 100 oppose result. That wouldn't matter here, since results would not be available during the election (presumably). Other candidates have withdrawn due to the timesink of the election (lack of time to be a proper candidate, realization that arbcom will take too much of their time, etc). If they can, they can - if not, I don't think it's a major issue. We did have one candidate exercise their right to vanish during the election, though - and their name was redacted from everything. That would prove difficult here. UltraExactZZ ^Claims ~ Evidence 00:45, 29 October 2009 (UTC)

The names on the ballot form can be changed (so an RTV candidate could be renamed to "vanished user"). Happy‑melon 11:03, 29 October 2009 (UTC)

The chances of a repeat RTV candidate are vanishingly small (see what I did there), but the fact that names can be changed on the ballot is good information to have. UltraExactZZ ^Claims ~ Evidence 12:17, 29 October 2009 (UTC)

Thanks for the replies. It was prompted by a bit of Wikilawyering I ran into from a minor. I know there is a statement from Jimbo somewhere that says he will not appoint anyone who doesn't meet foundation rules for CU/OS, which implicitly means you have to be over 18.

For reference's sake, The 2007 election states that eligibility required being over 18 and 1000 mainspace edits.I was travelling during the 08 elections so completely missed them, so I don't know why those conditions were dropped (not that it really matters). Manning (talk)

I think we caught the 1000 edit minimum last year, as well - but after nominations opened. It would be hinky to change the candidate requirements after candidates had already submitted, so we let it go. If there are no objections, I'll go ahead and add that back in today. We link to Jimbo's statement on eligibility somewhere, I'll have to track that down. The age restriction is listed (here, on the nomination page), as is the requirement to identify to the foundation (per a arbcom ruling in November 2007, discussed here). UltraExactZZ ^Claims ~ Evidence 12:14, 29 October 2009 (UTC)

"Single" general question

With virtually no publicity there are already a number of general questions - see Wikipedia:Arbitration_Committee_Elections_December_2009/Candidate_statements/Questions/General. However there are a number of cases where the "single" question rule has been ignored. In a few cases there are six or seven questions. I'm not sure of the accepted procedure here. Manning (talk) 04:39, 29 October 2009 (UTC)

I actually completely missed that prohibition, to my embarrassment. My notion would be to let the process run its course a bit longer, and then consolidate the repeat questions.--Tznkai (talk) 05:32, 29 October 2009 (UTC)

I am rather confused on what exactly this means. Does it mean one question per person, or one specific question to be asked just once (i.e. no duplicate questions). I first thought the former but if that was the case it was ignored last year too. Some clarification would be helpful, and if necessary the instructions should be updated to be more clear. Camaron · Christopher · talk 15:56, 29 October 2009 (UTC)

Is there a deadline for these questions? I'd like to submit a question, but don't want to influence an arb case that is currently open. SandyGeorgia (Talk) 16:00, 29 October 2009 (UTC)

I think, if I've understood the above, that the idea is for all questions to be in by the time candidates submit themselves as nominees. If so, the questions need to be in before November 10th. I doubt it'll influence the case if you want to ask, but then I'm not entirely sure which case you're referring to! :) Fritzpoll (talk) 16:02, 29 October 2009 (UTC)

Thanks, Fritzpoll. I do think it could influence the case, and I'm not sure the case will close before Nov 10 :( Perhaps one of the arbs following this page could advise me on how to proceed, or if I can submit a late question. SandyGeorgia (Talk) 16:11, 29 October 2009 (UTC)

The plan, I think, is to lock the General Questions page on 10 November. At that point, the list of questions can be posted on each candidate's questions page as soon as they nominate, and every nominee gets the same questions. You'll still be able to ask questions, but you'd have to ask each candidate individually. If you don't mind copying and pasting, that's your best option. UltraExactZZ ^Claims ~ Evidence 16:59, 29 October 2009 (UTC)

Thanks for the help; the top part of this page (the bottom is from last year still) is my solution. Hope this works, SandyGeorgia (Talk) 18:10, 29 October 2009 (UTC)

As far as the original issue - here's a possible solution.

• (a) notify the question writers now alerting them to the fact they can ultimately ask only one question.
• (b) let all questions stand for now until we get a sense of whether there are possible duplicates
• (c) On Nov 8 start a vigorous consolidation effort so as to get it all sorted by Nov 10.

Manning (talk) 00:09, 30 October 2009 (UTC)

Manning, to me it looks as though there are four issues already: (1) the sheer size of the body of questions, given that each candidate is expected to provide good responses and the voters to digest it all; (2) similarities/dupication; (3) a scattering of themes, such as desysopping; and (4) highly technical questions in which non-expert voters will have little hope of grasping the issue by reading either question or response (without expending considerable time and effort in reading link-targets). I am keen that the question/answer process be user-friendly for all parties. I'm less keen on a blanket restriction of one question per user, since some users have asked two good questions, and others have asked one that appears to be opaque. If the questions are to be recast with any semblance of thematic order, for easier digestion, I suggest that the headings be thematic and questioners' names be inserted in parentheses after the question. From a practical point of view, November 8 is too late to start rationalising, since back-and-forth communication with questioners might be necessary. Someone might be prepared to come up with a draft sooner than that, aiming for fairness and neutrality (BTW, I'm about to confer with LessHeard vanU on whether our two questions might be conflated). Tony (talk) 06:35, 30 October 2009 (UTC)

All very valid points Tony. I'm open to anything that works, frankly. With 43 questions (at my last count) we probably need to do something to make the process manageable, and soon. The "draft list of general questions" idea sounds very good, although it is a change of procedure. Are we also bold enough to suggest a cap on the total number of questions? Manning (talk) 06:55, 30 October 2009 (UTC)

What say I try a quick and dirty thematic presentation by cutting and pasting on a sandbox; then the similarities between the questions and the redundancies might be clearer, and we'd know whether/how to approach questioners with proposals for rationalisation. Further questions/themes could then be added to the sandbox.

It may turn out to be viable; it may not. Probably worth trying. Tony (talk) 08:28, 30 October 2009 (UTC)

The default questions was a problem for me last year. I had a major work project on at the time, and I found that a lot of the questions were duplicates, were not applicable to me, or were already answered in the mini-biog that was on my userpage. I tried to only answer questions which were directed at me[6], and I divided these questions into themes: about me, my candidature, my approach to Wikipedia editing practises and policies, my approach to Wikipedia administration, issues relating to the role of the committee, questions about specific arbcom cases, and "other". I gave up trying to add structure to the questions, and ended up trying to answer all of the general questions because people were opposing due to the fact I hadnt answered their question.

We should collaborate to build a single comprehensive questionnaire containing questions which everyone wants answered, and less emphasis on the person who posed the question. John Vandenberg ^(chat) 09:40, 30 October 2009 (UTC)

Yes. I've taken the first step by "coding" the questions into themes, here: now at least we know this is quite possible—in fact, it wasn't hard at all. The second task is to sift through the questions to identify savings, redundancies, the less important, the inappropriate. I'd have thought reducing it by half might be an aim; what do other people think? At the same time, the themes might change a little. Feedback would be appreciated on the thematic organisation. The themes are not yet in any particular order. Is it useful? Tony (talk) 10:13, 30 October 2009 (UTC)

• This is an excellent idea. Tony did a great job on the classification page. We definitely want to make the time-consuming process easier on candidates by eliminating duplicate questions! At the same time, we have to be sensitive to avoiding the appearance of inappropriately eliminating beneficial questions. A good balance in this matter is necessary, and I'm sure it will be achieved. —Finn Casey * * * 02:49, 31 October 2009 (UTC)

SecurePoll and approval voting

The past electoral method for ArbCom elections has been to give two options for each voter with regards to each candidate: support and oppose. However, the method of the AUSC SecurePoll which has been identified as a dry run adds a neutral option. Is neutral intended to be an option in the ArbCom election and if so what are the implications for determining the outcome?  Skomorokh, barbarian  00:51, 30 October 2009 (UTC)

As I understand it, the SecurePoll process requires a response beside the name of each candidate. The "default" vote is neutral, and voters actively determine whether to support or oppose a specific candidate. If they choose to do neither, then they leave the "vote" at neutral and it does not affect the outcome. Risker (talk) 01:03, 30 October 2009 (UTC)

We might consider changing it to "No Vote". Also, can one vote for a few candidates, then come back and vote for others? Say I'm not sure about one of them, and want to see answers to my questions, or whatever - can I change my "No Vote" to something else? UltraExactZZ ^Claims ~ Evidence 02:10, 30 October 2009 (UTC)

People are allowed to revise their votes. If you look at the list of voters for AUSC you will see greyed-out votes where people have submitted a subsequent vote; the system discounts the earlier ones and only counts the most recent. As well, each "ballot" lists the candidates in a different random order. The change of wording from "Neutral" to "No vote" sounds like a good idea. Risker (talk) 02:37, 30 October 2009 (UTC)

Responding to Skomorokh's original post, all voters had the "no vote" option as well in previous elections; literally meaning that they did not vote on the candidate's support/oppose page. With a ballot form to fill out, this just needs to be made explicit rather than implicit. Sjakkalle (Check!) 07:14, 30 October 2009 (UTC)

Risker is exactly right WRT the reason behind the third column. Perhaps "Abstain" would be better than "No Vote"?? Happy‑melon 07:42, 30 October 2009 (UTC)

I vote for "Abstain". John Vandenberg ^(chat) 09:41, 30 October 2009 (UTC)

Thanks all for the responses. There is I think significant difference between a neutral and an abstain/no vote option, and voters choosing "neutral" might well expect this preference to be represented in the outcome, when in fact if I understand correctly, neutral votes are not incorporated in the final tally (supports/opposes as a %). So I would agree with Happy-melon and John that "abstain" would be a much better wording. Another suggestion I would make for the same reason is to move neutral/abstain to the third position rather than its current place between support and oppose.  Skomorokh, barbarian  13:21, 30 October 2009 (UTC)

I should emphasise that these alterations are not possible now that the ballot has opened (otherwise what's to stop me changing "support" to "oppose" and "oppose" to "support"?), but they can easily be changed for ACE2009. Happy‑melon 17:34, 30 October 2009 (UTC)

No worries, ACE was my concern. I think most people who are addicted interested enough in the workings of the project to participate in the AUSC election will know the value of neutral when it comes to voting here. It would be great if we could put the rename on the agenda for whoever is handling the technical side of ACE2009/liasing with meta/MediaWiki (any volunteers?).  Skomorokh, barbarian  17:39, 30 October 2009 (UTC)

Voting method

On a project that frequently demands openness and honesty as an integral part of how things are run, I am struggling to see how secret ballots fit in with this. Personally, I am proud of my votes and I have nothing to hide in who I voted for. Why can't we keep what was not broken, instead of being all secretive about things? Majorly talk 17:24, 30 October 2009 (UTC)

I am also in agreement with Majorly that one of the core tenets of our project is openness, so secret ballots don't seem to fit in with that at all. I like public voting. I like seeing how other people vote, I like seeing their rationales, I like looking to see who is supporting whom. Especially when I don't have time to research a particular candidate in detail, I often scan for the votes of people I respect, to see how they are voting, and this helps me form an opinion of the candidate. It doesn't mean I'm going to vote the same way as my friends, but it absolutely helps me to gauge which candidates I want to do more research on, before proceeding with my own vote. With secret ballots I'm worried that we're going to degenerate to more of a system of, "I'm voting for this person because I like the way their name sounds" or "I'm not voting for this person because I've personally never heard of them." Secret voting provides LESS information about a candidate. And sure, public voting also has its problems, with canvassing and voting blocks, but I think the benefits of open voting far outweigh the negatives. --Elonka 17:50, 30 October 2009 (UTC)

I also agree with Majorly: I seem to remember people complaining that ArbCom wasn't transparent enough...changing the elections so that they're secret contradicts that and actually makes ArbCom seem less transparent. The old method worked fine, and things like canvassing, voting blocks, vengeance votes, etc. will still occur, except this time we won't be able to catch it like in the open elections. How is this a benefit? Acalamari 18:07, 30 October 2009 (UTC)

There is now a section on this at the new RfC: Wikipedia:Requests for comment/Arbitration Committee 2#Public v. Secret Voting, so that's probably the best place to move this discussion. --Elonka 18:41, 30 October 2009 (UTC)

Signpost article

Ok I've been in contact with Phoebe from the Signpost and so I'll be drafting the first "election report" starting in about 24 hours (it needs to be finished in approx 48 hours). Here's what I plan to cover:

1. General election news - timeline, number of candidates (see note below)
2. ArbCom RFC2 - the four issues on the current RFC are the biggest election news of the week
3. General questions drafting. Will discuss the more streamlined and consolidated method (ie. Tony's subpage)

If I've missed anything important (and crucial to this report) let me know. I'll post my draft for your comments/input as soon as it is done.

As regards the "number of candidates" I'm going to say eight, but just briefly acknowledge the Jimbo reassignment possibility.

Manning (talk) 02:44, 31 October 2009 (UTC)