Wikipedia talk:Requests for comment/Arbitration Committee 2

Voter eligibility
I think we should include the question of voter eligibility requirements in this RfC. It looks like the current plan is to allow votes from anyone with an account that was registered before 1 November and had at least 150 edits before then, but I think those requirements are too minimal; something like 500 edits and three months would be better, in my opinion. Everyking (talk) 02:34, 10 November 2009 (UTC)
 * I believe the current requirement is 150 mainspace edits before the deadline, which with a typical editing pattern, may be closer to what you are looking for. Newyorkbrad (talk) 02:43, 10 November 2009 (UTC)
 * I still think it's worth putting the matter before the community to see what level of eligibility requirements has the most support. Everyking (talk) 03:02, 10 November 2009 (UTC)
 * No objection from me, as long as people think there's enough time at this point to gather enough input and implement it before the election. Newyorkbrad (talk) 03:07, 10 November 2009 (UTC)
 * do you think there's enough time, brad? I reckon it's a bit late (without prejudice as to the merits of the idea) Privatemusings (talk) 03:22, 10 November 2009 (UTC)
 * That was my first reaction as well, but I'll leave it up to the volunteers who are organizing the election. Newyorkbrad (talk) 03:26, 10 November 2009 (UTC)
 * someone's organising this election? ;-) (are you sure we're not assuming that someone's at the wheel, when maybe we're all sat in the back?) Privatemusings (talk) 04:17, 10 November 2009 (UTC)
 * We are organizing it on typical wiki principles: whoever shows up and organizes. If everyking (or someone else) wants to change voter eligibility via RFC, the onus is on them to get that to happen.--Tznkai (talk) 05:47, 10 November 2009 (UTC)

< well ok, but what do you think the status quo is, Tzn? - if you're up for it, just make the call on how big the committee is going to be, who's eligible to vote, how the voting will work, and how long the folk elected's terms will be... I sort of feel these...um.. rather central details are a little vague at the moment, and that's a bit silly! :-) Privatemusings (talk) 06:09, 10 November 2009 (UTC)
 * Last time I checked, I had not staged a coup and declared myself Supreme Dictator of Wikipedia - as amusing as that might be for a few minutes at a time. As these are still technically advisory elections, Jimbo has the procedural ability to simply declare some things by fiat, and my best guess is his doing so would be roughly in accordance to the emerging consensus in the RfC (to which my best guess is 18 seats total, for 2 year terms, support/oppose voting, and possibly secret poll, but also possibly traditional public voting. There is a bit of a muddled mess there) and that there will be an acceptable bitching/moving forward ratio if and when he does so. Failing that, we'll simply just muddle through.--Tznkai (talk) 07:12, 10 November 2009 (UTC)
 * (e/c) What I can tell you is that Jimmy would much rather have all those things set and known in advance. He's been reading things up and consulting for a while, I expect he's been keeping an eye on this page and will eventually announce something on the topic &mdash; and I don't expect stunning revelations or surprises: most likely he'll go to two year terms and 18 seats as there is clearly consensus for both of those in the community and strong support within the Committee for both.  It seems also fairly clear that preferential voting is out given that it does not have very strong support (and that the necessary discussion about which preferential voting system to use&mdash;if any&mdash;has yet to take place at all). As for secret ballots...  well, this one is considerably more difficult to assess.  The community is pretty much split on the matter, so it's much less clear.  I know Jimmy starts with a slight preference against secret ballots, but I also know he's giving this serious consideration.  I expect that the fact that the AUSC election ended without a hitch, with good participation, and without problems will factor in any decision &mdash; he may be waiting for the announcement of results and postmortem from that before chiming in.  (Said announcement to come shortly, incidentally).  &mdash; Coren (talk) 14:40, 10 November 2009 (UTC)

I'm not too keen on adding another question so late in the game; it would not be proper if a handful of people got to decide the issue where scores of others participated in the older discussions. Is there reason to believe that suffrage is a contentious issue? Skomorokh, barbarian  14:30, 10 November 2009 (UTC)
 * A post-mortem (and results!) on AUSC elections would help a lot as far as settling the secret ballot/public voting question.--Tznkai (talk) 18:15, 10 November 2009 (UTC)
 * We just got the final okay from the scrutineers; I expect results to be posted today. &mdash; Coren (talk) 18:34, 10 November 2009 (UTC)

I've added the question to the RFC. As there's still a week before this RFC closes, I figured there's enough time to ask the question.  Rami  R  20:45, 17 November 2009 (UTC)
 * I'm not sure if we should be thinking of changing eligibility this late, but your question isn't quite accurate. You have current practice listed as 150 edits.  It's actually 150 mainspace edits.  Doesn't sound like much, but a significant difference.--Cube lurker (talk) 21:02, 17 November 2009 (UTC)
 * Indeed it is; fixed. It's true that it is late in the process, but this can be taken care of by issuing a sort of "last call to participate" in the RFC at the usual locations (AN, VP, etc.).  Rami  R  21:29, 17 November 2009 (UTC)
 * The requirement of 150 mainspace edits is perfectly fine. If it strikes you as "minimal", you've succumbed to editcount inflation. We need to set a basic minimum, but anything beyond that is an attempt to measure the worth of a Wikipedian by their edit count. rspεεr (talk) 03:19, 18 November 2009 (UTC)
 * Strongly agree with rspeer. —  Finn  Casey  * 04:01, 18 November 2009 (UTC)
 * But what about some time limit on these edits? Currently a user can make 150 edits and then never edit again other than to participate in an arbcom elections. Shouldn't it be required of voters to be at least semi-active community members? If secure poll is implemented, we won't even notice when retired users pop out of nowhere and vote. Such voters are not unlikely to be compromised accounts. And even if not, why should some who hasn't edited years or even almost a year be allowed to vote?  Rami  R  06:11, 18 November 2009 (UTC)
 * I can understand that concern. My objections to this proposal are twofold.  First, I believe it may be a solution in search of a problem.  Are there really people so interested in the election mechanism of our encyclopedia that they would go inactive, not even making 150 edits, but check in every year for the Committee elections? Really?   Second, I believe it is too late in the day to disenfranchise voters in this election.  However, I appreciate the serious consideration Rami R has given to this matter.  —  Finn   Casey  * 07:08, 18 November 2009 (UTC)
 * 150 mainspace edits strikes me as far too low. Tony   (talk)  07:10, 18 November 2009 (UTC)


 * I agree. It's a trivial matter to make 150 edits to articles. We've been plagued in the past by accounts with almost no edits turning up to vote, as well as by accounts with a substantial number of edits that haven't been used for ages. We should really increase the minimum number of mainspace edits, and also require that voters be active members of the community to some degree&mdash;so we'd be asking for a minimum number of edits overall, and a minimum number in the last year. And to answer Finn, yes, sadly, there really are. SlimVirgin  talk| contribs 08:50, 18 November 2009 (UTC)
 * Regardless, the requirement of 150 mainspace edits per year would disenfranchise me, Coren, Jimbo Wales, and many other Wikipedians. rspεεr (talk) 09:05, 19 November 2009 (UTC)


 * No opinion on the merits, but an element not discussed is the date on which an editor would have to have whatever level of edits is deemed franchise-worthy. For the present election cycle, the rule is 150 mainspace edits on or before 1 November - a date that has already passed. If we change the requirement, we risk having to tell editors that they were eligible, but now they're not. If we change the date, then we risk having editors rush to meet the requirment by editing the mainspace - and I'm not sure that's a great idea either. UltraExactZZ Said~ Did 19:17, 18 November 2009 (UTC)


 * Responding to what Finn said above: "Are there really people so interested in the election mechanism of our encyclopedia that they would go inactive, not even making 150 edits, but check in every year for the Committee elections? Really?" - yes, really. There were several editors who voted in the elections last year that had been inactive for all or most of the preceding year. Either they had continued editing under other accounts (and didn't want to vote with their now current account), or they had drifted away from active participation, and been drawn back to the election. One editor (once very high profile here, but now long inactive) explicitly stated (on an off-wiki site) that she only voted for a laugh, to stir things up. That is a weakness of the system that I think should have been addressed by a post-mortem of the election last year, or in this RfC, but (understandably) the drama of the elections last year led to a slump in activity surrounding these issues, and they were (by default) never dealt with. Carcharoth (talk) 06:25, 22 November 2009 (UTC)


 * The 150 edits (and it was any mainspace) is simply to avoid blatent socks. Less blantent are unlikely to be impacted by any reasonable edit count. The reason for the number is that there had been a number of recent AFDs whe requirment was 100 edits and wanted to knock out any left over socks from that. 150 is a lot of solid edits.©Geni 10:22, 29 November 2009 (UTC)

Tranch member replacement
Traches are much superior as they guarantee a stable transfer from one set of members to another. I'm not sure what is meant by "some arbitrators will be elected for one year instead of two" - all should be elected for the same term, the start times of those terms should be staggered into to groups. --rannṗáirtí anaiṫnid (coṁrá) 19:43, 13 November 2009 (UTC)
 * I believe it means that if open seats exist in because of a resignation, they're filled for only the remainder of the term, not for a complete new term. Same thing with expansion.  Of course its difficult to say what will happen because we don't know if the committee will switch to 2 year terms (or stick with 3 year terms) or how many arbs there will be.  Deserted Cities (talk) 20:23, 13 November 2009 (UTC)
 * I don't think either of those are in much doubt any more. The statement essentially notes that, if an arb is elected and vacates their seat within a year of their election, a by-election will be held to fill the seat at the next opportunity; with tranches, that by-election will be for a term of only one year, while without tranches, they would serve a full two-year term. That's the difference.  Happy ‑ melon  21:27, 13 November 2009 (UTC)
 * I'm even more confused now - but I don't think it matter, the essence is the principle of tranches, yes? So everyone (exceptions due to resignation aside) would serve for the same turation duration, the period of those terms would be staggered (presumably into two or three groups)? I think some thought would need to go into how to replace (I don't think it is a good idea for replacements to serve beyond the natural period of the member they replace, and maybe a system of co-opting member and/or leaving seats vacant depending on the length of time until the next election for that seat would be sometimes appropriate). --rannṗáirtí anaiṫnid (coṁrá) 21:38, 13 November 2009 (UTC)
 * Essentially there are three options when a seat becomes vacant very early in its term (within the first year for two-year terms, which we seem to be heading for). Option one is to leave the seat empty until its original term would have expired. On a committee chart, that would look something like this: The other options are to repost the seat for election at the earliest available opportunity (ie the next election, even though the seat would not normally be up for renewal then).  And here we have two options: either we keep every arbitrator's term the same length, which results in imbalances in the number of seats up for renewal at each election: Or, we insist on fixed tranches (ie the times at which a seat comes up for renewal are fixed), which results in arbitrators having variable term lengths: In a small example like this, none of the systems demonstrates a clear advantage over any other; a larger committee example would explore the possible permutations more thoroughly.  I hope this clarifies the distinction between them, however.  Happy ‑ melon  22:38, 13 November 2009 (UTC)


 * Yup. That's what I was thinking. It's something that can be left until later to decide anyway. It's not enormously pressing or would have a great influence on any decision - a similar problem exists whether there is a tranch system or not. --rannṗáirtí anaiṫnid (coṁrá) 00:20, 14 November 2009 (UTC)


 * It affects this election, though, because it dictates the lengths of the terms that are filled. If we retain a system of tranches, some Arbs elected this year need to serve only one-year terms, while if we abolish tranches, they should serve two-year terms.  It doesn't affect the way people will vote in this year's election, but it does affect the election itself, so we need to decide soon, preferably before people start voting.  Happy ‑ melon  08:23, 14 November 2009 (UTC)
 * Different term lengths are a needless complication, especially now that Jimbo has said he intends to play "a purely ceremonial role". The notion that a pre-existing "slot" has to be filled by a replacement until it expires involves another layer of decision-making as to which candidates fill the second year of a pre-existing two-year term, and which candidates will take up a fresh two-year term from scratch. This is all too much. The simplicity of fresh starts for every successful candidate is natural and compelling. Can you imagine getting consensus for a set of complicated rules as to who fills what "term space"? Tony   (talk)  10:25, 14 November 2009 (UTC) PS Why are different colours used in the charts above? That is tranche mentality, and makes the charts themselves complicated.  Tony   (talk)  10:27, 14 November 2009 (UTC)
 * The colours are purely to show the candidates appointed as a result of each annual election; otherwise the bars would all run into each other and be indecipherable. How on earth do they entail a "tranche mentality" when they can be equally-effectively used to illustate a non-tranche election process?  Happy ‑ melon  12:35, 14 November 2009 (UTC)
 * Colours: yep, you're right. Tony   (talk)  12:50, 14 November 2009 (UTC)
 * Yes, to tranch or not to tranch, that is the question - how do we fill posts left by resignations (regardless of the answer to that question) I mean is another question again. --rannṗáirtí anaiṫnid (coṁrá) 10:27, 14 November 2009 (UTC)
 * If the answer to "how do we fill posts left by resignations?" is "leave them empty" then you automatically have a tranche system; you're saying that the 'seat' has a term independent of the person who's sitting in it. If you elect to fill empty seats whenever possible, then you have the choice of keeping the tranche system, or dropping it.  So the questions are somewhat intertwined.  Happy ‑ melon  12:35, 14 November 2009 (UTC)

Canvassing with open versus secret voting
I've seen the issue of canvassing brought up in a couple of places that now seem to be archived. I'd had the thought that canvassing should be prohibited (and maybe even discouraged) only if the election is open. Mostly there is the issue that if the ballot is secret, and there is no good way to detect the effect of canvassing, then there is a problem that the policy becomes a more direct disadvantage for those who follow the rules. But, I think this issue may also relate to the size of an election; it seems that open voting is most often for small communities where you want accountability from the voters, and you’re concerned about the integrity of every vote. This contrasts with secret balloting, which seems to suggest more of an interest in getting the most wide-spread, unconstrained opinions. Also, there’s the thought that canvassing may be less influential in a secret ballot where people may feel freer to ignore it. Mostly the first point suggests to me that if Wikipedia goes fully democratic in these elections, with secret voting and the rest, then it should probably accept the prospect of campaigning as well. Mackan79 (talk) 00:12, 15 November 2009 (UTC)
 * To me, the restriction on canvassing seems to be motivated by a desire not to make of every small debate on a single issue a (personal) battleground, and to limit the potential influence of (small) groups of editors who might engage in "vote hopping". For the arbitration committee elections, where participation is much larger than in any particular AfD etc. discussion, these arguments, in my view, do not apply. I therefore see no reason to limit canvassing with regard to the arbitration committee elections. Cs32en  00:38, 15 November 2009 (UTC)

ArbCom responsiveness and workload
(In response to Requests_for_comment/Arbitration_Committee_2.) I will be very blunt and honest: People need to realize that arbitrators are volunteers with real lives and professions. The main problem is an absolutely ridiculous workload combined with the obligations of the position. Delegating work and establishing subcommittess can help, but it will in no way resolve the full problem (and to believe otherwise is a bit Pollyanna in my opinion). The oft-mentioned idea of using sub-panels for cases and the like sounds great, but in practice will almost certainly create more problems than it solves (including more delays and sprawling bureaucracy).

There are only two sure fire ways the speed and responsive issues will be handled, and both are extremely unlikely to occur.

First, the community can shoulder more responsibility and off-load substantial portions of the work that ArbCom bears. This potentially includes (but is not limited to or necessarily inclusive with) reinstituting some form of community ban process, instituting a community-based ban appeals process, permitting administrators greater leeway, implementing some form of binding mediation, increasing the robustness of our dispute resolution processes, and so on. If you have not noticed, it is an easier time pulling teeth from a lion than getting any such measures through to community approval.

The other potential solution is to pay arbitrators a firm, professional wage to allow them to devote themselves full time to the position. That has about a snowball's chance in hell.

Continuing down the path ArbCom has this year will help alleviate the issues, but the only remotely realistic solution is for the community to step up and develop better processes. Indeed, it is my opinion that the need for ArbCom at all indicative a failure in community processes. One cannot even state that privacy matters require an ArbCom to handle, as we have OverSighters and OTRS volunteers who deal with private, sensitive information day in and day out. Of course, all my opinion and you're welcome to some grains of salt. Vassyana (talk) 10:19, 17 November 2009 (UTC)


 * See also Areas_for_Reform and other sections trying to deal with those very topics.  In fact, most of that page is relevant to the discussion here. Collect (talk) 11:37, 17 November 2009 (UTC)
 * That is a very helpful link! Thank you. Vassyana (talk) 03:57, 18 November 2009 (UTC)


 * Vassanya, it seems to me like admins have been given a lot more authority in the past year or so. Stuff gets resolved on ANI every day by singlehanded admins and brief discussions, that would have taken arb cases not that long ago.  I'd hope that would decrease the arb workload.  I also very surprised when someone said the number of arbcom desysoppings of misbehaving admins has been steady at about 10 per year for the past 4 years.  It's obvious that the level of misconduct that it took to desysop someone was much higher 4 years ago, so if the number of desysoppings is staying steady, it means the admins are behaving better and therefore can be trusted with more responsibility.  As for paying arbs, that won't fly, but it had occurred to me that AUSC should perhaps have instead been a paid auditor from completely outside Wikipedia, hired under nondisclosure by the WMF so they are independent of any editor factions, and unlikely to leak private info.  It seems to me that AUSC is potentially subject to the exact same drama as the bodies that it's supposed to audit. I don't get what you're saying about privacy.  There will be disputes involving private info.  Are you saying OTRS or Oversighters should handle them?  That's just another arbcom under a different name.  69.228.171.150 (talk) 12:46, 17 November 2009 (UTC)
 * Vassanya: substantial, immediate savings could be instituted from January onwards by establishing a set of rosters to distribute arbs' time and effort more judiciously.
 * All rosters could rotate in a staggered fashion (i.e., the personnel gradually changes in each area throughout the year). Each area would have a designated number of arbs. Arbs could participate in more than one roster simultaneously, depending on Committee requirements and their own preferences.
 * Rosters should be adaptable—at short notice if necessary—to allow for unplanned burn-out, sickness, and RL exigencies of individual arbs. For this purpose, it is probably a good idea to have two or three arbs on stand-by or on a light roster (such as serving on just one subcommittee, or having a rest and contributing to WP in other ways). One Arb would probably need to run the roster (the Coordinating Arb or a delegate), to negotiate and update the rosters frequently. A Clerk could probably ease that task by taking some of the burden.
 * One roster for assessing applications for hearings (6–7 arbs?); one for the hearings themselves (7–9 arbs?); one for dealing with appeals (6–7); one for service on this subcommittee three on that subcommittee; one for liaising with CUs and other functionaries; some on stand-by.
 * This is simple in concept, but might take a little care in setting up and running. The rewards would be to dispense with the horrific workload and glasshouse atmosphere that has pervaded the Committee under the all-hands-on-deck approach, where the default is everyone doing most things at the same time ... and burning out and not enjoying WP. Tony   (talk)  13:22, 17 November 2009 (UTC)


 * I'd love to hear a more detailed analysis of what led you to the conclusion that The oft-mentioned idea of using sub-panels for cases and the like sounds great, but in practice will almost certainly create more problems than it solves (including more delays and sprawling bureaucracy), Vassyana. I very much support assigning cases to panels rather than try to do everything en banc, and I'm quite sure it's not only workable with a little care but should normally at least double or triple our capacity (as well as increase our resilience to absences). It's obvious that you're working from different assumptions and I'd like to compare.  &mdash; Coren (talk) 16:01, 17 November 2009 (UTC)


 * Finn Casey hits on part of it. It is worth bearing in mind that most real life analogues involve another layer of review/appeal and/or an avenue for en banc reviews. Thus, people are naturally going to desire a similar layer of bureaucracy if we switch in such a direction. It also would require significant and deep changes to the arbitration policy and process. It is simply one of those things that sounds wonderfully simple in the idea stage, but can be maddeningly complex to implement. My experience with similar divisions of labor within non-profits also motivates my deep skepticism. Certainly, this could be purely anecdoctal and not accurate as a predictor. I am certainly open to giving it a try. I am just extremely skeptical of its wisdom. (That is not to say I do not think it is a good idea, but rather that I am wary of its pragmatic reality.) Any such attempt would need to be crafted with extreme care and set up to be easily reversible in case it does more harm than good. Vassyana (talk) 03:57, 18 November 2009 (UTC)


 * I cannot comment on the practical convienience of panels, however I do note that many editors oppose them on principle. Many believe that the community elects the entire Committee to hear each case - any change to this principle would require a broad consensus that I do not believe exists. —  Finn   Casey  * 03:28, 18 November 2009 (UTC)


 * I think you are correct guessing there would be much hesitation from the community, but it's difficult to determine how much of that resistance is objection to the substance and how much is simple inertia. Wikipedia has become increasingly resistant to change of any sort (and that in my opinion, is the greatest danger that looms over us: fossilization).  &mdash; Coren (talk) 03:47, 18 November 2009 (UTC)
 * I would tend to agree about fosilization. Please note my comments about community reforms/processes and their likelihood of passing. Vassyana (talk) 04:02, 18 November 2009 (UTC)
 * Arbs are not elected on the basis of who will do what on the Committee. Arbs recuse, take time off, all the time. The community has had and will continue to have no reasonable expectation that some kind of all-hands-on-deck model will continue to be the norm. We elect ArbCom to do the job as best it sees fit within its constition (= policy page), which specifies nothing about the allocation of duties. What we do need is ways of keeping arbs in touch with the community and the business of editing and adminning out there, minimising burn-out, andensuring that arbs can do the job reasonably comfortably in their lives on- and off-wiki. A roster system that does this is urgently required. We do not need anything like 18 arbs on every single case. Tony   (talk)  07:17, 18 November 2009 (UTC)

Better committee organization, better buffers from harassment/trolling, and better Community policies are the answers to the arbitrator burn out. I think dividing the workload more as the Committee did this year is helping. I'm not at all sure that a panel for full cases is the solution. An ArbCom case rulings need to be consistent and be the final word on the matter. With panels that is not going to happen because appeals will be made to the full Committee, and different panels will have inconsistent rulings. And lack of action on one of the panels from arb inactivity could stall decision making more not less.

Some of the changes made this year will not be fully realized this year. The AUSC and Functionaries-l as working bodies are still finding their feet. Increasing the number of non-arb CU and OS is also slowly happening and will greatly help refocus arbitrator. Stricter enforcement of the arb inactivity policy this year has helped close cases that would not have been closed in the past. This still needs tweaking to make it more effective.

I can not emphasis enough the significance that the Community's poor policy making and policy review process has on Arbitration Committee work. The process is horrible broken and will continue to plague every area of the wiki including ArbCom. FloNight&#9829;&#9829;&#9829;&#9829; 10:23, 18 November 2009 (UTC)

Why only one ArbCom? Why not 2,3, or more? All of the arbs seem to agree that there is too much work, and I've read various comments that good candidates are dissuaded from running simply because the job is too drama-laden and is spotlit. Having multiple committees would lower both the profile and the workload of each arb. The process would become more mundane and more efficient. The post would become more appealing to those who simply want to get on and do the job, and less appealing to those attracted to the limelight. Arbcom is effectively a court. Is there any reason we cannot have more than one court sitting at a time?MoreThings (talk) 13:35, 18 November 2009 (UTC)
 * See my reply directly above your comment. I think having multiple bodies would cause another body to be formed to review the work of ArbCom. Instead, I think that more of the work can be spun off onto the Functionaries, admins, or other on site Community processes. We are on the path to this now. FloNight&#9829;&#9829;&#9829;&#9829; 13:43, 18 November 2009 (UTC)
 * FloNight, I'm not suggesting a dispersion into multiple bodies: rather, a roster to allocate arbs' workflow to minimise burn-out and maximise the reach of the Committee at the same time. It is not a good use of arb resources to have all hands on deck for every hearing. No judicial body does that. Tony   (talk)  06:59, 19 November 2009 (UTC)
 * Umm... U.S. Supreme Court :) —  Finn  Casey  * 17:16, 19 November 2009 (UTC)
 * The US Supreme Court has ?seven members. ArbCom has 18. Room to move. Tony   (talk)  09:54, 22 November 2009 (UTC)
 * US Supreme Court has 9 not 7 :) Davewild (talk) 10:06, 22 November 2009 (UTC)


 * Thanks, but nine is manageable, and I bet not all nine sit on all cases; even if they do, it's their full-time, well paid job. 18 busy WP editors is a crowd by comparison, and they have real lives to lead as well as expertise to continue giving to the project. Flexibility of assignment is entirely appropriate, and can be done in a way that should not attract accusations of stacking or cooking particular cases; arbs become inactive now, and recuse, and in principle a roster is the same thing, only on a larger, official scale. Tony   (talk)  10:14, 22 November 2009 (UTC)
 * They all sit on all cases (except when they are ill or recuse). Judges of the U.S. Circuit Courts on the other hand do not all sit on all cases. Ruslik_ Zero 15:04, 22 November 2009 (UTC)
 * I don't expect all 18 people to vote on all cases. People should move to inactive when outside events prevent them from participating. And if another arbcom task is keeping them very busy then not doing case work is fine as long as the arb move to inactive. But moving to smaller panels for case work is not going to stop either of these from happening on the panels, too. So we run the risk of having less excess help built into the system for particular case work tasks. And there is also the problem of different panels having conflicting decisions that would need to be reconciled through appeals. So, I don't think this really a strong solution. FloNight&#9829;&#9829;&#9829;&#9829; 19:30, 24 November 2009 (UTC)

Closing
Seeing as this was supposed to close 3 and a half hours ago, I was bold and did it myself. Good luck to whomever is asked to sort it out. Deserted Cities (talk) 03:35, 24 November 2009 (UTC)


 * See Wikipedia_talk:ACE2009.  Skomorokh,  barbarian   03:49, 24 November 2009 (UTC)


 * Who evaluates those polls of this RfC that don't have a consensus? A Horse called Man 16:01, 24 November 2009 (UTC)


 * I did, per below. For Eligibility, it defaults to status quo. For Mandatory minimums, we default to nothing (since there is no official status quo, though precedent exists, as I note). For Tranches, I think the question might be moot - but I left the issue open. UltraExactZZ Said~ Did 17:42, 24 November 2009 (UTC)

Requests for comment/Arbitration Committee 2 is now Closed
I have closed this Request for Comment. My detailed review of the issues and the results of that discussion may be found here. To summarize, I found that consensus exists as follows:


 * The Arbitration Committee shall consist of 18 Members elected to 2 Year Terms.
 * Arbitrators will be elected by Secret Ballot using the Securepoll extension.
 * Ballots will invite editors to Support or Oppose candidates.
 * Voters must have 150 mainspace edits before the election cycle to vote (Status Quo)

Questions or comments may be posted at The RFC's Talk Page. Thank you to all who participated. UltraExactZZ Said~ Did 17:42, 24 November 2009 (UTC)


 * 57% constitutes a measurable consensus? Good grief; that standard must've been invented sometime this year while I was inactive. Ncmvocalist (talk) 18:25, 24 November 2009 (UTC)
 * IMMA go w/ Ncmvocalist. 57% is not consensus.  If this securepoll thing was baked in already, then whatever, but last year was open voting, so how does 57%--> change status quo? Protonk (talk) 18:38, 24 November 2009 (UTC)


 * It comes on top of another RfC on the same subject, with even greater support for private balloting, so UltraExactzz has made the right call.  Roger Davies  talk 18:50, 24 November 2009 (UTC)
 * Um, what? The matter was reexamined in this RfC, and the results are different in this RfC. I'm not sure how it is even possible for an administrator, let alone an arbitrator, to absurdly jump to the conclusion that these results can be combined or are remotely the same - they're not. Ncmvocalist (talk) 19:13, 24 November 2009 (UTC)


 * Not that playing percentages is a worthy exercise, but one might ask in kind how 43% constitutes consensus for public voting. I don't have a position on the transparency issue either way, but this sort of inertia is at the root of the project's institutional sclerosis.  Skomorokh,  barbarian   18:54, 24 November 2009 (UTC)
 * No one said it was consensus for public voting; what is being said is that there is no consensus either way for a change - so the status quo stands until there is a consensus to change. That is the way this project has worked for many years, whether you like it or not, and it still works that way, unless, like I said above, there has been a revolutionary change to the entire system this year while I was inactive (which I'm unaware of somehow). Ncmvocalist (talk) 19:13, 24 November 2009 (UTC)
 * Inertia is overstated as a problem here. But you can go ahead and treat inertia in any organization as a cause of sclerotic behavior. But the response is that not changing mechanisms or social norms radically has a value in itself. Protonk (talk) 19:15, 24 November 2009 (UTC)


 * Agree with the close. And completely agree that stagnation comes from requiring unrealistic levels of support for change. In this pure up and down vote on a simple idea it would make no sense for 43% of the people voting oppose to prevent change from happening. FloNight&#9829;&#9829;&#9829;&#9829; 19:19, 24 November 2009 (UTC)
 * (e/c)If we flip this around, ultraexactzz will have to essentially declare that the discovered consensus of the previous RfC was invalid and support the 43% a more accurate gauge of consensus, which will still get irritated responses. Consensus is discovered not counted through the application whatever rules people have imagined there are. RfCs provide clues, not clear cut answers.--Tznkai (talk) 19:21, 24 November 2009 (UTC)
 * Another thought, related to the above. This is a wiki, not a government or civic society. - we change things by doing them, not by asking permission. I really don't know why the status quo gets any sort of deference.--Tznkai (talk) 19:23, 24 November 2009 (UTC)


 * One has to wonder why the turnout for this RfC was so starkly different to the previous one - perhaps this reflects a broader consensus by the community, especially given how soon the matter was reexamined. How often has that happened in the past - that a matter is reexamined again in another RfC so soon? That should provide a clue in itself. In any case, that doesn't change the Committee's appearance of wanting to change the way consensus has been interpreted at RfCs (let alone other processes), or a bid to change this into a mere poll that it is not. There's nothing "unrealistic" about expecting consistency. Ncmvocalist (talk) 19:31, 24 November 2009 (UTC)
 * What does the Committee have anything to do with Ultraexactzz? Nothing when I last checked.--Tznkai (talk) 19:37, 24 November 2009 (UTC)
 * Who said the Committee had anything to do with Ultraexactzz? No one, last I checked. I would think a general observation about how the Committee may appear is relevant to an RfC titled "Arbitration Committee 2". Ncmvocalist (talk) 19:41, 24 November 2009 (UTC)
 * Again, a preference for the sq reflects conservatism, not the characteristics of a governing body or a mandate from some social contract. I'm saying there is inherent value in deferring to the past simply based on continuity.  And honestly accepting 57% (hate returning to the percentages, but there isn't much recourse) as "consensus" doesn't square with any practice on en.wp I'm familiar with (and in relative terms this is even lower than the flagged revisions trial, which became a little notorious). Protonk (talk) 20:04, 24 November 2009 (UTC)

I suppose that if the result here remains disputed, we could refer the dispute to a panel of arbitrators ... but how would we pick them? Newyorkbrad (talk) 20:17, 24 November 2009 (UTC)

Consensus is largely arbitrary and inconsistently applied. Ultraexactzz made a closely reasoned and defensible close, which is all we can ask of an admin, notwithstanding that similarly closely reasoned, defensible, and legitimate closes could have been made for the opposite outcome. Support close as being within the discretion of the closing admin. Steve Smith (talk) 20:20, 24 November 2009 (UTC)
 * If you produced evidence to substantiate that (such as a previous RfC like this), that would make sense and there would be less of a dispute over this - that is how I'd worded my response. In particular, anything to invalidate the point Protonk has referred to above? If not, that isn't "within discretion"; that's a radical departure from Wikipedia norms. Ncmvocalist (talk) 20:33, 24 November 2009 (UTC)

Some thoughts.
 * When I was reviewing the RFC, the key point of contention was that, with the Secret/Open Ballot issue, a No Consensus result would default to the option supported by fewer editors, which was problematic. If this were a poll, then Secret Ballots had a majority in favor. Since it's a discussion, the weight of opinions matters more - and, on that point, I found a broader range of rationales for a secret ballot than for an open one. The fact that there were more editors favoring a secret ballot was not inconsequential, either, but it was not the critical factor - as I note in the close, it's the breadth of support that was key.
 * I very specifically did not use the previous RFC as a precedent, nor did I base my decision on it - but it did inform some of the arguments here, and several editors referred to it in their statements. This RFC was opened because it was felt that the other RFC did not have the level of participation necessary to effect change, and because other issues needed to be addressed (Term length, et al). It was widely announced, by myself and others, to get a broad view of the community's position. I believe it did that.
 * I'm unaware of any agenda with regard to consensus at RFC (Let's be honest - this closure brings the total number of RFCs I've closed to precisely 1), nor has any member of the committee counseled me on their positions or intentions with regard to the result of this RFC, or on how I should close it, despite the fact that I openly and publicly offered to close it a week ago. This was me sitting down and reading through it, making notes, looking at who said what, and distilling 270k of debate into a 5k statement. UltraExactZZ Said~ Did 20:37, 24 November 2009 (UTC)


 * I'll add one more - maybe it wasn't the greatest idea to include the 57% number in the close, as that seems to be the most controversial element cited above. In my defense, I presumed that you all could do math as well as or better than I, so I figured I'd save you the trouble. UltraExactZZ Said~ Did 20:48, 24 November 2009 (UTC)
 * I also have strong concerns that 57% is being declared as "consensus", especially as it's going to cause us no end of problems when we're trying to settle other disputes on Wikipedia. I can hear it now in AfDs, straw polls, RfAs, etc.:  "Look, 57% is consensus, all these arbitrators said so."  (sigh).  Or in other words, no, 57% is not consensus, and an RfC is not a "vote".  I would have thought that arbitrators who are the pinnacle of our dispute resolution system, understood this.  Let's not use some people's desire for a secret ballot for ArbCom, as an excuse to go trampling on some of our core tenets of how Wikipedia dispute resolution is supposed to work.  :/ --Elonka 21:06, 24 November 2009 (UTC)
 * Ultraexactzz has made it quite clear repeatedly that he did not consider a particular percentage of editors in favour of the secret ballot to be the determining factor; it would be a complete misrepresentation of the closure to characterise it as declaring 57% to mean consensus. The lining up of editors on this consensus non-issue along partisan secret-public lines does nothing to convince that this close was in any way improper.  Skomorokh,  barbarian   21:11, 24 November 2009 (UTC)
 * (e/c)The problem, Elonka, is that there is no line. Is 66% a consensus?  Is 75%?  (All those numbers have been rejected in the past as well, mind you).  88% isn't deemed enough for an RfB, but AfDs are routinely closed with slim majorities or (rarely) even minorities.  Consensus isn't about the numbers, but about the reasoning and the strength of arguments &mdash; though we'd be lying if we said we never also take into account the raw number of people in agreement or not.  If we insist on near-unanimity for any change, we might as well throw in the towel now. The point is, someone uninvolved (and, I'll insist, uncontested) closed the discussion and evaluated all those things  and reached a conclusion.  Unless we start accepting that results may sometimes be more marginal, the whole exercise of holding those discussions is completely and irremediably futile.  &mdash; Coren (talk) 21:15, 24 November 2009 (UTC)
 * Actually, UltraexactZZ has been engaged in this discussion, if you look at this page history. Note I'm not accusing UltraexactZZ of acting in bad faith...  But just as with AfDs, we prefer that a completely uninvolved admin close the discussion. Perhaps we should take this matter to the bureaucrats, who routinely close controversial discussions, and ask those who had nothing whatsoever to do with this page (if any), to weigh in with their opinions on consensus? --Elonka 22:05, 24 November 2009 (UTC)
 * That doesn't make sense to me. Bureaucrats aren't superusers. They've been entrusted with the ability to close RfA's, flag bots, and handle renames. My !votes at RfB are based solely on whether I trust someone's discretion in closing an RfA; I'm not voting to give those editors the right to adjudicate any controversial discussion across Wikipedia. I respect their opinions, since they are by and large experienced and trusted Wikipedians, but I don't think we need to turn this sort of thing over to them by virtue of the 'crat bit. MastCell Talk 22:56, 24 November 2009 (UTC)


 * a No Consensus result would default to the option supported by fewer editors, which was problematic - yet, the community clearly did not agree with this view in its application of consensus, be it at RFA, deletion discussions, or RfCs of this nature which is what is at the heart of this dispute. In such RfCs or community discussions, has a consensus been declared based on a percentage as low as 50 odd percent? So far, Protonk and myself have not been able to find one (either via searching or from memory) - but if you can, great; show us. That would help resolve the concern. However, in the absence of that, what we consider unacceptable is radically changing that norm here, and feel that the close needs to be amended accordingly in line with standard Wikipedia practice. Ncmvocalist (talk) 21:25, 24 November 2009 (UTC)
 * Ncmvocalist, Wikipedia is not a debating society, or at least isn't supposed to be one. We've had two discussions now, and both favored the secure polling option by substantial margins.  In a community this size 57%/43% is a statistically significant margin, and the closing admin weighed the quality of the reasoning.  For the smooth functioning of the project (yes, a project with productivity goals, not a community, nor a club), we need to respect decisions that are properly made by the administrators who serve.  There is no argument here that the closing admin abused their discretion in making this call. Jehochman Talk 21:31, 24 November 2009 (UTC)
 * What a spectacularly unhelpful response to a very simple request: "In such RfCs or community discussions, has a consensus been declared based on a percentage as low as 50 odd percent? So far, Protonk and myself have not been able to find one (either via searching or from memory) - but if you can, great; show us." Why do you evade underlying concerns instead of addressing them, Jehochman? I mean, I know that you tried to make out a community consensus for a sanction you unilaterally imposed (and I gave you the benefit of the doubt even then), but you're going too far by trying to change all Wikipedia norms in the same way. Ncmvocalist (talk) 21:47, 24 November 2009 (UTC)
 * AfDs routinely close one way or the other with thin margins. You're generating a lot of unnecessary text on this page.  There was a discussion.  A duly authorized party closed the discussion, and now it is over.  Let's move on to something productive, such as writing articles.  Jehochman Talk 21:51, 24 November 2009 (UTC)
 * It seems you have problems answering a simple question in an effective way, Jehochman. Until you can remedy that, I suggest you cease making unhelpful comments (and this reckless agenda that's unopen to review). Thank you. Ncmvocalist (talk) 05:30, 25 November 2009 (UTC)
 * My recollection was that WP:DATEPOLL closed at around 60%. Can anyone find the conclusion for that RFC?  I see 55% against date linking, 40% for date linking and 5% who didn't care. John Vandenberg (chat) 22:57, 24 November 2009 (UTC)
 * Was that an RfC or a poll? Either way, is there a link for that conclusion somewhere? Is there any others that you (or anyone else) can recall? Ncmvocalist (talk) 05:47, 25 November 2009 (UTC)
 * DATEPOLL was a poll, and this RFC was designed to be an RFC+poll; see Wikipedia talk:Requests for comment/Arbitration Committee 2/Archive 1 for the background. I put the poll above the RFC so that people didn't need to read lots of lengthy views before making some simple decisions.  Voting method (private vs public) alternatives were introduced soon after the poll was launched, and followed the same poll+RFC format. John Vandenberg (chat) 09:34, 25 November 2009 (UTC)


 * Whatever. Don't say I didn't tell you so when you guys fuck up the election like you fucked up the audit subcommittee election. Protonk (talk) 22:21, 24 November 2009 (UTC)
 * That would be difficult. The audit subcommittee elections were run by ArbCom, yes, but the ArbCom elections are being run by the community - and the volunteers taking a lead on this are those listed here as co-ordinators. Some ArbCom clerks are also helping out with this election (as they have done in the past), but they are helping out as individuals, not as clerks. It is important to be absolutely clear on this: ArbCom do not supervise or announce the results of this election in any way whatsoever. It is up to the community and the election co-ordinators to set up what is needed for the vote, and they need to get moving on that, as several aspects of a SecurePoll election need to be set up a fair bit in advance. Finally, I would like to note that several opposers of a secret ballot stated the need to be able to comment openly on why they are voting they way they are, and appeared to ignore the numerous comments that stated that discussion pages and endorsement pages are still possible even with SecurePoll. There are many ways for people who want to openly discuss and announce their voting choices to do so (endorsement pages, exit polls, discussion pages). If anyone wants the election co-ordinators to arrange for the SecurePoll page to have a link to such pages, for people to leave comments about candidates, or to an exit poll, I suggest they raise that point now, as the co-ordinators have a lot to arrange. Carcharoth (talk) 00:03, 25 November 2009 (UTC)

Ultraexactzz, thank you for your work on the RfC and I think your conclusions as the closing admin are sound. Cla68 (talk) 23:01, 24 November 2009 (UTC)
 * Yes, thank you Ultraexactzz for sticking your neck out there. I agree with your findings 100%.  Deserted Cities (talk) 23:20, 24 November 2009 (UTC)
 * I should throw in here that Ultra did a good job of attempting to close a contentious issue. Obviously doing so was a thankless job and at least as many people would have been upset had he closed it "differently". Protonk (talk) 04:07, 25 November 2009 (UTC)
 * Per Protonk's comments. I also don't think some users are willing or able to recognise the potential problems this causes for the future until they actually occur; it's sad that this probably that context. We've simply wanted to avoid that will be pointed to for all the wrong reasons in until the time where it isn't so much of an issue (and more discussion occurs to remedy those reservations), and perhaps some users don't users understand that we're not going to stand to personally gain or lose in this election: we aren't candidates, and we won't be members who are stuck with working with anyone for a year, 2 years or whatever duration they were elected for. Even if this message does not sink in, it's not our loss by a long-shot. Ncmvocalist (talk) 05:47, 25 November 2009 (UTC)
 * I think everyone here acknowledges that the consensus isn't unified, but the question is does this default to the status-quo or the majority? If this close comes under debate, who has the power to examine it? Jimbo Wales comes to mind: if he chooses to put in his two-cents on the issue, should that decision be binding? If so, should we invite him to take a look and close this for himself? I'd be supportive of that option if there is significant objection to this close.  Them  From  Space  05:56, 25 November 2009 (UTC)
 * FWIW, Jimbo already did opine, stating that he'll go along with the RfC result (including secret ballots, an option he is not fond of but agrees is worth a try). &mdash; Coren (talk) 11:45, 25 November 2009 (UTC)

Oh well, I'm too late, but I note here my preference for public voting. There is too much secrecy around arbcomm as it is William M. Connolley (talk) 11:21, 25 November 2009 (UTC)


 * 1) The post above by John Vandenberg, who conceived and launched this RfC page, clearly sets out the "poll+RfC" nature of the public vs. private voting question, with diffed evidence that was openly stated at the time. I think we owe it to the project to be true to this.
 * 2) There is nothing to suggest that UltraExcess, an uninvolved administrator, exercised anything other than neutral judgement in his closure of the RfCs.
 * 3) Jimbo has stated his acceptance of the use of SecurePoll for this election.
 * 4) I have complained about certain aspects of ArbCom in the past, so you'd expect me to be cautious in trusting the system. However, I have utter confidence in the capacity of the admins running the election and of the scrutineers to exercise fairness and neutrality. To me, the exclusion of ArbCom itself from the running of the process or the influencing of the outcome is transparent.
 * 5) We should assume good faith. Tony   (talk)  13:09, 25 November 2009 (UTC)


 * You've indented your reply as though you were replying to me, but what you're saying is all irrelevant. I can only assume you're responding to the bit about "too much secrecy". I think there is, and that switching to private voting increases secrecy (this can hardly be in doubt, can it?). I have no doubts that the count will be honestly done, but that isn't the point William M. Connolley (talk) 20:35, 25 November 2009 (UTC)

I'm late to this little party. Taking the most contentious point, we had 53 votes in favor of the current open balloting system, and 71 votes in favor of a secret ballot. (Yes, those are "votes", not "!votes", because that's how they're being treated here.) So by a margin of 18 users, we're going to make a major, major change in how we do things?

This should have been much more widely publicized as what it was. I think there may have been a banner, but it certainly wasn't widely publicized as a binding vote on future Arbcom staffing and election procedures. (When I saw it I thought "oh, another one of those endless RfCs" and almost didn't bother voting myself.) If we're going to make changes of this magnitude we need solid consensus behind them. Short Brigade Harvester Boris (talk) 21:31, 25 November 2009 (UTC)
 * However important ArbCom appears to be, it isn't all that important.--Tznkai (talk) 21:35, 25 November 2009 (UTC)


 * Although I've a preference for an open ballot, I don't see this as a big issue. I can see that the balance of commentary was for a secret ballot and Ultra judged the debate quite reasonably, imvho. It will be interesting to see how a secret ballot goes, but this is not a crucial change. If there are problems, we have plenty of time to return to an open ballot next year. It's not as though we're deciding the method for all time. --RexxS (talk) 22:27, 25 November 2009 (UTC)


 * As stated in my vote I can see the reasons for supporting a secret ballot and the change as such doesn't bother me much. I'd just like for things to be more above-board in saying "we're going to take a binding vote on whether to do X" right up front. Short Brigade Harvester Boris (talk) 22:33, 25 November 2009 (UTC)

57% is "consensus"? Shocking. And what is the benefit of steamrolling majority rule over the consensus process? Nadda. Regarding comments that UltraexactZZ is "uninvolved", this message on her talk page piqued my attention. --rannṗáirtí anaiṫnid (coṁrá) 22:30, 1 December 2009 (UTC)


 * I'm a little late for the after-party it seems, but this closure unnerves me as well (and this is not because I spoke against the option that ended up being the outcome). It appears from the above discussion there were two reasons for the close: the percentage and the strength of arguments. Obviously few (except possibly the closing admin, judging by one of their comments in this section) support the idea of determining a consensus for change on numbers alone when that number is 57%. That would, of course, be simple majority voting, not consensus, not even the blurry, complicated definition of "consensus" we operate under in Wikipedia. Ok, so if we treat that as a secondary reason for the close, leaving "the strength of the arguments" as the primary reason for deciding consensus, not the percentage of "votes", that may be valid and able to accept as being within a closing admin's discretion. However, it is usually considered proper to provide an explanation of how, in the closer's view, the strength of arguments showed a consensus. It is usually not enough to just say "the arguments for were stronger than the arguments against". Carcharoth provided a possible explanation, but the closer has not to my knowledge. Would it be too much to ask for the closing admin's reasoning? And yes, I do realize this disputed closure has been taken as gospel and the voting is now proceeding with SecurePoll; and so there is little chance for a different process being used for this election, no matter what is said here. I do, however, think too many controversial decisions like this have remained controversial, festering indefinitely, due to dissenting individuals being told to just walk away without their concerns being addressed; and we know this leads to editor burnout and a "chilling effect" on offering one's own minority views. So I hope this controversial closure can be resolved with a reasonable explanation? -kotra (talk) 01:48, 2 December 2009 (UTC)