User talk:MBisanz/ACE Draft

Open items
Thanks for taking a chance to look over this draft proposal for an Arbitration Committee Elections policy. It was mostly drafted by to reflect the present state of affairs of Arbcom elections, but contains some deviations where I identified areas of improvement. These are listed below. Now that I'm turning it over to the community, I'm hoping that people will edit it, debate it, and then after a week or so of editing and debate, actually start an RFC to discuss enacting it as a policy or a guidelines. Thanks.  MBisanz  talk 16:23, 17 December 2012 (UTC)


 * 1) Retains Special:SecurePoll as the means to vote;
 * 2) Retains Election Commissioners selected via RFC;
 * 3) Retains Jimmy Wales as the RFC closer;
 * 4) Removes Guides from WP: and WT: pages, but permits summary guides to be created during the election;
 * 5) Implements a timeline for when activities will be done to satisfy external needs;
 * 6) Retains Jimmy Wales at the person who ceremonially appoints the Committee;
 * 7) Adds an exclusion for certain types of questions with limited prejudicial relevance.

Terminology
Would it make sense to define Commissioners as automatically being coordinators, such that for tasks that either coordinators or commissioners should take care of, we can say coordinators, and for tasks reserved for commissioners, we say commissioners? Or would that result in adding more confusion then the language simplification would be worth? Monty 845  03:54, 13 December 2012 (UTC)
 * I thought that's what I was saying with "Commissioners are coordinators who . . ." but I'm open to ideas for other wordings that are less confusing.  MBisanz  talk 04:05, 13 December 2012 (UTC)
 * Ahh good point, you did. My focus was on trying to avoid repeating "Commissioners or other coordinators" repeatedly. Do you think it will be confusing to just say coordinators then? Monty  845  04:22, 13 December 2012 (UTC)
 * No, that would be a good simplification.  MBisanz  talk 04:26, 13 December 2012 (UTC)
 * Maybe we can find some new term for the "coordinators" who are not "commissioners." It's kind of confusing because, the way it has worked out, the "coordination" itself is mostly done by the "election admins", now the "commissioners", plus Monty has done a lot of coordinating by setting up pages, and maybe a few others, but many of the "coordinators" (including me this year, as I found very little to do except to fix one template and make random comments here and there) aren't actually coordinating anything.  I'm not sure what to suggest.  "Election assistants" sounds too menial.  "Election functionaries" or "election bureaucrats" are each suitably generic, but would be confusing because they each share a word with another role on Wikipedia (and I thought "election admins" was confusing as well because of the other kind of "admins.")  "Election workers"?  Those remind me a little too much of the 90-year old folks who sit at the table checking voters in on actual Election Days.  "Election staffers"?  Doesn't sound very appealing.  "Election volunteers"?  Hey... maybe that's it?  The "commissioners" are also "volunteers" of course, but their "selected" status puts them at a different level.  How about "election volunteers" for the non-commissioner "coordinators"?  Neutron (talk) 00:36, 16 December 2012 (UTC)
 * I rather like the name coordinators as opposed to volunteers on a purely personal basis. Without demeaning the efforts of the current crop of coordinators, I had been hoping for more active involvement from the coordinators who are not commissioners in the realm of voter eligibility and question management. While I know they didn't have the data to notify CSRF voters, I had been hoping to see them take the lead on checking and notifying people at Wikipedia talk:Arbitration Committee Elections December 2012/Log. In the end, I was reduced to yelling at people to get volunteers to do that while I studied for a finals. Also, while I know it was initially controversial, I had been hoping the coordinators would lend a hand in cross-linking the threaded discussions moved from question pages.  MBisanz  talk 17:59, 17 December 2012 (UTC)

Suggestions

 * 1) Consider whether or not a user who is subject to an existing community or Arbitration sanction may be a candidate
 * 2) Very very important - please end the election no later than December 5, with results available not later than December 15. Especially in years where there are many new arbitrators, it is very difficult to get everyone onboarded in the few days between the posting of results and the start of the Christmas break, when WMF staff who need to verify identification are (quite appropriately) unavailable.
 * 3) Why have Jimmy appoint the commissioners? For the first time, it may have made sense. Otherwise, I'd suggest simply the candidates who obtain the most support (using the same support/oppose metric as for candidates) be appointed. Yes, it makes it something of a popularity contest, but Jimmy would be very hard-pressed not to appoint the same people as would result from a direct appointment process.
 * 4) Remember that the speed by which individuals are "identified" to the WMF is partially dependent on WMF staff availability. For those who aren't already identified to the WMF, waiting until after the appointments are made will mean that they must submit their credentials via email scans of documents; some individuals have concerns about the security of this method. This may particularly affect the commissioner candidates, who have little time between close of the RfC and time that they must have completed the identification process.

Just a few suggestions. Risker (talk) 03:57, 13 December 2012 (UTC)


 * I have two concerns with #1. First, there are a variety of sanctions ranging from interaction bans to topic bans to site-bans, so it would be hard to determine what are disqualifying sanctions. Second, an individual could claim that Arbcom placed a certain sanction on them simply to disqualify them from the election and prevent them from getting on the Committee.
 * I can move the dates up accordingly for #2. I'll do that now.
 * It just seemed easier to carry over the present practice with a safety valve, but #3 can be changed.
 * I've changed the #4 provision to help ameliorate that problem.  MBisanz  talk 04:12, 13 December 2012 (UTC)
 * On point 1, I think the community has generally used "in good standing" to mean "not subject to active blocks or site-bans" so for clarity, we should just say that, so that "in good standing" is not interpreted to be stricter then that. Monty  845  04:19, 13 December 2012 (UTC)


 * Another suggestion: That guides may be deleted at the request of the author but should not be modified after voting has closed.  It seems unreasonable to expect that later readers would have to verify the page history of voting guides to make sure that they weren't altered afterward.  I don't think that protecting the pages would be appropriate. Admins would have the technical ability to make modifications while non-admins would not, and that seems to be to be an unfair situation.  Risker (talk) 23:00, 14 December 2012 (UTC)
 * What is the concern about modifying guides after voting has closed? First of all, did that happen this year?  And what were the nature of the changes?  But regardless of whether it has happened, I do not necessarily understand the harm in doing so.  (Nor do I understand what the guide-writer gains by doing so.)  Once the election is over, what's the difference?  If a reader really needs to know what a guide said at the time the election ended, I suppose they could look at the page history.  Neutron (talk) 00:36, 15 December 2012 (UTC)

Role of the Commission
My impression from the RFC was that the Election Commission would work with the elections similarly to how the arbitration committee works for the project as a whole. So the commissioners would essentially stand back and wait for either unforeseen problems to crop up that needed a determination or for a dispute to rise to the level that adjudication was required, and then to consider it as a commission. While not clearly outside the remit of the commission, I felt that the commissioners took a more active role in the election then at least I anticipated. The current draft also seems to envision a similarly active role for the commission in running the election, rather then as being a sort of emergency fallback position. It may well make sense to have a small, accountable, group to actively run the election, but I don't think that the last RFC really consider that to be what was proposed. When codifying the election process, I think we will want to discuss which vision of the commission we want going forward. Monty 845  04:15, 13 December 2012 (UTC)
 * At least for me, the impetus for taking a hands-on role was the lack of action from other parties. Between the Foundation not knowing it had to load SecurePoll, the Stewards not knowing what they had to do as Scrutineers, Candidates not knowing what statements they had to make, Questioners getting off tangent and creating conflict and Voters being uncertain of their eligibility, it seemed like Commissioners were needed to step up and coordinate things beyond the level of attention of voluntary Coordinators.  MBisanz  talk 04:25, 13 December 2012 (UTC)
 * I certainly don't intend to criticize your actions, and I thought that once the securepoll issue surfaced, the commissioners did an excellent job in handling the problem. I raise it because as a new feature of the election, I think it would be good to review its role while working on this policy. Monty  845  04:29, 13 December 2012 (UTC)
 * I understand and agree. It is a new feature and worth discussing. Particularly in the area of removing comments or questions that veered into conflict I felt as if I had no idea what I was supposed to be doing.  MBisanz  talk 04:32, 13 December 2012 (UTC)
 * To address Monty's first comment above, when Tznkai proposed an "election commission", what I anticipated (and of course I am just one person) was that the commissioners would assume the role of what had in the past been called "election administrators", of which there were usually three every year. I believe I asked Tznkai about that, and I seem to recall that his answer was something along the lines of "Let's see what happens" or "Let the commissioners decide" or some combination of the two.  (Or maybe I had that discussion with someone else.)  So rather than pressing the issue, I decided to see what happened.  When two of the four "finalists" turned out to be people who had been "election admins" in the past, I became fairly certain that I would turn out to be correct:  In addition to deciding issues that were brought to them, the "commissioners" would be the people who were actually "running" the election, and the "election admin" title would be retired.  And that is exactly what happened.  In addition to the more widely advertised role of "deciding disputes" (which was the subject of some concern on Jimbo's talk page), the commissioners also did what the election admins had done in the past, and maybe a little bit more.  I am fine with that.  If, for example, someone is going to decide that the comments in a particular "guide" are inappropriate, or that threaded discussions on a "questions" page have become more like a "debate" that belongs somewhere else, I would rather that decision be made either by the commissioners themselves, or perhaps by a somewhat wider group of people selected to do so by the commissioners.  I also believe the "commissioners" now have somewhat more of a "mandate" to do these things than the former "election admins" did, since there has been a selection process for the "commissioners."  I do not necessarily think the selection process used this year should be repeated -- I would actually like to see it be more like a "real election" -- but that is a more controversial topic and one for a different thread.


 * I would be interested in hearing whether this year's commissioners perceived their role the same way I did, or differently, and if so, how. It sounds to me like MBisanz, at least, has an understanding of his role that is consistent with my perception -- although perhaps he was a bit uncomfortable with parts of it.  But that is fine -- with more experience comes increased comfort.  :)  Neutron (talk) 01:00, 15 December 2012 (UTC)

A couple notes
Kinda busy atm, but at some point between now and next year, I intend to start an rfc concerning this process. In particular, the deprecation of using secure poll voting, and returning to the consensus process. Voting" can be replaced with an RfA-like process of support/neutral/oppose subpages. Just in this case (merely due to turnout), a separate sub page for each. (perhaps .../Support/0 .../Support/100 .../Support/200 etc.) We did this in the past, there's no reason to not return to it.

I've given a lot of thought about Sir Fozzie's stated concerns about secret balloting, and did support that this year, but honestly, we don't this in any other situation, there is no reason to do so here. Consider RfA. Admins requests are open and transparent. Why not arbcom? It's merely just another set of tools and responsibilities. And consider that admins receive adminship for life, not just 2 years, and can act at their own discretion, while arbs may only as a part of a 15+ member committee.

And besides, factionalism and such exist whether we use secret or open polling, and honestly secret polling would in my estimation merely serve the factionalism. Transparency should be the order of the day.

And guides of course should be allowed, as they are merely someone expressing their perspective. But they shouldn't be linked in the "official" notification template. It gives some editors more "weight" over others, and since this is a subjective determination (a question of trust) all Wikipedian comments (presuming, of course, NOTBATTLE/CIVIL/NPA/etc.) should be welcome. And that goes for questions too. I don't mind if threaded discussion about questions can happen on the talk page instead of on the questions page itself, as long as candidates are welcome to link to said discussion, since it relates to the question.

And questions should be open. Candidates are welcome to not answer questions at their discretion. And since discretion is one of the things that is looked for when attempting to determine trust, let's avoid CREEP along those lines.

And these are just a few things... - jc37 06:29, 13 December 2012 (UTC)
 * The fundamental issue is that we don't want ArbCom knowing who did and didn't vote for them. Not everything that is secret is bad. --Rschen7754 06:34, 13 December 2012 (UTC)
 * There are things which should be confidential. Community discussion is not one of them.
 * What is your concern? That some arbcom member will act in a biased way toward some Wikipedian? How is this any different than any editor? we all can comment in a community ban discussion. what makes one arbcom member so special? - jc37 06:56, 13 December 2012 (UTC)
 * I'll use an example. I voted in the December 2008, 2009, and 2010 elections (so in other words, I expressed an opinion on every arbitrator that was in office in 2011). In 2011 I was a party to the Racepacket arbitration case. If voting was public, how can you prove to me that my vote in the 2008-2010 elections did not affect the outcome of the case? Can you enter the homes of the arbitrators while they were doing research before voting and prove that they were not influenced by my votes in 2008-2010 when they chose what findings and remedies to vote for, even if they never said a word about it to anyone? Private voting is here for a reason, to ensure that there is not even a possibility of impropriety. (For the record, I have also written an election guide from 2008-2012, so my votes are public, but this is by my own choice; you get the idea). --Rschen7754 07:02, 13 December 2012 (UTC)
 * And again, how is that any different than a community ban discussion? If the community bans you or if arbcom bans you, you're still banned. If you are stating that people are like that silently, then they will be like that in everything, and by your argument all discussion on Wikipedia should be replaced with secure poll. Obviously it should not. WP:CON is the fundamental method we use for resolution. - jc37 07:25, 13 December 2012 (UTC)
 * Then I think that you miss the point of ArbCom. ArbCom is for when there is no consensus, and when there is a problem that the community cannot resolve. ArbCom resolves issues by voting, which is not consensus - 8/15, just over 50%, is enough for a motion to pass. Consensus is not voting. --Rschen7754 07:31, 13 December 2012 (UTC)
 * The point of arbcom is to as a committee to arbitrate in the stead of the community. With the obvious exceptions concerning private or confidential information (which they are trusted with just as they are trusted with CU and OS tools), and "emergency" situations (where by tradition they are acting in the stead of User:Jimbo Wales) there is not and should not be anything that arbcom can do that the community itself cannot. Period. Remember that arbcom requires the community to request that they arbitrate.
 * And arbcom internal bureaucracy, including voting on proposed decisions, is a slew of IAR codified into arbcom guidelines. It's what we have for good or ill. - jc37 07:49, 13 December 2012 (UTC)


 * I think we should limit the scope of the discussion regarding the draft Ace Policy to recording current practice as informed by past RFCs, making clarifications where there is ambiguity in current practice, and making minor refinements. Once an ACE Policy is adopted, we can focus discussion on RFC proposals to make bigger changes during the traditional October RFC, which is where editors would expect such bigger changes to be made. Monty  845  07:03, 13 December 2012 (UTC)
 * Note: my notes above clearly said I intend to start one or more RfCs, and I can assure you, I won't be waiting for next october. I just am waiting til I have resolved a few other RfCs etc that I have had on the back burner.
 * And incidentally, the best this page can be is a guideline, not policy, due to what it is and what it intends to cover. - jc37 07:25, 13 December 2012 (UTC)
 * I completely support the idea of more RFCs and certainly suggest you bring forward your ideas again when I make my proposal as they may be more convincing to the community. I'd rather leave the SecurePoll part in for now because that's something people are used to and I don't want what I propose to deviate too radically from prior practice. I had been hoping to get a policy in place, but if people think a guideline is better, I'll go along with that.  MBisanz  talk 12:47, 13 December 2012 (UTC)

It's worth noting that at least three consecutive RfC's have chosen the SecurePoll method. In these RfC's, I think it fair to say that every argument for an against non-public voting has been made, and the "consensus" has been consistent. Of course, consensus can change, and people can try to change it. I just don't know what new arguments anyone is going to come up with that are going to persuade the community to change it in this instance. (And in the interest of full disclosure, I have consistently "voted" in the RfC's for the current system. I would not vote in ArbCom elections if the voting was public.  And as for the comparison between ArbCom elections and RfA's, I don't think the RfA process is a particularly good one anyway, and I almost never (maybe never) participate in it.  I see no reason to "break" one process in order to be consistent with another broken (or at least, severely dented) process.  Neutron (talk) 01:09, 15 December 2012 (UTC)
 * Consensus can indeed change. Case in point, I've supported the closed system in the past.
 * But I just can't get past how this so specifically violates CON.
 * and incidentally, this process hasn't always been an "election". And there is no "need" that it be a vote. - jc37 01:29, 15 December 2012 (UTC)
 * Jc, we (mostly meaning you and I) have also been discussing this issue of "voting v. consensus" (as distinct from open voting and closed voting) in another section below. We should try to keep it all in one place.  As I said in that other section, I really don't think that the previous "open vote" elections for ArbCom were really "consensus" processes.  They were elections, with voting, the main difference between the 2006-2008 elections and the 2009-2012 elections being that previously, you voted on a wiki-page and everybody saw it, and now you vote on SecurePoll and everybody doesn't see it.  I did not go back before WP:ACE2006 to look at those elections, but I know that at least one of the previous election used BoardVote, so that was a closed ballot also.  So as far as I know, ALL ArbCom elections, and at least the last seven and at least one prior one (leaving one? or two?) have been "votes".  And in my opinion, it really has to be a vote, especially since the community has now repeatedly endorsed, in addition to closed voting, a 50 percent threshold, which is really inconsistent with "consensus."  I don't see how you would have a "consensus" process to elect candidates to a fixed number of seats (as opposed to an RfA, in which a negative result simply means that another admin position is not added -- not that I consider RfA to be a "consensus" process anyway -- nor do I consider it a process that works, or one that we would want to expand to other areas.)  And I guess the bottom line is, other than matters of philosophy (which is what WP:CON is), what is really WRONG with how we elect Arbitrators now?  Neutron (talk) 00:59, 16 December 2012 (UTC)
 * There is a difference between simply marking support/oppose/neutral and actually typing up a reason/rationale. And we shouldn't be seemingly relying on guide-writers alone to provide evidence (both positive and negative) concerning a candidate. That seems nonsensical, if only that one has to read over a dozen guides to get such information on candidates. Yes, it would be nice if every voter did the due diligence of checking contribution histories. But with several candidates (21 in this election) it makes sense to allow for the wiki-way of "many hands makes light work" and so to allow voters to add rationales which include evidential links. Allowing for both negative and positive.


 * Plus there is the question of transparency. As I have said about transparency elsewhere: if we can have policy discussions "out in the open", if we can have rfa "out in the open", then there is simply no reason to not have this "out in the open". I realise that secret balloting has had consensus in the past, but WP:CCC, after all.


 * The way we current do this would seem to me to do a disservice to our fellow Wikipedians. - jc37 08:01, 18 December 2012 (UTC)

A few points
Matt, well done. A few queries:
 * must "not have obtained less than 33% support in the immediately prior Arbitration Committee Election". Does this mean a third—that is, 33.33%? Should it read "support percentage", as elsewhere? I don't particularly mind this, but is it necessary?
 * "all periods begin at 00:00 UTC of the first date listed and end at 23:59 UTC of the last date listed"—Why not "00:01 UTC", to match the end-time. Four noughts is ambiguous.
 * Timeline: some of those times are "by ...", are they? It looks a bit inflexible, for example, that a coordinator has to file a notice in Bugzilla on November 5, exactly, not just by November 5.
 * "Commissioners compile a list of the voters from the previous elections"—it's not clear to me why.
 * "On the last day of the election, the commissioners post notice of the impending finish of the voting period to"—perhaps the "second-last day"?

Tony  (talk)  09:50, 13 December 2012 (UTC)


 * I've made #1 more precise;
 * I agree with your change on #2;
 * I've tweaked the heading to address #3;
 * I've tweaked the language of #4 to be more inclusive. My reason why for the entire provision is to try to retain voters from prior elections and increase turnout.;
 * I've tweaked the language to leave it to the commissioner's discretion as to the day of or the day before for #5. Thanks.  MBisanz  talk 12:44, 13 December 2012 (UTC)


 * Thanks, Matt. We've needed this page for a long time. Tony   (talk)  13:48, 13 December 2012 (UTC)

"Guides to other guides" strawman: Summary table at issue
The RfC failed to find consensus for this policy, which Monty inserted in the 2012 template before the end of the RfC and failed to remove when the RfC did not support it.

Nobody suggested a "guide to guides" and this strawman should be avoided. What you are trying again to ban is a "summary table" of other guides, which states only the practical conclusion as to recommendations for support, neutral, and oppose.

Kiefer .Wolfowitz  08:40, 14 December 2012 (UTC)
 * I've altered that provision in a way that I hope will remove disputes over the topic.  MBisanz  talk 19:18, 15 December 2012 (UTC)

CREEP?
This is quite ridiculous. Two annual RfCs. Officials to run the election of officials to appoint officials. Rules, more rules, which generate the need for more clarification. I understand this all has its own internal logic and is all offered in good faith, but it is clearly antithetical to the basic way Wikipedia (and wikis) are supposed to function. With declining involvement there is a need to streamline and simplify process so that less time is wasted on it.

I'd strongly suggest starting again, right from the beginning, and asking "what's the absolute minimum we need to stop Wikipedia blowing up?" A much larger community functioned for many years with a lot less red tape - and the more you legislate to cover problems the more you'll just discover more problems needing ruled on.--Scott Mac 23:00, 14 December 2012 (UTC)
 * To a point, I agree. I think there is a lot of plate spinning that just isn't necessary. Let's just default to WP:CON, and just toss out the whole elections bureaucratic process : ) - jc37 23:59, 14 December 2012 (UTC)
 * I disagree. Obviously Scott Mac is exaggerating a bit "for effect", but I think we need to look at what MBisanz is actually proposing here.  It is really just a codification of what has been done in the past, instead of necessarily having to have a from-the-ground-up reconstruction of the basics of the election every single year, which is what we have been doing.  Plus MBisanz has added a few proposals for improvements.  I do not necessarily go along with all of these these, and I might have other changes that I may suggest myself.  I will address my objections to one particular proposal in a new section.  But that is how I think this discussion should go; let's accept that we need a codification, and then discuss what particular items should or shouldn't be part of it, what needs to be changed, etc.  We are talking about an election, and like it or not, elections need rules, and the rules need to be enforced, and in a consistent manner.  The alternative is an election that a large part of the electorate will view as being illegitimate, and the "winners" of that election lacking in recognition as the legitimate holders of the position they occupy.  We don't need that.  Neutron (talk) 01:20, 15 December 2012 (UTC)
 * your comments presuppose that this need to be framed as a voting process instead of a consensual one.
 * That is not true, this could operate just fine as a consensual process, and indeed did so in the past. - jc37 01:32, 15 December 2012 (UTC)
 * Which of the past ArbCom elections would you classify as being a "consensual process" rather than a "voting process"? I am not saying it hasn't happened, but I am not aware that it has.  Since I assume that the discussion pages for all of the elections are still around somewhere, I would be interested in looking at the ones you are talking about to see how they worked.  Neutron (talk) 01:41, 15 December 2012 (UTC)
 * I'll have to go find the links, but my recollection is 2007 and earlier. - jc37 01:55, 15 December 2012 (UTC)
 * I did a little research of my own. While one or two of the earliest elections may have used confidential balloting, the elections have been run that way (using SecurePoll) consistently from 2009 forward.  I looked at WP:ACE2006, WP:ACE2007, WP:ACE2008 and some of the related pages (including some of the "voting" pages where people voted Support/Oppose/Neutral on each candidate, as they do on RfA), and what I see is NOT anything I would call a "consensus process."  They were elections, with voting, the same as they are now.  The candidates with the top "Support percentages" were appointed to the number of seats available (although in one or two elections the number of seats to be filled was not determined until after the elecion.)  If there was a mix of longer and shorter terms available (regardless of whether that was determined before or after the election), the higher-"scoring" candidates were appointed to the longer terms.  There was some talk that "community support" (which seemed to mean at least a "majority" Support percentage) was required for election, and nobody was appointed with less than a majority.  There was somewhat more discussion in those elections, especially since some editors tried to engage in threaded discussions on the voting pages, but those discussions were generally moved to a talk page.  What was left on the voting pages were Supports and Opposes, with some voters putting in very brief comments with their vote.  (See, for example, Arbitration Committee Elections December 2007/Vote/Newyorkbrad and Wikipedia talk:Arbitration Committee Elections December 2007/Vote/Newyorkbrad; NYB's first candidacy for ArbCom.)  Overall, the major difference between those elections and those of "today" are that in the earlier elections, voting was "open", and now it is "closed."  It was really not any more of a "consensus process" and I see no evidence that it worked any better than the elections do now.  In 2009, the participants in an RfC chose closed balloting, and they have ever since.  I see no reason to change it.  Neutron (talk) 21:45, 15 December 2012 (UTC)

Elections are fine. Securepoll is fine. Where I get off is with "the rules need to be enforced, and in a consistent manner" - no, that's not how we do things here. That leads to increasing bureaucracy, which is what we are getting. First there was an election (fine) then there as discussion as to the means of the election (fair enough), then we had folk to run it and folk to scrutinise it (ok, lets trust them), but then we needed "a commission" to oversee that, then we need rules for electing the commission, and now we need this. It isn't broke, so stop trying to fix it. All you will do is generate more rules, and then there will be some inconsistencies in them, and then you'll need more process to cover those eventualities. Remember, we are not electing a government, we are selecting a few folk to settle the odd thing that the community has trouble with, that's all. The mantra must be to keep any policies as minimalist as possible - and simply to deal with eventualities by discussion if and when they arise.--Scott Mac 00:49, 16 December 2012 (UTC)

33.33%

 * (v) not have obtained less than 33.33% support percentage, as calculated by the formula on this page, in the immediately prior Arbitration Committee election if they were a candidate in that election.

I disagree with the proposal numbered "v" in the candidate eligibility criteria currently on the page. This would require that someone who has run in an ArbCom election must have received at least 33.33 percent of the vote in order to run in the election. (Actually the current wording seems to say that NOBODY can run unless they have run previously and have passed the one-third threshold, but I assume that is not what it means and that it will be changed.) I think I understand the reasoning behind this proposal:  We don't necessarily want people running year after year, getting few votes but becoming a "perennial" or "joke" candidate. But why is this necessarily a bad thing? Having a few candidates who "everyone knows" isn't going to win does not cost anything. It does not disrupt the election. Why not just let the voters decide? (That question informs my opinion on a lot of subjects relating to the election, including such questions as whether someone who is the subject of an interaction ban or a topic ban should be eligible to run. My answer is, sure; and the guide-writers and talk-page writers can mention these things all they want, as a reason why the person shouldn't be elected, and then the voters can decide.)  We can afford our share of Harold Stassens. If people want to run just for the sake of running, and don't mind losing every year, how does that hurt the elections, or the process? I could see a problem if, for example, we had hundreds of candidates, 80 percent of whom were "perennial losers" or "joke candidates", because the voters would have to do a tremendous amount of work to separate the wheat from the chaff. But we don't. We have a few of these candidates, here and there, and the voters are capable of sorting them out from the "serious" candidates. And perhaps, every now and then, under the right circumstances, the voters might decide that a "perennial candidate" might make a good Arbitrator after all. I say, let them. Neutron (talk) 01:36, 15 December 2012 (UTC)
 * I don't think we can afford to tolerate gadflys of this nature, particularly when reviewing their perennially unsuccessful candidacies does absorb the time of new voters who may be discouraged from future participation by the extra exertion of effort required to review such candidates. And they can become a good arbitrator by waiting another year and running again.  MBisanz  talk 19:21, 15 December 2012 (UTC)
 * I dunno. Think about the NOTNOW situations at RfA. are we saying that a year later should automatically be regarded as NOTNOW too?
 * Sounds wrong somehow.
 * What if you drop it to 10%? Getting at least 90 out of 900 would seem to make more sense than needing over 300 out of 900... - jc37 20:21, 15 December 2012 (UTC)
 * By your 10% threshold, there are 3 people from the 2006-2011 elections who would not have been able to run the next year. By the 33.33% threshold, there are 36 who would not have been able to run the next year. No one who has ever gotten under 33.33% has gone on to serve on Arbcom and from my rough math, no one who has ever gotten under 33.33% has seen their support percentage improve in a subsequent election.  MBisanz  talk 22:07, 15 December 2012 (UTC)
 * Part of the reason AGK nudged me to run was "to give everyone more of a choice/selection". Setting this to 1/3 would seem to me to be preventing the possibility of a selection choice. I am not looking at the yearly numbers, but honestly, should we really disenfranchise the 300 who voted for these candidates?
 * I seem to remember that some US presidential debates required a potential debater to have at least 5% support to be able to join in the debate. While I can see pros and cons even for that, I don't think that 5% is too low as much as I think 1/3 is very much too high. - jc37 21:02, 15 December 2012 (UTC)
 * So rather than waste 900 people's time once every two years, it's better to waste 1,800 people's time? More candidate choice is better, but once it's certain a candidate found to be completely unqualified, I don't see how that's providing more choice to let them run.  MBisanz  talk 21:08, 15 December 2012 (UTC)
 * One could easily imagine an excellent new user, with only a few months experience, running to stimulate debate on various issues, and failing to get 33% because of a "not yet" oppose, and yet the same user being a very serious candidate 12 months later.--Scott Mac 21:13, 15 December 2012 (UTC)
 * In 5 years no one has even improved their support percentage from below that level, so it isn't easy for me to imagine what you describe.  MBisanz  talk 21:18, 15 December 2012 (UTC)
 * In 6 years, there have been 36 arbcom candidacies that have achieved less than 33% support. Of those, only 3 repeated the attempt, and would have been prevented by your suggested rule. So it would hardly prevent the "gadflies" you don't want to tolerate. Incidentally, among the sub-33% "gadflies" were three ex-members of Arbcom and a number of experienced admins (myself included) and one user who had scored 70% in a previous election.--Scott Mac 21:41, 15 December 2012 (UTC)
 * I just don't think that otherwise eligible candidates should be disqualified unless there is a really good reason, and lack of success in a previous year's election does not strike me as a really good reason. As mentioned above, there have only been a few "repeat candidates" who had previously received a very low percentage, so I really don't think it is going to make things easier on voters.  But if there is to be some threshhold for running in the following year's election, I think 10 or 15 percent would be more reasonable.  Neutron (talk) 21:51, 15 December 2012 (UTC)
 * Doesn't matter what the threshold is, the number of repeat unsuccessful candidates is, actually, very very low. If you want to reduce the number of "not a hope in hell" candidates, you'd need to have a threshold for first runs. You'd need to do something like requiring a significant number of people to endorse a nomination, but such a process may deter good candidates too. If you really wanted to eliminate no-hope candidates, you could prevent non-admins from running. No non-admin has ever been elected, and that's unlikely to change. But we know such a proposal would be widely resisted.--Scott Mac 22:12, 15 December 2012 (UTC)
 * I wouldn't favor either of those proposals. The "nominating process" sounds like it would be inviting a "party system", for real, not the ineffective attempt at a "party" that was made this year.  As for the second one, I don't even pay much attention to who is an admin and who isn't when voting.  I think one of the candidates I voted for this year was not one.  The voters should have that choice.  I really think that, generally speaking, the voters can handle this, with the criteria that existed this year.  If we want to make things easier on the voters, I think the way to do it is to limit the number of questions to the candidates, and therefore the number of answers.  If someone wants to ask a question based on old grudges and personal disputes with the candidates (or others), I think they should have to do it on the candidate's talk page and not on the "official" questions page.  But that is for another section.  Neutron (talk) 23:09, 15 December 2012 (UTC)
 * Oh, I'm not disagreeing with you. I'm only saying that if you want to reduce the number of no-hope candidates then you either need some form of pre-selection or you need to significantly raise the hurdle to qualify (or both). I can see the logic in what MBisanz is attempting to do here, I'm simply saying his 33% threshold does not achieve it - and that the only ways to realistically achieve it are going to be politically unpalatable.--Scott Mac 23:26, 15 December 2012 (UTC)

What if we changed it to be that any candidate who received less then 33% support in their last run, must receive 10 nominations from editors who meet the candidate eligibility criteria to appear on the ballot again. This would not apply to first time candidates or previous failed candidacies that occurred prior to the rule going into effect. Monty 845  00:54, 16 December 2012 (UTC)
 * Again since, in 6 years, only three people have ever run again having received 33% or less, what problem is it you are tying to fix?--Scott Mac 01:00, 16 December 2012 (UTC)
 * And isn't it a little too complicated? There are already complaints that the process is becoming overly bureaucratized, though I do think a solid body of rules are needed.  But can't we try to keep the rules themselves as simple as possible.  Monty, I appreciate the balanced approach you are taking, but as I read your suggestion, I start to see flowcharts in our future.  Neutron (talk) 01:04, 16 December 2012 (UTC)

I also disagree with the proposed 33.33% rule. I understand the purpose of the proposed rule&mdash;to prevent blatantly unsuitable candidates from running for years in a row. However, I can also imagine a situation in which the rule could disqualify a potentially good candidate. Imagine a relatively inexperienced editor who runs for ArbCom in his or her first few months of editing. He or she has good skills and instincts, but the voters decide that the candidate is just too new to Wikipedia to serve as an arbitrator. Another year goes by, in which the candidate makes lots of good edits, does good work as an administrator, participates helpfully in dispute resolution, and now seeks to run again. The candidate's misjudgment in having run prematurely the year before shouldn't deprive the community of the opportunity to elect that candidate now.

Another concern is the potential for deterring candidates from running: "If I run this year and don't do well, then I won't be able to run next year when I might do better."

I could understand if the risk of potentially losing a good candidate on occasion were more than balanced out by the voters' being overwhelmed every year by lots of weak or joke candidates, but as noted above, there's no evidence that anything like this is happening.

Finally, on a more process level, I think the overall purpose of the proposed draft should generally be to crystalize the elements of the election process on which a consensus has emerged, rather than to introduce controversial new rules. Newyorkbrad (talk) 00:55, 18 December 2012 (UTC)

Election of commissioners?
Does anybody other than me think the Election Commissioners should be chosen in an actual election? (Subject to actual appointment by Jimbo, I guess.) The RfC process this year seemed a little, I don't know, haphazard, which is how it had to be because there was no time for anything else. But in the future there would be time for an actual election. I realize there would be questions of when, how, etc., but I don't really want to figure out a whole procedure for that, only to be met with, Elections, we don't need no stinking elections. So before I spend a lot of time, is there any interest in this, or is the RfC process sufficient? (Actually, I think the Election Commissioners should be appointed by the elected Governing Committee of the English Wikipedia, but since I haven't noticed that body having been created and doubt it ever will be, our options are limited.) Neutron (talk) 01:32, 17 December 2012 (UTC)
 * Having gone through the process myself, I don't think more procedure is needed. Commissioners basically need to be WMF-identified people who aren't likely to let their personal preferences for candidates interfere with the functioning of the election and generally are trusted by other people to make good decisions. Given the need for a first actor to get things going, keeping their selection process simple is critical. Before I would support an actual election commissioner election, I would support even simpler methods like a random number selection from subscribers to functionaries-en.  MBisanz  talk 16:14, 17 December 2012 (UTC)
 * Instruction creep is beginning to set in here. Several years ago, I became concerned that too much time of the core members of the community was already being taken up by multiple sets of elections. (In one year, I think we had the ArbCom elections, the Audit Subcommittee elections, two sets of Oversight/Checkuser elections, the Stewards elections, and the Foundation Board of Trustees elections.) I really don't think we can afford the luxury of having a formal election just to decide who runs the election. (Not to mention the obvious question of who would run the election to elect the people who run the election....) Newyorkbrad (talk) 01:00, 18 December 2012 (UTC)

"failure to answer questions may be perceived negatively"
I wonder if we should reword the sentence stating that a candidate's failing to answer a question may be perceived negatively by the community. In many circumstances that may be true, but in other circumstances (unfair questions, irrelevant questions, badgering) it might not. Perhaps stating that a failure to answer questions "may be taken into account by the voters" would be more neutral? Newyorkbrad (talk) 01:00, 18 December 2012 (UTC)


 * Odd cavil -- I woulda thunk it obvious that the reason for questions at all is for voters to consider the answers thereto. And as a corollary, if a person does not answer a question, isn't it also kind of a "d'oh"  sutuation that someone might consider such avoidance of a question in a negative manner?   Changing the wording here is not going to affect much of anything at all, as far as I can tell. Collect (talk) 02:00, 20 December 2012 (UTC)

Community discussion

 * Candidate statement
 * The candidate's statement may not exceed 400 words in length, although candidates are free to link to a longer statement if they wish. The mandatory disclosure of alternate accounts and declaration of intent to comply with the WMF identification policy are exempt from the 400-word limit, although candidates are encouraged to be concise.
 * The current process for creating candidate statements is to use the inputbox on the candidate page, by appending the candidate's username to the existing text, clicking the button, and following the instructions.


 * Questions
 * If there are no general questions submitted, the previous year's question set is used.
 * Once a candidate has declared their candidacy, voters may ask them specific questions on their question page in the format determined by previous consensus.
 * Voters may ask follow-up questions, but may not engage in threaded discussion on the questions page, which should instead be on the talk page. Relevant discussion on the talk page may be linked at the question as appropriate.
 * Voters may ask an unlimited number of specific questions, but are urged to avoid asking duplicate questions or irrelevant questions. Voters may not ask questions that reference or concern their own previous interactions with the candidate or that reference the voter's past conduct or behavior.
 * If there is a dispute regarding the inclusion of a question, it is referred to the election commissioners for determination.
 * Once a candidate has made their statement, they may proceed to answer the general and individual questions. In responding, candidates are advised to answer each of the general questions completely but concisely.
 * Candidates may refuse to answer any questions that they do not wish to, with the understanding, however, that choosing to not respond to a question may potentially be perceived negatively by the community.


 * Discussion and guides
 * To facilitate their discussions and judgements, voters are encouraged to familiarize themselves with the candidates. This can be done through reading the candidate statements, the answers to the questions put to each candidate (linked from their candidate statements), and the discussion of each candidate (a centralized collection of which is made available at the discussion page). A summary guide to candidates is made available, and may be augmented by a set of personal guides by individual voters.
 * Voters may write guides describing their individual opinions of the candidates. Voters may write guides that evaluate one or more other guides. Voters may create charts and tables summarizing guides. No guide, guide to guide or summary of voter guides may be linked from the projectspace, talk pages of the projectspace, or a template which can be reasonably expected to appear in the projectspace or project talkspace.
 * 1) Voters are encouraged to not edit their guides subsequent to the close of voting.
 * Voters may discuss the candidates on each candidate's discussion page and on the general discussion page for the election.
 * The coordinators work with other users to compose a summary guide listing basic biographical information of the candidates such as account age, userrights, and previous position requests.
 * All parties must abide by the general policies and guidelines for user conduct in their participation in the election.

I removed the above from the proposal.

It occurs to me that this should stay focused specifically on the "nuts and bolts" of the election. Eligibility requirements for candidates, voters, scrutineers and commissioners; assessment/review of the votes; and the posting of the results.

Guidelines covering interaction between the candidates and the community, including discussion with and/or concerning candidates, either through statements, questions, threaded discussion, or guides, probably shouldn't be part of this particular page. - jc37 07:43, 18 December 2012 (UTC)
 * Where would you suggest they go then?  MBisanz  talk 17:36, 18 December 2012 (UTC)
 * I think that interaction guidelines should be separate from election process guidelines. And should probably be a separate proposal/RfC.
 * I also think that some of the above could be combined into a general "best practices for interaction" or some such. - jc37 23:26, 18 December 2012 (UTC)
 * I understand your thoughts, but I'd like to present it as one unified proposal for discussion in the projectspace, so I've gone ahead reinserted a lot of the material until I can start a bunch of mini-RFCs on elements when I move it over.  MBisanz  talk 02:06, 19 December 2012 (UTC)
 * Nod. No worries. It's your proposal. I was merely attempting to help. - jc37 02:09, 19 December 2012 (UTC)
 * Thanks and I appreciate it. I did retain the removal of the 33% provision; that's just too controversial to be worth including.  MBisanz  talk 02:18, 19 December 2012 (UTC)

Question on Questions
In the section on questions, it says "voters may not ask questions that reference or concern their own previous interactions with the candidate." In principle this makes sense, but I can think of a few instances where a question could be considered pertinent. Say a candidate blocked a user and the block was overturned with community consensus,for example.  Hot Stop     (Talk)   16:29, 22 December 2012 (UTC)
 * I have two aims with this provision. The first is to eliminate questions of extremely limited relevance. If an admin blocked a user and the block was overturned and no one other than the blocked person cared enough to remember/research it to ask a question, I don't see the value to the broader electorate in that question. The second aim is to reduce conflict/bias in the questions. If a user asks a question about their own prior interaction with the candidate or their own behavior, it is far more likely to contain bias or be a veiled attack on the candidate than if it is asked by a third-party. Assuming there is something of relevance to the electorate, I have great faith in third-parties asking the candidates about it.  MBisanz  talk 17:35, 22 December 2012 (UTC)