Wikipedia:2015 administrator election reform/Phase I/RfC

Introduction

Purpose

The intent of this RfC is to broadly identify the problem(s) with our administrator election process. As was mentioned just now, we are focusing on broad issues here. Assuming that any of these issues obtain consensus (as described in the "Instructions for closers" section), there will likely be a second RfC that will follow this one, so that we can narrow our range of focus and expand on the issues discussed here.

Instructions for voters

There are eight proposals to start with, but additional proposals may be added within the first three days of launching. This narrow window exists to prevent proposals from getting less attention, as would likely happen if they were added near the end of the RfC. (For instance, a proposal existing from the start would likely obtain more attention than a proposal added a few days before the end of the RfC.) Editors are to support, oppose, or simply comment on (in the "Comments" section) the proposals accordingly. To represent the highest level of consensus possible, participants are encouraged to vote on as many proposals as they are able.

Instructions for closers

The panel of closers, comprising three editors, shall have the discretion to determine which proposals attained consensus. They should, however, be lenient when determining which proposals should advance to formal consideration, since it would be beneficial to discuss as many potential issues as possible (except those which clearly did not attain consensus).

Voting

A: High standards

Participants at RfA generally expect too much of candidates (e.g., have excessively strict criteria). Those who !vote at RfA should lower their expectations and understand that the majority of candidates will be less than perfect.

Support A

1. Support The current expectations of !voters are unrealistic. Several years of experience, tens of thousands of edits, accurate participation in all admin areas, the right "hit rate" at AfD. It's nonsense. My idea of the right admin candidate is simple: the candidate has been around for a time (about 9 months to a year), shows no signs of habitually making personal attacks, and has demonstrated knowledge of basic policies. Some oppose for extremely petty reasons not grounded in policy. We were much less strict previous years, and Wikipedia didn't plunge into ruin. --Biblioworm 00:20, 15 October 2015 (UTC)
2. Support – in general, I find this to be true. The most extreme version of this is those voters who have already expected candidates to have participated in virtually every facet of the project before supporting for RfA. --IJBall (contribs • talk) 01:52, 15 October 2015 (UTC)
3. Perfection is impossible. 100s and 100s of good edits, and an opposer will pin point one bad CSD or AfD nomination. That doesn't really make sense. And the next thing is asking for GAs and FAs. It should be understood that not all editors are native or professional English speakers and it is hard for non-native or non-professional English speakers to promote articles to GAs or FAs. Plus less or no content creations doesn't necessarily mean they have little or no idea on policies. Regards—☮JAaron95 ^Talk 04:43, 15 October 2015 (UTC)
4. Support. The "requirements" people look for are the source of the problem. Adminship is not about content creation and is not about being "right" in a discussion. —⁠烏⁠Γ ^(kaw), 05:17, 15 October 2015 (UTC)
5. Support. Well, alright, I'll support this. But what effect will supporting this have? It seems like a petition, or maybe an Arbcom principle. Are we going to vote on the "purpose of Wikipedia" like Arbcom? Well, anyway, I agree. Let's just focus on one aspect of high expectations: numbers of edits. Someone with 6500 edits would probably be told they need more experience, but that's enough to put you on the list of 10,000 most active Wikipedians of all time. There are only 2100 people have more than 30K edits total, and there are only 3300 people who make more than 100 edits per month on English Wikipedia. I think people probably should have some kind of experience with content creation, voting at AfD, new page patrol, and speedy deletion. But these should be indicators of suitability, not a checklist of "must have" qualifications. NinjaRobotPirate (talk) 05:39, 15 October 2015 (UTC)
6. Not sure I agree with the premis for promotion suggested here but overall we do have unreasonable expectations for candidates. Spartaz ^Humbug! 08:16, 15 October 2015 (UTC)
7. Support. Agree with everything stated by Biblioworm and NinjaRobotPirate, above. Also, the backlogs on Wikipedia are getting worse. — Cirt (talk) 08:53, 15 October 2015 (UTC)
8. Support. One year and 5000 edits are more than enough if they come with clue, and perfection is something we should not expect from anybody. The ability to achieve something great without asking for immediate perfection is central to Wikipedia's success. —Kusma (t·c) 09:45, 15 October 2015 (UTC)
9. we have moved from a system where people check the candidate's edits and come to a view as to whether they would make a good admin to one where few really check edits, not least because the expectation is that candidates should have too many edits to be checked. So some expectations are higher and unrealistic, others are lower. I don't believe that overall the standards are as effective at screening out people who would make bad admins, but some arbitrary bits such as edit count and tenure are higher. ϢereSpielChequers 12:48, 15 October 2015 (UTC)
10. support, though I have no idea how this will lead to an actionable recommendation. Protonk (talk) 13:38, 15 October 2015 (UTC)
11. Support: the "NOTNOW" opposes in this RfA (a recent prominent example) show just how unreasonable people's base expectations are. But in RfAs that are close to passing, we hit a different problem: lots of people have very high standards in a specific area. So some people expect unreasonable levels of 'content creation', others value AfD stats too much, others expect candidates to make X posts to random admin boards Y and Z etc. These are significant enough minorities to cause an RfA to fail, so for a candidate to pass they have to appease every single one of these groups, leading to a requirement for them to pass every single high standard if they want their RfA to succeed. — Bilorv_(talk)^(c)(e) 15:25, 15 October 2015 (UTC)
12. Support All a candidate has to show is a good level of experience, competence and confidence. That doesn't translate into perfection or haughty standards. ^{My name is}not^dave (talk/contribs) 15:34, 15 October 2015 (UTC)
13. Support - the main question we need to answer, in order to vote in an RFA, is: Are we better off with this user handling the issues (s)he claims (s)he will, or without? Obviously, if the user would do the job badly, then we're better off without. However, a user needs not to have any experience beyond what's necessary to understand the policies behind his/her planned areas. For example, having written an FA makes a user no better at recongizing edit warring or sockpuppetry, or at assessing the level of notability of the topic of a specific article, etc. And age is certainly of little value - I doubt a 5-year-old could possibly edit on an adult level well enough to pass an RFA, but a 15-year-old who can certainly should be able to run with no disadvantage due to age. עוד מישהו Od Mishehu 19:30, 15 October 2015 (UTC)
14. Support. While trusting a user with the tools is a subjective judgement, there are cases where the standards of voters are simply too high. Requiring featured articles, 10K+ contributions, a certain number of contributions to AN & ANI, digging into problematic edits from several years back, are standards that show up at RfA and are generally higher than should be necessary. ---- Patar knight - ^chat/_{contributions} 02:09, 16 October 2015 (UTC)
15. Everyking (talk) 14:13, 18 October 2015 (UTC)

Oppose A

1. I want admins to be seasoned. A year is a minimum. Usually, usually, a significant number of article edits, some article creations (not redirects or stubs), that's a minimum too. Come on--we want an admin to have judgment, to have a feel for the community and for its members, to know what it's like to write an article so they know what it's like to get one's work deleted, etc. Drmies (talk) 00:53, 15 October 2015 (UTC)
2. Current policy WP:Consensus ("try to persuade others, using reasons based in policy, sources, and common sense"), WP:Administrators ("Any editor can comment on a [RfA] request, and each editor will assess their confidence in a particular candidate's readiness in their own way.") is extremely broad in considering individualized factors gauging judgement of "trust" and "common sense" -- so are the instructions at WP:RfA. By measure of policy then, the support arguments for this proposition must fail. Alanscottwalker (talk) 01:34, 15 October 2015 (UTC)
3. RFA has no shortage of editors who will comment upon (and ridicule, if appropriate) overly strict and/or unreasonable criteria for adminship. Neil916 (Talk) 01:45, 15 October 2015 (UTC)
4. While we certainly shouldn't expect prospective admins to have the gift of the gab or a golden tongue, we should certainly expect them to have experience in mediation, negotiation and conciliation especially those that work in contentious discussions. Blackmane (talk) 01:53, 15 October 2015 (UTC)
5. The idea that becoming an admin is "no big deal" is a remnant from a different age, and is not relevant today. Admins need to be well vetted. Although it's easy enough to find faults in the current system, this is not one of them. BMK (talk) 02:13, 15 October 2015 (UTC)
6. WP:NOBIGDEAL is a big deal now. If you want to talk about being too strict you can review our GA and FA criteria, and the effort it takes for a sole person to achieve those goals in comparison to what it took many years ago. The same is true with adminship, its criteria has grown stricter as the project grew large. Candidates should know the ins and outs. Overall I don't think our expectations are excessive or unreasonable given the circumstances, but that's not to say an RfA won't be a harrowing experience, but it can be, but for different reasons — MusikAnimal ^talk 05:45, 15 October 2015 (UTC)
7. It's not clear how this is supposed to work. For example, Biblioworm opposed Montanabw's recent RfA while I supported. We were both aware of a general issue - her feisty behaviour - but took a different view about it. Was our judgement too strict or too lenient? It seems impossible to say because there are no standards or measures for this. My impression is that, if we were to be more relaxed about this then we'd use a significantly different system which would be more automatic - like autoconfirmed or autopatrolled status. That would be more efficient but we'd then need term limits and a better process for removing the tools. Andrew D. (talk) 11:44, 15 October 2015 (UTC)
8. Cats do not go back into bags - Wikipedia is no longer small, irrelevant potatoes, and we can't treat it as such. WilyD 12:36, 15 October 2015 (UTC)
9. Since administrators are effectively lifetime positions, removable only in case of egregious misconduct, it's only natural that users are difficult and exacting in RfAs. Even without that factor, users have a right to expect editors who have some experience in content creation. If the tools were ever unbundled, that might not be necessary. If the procedures for removing admins were made less onerous on all concerned, the process itself might ease. Coretheapple (talk) 15:21, 15 October 2015 (UTC)
10. While the "high expectations" are somewhat of a problem, the solution is to respectfully ask voters with these expectations to reconsider them. If they continue to vote with high expectations, it's their standard, and since RfA is based on community consensus, they should be allowed to vote with such expectations. I personally believe that the general standard (~10,000 edits and 2 years of editing on average) is fine. Esquivalience ^t 21:13, 15 October 2015 (UTC)
11. RFA can be strict at times but for the most part I'd say overall it's well balanced, Whilst I don't expect every potential admin to know everything here I do expect them to have atleast some knowledge of things. –Davey2010^Talk 00:25, 16 October 2015 (UTC)
12. Too often in this Trophy Generation, people equate "having standards" with "having high standards." It's OK to have standards. Townlake (talk) 01:58, 16 October 2015 (UTC)
13. Makes little sense, as drafted. Telling people who !vote that they have to lower their standards when there is no benchmark standard doesn't get any where near the "is this practical" test. It isn't. My standards might already be low. I'm not going to agree to lower them further just so that a few popular editors can be scrape through. Leaky Caldron 15:50, 16 October 2015 (UTC)
14. By and large, those who vote regularly, intelligently, and objectively at RfA practice a code of criteria that a vast number of editors of the right calibre for adminship easily exceed. We must be wary however of those who apply ludicrous barriers to adminship, vote against the system rather than against the candidate, and those who just never seem to get it right. Kudpung กุดผึ้ง (talk) 16:49, 16 October 2015 (UTC)
15. The proposal has no practical effect because those opposing a candidate do so because they do not think that candidate is suitable at this time—many would be quite happy to accept less than perfect but they would still think that a particular candidate was not suitable. RfA does not need motherhood statements. Johnuniq (talk) 09:21, 17 October 2015 (UTC)
16. I believe most !voters have reasonable standards. I may not agree with those standards, but I can respect them. I don't want to impose my judgment values on others, and I don't want others to impose theirs on mine. Glrx (talk) 17:44, 17 October 2015 (UTC)
17. Oppose lowering standards absent clear evidence that qualified individuals are regularly failing to pass their RfA. RO^(talk) 17:57, 17 October 2015 (UTC)
18. Oppose What defines "excessively strict criteria"? I feel this question is badly worded. - Knowledgekid87 (talk) 21:34, 18 October 2015 (UTC)

Comments on A

• 1) Unless RfA is changed from "consensus about whether other editors trust the candidate" to "consensus on whether the editor fulfills a community-crafted checklist" every editor will have his own standards and some editors will have standards that others consider unreasonably high. 2) It's very very unlikely that RfA will be anything other than a "consensus about whether other editors trust the candidate" any time soon. davidwr/(talk)/(contribs) 02:34, 15 October 2015 (UTC)
• I have always found it very amazing indeed that this is so difficult to see for some. We rely on the admins elected back in the day when the standards were lower. There is no question that they were. As recently as a few years ago, we still had admins being elected with only about a year of experience and less than 15k edits. Wikipedia did not crash amist all the alleged abusive (or incompetent) admins that are supposed to result when we're not incredibly strict, right? In fact, without them, Wikipedia would have tanked a long time ago. Several admins that were elected back in that time are still active today, so the effect of standards increases may not be extremely obvious. But, inevitably, these old-timers will retire, and with the strictness of today's process, we will not have anywhere near enough to compensate for our losses. Mark and remember my words: eventually, in a few years' time, this will happen and everyone will see that we annoying "gloom and doom" predictors were right after all. But really, is there any evidence to suggest that we had loads of more abusive admins from the older period than we have now? Bad apples will always get through, even if we set the bar at 90% and made our standards even stricter. Of course, it should be realized that there will almost certainly be more desysopped admins from the older period, since there have been more years since they were elected and therefore they have had more time to do something wrong and lose their tools. But the same thing will happen in a few years to those at least a few of those who we elect now. I suspect that over time, the ratio of admins elected in a given year that are desysopped to the total number elected in that year stays even. In fact, I'm compiling data on this offline right now, looking for an even way to present the data. --Biblioworm 03:13, 15 October 2015 (UTC)
• It seems like the two "sides" here have extremely different ideas of what "unreasonably high" is. shoy (reactions) 12:33, 15 October 2015 (UTC)
• @Davidwr: nails it: every editor has their own standards. Often that standard is, "did this candidate piss me off personally?" Another standard is, "what is the position of others that I can adopt without looking into the situation myself and doing my own analysis? I'm not saying this is entirely wrong, but the reality is that RfA cannot survive with no changes at all. As it sits, any admin who has been on-wiki will, almost inevitably, have things in their "past" that can be targeted and then dogpiled. Some sort of decision must be made as to what is allowable discussion of a past record and what is impermissible trolling, dogpiling and character assassination by the disgruntled, some of whom had a WP:BOOMERANG hit them and are out for revenge?(Full disclosure: I had an unsuccessful RfA) Montanabw^(talk) 01:22, 17 October 2015 (UTC)
• Is there any evidence that lowering the standard will increase the number of suitable admins? Many users with ≥10,000 manual edits and reasonable tenure pass RfA easily; there were two 2015 admin promotions by editors with less than that (~9,500 manual edits including old blocked sock; 7,000 edits at RfA and 6,000 edits). Esquivalience ^t 23:10, 17 October 2015 (UTC)
• I, personally, don't recall ever looking at a valid oppose at an RFA and thinking "this standard is far too high". No one has unreasonable standards; many have reasonable standards that, when taken as a whole, become unlikely. So for editor A to expect X is reasonable, and for editor B to expect Y is reasonable, but X and Y may be in vastly different areas, so few editors will have both. Howicus (Did I mess up?) 23:22, 17 October 2015 (UTC)

B: Undefined standards

Participants can support or oppose candidates using any criteria they like, some of which are not based in policy but rather personal opinion. This system is not used in other areas of Wikipedia. Arguments for keeping/deleting an article at AfD, for instance, are given weight according to their basis in policy

Support B

1. Support – minimum standards, either in terms of edit count, or time of active editing, would be an improvement to the process to my mind (as it would mostly eliminate WP:NOTNOW candidacies). --IJBall (contribs • talk) 02:20, 15 October 2015 (UTC)
2. What I think is a problem is that expectations keep shifting and its difficult for a prospective candidate to gauge whether their face fits at the time they are planning a run. I agree opinion is good but we should offer more consistency. Spartaz ^Humbug! 08:18, 15 October 2015 (UTC)
3. Support. Agree with Spartaz, above. More specificity and increased standardization and increased uniformity would be a great thing to provide ease of mind to both our candidates and the community. It would also set certain standards for the future. — Cirt (talk) 08:54, 15 October 2015 (UTC)
4. This is the heart of the problem. We don't have this issue with other rights such as Rollback where we have agreed a criteria. Frequently RFAs degenerate into discussions not about whether the candidate would be a good admin, but whether admins need to have x months tenure, an FA or be an adult. If this was a job interview you would have some things set out in the job ad. You'd know if the candidate had to have a clean driving license or other qualification and you could then concentrate on other issues. ϢereSpielChequers 13:01, 15 October 2015 (UTC)
5. Support...sort of. If ATA is the model, I'm not super hopeful for a resolution, since I don't really want the oppose section of an RfA to be filled with meta comments about ATARFA. The alternative is to have roving clerks/crats remove opposes based on this, but I can't see that ending in any other way than a huge shitstorm. Protonk (talk) 13:40, 15 October 2015 (UTC)
6. The community needs to decide what it wants in an admin and set those standards. Right now a minority can force their view of what admin should be even though the community at large rejects the idea. Arguments that goes contrary to the expectations of the community should be given less weight like everywhere else one Wikipedia, otherwise we are picking admins by voting, not by consensus. I disagree with those that say this would be hard to define, I am sure for example we could come to a clear consensus of the age issue. If a topic is not covered at all then it is naturally left to opinion, if it is divisive then we can have a nice talk about it and amend our expectations. HighInBC 14:39, 15 October 2015 (UTC)
7. I think this would certainly help. Comparable processes involve judging specific situations against generally agreed upon policies and guidelines. Without that people are free to make up whatever standards they like, however unreasonable. This would reduce the number of such comments at RfA and give bureaucrats clear authority to discount them. Hut 8.5 20:08, 15 October 2015 (UTC)
8. Support minimum standards, to cut down on WP:NOTNOW nominations, and to highlight that the standards are not impossible. ---- Patar knight - ^chat/_{contributions} 02:13, 16 October 2015 (UTC)
9. Support: This is the only one of the multiple “contentions” that this RFC seeks to evaluate that I will address because I believe it is the root of all the ills of RFA. The other contentions are merely symptoms of a lack of agreed upon standards (no matter how objective or subjective, no matter how rigorous or liberal) for selecting or not selecting an editor for Admin tools. Until there is an agreed upon set of standards, all the other things perceived to be wrong about RFA will continue and there will be no solution, no matter how hard we try, to those ills.

   Freedom of men under government is to have a standing rule to live by, common to every one of that society, and made by the legislative power vested in it; a liberty to follow my own will in all things, when the rule prescribes not, and not to be subject to the inconstant, unknown, arbitrary will of another man.

   — John Locke

   If the ultimate qualification to become an admin is that the community has confidence it can trust the editor with Admin tools, the community must decide what are reliable indicators (objective and subjective) that engender that confidence and trust?; and hold all admin candidates to the same standards. As John Locke reminds us, arbitrary standards and norms are contrary to the health of any enterprise.

   Apart from the readily apparent ills of RFA that a lack of standards causes, it also causes editors who aspire to be admins to possibly behave in ways or venture into elements of WP and make big mistakes which are eventually held against them in RFAs. My contention is that those kinds of aspirational mistakes might never happen if we had a consistently applied set of standards that editors must meet to be selected as an admin. I would relish to opportunity to participate in a standards discussion about this. --Mike Cline (talk) 14:52, 16 October 2015 (UTC)

10. Support: Clear standards will benefit both candidates and !voters. Also per Mike Cline Montanabw^(talk) 01:27, 17 October 2015 (UTC)

Oppose B

1. If there's any place where Wikipedia should allow for "opinion" it's this. And this question is problematic: you can't say "I support candidate X because they have reliable sources". You can say "I support candidate X because they seem to have the right temperament for the job", and that's the kind of thing one should say. If RfA turns into a rubric...well, that will be the end of Wikipedia. Drmies (talk) 00:46, 15 October 2015 (UTC)
2. Current policy WP:Consensus ("try to persuade others, using reasons based in policy, sources, and common sense"), WP:Administrators ("Any editor can comment on a [RfA] request, and each editor will assess their confidence in a particular candidate's readiness in their own way.") is extremely broad in considering individualized factors gauging judgement of "trust" and "common sense" -- so, are the instructions at WP:RfA. By measure of policy then, the support arguments for this proposition must fail. Alanscottwalker (talk) 01:34, 15 October 2015 (UTC)
3. I see no problem with each editor determining for themselves what's important in an admin. For one thing, it allows for the possibility of more diversity in elected admins, which would not be the case if a single standard were imposed. BMK (talk) 02:16, 15 October 2015 (UTC)
4. This is a consensus-driven project, and applies to every corner of the wiki. Adminship accounts for a suite of tools and responsibility that I don't think can be put into a definitive set of standards, and requires broader input. If you want a rough guideline you can refer to WP:RFAADVICE, which I fear many candidates fail to review — MusikAnimal ^talk 06:08, 15 October 2015 (UTC)
5. Editors need to apply their own life experiences and common sense to the process. There should be no arbitrary "standards." Coretheapple (talk) 15:23, 15 October 2015 (UTC)
6. Common sense should apply - Personal opinions are IMHO fine if they're relevant to the candidate but Supporting this would only see stupid reasons!, –Davey2010^Talk 00:36, 16 October 2015 (UTC)
7. The question everyone must answer for every RFA is "Should this user be trusted with the tools?" Everyone gets to their answer a different way, and that's fine. "Is this person trustworthy" does not lend itself to rigorous criteria that applies and is conclusive in every circumstance. Townlake (talk) 02:00, 16 October 2015 (UTC)
8. If we could define an objective standard for RfA, we wouldn't even need RfA. The problem is: it's too hard to objectively judge a candidate under a defined standard; it may lead to wikilawyering; and community consensus isn't adequately represented. Esquivalience ^t 20:10, 16 October 2015 (UTC)
9. WP:BURO is a core part of Wikipedia—there can be no set of rules to define the features needed for a candidate to be successful. For one thing, if such rules were enforced, the hat-collectors would go crazy ticking all the boxes without showing any clue about how things actually work. A set of rules could be developed, but they would be far too generic to be useful ("candidate must have demonstrated they have a clue" would be near the top of my list). Are there any known examples of failed RfAs where this proposal would have helped? There are probably many such examples where a candidate would have been successful with this proposal, but who is to say that would have been a desirable outcome? Johnuniq (talk) 09:31, 17 October 2015 (UTC)
10. I'm confused by the statement (what is support supposed to mean? any standard is OK? there must be a standard? no standards?), but I'll come here with the notion that there should not be a uniform policy decision. A significant portion of my judgment is about the candidate's skill, personality, and judgment. Some of that is objective, but a lot is subjective. A candidate might be exceedingly polite, but may not let something go. Another might be polite most of the time but blow up once in awhile. Are those enough to oppose? Possibly. Sometimes the words are mostly right, but something just does not seem right. Glrx (talk) 17:57, 17 October 2015 (UTC)
11. Oppose This is begging the question, as community consensus determines standards, and that's how it ought to be. RO^(talk)
12. Oppose. Everyking (talk) 14:17, 18 October 2015 (UTC)
13. Oppose - you can't compare content issues to user issues in that way. All content issues have core policies where each voter simply has to express his/her opinion on how the policies apply to the article in question; with user issues, the question is the level of trust the community members have with the user - and these can't be reduced to a list of policies. עוד מישהו Od Mishehu 18:54, 18 October 2015 (UTC)
14. Oppose - Community consensus is a key part of Wikipedia, without it many policies that we have wouldn't even be in place. - Knowledgekid87 (talk) 21:38, 18 October 2015 (UTC)

Comments on B

• There will always be cases of relevant non-policy reasons why someone would support ("has been an admin on the Commons and French for years and was elected a Steward two years ago, no history of problems on those other projects") or oppose ("Met him at Wikimania last year, he was supposed to give a talk but bailed with a lame excuse, plus he got drunk and got a DWI, it was in the local papers[citation goes here]. Therefore not trustworthy.[link to candidate's own publication of his real-life identity goes here so there are no "outing" issues]"). So, even if I were to generally support this I cannot absolutely support it as written. davidwr/(talk)/(contribs) 02:39, 15 October 2015 (UTC)
• @Biblioworm: I actually cannot tell whether this section is intended to be an affirmation of current fact or of what the goal should be. Once I can tell whether I should be supporting or opposing, I will respond there with the following statement: Any argument that is not primarily policy-based should have little to no value; that is all that adminship should represent. —⁠烏⁠Γ ^(kaw), 05:21, 15 October 2015 (UTC)
• Instead of community standards, why don't we encourage individual editors to state their personal standards? This would let candidates get more of an idea of what RFA participants are looking for before they take the plunge. Howicus (Did I mess up?) 23:22, 17 October 2015 (UTC)

B.1: Standards or Criteria - Potential Development

The community requests that interested Users create a WP:Essay at Wikipedia:RfA standards and criteria laying out standards and criteria for RfA participants consideration. Possible future action may then include, in no particular order:

• • Normal editing consensus.
  • Possible borrowing or repurposing of text from other pages (including, for example an individual user's written criteria). It is also suggested that the essay readily cite to Wikipedia policy and guidelines.
  • WP:Dispute resolution with respect to content of the essay.
  • Seeing if people actually find it useful, perhaps by looking at links to the essay in RfAs or elsewhere.
  • Gaining consensus to include appropriate reference to the essay at the WP:RfA page.
  • Gaining consensus to promote the essay to guideline or policy (or guideline, then policy).

Support B.1.

• I can support further discussion and the B.1. path is a Wikipedia way to do that (I cannot support the conclusory B., however.) Alanscottwalker (talk) 18:16, 18 October 2015 (UTC)

C: Hostile environment

The environment of RfA is hostile toward candidates and is discouraging for those who are contemplating a candidacy. Incivility, lack of WP:AGF, and excessive (or intentionally deceiving) interrogation are too common. Furthermore, candidates are unable to defend themselves and will garner opposition whenever they attempt to do this.

Support C

1. Support The general environment at RfA seems to be "assume bad faith" rather than "assume good faith". (The latter is a community-approved guideline, for the record.) This is shown by the skeptical, excessive asking of questions and the tendency to oppose for the slightest issue. --Biblioworm 00:20, 15 October 2015 (UTC)
2. Support While there are certainly many times voters will cast a !vote to support/oppose per Editor X, the RFA environment is nonetheless hostile. RFA regulars may view that as fairly standard but at least, in my eyes, some RFA's descend into something that is just this side of a brawl. Blackmane (talk) 02:07, 15 October 2015 (UTC)
3. Support – I also find it to be, on balance, an "assume bad faith" environment among far too many of the participants. --IJBall (contribs • talk) 02:24, 15 October 2015 (UTC)
4. Support This is the biggest fix that I think is needed; the questions and votes too many times do not assess the suitability of the person for the job, but are veiled personal attacks and attempts to play "gotcha". I don't know yet what to do about it, but to say that the general tenor of RfA is not "lets find suitable candidates for granting admin tasks to", it's "Someone wants to be an admin, that means there must be something wrong with them. Expose them now!" It has to stop if we're going to keep the admin corps going. --Jayron32 02:48, 15 October 2015 (UTC)
5. We are here to build the encyclopedia. One good faith contributor means a lot. And an experienced editor, who is ready for an RfA, means more. Even if he's not ready for the mop, hostility or incivility or ABF is not really a way to address them. Facing all those nonsense, he'll retire or loose interest in contributing. Congrats community, we've lost one valuable asset. —☮JAaron95 ^Talk 04:51, 15 October 2015 (UTC)
6. Support. This is a corner of the site that I have stayed away from participation in simply because of all of the toxicity reported, and though I've been silently watching, I regret not arriving sooner. If there's only one thing that needs to change, it's the desire to sharply oppose for small things, and I additionally agree with the proposal's words in regard to self-defense. —⁠烏⁠Γ ^(kaw), 05:33, 15 October 2015 (UTC)
7. Support. We should better enforce civility at RFA. It's better than it used to be, but that doesn't mean that we should allow anything to go. I'm not talking about trick questions or assuming bad faith of the candidate. I don't really care about that. I mean passive-aggressive hostility, outright personal attacks, and aggressive badgering. You can ask as many bad-faith questions as you like, but remain civil while doing so, that's all. NinjaRobotPirate (talk) 06:02, 15 October 2015 (UTC)
8. Partial support It can be hostile, very hostile, but that's expected anywhere there's community-wide discussion. Here it's just a matter that we have a sole individual at the center of it, who has to take a beating if the discussion goes sour. It probably wouldn't hurt to have some additional ground rules on conduct at RfA, or rather we just better exercise the existing fundamental policy on civility. However I think the ability to see past the imminent nonsense and have the strength to deal with it is yet another expectation or valued asset of an aspiring admin – this prudence shared with the bureaucrats who should discount frivolous opposition. Genuine opposition doesn't collect on it's own, though, there's reason for it, and if the candidate is well-suited there should be plenty of support to make up for it in both the eyes of the community and the closing 'crat — MusikAnimal ^talk 06:33, 15 October 2015 (UTC)
9. Support, strongly, per Biblioworm and NinjaRobotPirate, above. Unfortunately, RFA can be a poisonous, toxic environment. Increased standards for civility and professionalism that are standards which would be uniformly applied to increase standardization — would go a long way towards ameliorating this significant problem. — Cirt (talk) 08:58, 15 October 2015 (UTC)
10. Jayron32 says it well. —Kusma (t·c) 09:56, 15 October 2015 (UTC)
11. The most effective way to derail an RFA is to give a clear civil oppose with two or more reasons supported by recent difs, showing why the candidate is not yet ready. There are some who seem to believe that adding invective and overstating their case makes it stronger, at worst they then ramp up the invective instead of taking the smarter route of making their oppose more focussed on reasons why the candidate is not yet ready. I've seen at least one RFA sail through because the opposer was so incivil they were counter productive. ϢereSpielChequers 13:12, 15 October 2015 (UTC)
12. Support, RfA should not be a place where editors take their hatches to. Fix the hostile environment will allow more candidates to be willing to come forward. - Mailer Diablo 13:18, 15 October 2015 (UTC)
13. RFA can be hostile - Sometimes RFAs can go well and at times they can get completely derailed, Not sure how but I too a point think this perhaps needs fixing. –Davey2010^Talk 00:39, 16 October 2015 (UTC)
14. The bureaucrats foster and encourage a hostile environment for close-call RFAs. Townlake (talk) 02:02, 16 October 2015 (UTC)
15. Support, going through an RfA is frequently described as running a gauntlet, and very good users are put off by the hostile environment. ---- Patar knight - ^chat/_{contributions} 02:15, 16 October 2015 (UTC)
16. I support this as a statement because it is one of the two reasons why I launched WP:RFA2011. It's actually reundant as a statement though because it's something we already know and it's the main reason why all candidates (except the hardest nosed and those who just cannot fail to pass) are staying away. For some odd reason which no one has ever ventured to explain, RfA has traditionally been the one accepted venue where editors and admins alike can be as spiteful, vindictive, and downright insulting as they like with total impunity. I don't know why this is either, but if progress is going to be made, we have to find a way to put an end to it. Probably by either potty training the participants, or showing them the door. Kudpung กุดผึ้ง (talk) 17:05, 16 October 2015 (UTC)
17. Support: No one in their right mind would want to go through what I did. (raising the question if anyone accepting a RfA nom in the current climate is in their right mind, but I digress...) This per my own experience at RfA which, my personal strengths and weaknesses aside, included doxxing, on and off-wiki canvassing, stalking, false accusations, character assassination, raising issues related to newbie mistakes many years in the past, and so on. Relevant mistakes, particularly when recent and supportable by on-wiki diffs, fair game. The rest, only for the truly iron-spined, and even then, it's an open question if the heat tempers the metal or warps it... Montanabw^(talk) 01:33, 17 October 2015 (UTC)
18. Some RfA candidates are universally acclaimed, but for others, insults, chasing, and incivility are the status quo. Montanabw's RfA is an excellent case-in-point; although I had to oppose due to valid concerns, I considered weakly supporting just for enduring the seven days of angry mobbing. Esquivalience ^t 22:57, 17 October 2015 (UTC)
19. Support. Everyking (talk) 14:20, 18 October 2015 (UTC)

Oppose C

1. Sorry Biblioworm, but the "tendency to oppose for the slightest issue" is more than balanced by the tendency to "support per editor X". I should know; my own RfA was full of those (and I thank you all very much for them). That editors Y or Z got a hard time at RfA may well mean that they deserved to get a hard time at RfA. Drmies (talk) 00:40, 15 October 2015 (UTC)
2. Current policy WP:Consensus ("try to persuade others, using reasons based in policy, sources, and common sense"), WP:Administrators ("Any editor can comment on a [RfA] request, and each editor will assess their confidence in a particular candidate's readiness in their own way.") is extremely broad in considering individualized factors gauging judgement of "trust" and "common sense" -- so are the instructions at WP:RfA. By measure of policy then, the support arguments for this proposition must fail. AGF cannot require someone to assume a particular person should be an administrator - moreover, the framing of this proposition violates the neutrality requirement of WP:RfC - in fact the proposition, itself, generally assumes bad faith by those who pose questions. Alanscottwalker (talk) 01:34, 15 October 2015 (UTC)
3. In actuality, my experience is that the atmosphere is not hostile to the candidate, per se, it's hostile between those voting "support" and those voting "oppose". That's one of the primary things which makes the process so unpleasant. BMK (talk) 02:18, 15 October 2015 (UTC)
4. Even if it were the case, its fair to say that being an Admin can be very challenging and we need people doing the job who are robust enough to deal with the pressure that even the most routine admin actions can engender. I personally think the level if unpleasantness depends on the level of controversy around the candidature and the more marginal the candidate the harder the ride. Whether that is a good thing I'm not sure but it does serve to test the level of resilience. Spartaz ^Humbug! 08:15, 15 October 2015 (UTC)
5. It's the opposition that gets too much heat, in my experience. This can be quite intimidating and this risks distorting the process. Notice that, when you have a secret ballot, as in the arbcom elections, you tend to get a higher level of opposition. Andrew D. (talk) 11:58, 15 October 2015 (UTC)
6. There is a "hostile environment" for editors who probably shouldn't be up for RfA in the first place. I've seen many slide right by. Coretheapple (talk) 15:25, 15 October 2015 (UTC)
7. There is no "hostile environment". If there were, all RfAs would be fought over and lead to lots of discussion, but many don't. See Wikipedia:Requests for adminship/Ian.thomson (139/9/3), Wikipedia:Requests for adminship/Opabinia regalis 2 (118/22/5), Wikipedia:Requests for adminship/Abecedare 2 (119/0/0) etc. Especially in the last one, which was unopposed, where is the hostile environment? Hostility ensues, for example, when blatantly unsuitable candidates appear, and get support !votes from their cronies, and then the community makes an heroic effort to avoid an exponential increase of future drama... Kraxler (talk) 11:17, 16 October 2015 (UTC)
8. An admin must be able to engage with people who are in a heated battle without emotional stress. If people say nasty things, the solution is to respond to the core issue and ignore any over-excited rhetoric—exactly what an admin would have to do. Johnuniq (talk) 09:39, 17 October 2015 (UTC)
9. A difficult RfA is the first test to see if an admin can handle the stress of dealing with difficult editors and difficult situations. If RfA is too tough for the candidate, that candidate is a unsuitable. RO^(talk) 18:00, 17 October 2015 (UTC)
10. Yes, it can be rough and several recent ones have been toxic: lots of bickering has been moved to the talk page. Lots of heat has been directed at the !voters rather than the candidate. Overall, I expect candidates to cope well with conflict. Glrx (talk) 18:14, 17 October 2015 (UTC)
11. If someone has skeletons in their closet, it isn't assuming bad faith to let others know about it. As an admin you have to take responsible for your past mistakes, say what you learned from them if asked, and move forward. I agree with the above as well, that as an admin you have to be able to stay calm through a heated discussion. - Knowledgekid87 (talk) 21:44, 18 October 2015 (UTC)

Comments on C

• While I acknowledge that this is true, I also acknowledge that it can be necessary to a degree. A few years ago an RfA was nearly passing until about day 5 or 6 when someone dug up some policy-related "dirt" which rightfully caused the nomination to tank. By "rightfully" I mean if the info had been stated up front the person would probably never have crossed 50% much less 75%. I don't remember the details but the revelation itself - an the feeling that participants had been duped - had inherent elements of hostility in it. If "hostility at RfA" were banned, such an editor may be promoted because late-arriving RfA participants who know there is relevant "dirt" to be found may be afraid to bring it up. Having said that, we should expect everyone to behave as politely as reasonably possible under the circumstances with allowances for human emotion to intrude a bit in certain circumstances such as the one from a few years ago (sorry I can't remember the particular RfA, and it wouldn't surprise me if such things have happened more than once). davidwr/(talk)/(contribs) 02:46, 15 October 2015 (UTC)

  This kind of thing is a common problem in RfA-related commentary. "Something bad happened this one time, I don't really remember exactly what, but it was definitely bad, and we should definitely make sure something like that doesn't happen again! Or something." I appreciate this is just a thinking-out-loud comment here, but really. Everything on this project is logged. Evidence-free speculation shouldn't be necessary. If it's not worth the time to search for what you're remembering to refresh your memory, is it worth the time to write the comment? Opabinia regalis (talk) 22:15, 15 October 2015 (UTC)

• I'm not sure that "hostile environment" is even correct. Some admin candidates are challenged more than others, but that's because their qualifications are questioned and there is concern. Coretheapple (talk) 18:24, 15 October 2015 (UTC)
• I make these comments as a matter of opinion or different perspective, not for the purpose of dispute or argument. I am not sure how the "tendency to oppose for the slightest issue" is more than balanced by the tendency to "support per editor X" unless editor X is supporting for some minor reason or slight issue or the later supporter is simply relying on the previous supporting !voter or !voters, due to reputation, friendship or whatever, and not on their reasons. If one or a few editors write comprehensive, well-written comments and give good reasons, is it incumbent on later !voters to write these out again or can they simply say: editors X, Y and Z express the same position and give the same reasons as I have, or in shorthand, per editor X, without their vote being questioned in some way? Oppose !votes also are made per another earlier !voter's reasons. In the final analysis, this point may not be important with respect to the value of this proposition or overall, but I think it is incorrect to suggest that !votes should be discounted, whether support or oppose, if they are per previous !voters' well-written and persuasive comments. And I do not think that !votes per another editor's position necessarily balance !votes on the other side that are based on a narrow issue - especially when comparing oppose !votes to support !votes when oppose !votes are worth 3 to 4 times the value of support !votes. Perhaps an affirmation of the later !voter having given an independent look or a statement that the previous expressions cover the later !voter's position would help the credibility of such !votes per earlier ones, if that is necessary. Yet, the need for such statements would not seem to be something that could become well-known, much less required. This may be a minor point in the overall scheme of things but I would not like to think that the original comment on this point would be taken as a given without some further discussion. Donner60 (talk) 02:30, 16 October 2015 (UTC)
• I should note that Kraxler's examples conveniently glide over an important fact. This fact being, two of examples he chose are examples of successful RfAs from very highly experienced (both tenure and edit count) candidates. The other, although she admittedly had a much lower edit count than average, got more objections as noted in the final tally. But, she also had previous experience as an admin, so I'm sure that worked in her favor. However, we're talking about candidates who didn't have it so easy, and of these there are many. In fact, the whole point of discussion is that it shouldn't be only near-perfect candidates who have it so easy. There are many candidates who might have been good admins but failed and had a very bad experience at RfA, and we see candidates fail much more often than not. It should be a civil, productive experience for every candidate, not just the candidates we decide are "good" with our unreasonable expectations. In fact, according to policy, an admin candidate simply should have been active for at least a few months and have shown that they respect and understand the policies and practices of Wikipedia. --Biblioworm 15:45, 16 October 2015 (UTC)

• This comment above If "hostility at RfA" were banned … (and similar opposes), the implication of which is that we should tolerate hostility at RFA, brought this famous America orator’s 1978 speech to mind:

But you can't hold a whole fraternity responsible for the behavior of a few, sick twisted individuals. For if you do, then shouldn't we blame the whole fraternity system? And if the whole fraternity system is guilty, then isn't this an indictment of our educational institutions in general? I put it to you, Greg - isn't this an indictment of our entire American society? Well, you can do whatever you want to us, but we're not going to sit here and listen to you badmouth the United States of America. Gentlemen!

— Otter, Animal House, 1978

If we are willing to tolerate "hostility" or uncivil behavior in any form in RFA discussions regardless of target or topic, then we can hardly expect the community at large to behave differently. In my dead serious view any tolerance of "hostile", uncivil discussion in RFA is contrary to Wikipedia:Civility, one of the WP:FIVE --Mike Cline (talk) 15:00, 17 October 2015 (UTC)

D: More participants

RfAs should be advertised more widely than they currently are. Different !voters, rather than constantly the same ones, would result in a wider and fresher range of opinions. This would also make RfAs more representative of the community's opinion, rather than a small subset of it.

Support D

1. Support Although I don't think this will fully "fix" the process, it would be good to have a wider spectrum of opinions from the wider community. --Biblioworm 00:20, 15 October 2015 (UTC)
2. I wish we had a quick and easy bulletin board where we could see this stuff--this, and major changes in the interface, ArbCom elections, what not--and not those irritating pop-up kinds of things. Drmies (talk) 00:42, 15 October 2015 (UTC)
3. Support with the obvious reservations about making things too open, and possibly find !voters drawn in from some other site to support of oppose a candidate. John Carter (talk) 00:48, 15 October 2015 (UTC)
4. Strong support More often than not, the voters tend to be those who are familiar to regulars on the admin boards. The consistency with which some names appear at RFA to support/oppose certainly give the impression of a cabal promoting their friends, even if this is not the intention. Blackmane (talk) 02:10, 15 October 2015 (UTC)
5. There's never going to be anything wrong with getting as many voters out to participate as possible, although I will say that I'm constantly surprised that many of the names I see voting on RfAs are of editors I've never heard of before. BMK (talk) 02:21, 15 October 2015 (UTC)
6. Support RfA has become a walled garden of regulars who have axes to grind with adminship in general, and we need fresh voices to fix the system. Not sure how to do that, but having more people invovled would certainly help a lot! --Jayron32 02:50, 15 October 2015 (UTC)
7. Strong support - I would go so far as to have random messages go out to experienced users, such as "one out of every 100 editors with over 3 months of editing and 1000 edits total and 10 edits in the last 90 days will get a message about a current RfA or RfB candidate, assuming there are any RfA/RfB's in progress," with a method to turn this off in the user's preferences. davidwr/(talk)/(contribs) 02:53, 15 October 2015 (UTC)
8. Support - Better outreach and advertisement with the RfA process would do many good things (as stated in the initial proposal). It would encourage a wider and more diverse range of voters that would provide new/different opinions and insight. An even more important reason to support expanding outreach and encourage new/more !voters is because it would help "shift" the norm that the RfA is comfortably used to sticking by - something that usually does not occur when the same long-term participants, and only the same long-term participants, cast !votes. All you need to do is simply look at the last 100 RfA's, as well as the many attempts to change the system in RfC. A majority of RfA's close with (generally) the same reasons and with many editors mentioning how badly it's broken, and when a well-thought-out RfC is started that would change things with RfA, it fails... every time. I believe that expanding outreach and encouraging more editors to participate and vote in the RfA process is a solution that would begin to break this entire cycle. If anything, it's a perfect place to start and we really have nothing to lose. ~Oshwah~ ^{(talk) (contribs)} 03:35, 15 October 2015 (UTC)
9. Support I think that getting more opinions is generally a good idea, and I don't see any harm in better advertising RFAs. Spirit of Eagle (talk) 04:21, 15 October 2015 (UTC)
10. Support, with one suggestion being to add RfA as a category of the feedback request service. —⁠烏⁠Γ ^(kaw), 05:36, 15 October 2015 (UTC)
11. Support - "New blood" in these kind of important discussions on Wikipedia is almost never a bad idea IMO. Guy1890 (talk) 05:57, 15 October 2015 (UTC)
12. Support The more the merrier. We could consider a watchlist notice, or for starters transclude the RfX report on more visible pages. — MusikAnimal ^talk 06:38, 15 October 2015 (UTC)
13. Support, strongly, per Bibliworm. This is an excellent idea, for excellent reasons as set out, above. — Cirt (talk) 08:59, 15 October 2015 (UTC)
14. Strong support, in international comparison our participation rates are embarrassingly low. If at least half of active admins could show up at RfA, that might help. —Kusma (t·c) 09:59, 15 October 2015 (UTC)
15. Support I agree, getting more opinions by advertisements will be a net profit. Jim Carter 12:34, 15 October 2015 (UTC)
16. More community input means a more diverse opinions. It will bring up great qualities of a candidate which other never know and vice versa. Regards—☮JAaron95 ^Talk 13:44, 15 October 2015 (UTC)
17. Support - It makes sense. More participation means more varied opinions. This has many plus points and would a net positive change. Yash! 14:12, 15 October 2015 (UTC)
18. Yes, there needs to be a mechanism to inform people of pending RfAs. On several instances I've learned of RfAs that just ended, purely by happenstance, and was sorry I hadn't had a chance to participate (mainly to vote favorably, by the way). Coretheapple (talk) 18:25, 15 October 2015 (UTC)
19. Advertising this in a larger scale should hopefully bring in new editors so thus you'd have new & varied comments - Sounds a great idea. –Davey2010^Talk 00:43, 16 October 2015 (UTC)
20. Support I echo Coretheapple and Davey2010 (above). I usually find out after the fact that editors I know have run the gauntlet, wishing I'd voted them up. I'd welcome a notification system for users with a certain number of edits, or who have shared editing history and/or interaction with the admin candidate (as discussed elsewhere). Notification has the potential to both enlarge and democratize RfA culture; it might help people understand what admins do, improve their own behavior, and encourage more to consider doing the job. Vesuvius Dogg (talk) 01:04, 16 October 2015 (UTC)
21. Each RFA should be linked in the Centralized Discussion box. This process absolutely needs more eyes on it. Townlake (talk) 02:03, 16 October 2015 (UTC)
22. Support. Not everyone checks WP:RFA, so other ways of spreading the message should be considered (CENT, Signpost, a notification system, etc.) ---- Patar knight - ^chat/_{contributions} 02:17, 16 October 2015 (UTC)
23. Support I strongly feel that if fresh eyes come to RfA they will quickly settle one way or another many of the perrenial disputes the usual crowd(me included) keep rehashing. Having a greater cross section of Wikipedia will give us fresh input on the community's view of administrators. It will reduce the amount of power minority views have to derail an RfA. Most importantly I hope it will replace the repetitive arguments at RfA discussions with fresh new ones. HighInBC 14:14, 16 October 2015 (UTC)
24. Support per my comment below where I suggest a candidate to self-identify any WikiProjects where they work and thus an automated, neutral message that says that "editor foo, a member/participant of this wikiproject, has an RfA" would be generated. Likewise, others could also note this RfA on any wikiprojects where the candidate had a "history" - thus making it neutral (in my case, pinging WikiProject Horse racing may have generated support, but pinging WikiProject Opera may have generated opposition - arguably, pinging both projects would have been appropriate). Montanabw^(talk) 01:47, 17 October 2015 (UTC)
25. Support Coupled with a "de-weighting" for editors with under 250 total edits under their belts as being unlikely to have had enough administrator contact to have the slightest idea about the process. Historically under 2% of the voters fall in this category (too much time to go back to the beginning though), but in the last case it represented 4 out of the 58 "opposes" which was a significant percentage in that case. (a difference between 67% yea and 73% yea) Collect (talk) 14:10, 17 October 2015 (UTC)
26. Support This is hands down the best idea on this page. Increased participation dilutes block voting that is currently too influential at RfA. We need the whole community to decide RfA, not a small group of regulars. RO^(talk) 18:02, 17 October 2015 (UTC)
27. Support The more opinions that are cast upon a candidate the better as there is more to take away from the discussion. - Knowledgekid87 (talk) 21:46, 18 October 2015 (UTC)

Oppose D

1. RFA is not going to hold widespread appeal for most of the users of Wikipedia. Editors who have the familiarity with issues that matter at RFA are going to be aware of the process, they may just not choose to participate, just as an editor may be aware that there is a process of deletion, but it's not a process that interests them. I don't think RFA participation rates are unreasonably low. Don't overlook that a count of the number of participants only includes the number who have actually commented on a candidate, not the number who have reviewed the RFA, considered the supporters and opposers, and chose not to comment because the RFA was already going in the direction they would have preferred anyway. Neil916 (Talk) 01:58, 15 October 2015 (UTC)
2. People who are interested will watch WP:RFA and those who aren't won't. Advertising is simply going to randomise the expectations even more then they already are and encourage inexperienced users to dive in with out of field thoughts and opinions. There is very litter evidence that involving more random people has ever solved anything. Just look at the reactionary cesspit that is AN and ANI. Spartaz ^Humbug! 08:22, 15 October 2015 (UTC)
3. I thiunk that lack of participation is a result, not cause, of the RFA trouble. When fewer users run at RFA, fewer users will keep an eye on it and vote in the new discussions. עוד מישהו Od Mishehu 19:33, 15 October 2015 (UTC)
4. A short lesson in History: To increase the overall voter base has not led to any significant change in history. When women conquered the right to vote in New York in 1918, politicians were afraid that they may vote for new parties, bad candidates, or women, but actually nothing changed. Although the total number of votes doubled, they voted for the same people as the men used to vote for, and only two women were elected for 1919, one Republican and one Democrat. At some time the voting age was lowered from 21 to 18 in Germany, with the same result: no big change. On the other side, when Jacob D. Fuchsberg (a lawyer without experience as a judge) started to campaign for a seat on the New York Court of Appeals, which had never before been done (the judges were nominated by the parties, and voted for without campaigning) and defeated the party nominees, the seats were made appointive. Apparently the voters are won over by the candidate who spends more money, and makes the funnier TV commercials, not by the more experienced and qualified candidate. Kraxler (talk) 11:35, 16 October 2015 (UTC)
5. The turnout at RfA has never been so high. We don't want extra publicity to attract even more trolls to the process. Kudpung กุดผึ้ง (talk) 17:09, 16 October 2015 (UTC)
6. Quality over quantity of voters. Quality of voting needs to be addressed before quantity. Esquivalience ^t 20:15, 16 October 2015 (UTC)

Comments on D

• Although I support this idea, care should be taken as this invites SPAs, and votes rather than discussions i.e., Supports and Opposes not based on contributions.—☮JAaron95 ^Talk 04:56, 15 October 2015 (UTC)
• @Neil916: If they don't comment, then they shouldn't be called a "participant". —⁠烏⁠Γ ^(kaw), 05:38, 15 October 2015 (UTC)

  @KarasuGamma: That was sort of the point I was trying to make, I probably didn't word it as clearly as I could have. Restating the point: You can't look at a completed RFA, and determine that only 126 users determined the outcome. Many more people would have seen it, may have formed an opinion about the candidate, and did not post a comment about it because it did not add to the discussion, and it was clear that the apparent where the candidacy was headed. This is especially true if a candidacy is heading for defeat, as many experienced contributors don't like to pile on in opposition. Neil916 (Talk) 19:40, 15 October 2015 (UTC)

• I don't know whether to support or oppose: I oppose the header but support the statement underneath it. We need different participants. We do not need more. Roughly 100-200 !voters per candidate is more than enough. — Bilorv_(talk)^(c)(e) 15:33, 15 October 2015 (UTC)
• How about having an automatic watchlist notice every time an RfA or RfB starts? This should be possible using Lua without having to resort to a bot. — Mr. Stradivarius ^{♪ talk ♪} 01:43, 16 October 2015 (UTC)

  Could the automatic watchlist notice wait an hour or two after the RfA has been transcluded to avoid bothering people with obvious NOTNOW RfAs? — Bilorv_(talk)^(c)(e) 14:53, 16 October 2015 (UTC)

  Yep, that shouldn't be too hard. — Mr. Stradivarius ^{♪ talk ♪} 05:14, 17 October 2015 (UTC)

• Regarding SPAs, they find their own way to RfA if their "single purpose" includes that. As for the arguments that we don't need greater numbers of people I agree, but what we do need it not to have the same people every time. Right now there are more than a few people that go to RfA to oppose the process of RfA itself or to push a minority view by opposing at RfA. More people will reduce the power these minority views have to disrupt RfA(no I will not mention names). HighInBC 14:38, 16 October 2015 (UTC)
• Per JAaron95, I would echo the caution about anon IPs and socks. I had a clear anon IP and a SPA try to !vote or comment on my RfA, and I strongly suspect the came from one or two now-banned sockpuppets where I had a role in "busting" them. That said, I also think there needs to be some way to "advertise" an RfA to those who might actually know the candidate's work and be in a position to offer clueful comments. One idea might be for a candidate to self-identify any WikiProjects where they work and thus an automated, neutral message that says that "editor foo, a member/participant of this wikiproject, has an RfA" would be generated. Likewise, others could also note this RfA on any wikiprojects where the candidate had a "history" - thus making it neutral. Montanabw^(talk) 01:44, 17 October 2015 (UTC)

E: Difficult to remove admins

The current process by which we remove admins (namely, WP:ARBCOM) is too long/bureaucratic/difficult. The community should have an easier and more community-based method with which admins can be desysopped.

Support E

1. ...Weak support maybe policy or guidelines could be adjusted in such a way as to make WP:AE or something similar a way to remove admins, probably with a fair number of opinions or !votes on the removal required before action could be taken? John Carter (talk) 00:49, 15 October 2015 (UTC)
2. I think this point can be worded better to illustrate that "RFA is hard to pass because participants know it is so hard to remove a seated admin". I agree with this, and in the current system, it causes participants to be very cautious about who makes it through the process. Neil916 (Talk) 02:01, 15 October 2015 (UTC)
3. Support. I agree. I think that, if bans and editing restrictions can be imposed upon a user by community consensus (such as what you will see occur regularly on ANI), then surely there can be the same process of establishing community consensus to have an administrator desysopped (given that legitimate concern is brought forward with supporting evidence, enough editors are involved, there is enough community participation to comfortably assert that consensus has been reached, etc). Or, maybe an "Request for De-adminship", or something similar? Where filing one against a current admin would require evidence of numerous failed attempts to resolve the matter in question and/or evidence of numerous failed attempts to correct the action(s), and maybe 10 supporting nominations before moving forward. If it does, then there are questions that the admin can answer in order to explain the concerns. If enough support votes are attained to establish consensus then the "RfD" closes as successful and a crat proceeds with desysopping. Just another idea... ~Oshwah~ ^{(talk) (contribs)} 03:20, 15 October 2015 (UTC)
4. Support, though I don't have much to add; my opinions are closely aligned with those of Neil916 and Oshwah here. —⁠烏⁠Γ ^(kaw), 05:46, 15 October 2015 (UTC)
5. Support - There needs to be a more widespread community recall procedure for administrators on Wikipedia. No where near enough current administrators are really open to voluntary recall efforts, and I think that the community would get better general behavior from our administrator corps (and improved confidence in them over time) if we all knew that being an administrator wasn't a "lifetime appointment". Even a few Wikipedia administrators that feel like they are above reproach can cause a lot of damage to this project over time. Guy1890 (talk) 06:05, 15 October 2015 (UTC)
6. Support, with a good idea put forth by John Carter, above. — Cirt (talk) 09:00, 15 October 2015 (UTC)
7. Support without reservation. IMO one of the big causes of RfA problems is the eventual realization by the community that adminship is essentially indefinite. We can talk about how it isn't a big deal or how opposes should be moderated on the basis of this or that, but so long as people understand the structure of the underlying game, they're going to be strongly risk averse. Protonk (talk) 13:42, 15 October 2015 (UTC)
8. Yes, this is a major underlying reason for the reluctance of users to elevate editors: that being an administrator is a roach-motel lifetime position. The process for removal needs to be made more civilized, less drama-ridden for all concerned. Also, within reason, users bringing complaints against admins in such proceedings need to be given "whistleblower" protection and not subject to retaliation unless their complaints are utterly groundless and made in bad faith. Coretheapple (talk) 15:28, 15 October 2015 (UTC)
9. I’ve seen this proposed a lot, and I think that allowing a more streamlined, community centered way to remove admins is a good idea. I think it would also increase the number of successful RFAs, since the community would be more willing to give the mop to editors if they knew they could take it back with relative ease if the editor started to misbehave or failed at his or her duties. I would like to make two caveats though. First, I would like to see some protection put in place to ensure that admins aren’t being desysoped for purely political motives, and that outside groups such as GamerGate aren’t able to hijack the recall system to get revenge against admins who make sound, policy-based decisions that are to the disliking of the outside group. Additionally, I oppose limited terms. I feel that if an admin is doing their job well, then they should be allowed to keep the mop as long as they keep up their good behavior. Spirit of Eagle (talk) 20:39, 15 October 2015 (UTC)
10. A community based board/discussion sounds a brilliant idea - Not everything needs to go to Arbcom. –Davey2010^Talk 00:47, 16 October 2015 (UTC)
11. Support - but it will never fly because a) nobody wants to come up with an idea and launch it at RfC, and b) if they do, the professional anti-everything mob turns up to derail it including the noisy back benchers and those former functionaries who believe that bullying the participants is the best tactic.Kudpung กุดผึ้ง (talk) 15:54, 16 October 2015 (UTC)
12. Support. I've been unsure about this one, but after considering it I think it would be best for the community to have a more direct voice in dealing with admins who need to be dealt with. --Biblioworm 17:05, 16 October 2015 (UTC)
13. Support Taking the mop back should not be more difficult than giving it away. RO^(talk) 18:03, 17 October 2015 (UTC)
14. Support Everyking (talk) 14:23, 18 October 2015 (UTC)

Oppose E

1. RfA is too hard by desysopping is also too hard...? The recent past has seen a number of admins desysopped, and while that usually takes a bit of time, that's the proper way. And if desysopping is so hard, I wonder what one makes of the case of Malik Shabazz, who got his tool within a few hours (or less?) yanked after one tiny little infraction, being baited by racist commentary. Drmies (talk) 01:56, 15 October 2015 (UTC)
2. I'd say this has it 100% backwards. The reason why adminship is too hard to take away is that the community treats it as too precious. If it were already easier to become an admin, we'd have more better admins, and wouldn't feel the need to make removing it so hard. --Jayron32 02:51, 15 October 2015 (UTC)
3. The evidence is that admins who step over the line are having their bit removed - if the Malik Shabazz case shows anything, its that Arbcom are too quick to yank the bit. People who claim abusive admins are inviolate rarely provide the names and evidence to back up their assertions. Spartaz ^Humbug! 12:32, 15 October 2015 (UTC)
4. Oppose: ~97% of administrators who have passed RfA have not been removed from their posts by ArbCom. Unless a body of evidence can be provided showing there is some significant number of admins that should not be admins that have not been de-adminned by ArbCom, then this idea is moot. Everyone likes to think that a community based de-adminship would solve what ails RfA. There's no evidence to suggest it would. Meanwhile, there's significant evidence to suggest it would decimate the standing admin corps and reduce the ability of this project to function. I am not opposed to a community based de-adminship process. I am opposed to attempting to implement one without evidence to support there is a need for one. Taking a poll is not evidence, just opinion. --Hammersoft (talk) 16:20, 15 October 2015 (UTC)
5. Oppose - In theory, there are only 2 options for desysoping in case of misbehavior: community or a selected committee. I believe that in this case, the community would tend to remove the very best sysops (the ones who handle the most controvertial issues), so we need a selected committee - and ARBCOM seems like the best option we have for it. עוד מישהו Od Mishehu 19:18, 15 October 2015 (UTC)
6. Desysopping isn't exactly rare. Five admins have been desysopped for cause this year, and more resigned under a cloud. Nor does it really have much to do with RfA: the promotion rate has taken a nose dive while it hasn't got any harder to desysop people. Most concrete proposals for an alternative desysop process have major flaws. Hut 8.5 20:13, 15 October 2015 (UTC)
7. This is a popular meme with no apparent basis in fact. I'm not aware of recent difficulties removing problematic admins. Townlake (talk) 02:14, 16 October 2015 (UTC)
8. The statement is just wrong. Several admins have been desysopped by motion within a few days, lately. You better withdraw this, or reword the statement. Anyway, any admin who is active in AfD would be mobbed by those who got their articles deleted, if there was some community desysop procedure. There's a reason why in civilized places people who are suspected of breaking the law are tried in a court of law, not by their neighbors. Kraxler (talk) 11:44, 16 October 2015 (UTC)
9. Desysopping has become significantly easier (and more common) over the last ten years, while becoming a sysop has become harder. (Maybe we should make desysopping harder? Would that make becoming a sysop easier again??) —Kusma (t·c) 13:32, 16 October 2015 (UTC)
10. It's not really that hard to remove unsuitable admins; it just may take a while. Esquivalience ^t 20:17, 16 October 2015 (UTC)
11. Weak Oppose per several of the above comments and my own observations that many admins who make serious mistakes voluntarily resign the mop or quit wikipedia altogether. It isn't that difficult to deysop, actually it's easier than getting the mop in the first place these days. My thinking is that if there is any actual problem, just clarify the grounds for a desysop - any weakening of the process is just bait for the disgruntled trolls who love drama. Montanabw^(talk) 01:50, 17 October 2015 (UTC)
12. I'm OK with hard to get the bit and modestly difficult to remove it. My impression is that wheel wars were very common in the past and have become less common in the present. I don't buy into the theory relaxing standards to give more editors the bit. Glrx (talk) 18:20, 17 October 2015 (UTC)
13. Oppose, in the current climate of RFA. A side effect of the demanding nature of RFA is that those who are promoted are *very* good candidates. If getting the mop becomes easier, then we should talk about making taking it away easier. Howicus (Did I mess up?) 23:22, 17 October 2015 (UTC)
14. Weak Oppose I don't think leaving this in the hands of the community is the best idea. For all the bad admin out there, what is stopping a lynching for an admin that might have done something minor? The people voting to desysop an admin would have their votes weighed in by strength of argument right? If so then by who? - Knowledgekid87 (talk) 21:54, 18 October 2015 (UTC)

Comments on E

• My solution to this is for admins to serve limited terms, say 5 years, after which they would go through a process similar to RfA, but where the passing level would be pretty low, say 45%. (This is to compensate for any accumulated grudges.) For the occasional admin who oversteps his bounds, or approaches their work with a bad attitude, knowing that a re-confirmation process is coming up might be a impetus to behave better. Or, admins who just slip by could serve "probationary" terms of 2-3 years, while those who pass with high percentages perhaps serve 7 years before having to be confirmed. The point is that a lifetime appointment does not seem to me to be a good idea. Certainly it protects the admin from unwarranted reprisals, but it also shields them from the displeasure of the community. BMK (talk) 02:28, 15 October 2015 (UTC)
• Endorse limited-term adminship, BMK's idea of retention elections is not the best but it is better than we have now. I do like his idea of shorter terms for those who start off with lower support ("admins who just slip by"). I think another Wikipedia (German?) already has fixed-term admin elections. davidwr/(talk)/(contribs) 02:56, 15 October 2015 (UTC)

  If the problem with RfA is that the hostility of the environment makes it too hard to add new admins to the admin corps, the solution to the problem is to now take the existing corps and subject them to the same unwarranted hostility so they can be allowed to continue responding to RFPP requests? And that's going to fix the core problem? --Jayron32 03:02, 15 October 2015 (UTC)

  If you are referring to those elected prior to any change to a term-limit, we would have to handle "grandfathering in" existing admins in some form or fashion. This may mean leaving them with the "lifetime bit," it may mean leaving them with it as long as they remain active as an editor and/or as an admin, or it may just mean "declaring them elected for a full term" so they won't have to worry about running again for awhile. But that's a discussion for another day. davidwr/(talk)/(contribs) 03:43, 15 October 2015 (UTC)

  Our problem is too few admins, we need to encourage more candidates to run, the last thing we need is to get rid of good experienced admins. Most of our currently active admins have been admins for more than five years, and with RFA so nasty these days, this is a disruptive process that would do a lot of harm for no discernable benefit. ϢereSpielChequers 18:52, 15 October 2015 (UTC)

• Re Hammersoft's concern: I am not surprised that 97% of admins, if that accurate, don't get kicked out. I hope not, nor do I expect that not to be the case. I don't believe that proponents of making the desysop process easier believe that admins ranks are saturated with bad apples. However, all you need are a few to cause real damage and to create a burden for other editors. Right now at RfArb a case is coming up, still active at last look though probably not to be accepted, revolving around an admin who supposedly abused his tools. Such occasions are tough on all concerned, and should be handled in a less drama-ridden process. Coretheapple (talk) 20:23, 15 October 2015 (UTC)

• I've been working on some evidence over the last week. I've been seeing some interesting results. I don't have anything hard at this point, but there seems to be a case to be made that RfA is an extremely poor predictor of later failure of admins that are forced out. This makes me wonder if the inverse, a system to remove admins, would also be a very poor predictor of which admins should be removed. --Hammersoft (talk) 20:31, 15 October 2015 (UTC)

Of course it's a poor predictor, just as it was in the case of Pastor Theo, Wifione, and some other prominent ones whose names I can't mention because they are still with us and for some reason are still allowed to participate at all. Those who loudly insist that all admins are bad are doing more harm than good and most of their complaints are just righteous indignation and they don't have long lock logs for no reason; that's why although there needs to be a fast track system for making admins more accountable, it doesn't always need to end in a removal of the tools and I'm very much averse to the notion that an easier desysoping process should be used as a witch hunt to whittle out more admins under the flimsiest of accusations. I thought that kind of thing went out with McCarthy. Kudpung กุดผึ้ง (talk) 16:29, 16 October 2015 (UTC)

F: Unbundle the tools

Currently, it is too difficult for candidates to pass because they are being trusted with a large toolset. If we unbundled the tools, candidates could simply apply for the particular right(s) which they would personally find most useful and/or have the most non-admin experience in.

Support F

1. Support – Of course. It's inevitable. It's only a question of when. That said, this is tangential to the overall thrust of Biblioworm's RfC, and should probably be dropped as it's a separate topic. --IJBall (contribs • talk) 02:22, 15 October 2015 (UTC)
2. Strong support. Some candidates could easily be trusted with page-delete/undelete but their past behavior (e.g. interaction bans, short temper, etc.) means they probably can't get 70% support to have the "block user" user-right, and vice-versa. davidwr/(talk)/(contribs) 02:59, 15 October 2015 (UTC)
3. Support, but depends on which tools, and what are the standards, for each. — Cirt (talk) 09:01, 15 October 2015 (UTC)
4. Support AS ALWAYS. It's silly to keep tools bundled beyond the minimum necessary for legal issues. block/protect/delete are different classes of tools and can be meted out through different processes. Protonk (talk) 13:44, 15 October 2015 (UTC)
5. Weak support I support unbundling in principle, but I don't think it should be unbundling just for the sake of unbundling. I don't think it would make sense to unbundle something that is almost always used with other admin tools during normal admin duties, but if there are cases where the tool is usually used by itself, then unbundling would be a good option. One example that springs to mind is closing AfD discussions, which normally only requires delete and viewdeleted. Of course, viewdeleted is the right that requires community vetting due to the legal implications for the WMF, but it still may be more attractive for some candidates to run for a deletion-only RfA than to run for the current incarnation of RfA. There may be more scenarios like this that I haven't thought of. — Mr. Stradivarius ^{♪ talk ♪} 15:07, 15 October 2015 (UTC)
6. Yes, I think that if the tools were unbundled it would make the whole process much easier on everybody concerned. An editor who can recognize and block vandals might not be qualified to engage in intricate sockpuppet investigations or rule on complex ANI complaints. Coretheapple (talk) 15:31, 15 October 2015 (UTC)

   You're correct that being competent at vandal blocking doesn't require competence at ANI and sockpuppetry-related issues, but it does require the same technical tools. עוד מישהו Od Mishehu 19:06, 17 October 2015 (UTC)

7. Delete, block and protect. People might think these are roughly the same level of importance, and therefore that they should be given out together. But, as an example, people often oppose RfAs for reasons relating to how someone deals with deletion (i.e. CSD and AfD), when actually the user just wants to deal with vandal fighting (block, protect, and the occasional delete). Yes, some jobs need mixtures of these tools, but then you can post to the relevant noticeboard—that's what happens anyway when the user has none of the three rights. I would be open to a variety of ideas about how these tools could be given out individually (based on an RfA, some kind of mini-RfA, something closer to requesting rollback etc.), but I definitely support the idea of unbundling them somehow. — Bilorv_(talk)^(c)(e) 15:40, 15 October 2015 (UTC)
8. We've had several successful unbundles in the past, and could do more. We need some strong content creators on the DYK queues, and it might probably makes sense to make that a separate right or role similar to the way FA runs. There is a proposal elsewhere on this page to unbundle a very limited delete to everyone. I'd also support unbundling block providing it didn't work on accounts with more than 100 edits. But each unbundling needs to be thought through, not everything can be unbundled ϢereSpielChequers 18:42, 15 October 2015 (UTC)
9. Support Some of the admin rights (suppressredirect, tboverride) have less potential for disruption than rights available to autoconfirmed users or at Requests for permissions, others such as undelete should only be available to users who need them, not just because they are only available bundled together. Peter James (talk) 22:59, 15 October 2015 (UTC)
10. Support: At the very least, there should be a split between the "people" toolset (block, unblock) and the "pages" toolset (protect, delete, etc.). --Carnildo (talk) 23:30, 15 October 2015 (UTC)
11. Support to the extent that it complies with Wikimedia Policy. The admin tool bundle is huge and many admins really just want to specialize in a few areas (e.g. updating sections on the Main Page). ---- Patar knight - ^chat/_{contributions} 02:20, 16 October 2015 (UTC)
12. Support per Bilorv and WereSpielChequers. Carnildo also makes a good point about categorizing the rights. —⁠烏⁠Γ ^(kaw), 02:59, 16 October 2015 (UTC)
13. Support but as any admin who has worked at WP:PERM will know, it will add another shelf to the millinery. Kudpung กุดผึ้ง (talk) 16:36, 16 October 2015 (UTC)
14. Unbundling the tools will reduce the need for admin privileges, but for more controversial actions (e.g. deleting, blocking), confirmation by another user with the same right should be required. Esquivalience ^t 20:20, 16 October 2015 (UTC)
15. Support: There are tools related to wikignoming tasks related to article and project content that would be very nice for trusted content creators to have, particularly regarding article protection, moves, deletion and so on. The other set of tools, and in my own experience the source of the drama, are those with the ability to block users. Split the content from the behavior. I sometimes wish —desperately—that I could protect, delete, undelete or move articles over redirect (and load DYK, ITN and so on). Though trolls are annoying, AIV actually is a pretty fast process for the truly harmful or dangerous users most of the time. I wanted the wikignoming set, could live without the banhammer if I could simply do my content work with fewer hassles. Montanabw^(talk) 01:55, 17 October 2015 (UTC)
16. Support in principle, but it remains to be seen if any specific tools are suitable for unbundling. RO^(talk) 18:05, 17 October 2015 (UTC)
17. Support especially for IP-blocking, seeing deleted pages, and placing articles under semi-protection. Collect (talk) 23:08, 17 October 2015 (UTC)

Oppose F

1. Nope. Any level of trust required to grant one of the tools would be required for all of them. If I trust a user to block someone, I'd also trust them to delete an article. If I don't trust them with the ability to block someone correctly, I also wouldn't trust them to use protection correctly. --Jayron32 02:53, 15 October 2015 (UTC)
2. Admin tools are for Admins. We currently have some of the unbundled tools and that is sufficient. Unbundling all of 'em will lead to more chaos and confusions. Regards—☮JAaron95 ^Talk 05:01, 15 October 2015 (UTC)
3. Oppose I think, we give the admin bits on the basis of competence and trust. A candidate applying for a single tool must have both, as well as a candidate applying for a complete tool set. Simply in the latter case, it is AGFed (based on trust) that the candidate won't use the tools in which he is incompetent haphazardly. Secondly, all the tools are needed in tandem, unbundling will be chaotic as more semi admins would be required to take on an IP hopping vandal, moving and vandalizing pages. --Fauzan^{✆ talk}_{✉ mail} 05:05, 15 October 2015 (UTC)
4. Oppose I think the ones that are worth unbundling have been. Delete, block, protect, etc, if you can be trusted with one of those you can be trusted with the others. We've been through this far too many times anyway to no avail, let's focus on other solutions to improve the RfA process — MusikAnimal ^talk 06:42, 15 October 2015 (UTC)
5. Oppose - if you believe RFA has problems, proposing replacing it with multiple RFAs is, frankly, stupid. WilyD 12:33, 15 October 2015 (UTC)
6. Oppose, vandal fighters need block and protect and delete and viewdeleted to work properly, speedy patrollers need delete and block. RFPP patrollers need block and protect. I can't think of a major admin area where people with access to only one of the three key tools would be all that helpful. —Kusma (t·c) 12:41, 15 October 2015 (UTC)
7. Oppose - One of the main areas we need more admins for is deleting. And with deleting comes undeleting (necessary to allow a user to undo his/her actions), and we run into an issue where the Wikimedia Foundation has said absolutely not. עוד מישהו Od Mishehu 19:25, 15 October 2015 (UTC)
8. Oppose - No need to unbundle - If you wanna delete, block, whatever then become an admin .... –Davey2010^Talk 00:50, 16 October 2015 (UTC)
9. I'd only support this if every current admin also had to re-stand for their police powers, and that ain't happening. Townlake (talk) 02:15, 16 October 2015 (UTC)
10. The main areas of contention seem to be block, delete, protect. Often an admin uses 2 out of three of these on any given case. I think ungrouping these three would be an ineffective use of resources (editors), if someone can delete a recreated page, but then has to go to RFPP to request its protection (for example). I can't think of any other commonly used tools I have as an administrator that would be worth unbundling, not to say there isn't. --kelapstick^(bainuu) 15:14, 16 October 2015 (UTC)
11. Be careful for what you wish. I think this adds complexity without solving the problem. Except for technical admins, I use the same judgments for the tools. If you rate one, then you rate them all. Someone who can close an AfD can also recognize vandals. Glrx (talk) 18:27, 17 October 2015 (UTC)

Comments on F

• @Od Mishehu: regarding his "oppose" of 19:25, 15 October 2015 (UTC) in which he said "And with deleting comes undeleting (necessary to allow a user to undo his/her actions), and we run into an issue where the Wikimedia Foundation has said absolutely not." If I recall, the Foundation said editors could not see deleted material without going through an RfA-like process. This would not preclude an "undo delete" button. Having said that, any "undo delete" tool would have to be very constrained in order to prevent abuse. I don't see the Foundation having a problem with allowing a vetted editor who has not gone through an RfA-like process to delete pages and then being allowed to "blind undelete" them provided the undelete 1) was done by the same editor within a few hours of the deletion and 2)there were no changes to either the page title or the deleted revisions since the deletion. Such a "pseudo-userright" could be implemented via a proxy admin-bot now if the community desired it. davidwr/(talk)/(contribs) 20:57, 15 October 2015 (UTC)
  • How about the following: A page deleter deletes an article under A7 or G11; several hours later, someone asks the user to explain his/her actions; the user may not see it for several more hours (or even a day or 2). How is the user supposed to respond to it? עוד מישהו Od Mishehu 04:58, 16 October 2015 (UTC)
    • Good point. The most expedient way I can think of would be through a special use of WP:Request for undeletion: If the page-deleter can't explain his actions because he forgot the details, either he, or if he doesn't reply within a day or so, the person who objected should go to WP:Request for undeletion. Any administrator there can look at the page and say "bad call, I'll un-delete it for you" or "good call, and here is why I agree." Presumably, the vast majority of such requests will be turned down, and any page-deleter gets too poor of a track record he may find himself losing that privilege. If this became a frequent issue I would expect someone to write a script to assist page-deleters, so they could type in an extended reason why they believed the page met the criteria and record it in a log and/or on the page-creator's talk page. davidwr/(talk)/(contribs) 05:18, 16 October 2015 (UTC)
• oppose Blindly unbundling the entire toolset is just about the worst idea ever for reforming RFA. The tools are indeed a set, they work together. Any experienced admin can tell you that. If all you can do is block, all you will do is block, even if deletion and/or page protection is the right response. A real admin with all the tools can choose which is best for the job and act appropriately. Beeblebrox (talk) 15:34, 16 October 2015 (UTC)

• What your argument conveniently ignores is that an editor that just wants to "Vandal fight" will often garner more than enough opposes to sink their RfA candidicy on the basis of "They only ever work at AIV – they have no XfD experience, no ANI experience, have 0 GA or FA's under their belts, and have only created 3 stub articles." IOW, they'll never be granted the toolset in the current environment because, to use BMK's oft heard argument (i.e. he's not the only one who says it), "They aren't applying for a portion of the toolset, they're applying for the full bit, and [we] won't vote for someone who hasn't shown experience to the full range of tools." It also conveniently ignores the ton of us editors who will never run in an RfA because we'd only want a portion of the toolset, not the whole thing. But, hey – if the current Admin corps is actually convinced that the current system is the only system that should exist, then I guess they really don't need our help after all – get to those backlogs, fellas!! --IJBall (contribs • talk) 16:31, 16 October 2015 (UTC)
  • That many opposers have the wrong approach to adminship is a reason to educate the opposers, not a reason to rip apart tools that belong together. —Kusma (t·c) 20:16, 16 October 2015 (UTC)
• And sometimes an edit filter or checkuser is the right response, and the administrator needs assistance. Similar an editor responding to AIV or UAA reports, or patrolling new pages, can request assistance if deletion, checkuser or other action is needed. Peter James (talk) 23:02, 16 October 2015 (UTC)

G: RfA should be replaced

The current process by which we choose our admins is fundamentally broken. No amount of reform will fix it. We must discontinue it and replace it with a different system.

Support G

1. See User:Application Drafter/Sysop applications draft. Agent 73124 (talk) 15:27, 15 October 2015 (UTC)

Oppose G

1. Oppose I support the concept of RfA, which is namely that the community have a direct voice in electing its admins. I can't think of any other alternative to the basic structure of admin selection that would be more bureaucratic or "elite" (in that the selection of admins is put in the hands of a committee). --Biblioworm 00:20, 15 October 2015 (UTC)
2. Opposeish I think the current system should be replaced, but I'm afraid the replacement would be (on paper) pretty much exactly like RfA is supposed to work on paper. That is, if I were to create a set of rules to select admins, it would look like what RfA's rules look like (for the most part). The problem is that what we need to do is flush the entire current system, and create the same system we have now, minus all of the bitterness and negativity. --Jayron32 02:55, 15 October 2015 (UTC)
3. Oppose I'm with others in that we can focus on making the experience a little calmer. I'm not talking about legitimate opposition perceived as being harsh, but the unbridled negativity some bring to the table that can steer an otherwise promising RfA off course — MusikAnimal ^talk 06:47, 15 October 2015 (UTC)
4. It used to work ten years ago, so it can work, but maybe not with the current Wikipedia community. —Kusma (t·c) 18:31, 15 October 2015 (UTC)
5. RfA isn't particularly unusual: people leave comments supporting or opposing a proposal, with some scope for them to have discussions, and the debate is closed using headcount with some weight given to strength of argument. Most of our other decision making processes work in a similar way without attracting nearly as many complaints. Hut 8.5 20:16, 15 October 2015 (UTC)
6. This is a non-starter. RFA is the worst possible system, except for every other system that's been suggested. Townlake (talk) 02:16, 16 October 2015 (UTC)
7. Oppoe. Absolutely not completely broken. Any system that could replace it will be blighted by the same phenomena unless they can be controlled. And if they can, we don't need to replace the current system. Kudpung กุดผึ้ง (talk) 17:13, 16 October 2015 (UTC)
8. Oppose RFA works, even if it has become a somewhat flawed process. If we replace it, we have no guarantee that the new process won't be even more broken. We should be focusing on reforming the existing system rather than replacing it. Spirit of Eagle (talk) 05:30, 17 October 2015 (UTC)
9. Oppose Without a specific example of what would replace it, this question is too open-ended and malformed. RO^(talk) 18:06, 17 October 2015 (UTC)
10. Oppose - The alternatives to a community-based process for selecting admins would be either an automatic assignment (any user who meets certain criteria); an administrator selection committee (how would this committee be selected? We run into the same issues here; alternatively, we let ArbCom do it - but I think that the community is generally better at deciding these things, and ArbCom has more important jobs to do); or let the administrator group be a self-selecting group (such as the Mediation Committee). I think that the best choice here is a community-based process - that is, a reformed version of RFA. עוד מישהו Od Mishehu 18:16, 17 October 2015 (UTC)
11. Difficult, selective, or hostile does not imply broken. Are the results bad? Some get promoted and then quickly desysoped. Those are failures, but I don't think there have been a lot of them. Those who don't get promoted usually have more than 25% opposing. Some of them should have been promoted, but which ones? There's a difference between disappointment in a result and something being broken. Glrx (talk) 18:36, 17 October 2015 (UTC)

Comments on G

• There are a few user-rights, such as seeing deleted pages, which the Foundation will require and RfA-like process, so for editors seeking this user-right, we are kind of stuck with it. For users seeking other user-rights like block, delete-to-make-way-for-page-move, delete-and-"content-blind-undelete" (useful in history merges), and the recently unbundled edit-filter-manager and template-editor user-rights an RfA process is not required by the Foundation and, as we have seen with edit-filter-manager and template-editor, may not be required by the community either. Disclaimer/selfish conflict of interest: If the tools needed to do a history-merge are ever available without going through RfA, sign me up. davidwr/(talk)/(contribs) 03:07, 15 October 2015 (UTC)
  • Of course, this only means that there is no process that they could come up with, other than RFA-like, which would be acceptable; however, it's theoretically possible that some user will surprise them with a totally different process that they would approve of. עוד מישהו Od Mishehu 18:44, 17 October 2015 (UTC)
• I'm neutral on this one. I think that RfA could be reformed successfully while retaining something close to its current incarnation, but I would also support well-though-out proposals that look nothing like the current RfA. The devil is in the details - just because we ditch RfA doesn't automatically mean that we will get a system that is better. — Mr. Stradivarius ^{♪ talk ♪} 14:40, 15 October 2015 (UTC)
• ArbCom elections seem to work quite well. I suggest a (say quarterly) election where any interested candidates can put themselves forward to be a life-term admin. People vote, and anyone with a specified percentage of support gets the bit. There is no quota of new admins to fill, just a ballot where they can be elected democratically. Only an idea. Rcsprinter¹²³ (tell) 20:11, 15 October 2015 (UTC)
• It's not fundamentally broken, but the "chumming the water for the sharks" approach needs to be fixed. Montanabw^(talk) 01:56, 17 October 2015 (UTC)

H: Leave as is

The process with which we select our admins is completely fine as is. There are no problems with it, and no change is needed at all.

Support H

1. This is a loaded question, but RfA in itself isn't broken. It's being trolled, sure, but that goes for every page on Wikipedia. There is no better process than community approval: this is what we are, this is what working in a collaborative environment is all about. There is no metric for determining whether someone is going to be a good admin or not; it cannot be automated. Drmies (talk) 00:51, 15 October 2015 (UTC)
2. I'll put a vote in here. I've been participating (off and on) in Dank's RfCs about RfA for a while now, and in my opinion a lot of the collapse in RfA is due to a shrinking pool of long term editors. The decline in long term editors can be more moderate than RfA decline, in fact I'd expect it to be, and can drive both candidate and voter behavior. We can and should reform RfA, but no reform will be magic. Protonk (talk) 13:48, 15 October 2015 (UTC)
3. Definitely. Those who say otherwise are failed candidates, and voters who didn't get it their way, a small minority of users, overall. That's the reason why all proposals of reform have been voted down so far. Kraxler (talk) 11:50, 16 October 2015 (UTC)
4. Qualified Support The way this is drafted I don't quite know whether to support or oppose, but to be on the safe side I support. I favor various methods of unbundling as proposed here and making desysopping easier. But if that doesn't happen then no, I don't favor any change in the current system. Coretheapple (talk) 14:31, 16 October 2015 (UTC)
5. Support Not totally convinced that anything really needs to be done, except of course increasing participation, which isn't really a change to RfA so much as a renewed effort to advertise them. RO^(talk) 18:07, 17 October 2015 (UTC)
6. We haven't given as many people admin rights in recent years because we haven't had as many suitable candidates wanting to become adminsitrators and we have perhaps been more careful as a community regarding who is given admin rights than in the early years. No amount of 'reform' of RfA is going to change that and I'm not convinced that the RfA process is a significant deterrent to suitable candidates wishing to become admins. --Michig (talk) 22:23, 17 October 2015 (UTC)
7. Strict support. Almost every grievance is misattributed to RfA. They should be attributed to a subset of the participants. Esquivalience ^t 22:47, 17 October 2015 (UTC)

Oppose H

1. Oppose The very obvious consensus amongst the community is that RfA needs some sort of change. I think enough has been said. --Biblioworm 00:20, 15 October 2015 (UTC)
2. Oppose While RFA itself should continue to exist, the existing process needs change. An RFC was recently proposed to look into adopting a DE-WP style system. While on merits it looked good, it wouldn't work here, but it certainly sparked, along with a large number of other RFC's, a great deal of discussion about the whole RFA process and the subsequent management of administratorship. Blackmane (talk) 02:20, 15 October 2015 (UTC)
3. Oppose-ish It has its problems, but do not take this "oppose" to mean the status quo is worse than whatever will come out of this discussion or that the status quo will automatically have less community support than any particular proposal here. In other words, even if 80% of the community says "the current system is broken" but no single proposal or group of mutually-compatible proposals has more than 20% support, then we must acknowledge that the current system is still "more popular than any other viable proposal" and either leave it as is or have a second "up or down vote" RfC to see whether the community preferred to keep the current RfA process or the new proposal with the most support. Do not "close" the discussion by saying "oh, the current RfA has only 20% support so we will have to ditch it, let's see, the highest-supported proposal has 18% support so we'll go with that one." davidwr/(talk)/(contribs) 03:13, 15 October 2015 (UTC)
4. Something has to change. The current process is broken and we *do* need a reform. Regards—☮JAaron95 ^Talk 05:12, 15 October 2015 (UTC)
5. Oppose, per Biblioworm, above. Change is needed to improve this site. — Cirt (talk) 09:02, 15 October 2015 (UTC)
6. Oppose I don't agree that RfA is totally a broken process but I think there are things which can be reformed/changed. Jim Carter 12:42, 15 October 2015 (UTC)
7. While it worked in the past, it seems to no longer work for the Wikipedia we have today. —Kusma (t·c) 12:48, 15 October 2015 (UTC)
8. Oppose We need more admins, and the current system isn't creating them. Something has to change. — Mr. Stradivarius ^{♪ talk ♪} 14:02, 15 October 2015 (UTC)
9. Oppose - A change is certainly needed although to be totally honest right now I'm not sure what does need changing. –Davey2010^Talk 00:53, 16 October 2015 (UTC)
10. Opposing this since I supported a couple things above. Townlake (talk) 02:17, 16 October 2015 (UTC)
11. Oppose per everything stated above and my comments and !votes in earlier sections. Kind of a silly proposal as anything can be improved. ;-) Montanabw^(talk) 01:59, 17 October 2015 (UTC)
12. Oppose - RFA isn't producing enough admins, as can be seen by the backlogs getting bigger. עוד מישהו Od Mishehu 18:47, 17 October 2015 (UTC)

Comments on H

• Question: What is the German system? BMK (talk) 02:29, 15 October 2015 (UTC)

  @Beyond My Ken: See Wikipedia:Wikipedia Signpost/2012-10-22/Special report. — Mr. Stradivarius ^{♪ talk ♪} 13:58, 15 October 2015 (UTC)

  Thanks. I read it, but I'll have to digest it. BMK (talk) 15:16, 15 October 2015 (UTC)

  Sorry I was a bit vague with my remark. I read the RFC that concerning it, but I couldn't remember exactly where it was. I'll have to do some digging for the RFC. Blackmane (talk) 03:54, 16 October 2015 (UTC)

• Eh, no. The number of users who believe that no changes whatever (no easier desysopping, no unbundling, no anything) are a very small minority indeed. Anyone will be extremely hard-pressed to show serious evidence that a majority of users believe that our system is perfectly fine as is. Reform RfCs are voted down because no can agree on how to fix the issue, not because they think we don't need any reform at all. --Biblioworm 15:51, 16 October 2015 (UTC)

I: Ease the load on admins

The admin load can and should be lightened with carefully considered changes to procedures and/or by creating bots and scripts to assist and possibly in a few specific cases replace admins.

We've already done this with Wikipedia talk:Templates for discussion/Archive 19#RfC: Proposal to allow non-admin "delete" closures at TfD 3 months ago, effectively allowing editors to make all of the decisions and do all of the work except actually hitting the "delete" button.

Here are two hypothetical examples:

• Encouraging established editors to "second" speedy-deletes and giving admins a tool to flag those CSDs as "seconded by an established editor" would free the admin up to do a 2-second sanity check before hitting "delete," freeing up his time for more involved tasks.

• Creating admin-bots to allow editors to delete a page they created in their own User: space if it meets certain other conditions (e.g. never been edited by anyone else, never been moved, etc.) would free up admins to do other tasks.

Note: "support" or "oppose" on the general idea that we should work towards lightening the load of admins, not on any particular method. You can be dead-set against having a special group of "CSD second-ers" or having admin-bots to delete user:-space pages but still support lessening the load of admins.

Support I

1. Support as creator of this section. davidwr/(talk)/(contribs) 03:36, 15 October 2015 (UTC)
2. Support. We have the technology. —⁠烏⁠Γ ^(kaw), 05:53, 15 October 2015 (UTC)
3. Support, most sensible ideas. — Cirt (talk) 09:03, 15 October 2015 (UTC)
4. Support I like the idea, I can't imagine any major problem with this idea if applied. Jim Carter 12:47, 15 October 2015 (UTC)
5. Support In the long run, this is one of the better solutions, but very few bots come out of committees. :| Protonk (talk) 13:48, 15 October 2015 (UTC)
6. Support Sure, these are good ideas. I don't think they should be should be a substitute for reform of RfA itself, however. — Mr. Stradivarius ^{♪ talk ♪} 13:57, 15 October 2015 (UTC)
7. Support - Why not? These are amazing ideas. Yash! 14:09, 15 October 2015 (UTC)
8. Support: this can and should be done, even if it will not solve the problem in its entirety. — Bilorv_(talk)^(c)(e) 15:47, 15 October 2015 (UTC)
9. Support we should certainly allow users to delete their own work without needing an admin. A large proportion of U1 and G7 requests could be automated, computers can easily scan through hundreds of edits to see if a page has been moved. ϢereSpielChequers 18:29, 15 October 2015 (UTC)
10. Support - These are all great ideas. Automation would reduce mundane button pushing work while setting up a vetted queue of actions seconded by established users would both speed up admin efficiency and give those who seek adminship an additional chance to prove their worth. ---- Patar knight - ^chat/_{contributions} 02:04, 16 October 2015 (UTC)
11. Support Yes, good idea, and makes sense especially in a package with unbundling. Coretheapple (talk) 14:27, 16 October 2015 (UTC)
12. Support Of course! If the load can be lightened it should. RO^(talk) 18:08, 17 October 2015 (UTC)
13. Support - there are probably some classes of tasks which can, at least to a significant degree, be handled by bots with a reasonably low false-positive rate. Tasks like self-evident G7 and U1 deletions are an obvious example. עוד מישהו Od Mishehu 18:20, 17 October 2015 (UTC)

Comments on I

• I like the WP:U5 adminbot idea, but that would require considerable input and well-established consensus. We do have ProcseeBot, who makes about 300-500 blocks a day, so this kind of powerful full automation is not unheard of. From my own experience, however, what we really need is more recent changes patrollers. This task is ongoing and of extreme urgency. A TfD that stays open for an extra week is absolutely fine, but the wiki will be useless if we can't keep it from being defaced. If all I had to do was clean out AIV than it would be an amazing, rewarding day and I'd have a few less gray hairs. Instead most of my blocks are from patrolling, since during peak hours we seem to be lacking that much needed help. Meanwhile the unfounded reports at AIV grow and grow, but I digress — MusikAnimal ^talk 07:02, 15 October 2015 (UTC)
• I am not convinced that taking away more of the "easy" part of the workload is going to make a day of admin work any more fun. As MusikAnimal said, it may help with some less important backlogs. However, to deal with user-requested deletions is actually the relaxing part of CAT:CSD patrol (you sometimes even get nice thank you notes for it). If you take that away, only the stuff remains that will make people shout at you or that is difficult to sort out. When I do speedies, I like it that some things are easy and some are hard. Probably just me, though. —Kusma (t·c) 10:08, 15 October 2015 (UTC)
• I'm not going to oppose this because I don't think it would be a bad idea, but I don't think it would make much difference either. The workload of processing U1 deletions isn't very much, and any other tasks which could feasibly be replaced with adminbots would also be fairly low workload. I don't see this approach having significant impact on the total demand for admin time. Hut 8.5 20:19, 15 October 2015 (UTC)
• I worked with the TfD regulars to write the original version of the TfD RfC, so obviously I support this in principle, and I especially support it in cases where it's essentially tradition and social expectation rather than a genuine need for admin tools that puts a particular task into the admin queue. But I'm not convinced it would be a win to turn low-volume tasks currently done in a decentralized way by whoever has ten seconds to spare into tasks that require small groups of active bot developers and maintainers. And as the others above say, there's a psychological effect at work. Most admins get their warm fuzzies from doing things that help other editors be more productive, though we may disagree on the best ways to do that. If we think of 'admin time' as an undifferentiated mass resource, and try to reallocate it toward the most difficult of the current admin tasks, we'll just end up with less aggregate admin time because most of those things are no fun. Opabinia regalis (talk) 23:05, 15 October 2015 (UTC)

  Question: what are some other examples of cases where this could be done by process changes rather than by development of new tools? Opabinia regalis (talk) 19:26, 17 October 2015 (UTC)

• No substitute for human eyes. Bots are wonderful, and they will continue to be developed, but the blocking and revdel functions require human judgement. Montanabw^(talk) 02:01, 17 October 2015 (UTC)

J: Oppose votes carry too much weight

As Biblioworm said in their Signpost op-ed, due to the average of 75% support necessary to pass an RfA, oppose votes carry about three times as much weight as support votes. That is, if an editor gets six oppose votes, it will require at least 18 support votes to cancel them out. Most voting systems only require 50% plus one vote for a pass, and as Biblioworm also points out, at 67%, even getting the U.S. president's veto overturned requires a lower percentage of support than RfA. Closer to home, the German Wikipedia require 50% + 1 67%, and it doesn't seem to have blown up their wiki. For a fairer system we should reduce the discretionary zone from 70%–80% to something lower.

Support J

1. As creator of this section. — Mr. Stradivarius ^{♪ talk ♪} 13:54, 15 October 2015 (UTC)
2. Support - Something that should have happened a long time ago. Yash! 14:05, 15 October 2015 (UTC)
3. Definitely. A candidate with 2/3 majority (or even 3/5 majority) is still going to be widely trusted. —Kusma (t·c) 14:13, 15 October 2015 (UTC)
4. Support. Obviously. --Biblioworm 14:56, 15 October 2015 (UTC)
5. Yes, we need to lower the majority level.—☮JAaron95 ^Talk 15:54, 15 October 2015 (UTC)
6. Support. I think we should determine if there's consensus to change the values in this RfC, then hold another RfC to determine what, exactly, the new numbers should be. Otherwise, we'll be endlessly arguing back-and-forth about minor details when there's consensus for change. I think that if someone can gain a 2/3 majority, they have consensus on their side. NinjaRobotPirate (talk) 17:51, 15 October 2015 (UTC)
7. Support per NinjaRobotPirate. The target numbers don't matter yet; the fact that they need to change does. —⁠烏⁠Γ ^(kaw), 02:57, 16 October 2015 (UTC)
8. Support the 67% or 2/3 majority is plenty, and worth an experiment to see if it fixes the "not enough admins" problem (and FWIW, my RfA/1 ended at 60% so if I really am an Evil reptilian kitten-eater from another planet I would have failed under this standard as well—LOL). 75% is the "states vote to amend the US Constitution" standard and really far too high for an experienced user—those of us who have been around for awhile will most likely have pissed off a few people, let's not give the trolls so much power. I'm OK with a supermajority (though I personally would have loved a 50% standard, of course!) Montanabw^(talk) 02:07, 17 October 2015 (UTC)
9. Support - I think we can reduce the threshold a bit. I think that 60-67% is a good range to aim for. עוד מישהו Od Mishehu 18:38, 17 October 2015 (UTC)
10. Support. Everyking (talk) 14:26, 18 October 2015 (UTC)

Oppose J

1. This is just a backhanded way of stacking the deck in favor of getting more admins. Since people have well-grounded concerns, change the rules. No dice. Coretheapple (talk) 15:32, 15 October 2015 (UTC)
2. It is untrue that it is 75% to pass (it is about 70%), so per WP:RfC (the requirement for neutral statement) this RfC should fail for being misleading. Admins need very wide demonstrated support, if anyone can believe, in the least, they are trusted by the community. It is nonsense to suggest that 70 is somehow unfair but 67 is not. This proposal displays fundamental dislike of deciding things with even a semblance of WP:Consensus("A consensus decision takes into account all of the proper concerns raised. Ideally, it arrives with an absence of objections, but often we must settle for as wide an agreement as can be reached. When there is no wide agreement, consensus-building involves adapting the proposal to bring in dissenters without losing those who accepted the initial proposal" (emphasis added)), because apparently 'consensus is hard', which is the usual claim of people who just do not believe that Consensus matters. Alanscottwalker (talk) 16:11, 15 October 2015 (UTC)
3. Pre-RFA vote-stacking / canvassing via "nominations" is routine. The solution to this issue is not to make the routine cheating more effective. The simple fact is, good candidates still pass RFA. From a "hasten the day" perspective it would be great for massively problematic candidates to pass at 60%, but that's not really good for the project. Townlake (talk) 02:24, 16 October 2015 (UTC)
4. Opposing !votes attract opprobrium. To cast them, an editor has to have some courage and also some technical skill in finding and presenting supporting evidence. The supporting votes tend to be comparatively lightweight. For example, consider the recent RfA for Thine Antique Pen. Yngvadottir presented good detailed evidence but did not have the courage or confidence to make this an oppose. On the support side, we had lightweight support !votes such as "I owe him one". In such circumstances, it seems properly prudent that the final score of (118/58/11) was found to be inadequate. In such close cases, the behaviour of the participants would change, if the threshold were lower. Candidates would not withdraw so quickly when their RfA wasn't going well. There would be more pile-on votes to make sure of a negative result. And the support votes would be challenged more often if they were given more weight. The result would be to raise the temperature so there would be more bad-feeling and drama. Andrew D. (talk) 07:27, 16 October 2015 (UTC)

   Forgive me, but this made me laugh. Neither courage nor technical skill are required, and in fact I have seen many RFAs opposed by persons clearly entirely lacking in both. Beeblebrox (talk) 15:38, 16 October 2015 (UTC)

5. THis is one of those suggestions that is often made by, and certainly relished by those who just wanr it made easier for them to get the bit and grab some power - or have something to boast about in the schoolyard. Kudpung กุดผึ้ง (talk) 17:21, 16 October 2015 (UTC)

   I don't quite understand -- isn't the entire point of RfA reform to make it easier for people to get the bit? —Kusma (t·c) 20:19, 16 October 2015 (UTC)

   No. For some of us the point is to get more admins, and most of those extra admins would be people who won't currently run not because they aren't qualified, but because RFA all too often is a demeaning hazing ceremony that they don't want to dignify with their presence. ϢereSpielChequers 09:06, 17 October 2015 (UTC)

   "Make it easier to become an admin" and "make it so that more people become admins" are almost the same in my book. —Kusma (t·c) 21:27, 17 October 2015 (UTC)

6. Lowering the percentage would probably mean alot more canvassing & god knows else what, It's fine as it is IMHO. –Davey2010^Talk 21:55, 16 October 2015 (UTC)
7. 'Oppose lowering the threshold to allow more borderline cases to pass. We should be passing only the very best candidates. RO^(talk) 18:09, 17 October 2015 (UTC)
8. Admins are a small population, so a large supermajority is appropriate. I wouldn't want to be blocked by somebody who one-third of the community did not trust with the bit. Glrx (talk) 18:49, 17 October 2015 (UTC)
9. Oppose. We don't vote. We !vote. About time people understood the difference. Leaky Caldron 15:18, 18 October 2015 (UTC)

Comments on J

• Looking at recent German de:Wikipedia:Adminkandidaturen, it seems to me that they use 2/3 or 70% as the cut off point, not 50%+1. There is no bureaucrat discretion involved. —Kusma (t·c) 14:12, 15 October 2015 (UTC)

  @Kusma: Has this changed since 2012 when the Signpost op-ed I linked above was written? I may need to strike that part of the proposal if I have outdated or wrong information. Unfortunately, my German isn't good enough for me to check on the German Wikipedia myself. — Mr. Stradivarius ^{♪ talk ♪} 14:16, 15 October 2015 (UTC)

  Actually, it appears that I just misread the op-ed. If I am understanding this correctly, the percentage has always been 67%, and the thing that changed in 2009 was the addition of a re-election and recall procedure. — Mr. Stradivarius ^{♪ talk ♪} 14:25, 15 October 2015 (UTC)

  From de:Wikipedia:Adminkandidaturen/Statistik it appears that it has always been 2/3 (sort by "Prozent"; "wurde Administrator" and "erfolgreiche Kandidatur" are passes, "wurde nicht gewählt" or "... nicht bestätigt" are fails, while "Abbruch" is a withdrawn candidacy). I also understand it that the thing that happened in 2009 was a forced recall procedure that leads to a new election (there have also been many voluntary re-elections). I don't have a lot of time right now, but can dig to answer specific questions if you have any (I am a native German speaker, but almost inactive on the German Wikipedia). —Kusma (t·c) 14:36, 15 October 2015 (UTC)

• Note that 70% is near the rock bottom of the discretionary range. An RfA almost never succeeds in this range without a 'crat chat, and it is not uncommon for such RfAs to be closed as unsuccessful or no consensus. I say that 75% is the bar because, reviewing RfAs from previous years, I have seen virtually no RfAs that have failed above 75%, while the number that passed between 70-75% is split quite evenly. Under 75%, closes are typically very drawn out and controversial. --Biblioworm 16:18, 15 October 2015 (UTC)

  If they are "controversial" there is little point in claiming a consensus that does not exist - if the crats find an oppose so compelling that they declare it prevents consensus on a candidate, it is because the oppose is compelling by community standards. Alanscottwalker (talk) 21:22, 15 October 2015 (UTC)

  That's a kind of circular argument. It is controversial because we have made successful RfAs in the low 70s controversial by grabbing our bar randomly from the ancient VfD process. Almost any other group outside Wikipedia would not imagine the low 70s to be controversial by any stretch of the imagination. If a presidential candidate were to win the popular vote by 70%, it would go down in history as a complete landslide win; the same applies for virtually any other candidate of whatever sort. Also, the Supreme Court makes verdicts by simple majority; it doesn't matter if the tally is 5-4: it still passes. --Biblioworm 16:13, 16 October 2015 (UTC)

Those are ridiculous analogies, none of those instititution work by WP:Consensus. It just goes to demonstrate the hostility to WP:Consensus in this proposal - on Wikipedia for example, not a single one of these RfC questions on this page will be adopted without a super-majority. Alanscottwalker (talk) 21:54, 16 October 2015 (UTC)

The question of which proposals will be adopted is in the hands of the closers. And if a person really cares about consensus, as you claim to, they wouldn't be so obsessed with insisting that we keep and strictly adhere to the current bar for passing; rather, they would be flexible and willing to consider a wide range of possibilities. (Look, really: does giving a narrow 5% window for discretion really make something a consensus-based discussion?) Resolutely defending the current bar without the will to even consider anything less does not show respect for consensus: that's voting. Consensus, by its very nature, as defined by the policy you keep getting selected quotes from, is flexible and varies on a case-by-case basis. You cannot "pick and choose" from the policy; you must accept all of it. --Biblioworm 22:53, 16 October 2015 (UTC)

Apparently, you have a fundamental misunderstanding of Consensus. It requires no one to agree with you. I'm sorry if that bothers you. I've seen enough closes of policy RfC's to know how they work. The preliminary concern addressed is whether participation is broad enough for the matter addressed, and then if in the aggregate there is an agreement by the users, based in current policy, and common sense. -- Alanscottwalker (talk) 23:57, 16 October 2015 (UTC)

Where is my fundamental misunderstanding of consensus? One of the central tenets of the policy is that arguments are judged on their strength. It doesn't reflect very well on a person to constantly get selected quotes from a policy and then quickly reverse their opinion when a central point of the policy becomes inconvenient. In my comment, I said that because consensus is a relative and dynamic process that relies on the strength of arguments rather than percentage bars, it makes no sense to claim to defend the policy and then turn around and insist upon keeping a strict, narrow percentage-based bar and so-called "discretionary range" (about 5%), which is not consensus, but voting with a small window of "discretion" just so that people can have an excuse to call it a "discussion". Real consensus is much more open-ended. So, I'm very much at a loss to understand how I am showing a fundamental misunderstanding of a policy by using one of its central principles to make a point. --Biblioworm 00:36, 17 October 2015 (UTC)

Your fundamental misunderstanding is that "strength of argument" is not judged by some omniscient alien, it is fundamentally judged by the community in what they see as good to accept and not good to reject. One of the very reasons policy is important is it is recordation of consensus of what the community has accepted and rejected. Alanscottwalker (talk) 00:53, 17 October 2015 (UTC)

I'm sorry, but someone must close the discussion (they're not "omniscient aliens", by the way), whose job is in fact to judge the strength of the arguments. The process described in WP:CONSENSUS is that editors make arguments for their view, and then someone closes the discussion according to the strength of the arguments. So, it seems to me that by saying "'strength of the argument'...is fundamentally judged by the community", you're suggested that the community should be judging the strength of the arguments they make, which would result in all kinds of disagreement, bias, and circular chaos. The community's role is to voice comments on the proposal; the closer's role is the determine the consensus. And finally, consensus can change. It's not set in stone forever. --Biblioworm 15:25, 17 October 2015 (UTC)

If you don't want understand the community dynamics, that is your loss. Strength-of-argument is not actually the platonic form magic, you surmise. Any closer, who does not see a split community, where and when it is staring them in the face, is in for one heck of a time. Your citation to consensus-can-change just makes the point that the community is in charge of the changing and the consensus, no one else - most particularly by virtue of function, the closer is not in charge. (I think, we have discussed this meta-stuff sufficiently, so I'll move on). -- Alanscottwalker (talk) 16:47, 17 October 2015 (UTC)

Yes, and after this comment, I will move on as well. I happen to be getting quite tired of this discussion. But, to be quite frank, I see this as an attempt to say "I'm sorry that you don't understand" as a blanket statement so that challenging points can be avoided. I do understand community dynamics; I simply interpreted the consensus policy that you kept getting quotes from. I don't see how your comments disprove my point. I think the interpretation of the consensus policy is rather clear: in proposals, the members of the community voice opinions on the proposal, and the job of the closer is to determine the consensus of the community. Of course, the closer is not the consensus, but rather the interpreter of the consensus the community reached, according to the strength of the arguments presented by the proponents of the respective positions. This is what I was saying all along, so I don't see where I ever denied it. In the spirit of true consensus (not a very narrow and artificial range of 5%), according to the policy, bureaucrats should have much more leeway to weigh the strength of arguments on both sides "as viewed through the lens of Wikipedia policy." There is a tendency at RfA to treat as a vote outside the discretionary range of approximately 70–75% and only allow the weighing of arguments in that narrow range. This is not a true application of the consensus policy, which states that arguments should always be weighed when closing a discussion, not just within a narrow percentage range. The point is, RfA is much more vote than it is discussion and true consensus. That must change. I rest my case. --Biblioworm 18:35, 17 October 2015 (UTC)

• I'm also getting very tired of hearing this complaint about the "neutrality requirement" for the RfC and that the RfC should therefore fail. First of all, it's a logically invalid argument in that it attempts to cheaply discredit the argument by criticizing the way it is written rather than addressing the substance of the issue. By this logic, all RfAs should fail, because the nominator doesn't write a neutral statement; all RMs should all fail, because the proposer is "advocating" that it be moved to a new title and is therefore not neutral; finally, pretty much all other RfCs should fail, because the proposers often write at least a few lines showing why he thinks the proposal is needed. I should also note that I wrote all the proposals here favorably for the side concerned, so I did not write it well for my "side" and write it poorly for the position that we should leave RfA as is. --Biblioworm 16:31, 15 October 2015 (UTC)

  No. RfC policy is clear you can't stack the deck, just because you want to ask leading questions, include misinformation and get a skewed answer that supports your apriori conclusion - it makes no sense to do it your way, in light of what an RfC is actually suppose to do. Alanscottwalker (talk) 16:44, 15 October 2015 (UTC)

  I put headers for each of the topics that were to be voted on, and I wrote a brief summary of the beliefs which the holders of said position possess. It was not written in a "this is the way it certainly is" manner, but rather written to describe the beliefs of the position's proponents. I wrote it even-handedly for all the proposals, and represented the positions involved as well as I could (even those I disagreed with). How else was I supposed to write it? I had to put some description there, because a simple header might not be sufficient for voters to fully understand what they're voting on. By your logic, proposals G and H should be getting more support than they are, because the "skewed" opening summary supposedly biased in favor of the position should influence the result. But they're not. The positions are obviously not gaining consensus. This shows that the opinions being expressed are actually the opinions of the community, and not the "skewed" result of the opening summary. This is precedent for this, you know: WP:RFA2013 did it the same way. And finally, Mr. Stradivarius started this section anyway. If you object to the way the opening summary is written, go talk about it with him, not me. --Biblioworm 00:40, 16 October 2015 (UTC)

  • I should also address Alan's constant citing of WP:CONSENSUS, namely his implication that we cannot lower the bar because proposals should ideally have complete support. But the wording there is very vague and allows a broad interpretation; it nowhere says "somewhere in the 70s is acceptable". In fact, I think I once read that our bar for RfA was actually just arbitrarily grabbed from the old VfD process. I will offer a quote of my own. WP:CONSENSUS says that: "Consensus is ascertained by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy." Therefore, since many oppose votes are not based in policy but rather personal, arbitrary criteria, their arguments are being given too much weight. Therefore, the proposition that we should allow all editors to oppose as they please "must fail". --Biblioworm 16:42, 15 October 2015 (UTC)

    Well, no. According to your analysis we must throw out most the supports in RfA. But you are wrong. They are based in policy - policy explicitly asks for the individual User's judgement of "trust", based among other things, "common sense". Alanscottwalker (talk) 17:28, 15 October 2015 (UTC)

    Well, no. If the support votes are based in policy (Example: Support User follows the CSD policy/is civil and competent/meets all the expectations set forth in WP:ADMIN/etc.), then they should not be discarded. Supporting per the rationale of a user who made a policy-based argument would also be legitimate, since the supporter is saying that he agrees with the policy-based reason the supporter in question gave and that he simply sees no reason to write it all over again when it's already been said. On the other hand, many oppose !votes are based upon purely personal and sometimes unreasonable expectations (doesn't have at least 20,000 edits/hasn't been around for at least 2 years/hasn't written at least two GAs or an FA/etc.), which have no grounding in Wikipedia's policies. A proper oppose vote should show relevant examples of the candidate actually violating policy. --Biblioworm 00:40, 16 October 2015 (UTC)

    I see this as Alan simply going in circles with an argument that's basically "policy can't be changed because it's policy". —⁠烏⁠Γ ^(kaw), 02:56, 16 October 2015 (UTC)

• If 67% were the "bar", the following additional RfA's would have passed this year: Cyberpower678 and Thine Antique Pen; two more are just outside this range: EuroCarGT (65.83%) and Rich Farmbrough(65.97%). While this may be a good idea in its own right, we need to be realistic that it's not a "magic bullet" that is suddenly going to produce "dozens" of extra Admins each year... --IJBall (contribs • talk) 16:25, 15 October 2015 (UTC)

  Part of the idea of lowering the pass percentage is to make it more attractive for candidates to run. Hopefully the effect wouldn't just be two extra admins, but several, due to the increased number of candidates overall. Also, I'm not proposing that the bar be 67%, just that it be something lower than it is now. Actually, my personal preference would be 50% + 1, although I also like the idea of dropping the vote counting altogether and deciding things purely on the strength of the arguments. — Mr. Stradivarius ^{♪ talk ♪} 23:38, 15 October 2015 (UTC)

• I don't have an opinion one way or the other on this proposal, but I do have an important response for the comments in the Oppose section made by @Alanscottwalker:. He brings up an important point about Consensus building, but there's a fundamental difference between the sort of consensus building being discussed in the text he brings up, which speaks to "adapting the proposal to bring in dissenters without losing those who accepted the initial proposal"; this is all about finding reasonable compromises to allow article text to meet standards acceptable to differing viewpoints. RFA's don't fail because those who oppose a candidate and those that support one can't come to an agreement on how to promote the person properly. As much as we want RFA to be a "consensus building process", it simply isn't. It's a vote-with-rationale. But we shouldn't pretend like there's a middle ground here that the traditional "consensus building model" is designed to achieve. This is a rationale-weighted vote, and nothing more, because there is no "third way" on an RFA vote: either a candidate becomes an admin, or they do not. It's like the old canard about being "partially pregnant". You either become an admin or you don't. There's no compromise to be made. So the "consensus-building model" noted above is woefully inadequate for dealing with the nature of these votes. --Jayron32 20:25, 15 October 2015 (UTC)

First of all, no, many opposses explicitly (and implicitly) argue that the candidate is not suitable at present, not for all time. Moreover, the consensus for passage does emerge sometimes at a later date (sometimes even within the seven-day period, but sometimes at a second or third proposal of the candidacy) when the concerns are addressed. Alanscottwalker (talk) 21:04, 15 October 2015 (UTC)

That's not my point at all, so let me restate it in different terms. I'm sorry it was too obtuse. A single vote, for a single candidate, is not about consensus building because there is no middle ground to build consensus towards. You can't "part way" make someone an admin. That is all. If anything else I stated made you think I was discussing anything differently ignore it like it was never said. All that matters to my point is "there is no compromise between promoting a candidate or declining to promote them". All else is irrelevant to my point. --Jayron32 21:51, 15 October 2015 (UTC)

But that way of looking at it is not how people pass or fail, they do so in the aggregate of all comments (not the individual support or oppose). There is a compromise of concerns, and the community either finds it or it does not. And even with the individual support or oppose or neutral, the particular raised concern may be ultimately compromised enough, in the course of that discussion or in a later proposal of the candidacy. Alanscottwalker (talk) 22:07, 15 October 2015 (UTC)

• I should note that not a single opposer of this view has ever directly addressed the rather obvious fact that we give opposers triple power (or, at the very least, almost triple power) over the supporters. Why in the world should naysayers be getting so much weight? Think about this: only a little more than 1/4 of the participants must oppose to sink the RfA. If there were ten participants, only three need to oppose. (A bit less, actually, but I just rounded up the number since we obviously can't have a fraction of an oppose.) Furthermore, it doesn't matter how outrageous the argument is; it still counts. Aren't we sort of assuming that the candidate would be bad? Isn't paying so much attention to the negative side a violation of WP:AGF? --Biblioworm 16:13, 16 October 2015 (UTC)

• Supermajorities are normally required for certain important decisions, and RfAs understandably require a very high one because it is a powerful and lifetime post. That does inherently give the opposition power. The supermajority custom in RfCs is one reason admins are able to block, as they did quite recently, proposals to make it easier to desysop them. I don't like it, but supermajorities are not unusual and that's how it is. Coretheapple (talk) 17:42, 16 October 2015 (UTC)
• At least make it a 2/3 supermajority (what Congress uses) and make the discretionary range about 60–65%, which would be more reasonable, instead of the unrealistic and Wikipedia-unique approximate 3/4 supermajority required to pass an RfA. --Biblioworm 19:02, 16 October 2015 (UTC)

• Mr. Stradivarius makes a point with which I agree though I suppose there can be no evidence for it unless it is tried. Lowering the bar might encourage more candidates to apply. So such a change would not just have resulted in two or a few recent candidates passing but more candidates applying. I don't agree with 50% + 1 as a passing percentage nor do I think that a substantial lowering of the passing percentage would be needed. If all candidates receiving 75% or above (not "most" as the RfA page now states - implying something more such as 80% is needed to pass) were assured of approval and the discretionary range were 65% or even 67% to 74%, I think the bar would not appear to be so high. In fact, I would favor lowering the assured passing percentage to a lower number such as 70% while making the discretionary range a narrower 65-69% but even the slight change that I led with might have a favorable impact without unduly disturbing most of those who want to keep a high bar. Donner60 (talk) 01:45, 17 October 2015 (UTC)
• I'm OK with leaving the support percentage at 70-75% provided we have clerks to get rid of the most unpleasant opposes as per proposal M. My fear if we leave the system as is but lower the required support threshold we could get more unpleasantness and more off site canvassing on badsites. Also there have been some candidates who have got 60% or more who I don't think were ready or suitable. ϢereSpielChequers 14:04, 17 October 2015 (UTC)
• We should only be passing the very best candidates. Well, that won't make us many admins, will it? And, as I have mentioned before, there is no evidence to show any correlation between the percentage of support in the RfA and the candidate's subsequent performance as an admin. Candidates with unanimous or near-unanimous support were desysopped for tool misuse. Candidate who just barely made it through and had many supposed "problems" turned out just fine. Percentage means nothing. --Biblioworm 18:41, 17 October 2015 (UTC)

  Well, it could be true that candidates who had difficult RfAs take the feedback seriously and work more cautiously than they might have if they'd been waved on through. The fact that percentage and performance are uncorrelated doesn't imply that percentage is meaningless, either. Opabinia regalis (talk) 19:50, 17 October 2015 (UTC)

• In regard to Leaky's comment, we do not !vote: we vote. RfA is a vote, except when it falls within a tiny little 5% range. This is not really a discussion, but rather a 95% vote a tiny little window of "discretion" for bureaucrats so that people have an excuse to call it something that it's not. Note that this system does not follow the process described in the WP:CONSENSUS policy or the Wikipedia:What Wikipedia is not policy. (WP:NOTDEM, in particular.) Consensus is supposed to apply at all times, not when it falls within a narrow percentage range, which in itself still relies on voting since the whole thing is based upon percentages. --Biblioworm 18:57, 18 October 2015 (UTC)
  • Nothwithstanding the clarification, the real issue with current RfA voting is the initial pile-on of popularity based friends blindly adding support before a question has been answered. Leaky Caldron 19:21, 18 October 2015 (UTC)
    • That might very well be a problem, but pile-on opposition sparked by poor and/or trivial oppose rationales (sometimes "supported" by cherry-picked diffs that ignore the context) is also a problem. Good support votes should show that the candidate follows policy, or cite rationales that do so; good oppose votes should show where the candidates has truly violated policy. Since we are to WP:AGF, the burden of proof should be on the opposition to show that the candidates habitually violates policy and would therefore be a poor admin. ;) --Biblioworm 19:27, 18 October 2015 (UTC)
      • the burden of proof should be on the opposition to show that the candidates habitually violates policy and would therefore be a poor admin. One could be a bad candidate for adminship without "habitually violat[ing] policy", and oppose voters are not required to demonstrate this beyond reasonable doubt as if this were a court proceeding. Why is the burden on opposes? The burden should be equally on supporters to demonstrate that the community should trust the candidate. In fact, that's why RfA's get so negative. Supporters demand so much proof that they end up badgering opposers into piling it on to make their case. RO^(talk) 19:45, 18 October 2015 (UTC)

K: Allow limited-charter RFAs

Allow candidates to run on a "limited-charter" RFA in which they ask for permission to use only certain tools or only use them for certain purposes, subject to sanctions (summary desysopping in most cases) if they exceed their "RFA charter."

If they want permission to use additional "sysop" rights in the future they can run again to expand their user-rights.

Procedurally, this would not be "RFA light" - it would run the full 7 days and would have the same "process" of a full RFA, but it would eliminate the "Oppose: the candidate says he will focus on deletions and I trust him for that but I don't trust him not to block someone incorrectly."

In the future, if and when tools are un-bundled or if the community decides that "customized userrights packages" are desirable, those who hold the sysop bit merely because they passed a "limited-charter RFA" will lose the sysop bit in favor of less powerful user-rights unless doing so is technically impossible (e.g. if they still need a user-right that hasn't been unbundled). Since the unbundled rights may include rights that aren't needed in their charter, tey will still be limited to "admin-type behavior" listed in the "RFA charter" unless they apply for additional rights using whatever process exists at the time.

Support K

1. Support as creator of this section. davidwr/(talk)/(contribs) 18:00, 15 October 2015 (UTC)
2. Weak support – I preferentially support unbundling, but failing that, this is the next best option. No matter how much people around here want to bury their heads in the sands about "all encompassing Adminship", there is a substantial subset of productive veteran editors who aren't interested in the whole "bit" but in just a "subset" of the tools (e.g. moving or deleting, but not blocking), and who will never submit themselves to a "full" RfA because they know they will be rejected for exactly that reason ("But you have NO WP:AIV experience!!1! J'OPPOSE!!"). --IJBall (contribs • talk) 23:27, 15 October 2015 (UTC)
3. Support, either in lieu of or in combination with any level of unbundling. —⁠烏⁠Γ ^(kaw), 03:01, 16 October 2015 (UTC)
4. Support. Yes, I think this is a very good and creative idea. Coretheapple (talk) 14:25, 16 October 2015 (UTC)
5. Weak support: If they can't unbundle the tools, this would be a simpler version of "open to recall" - the candidate declares the tools they need, if they use any others, then smackdown! I could live with that. Montanabw^(talk) 02:10, 17 October 2015 (UTC)

Oppose K

1. I don't want to vouch for another ~RfA when the current process is badly damaged. This only leads to more chaos and troubles. If this happens, ArbCom and AN/I will have more work to do. And as I said in section F, Admin tools are for Admins. —☮JAaron95 ^Talk 18:17, 15 October 2015 (UTC)
2. Oppose: As noted above, if I don't trust the user to block someone, I wouldn't trust them to know when to properly protect or delete articles. The suite is packaged together for a reason. --Jayron32 20:16, 15 October 2015 (UTC)
3. Too complicated. Townlake (talk) 02:25, 16 October 2015 (UTC)
4. Too complicated. --Kudpung กุดผึ้ง (talk) 17:22, 16 October 2015 (UTC)
5. This effectively proposes a hat shop. Esquivalience ^t 20:25, 16 October 2015 (UTC)
6. Too complicated. –Davey2010^Talk 21:48, 16 October 2015 (UTC)
7. Counterproductive. We want admins who as they get experienced move on into various other areas of adminship, we should be encouraging trusted editors to help out in other areas not trying to confine them to what happened to interest them at the time of their RFA. ϢereSpielChequers 13:58, 17 October 2015 (UTC)
8. Too complicated and counterproductive. RO^(talk) 18:09, 17 October 2015 (UTC)
9. This is bit splitting. Glrx (talk) 18:52, 17 October 2015 (UTC)
10. Mistaking the map for the territory. Opabinia regalis (talk) 19:35, 17 October 2015 (UTC)
11. Totally against dilution of accountability. Leaky Caldron 15:23, 18 October 2015 (UTC)

Comments on K

• @Townlake: Why do you feel it's complicated? One simply deconstructs admin functions, so that Category 1 allows, say, only page moves and Category 2 allows page moves and CSD deletions. Etc. It doesn't have to be complicated. Coretheapple (talk) 16:03, 16 October 2015 (UTC)

  Monitoring compliance would be a nightmare, and WP:IAR would be an easy excuse for breaching charter. Townlake (talk) 00:14, 17 October 2015 (UTC)

• Comment: If there is a technical reason the tools cannot be unbundled, I see the potential for this to be akin to a "probationary" period; if there is a clear criteria (checkboxes would be handy) and the candidate is very specific about what tools they need and agree not to use any others, perhaps with an automatic desysop if they exceed the parameters of their agreed-upon toolset, I think this could be a very good idea; it addresses the "trust me" versus "can't be trusted with the tools" !votes. Give them the mop (or the WP:ROPE) and see if they can earn the trust of the community. Montanabw^(talk) 02:14, 17 October 2015 (UTC)
  • You're supposed to earn the trust of the community before you ask for the tools, not after. Townlake (talk) 14:40, 17 October 2015 (UTC)
    • Exactly! RO^(talk) 18:22, 17 October 2015 (UTC)

L: Minimum requirements for voters

Currently all but IPs can vote at an RFA. This leads to inexperienced editors and newbies influence a process of which they have no idea about. There ought to be a minimum requirement for voters

Support L

1. Currently we have an unclear but effective minimum requirement for voting, and newbies who vote are at risk of being treated as SPAs. Effective in the sense that it deters the vast majority of newbies, but I would argue it is far too effective and deters lots of people who we would want to participate at RFA. A clear but low voting requirement such as 100 edits would be less intimidating to newbies. DE wiki has a written rule and in my view that explains why they have a higher turnout in their RFAs. ϢereSpielChequers 09:01, 17 October 2015 (UTC)
2. Yeah, I also think this would be a good idea. It's akin to an "age limit" in regular elections. But, if such a level is introduced, it should be mild – perhaps 200(?) edits to vote. All that said, I don't think "newbie voting" is really all that prevalent, and it's certainly way down the list of things that are causing RfA's to be such a relative cesspool. --IJBall (contribs • talk) 20:36, 18 October 2015 (UTC)

Oppose L

1. Oppose. Personally, I don't often see newbie voters at all, and I don't think they're really a problem. The true problem lies in the group of experienced voters who get away with incivility and their completely made-up and unreasonable criteria. --Biblioworm 05:30, 17 October 2015 (UTC)
2. Oppose I don't think that newbie voters are really a problem, and fail to see a need to effectively disenfranchise them. Spirit of Eagle (talk) 05:39, 17 October 2015 (UTC)
3. Solution in search of a problem. Townlake (talk) 14:42, 17 October 2015 (UTC)
4. Oppose We need increased participation, and I reject the notion that newer people are inherently unable to provide meaningful feedback on community leaders. RO^(talk) 18:10, 17 October 2015 (UTC)
5. Oppose - if an anomymous user has some comment with merit - we should allow him/her to express it. As long as there is no suspicion of sockpuppetry, I see no reason to stop it. And sonce the newcomers rarely vote at RFA, it wouldn't solve much, anyway. עוד מישהו Od Mishehu 18:34, 17 October 2015 (UTC)
6. Oppose I'd be more in favor of an activity guideline, meaning the voters had to have made an edit in the past 12 months. It's a little nuts when someone who hasn't edited from their account for years, suddenly pops up to (usually) oppose a candidate and then disappears again. Liz ^{Read! Talk!} 19:50, 17 October 2015 (UTC)

Comments on L

• Yes and no. We need to indent, strike out or remove votes that come from single-purpose accounts, which already happens. So we already have a minimum requirement, which is "anyone that probably isn't an SPA". But I don't believe we need to go further than that, because I have never seen any evidence that "newbies influence a process of which they have no idea about". I also don't think this would solve any of the problems with RfA discussed above (e.g. too high standards), which seems to be mainly what we're debating. — Bilorv_(talk)^(c)(e) 08:27, 17 October 2015 (UTC)
  • But as a newish editor how do you know whether your vote would be accepted or you would be seen as too new? ϢereSpielChequers 13:53, 17 October 2015 (UTC)
    • As a newish editor I'd be very surprised if you knew about the SPA rule. But I think this is a pretty rare situation anyway, and if the user thinks they stand a chance of their !vote being 'accepted' then they might as well try. — Bilorv_(talk)^(c)(e) 17:30, 17 October 2015 (UTC)

M: Active clerking at RFA

We have clerking at other contentious areas such as Arbcom, active clerking at RFA would do much to take heat and incivility out of RFA. There is the argument that crats can ignore incivil votes, but simply leaving them to stand worsens the atmosphere at RFA. It could also resolve the problem where people sometimes try to change policy by opposing RFA candidates who would follow or enforce a particular policy rather than try to get an RFC to change that policy. So RFA clerks should be appointed, perhaps by election, with a remit to:

1. Remove incivil votes including unevidenced personal attacks (leaving the word redacted if there are already responses to them). Editors who had made unevidenced personal attacks would be told their vote had been removed and they could still vote, providing they did so civilly and/or with a diff.
2. Remove votes and vote rationales that are trying to change policy by opposing those who would enforce a particular policy.
3. Deal with the recent problems we had at a recent RFA where people were removing or editing their rebutted comments instead of striking them or responding to people who had commented on them.
4. When editors have !voted per another editor and that editor subsequently changes their vote, neutrally inform them of that and invite them to review and reconfirm or alter their vote and rationale.

Note there is a difference between opposing because you think a candidate is not applying a policy correctly and opposing because you disagree with the policy that they are following.

Support M

1. ϢereSpielChequers 13:49, 17 October 2015 (UTC)
2. Support, strongly. Civility must be enforced and those who make abusive comments should not be able to get their way. --Biblioworm 15:04, 17 October 2015 (UTC)
3. Support, strongly. It's an excillent idea! To maintain civility on RfAs I think clerking is the best idea. I really think this particular proposal will get opposes only from those who has a background of making uncivil or invalid !votes/comments at RfAs because other than those people, I don't see any problem with this. And I agree civility applies at every corner of Wikipedia, including RfA. I really hate when people says, "leave it to 'crats". Semi-real time patrolling by clerks at RfA pages will be more helpful than "leaving it to 'crats." Jim Carter 16:37, 17 October 2015 (UTC)
4. Support - having a dedicated group to keep the RFA process civil and reasonable would probably do a lot to improve the atmosphere there, bringing in more candidates. By the way, perhaps the 'crats should serve in this job? עוד מישהו Od Mishehu 18:23, 17 October 2015 (UTC)
5. Support This would make everything easier to follow, and Jim Carter makes good points about why this shouldn't be opposed. —⁠烏⁠Γ ^(kaw), 06:04, 18 October 2015 (UTC)
6. Support RfA is no exception to the rules that are civility. ^{My name is}not^dave (talk/contribs) 12:55, 18 October 2015 (UTC)
7. Support, we should have an enforced way to keep the discussion on track. This would take some of the heat out of RfA if we manage to choose wise clerks. —Kusma (t·c) 13:30, 18 October 2015 (UTC)

Oppose M

1. Civility policing has a place, but it's not at RFA, where the 'crats expect oppose vote rationales to be candid and on point. One person's candid is another person's incivil. Townlake (talk) 14:50, 17 October 2015 (UTC)
2. The community has all but rejected enforced civility, so the RfA process ought not be different from typical community interactions. RO^(talk) 18:11, 17 October 2015 (UTC)
3. Elections are flawed, so let's have more elections? I think the 'crats can think for themselves. Any clerk would essentially be telling the 'crats what they could consider. Glrx (talk) 18:56, 17 October 2015 (UTC)

   How about if the 'crats are the clerks in question? And they be allowed to consider votes removed by other 'crats? עוד מישהו Od Mishehu 18:40, 18 October 2015 (UTC)

   Uh, isn't that the way it is now? 'Crats looks at the !votes, throw out the pink elephants, and then consider the result. Truly outrageous material gets snipped before the close. The problem isn't the clerking but rather disagreements about whether some comments should be relevant. Age was a recent example; some think it relevant and others do not. A clerk won't help if the criterion is in doubt. Some !voters post vendettas, but the community as a whole is capable enough to see the merits and toss the dross in such posts. What would a clerk do? Toss all vendettas even if there is some truth in them? If the claims are unsupported, other editors quickly demand diffs. I also think that most of the community is smart enough to look at !voter bickering and ignore it. Glrx (talk) 17:02, 19 October 2015 (UTC)

4. Yes, the thing that would make RfA better is more fights about civility "enforcement". Opabinia regalis (talk) 19:56, 17 October 2015 (UTC)
5. The stuff about policy disagreement has nothing to do with civility and shows how the clerk position would be used to improperly influence the process, by discounting votes on such specious grounds.Andrew D. (talk) 20:14, 17 October 2015 (UTC)
6. No, this is a recipe for more drama than there is already. The dividing line between strong and incivil opposes can be narrow, considering that we are dealing with the attributes of individual editors, and we don't need squabbling about where that line exists. Coretheapple (talk) 13:24, 18 October 2015 (UTC)

Comments on M

• To the contrary of what Townlake said, civility applies everywhere. We have a clear standard as to what incivility is: WP:CIVIL. RfA should not be the exception, and there is no place in the policy I just linked that makes it the exception; making it the exception and permitting ABF and veiled verbal abuse is what makes the place so unattractive and nasty for candidates, especially ones that might be unperfect and have some enemies with an axe to grind. --Biblioworm 15:04, 17 October 2015 (UTC)
  • RFA is unique in that editors are asked to comment on other editors' entire history of work and temperament. Personal comments will be made. You have to allow some latitude for feelings to get hurt. The alternative would be milquetoast nonsense where bullies and intransigent oafs who are popular with the cabal get free passes through RFA because nobody is allowed to call them out. Townlake (talk) 16:04, 17 October 2015 (UTC)
    • Don't confuse effective opposes with incivil ones. Requiring diffs and civility does not stop us from opposing bullies and oafs, it just makes life difficult for bullies and oafs who want to oppose people without bothering to do research. ϢereSpielChequers 00:38, 18 October 2015 (UTC)

N. Determine granting of adminship based on consensus of pros and cons

Current general practice for determining the outcome of a Request for Adminship proposal treats the submitted votes as a straw poll. If the support percentage is below a certain threshold, then reasons are sought to adjust the percentage by changing the weight given to specific votes. This is not reflective of a true consensus process, though; it treats the RfA process as a vote.

In the real world, when a group is deliberating on a consensus decision, typically the group will determine what criteria will be used to evaluate the possible options, and the importance to be given to each criterion. The options are then ranked based on these criteria.

Each candidate for adminship has his or her specific strengths and weaknesses, and they should be weighed based on how strongly the community values each. Accordingly, the Request for Adminship process should be structured to determine consensus views on the pros and cons of a given candidate, and the relative weight that should be given to each. Once these have been determined, closers can judge if a candidate would be a net positive in the role of administrator.

By focusing on pros and cons and how they should be traded off against each other, the process is depersonalized, which reduces acrimony. In addition, much redundancy can be eliminated from the conversation, since each pro or con can be discussed in a single, consolidated thread. This will make easier for community members to participate, as it will take up less of their time and simplify following an ongoing discussion. As a result, it will be easier to get more people to participate.

Oppose N

1. Idea sounds great in the abstract, but there's a reason the votes are numbered and a consensus % range has been established. Consensus-based decision making doesn't scale; it works at AFD, but it does not work at RFA, and the process has had to evolve to account for that. Townlake (talk) 17:08, 17 October 2015 (UTC)
2. Reinventing the wheel? RO^(talk) 18:11, 17 October 2015 (UTC)

Comments on N

• Regarding whether or not the Request for Adminship process is a vote: if it is indeed deemed to be a vote, then the process should go all the way: no individual discussion of votes, and no re-weighting of votes. Make it like the arbitration committee voting: each participant simply votes support or oppose. Discussion can be held on specific pros and cons raised, again avoiding redundancy by consolidating conversation. isaacl (talk) 17:42, 17 October 2015 (UTC)
• Regarding the problems with using consensus decision-making in a large group, I fully agree and have written about the issues. It is unfortunately a catch-22: a consensus decision is required to divert from general community practice, but the problems with consensus decision-making hinders this. As long as the community wishes to continue to attempt using consensus, though, decision-making can be broken down to make it more amenable to this method. Focusing on reaching consensus for more concrete, smaller scope items is easier: is person A strong in X? is person A weak in Y? isaacl (talk) 17:25, 17 October 2015 (UTC)
• Regarding the comment on reinventing the wheel: since this Request for Comment page is soliciting views on problems with the Request for Adminship process, I think it's within its purview to discuss how the process is problematic in trying to fit different purposes. It is structured partially as a vote and partially as a multi-threaded discussion, which does not do a good job at supporting either a straight vote or a consensus-based decision. And the way votes are re-weighted now places too much emphasis on arguing with individuals to change their vote, instead of focusing on the pros and cons of the candidate. isaacl (talk) 18:24, 17 October 2015 (UTC)

O: The discretionary range is too narrow

The discretionary range at RfA, which is roughly 5% (70–75%) is too narrow. In the spirit of making RfA based upon true consensus, we should widen this range and give bureaucrats more discretion to weigh arguments.

Support O

1. Support. I think this is good alternative to lowering the bar; instead, we give less importance to percentages and structure RfA more in the spirit of WP:CONSENSUS. The policy (yes, it is a community-approved policy which has gained consensus itself) states that discussions are to be closed according to weight of the arguments, but in RfA this only occurs within a very narrow 5% range. This is not true consensus, but is rather 95% voting with a tiny window so that it can be called a "!vote" (not vote) and "discussion". Really, we are deceiving ourselves and called RfA a discussion when it is, in fact, mostly a vote. Do we want this? People repeatedly assert that Wikipedia is not a bureaucracy, and voting is a feature of a bureaucracy, so I think we need to make RfA much less about voting and percentages and more about the strength of arguments. (WP:NOTDEM is also applicable, which clearly states that decisions are to be reached by discussions and not votes.) Both of these are also policy. Note that this is not about "lowering the bar", but rather permitting wider discretion. See this comment for more detail on this topic. In summary, this idea is thoroughly supported by policy. --Biblioworm 18:52, 17 October 2015 (UTC) Edited @ 18:56, 17 October 2015 (UTC)
2. Support per Biblioworm, especially the comments below. —⁠烏⁠Γ ^(kaw), 06:05, 18 October 2015 (UTC)
3. Support Very good points. --Fauzan^{✆ talk}_{✉ mail} 11:26, 19 October 2015 (UTC)
4. Support 5% is an extremely tiny margin and turns the "discussion" into a vote in the vast majority of cases. Decision-making processes on Wikipedia are, as a rule, not votes. This gets even worse if the bureaucrats interpret RfAs towards the lower end of the margin as being unlikely to pass, and ones to the higher end as likely. Hut 8.5 21:59, 19 October 2015 (UTC)

Oppose O

1. We just recently narrowed the range from 70-80 to 70-75. I should know, I'm the one who introduced the edit to the page. This suggestion is just a sad attempt to backdoor in a lowered passing threshold. Cute. Townlake (talk) 22:24, 17 October 2015 (UTC)
2. No to discretion, no to weighing arguments, no to super-votes. Everyking (talk) 14:48, 18 October 2015 (UTC)
3. Oppose per Townlake. The problem isn't the threshold, it's the unsuitability of some of our recent candidates. RO^(talk) 21:14, 19 October 2015 (UTC)

Comments on O

• Cute? I did not suggest that we lower the bar for passing, I suggested that we give bureaucrats more ability to do what policy says closers of discussion should do. (Which is namely to weigh arguments according to strength: [1][2]) An exemption for RfA is not included in any of those policies. We can start small; sudden, sweeping changes are not required. How would expanding the discretionary range by a little to, say, 65% or so (a mere change of 5%) cause any fatal harm to the project? We could experiment and see how it goes; we'll never know until we try. I do not see, Townlake, how defending clearly worded community-approved policy about consensus is in any way "cute". This is perhaps one the of the suggestions that has the most policy-based support on its side. --Biblioworm 00:09, 18 October 2015 (UTC)
• 'Crats can and in very rare cases should WP:Ignore all rules and, after a serious and sober 'crat-chat, reject or re-list/extend the discussion for candidates with over 75% or even over 90% support if circumstances warrant (usually this would be a late-breaking "show-stopper" evidence that would result in a desysop or ARBCOM hearing or, if known at the start of the RfA, would have doomed the nomination anyway). Likewise they can and in very rare cases should follow that same policy and promote or re-list/extend the discussion for those with below 70% or (in the face of strongly suspected-but-unproven large-scale sock/meat-puppetry and/or unproven-but-strongly-suspected off-wiki canvassing or similar abuse-of-process to stack the !votes) even below 50% if it's clearly the right thing to do. By "very rare" I mean "I wouldn't be surprised if such things haven't happened at all in the last few years, and I wouldn't be surprised if they don't happen in the next few years either." davidwr/(talk)/(contribs) 04:19, 18 October 2015 (UTC)
• Townlake, you probably mean you just recently edited the documentation to reflect the fact that the de facto discretionary zone is 70-75% and has been so since at least 2008. We had a case of documentation lag, not a recent substantive change. Opabinia regalis (talk) 04:34, 18 October 2015 (UTC)
  • Yes, the change was to reflect a reality that the community had clearly already accepted. Townlake (talk) 06:05, 18 October 2015 (UTC)
• I am unsure about this one. It really depends on how the bureaucrats interpret their jobs. Recently they have seemed to take a fairly conservative approach to promotions, which means that arguments are only looked at when the numbers fall into their narrow "discretion" range, so it might be useful to tell them their discretion range should be larger. (For example, I think it would have been entirely defensible to promote Thine Antique Pen, as there were quite a lot of purely age-based oppose votes). Also, bureaucrats exercising unexpected discretion has historically been a way to prolong the flamefest that some RfAs are for another week, and I can see why bureaucrats would not want to do that. But then I also feel that the "strength of argument" thing is so inherently personal and hard to measure objectively that pure voting would actually be preferable, and then there should be no bureaucrat discretion at all. —Kusma (t·c) 09:55, 18 October 2015 (UTC)
• The problem isn't the threshold, it's the unsuitability of some of our recent candidates. First of all, as I have clearly stated several times, this is a proposal that is supported by policy. It shouldn't matter whether or not you think it's a problem; our policies say that our current system is a problem. (Not explicitly, of course, but considering the extremely vote-based nature of RfA it's plain that it is not in line with the procedures for determining consensus described in the relevant policies.) Secondly, why were all the failed candidates unsuitable? Because they failed RfA, and therefore they were unsuitable? That would be a circulus in probando fallacy, because it reasons that the candidate must be unsuitable because they failed RfA, while ignoring the fact that the very point of discussion is that RfA is not working properly. That's defending RfA with its own results, although there is widespread consensus that something is wrong with the process (there may not be widespread agreement on what is wrong, but there is widespread agreement that something is wrong). Therefore, your argument defends a broken system with its own output. How do we know that the candidates were unsuitable? Might they not have been good admins, if they were given the chance? Pile-ons in the oppose sections are not uncommon, so surely some RfAs have failed when they shouldn't have. Why should you have the authority to say that recent candidates were unsuitable? Given my argument just now showing the flaws in it, don't I have just as much right to say that many of our failed candidates were not unsuitable? In summary, there are many problems, both with policy and logic, in that one sentence. --Biblioworm 22:19, 19 October 2015 (UTC)
  • I think you're going a bit too far in your incessant badgering of commenters here, particularly when you overanalyze their statements to the point of calling them fallacies. Don't forget that to invalidate every oppose comment by calling them fallacious is in itself a fallacy (the fallacy fallacy). For example, it's not faulty logic to say many of the recent candidates failed RfA because they were unsuitable. Many of them were unsuitable. I didn't say or intend to imply that every single one that failed was demonstrably unsuitable, which is where you had to go to invalidate my input. A major reason RfA I so confrontational is the badgering of editors who dissent, just as you've been doing here. I suggest you take it down a notch, and let this unfold as it should. RO^(talk) 22:36, 19 October 2015 (UTC)
    • How is simply calling a fallacy a fallacy a fallacy? (*head spinning*) A "fallacy" which says that it is incorrect to question an argument because it contains a circular fallacy (one of the worst and most obvious of all fallacies) is itself fallacious. But you're simply reiterating your same argument. Why were they unsuitable? Because you decided they were? And I'm also not a believer in "badgering" at RfA. Ones who oppose candidates should have to explain and clarify their rationales appropriately when questioned. In fact, I think a major issue at RfA is actually the stigma that comes with asking questions of opposers. I apologize if a can sound too aggressive at times, since that is never my intent (it's difficult to convey tones through a text medium), but I'm a very logically-minded person and spend a good part of my life dealing with it and analyzing the validity of propositions, so perhaps it spills over a bit more than it should here. --Biblioworm 22:54, 19 October 2015 (UTC)

General comments

• What a mess of a discussion. If one is truly trying to design a survey, this is a poor effort. Carrite (talk) 05:43, 19 October 2015 (UTC)
  • Well, maybe the discussion could help Biblioworm and the other folks interested in actual change to design something useful on Survey Monkey or something... ? Montanabw^(talk) 07:23, 19 October 2015 (UTC)
  • One day, the community will have to realize that consensus doesn't work well with a large group, and even under the best of circumstances, consensus requires patience. So yes, if everyone wants to be able to weigh in, it's going to be a lengthy, verbose conversation that takes time to work through. And unless the participants agree to follow a moderator's direction, it's also going to spread out into multiple parallel branches. For better or worse, most of the people who are active in these threads seem to prefer this form of discussion over various other formats that have been tried or proposed. isaacl (talk) 19:01, 19 October 2015 (UTC)

Recap

Collect added this section immediately after this revision of 11:26, 19 October 2015 (UTC) and reflects the state of the discussion as of that time. davidwr/(talk)/(contribs) 20:49, 19 October 2015 (UTC).

OK -- the following would seem to be a fair recap of all the mass of material above:

1. RfA tends to be viewed as a "hostile environment" for candidates.
2. More participants in RfA discussions would improve the process, we hope.
3. The current process is not ideal at all, so it should be changed.
4. Any person who expects to do everything we expect and enable of administrators would face an impossible workload, and we should consider making some of the routine tasks be done at a bot or software level.

As far as I can tell, nothing else above gets overwhelming support at this point. I rather suspect that starting from these generally accepted points would work better than belabouring new issues - four is more doable then umpteen discussions at a time. Collect (talk) 11:41, 19 October 2015 (UTC)

Yes, the focus should be on RfA reform (specifically reforming the process to make it less hostile, and possibly encourage a wider voting pool), and nothing else. Your #4 is a separate issue, and should be split off, and handled separately (possibly at WT:Administrators). --IJBall (contribs • talk) 18:45, 19 October 2015 (UTC)

I sought to note each and every topic which has broad consensus above - not to say on my own that any were unfeasible or unrelated to RfA reform. Collect (talk) 19:12, 19 October 2015 (UTC)