Wikipedia:Requests for comment/Arbitration Committee Elections December 2018

-

The purpose of this request for comment is to provide an opportunity to amend the structure, rules, and procedures of the December 2018 English Wikipedia Arbitration Committee election and resolve any issues not covered by existing rules. 05:18, 4 September 2018 (UTC)

Background: In the case of proposals that change existing rules, or that seek to establish new ones, lack of consensus for a change will result in the rules from the 2017 election remaining in force. Some issues are not covered by the existing rules but will need to be decided one way or another for the operation of the election, in those cases it will be up to the closer to figure out a result, even if there is no clear consensus, as they have had to in the past.

Structure: This RfC is divided into portions, each of which contains a discussion point for the community. The standard RfC structure will be used, in which any user may make a general statement that other users may endorse if they so agree. The points will be listed in the table of contents below, along with the users who have made statements.

Per the consensus developed in previous requests for comment, the electoral commission timetable is as follows:


 * Nominations: Saturday 00:00, 6 October – Friday 23:59, 12 October (7 days)
 * Evaluation period: Saturday 00:00, 13 October – Friday 23:59, 19 October (7 days)
 * Commission selection: completed by Friday 00:00, 26 October

Per the consensus developed in previous request for comments, the arbitration committee election timetable is as follows:


 * Nominations: Sunday 00:00, 4 November – Tuesday 23:59, 13 November (10 days)
 * Setup period: Wednesday 00:00, 14 November to Sunday 23:59, 18 November (5 days)
 * Voting period: Monday 00:00, 19 November to Sunday 23:59, 2 December (14 days)
 * Scrutineering: Begins Monday 00:00, 3 December

Anyone is free to raise any new topics that they feel need to be addressed by adding them as level two headers.

Duration: Normally, these RfCs begin on September 1 and last for 30 days. However, these elections are volunteer-run, and this year, the RfC is starting a bit behind schedule. In order to preserve the timetable above, we'll see whether a consensus develops on or after September 30. The results will determine the structure, rules, and procedures for the election.

Use the following format below; post a new statement at the BOTTOM of the section in which you want to make a statement. Endorse by adding a hash symbol (#) and your signature.

Statement #N by USERNAME
Comment ~

Users who endorse statement #N:


Users should only edit one summary or view, other than to endorse.

= Points of discussion =

Statement #1 by Rschen7754 ( 14, then 13 arbitrators)
Only elect enough arbitrators to bring the committee to a size of 14 for the 2019 year, and 13 for the 2020 year and beyond. 13.
 * Proposal


 * Rationale

With the caseload decreasing (from 100+ cases a year in the 2000s to 11 cases in 2012 when the size was reduced to 15, to 4 in 2017), ArbCom no longer involving itself in most ban appeals, and WMF handling some matters including child protection and global bans, the apparent workload of ArbCom is decreasing. Meanwhile, fewer candidates are running (12 in 2017 and 2/3 of them being elected - compared to 17 candidates in 2011 and just under 1/2 of them being elected), increasing the chances that an "unsuitable" candidate would be elected. Therefore, I think that we should at least put on the table the possibility of reducing the number of arbitrators. Rschen7754 06:05, 4 September 2018 (UTC)


 * Postscript: Alex Shih resigned, which changes the math a bit. I was concerned about only electing 5 candidates this year, but now that it is 6, we could just go straight to 13 and ignore the phased-in part. --Rschen7754 18:17, 4 September 2018 (UTC)

Users who endorse statement #1:

 * 1) Rschen7754 06:05, 4 September 2018 (UTC)
 * 2) Makes sense. It's not good to have to elect so many arbs from such a small pool of candidates. Bishonen &#124; talk 11:06, 4 September 2018 (UTC). Added: I also support going directly to 13. Bishonen &#124; talk 01:01, 5 September 2018 (UTC).
 * 3) Agree with basically all of the above, plus the fact that if ArbCom bans are all to have a "may not be appealed for X months", then waiting a month and a half just for enough members of the Committee to log in and vote, the way they were obviously always going to vote, is just a waste of time. Hijiri 88 ( 聖やや ) 11:54, 4 September 2018 (UTC)
 * I'm cool with just dropping straight to 13. If trends of decreasing numbers of candidates continue, we can pull it down further later on. (Although Tony's concern in the section below is compelling, I don't think 13 would be all that much worse than 15, at least in the short term.) Hijiri 88 ( 聖やや ) 01:00, 5 September 2018 (UTC)
 * 1) I think I agree as well. There is some activity which is invisible for us (such as dealing with unblock appeals, privacy violations etc), but from the visible part (cases etc) my conclusion is indeed that it is now significantly reduced, and we are ready to some reduction of the number of the committee members.--Ymblanter (talk) 12:37, 4 September 2018 (UTC)
 * Also fine with going to 13.--Ymblanter (talk) 05:30, 5 September 2018 (UTC)
 * 1) Strong endorsement.    --  Green  C  13:57, 4 September 2018 (UTC)
 * 2) Since Alex resigned from ArbCom, no need to elect one more, so it is better to elect 7 6 this year Hhkohh (talk) 15:36, 4 September 2018 (UTC)
 * Also support directly reduced to 13 Hhkohh (talk) 16:12, 4 September 2018 (UTC)
 * 1) Huh. I planned to propose this, but I guess I don't have to. I want to make clear activity has not significantly reduced. The Arbitration Committee had over 1,200 emails on the ArbCom lists in August alone, not counting functionaries-en, checkuser-l, oversight-l, or clerks-l. Having said that, I think 15 voices is unmanageable at times. We spend a lot of time with a decision basically made but waiting on the votes because it's difficult to wrangle a majority on any issue. A big upside to lots of voices is lots of perspectives and a diversity of opinions, but I don't think we lose that by dropping from 15 to 13 arbs. ~ Rob 13 Talk 15:43, 4 September 2018 (UTC)
 * Also, we could reduce straight to 13 this year if we wanted to. Don't replace Alex and remove a seat. We'd have 6 seats this cycle and 7 seats next cycle. ~ Rob 13 Talk 15:55, 4 September 2018 (UTC)
 * That 1,200 emails figure is interesting. I fully believe it, and I know that most of ArbCom's work is outside the public eye (and probably always has been). But I've seen a somewhat alarming trend of editors proposing "send it to ArbCom -- they don't seem to be all that busy at the moment, as there have only been X cases in the last Y months" on ANI threads that really could have been easily solved by the community if it weren't for the "send it to ArbCom" chorus; is the 1,200 emails in August figure publicly available anywhere other than your above comment? If not, I'm kinda thinking it should be, so the "ArbCom don't have much on their plate at the moment" non-argument could be shot in the bud. Hijiri 88 ( 聖やや ) 03:02, 5 September 2018 (UTC)
 * I've queried the Committee about releasing the monthly numbers from January 2014 forward. If no-one objects, I'll publish the totals shortly. As for the "send it to ArbCom" chorus when the community could solve it, I really want to push back with a "send it to the community" chorus when things come to us without bona fide dispute resolution attempts. In my opinion, GWE was a perfect example of something ArbCom should never have taken on due to a lack of prior dispute resolution. I plan to question candidates this year on their opinions on the importance of prior dispute resolution. ~ Rob 13 Talk 06:58, 5 September 2018 (UTC)
 * I've posted similar figures before, but only in talkpage chatter, never anywhere official - if memory serves it came out to ~35 messages a day last year, well down from the peak around 100ish in 2009 (when the committee was newly expanded). Of course many of those barely take any time to read - "Agree", "+1", "decline", "will be out of town next week", "need two more votes on motion X", etc. Opabinia regalis (talk) 07:41, 5 September 2018 (UTC)
 * 1) Support reducing to 13 per Rob’s statement on workflow and Alex’s resignation. The unsuitable argument can go either way, and the workflow argument is pretty strong. Also, not having a year of an even committee is a positive., you may want to update the proposal. TonyBallioni (talk) 16:03, 4 September 2018 (UTC)
 * Hmm. I'm okay with that, though it would mean that we are only electing 6 seats this year. I also don't know if it's proper to change my proposal after people have voted. --Rschen7754 18:13, 4 September 2018 (UTC)
 * I think it's fine, especially given how many people below support going directly to 13. Just ping everyone who voted early if you change it. ~ Rob 13 Talk 19:44, 4 September 2018 (UTC)
 * Done. --Rschen7754 00:54, 5 September 2018 (UTC)
 * 1) I am also in favour of a reduction. While there are fewer cases that come before ArbCom, the dissolution of the Audit Subcommittee, ARCA filings, and other factors has meant that serving on the Committee still feels quite busy, if active on all matters . Bringing the Committee to a smaller size will likely improve response times by reducing the time for consensus building, which takes 90% of the time. It will mean for the Community that selecting who is on the Committee will be more important than ever, but I believe it will be a worthwhile trade-off considering how frequent elections occur. Mkdw  talk  16:44, 4 September 2018 (UTC)
 * 2) Per Mkdw and Rob13 - I was thinking of supporting this but was waiting for some comments from Arbitrators, especially wondering if reducing the number would increase workload for the arbs (and thus make it even more off-putting..), but per their comments it seems reducing the number would even be beneficial re workload due less time being spent on consensus building. Reducing it means the opportunity to have little more actual choice in who is elected. Doesn't make sense to have such a huge election if the the number of acceptable people is something like two more than the number of people who are elected. Also support directly going to 13. Galobtter (pingó mió) 17:07, 4 September 2018 (UTC)
 * 3) There's a balance here, and I may be saying something different than I've said in the past, but these days I come down firmly on the side of reducing the size of the committee. Reason 1 is one of Rschen's original justifications: fewer positions reduces the likelihood of borderline candidates getting in because people feel compelled to vote for enough candidates to fill the vacancies. Reason 2 is that, per Mkdw, the reason ArbCom is often glacially slow is not because no one knows what to do or no one can agree, but because they're waiting for everyone to chime in on stuff.  The fewer cats to herd, the better. I'd be happy going directly to 13 this year. --Floquenbeam (talk) 17:32, 4 September 2018 (UTC)
 * 4) Echoing the above. The committee needs redundancy built in, as you regularly have period where half are inactive to some extent - however, there is definitely scope for reduction and I think going straight to 13 is the best option. WormTT(talk) 22:03, 4 September 2018 (UTC)
 * 5) Per above. -  F ASTILY   22:12, 4 September 2018 (UTC)
 * 6) Support going straight to 13. Don't agree with the sentence re "unsuitable" candidates - who is suitable or not is a matter for the voters. The real issue is the extensive delays in corralling enough votes for routine matters - many Arbcom issues currently gather 4-5 supports and no opposes, and then stall for weeks while a few more votes come in. No criticism there - people are busy and time zones difficult. However a smaller Arbcom makes it slightly easier to keep things moving along. Further, as someone first elected in 2014 I can attest to a smaller functional workload for the committee of today. We used to do child protection issues; we used to do BASC, we used to do more cases - so we used to need more Arbs. Times have changed, probably for the better. We should feel free to change Arbcom to match. -- Euryalus (talk) 22:45, 4 September 2018 (UTC)
 * 7) Support 13. Ealdgyth - Talk 00:04, 5 September 2018 (UTC)
 * 8) Support: electing only 6 new arbs (to 2 year terms), going to a committee of 13. power~enwiki ( π,  ν ) 01:01, 5 September 2018 (UTC)
 * 9) Support unless the current members of arbcom come out and say this is a bad idea (so far the opposite seems to be happening) Tazerdadog (talk) 03:09, 5 September 2018 (UTC)
 * 10) Seems sensible, based on the observation from a few current arbs that the workload is less than it was in years prior and the majority of the delay in getting things done is coaxing enough arbitrators out to get a majority to pass something. A smaller committee means that a smaller number of arbitrators is needed to get things moving. Going down to 13 seems like the best way to go. Mz7 (talk) 07:22, 5 September 2018 (UTC)
 * 11) Largely per above. Vanamonde (talk) 13:10, 5 September 2018 (UTC)
 * 12) Support going to 13. This would help with the process of getting to consensus a lot quicker. RickinBaltimore (talk) 14:07, 5 September 2018 (UTC)
 * 13) Support 13 given the arguments put forward here by past and present Arbs. Best, Barkeep49 (talk) 22:51, 5 September 2018 (UTC)
 * 14) Support, even during my time on the Committee, when BASC/AUSC were still active, we rarely suffered from a shortage of input on a matter. Rather, we often found ourselves waiting for just enough votes to pass something, even though we all essentially agreed on it. I don't see any substantial harm in the Committee being reduced to 13 members, and if it does prove unworkable for some reason or future events raise the workload, the Committee can always be expanded again at a future election. Seraphimblade Talk to me 03:56, 6 September 2018 (UTC)
 * 15) Support going to 13; seems reasonable. K.e.coffman (talk) 12:41, 6 September 2018 (UTC)
 * Yep, seems very sensible to me too to go straight to 13, and a step towards the day when we can reduce the number to zero ;-) Boing! said Zebedee (talk) 19:43, 6 September 2018 (UTC)
 * 1) Support --Joshualouie711talk 23:59, 6 September 2018 (UTC)
 * 2) Support, seems like a reasonable move to streamline things. the wub "?!"  22:00, 10 September 2018 (UTC)
 * 3) Support. Abequinn14 (talk) 21:00, 12 September 2018 (UTC)
 * 4) I can see the argument that it's more important to have diversity via more members, as well as the argument that a change of this size won't actually make that much of a difference. But the way that I ultimately see it is that the biggest procedural problem encountered by the Committee really is the delay that results from waiting for one more member to chime in. From the outside, this has long looked to me like a significant issue. So I'm concluding that this is a good idea – with going straight to 13 this year – and that the negative impact on diversity won't be too bad. (And I'm also saying this with the expectation of going to a 60% threshold for two-year seats.) --Tryptofish (talk) 19:13, 29 September 2018 (UTC)
 * 5) Endorsed. The reduction seems small enough to not weaken operation and reasonable decisionmaking, yet may speed up the process. ...GELongstreet (talk) 21:48, 30 September 2018 (UTC)
 * 6) Support straight to 13, largely per Floquenbeam and Euryalus among others. Callanecc (talk • contribs • logs) 11:01, 1 October 2018 (UTC)
 * 7) Support straight to 13. Probably the biggest problem I've seen in my 4 years is how slow we are. I don't know if this will help, but I don't think it will hurt and it's worth trying. Doug Weller  talk 16:09, 1 October 2018 (UTC)

Statement #2 by Rschen7754 (status quo of 15 arbitrators)
Providing this here as an alternate option, for those who want to keep the status quo. Possible arguments: more stability in case of inactivity, more checks and balances. Rschen7754 06:05, 4 September 2018 (UTC)

Users who endorse statement #2:

 * 1) What's wrong with having more arbitrators than necessary? If they have the confidence of the community (as is required to win the elections) I don't see any reason to artificially restrict the number. To have a restriction would be similar to, "we don't need any more people with _____, so let's not give more people ____ even if everyone thinks they're qualified". Banedon (talk) 10:10, 4 September 2018 (UTC)
 * Everything the Committee does requires a consensus. The more people we have, the harder that consensus is to obtain. The community has repeatedly complained of our response times, which is fair, but those long response times are an artifact of the size of the Committee. ~ Rob 13 Talk 19:46, 4 September 2018 (UTC)
 * I still don't see why that's a problem. Arbitration as far as I can see is also a straight vote, so to reach a 50% majority is easier if there are more voters (e.g. to get a majority vote with 3 people is harder than with 30,000). Just set a deadline, and count all votes by that deadline. Banedon (talk) 23:07, 4 September 2018 (UTC)
 * There are no deadlines to our votes/discussions, and implementing any would require community ratification of a change in WP:ARBPOL, I think. That probably isn't feasible. ~ Rob 13 Talk 07:05, 5 September 2018 (UTC)
 * Then let's make that change in ARBPOL instead of this one. Banedon (talk) 12:35, 5 September 2018 (UTC)
 * That requires ratification by 100 editors, and many would disagree with imposing deadlines on the Committee. Sometimes, we need time to get to the right solution, and deadlines would be actively harmful. Other times, we're just waiting on people. It's hard to know which is the case a priori. It would be very difficult to get 100 editors in support of your proposed change. ~ Rob 13 Talk 14:25, 5 September 2018 (UTC)
 * Regardless, we should fix the source of the problem, not the symptom. If this proposal passes for example and half the arbitrators are inactive, then we're right back where we started. With deadlines, the "lack of responsiveness" issue is resolved once and for all. Banedon (talk) 00:26, 6 September 2018 (UTC)
 * While I’m very sympathetic to the “unsuitable” argument above, there have been times the past year where the bench of active arbitrators was as low as 9/15, and I suspect if we lowered the number the number of active arbs would get lower. This also has an impact on the unsuitable issue: currently 1/15 with a bunch inactive isn’t a big deal. 1/13 with a bunch inactive is much more powerful. I don’t see a larger committee creating harm and I see some benefits. TonyBallioni (talk) 13:55, 4 September 2018 (UTC)
 * 1) While more arbitrators increases the chance of an unsuitable one getting in, it also mitigates the effect of an unsuitable one does succeed in getting in. The other justifications while valid, don't seem to actually do any harm if left as-is, so "closing-up" as it were seems to have no inherent benefits or negatives in that sense, leaving the unsuitability viewpoints as the most relevant ones (either way). Nosebagbear (talk) 15:09, 4 September 2018 (UTC)
 * 2) I concur with Nosebagbear that a higher number of arbs is better for mitigating the impact of a bad arb. A smaller committee means that more power is concentrated in the hands of fewer people. While a smaller committee would presumably take less time to process a case, it seems unlikely that the removal of two arbs will make a significant difference. Lepricavark (talk) 03:47, 5 September 2018 (UTC)
 * 3) We just had an arb resign unexpectedly. Best to keep plenty of spares. Andrew D. (talk) 06:47, 5 September 2018 (UTC)
 * Worth noting that Alex had been inactive for large stretches of time. Usually, the arbs that resign are the ones that are not fully active anyway, and they're often among those that would not have been elected had the total number of arbitrators been lower. If we had 13 arbs last election cycle, Alex wouldn't have made the cut. If something truly exceptional were to happen and many arbs resigned, the Arbitration Committee does have the ability to call a special election under WP:ARBPOL. I see that as extremely unlikely, though. ~ Rob 13 Talk 07:00, 5 September 2018 (UTC)
 * , I would appreciate if you would be slightly more sensible to not paint me in a possibly negative light. If we had 13 arbs last election cycle, yourself wouldn't have made the cut too? So please just stick to the original point, in which I have no strong opinions of. Thanks, Alex Shih (talk) 07:49, 5 September 2018 (UTC)
 * Inactivity isn't negative. It's a function of real-life responsibilities, which certainly take precedence over on-wiki matters. Most arbitrators go through periods of inactivity, myself included. It's just a fact that some arbitrators spend more time on the inactive list than others; Ks0stm and DeltaQuad, both good friends of mine, have spent most of this term there. (And last election cycle would have been the year of 7 seats under a 13 arb system, so I would have been elected, but that's not terribly important.) ~ Rob 13 Talk 07:59, 5 September 2018 (UTC)
 * Mildly, the topic isn't a Monday morning quarterback about previous results; it's about the utility or otherwise of a smaller committee going forward. Let's all leave personalities out of it. -- Euryalus (talk) 03:36, 6 September 2018 (UTC)
 * 1) More Arbs mean more diversity. It's good to have Arbs with different knowledge and skill-set. Some "backbencher Arb" can have his time to shine in a special case he is knowledgeable about, and I don't see harm in having more arbs in a straightforward case either. Also since the German War Effort case for instance took 3 months to complete, it's not like the drafters can have a higher workload. --Pudeo (talk) 06:57, 5 September 2018 (UTC)
 * 2) Diversity is good, resilience is good in the case that not all arbs are available at a given time or some arbs are unable to complete their term. A way to speed up some aspects of cases is required but I don't think reducing the number of arbitrators is going to have any impact on this by itself. Thryduulf (talk) 12:06, 7 September 2018 (UTC)
 * 3) Support status quo. It is not unusual for four or five arbs to take themselves out of the rotation due to real life activities and other constraints upon their time. Thirteen minus five equals eight, and that's starting to get dangerously thin, which increases the likelihood of bad decision-making, particularly if there are abstentions. Carrite (talk) 15:39, 7 September 2018 (UTC)
 * 4) I am also not comfortable with the idea of reducing arbs for the sake of speeding up case work. Less people, less diversity and more risk of groupthink. Find better way to speed up the process, either through deadline or whatever way is better, but not through reducing them by 1 every year. –Ammarpad (talk) 05:02, 8 September 2018 (UTC)
 * 5) I'm not convinced a reduction is needed. We may have an unexpected number of drop outs / real life emergencies one year. I agree with Pudeo regarding diversity - a bigger committee allows for a wider range of opinions/approaches. If people are worried about low quantity of candidates leading to weaker appointees, I would prefer to raise the minimum percentage support from 50% to 60%. WJBscribe (talk) 15:16, 11 September 2018 (UTC)
 * 6) Most years, not all members remain active throughout the year; for some types of discussions, not all members participate; for some cases there are a substantial number of recusals. . Having only 13 on the committee would probably mean an average of only 11 active members at any one time,and based on experience might give only 7 or 8 active members on some discussions, and in some full cases. Although one might think this couldbe solved by selecting people who would remain active, this cannot be ensured, as nobody coming onto the committee will initially be aware of the extent of the workload (which is much higher than what appears on-wiki)  DGG ( talk ) 01:48, 16 September 2018 (UTC)
 * My understanding of the committee's workings is that most tasks are selected and handled by individual committee members, and are never presented to the full committee. If that's indeed correct, that most things don't require a quorum, having extra committee members does not create layers of redundancy that slow everything down, at which point the only real argument (that the committee isn't selective enough) can be resolved by other means. —Compassionate727 (T·C) 00:08, 17 September 2018 (UTC)
 * That's very incorrect, actually. Everything requires a vote of the entire Committee, basically. Sometimes arbitrators will take on individual tasks, but only in the sense of being the one that pushes it along. It still needs a vote. We routinely struggle with getting enough people commenting to even have a quorum. ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 14:45, 18 September 2018 (UTC)
 * Thanks, struck while I reconsider. —Compassionate727 (T·C) 15:37, 18 September 2018 (UTC)
 * 1) Support I prefer more diversity in the Arb tranche, and that is more likely with more Arbs, also reduces the impact of an unsuitable Arb being elected. I see the increased threshold working well with the current numbers. Peacemaker67 (click to talk to me) 09:30, 19 September 2018 (UTC)
 * 2) Support status quo. I am not convinced a reduction is needed for an odd Arb will be inactive due to real life activities and some can resign in the middle .Hence do not see why it is needed.Further diversity is better.Pharaoh of the Wizards (talk) 18:18, 25 September 2018 (UTC)
 * 3) Support status quo, and I wouldn't object to expanding it to accommodate absentees and recusals. Any editor who agrees to serve on ArbCom needs to decide beforehand if they actually have the time to devote.  Some of the concerns raised here and here did not result in any further clarity than what we had when the discussions first began.  Is it not a sign that we need more arbs participating in these discussions rather than fewer?  Ambiguous remedies for DS may explain some of the cases that end up in arbitration or at ARCA, which to me indicates some of these remedies are far more complex than the original case, not to mention the occasional admin misinterpretations and/or errors, or lack of time to thoroughly investigate a case before enforcing a t-ban or block, and that does more harm than good to the project.  Yet, we now have instances where individual admins are determining the fate of an editor in complex cases, especially when the complexity is caused by DS, and requires a decision by a committee, not an individual, so no - let's not shrink ArbCom.  While we'd all like to think our admins are perfect, not all hold the same level of trust & respect by the community.  It is simply not fair to expect admins to use their sole discretion in deciding complex cases that should be decided by ArbCom. <span style="text-shadow:#F8F8FF 0.2em 0.2em 0.4em,#F4BBFF -0.2em -0.3em 0.6em,#BFFF00 0.8em 0.8em 0.6em;color:#A2006D">Atsme ✍🏻📧 00:48, 30 September 2018 (UTC)
 * 4) Support status quo. IMO. the current number of arbs is not broken and does not need to be fixed. --Guy Macon (talk) 10:01, 2 October 2018 (UTC)

Comments

 * I have mixed feelings about this, and may just not vote in the end - but some general thoughts:
 * Awhile back I did some digging on past mail volumes and noticed there really was a substantial increase in mailing list traffic during the 18-arb era, especially when the committee was newly expanded in 2009 (and despite the fact that inactivity and resignations meant there weren't really 18 arbs most of the time). I think there's reasonable evidence that there really is significant communications overhead associated with even a slightly larger group.
 * But, if less is better than more, you'd predict that the current committee would be more efficient or effective in periods of higher inactivity. I don't have numbers on that but it's subjectively not the case. Because of inactivity, 12-13 active members is about the steady-state size anyway.
 * Arb time isn't really fungible; we're not interchangeable work units. We're not just aiming for a certain amount of aggregate arb time to get stuff done, but also for representatives of a range of skillsets and backgrounds. That's necessarily reduced with fewer people.
 * In my first year and a half or so I was on the higher end of average activity-wise, and for the last year I've been on the lower end. So I've had both the "OMG why are you all so slow, we've had this for a week, hurry up!" and the "OMG where's the fire, we've only had this for a week, slow down!" reactions to arb business. From either end of the scale, I'm not sure it's "consensus-building" that takes up all the time, or even waiting for enough noses to show up and be counted. It's really the fact that it's a group of volunteers, with little internal structure, who self-select their tasks - that means tedious or time-consuming stuff sits around while interesting or urgent stuff gets managed quickly (maybe too much so). Over a decade of discussions, internal and on-wiki, have failed to produce any significant amount of stable internal organization. So I predict that the less-desirable tasks will be just as slow with 13 as 15, and the more interesting ones will be just slightly faster. Opabinia regalis (talk) 07:58, 5 September 2018 (UTC)
 * I agree with the observations made by Opabinia regalis, particular the part about whether or not it is the consensus building or the volunteer nature and the lack of any real structure that takes up majority of the time. While reducing the number to 13 is the sensible thing to do, it wouldn't really change much in terms of general response time. Alex Shih (talk) 08:27, 5 September 2018 (UTC)
 * On the topic of internal organization, I think we should be having bi-weekly virtual meetings where we can clear out appeals and other matters quickly. Discussions that would take weeks on arbcom-l can really be handled in 10-15 minutes of conversation, from what I've seen. This is made more difficult on a larger Committee, since finding a time that works for everyone becomes impossible, but we should be able to find a time that works for most if people are willing to commit to a specific hour each week. The hours saved elsewhere would make it worthwhile. Perhaps I'll take this up after the elections. It doesn't make much sense to do it before, since any organized meeting would have to be agreed to by the new guns as well. ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 08:36, 5 September 2018 (UTC)
 * Specify minimal special elections? - If the number of arbitrators being elected does get reduced (as is looking likely), then do we need to think about putting in at least some minimal points at which special elections must be called (if more than a certain time from a standard one, obviously)? The minimal number of active arbitrators that we consider acceptable now will still be that number with a smaller max size. Between resignations and inactive 3 or 4 becomes a critical number. Nosebagbear (talk) 12:38, 5 September 2018 (UTC)
 * We actually can't do that here. It would require an amendment to the arbitration policy, which needs either 100 editors in support of the change or a motion from the Arbitration Committee to initiate. Further, keep in mind the Committee can call special elections by a simple majority vote at any time. If, for some reason, the Committee were unable to perform its duties due to resignations, I don't think we'd have any trouble knowing to initiate a special election. I highly doubt this would prove necessary. ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 14:28, 5 September 2018 (UTC)
 * I don’t think the difference between 13 and 15 arbs would have a major impact one way or the other. Newyorkbrad (talk) 16:26, 7 September 2018 (UTC)

Easier access to vote
I wish for the voting system to be changed back to when one could merely vote without having to give up personal information. GoodDay (talk) 10:34, 4 September 2018 (UTC)
 * Can you elaborate on this? We are currently using SecurePoll and have been since 2009.  Prior to that we used open on-wiki voting (sample votepage). —  xaosflux  Talk 11:44, 4 September 2018 (UTC)
 * GoodDay might be referring to the increased use of Checkuser rights to investigate not only questionable votes, but basically all votes. My memory might be failing me, but this seems new within the last 3-4 years. --Floquenbeam (talk) 17:35, 4 September 2018 (UTC)
 * If people edited, that would leave CU info too... --Rschen7754 18:10, 4 September 2018 (UTC)
 * I imagine the issue isn't whether there is CU info available, so much as its (relatively new?) widespread use on every voter. But the more I think about this, the more I think I could be completely off base, so I'll stop guessing and let GoodDay reply. --Floquenbeam (talk) 18:44, 4 September 2018 (UTC)

Yes ago, one could go to the Arb election page & merely dot those you supported, opposed or abstained from. A design came up with 3 circles by each candidate. Only one of 3 circles per candidate, could have a dot (vote) placed in it. GoodDay (talk) 19:27, 4 September 2018 (UTC)
 * But... that's basically how we do it now. And what does this have to do with personal information? Xaosflux, Rschen, ignore me, I guess I misinterpreted, but now I'm even more confused. --Floquenbeam (talk) 19:42, 4 September 2018 (UTC)
 * This may be relevant, although it did not really make things any clearer to me. Beyond My Ken (talk) 23:31, 4 September 2018 (UTC)
 * I think I got it: The scrutineer instructions page seems to imply that it is normal to run mass CheckUser on voters in SecurePoll to flush out double votes or the like. RfA and RfB voters are not to my knowledge scrutinized in this fashion. Pinging last year's scrutineers to ask whether this is an accurate description of the situation. Jo-Jo Eumerus (talk, contributions) 07:45, 5 September 2018 (UTC)
 * Nothing to do with GoodDay's "clarification". As to the private data, the original statement made was "give up personal information". No more and no less information are recorded whether you make an edit, or whether you submit a vote via SecurePoll. -- KTC (talk) 09:54, 5 September 2018 (UTC)
 * These things are not equivalent when they aren't used in the same way. Private information submitted by an edit is unlikely to be viewed by anyone before it's auto-discarded - 51402 CheckUsers were carried out by enWiki CheckUsers in 2017 many of which are probably separate checks of the same (few) accounts - whereas with SecurePoll there is the implication that in many cases someone will view the private information. Jo-Jo Eumerus (talk, contributions) 14:59, 5 September 2018 (UTC)
 * That wasn't the complaint/request. GoodDay said they would like to be able to "vote without having to give up personal information". No statement were made regarding who had access or how it's handled. The information being given up/recorded is the same whether you are making an edit, or submitting a vote via SecurePoll. -- KTC (talk) 15:05, 5 September 2018 (UTC)
 * Yes, but it is far easier for a scrutineer to access the data than it is for a CheckUser. As an ElectCom member you know this - it lists a table of everyone's usernames and the IP that they used to vote, without any second step needed to access the information. The scrutineers here are trusted stewards so it's not a huge issue IMO, but it's a fair concern that the private info is more openly revealed to the small group here. -- Ajraddatz (talk) 21:24, 5 September 2018 (UTC)

Statement #1 by TonyBallioni
Proposal: A mass message will be sent out to all eligible voters that have been active within the last 12 months. If this cannot be done within 72 hours of the voting starting, the election committee may elect to notify all eligible voters.

Rationale: Given last year’s difficulty with this, I think we should codify giving the election commission more authority to deal with any potential issues which arise. I propose keeping last year’s standard of sending a message to eligible voters who have been active in the last 12 months, and also authorizing them to send a message to all eligible voters if it proves too difficult to send out that message within 72 hours of the voting period opening. While there is the possibility of sending this out server side (which I think we should do), giving the election commission more discretion on how to handle this if issues arise would prevent some of the drama we saw last year. TonyBallioni (talk) 14:03, 4 September 2018 (UTC)

Users who endorse statement #1:

 * 1) TonyBallioni (talk) 14:03, 4 September 2018 (UTC)
 * 2) With a strong note that the Commission should be identifying this list early on. We can get a list in the next two months, guaranteed. If needed, we can involve the WMF to help generate it. The reason this was a mess last year was because it was delayed until the end. ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 15:54, 4 September 2018 (UTC)
 * 3) Support allowing the commissioners to make decisions on how to handle any messaging challenges, and encouraging work to begin as early as possible. —  xaosflux  Talk 16:59, 4 September 2018 (UTC)
 * 4) What BU Rob13 said. --Floquenbeam (talk) 17:37, 4 September 2018 (UTC)
 * 5) Per BU Rob13, definitely should be able to get a list. Galobtter (pingó mió) 17:46, 4 September 2018 (UTC)
 * 6) This is a sensible clarification establishing a failsafe in the event that unexpected problems arise as unexpected problems do. Mz7 (talk) 18:11, 4 September 2018 (UTC)
 * 7) This will make things clear on what should be done and who's responsible to avoid last year's problems. –Ammarpad (talk) 20:42, 4 September 2018 (UTC)
 * 8) Support given the challenges in getting this done last year. On the general principle of the notifications: More voters makes for a more democratic outcome, and a one-off talkpage alert is a pretty minor price to pay for getting an Arbcom that reflects more than the "insiders." Euryalus (talk) 22:48, 4 September 2018 (UTC)
 * 9) Endorse - reaching a wider voter base is a good idea, as the Arbcom election results could affect all Wikipedians, immediate or otherwise. Kirbanzo (talk) 18:04, 5 September 2018 (UTC)
 * 10) Support: would hopefully improve participation. K.e.coffman (talk) 12:41, 6 September 2018 (UTC)
 * 11) Yep. Boing! said Zebedee (talk) 19:51, 6 September 2018 (UTC)
 * 12) Support --Joshualouie711talk 23:58, 6 September 2018 (UTC)
 * 13) Support this, the past mass messages have been great at increasing participation. Wouldn't want them to be missed out. the wub "?!"  22:05, 10 September 2018 (UTC)
 * 14) We need to improve on last year. Pawnkingthree (talk) 23:58, 11 September 2018 (UTC)
 * 15) —Compassionate727 (T·C) 00:02, 17 September 2018 (UTC)
 * 16) Support Anatoliatheo (talk) 09:23, 21 September 2018 (UTC)
 * 17) While I still retain the experienced Wikipedian's skepticism about the average vote, I still believe this to be a good thing which needs to be done as a matter of principle. Vanamonde (talk) 16:00, 22 September 2018 (UTC)
 * 18) Support as seeing these messages is how I found out about much of Wikipedia's internal governance. Also, I would suggest allowing the election committee full discretion to vary this if there is a last-minute scramble. GreyGreenWhy (talk) 10:45, 23 September 2018 (UTC)
 * 19) yep. Clearly needed. Hobit (talk) 12:45, 24 September 2018 (UTC)
 * 20) This seems reasonable to me, although I share the opinion of some other editors that it gets to be rather spammy, and that a watchlist notice might be more appropriate. --Tryptofish (talk) 19:07, 29 September 2018 (UTC)
 * Yep, good idea. Callanecc (talk • contribs • logs) 11:03, 1 October 2018 (UTC)

Discuss
if I recall correctly, the "problem" wasn't "sending the message" so much as generating the list of recipients (c.f. Wikipedia_talk:Arbitration_Committee_Elections_December_2017/Coordination). I think this issue was with both the "full list" as well as the "active in last year" list. I'm generally supportive of letting the commission decide to message more people if they deep it necessary though. Ensuring that a resource is selected to actually perform this list making early on is key here. — xaosflux  Talk 15:00, 4 September 2018 (UTC)
 * I thought the issue at the end was the filtering process for the “active list” that added additional delays (and that’s still what that read like to me, but I could be misreading.) Regardless, like you say, giving the commission the discretion to alert more people if there is an issue with the filtering is ideal. Hopefully it won’t come up this year if we can run it server side (ping ), but having contingency plans around beforehand is good. TonyBallioni (talk) 15:09, 4 September 2018 (UTC)
 * that was the final issue (its a bit hazy now, I did the "send messages" part once the list was good - but I didn't build the lists) - I think another issue was bad encoding for everyone who didn't have plain AaZz09 usernames. The "active" list last year was ~60,000 recipients, I'm not sure what the "total eligible" count was.  One item that was discussed that I don't think is being clearly covered here yet is: If the total list can't be reduced to the "active" users what can be done.  You've proposed allowing the commission to expand it, but what if they choose not to - just send no message? —  xaosflux  Talk 15:31, 4 September 2018 (UTC)
 * If it’s impossible to send, then it can’t be sent, but I’d want them to try. The controversy last year was to send to a larger list than discussed. I’m hoping this won’t be an issue this time around, and I’m fine filing a phab ticket about it when this closes, which should hopefully make it moot. TonyBallioni (talk) 16:08, 4 September 2018 (UTC)

Mass-messaging 60k+ users strikes me as spammy; IMHO a watchlist notice would better serve this purpose - F ASTILY   22:17, 4 September 2018 (UTC)
 * This is what we’ve done for the last several years. The only thing this changes from last year is giving the election commission the ability to send it out to the easier to produce full list if there are issues with creating a filtered list. The reason this goes out in addition to watchlist notices is that ACE really is an inside baseball thing that impacts many users who might not even know what ArbCom is, so wouldn’t click a watchlist notice. Their voices mater to, so we send the message out to let them know they can vote. TonyBallioni (talk) 22:30, 4 September 2018 (UTC)
 * the 60000 talk sends were a bit "spammy" but they are sent with MMS that uses the bot flag so it is minimally bothersome to others, we had a lot of outcry for not sending it last year, and participation greatly increased after they were sent. —  xaosflux  Talk 17:16, 5 September 2018 (UTC)


 * As a general point of business, how are we determining whether this mass messaging proposal will reach a consensus? <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 15:31, 5 September 2018 (UTC)
 * At the end of the RFC period, an uninvolved admin (me, the last 3 years, but not this year) determines whether there's a consensus for each section, and if so, what the consensus actually is. --Floquenbeam (talk) 17:47, 5 September 2018 (UTC)
 * My question is not about who will determine the consensus, but how will a consensus be determined? If it is a given that this RFC is to follow the standard closing process, it should have been structured along the same standard Wikipedia conventions for RFC consensus building, e.g. the inclusion of an oppose section. The mass message proposal only has a 'support' and a 'discuss' section. There are no clear instruction that any opponents to the proposal should comment in the 'discuss' section. Meanwhile, the structure heavily favours collecting support opinions only. I do not see how any closer could realistically determine whether a fair consensus has been reached based upon the structure and process so far. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 20:17, 5 September 2018 (UTC)
 * , fixed. My fault. I was basing it on above, which didn’t have oppose headers. I knew traditionally these are position statement RfCs, and there is a format of those where anyone opposing would create a 2nd statement for people to endorse. I looked back at last years RfC, and it was position statements plus endorse/reject, so I added it to all the current statements. TonyBallioni (talk) 20:34, 5 September 2018 (UTC)
 * (e/c with Tony) These RFC's have, in my experience, not been perfectly organized and laid out. But if people are opposed, they should be bold and add an oppose section (which I see Tony has done while I was writing this).  If their opinion is more nuanced that a simple oppose, they should add a "Statement by User:X" section that has its own support and oppose subsection. In any case, if neither of those two things is done, an experienced closer can still read the discussion and figure out what there is a consensus for; they aren't going to just count the supports, note that there is no oppose section, and call it unanimous. --Floquenbeam (talk) 20:36, 5 September 2018 (UTC)

Write a short general guide to voting
Covering topics such as "what is arbitration", "what do arbitrators do", "what to look for in a candidate", etc. Reason is in view of this from the last election. Banedon (talk) 23:36, 4 September 2018 (UTC)
 * I agree. In addition to voter's guide which generally isn't really all that helpful for new editors, a short general guide would certainly be positive. I will be interested to write one if nothing is submitted by then. Alex Shih (talk) 07:57, 5 September 2018 (UTC)
 * It's worth noting that "what to look for in a candidate" is highly subjective. If there is to be a general guide that is held up as a neutral guide to what people are voting for, this part should be omitted. ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 08:38, 5 September 2018 (UTC)
 * Could still write something like "here're some things people look for in candidates, but your mileage may differ. You'll have to come to your own conclusion." Banedon (talk) 12:30, 5 September 2018 (UTC)
 * Or just link to the voting guides for that. I think there's serious potential for bias/disagreement if anyone even summarizes some things to look for in candidates in Wikipedia's voice. ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 14:23, 5 September 2018 (UTC)
 * The linked thread illustrates the problem with voting guides however. They mostly say who the author is voting for, and are intended for people who already have some familiarity with arbitration. Banedon (talk) 00:24, 6 September 2018 (UTC)
 * The general guide seems like a good idea though. What is Arbitration, what extra rights, how important access to non-public information is and what does it actually mean are all questions people should be able to ask. It may be worth covering some of the general controversial topics - like "admin status", editing history, previous arb cases and so on - we have Arguments to avoid in adminship discussions which sort of does that at RfA level - not that I think people care about that essay any more! <b style="text-shadow:0 -1px #DDD,1px 0 #DDD,0 1px #DDD,-1px 0 #DDD; color:#000;">Worm</b>TT(<b style="color:#060;">talk</b>) 16:43, 5 September 2018 (UTC)
 * Arbitration/Guide to arbitration exists and is pretty readable (or at least as readable as a guide to ArbCom is likely to get) in regards to the on-wiki stuff. TonyBallioni (talk) 16:53, 5 September 2018 (UTC)
 * It's quite long however, and one can certainly write a guide that's more relevant to the elections. Banedon (talk) 00:24, 6 September 2018 (UTC)
 * Some thought should be given to how much vetting by a wider community should happen before the adoption of such a guide. --Tryptofish (talk) 19:05, 29 September 2018 (UTC)

I have started a draft at 5 Minute guide to ArbCom Elections. Everyone is explicitly invited to edit it into something that will serve the purpose. I have tried to include a one-sentence neutral bit about administrators typically being selected for arbcom, but mostly I just kicked users back to voting guides for much of the candidate selection portion. Tazerdadog (talk) 03:17, 9 September 2018 (UTC)

Statement #1 by WJBscribe (raise from 50% to 60%)
The threshold for appointment should be raised from 50% to 60%. I note the concerns above regarding numbers of candidates vs. number of available seat leading to a decrease in quality of appointees. I suggest, regardless of whether or not the total size of the committee is reduced, that the threshold for appointment is raised from 50% to 60%. Looking back over the years, invariably all appointees have had this level of support with only a couple of exceptions (one of which was only fractionally below 60%). Based on past election years, it appears we can staff the committee adequately without looking at candidates who have <60% support. <strong style="font-variant:small-caps">WJBscribe (talk) 15:29, 11 September 2018 (UTC)
 * Proposal
 * Rationale

Users who endorse statement #1:

 * 1) Without this change, there is a risk that we could end up with a smaller committee and weaker candidates, giving greater influence to those appointed with a bare majority of the community. <strong style="font-variant:small-caps">WJBscribe (talk) 15:29, 11 September 2018 (UTC)
 * 2) I'd support this. With respect to the the half a dozen arbs who didn't quite make the 60% mark over the past 10 years, I do not have any issue with setting the level higher. <b style="text-shadow:0 -1px #DDD,1px 0 #DDD,0 1px #DDD,-1px 0 #DDD; color:#000;">Worm</b>TT(<b style="color:#060;">talk</b>) 15:36, 11 September 2018 (UTC)
 * 3) I definitely support this. I'm not convinced that if someone can't get at least 60% that they should be given the responsibility and tools. Doug Weller  talk 16:36, 11 September 2018 (UTC)
 * 4) Absolutely. Per Doug Weller. Bishonen &#124; talk 19:00, 11 September 2018 (UTC).
 * 5) Yes, I'm happy to support this too. Boing! said Zebedee (talk) 19:03, 11 September 2018 (UTC) I've changed my mind in favour of the option below which is now my first choice, but I'd support this as second choice. Boing! said Zebedee (talk) 09:01, 16 September 2018 (UTC)
 * 6) Although I'd suggest 60.67%, I like this in combination with decreasing to 13. But I support both ideas independently as well. --Floquenbeam (talk) 19:06, 11 September 2018 (UTC)
 * 7) The proposal states it is "regardless of whether or not the total size of the committee is reduced" and therefore agree the bar should be raised as this metric will become increasingly more important. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 19:52, 11 September 2018 (UTC)
 * 8) Actually having 50% is a mediocre performance in every decision-making/assessment system. If this election were to be determined through discussion-mode we know approval of 50% will never be acceptable bar for promotion. –Ammarpad (talk) 21:00, 11 September 2018 (UTC)
 * 9) I am very happy to support this proposal as well. Alex Shih (talk) 22:25, 11 September 2018 (UTC)
 * 10) Per above. As this is a vote, I think an overwhelming majority is beneficial for the project.  With Neutrals uncounted, I'm not worried about a huge lack of candidates. ~  Amory <small style="color:#555"> (u • t • c) 01:38, 12 September 2018 (UTC)
 * 11) Strongly prefer this to statement 2. This is a question of whether we trust someone to be on the committee. If more than 40% of voters don't trust you, you shouldn't be deciding the some of the most important (and least visible) happenings on the project. —Compassionate727 (T·C) 00:12, 17 September 2018 (UTC)
 * 12) Combined with maintenance of the current committee size (which I support), this will give greater assurance against unsuitable Arbs being elected. If it results in reduction of the committee size, I still think it is worth it. Peacemaker67 (click to talk to me) 09:35, 19 September 2018 (UTC)
 * 13) I am supporting this and would also consider supporting higher percentage. Anatoliatheo (talk) 09:25, 21 September 2018 (UTC)
 * 14) As first choice. To be completely honest, I do not follow the reasoning behind the proposal below. The concern described above is that a small candidate pool and a threshold of 50% will allow an unsuitable candidate to be elected. I find this a reasonable concern, particularly for our one process that is not consensus based, and which is therefore more open, theoretically, to gaming than RFA, for instance. However, if a candidate is unsuitable, then they are unsuitable to serve for any period of time. I'd rather have a smaller committee than a committee with an unsuitable candidate. Vanamonde (talk) 16:08, 22 September 2018 (UTC)
 * 15) Endorsed as per above. If the committee gets smaller because candidates don´t get supported by the majority of the voting community then so be it. ...GELongstreet (talk) 21:44, 30 September 2018 (UTC)
 * 16) Support as second choice over #2. Callanecc (talk • contribs • logs) 11:07, 1 October 2018 (UTC)
 * 17) First choice. Doug Weller  talk 16:19, 1 October 2018 (UTC)
 * 18) First choice. We require much higher consensus for admins, and I would rather a seat be empty than sit someone without a strong consensus of support. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 06:48, 2 October 2018 (UTC)

Users who reject statement #1

 * 1) Would likely have a similar impact as changing the number of seats without the potential for effeciency increases. Also noting I’m not opposed to it in itself, but oppose it as an alternative to lowering the number of seats, which is what it appears to have been introduced as. TonyBallioni (talk) 15:33, 11 September 2018 (UTC)
 * Just to note that, although I personally prefer this to reducing the size of the committee, the proposal I made was to increase the required support needed "regardless of whether or not the total size of the committee is reduced". <strong style="font-variant:small-caps">WJBscribe (talk) 11:47, 12 September 2018 (UTC)
 * , fair point. I think there is a concern that what could happen at some point is you get a year with only 3 or 4 people elected if this is the requirement. I’m concerned with the “unsuitable” candidate issue, but I also think a reduction has the impact of making that less likely while keeping an adequately sized committee. If this passes and the reduction passes, then we’d run the risk of having 10 arbs depending on the year’s candidates. I think a reduction is the most straight forward way of improving the committee currently. That being said, I’m not super opposed to this, I just think a reduction without an increase in minimums is better. TonyBallioni (talk) 18:00, 12 September 2018 (UTC)
 * 1) Even back in 2013 and 2014, when there was no dearth of candidates, this would have resulted in a seat remaining unfilled. I consider those arbitrators to have been quite good. Whereas lowering the number of seats provides the Committee with flexibility in calling a special election to almost certainly fill some seats, this would provide us little flexibility if the community did not find any candidate to be suitable to this margin. I would support increasing this to 55%, perhaps. ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 22:20, 11 September 2018 (UTC)
 * 2) Essentially per TonyBallioni, with the addition that I'd be concerned with a prospect of ArbCom Being unable to fill its seats long-term if this is passed. I'd rather have the controlled reduction to 13, instead of an uncontrolled reduction to who-knows-how-many candidates who pass the bar.  I'd also be concerned about changes in voting patterns this change will cause. Tazerdadog (talk) 00:57, 12 September 2018 (UTC)
 * 3) I much prefer the proposal below. Thryduulf (talk) 14:35, 15 September 2018 (UTC)
 * 4) Per Tazerdadog, I am concerned this could make ArbCom's membership very unpredictable. GreyGreenWhy (talk) 10:45, 23 September 2018 (UTC)
 * 5) Too worried about not enough members. And I feel this is solving a problem we don't have yet.  Hobit (talk) 12:43, 24 September 2018 (UTC)
 * 6) Tactical voting is allowed .Hence users can oppose candidates not on there list for no reason .Hence oppose this .Pharaoh of the Wizards (talk) 18:29, 25 September 2018 (UTC)
 * 7) I much prefer the alternative below, because I think there is a very real risk of electing too few members for the Committee to work well, with multiple points of view. --Tryptofish (talk) 19:01, 29 September 2018 (UTC)
 * 8) 50% is fair play regardless of how many or how long. We should probably adopt the same principals to adminship...to look more like a jury of our peers. <span style="text-shadow:#F8F8FF 0.2em 0.2em 0.4em,#F4BBFF -0.2em -0.3em 0.6em,#BFFF00 0.8em 0.8em 0.6em;color:#A2006D">Atsme ✍🏻📧 21:57, 30 September 2018 (UTC)
 * 9) Statement #2 is my only choice. Kevin ( aka L235 ·&#32; t ·&#32; c) 01:24, 1 October 2018 (UTC)
 * 10) Another attempt to fix something that isn't broken. --Guy Macon (talk) 10:05, 2 October 2018 (UTC)

Discussion
I doubt this will matter this year; if there are only 6 seats up for election, it's likely the elected candidates will all have >60% approval. Looking at historical results, one option could be that the requirement for a 2-year term is 60%, with people getting less than that amount only being elected to a 1-year term. In practice, I don't think that "compromise" solves any problems at all. power~enwiki ( π, ν ) 01:04, 12 September 2018 (UTC)
 * I'd support this, it provides a safety valve for filling seats if we otherwise would be unable to under this proposal.Tazerdadog (talk) 01:48, 12 September 2018 (UTC)
 * I would very much support this. My biggest concern is flexibility if circumstances change. For instance, what if, in the future, we get 20 candidates after the rules of the election are in place? Many voters just vote for the number of seats available (myself included), since that is the optimal voting strategy that maximizes the chances of the seats being filled with the candidates you want. If many voters behaved this way, we would struggle to get enough candidates above 60% support. Historical data seems to suggest a significant number of voters do behave this way, since support percentages were, on average, lower when we had a larger number of candidates. Having a "safety valve" by which we can make one-year appointments of those in the realm of 50-60% (or even 55-60%) would greatly increase my comfort with this proposal. ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 04:43, 12 September 2018 (UTC)
 * I guess it depends on whether you think maintaining the numbers matter more than the level of community confidence. I'd rather have a committee of only, say, 7/8 Arbs with 60%+ than a committee twice that size where a majority had less than 60% support from the community. Based on past elections, I really don't see a problem filling (or near filling) the committee with those with 60%+ support. If that causes problems, no doubt there'd be an RfC proposal the following year to reduce the threshold. Also, I don't think there's any harm in discouraging the kind of tactical voting you describe - voters should be supporting those they believe suitable to serve as arbitrators, not their favourites, in my opinion. <strong style="font-variant:small-caps">WJBscribe (talk) 11:50, 12 September 2018 (UTC)

How about a safety valve based on the mininum size of the committee that is thought feasible? Something along the lines of, "in the event that appointing only those candidates with more than 60% support will result in a committee size of less than [11], those candidates with the highest support between 50% and 60% or more shall be appointed to serve 1 year terms to bring the committee to a total of [11]." <strong style="font-variant:small-caps">WJBscribe (talk) 11:55, 12 September 2018 (UTC)
 * I think a Committee of 11 members is already far too small. People get busy and go inactive. With a Committee of 11, we could realistically have cases decided by 5-6 arbs, which I think is below optimal. We'd also be in the position where a single long-term inactive arb could justify a special election, which isn't ideal. If you would support this, I guess I don't get why you wouldn't support similar one-year terms to bring the threshold up to the normal amount of arbs. This is especially true since we'd be experimenting with a 13-person Committee for the first time. Maybe do a 13-person Committee with 60% support required and one-year terms to bring it up to 13 in the first year, then revisit making that requirement stricter after we know for sure that size works well? ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 17:27, 12 September 2018 (UTC)
 * This safety valve is sufficient to address my concerns. Tazerdadog (talk) 00:45, 13 September 2018 (UTC)

Statement #2 by WJBscribe (raise from 50% to 60% for two year terms only)
The threshold for appointment to two year terms should be raised from 50% to 60%, with candidates who are appointed with >50% but <60% receiving one year terms. A more nuanced change than my previous proposal, designed to address the risk that the committee becomes too small. This would mean that only candidates with 60% support can be appointed to two year terms, with any remaining vacancies to be filled with candidates who have 50%+ support. <strong style="font-variant:small-caps">WJBscribe (talk) 10:25, 13 September 2018 (UTC)
 * Alternative proposal
 * Rationale

Users who endorse statement #2:

 * 1) Happy with this, but my preference would be statement 1 (i.e. raising threshold for all appointments). I note that (at least since we started voting through SecurePoll in 2009), no candidate has been appointed to more than a one year term without 60%+ support. <strong style="font-variant:small-caps">WJBscribe (talk) 10:29, 13 September 2018 (UTC)
 * 2) Similiarly, happy with this, but prefer the blanket threshold. <b style="text-shadow:0 -1px #DDD,1px 0 #DDD,0 1px #DDD,-1px 0 #DDD; color:#000;">Worm</b>TT(<b style="color:#060;">talk</b>) 10:43, 13 September 2018 (UTC)
 * 3) Support this. Would even support the threshold for a one-year term being 55-60%. ~ Rob 13 <sup style="margin-left:-1.0ex;">Talk 16:09, 13 September 2018 (UTC)
 * 4) Support as second choice, prefer all-round 60%. Boing! said Zebedee (talk) 16:12, 13 September 2018 (UTC) Support as first choice, all-round 60% now second choice. Boing! said Zebedee (talk) 09:02, 16 September 2018 (UTC)
 * 5) I’d be fine with this. TonyBallioni (talk) 16:19, 13 September 2018 (UTC)
 * 6) Support this (of course.) Prefer Statement 1, however. –Ammarpad (talk) 16:58, 13 September 2018 (UTC)
 * 7) I'd support this only as an alternative to #1, in case support for that one tanks for some reason. Better than nothing, but worse that 60% for everyone. --Floquenbeam (talk) 18:00, 13 September 2018 (UTC)
 * 8) Support - Only choice. Tazerdadog (talk) 01:01, 14 September 2018 (UTC)
 * 9) I support this – jury's still out on #1, but this would be my preference even if I supported #1 more. Kevin ( aka L235 ·&#32; t ·&#32; c) 22:16, 14 September 2018 (UTC)
 * 10) No problem with this. -- KTC (talk) 22:29, 14 September 2018 (UTC)
 * 11) Support this. Thryduulf (talk) 14:35, 15 September 2018 (UTC)
 * 12) Definitely better than 60% for everyone.  DGG ( talk ) 01:52, 16 September 2018 (UTC)
 * 13) Tactical voting is allowed .Hence users can oppose candidates not on there list but as we saw in the last elections 2017 there were only 12 candidates for 8 seats in 2017.Hence support this.Pharaoh of the Wizards (talk) 18:27, 25 September 2018 (UTC)
 * 14) Strong support for this approach. It's a good idea to have a reasonably high bar for two-year terms, but we need some flexibility for when there simply aren't enough candidates who reach that level. --Tryptofish (talk) 19:03, 29 September 2018 (UTC)
 * 15) Support as first choice over #1. Callanecc (talk • contribs • logs) 11:06, 1 October 2018 (UTC)
 * 16) Second choice. Doug Weller  talk 16:20, 1 October 2018 (UTC)
 * 17) Second choice. I would prefer those elected to arbcom have greater consensus for their seat than simple majority no matter the term length, but it's a very good compromise that I can live with. Wugapodes [thɔk] [ˈkan.ˌʧɻɪbz] 06:50, 2 October 2018 (UTC)

Users who reject statement #2:

 * 1) Still don't see a need for this and think it could do more harm (making the committee too small) than help (keeping off riff raff?). Hobit (talk) 12:44, 24 September 2018 (UTC)
 * 2) 50% - and holding. Popularity has too many variables. I'm of the mind that keeping the batting averages at 50% we get a committee that looks more like a jury of our peers.  Exemplary candidates are going to garner the most support anyway, so it's probably not worth changing. <span style="text-shadow:#F8F8FF 0.2em 0.2em 0.4em,#F4BBFF -0.2em -0.3em 0.6em,#BFFF00 0.8em 0.8em 0.6em;color:#A2006D">Atsme ✍🏻📧 22:10, 30 September 2018 (UTC)
 * 3) Conditional oppose. In looking back at WP:ACE2017, WP:ACE2016, and WP:ACE2015, none of the elected candidates would have been affected by this change. If the total number of appointments in ACE2018 is reduced, I think the likelihood of a situation where the Committee is faced with multiple vacancies due to the change in criteria is low. Very low. Any unfilled seats would be up for re-appointment in the following election. I think it is far more important to have individuals appointed to the Committee with who the community has a higher degree of trust and should not be traded for a very unlikely scenario in vacancy. If the Committee remains the same size, then I suppose Statement #2. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 19:33, 1 October 2018 (UTC)
 * 4) Another attempt to fix something that isn't broken. --Guy Macon (talk) 10:06, 2 October 2018 (UTC)
 * 5) Per my comment above. Vanamonde (talk) 14:52, 2 October 2018 (UTC)