Wikipedia talk:Requests for adminship

From Wikipedia, the free encyclopedia
Jump to: navigation, search
RfA candidate S O N S% Ending (UTC) Time left Dups? Report
RfB candidate S O N S% Ending (UTC) Time left Dups? Report

No RfXs since 21:13, 30 August 2015 (UTC).—cyberbot ITalk to my owner:Online

Latest RfXs (update)
Candidate Type Result Date of close Tally
Wbm1058 RfA Successful 30 Aug 2015 121 17 3
Doniago RfA Withdrawn 22 Aug 2015 14 17 1
Ceradon2 RfA Withdrawn 21 Aug 2015 54 27 11
Nrwairport RfA WP:NOTNOW 17 Aug 2015 0 7 1
Cyclonebiskit2 RfA Successful 11 Aug 2015 121 12 9
Liz RfA Successful 6 Aug 2015 200 72 9
Primefac RfA Withdrawn 30 Jul 2015 12 17 7

Current time: 01:34:26, 31 August 2015 (UTC)
Purge this page

RfC: Is RfA a broken process?[edit]

In March 2011, Jimmy Wales stated that RfA was "a broken process". There has since been much discussion about fixing RfA, and some appear to believe that it no longer needs any changes and that Wales' comment is no longer applicable. For those who might comment on such things, I am well aware of the work done at RFA2011, but it has been admitted that no formal RfCs were ever presented to the community as a result of it. RFA2013 also did not appear to include surveys of this nature. (I might be wrong, but in any case it doesn't hurt to get an updated opinion from the newer members of the community.) Before we can ever hope to proceed with any sort of serious reform, the community as a whole must first formally show reasonable agreement on the most fundamental underlying principle of any reform: Is RfA really a broken process (i.e., does RfA need to be reformed or even revolutionized in some manner)? This RfC is not for discussion on methods by which this might occur. I will emphasize this: This RfC is only for the purpose of reaching a consensus on whether or not RfA needs changing in some manner. It does not matter how that might occur at this stage. If this RfC gains traction after a day or two, we might consider adding it to WP:CENT and on the watchlist notice. --Biblioworm 20:02, 5 August 2015 (UTC)


  1. Support. The expectations at RfA are currently much too unreasonable. Almost complete perfection and balance is required to obtain the bit. --Biblioworm 20:02, 5 August 2015 (UTC)
  2. Support. Is it broken? No, the fundamental process of RfA is sound. Is it currently working properly? Also no, and the reason appears to be that the stakes of granting admin tools for life is perceived to be quite high. It can be made better by reducing the perceived stakes. ~ RobTalk 20:06, 5 August 2015 (UTC)
  3. Support. There have been no real changes. The number of active admins continues to decline, because only a few experienced editors, who have the perfect balance between article creation and other processes, will pass. As a result, the number of active admins continues to decline. Something needs to be done. Wales' comment is just as true as it was in 2011. My own thought is that part of the problem is that some editors have enemies lists and will vote against editors who have made their enemies lists, and that this excludes editors who have been bold, but lack of boldness is seen as lack of productivity, but that is only my opinion. The process is broken. I am not convinced that it can be incrementally reformed. At the same time, I do not favor the occasional idea to delete it and leave the English Wikipedia with no RFA process to provide an incentive to develop a new. A significantly new process needs to be proposed as an alternative to the current broken one. Robert McClenon (talk) 20:13, 5 August 2015 (UTC)
  4. Support per my comments in Liz and Cyphoidbomb's recent RFAs. Not only are the expectations of the community unreasonably high to promote enough good admins to stem attrition, the community's expectations vary widely from candidate to candidate, from day to day, and from !voter to !voter. I've frequently compared the process to running the gauntlet or trial by ordeal, and the result here is that editors who would make good admins don't step up, and those that do get shot down seemingly more often than not, and for silly reasons. Some editors have suggested below that this RfC is pointless if we're not going to discuss solutions (again) but I disagree: periodically restating the community's dissatisfaction with the process is a step in correcting it, and I believe there's no process more universally viewed as broken than RFA at this time. I'm generally with Robert McClenon that the current process can't be fixed incrementally, but I have been willing in the past to support oddball proposals to try. To memory, none have been implemented regardless of support, and the community's reluctance to try to fix it is as much a problem as the process itself. Ivanvector 🍁 (talk) 20:31, 5 August 2015 (UTC)
  5. Support When I see reasonably decent editors end up in endless conversation about whether or not there is a consensus, despite a 74% support, I can't help but wonder what the meaning of consensus really is? We have a system where the reasoning behind the !vote seems to count less than the tally, and where the !novote seems to hold 3x the weight of the !yesvote. In most systems I'd say that if both sides make reasonable arguments, but neither is overwhelming, but the !vote is 70%+ in favor that this is indeed consensus. AliveFreeHappy (talk) 20:42, 5 August 2015 (UTC)
  6. Speaking as one of the coordinators of RfA2011 (which was prompted by Jimbo's comment and similarly shameful conduct at RfA), I think it's widely agreed upon and understood that RfA is inherently dysfunctional to some extent and has been for years. However, based on my experience, there's not even any point in trying to introduce reforms through the RfC process. There's simply too much backlash and trolling from the vocal minority of dissenters who want to keep it as difficult as possible for the administrative corps to expand. The pro-reform advocates are unorganized and not as motivated to push back because every bit of fuel added to the fire makes it more and more difficult. Even with a massive task force behind RfA2011, we simply couldn't get them to come out and support our actual proposals or help improve them. No one wanted to do the work. Even the WMF supported our project but was unwilling to implement even the most moderate and minor improvements from the top down. Those of us who tried to work on reforms were derided by the trolls until we simply gave up. I'm now of the opinion that RfA reform is a hopeless endeavor and that the only thing we can do is stand up as individuals to those who disrupt the process as much as possible. If they want to make it worse, they have to deal with the people who call them out on it. Swarm 20:50, 5 August 2015 (UTC)
  7. Support Along with what Swarm and others state above, I have limited experience but it appears that this process is failing and needs to be addressed. This project is too large with too many things needing done and not enough qualified people to do them. In case your wondering how I stumbled onto this niche page as such a new editor I have been watching the recent changes button to learn what types of edits are being done to learn how to do things here. This page keeps popping up. This is clearly not a Noob area so I wanted to set your minds at ease. :-) RingofSauron (talk) 21:11, 5 August 2015 (UTC)
  8. Support (but with a caveat) – I fully agree that the RfA process is "broken". I find it to be a mean-spirited, dehumanizing hazing process for the experienced editors who run as candidates (ironically, I'd argue that the WP:NOTNOW and "Withdrawn" cases generally have an easier time of it – but I found the last 3 RfA's then have gone to "Crat Chats" to be generally beyond the pale in how they went down...). All that said, I'm not sure that the RfA process can be fixed "in isolation" of also tackling the general concept of "Adminship", which I am increasingly thinking is an antiquated position that should have been wound down, c.2010–12. In other words, I don't think "RfA's" can be fixed without also "fixing" the very idea of who should run this place (and how). So, while I support anything that takes a hammer to this noxious RfA process, I am dubious of its success if it simultaneously ignores the need to reforming the governance structure of this entire Wiki. In other words, this is a much bigger problem than just "RfA's". My $0.02, FWIW... --IJBall (contribstalk) 22:02, 5 August 2015 (UTC)
  9. Generally agreed with IJBall is who is more concise than I am.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  00:41, 6 August 2015 (UTC)
  10. Support. Swarm and IJBall hit the nail on the head. Additionally, the lack of editors promoted or even trying to go through the process speaks for itself. Calidum T|C 02:06, 6 August 2015 (UTC)
  11. Support. Even though I opposed Liz's RfA, I'm glad the crats found that there existed consensus to promote, as I felt that some of the other opposes were quite weak, mean-spirited, and/or bad-faith. IJBall puts most of my concerns much more eloquently than I could. I still think unbundling of the tools is the ideal solution. Since candidates can have a need for some of the tools while having no need for others, and the requirements for different tools differ so substantially from each other that the all-or-nothing procedure we have now sinks candidates who would do well with and need one or more tools but not the whole package. StringTheory11 (t • c) 04:07, 6 August 2015 (UTC)
    Exactly. For example, all I would want of the toolset (at least, right now) would be the equivalent of "article mover" tools, to help out Jenks24, Fuhghettaboutit, et al. Why should I have to run through an RfA "gauntlet" for the "full" toolset, when all I want is a subset of the tools?! --IJBall (contribstalk) 04:48, 6 August 2015 (UTC)
  12. I disagree with the term "broken", however the process could use improvements. I would like to see the community come up with clear criteria and that like everywhere else on Wikipedia votes should be weighed based on their basis in that criteria. Right now it is "I like" and "I don't like", imagine if we ran AfD or TBAN discussions that way? Chillum 04:44, 6 August 2015 (UTC)
  13. RFA has become dysfunctional because it no longer works for the community we now have. As Chillum has pointed out we need to agree a criteria for adminship, the drawback of the current system is that it takes less than 40% of the community to impose an additional or more stringent criteria for adminship. This would be problematic even if there were only one criteria that was ratcheted up by this effect, but it happens on things as diverse as edit count, tenure, time since last block and area of goodfaith contributions. If we take the analogy of a job interview, an interview panel would decide whether a clean driving license was required when they drafted the job ad, they wouldn't argue the criteria during the interview. ϢereSpielChequers 11:00, 6 August 2015 (UTC)
  14. Support The process is ridiculous. We seem to expect candidates to be able to hit the ground running and ready to perform 100% perfectly the moment they get the tools. Unless we think a candidate is likely to intentionally misuse or is so green they'll make tons of work for others correcting their mistakes, admin tools should be WP:NOBIGDEAL valereee (talk) 19:37, 8 August 2015 (UTC)
  15. Support. Everyking (talk) 18:12, 10 August 2015 (UTC)


  1. I find it odd that I'm here, but I have to oppose because I feel the RfC is fundamentally flawed. "Something is wrong with RfA" is a perennial high level statement and may well be true. Unless you can get consensus that something specific is wrong with RfA (eg, the votes are too harmful, the wrong people are getting in etc) then the statement is meaningless. The reason I find myself at Oppose, rather than Mu (below) is that I believe that a consensus on the statement "Something is wrong with RfA" would actually be harmful, as the interpretation will be personal and therefore lead to many different discussions and lose any chance of a coordinated discussion with a positive outcome. Finally, at a very high level, RfA is not broken, it keeps the bad candidates out and lets the good candidates through. It shows that admins have the trust of the community. It can be improved I'm sure, but no, it's not broken. WormTT(talk) 09:28, 6 August 2015 (UTC)
    I don't intend to "badger", but saying that RfA "keeps the bad candidates out and lets the good candidates through" is a completely relative statement. Who decides what candidates are "good" and "bad"? Candidates we thought were bad might have been good admins, and it might be an offense of a sort to good-faith failed candidates, since it implies that they were "bad". It also doesn't keep all the bad apples out, and unless we make it impossible to pass it never will. --Biblioworm 15:14, 6 August 2015 (UTC)
  2. Oppose I agree with Worm That Turned. It is trivially easy to get people to say there's something wrong with RfA. I'm not sure of the intent of this RfC. Getting everybody to agree there's bad things about RfA does nothing. I think most, if not all, of us have a pet peeve about RfA and see something that we think could be improved. We already know this. I mean no disrespect to Biblio with this. Paraphrasing Churchill "RfA is the worst form of selecting admins...except all those other forms that have been tried". I argue that RfA is the best system we have, and it is flawless...until we produce a system that is clearly superior. To date, despite many years of effort, despite dozens if not hundreds of proposals, despite the combined efforts of hundreds of contributors to RfA, no system has come forward that is better. That, to me, is the purest testament that RfA is perfect as is. Think I'm crazy for saying that? Ok, produce a system that works better than RfA. --Hammersoft (talk) 12:30, 6 August 2015 (UTC)
  3. I see this as misattributing the problem and focusing on wrong issue. "RfA process" as it is today simply reflects the fact that it grants very wide toolset, including power to block editors and delete articles, for lifetime, and requires a full arbcom case to revoke. It is infinitely easier to err on the side of caution during RfA, and prevent any even somewhat uncertain candidates from getting it, then try to fix the mess afterwards if an asshole or incompetent made it through. While I personally believe that radical changes, like unbundling toolset or introducing term limits, could be a good idea, having seen previous RfA reform attempts I don't believe that consensus for such changes can be achieved. At best some minor cosmetic changes will be agreed, and those are not going be really relevant in big picture.--Staberinde (talk) 15:19, 6 August 2015 (UTC)
  4. RfA is not broken, it's not ridiculous, and it does not expect candidates to hit the ground running. What it needs however is bringing into line with the needs of Wikipedia today. Everything about Wikipedia peaked in or around 2007 and has been dropping off since but there are logical reasons for this which are not the sbuject of this discussion. RfA does what it is expected to do: decide whether or not a candidate should be granted the tools. With very few exceptions which certainly are not a sign that anything is fundamentally wrong with the process, it actually does this quite well. There is a vast difference however between a highly participated RfA that falls into the discretionary zone but where the participants have been exceptionally polite, and one that passes with flying colours but where the candidate has been called a liar, an obtuse jerk, and a child hater. The problem is not with the system but one of the people who are allowed to vote on it and in such a manner that the process has become rather unpleasant - but fortunately not quite as much as in the days when when it was perfectly permissible to voice the nastiest of PA (that users would be blocked for anywhere else} and candidates were told to shut up and put up if they even dared to defend themselves aagainst inappropriate votes. It's the voters who need to change and unless their behaviour changes for the better, nothing will change in the forseeable future. --Kudpung กุดผึ้ง (talk) 10:44, 11 August 2015 (UTC)
  5. RfA can be viewed as broken, but if the community has not came up with a partial or full solution to this problem in 10+ years (no deadline, no excuses), then it can be concluded that the process itself is not broken. The problem is not the process; the problem is the editors participating in the process. When people talk about the fire and brimstone of RfA, they really should talk about the fire and brimstone thrown by (some) of the participants at RfA. The real solution is to educate editors about how to vote at RfA, deter incompetence from hitting the shores of RfA, and preventing trolls from editing Wikipedia. Esquivalience t 00:57, 14 August 2015 (UTC)
    @Kudpung and Esquivalience: Yes, I agree with you that the primary problem is the attitudes of some participants, not the system itself. It's axiomatic, in fact, as far as it goes. The problem with "the participants are the problem" observation is that it is a disease with no known cure. If there is one thing I have learned in 20 years of drafting governing documents for business organizations and not-for-profits, it's that you can design your governing system for the people, but you cannot redesign your people to fit your ideal governing system. The Russians tried that; it was called the New Soviet Man. I think most of us know how that little experiment in social engineering turned out. In the short and intermediate runs, we don't get to pick or even educate our RfA participants, but we can fashion a system of community norms, expectations and, yes, written guidelines to restrain some of the more egregious RfA behaviors. Arguing whether RfA is broken or not is academic; it has real problems and most serious observers of the process know it. Dirtlawyer1 (talk) 01:45, 14 August 2015 (UTC)
    @Dirtlawyer1: The goal of educating voters and encouraging appropriate voting is less "get every voter to be a certain archetype" than it is to ensure that voters participate properly and orderly. Although that I agree that RfA still needs some changes (more order, specifically); it is still necessary and within our ken to educate voters - making new guidelines and written norms and expectations is actually one way of education, impose limits on voting, and give bureaucrats ample discretion to discount trolls. We can't fully cure the problem, but we can alleviate it. Esquivalience t 02:14, 14 August 2015 (UTC)
    @Dirtlawyer1:, @Esquivalience:, I've written thousands of words on this issue since I started WP2011 and Worm echoes them perfectly with his own observation above. It all boils down to simply requiring objective voting and an acceptable level of decorum at RfA. My mantra has always been "Fix the voters, and RfA will fix itself." The only way to start is by either introducing minimum qualifications for voters as is done on the other major Wikipedias, or systematic clerking. Preferably both. Kudpung กุดผึ้ง (talk) 23:15, 14 August 2015 (UTC)
    @Kudpung: I agree that new RfA "decorum" guidelines must be part of any systemic fix of the sometimes free-for-all atmosphere of RfA. That's a big part of what I meant by "a system of community norms, expectations and written guidelines." As for the minimum voting qualifications for participants you suggest, I am much less sanguine about their ameliorative impact. If you review the wildest RfAs from the last 3 to 5 years, you will find that some of the harshest comments and openly hostile campaigning come from long-time editors (and some administrators) who will easily satisfy any reasonable minimum voter experience standards. I am happy to provide examples outside an on-wiki forum -- but you quite probably remember them from having been there. The harshest comments and open campaigning are not the province of newbie editors. Cheers. Dirtlawyer1 (talk) 00:20, 15 August 2015 (UTC)
    @Dirtlawyer1:, Swarm describes with penetrating accuracy in his support vote above the reason why discussions, research,and proposals of this kind fail. The need for such measures is actually twofold: Yes, correct in that generally the nastiest of comments come from older users who are bold enough to believe that RfA is one venue where they can be as toxic as they like with impunity, and this is the part that needs systematic clerking, while introducing minimum voting requirement would stop a lot of the trolling, sour grapes from block evading IPs, SPA, amd new, but perhaps nice users who are still totally ignorant of what adminship is all about. Kudpung กุดผึ้ง (talk) 01:25, 15 August 2015 (UTC)
    I agree that the most caustic and personally disparaging comments in an RfA come not from newbie editors (who usually just vote Yea or Nay or write brief rationales for their votes) but from veteran editors who have serious concerns about who is or is not qualified to become an administrator. Most casual editors are not that invested in specific admin criteria and just go with their gut feeling. This is just my impression though, it would be interesting to see the data on recent RfAs (say, the past five years). Liz Read! Talk! 02:24, 15 August 2015 (UTC)


  1. Mu. RfA will always suck, in exactly the same way that the weather is always lousy and traffic is always terrible and everyone's boss is a jerk. It's simple, it's salient, everyone has an opinion to share when the topic comes up, and it's a long-standing socially acceptable way to complain. Holding an RfC about whether we should do "something" to make it suck less is a lot like holding an RfC about traffic. Everyone says yes, I support, traffic is the worst! Only after there is "consensus" on whether to do "something" about traffic do we discover that we don't even agree about which side of the road to drive on. Endless time-suck RfCs are not going to help; the only thing you can do about the fact that traffic sucks is recognize that you are traffic. And frankly, some of you drive vote like jerks. Opabinia regalis (talk) 22:25, 5 August 2015 (UTC)
    Every voter at RfA is quite certain they aren't the jerks. Similarly, every proposer of an idea at RfA is quite certain their idea is the answer. Of course, no one really knows what the question is. The only certainty is that the inventor of RfA's idea to create RfA has created a lot of kerfluffle and has generally been regarded as a bad idea. So, since it is impossible to replace RfA, let's wrap things up and have some breakfast. --Hammersoft (talk) 13:08, 6 August 2015 (UTC)
    If someone spontaneously decided to invent it, then let someone spontaneously decide to close it and boldly invent a new way electing admins. ;) I don't think it's fully "impossible" to change things (although indescribably difficult), since it was once possible in ye olde Dark Ages. --Biblioworm 17:07, 6 August 2015 (UTC)
  2. (「Mu」は日本語で「無」、けど...) This strikes me as the real world equivalent of a job interview. Does everyone agree they're awful, yes (my RfA went about as well as anyone's, and of the many adjectives I'd use "enjoyable" isn't one of them). Does everyone want to do something about it, yes. Does anyone other than the creator of a particular solution want to try said solution, no. As long as I have a moment, I'll say something that's probably politically incorrect but that at least one admin needs to say. I have absolutely no intention whatsoever of making myself open to recall in any way (obviously I'd have to answer to ArbCom if it came to that, and I have in the past). I'm not beholden to anyone to do what I most want to on Wikipedia or to be popular, and I want nothing to do with a process which would change that. Given how toxic everyone seems to perceive RfA, which I think is not at all without cause, I have no idea why anyone would have confidence that a chance to forcibly remove adminship from someone would be 1. somehow less contentious than an RfA or 2. make RfA into a better environment. If anything, it will lead to people obsessively tracking admins for a chance to pounce and make our Wikipedia experience as miserable as possible. For people who think this can be prevented and/or patterns are easily recognized, I invite you to look through AE for a while and see how often the lines are extremely murky. Whenever this gets to be too much for me I do some actual writing, instead of whining and bitching about how no one else does, and my topics of choice (especially my current one, but last ones too) shake things back into proper focus. The Blade of the Northern Lights (話して下さい) 00:29, 6 August 2015 (UTC)
    RfA is not like a "job interview". Beeblebox's (I think it was Beeblebox's...) proposal for a "panel" to evaluate candidates is a lot closer to the "job interview" idea – which is why, failing the kind of systematic overhaul that's needed, Beeblebox's proposal for "RfA reform" is the idea I'm coming around to. --IJBall (contribstalk) 02:10, 6 August 2015 (UTC)
    I mean that it's similar to a job interview in the sense that as a candidate you're essentially subjecting yourself to the will of a person/group of people, and those people are in turn trying to figure out your fitness based on that very limited interaction. There's no way I know of to make a process like that feel smooth in any way. How the process itself relates to RfA is based on your experience, at least in mine I had a successful job interview that certainly felt similar (and I was on a board that conducted one, so I know it from both sides). The Blade of the Northern Lights (話して下さい) 02:38, 6 August 2015 (UTC)
  3. What they said. ^^^ GregJackP Boomer! 00:30, 6 August 2015 (UTC)
  4. RfA isn't perfect, but this does nothing to fix it. Actually, the events of yesterday probably took a spanner to what remained working and gave it a good few thwacks.--Wehwalt (talk) 02:45, 6 August 2015 (UTC)
  5. This RFC is a complete waste of time. Townlake (talk) 04:10, 6 August 2015 (UTC)
  6. I am traffic. Andrevan@ 15:07, 6 August 2015 (UTC)
  7. Not again, surely? Hasn't this been discussed to death already?  Philg88 talk 15:51, 6 August 2015 (UTC)
  8. RFA has problems, sure. But getting a mandate to impose a change without revealing the change is an obvious way to end up with a worse process. WilyD 10:54, 11 August 2015 (UTC)


I'm not sure I see the point of this discussion - it isn't proposing any specific change, and even if we were to all agree that the current system is 'broken' (a vague description), it is no indication that we will ever agree on any specific alternative. Or even over how it is 'broken'.AndyTheGrump (talk) 20:07, 5 August 2015 (UTC)

This has been done to death and the consistent consensus over the past few years is that change is wanted. I see no reason to have another Forum on the Potential for the Desire of a Future Working Group to Establish Community Involved Nondescript Change. GraniteSand (talk) 20:09, 5 August 2015 (UTC)

RfA is a living ever-changing process, as Wikipedia is an ever-changing entity. It can, and should, always strive to adapt and improve so as to better support the project. Therefore, I too don't see the net benefit of this RfC. We can all accept that RfA should improve—HOW is a much more important question than IF. -- Avi (talk) 20:12, 5 August 2015 (UTC)

I'm with GraniteSand. We might as well make this RfC "Who Doesn't think RfA is a broken process?" because that would be interesting to hear from people. The RfA is one of Wikipedia's biggest punchlines, going on longer than VisualEditor. What else is there to say but not saying anything and just do something about it? GamerPro64 20:25, 5 August 2015 (UTC)

Actually, this RfC does include that (e.g., the "oppose" section), and I agree that it will be interesting. --Biblioworm 21:01, 5 August 2015 (UTC)
Good point. Though I'm more inclined to wanting stuff to happen than just voting to show that I want RfA to be less shit. GamerPro64 21:04, 5 August 2015 (UTC)

I of course knew as a fact that there were going to be comments like this. However, I have no intention of withdrawing it, because there's no harm in having it, and although it doesn't propose any specific change, I feel it would be useful as a reference, so that it can at least be used in a baseline "proof of concept" sort of manner, if or when any sort of serious reform project starts again. --Biblioworm 20:58, 5 August 2015 (UTC)

So we're we're having a meeting to discuss the next meeting? Wow. §FreeRangeFrogcroak 18:16, 10 August 2015 (UTC)
Precisely, FreeRangeFrog, the irony is that irrespective of which section they are placing their comments, they are all saying that we should have that meeting. That said, this RfC is as valuable as a debate to confirm that rock is hard and those who complain loudest are often the ones who have put RfA between a rock and a hard place. Kudpung กุดผึ้ง (talk) 23:37, 14 August 2015 (UTC)

Arbcom to take over RFA?[edit]

Random thought, if Arbcom is clearly only ones suited for removing adminship (as many seem to believe), then maybe they would be also best suited for granting it? That way both sides of process would be responsibility of same body, unlike current unbalanced system where only community can give and only Arbcom can remove.--Staberinde (talk) 17:09, 6 August 2015 (UTC)

Aside from Arbcom being busy, and many desysops requiring non public info, this makes as much sense as combining the driving test examiners who give out driving licenses with the magistrates who take away the licenses of drunk drivers. RFA gives out mops to trusted clueful editors who are ready to use the mop, arbcom takes it away from those who have lost the community's trust, there is little overlap in this, except for former admins desysoped for cause. I can see some logic in giving arbcom the job of reappointing former admins desysopped for cause - but I'm pretty sure they won't want the role, and for perhaps illogical reasons neither I nor I think the community would be happy with such a change. ϢereSpielChequers 19:20, 6 August 2015 (UTC)
I find your analogy a bit flawed as driving licence is basically just a permit, while adminship would be better compared as a job or assignment. My own analogy would be that currently admins get voted in by masses like they were politicians, but they have no term limits and can be recalled only by higher body like they were simply specialists assigned to job. I suspect that system that goes full way to politician or specialist route would be more balanced. Wikipedia has examples of both, Arbcom are basically elected like politicians and have term limits, while on other hand Checkusers are specialists appointed by arbcom and also recalled by arbcom. Not saying that making current Arbcom appoint admins tomorrow would be best solution, but I do think this aspect deserves consideration in relation of RfA problems.--Staberinde (talk) 16:27, 7 August 2015 (UTC)
Er... adminship is exactly like a permit, and in no way like a job or assignment. That I am an admin means I can access a couple of extra functions in the software, nothing more, nothing less. I am just as much a simple volunteer helper here as everyone else, and actually there are many non-admins who command more respect (or hold political power, if you will) than me. —Kusma (t·c) 19:02, 7 August 2015 (UTC)
So far as desysopped admins go, that actually did used to be the process. Desysops were either temporary and adminship automatically restored once they expired, or (as discussed here) adminship could be restored by appeal to ArbCom. There's really no reason for that to have to happen, though. The community will certainly be made aware that someone has been previously desysopped at a future RfA, and each participant can decide whether with knowing that they're prepared to trust the candidate to regain the tools. That ultimately should be a community decision. Seraphimblade Talk to me 19:09, 7 August 2015 (UTC)
@Staberinde Political elections are about what changes a politician might make such as policies they want to change. Political voting is about whether one likes the individual or the party/faction/set of policies they represent. RFA is more like a driving license in that people assess a competency and that means sometimes opposing friends who are not ready and supporting past opponents who are. ϢereSpielChequers 19:39, 7 August 2015 (UTC)
If that were to be done, adminship would become, or at any rate become seen as, a function of government and not as a community post. Giving an elected body control over the admin class would end effective community scrutiny of candidates and put what authority there is on Wikipedia into ArbCom's hands. See User:Ohconfucius/Userboxes/Arbcom, not Govcom. BethNaught (talk) 19:54, 7 August 2015 (UTC)
  • Oppose- ArbCom's powers should be curtailed, rather than expanded. Reyk YO! 19:59, 7 August 2015 (UTC)
  • LOL Candidates could die waiting. Popes are elected in less time than it takes Arbcom to pontificate. Leaky Caldron 20:47, 7 August 2015 (UTC)
  • Oppose. It's a separation of powers problem.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  11:18, 9 August 2015 (UTC)
I would note that this definitely wasn't intended as concrete proposal worth support/oppose voting, but just an idea that popped in mind in relation to discussion at previous section. Someone else decided that this should be moved to separate section though.--Staberinde (talk) 17:53, 10 August 2015 (UTC)
  • Oppose - Arbcom is rarely able to keep its own house in order, rather not entrust them with another. Tarc (talk) 01:02, 18 August 2015 (UTC)

Split up the powers[edit]

We have had a number of unsuccessful RfAs where the person really only sought one of the "powers" of admins - perhaps we should separate some of those powers now?

I am not saying any specific devolution would be perfect, but, arguendo, let's look at having the "protect" right being extended to those whom the community feels would not abuse such a right. The "block" right should be separated into "block IP" which would be a lesser right than that of "block established editor" (which I would think would be more highly restricted than it is currently). "Viewing deleted pages" would be a lesser right that that of "restoring deleted pages." Viewing "special:unwatched pages" should clearly be a fairly minor right, as far as I can tell.

The concept would be to reduce the angst involved where a person "really" only wants or could use a minor right, thus making it moot as to whether they would misuse a major right. I could even see a minor right being given by the community for those who specialize in certain policies or guidelines - such as having an MoS flag for those who would benefit from the devolved flag, or a template flag for those who are best equipped to handle template issues.

Right now, we are pretty much required to assume any admin-candidate will use all the powers vested in the current admin system - and it is rational that people who write articles feel that an admin who does not fully understand that task may not handle conflicts wisely regarding such content editors. It is rational that editors who are familiar with XfD discussions would prefer that admins be familiar with that process, and not be rubber stamps for it.

Lastly, making "blocks" of established editors a "major flag" would cut down on the "first mover advantage" currently the source of many problems - by restricting it to a subset of admins who would be chosen for their reticence to block where others demur, allowing actual discussions before the "first mover" Zugzwang and the resultant (all too often) accusation of "wheel warring". Collect (talk) 13:43, 6 August 2015 (UTC)

Not on point - this is not a "hierarchy" but a recognition that different sysops can have partially overlapping or non-overlapping spheres of authority. Prior proposals for "sub-admins" did not succeed for a very good reason - it made the "full admins" a superior class. Kindly note that this current suggestion is not the "perennial failure" one at all. Cheers. Collect (talk) 15:17, 6 August 2015 (UTC)
  • One, I'm not sure a "hierarchical" structure is by definition "a bad thing" – the internet likes to pretend that it's somehow "reinvented human nature", but that's never been the case going all the way back c.1994, and hierarchical structures are implemented because they work. But the second part of what you've linked to is that it's a strawman – the way you implement "unbundling the bit" is basically via a WP:PERM-, or "super" PERM-, like mechanism in which an Admin or a group of Admins (yeah, there's "hierarchy" part – deal with it!) dole out various portions of the "tool bundle" to trusted editors (presumably who have a demonstrated record of trust, and who have achieved some "benchmark", a la the number-of-article-created benchmark for Autopatrolled status). And anyone who abuses their portion of the bit would lose (again, likely via Admin action) their "bundle" of privileges, just like users can currently lose Rollback or AWB rights for abusing them. The idea that you'd need to recreate an "RfA process" to dole out portions of the bundle is nonsensical. And the idea that we can't trust the vast majority of experienced, longer-term editors with some portions of the "bundle" is ridiculous. --IJBall (contribstalk) 16:16, 6 August 2015 (UTC)
  • Special:ListGroupRights lists 79 rights that administrators have (as well as the ability to add/remove users from certain groups), although a couple are associated with Flow, so I would assume they are currently useless. I'm sure most administrators don't use very many of those 79 permissions and that most people who want the mop aren't thinking about using (ep-token), (editusercss), (apihighlimits) etc. Most rights are already duplicated (e.g. (skipcaptcha), (autoreview)); I understand some of this was done as a conscious result of consensus deciding to unbundle tools (e.g. (rollback)). The following are the 33 rights that are not given to anyone in any category other than administrators [or bureaucrats, checkusers, oversighters, stewards, researchers and/or bots]:
ep-bulkdelorgs, protect, stablesettings, managechangetags, deletelogentry, deleterevision, delete, globalblock-whitelist, editusercss, edituserjs, editprotected, editinterface, gather-hidelist, markbotedits, nuke, mergehistory, move-subpages, suppressredirect, reupload-shared, abusefilter-revert, browsearchive, unblockself, undelete, apihighlimits, transcode-status, unwatchedpages, abusefilter-view-private, deletedhistory, deletedtext, spamblacklistlog, titleblacklistlog, jsonconfig-flush, proxyunbannable.
A few of those seem to be useful to those who are involved in deletion, a few are designed for vandal fighting, a few for fairly specific tasks (e.g. (mergehistory)) and many seem to be convenient for general editing / only necessary very rarely (e.g. (unwatchedpages), (suppressredirect), (ep-bulkdelorgs)). I have absolutely no idea what (jsonconfig-flush) is.
And while the concept of bundling is listed at "Perennial proposals", that doesn't prevent consensus from changing and I don't see the harm in discussion. Bilorv(talk)(c)(e) 16:25, 6 August 2015 (UTC)
Ooo! That's one I keep forgetting: "article protector"!... Basically, I see at least four "jobs" that can be created from this (and some of these will have "redundant" GroupRights – that's just inevitable): Vandal fighter (blocking IP and un-autoconfirmed vandals), AfD supervisor (article deletor), Article mover (for WP:RM), and Article protector (for WP:RfPP). And I bet there are a couple more that I'm not thinking of... Of course, we'd still need Admins-Bureaucrats for the others stuff (giving out group permissions, blocking long-term editors, etc.), just not as many of 'em... Face-wink.svg --IJBall (contribstalk) 16:42, 6 August 2015 (UTC)
Also, whoever has the ability of seeing deleted edits would have to run through a RfA or a similar process. Jo-Jo Eumerus (talk, contributions) 16:45, 6 August 2015 (UTC)
I'd be wary of splitting up the core functions. In a few cases I've had to delete, protect, block, and look at deleted edits all for one incident (typically a page protect request that leads to a SPI). Each action has to be thought through. Having the tools to do only half the job means another admin will have to familiarize themselves with the situation. --NeilN talk to me 17:01, 6 August 2015 (UTC)
Remember that, in my view of this, there will still need to be "Admins" (or something like "Admin-Crats"). And, yes – I will bet there will still be cases where an Admin-Crat will have to do the kind of job of what you're describing. But it will still reduce the workload on you guys if >90% of the "simple jobs" can be done by experienced editors with a subset of the tools. --IJBall (contribstalk) 17:19, 6 August 2015 (UTC)
Neil, I'm interested to hear your experience. I certainly think that admin's comments [and former admin's comments] should bear more weight, as they have actually used the tools being considered for unbundling. You say "a few cases". Is this a large percentage of cases? Only rare situations? Would unbundling, for example, page deletion and blocking cause so much duplication of effort that it would not be worth doing it? If an experienced non-sysop comes across a situation where a vandal has created several attack pages, is it not better if they can administer the block and CSD tag the pages (or if they can delete pages and report to AIV) than if they have to CSD tag and report to AIV? Bilorv(talk)(c)(e) 17:33, 6 August 2015 (UTC)
They can't do either if they can't get any of the rights because some editors place a high bar that inhibits editors who do gnome work from getting the bundle of rights. bd2412 T 17:40, 6 August 2015 (UTC)
Bilorv, the situation I described where four rights are used for a single incident happens maybe once every couple weeks (and I'm a highly active admin). However, if we do unbundle these rights, we're going to have a "if all you have is a hammer, everything looks like a nail" environment. Every day, there are reports to AIV that are better served by page protection or page deletion than a block. Likewise, every day there are reports to RFPP that are better resolved by blocks or deletions. If the editors patrolling these boards only have access to one tool, we run the risk of implementing sub-optimal resolutions. --NeilN talk to me 19:05, 6 August 2015 (UTC)
  • There are some rights in there that some to be only for sysops purely due to lack of trust: for instance, "unwatchedpages" sounds like a list that a lot of the subtler vandals, hoax creators and LTAs would take advantage of if they could view it. "suppressredirect" would also be useful; I can't actually understand why that would be open to abuse any more than the move function is. I think anyone with several months of editing activity and a few thousand edits [these are higher standards than, for example, rollback] should be able to have these if they request them, although I'm no expert and can't actually use either of those rights, so it's possible there's a good reason for a firm restriction of "admin only".
As an aside, I was slightly confused by the duplication of some rights: there is a clear hierarchy which goes IP ("all"), user, autoconfirmed, admin, 'crat. [Oversighters could also follow from admin.] There's no chance anyone could ever become an admin before becoming autoconfirmed; I can't find anything that says 'crats have to be admins but I would think it to be common sense (especially since they can grant the admin right anyway), and would be very surprised if there's ever been a 'crat that isn't an admin. So why do these rights need to be duplicated? Is it an inherent, non-overridable function of MediaWiki software? (Even then, it begs the question of why the MediaWiki developers think this redundancy is necessary.) Bilorv(talk)(c)(e) 17:33, 6 August 2015 (UTC)
  • Support, as I do, and will, perennially. bd2412 T 17:04, 6 August 2015 (UTC)
  • Tentative oppose I might support unbundling if I saw a well defined set of roles to create, I don't in the above. But any candidate for any role that would include blocking, unblocking, deleting pages, or viewing deleted pages or revisions, would need to go through more or less the current level of RfA scrutiny, IMO. Given that, I'm not sure that that unbundling would actually help much, and I think that Bilorv has a point above. DES (talk) 19:09, 6 August 2015 (UTC)
  • Strong support. I will quote verbatim my comment in the above poll: " I still think unbundling of the tools is the ideal solution. Since candidates can have a need for some of the tools while having no need for others, and the requirements for different tools differ so substantially from each other that the all-or-nothing procedure we have now sinks candidates who would do well with and need one or more tools but not the whole package." StringTheory11 (t • c) 02:51, 7 August 2015 (UTC)
  • Oppose: If an editor can't be trusted to not use functions that they don't know how to use, then we can't trust them with a "unbundled tool". We should not add more hats for collectors to collect. Rollback and reviewer are pretty much unused by most,Esquivalience t 18:01, 8 August 2015 (UTC)
  • Of course. This has been needed for years. Lots of panic about it, but when it happens, we get good results. The template-editor bit proves this, in a long line of other tool-splitting.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  11:17, 9 August 2015 (UTC)
  • Oppose - some of the tools have been unbundles already (WP:TPE) and that's fine, but for real admin powers the old mantra 'if you can't be trusted with one, you can't be trusted with any' applies. GiantSnowman 11:20, 9 August 2015 (UTC)
  • Comment I consider that a person who states he would like to operate in an area of (say) "cutting wood", and asks to be given a "saw" rarely needs to be given a "soldering iron" or "machine gun." People who want to (say) edit templates do not need permissions to revdel edits. People who wish to see deleted articles in order to determine possible new articles for creation do not really need blocking abilities. If a person wishes to work in an area where he or she is comfortable with regard to Wikipedia policies and standards, then insisting on "all or none" makes precious little sense. This is not "hierarchy" it is simply giving tools appropriate to what the person says they wish to do. Collect (talk) 13:18, 9 August 2015 (UTC)
  • Say we get an admin who wants to close XFD - so we give them the power to delete. But what about if their interests change, and they then want to start blocking vandals? Do we need a second RFA? And then if they need to protect pages? A third RFA? etc. ad infinitum - WP:BUREAUCRACY. GiantSnowman 13:47, 9 August 2015 (UTC)
You seem to overcomplicate the concept - the concept is that individual rights would be handled with a great deal less formality than the current RfA gantlet - that is, each right would not require the extensive and daunting process currently used. If the person has one of the flags and seeks another, the proposal would be to give them the added flag with a minimum of discussion, not to multiply the current process ten-fold at all. Looking at "template editing" as an example - do you see us holding long involved debates about such editors? Do you view the "template editor" process as being an "RfA"? Cheers. Collect (talk) 15:07, 9 August 2015 (UTC)
  • I'm with GiantSnowman. Adminship is, in all reality, largely learned on the job; invariably, new admins start out in their comfort zones and branch out from there. That's how we get a diverse admin corps. It seems silly at best and obstructive at worst for people to submit a new application every time they summon the courage to pick up a new tool. I don't agree with the above saw/machine gun analogy, as a logger is typically thousands of miles away from the front lines, but every last user here is on one language edition of one website. Patrolling and clearing one specialty noticeboard vastly increases the chances that other tools will be required along the way, since many (most?) of the admin tools deal with one broad task: countering editing abuse. Blocking, page protection, and deletion—among others—are all interwoven so that, as NeilN said, you generally can't just pick up one tool and clear a backlog with it. More acutely, I can't trust somebody to view deleted pages if they can't be trusted to restore them, or to block one class of editor but not another. Adminship is entirely a volunteer role, and we should therefore be as accommodating as possible, but sysops are serving the website and not themselves. It's not too much to ask of them to be willing and able to participate outside of their specialty areas when required. There are lots of other things that are bothering me about this proposal (how difficult will it be for new users to find somebody able to offer the specific services they need?), but these are my most fundamental concerns. – Juliancolton | Talk 14:49, 9 August 2015 (UTC)
  • I once asked a friendly admin his opinion on whether to propose splitting off the move-over-redirect capability, which is after all one of the least dramatic. He pointed out to me that this and many other admin functions require the ability to see deleted content, and this makes it difficult to split things off. As I understand it, the WMF require an RFA-like process to ensure that the community trusts anyone wanting to do things that require them to see deleted content.

    However, I think we're missing a trick by not separating hard and soft restrictions. I mean, wheel warring is forbidden to all admins - right? - but it isn't physically prevented by any technical mechanism. It's simply expected that admins don't do it, or that they get into trouble if they do. Well, why not have a form of adminship in which certain of the more drama-causing actions are not permitted: I suggest no blocking of anyone except for clear vandalism and clear, bright-line edit warring, and no reversing of any blocks imposed by any other admins. These restrictions would be soft: the block and unblock buttons would still be given, but with restrictions on their use. This would effectively create two tiers of adminship. With this facility, I'd certainly have supported for the lower level some RFAs that I opposed on demeanour grounds or out of a suspicion that a person would cause drama with the buttons. --Stfg (talk) 15:22, 9 August 2015 (UTC)

  • Obvious oppose - if you think RfA is a pain, replacing it with a triathelon of RfAs makes no sense. WilyD 10:52, 11 August 2015 (UTC)
  • Support and let admins grant the unbundled privilege but removal is through ArbCom. template editor is excellent example. Most admins can't make proper template edits. Those that can. should have the ability. Not winning the prom queen process of RfA shouldn't gate that. --DHeyward (talk) 21:06, 17 August 2015 (UTC)
  • Oppose. I think I've said this a few times before in similar discussions: " If it ain't broke, don't fix it".  Philg88 talk 21:45, 17 August 2015 (UTC)

RfA: The Good, the Bad, and the Ugly Future[edit]

Bumping thread for 30 days. Avi (talk) 14:58, 6 August 2015 (UTC)

I think it fair to say we have a consensus that RfA, as it is currently implemented has much room for improvement. There is a lot of passion and energy that it engenders, which is a good thing. Unfortunately, much of this energy is wasted (like friction) and there is no question that it is very hard on the candidate. What I would like to see now is for people to channel that energy and passion into listing what they like about the current process—what they would want to see continued in some form in any evolution of the process(the "Good"); what they do not like about the current process—what they would want to see removed some form in any evolution of the process(the "Bad"); and what is currently not done and should be done in any evolution of the process (the Future). I'd request that people add items to the list, and preface them with at most a few-words description for ease of parsing. Long discussions would preferably be placed in the fourth section. I would like that this be used as a way for us to identify distinct ways to change RfA, be it officially through an RfC, or unofficially, through self-improvement or how bureaucrats (or any editor for that matter) approaches it. Thank you. -- Avi (talk) 14:57, 6 August 2015 (UTC)

The Good[edit]

  • Discussion, not vote: I think the fact that we do not treat RfX as a vote, but as a discussion allowing for back-and-forths and the ability to try and explain different points of view as persuasively as possible is a feature. -- Avi (talk) 14:57, 6 August 2015 (UTC)
  • Since RfA is such a high traffic place, anyone who gets a clear pass has obviously got a very solid consensus to be one. Ritchie333 (talk) (cont) 15:20, 6 August 2015 (UTC)
  • As the process has remained static, it is easy to find advice of how to get through an RfA unscathed via numerous essays. Ritchie333 (talk) (cont) 15:20, 6 August 2015 (UTC)
  • No firm criteria: people are free to support or oppose based on all kinds of different (and valid) forms of "evidence" about the user's behaviour, weighing up the pros and cons from situation to situation, rather than applying an inflexible rule that can't adapt to unique circumstances. On the flip side, this means people can make poor !votes based on grudges and other bad reasons. Bilorv(talk)(c)(e) 20:08, 6 August 2015 (UTC)
  • comment Ha! I just posited this exact thing as a 'Bad.'  :) valereee (talk) 20:15, 8 August 2015 (UTC)
  • The community wants it: Numerous efforts over the years to "reform" RFA have failed, and somehow the project has survived, thanks largely to the fact so many admin tasks are now automated. Last year a couple of us tried to get the community to close RFA and thus have no admin-appointment process for a while so something would have to take its place. But the community voted to keep the current process. And that's fine. Townlake (talk) 03:33, 7 August 2015 (UTC)
  • Discussion is good. At one time I thought RfA would be improved if the servers silently swallowed any post containing the word 'badgering'. Robust discussion is now common. Avi is right, this is good, even if stressful for the candidate. Opabinia regalis (talk) 04:14, 7 August 2015 (UTC)
  • Testing under real-world conditions. Driving tests take place on the road, not on closed courses, even though it increases the risk; sometimes, there is no substitute for testing in the field. Most people are poor judges of their own character, and the stressful environment of RFA provides a real-world test in a controlled environment of how a candidate will actually react under pressure, rather than how they say they'll react. If an RFA candidate lashes out when someone questions their judgement in the "oppose" column, they're also likely to behave poorly when hauled before AN/I on what they consider spurious grounds. – iridescent 07:44, 7 August 2015 (UTC)
Agreed. I have been a casual observer of the RFA process for a while and I must say that while it is often unpleasant, it is an accurate reflection of the atmosphere that administrators must operate in once they get promoted. If anything is broken it is the Wikipedia environment in general, what happens in RFA is just a manifestation of that. Vrac (talk) 13:50, 7 August 2015 (UTC)
I've always disagreed with this stance. It's simply cruel and unfair to put a person through stressful and unpleasant conditions using the excuse that "they'll experience this if they get the bit". Although it might be a somewhat extreme analogy, here's an example: Police officers sometimes have to deal with violent and dangerous criminals that won't hesitate to physically harm them. They of course must know how to deal with this, but does this mean that we should actually put them through this severely dangerous experience when they're training? Of course not. My point here is that, as I stated above, we shouldn't make people go through unpleasant conditions for the sake of "showing them how it is". --Biblioworm 18:29, 8 August 2015 (UTC)
Forgive the statement of the obvious, but that is exactly how they train police officers (and soldiers etc), although they're now moving towards immersive interactive systems rather than live-action exercises. Wikipedia unfortunately doesn't have the facility at present to have "virtual assholes" who deliberately behave obnoxiously to a prospective admin in a closed test environment to see how they'll react; the alternative would be some kind of admin probation but that suggestion has always (rightly) been shot down, since failing probation would cause significantly more ill-feeling than failing an RFA. – iridescent 19:47, 8 August 2015 (UTC)
It would be surreal if everyone pretended to be nice and civil at RFA while continuing to bash elsewhere. Am I the only one who thinks these problems are not limited to RFA? Vrac (talk) 19:56, 8 August 2015 (UTC)
Apologies, as I'm not at all familiar with the details of police officer training. I did guess that they might put them in a simulated environment, but I'm quite certain they don't actually put the trainee at the risk of severe injury or death, as might happen in actual practice. And in any case I still stand by my position that we shouldn't make candidates suffer simply to "teach them how it is"; it's illogical and contrary to our policies. --Biblioworm 20:00, 8 August 2015 (UTC)
  • Emphasis on deletion knowledge. Overzealous deletions, A7s of non-eligible subjects, and not doing due diligence at AfD are some of the problems behind how editors treat deletion. This erroneous treatment has caused many newbies to leave, thinking that they are not welcome here. Such problems are even more problematic when done as admins. RfA's current heavy scrutiny on deletion matters has at least somewhat prevented this problem from growing, as editors must have a good knowledge of deletion before they are trusted by RfA to, well, delete. Esquivalience t 17:43, 8 August 2015 (UTC)
    • I counter that although sufficient knowledge of deletion processes are required, this can sometimes become much too overzealous to the point where any mistakes become unacceptable. For instance, I think one of the most ridiculous opposition rationales is the "win/loss" ratio at AfD (i.e., editor has only a 65% "hit rate" at AfD!), which implies that editors !vote at AfD to "win" rather than state their case based upon policy. I also have to ask how not getting a high hit rate at AfD (which can be volatile and occasionally deliver unexpected results) would affect their ability to judge consensus. And, finally, should we fail candidates over a few CSD mistakes? --Biblioworm 18:22, 8 August 2015 (UTC)
      • Yes we should fail candidates if they have a few CSD mistakes, provided those mistakes are recent (I'm happy if recent edits show that tagging has now improved). However RFA has not got to the point where any mistakes have become unacceptable. A few serious and recent mistakes at CSD can and should derail an RFA, but a single isolated mistake is not sufficient to evidence the sort of pattern that can derail an RFA. A candidate who hasn't yet worked out the difference between vandalism and goodfaith edits would also fail RFA. ϢereSpielChequers 21:09, 10 August 2015 (UTC)
  • It stays at RFA I've certainly had discussions and the occasional disagreements about certain support/opposes with other editors but rarely have I run into a situation where what's happened at RFA has been held against me or cited against me in other places on the project. Perhaps I'm totally out of touch or my experiences differ from the majority, but people seem to understand that RFA is a bit of a battleground and once we leave the field its business as usual. Mkdwtalk 02:02, 24 August 2015 (UTC)

The Bad[edit]

  • Incivility: I think people get too caught up in the discussions which results in too many personal attacks, adding tension and affecting the tone of the discussion. -- Avi (talk) 14:57, 6 August 2015 (UTC)
  • Improper Intensity: I think that the candidate is subject to too intense scrutiny. Serious concerns need to be raised, but not enough good faith is applied. -- Avi (talk) 14:57, 6 August 2015 (UTC)
  • No AGF: There is no effort to discourage and/or discount !votes that assume bad faith (especially those that pull quotes out of context to make it seem like the candidate has an agenda). --Ahecht (TALK
    ) 15:15, 6 August 2015 (UTC)
    • Agreed. I've seen too many recent RfA's where the reverse was true: assume bad faith. --IJBall (contribstalk) 17:05, 6 August 2015 (UTC)
  • High standards. I concur with the issues raised above, but I think the single worst problem with RfA is the unreasonably high standards, which are impossible to pass except for almost-perfect candidates. The declining number of promotions is clear evidence of this. --Biblioworm 15:39, 6 August 2015 (UTC)
    • Standards also often conflict with each other. That would not be a problem if voters did not oppose candidates who "fail" these standards instead of simply not supporting them. —Kusma (t·c) 18:45, 6 August 2015 (UTC)
  • Discussion, not vote. Throwing this out as a talking point. This definitely also belongs in the good column, but it has its downsides. Requiring users to justify their positions, especially opposes, leads to quite hurtful feedback for candidates. Even if delivered in a civil manner, editors explaining why they don't trust the candidate with extra tools is going to make hard reading. The more we encourage opposers to give detailed rationales, supported by diffs, the more we may unintentionally magnify this aspect of RfA. Whilst it would not allow the same discretion in determining the outcome, a pure vote might be easier on the candidates. WJBscribe (talk) 18:32, 6 August 2015 (UTC)
    • Rationales attached to votes often are like campaigning right at the ballot box, which is forbidden in most democracies. —Kusma (t·c) 18:45, 6 August 2015 (UTC)
  • Low participation. Typically less than 200 votes on a project with 585 "active" administrators shows that RfA has little participation from the wider community. —Kusma (t·c) 18:45, 6 August 2015 (UTC)
  • No ability to defend oneself. Candidates who attempt to point out incorrect or otherwise misleading information presented by those opposing are often accused bludgeoning the process and have that held against them by subsequent voters. The same problem exists if others attempt to correct the misinformation. Calidum T|C 18:47, 6 August 2015 (UTC)
    • Not only that, but I've seen candidates opposed because other people disagreed with previous oppose votes. It's the most boneheadedly stupid rationale I've seen at RfA since the "primo hungry power-face" guy. Reyk YO! 18:55, 6 August 2015 (UTC)
  • Editcountitis in various guises. Besides unreasonable expectations on number and distribution of edits, some voters seem to rely on automated statistics (e.g. AfD stats) instead of trying to answer the question whether the candidate will make a good administrator. —Kusma (t·c) 18:51, 6 August 2015 (UTC)
    • Fully agree. I'm rather disturbed and amused at the same time when I see people comment that only several thousand edits is not enough. For the average person who edits as a hobby and doesn't want (or have the ability to) spend all their time here (there, I've really done it now...), getting even a few thousand edits isn't as cheap as some seem to think it is. --Biblioworm 19:53, 6 August 2015 (UTC)
      • Yeah, but the flip side of that is that (IMO) somebody can't just "walk in off the street" and "be an Admin". I'm not sure how true that was (or not) in 2004, but it's definitely not true in 2015 – in order to understand this place (and how to run it) takes time and experience... But this gets back to the point that being an Admin is too big a job for >99% of our editors, and that's probably why "smaller bites at the apple" (i.e. smaller, unbundled toolsets) makes sense as, you're right – most editors probably don't have the time to give to do what it takes to make it through an RfA and be an Admin. --IJBall (contribstalk) 00:54, 7 August 2015 (UTC)
        • No, the flip side of that is alienating people who do good work and have the right skill set to be an administrator because they've failed to meet some arbitrary expectations that are not evidence-based. There is no evidence that having featured content under one's belt makes one a better admin; there was a point in time where everyone built a featured portal to meet that 'qualification' - hilarious given the fact that even back in those days nobody actually used portals. There is no evidence that people with no contributions to template space are going to be problem admins. There is no willingness to consider that different editing styles will directly affect total number of edits; people tend not to "save" nearly as often when using VE as they do with wikitext, for example, even if they are editing a page more extensively. Risker (talk) 05:21, 7 August 2015 (UTC)
          • None of which are arguments I've made. I'm not talking about "GA" or content creation experience – I'm talking about the breadth of policy knowledge (not to mention knowledge of the various areas like WP:AIV, WP:RfPP, etc.) required for Adminship. There are a lot of different, moving parts to the toolset. I can't speak for anyone else, but I don't want someone who isn't well-versed in much of that getting Adminship, and I don't think you can get that in just 6 months here. --IJBall (contribstalk) 06:26, 7 August 2015 (UTC)
            • I've done quite a few edits here and on commons, some of the things I do differ in edits per hour by not just a hundred times but if we include Catalot edits on commons, my fastest editing rate is a thousand times my slowest. But I probably learn more when I'm spending an hour doing something complex. Of course there are editors with less than a tenth my edit count who have put in more time than me and know more about this place. So I'm living proof that editcountitis is a distraction from RFA. ϢereSpielChequers 18:05, 8 August 2015 (UTC)
  • Hard to tell how much research has gone into a !vote: this might be an unsolvable problem. But imagine two people, both of whom have written "Oppose per User X". The first person who supported was just blindly accepting X's judgement; they looked at very little themselves but thought "I'm sure X did their research". This rationale should get less weight. The second person spent hours researching, then came to the !voting section and thought "User X has said exactly what I want to", and just wrote "per User X" to avoid wasting words (X said it better anyway) and avoid hurting the candidate's feelings any further. This !vote should get as much weight as User X's original message. But there's no way to tell these two people apart. Bilorv(talk)(c)(e) 18:53, 6 August 2015 (UTC)
  • Axes to grind. There is no expectation that editors should recuse themselves from participating in an RFA when they may have an ax to grind with the candidate; said participants aren't even required to note any potential previous interaction with the candidate. There is a reason why admins are forbidden from acting when involved and arbitrators have to recuse themselves from time to time. Interestingly enough, this is often one of the issues that is brought up in nearly every discussion about a community based method of removing the administrative privileges. Edit to add: I also think there may be instances where a participant votes for or against a candidate based on a grudge he/she holds against someone who voted the opposite way previously. Calidum T|C 22:32, 6 August 2015 (UTC)
    • On that note, it appears to me that a number of !votes are about promoting particular wiki-ideologies as opposed to assessing the suitability of the candidate. Arguments should be framed strictly around the suitability of the candidate, since we are at RfA here not VPP. Jo-Jo Eumerus (talk, contributions) 22:42, 6 August 2015 (UTC)
  • Too many questions and comments. We treat candidates as though they're already experienced admins. There are too many questions, often asking about hypothetical situations that would be hard for anyone to address, and there's too much threaded discussion. I'd like to see a maximum of two comments per person, unless the candidate responds, in which case responses to that reply would be okay. Sarah (talk) 00:21, 7 August 2015 (UTC)
    • This is an excellent idea, Sarah. Risker (talk) 05:21, 7 August 2015 (UTC)
  • Evidence-free criteria. The crat chat for Liz included a discussion of 'substantiating diffs'. Several people objected to this, thinking it devalued holistic evaluation and made picky, hypercritical oppose votes more likely. But nobody ever seems to suggest 'downweighting' votes that simply state that a candidate hasn't reached some arbitrary numerical threshold. User has only been here for two years, Can't support under 10,000 edits, AfD hit rate isn't even 75%, Wants to work at AIV but only has 20 reports, I really prefer to see article+talk edits at least 50%, and on and on. None of these thresholds have been validated at all. We shouldn't encourage pseudostatistics. Opabinia regalis (talk) 04:20, 7 August 2015 (UTC)
  • Some voter criteria induce candidates to refrain from activities that would improve the encyclopedia. I think of this in particular with relation to the statistics and expectations surrounding deletion. Woe betide the admin hopeful who proposes too many articles (even spammy articles) for deletion and gets turned down (often because there are so few AfD votes nowadays and the article gets kept by default), or votes to delete too many articles, or sends too many redirects (even implausible ones) to RfD. This also applies to potential candidates keeping their heads down and not participating in discussions that may have any potential to be contentious. And perhaps most harmfully, it discourages potential candidates from working on content in "contentious" areas, or to simply walk away from poor quality content rather than argue with another editor who's got WP:OWN issues. Risker (talk) 05:21, 7 August 2015 (UTC)
  • Trick questions - Trying to trick candidates is a bad faith tactic. - MrX 20:37, 7 August 2015 (UTC)
  • Soapboxing in the guise of authority. The general discussion area should only be used for neutral discussion about the RfA, not as a platform for dredging up past issues. This is particularly harmful when done by current and former Arbcom members. - MrX 20:37, 7 August 2015 (UTC)
  • Popularity: As a candidate's popularity grows, the amount of research voters actually do approaches zero. And this is equally true for candidates that are liked or disliked. Esquivalience t 17:49, 8 August 2015 (UTC)
  • Incredibly detailed critieria I have been to user pages where the editor has parsed their critieria to the point of torture. If it's so important that you need a thirty-cell table to decide if someone is qualified, how can you call it no big deal? The criteria need to be set in policy, and IMO they need to be: 1. Is this person likely to intentionally misuse tools? and 2. Is this person so green they're likely to, net, cause more work than they accomplish? If the answer to both is no, then under WP:NOBIGDEAL they should be trusted with the tools. valereee (talk) 20:08, 8 August 2015 (UTC)
  • Single metric voting Using a single metric/criteria to support or oppose a candidate such as AFD participation, editcountitis, article creation, or featured content to name a few. Especially when the candidate hasn't expressed an interest in working in that particular area and has shown a long and good standing track record of working in the area they have expressed an interest. Mkdwtalk 01:49, 24 August 2015 (UTC)
  • Inverse relationship between experience and success There is the occasional exception but it would seem that there is a threshold where by individuals reach a point where they have too many enemies, have become involved in too many controversies, and have made too many citable mistakes to have a successful RFA -- even if those number of mistakes, enemies, and controversies are proportionate in scale to those who have run a successful RFAs. Mkdwtalk 01:55, 24 August 2015 (UTC)

The Future[edit]

  • Improved tone: I would want more emphatic implementation of civility mandates in the discussion. If one cannot say something without ad hominem attacks, don't say it. If it is said, it can be removed by an uninvolved editor/admin/crat (to be determined). -- Avi (talk) 14:57, 6 August 2015 (UTC)
  • Lower the standards. !Voters must understand that candidates will not be perfectly well-rounded. If a candidate is acting in good faith, shows decent competency with policy, and is reasonably civil, they should be admins. Humans are not incapable of learning new things as they go along. --Biblioworm 15:39, 6 August 2015 (UTC)
  • It is essential to have a more deliberate effort to recruit more admins with a wide variety of backgrounds, interests, skills, and experience (both on and off Wikipedia.) Sydney Poore/FloNight♥♥♥♥ 22:07, 6 August 2015 (UTC)
  • Steady as she goes: Don't change a process that the community wants and that gets the "right result" almost every time. I realize some users are getting their feelings hurt, but oh well, honesty hurts sometimes. Townlake (talk) 03:37, 7 August 2015 (UTC)
  • Delete everyone's "RfA criteria" subpages. Only half joking. Opabinia regalis (talk) 04:22, 7 August 2015 (UTC)
  • Delete everyone's "RfA criteria" subpages. Or more precisely, any vote with comments like "doesn't meet my [wikilinked] criteria #5" or similar should be immediately discounted, perhaps even to the point of indenting. Same for deprecating votes that refer to essays about RFA including the frequently misused WP:NOTNOW, which is intended to dissuade users with obviously insufficient experience from a candidacy, but I've seen it being used on people with more than 5000 edits who've been editing for more than a year. Risker (talk) 06:15, 7 August 2015 (UTC)
  • Lower the threshold for consensus. Most other large projects have their RFA success threshold at closer to 2/3, and that's the threshold that's been used for genuinely major changes to our project in many cases. Risker (talk) 06:15, 7 August 2015 (UTC)
    I strongly concur. I've often wondered why the bar is set so high.- MrX 20:40, 7 August 2015 (UTC)
    70-75% does seem high. It may also be too high for bureaucrats. People are having a hard enough time passing RFA -- it's no wonder we haven't had a bureaucrat nomination in over a year. Coming up on two years in January if I'm not mistaken. Mkdwtalk 02:06, 24 August 2015 (UTC)
  • More active management of RFA by bureaucrats. There are more bureaucrats now than there were back in 2005-2009 when there were frequently 5 or more RFAs occurring at once; 'crats didn't have time to keep an eye on the RFAs. That's changed. I'd like to see bureaucrats take a more active role in the flow of RFAs to improve the tone, to prevent candidate abuse (and in some cases voter abuse). This could take the form of one or two 'crats taking on a "supervisory" role for a particular RFA, while leaving that RFA to be closed by a colleague; if there is a 'crat chat, the supervising 'crats would not vote. Risker (talk) 06:15, 7 August 2015 (UTC)
That's a very old, and one of the most perennial suggestions, Risker. You know as well as I do that the bureaucras are loathe to take on any tasks that are not firmly written into their job description already, and that if the community as much as dared to reach a consensus for it, the crats woud simply still not feel compelled to do it. So where you've pretty well said as much elsewhere, I'd like to know just which camp you are really in. Kudpung กุดผึ้ง (talk) 08:16, 7 August 2015 (UTC)
<sarcasm>I assume, Risker, you are saying they should have more effectively managed the crat chat talk page, and deleted comments like accusing another administrator of misrepresentation, and "There's definitely a bias here; whether it's male vs female, or "long time admin" vs. upstart, or something else, or "let's put as many obstacles in the way of any new admins" or something else, I'm not sure."</sarcasm> Sauce for the gander is sauce for the goose. Are you proposing, Risker, a standard that allows comments like that one you made in the crat chat talk, or something stricter?--Wehwalt (talk) 08:55, 7 August 2015 (UTC)
Wehwalt, I was entirely and completely able to show that there was misrepresentation, simply by directly and fully quoting the exact words of the candidate. I don't know that it was *intentional* misrepresentation, but it was misrepresentation. I have no problem with that being the standard.

Kudpung, it seems that the bureaucrats as a group are more receptive to reconsidering what their jobs actually entail right now; in fact, a 'crat has joined you in a proposal that would significantly alter their responsibilities. This discussion was initiated by bureaucrats, and I assume good faith that they are genuinely concerned about the mess that RFA has become. I don't have a "camp", Kudpung. I'm a working admin who knows darn well that we're already very short of working admins, that we need more, that we're training potential admins not to do a lot of things that the encyclopedia needs to have done because it will adversely affect their chance of success at RFA, that perfectly reasonable candidates have been rejected for reasons that are not relevant to whether or not they'd do the job properly, and that very good potential admins routinely take one look at *RFA as it exists today* and say "no thanks". Risker (talk) 12:19, 7 August 2015 (UTC)

This would be a very good move. ϢereSpielChequers 20:21, 7 August 2015 (UTC)
  • Don't allow threaded discussions. Supports comment in their section, opposes in theirs, neutral in theirs, and discussion/debate is limited to the talk page. GregJackP Boomer! 08:31, 7 August 2015 (UTC)
That's a useful suggestion. I too often see, going down the list of opposes, one or two prime supporters of the candidate trying to answer every point. It's tedious and repetitive. Do it like an ArbCom request, keep all your comments together. I bet if someone has to dilute their comment with 20 replies, they won't do so.--Wehwalt (talk) 08:55, 7 August 2015 (UTC)
The threaded discussions are one of the few bits of RFA that works reasonably well. People with untrue assertions or who try to shift consensus as to what the criteria for adminship will be will usually get their position challenged, often that results in people correcting their !votes. ϢereSpielChequers 19:45, 7 August 2015 (UTC)
Discussion works well; the multi-branching threading that occurs in most English Wikipedia discussions means that an interested party may need to monitor tens of conversations. This approach prioritizes the convenience of the person who made the initial comment over everyone else, which makes it harder to keep people engaged in the conversation and so makes reaching a group decision harder. isaacl (talk) 20:27, 7 August 2015 (UTC)
  • Admin incubator - Let's have a venue where users interested in becoming admins can get advice and be directed to areas where they can hone their skills under supervision (AfD, NPP, SCV, AN/RFC, etc.). I envisage it being something like the WP:ADOPTION with a little more structure and formal mentoring. Users who WP:HATCH (graduate) from the program should be a shoo-in for adminship.- MrX 20:54, 7 August 2015 (UTC)
This sounds like Wikipedia:Admin coaching to me. Besides, I doubt that graduating from this program would actually make the editor an adminship shoo-in, because they would likely be opposed due to the fact that they "desperately want adminship" and are "power hungry". --Biblioworm 20:59, 7 August 2015 (UTC)
Sad. The ones who are actually power hungry are likely clever enough to disguise it.- MrX 21:15, 7 August 2015 (UTC)
There's no evidence that newbie admins systematically make more errors of judgment in basic tasks. Despite some of the comments in these threads that being an admin is Very Serious Business, the overwhelming majority of admin tasks are really not difficult. Opabinia regalis (talk) 21:30, 7 August 2015 (UTC)
  • Set standards in policy Clarify what we're looking for rather than letting every editor decide for theirself what's the criteria. The things some editors are looking for are insupportable w/re: WP:NOBIGDEAL. If it takes a table to lay out your criteria, you are treating it as a very big deal indeed. valereee (talk) 20:12, 8 August 2015 (UTC)
Something like WP:FAC, maybe? That is, in a RfA a candidate is evaluated against expectations and a bureaucrat then assesses if they meet the expectations, based on the discussion? Jo-Jo Eumerus (talk, contributions) 20:29, 8 August 2015 (UTC)
I don't we need a bureaucrat to answer that question? Can't editors answer them? I guess what I was going for is more that if you oppose, you do so for a SET reason. That is, the reason(s) for the oppose has to fall within one or more of a set number of categories, such as "Likely to misuse tools" or "Unwillingness to learn even after having been pointed in the right direction" or "Tends to foment drama" or "Difficult to work with" or whatever. Things that actually mean a person is already a problem and likely to be more of a problem if given admin tools. Rather than "Hasn't learned EVERYTHING" or "Hasn't done enough (whatever the editor's hobbyhorse)" or "Only interested in X when I personally think Y is more important." valereee (talk) 21:50, 8 August 2015 (UTC)
  • Have a discussion phase before any support or oppose opinions are given: for the first three days, have editors comment on the relative advantages and disadvantages of a candidate's qualities regarding suitability for the administrator role. Then open up the registration of opinions for or against the assignment of administrative privileges. This will present a more complete view to everyone before they start weighing in, and hopefully result in more effective opinions, particularly amongst those who tend to only visit a given request for adminship page a small number of times. isaacl (talk) 02:35, 15 August 2015 (UTC)
    • I think someone tested that, aside from making the whole thing longer it added emphasis to the question section. Also a three day phase rather discriminates against regulars who only only edit one evening a week. ϢereSpielChequers 19:18, 15 August 2015 (UTC)
      • "Three" is of course a number pulled out of thin air; the length of each phase can be determined by consensus. Naturally if there is a desire to cater to those who edit once a week, the process would have to be extended to a minimum of two weeks (maybe a fixed day of the week for starting requests for adminship should be designated?). Yes, having an initial discussion phase would, by design, increase the importance of learning about the candidate via responses to questions, as well as provide a set period during which interested parties could report the results of their analyses of the candidate's abilities. The end goal is to improve the quality of the expressed opinions of participants, by reserving time for discussion and analysis to take place first. isaacl (talk) 21:28, 15 August 2015 (UTC)

Discussion (GBF)[edit]

I think that the intensity is something at every admin needs to be okay with going through. That being said I think we should come to a consensus as a community what our criteria for being an admin is.

One of the reasons too many candidates fail is a combination of allowing any criteria and allowing a minority to fail an admin. Just like at AfD you will not be given much consideration if you vote "keep I like kittens", votes that have no basis in policy or criteria should be given little weight. Until we agree on that criteria we cannot do this.

The current situation allows 10% to oppose for lack of content contributions, 10% to oppose for too much experience with drama, 10% to oppose for not enough experience with drama and so on. It all adds up to a fail even though no specific objection favours consensus. By making clear what our criteria are we at least make a reasonable target for candidates to aim for.

If our admin to vandal ratio falls too low you will see grumpy admins who are overworked and jumping at shadows. The less admins there are the more the power and burden will be concentrated in the few remaining admins. Many seek to improve our desysop process, but if we cannot first get RfA passing people at a reasonable rate first we will end up with a ship with nobody at the wheel. Chillum 15:00, 6 August 2015 (UTC)

Chillum, why do you think the intensity is something every admin needs to be okay with going through? I would imagine there are admin jobs that have very little drama associated with them. So little that very likely many intensity-tolerant admins would be bored by doing them. Protecting pages that get short-term casual vandalism, for instance. Wouldn't it be good to have some admins out there who are happy to do those jobs? And if it would, and those are the very people turned off by the high-intensity of the current RfA process... valereee (talk) 21:59, 8 August 2015 (UTC)
Admins will experience intense people, it is important that they can keep their cool in such situations. While I imagine it is possible for an admin to focus solely on routine non-controversial decisions that never upset anyone, what we need is admins willing to take on the difficult tasks. Chillum 06:13, 11 August 2015 (UTC)
@Chillum: -- I think we need both. valereee (talk) 02:32, 15 August 2015 (UTC)
Can you show me a single admin that has done their job for more than 6 months that has never had to deal with a stressful situation? Chillum 03:42, 15 August 2015 (UTC)
I am not sure that setting strict criteria for an admin is a good idea. There are many kinds of admins: the more content focused ones, the more janitorial ones, and the more judicial/enforcement ones. Each helps the project, and each has different skills. I'd rather we each individually decide about each candidate. Should enough people agree that the skills brought by the candidate are accpetable, that person gets the mop. -- Avi (talk) 15:05, 6 August 2015 (UTC)
Specifically I think that we should define what is not acceptable in an admin. If it is not acceptable for a candidate to have never written a featured article for example, then that should be shown by the consensus of the community. If the community does not agree this is a requirement then it should not be given much weight as the only reason for an oppose.
I also do not think the criteria should be strict or exhaustive. We can still employ common sense while describing best practices like we do in many other policies. Chillum 15:08, 6 August 2015 (UTC)
(edit conflict)I think you may have an entry for the future section, good sir (or ma'am or howsoever you wish to be addressed) Face-smile.svg. -- Avi (talk) 15:11, 6 August 2015 (UTC)
  • Well obviously the administrators who focus on janitorial skills rather than content are flippin' incompetent when it comes to content, and shouldn't be trusted with the tools. Maybe we should have three classes of admins; janitors, content managers, and enforcement. Hummm...Hummm.... <insert smiley faces liberally, as necessary> --Hammersoft (talk) 15:10, 6 August 2015 (UTC)
    • If that is what you want to see in the future, please place it there. It's much easier to parse a list of sentences than reams of paragraphs (Having done so for the past 37 or so hours, please have mercy Face-wink.svg). -- Avi (talk) 15:13, 6 August 2015 (UTC)
I'm reminded of what ThatPeskyCommoner said back in 2011 on this very talk page: "yes, we need people to build wonderful new roads - but if we're building wonderful new roads, we also need people who are happy to make sure the signposts are easy to read, people to create new maps, people to make sure that potholes are filled in, people who pick up the litter so road-users don't break their necks in it, people who make sure that bandits aren't lurking in ambush along the sides of our wonderful new roads, people who clear the road quickly when there's a crash ... we're all important." archived here -- Gogo Dodo (talk) 06:09, 7 August 2015 (UTC)
  • We should compile a list of reasonably experienced editors and choose randomly choose 5-10 to be admins, and see how it turns out. I don't think anything catastrophic would happen. --Biblioworm 15:39, 6 August 2015 (UTC)
    <beavis>Hhuhuhuhh he said CHOOSE</beavis>. And there's the rub. How do you choose? Send them to RfA. Oh wait... --Hammersoft (talk) 15:43, 6 August 2015 (UTC)
    <butt-head>Ehhh heh hehh, you said "rub" hehh heh eh heh<butt-head> Anyway, that might be a fun experiment, but the baseline criteria would have to be pretty high.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  16:37, 6 August 2015 (UTC)
    We might as well do it now before we're forced to do it when (yes, I said when, not if) RfA becomes impossible to pass and we have an admin shortage crisis. --Biblioworm 16:58, 6 August 2015 (UTC)
  • I honestly think the biggest problem with RFA is that currently AGF and NPA don't apply, which both lead to incivility. If we sanctioned editors who made post attacking the candidates personally and not their edits, and if we discounted !votes that assume bad faith (especially those that pull quotes out of context to make it seem like the candidate has an agenda), I think RFA would be a much nicer place. I honestly found it ironic that the most recent RFA candidate was opposed because she spent too much time on the "Drama Boards", but RFA has become the biggest drama board of them all. At least AGF and NPA exist at AN/I. --Ahecht (TALK
    ) 15:15, 6 August 2015 (UTC)
    Could you please post some diffs about attacking the candidate personally? Saying that I don't like her hair or he is too fat, or has unpleasant habits? Commenting on the positions a candidate has taken or where the candidate has spent his time while participating at Wikipedia aren't personal attacks, because those are, by long practice, some of the criteria. Mentioning them in an oppose is only a personal attack if you are emotionally committed to the success of the candidate. If we are going down that road, then something should be done about "Support. no problems" votes. Go on, give us some diffs then to show us why you think the candidate should be an admin. Show us the candidate being thoughtful, or intervening helpfully in a dispute, or other evidence of qualification. But don't seek to use a contentious case to impose standards on only some, and thereby shift the burden from the candidate to opposers.--Wehwalt (talk) 08:44, 7 August 2015 (UTC)
    Wehwalt, it's not possible to provide "diffs" of no problems. The entire point is that the voter has reviewed the candidate and has not found anything that is concerning. There is no way to demonstrate the absence of concern by diffs. Risker (talk) 12:24, 7 August 2015 (UTC)
    But I think it's possible to provide diffs of excellent or even good behavior. "He reacted in a positive fashion in X dispute[1]" or "he displayed a good understanding of the MOS with this edit[2]", and so forth. These two rationales ("found no problems" and "found good behavior") may differ (possibly substantially), but such diffs could generate the same kind of discussion that diffs off problematic edits ("is this actually an example of a good/bad thing?") and so forth. --Izno (talk) 12:47, 7 August 2015 (UTC)
    Just so. We like to see participation at AFD; difs might be of particularly thoughtful interventions there that were not duplicates of what people had said previously, or that calm down the situation. In other words examples of what we want to see in an administrator. Otherwise it's just "Support. I like kittens."--Wehwalt (talk) 13:11, 7 August 2015 (UTC)
    I disagree very much with this, and believe that your comments are rooted in the assumption that all users are inherently *unqualified* to be administrators and must demonstrate some special, secret qualities in order to be granted this amazing gift. It's a mop, Wehwalt and Izno, not a sceptre. This presumption that someone must be exceptional in order to pitch in on cleaning up CSD or AIV is a very major part of the problem with RFA. It's only really taken root in the last couple of years. It is exactly what I believe is wrong with RFA. You're basing your ideas on the assumption that nobody's qualified unless proven otherwise, which is a dramatic change from the basis on which the idea of RFA consensus was developed. Risker (talk) 13:24, 7 August 2015 (UTC)
    rooted Not consciously! My thought isn't that this would be used to show an absence of unqualification, but to show that these editors really are just normal people who want some extra buttons, by providing counter-evidence to an opposer's diffs. It's the idea that "this user isn't perfect, but just as you [an opposer] can show that this user may have times at which he has displayed negative traits, I also can show that he has displayed positive traits, and think that because of this he's trustworthy enough/[insert phrase of choice]". Really, it's sharing the analysis a more expert support !voter at RFA has done. --Izno (talk) 13:35, 7 August 2015 (UTC)
    Again, I agree with Inzo. As for adminship, it is what the community says it is. Yes standards have increased. Rightly so. They've increased for all sorts of things on Wikipedia from articles to editors. Perhaps 11 years ago the mop was handed out with the rations at the time when articles without a single citation in them were featured. The community's collective wisdom is determining how many administrators we should have. The practice seems to be that there's no desire to add too many administrators, and the candidates must meet a high standard. Until the number of voters who think that drops significantly, I doubt if there will be any change. Really that's not Wikipedia works. The community does things its way and sometimes we think we know better. --Wehwalt (talk) 14:00, 7 August 2015 (UTC)
Perhaps you're not consciously doing it, Izno, but others definitely are (one only has to look at some of the lists of "criteria" to know that), and I think you are probably one of the, Wehwalt, based on what you are writing here. The reality is that 98% of people who have *ever* passed RFA, including a significant percentage from back in the day where all it took was "can I haz tools" and [almost] nobody said no, have caused no harm with those tools. (Some of them never did very much with them, but that's a different story.) We could take a random group of 100 editors with 5000 edits (at least 30% in mainspace) and 2 years of editing experience, give them the tools without any kind of vetting process, and the odds are strongly in favour of them doing absolutely no worse with having those tools than the admins who went through RFAs of varying degrees of nastiness. What I'm saying is that the tools should go to people who are operating normally within our community's behavioural and editing quality expectations, and that there is no way to give diffs for "acts just like a normal editor" because it's, well, normal. Of course, we have a problem in that anyone who puts their head above the parapet nowadays has it handed to them on a platter if they subsequently run for RFA, and the social networks that helped people to develop useful skills are significantly eroded, but my point remains. Until around 2009-2010, unless someone did something unusually problematic (either specific incidents or a pattern of behaviour), they were likely to get the tools, and they were likely never to create a problem as a result of having those tools. That has changed, and now there seems to be on the part of many the belief that candidates need to be much more than useful contributors who are likely to help out in areas where these tools are useful. They must be significantly above-average editors (the GA/FA issue) or have reflected the voter's particular views at deletion discussions (the inclusionist/deletionist dichotomy), or met some other abstract expectation. The starting point used to be "will be okay" and opposers had to prove otherwise. What you're suggesting is the starting point should be "isn't qualified" and supporters must prove otherwise. That is an inherent assumption of bad faith, there is no evidence that there is even a trust angle involved let alone any real problems with the candidate having the tools, and it is having a very significantly negative effect on the ability of the project to cope with workloads. Risker (talk) 14:15, 7 August 2015 (UTC)
Wikipedia has changed. It's not what it was in 2004, four 2009, or even 2013. And it's less intimate, there's been a major turn over in the community, and we are getting respect from outside. What suited us then may not suit us now. Being here for six months or year and being relatively inoffensive is no longer a qualification. People want the best administrators they can get because we have seen what's the worst can do. And turning back the clock to the way things were in the good old days which really weren't isn't practical anymore. It is sad in a way. The community is cautious when it comes to these candidates, and whether the trend changed in 2005 or 2009 or 2013 and is really a matter for the historians of Wikipedia in the unlikely event we ever have any. What is expected are attributes that would show that the candidate would make a good administrator, not that it is her turn.--Wehwalt (talk) 16:01, 7 August 2015 (UTC)
We aren't getting the best possible administrators, Wehwalt, because we've warped the notion of what makes someone a 'topflight' administrator to the point that their actual likely activity as an administrator is irrelevant. The result of this is that we are barely getting any administrators, and certainly not enough to carry out the necessary tasks. And one of the main reasons that Wikipedia isn't what it used to be is this concept that "assume good faith" is a quaint and passé notion that has outlived its usefulness. We have lots of good evidence (years of experience, in fact) that candidates with the RFA equivalent of a 3.0 grade point average not only are acceptable, but that they are probably the lifeblood of the active admin corps. We have no evidence to support escalating the requirements to the RFA equivalent of Rhodes scholar, yet that is what we have done. There is no evidence to support this absurd escalation in expectations, and lots of evidence to discount it. In other words, we have replaced a system that actually worked with one that doesn't, without any reasonable basis for the change. I do trace this change back to the point where people were (completely incorrectly) going around saying that it was impossible to get bad admins desysopped. It's always been possible, and it became much more possible from 2009 on, when Arbcom increasingly took into consideration the activities of administrators that weren't specifically dependent on the use of admin tools. It has never been that hard to get problem admins desysopped. "We have seen what's the worst can do" - what we see are individuals affecting usually a small number of articles or pursuing a dispute with a small group of editors, as often as not without using admin tools. The number of "bad apples" in the admin corps (and bear in mind, the overwhelming majority of current administrators passed RFA more than 5 years ago) is minuscule; you can dispute that statement until the cows come home, but it it brutally unfair to the encyclopedia to be taking out that personal belief on the back of editors who'd do a perfectly good job just because you think that Admin A isn't very good but you won't do anything about it. Risker (talk) 16:34, 7 August 2015 (UTC)
You may believe things were better in the old days. Possibly so. (And I should add I can think of some administrators, fortunately no longer active, who did serious thing and never faced the wrath of ArbCom) But the system exists by the collective will of the participants and trying to impose a change strikes me as futile. Whatever you may think about it, a sizable group of people, judging by the RfA just completed, feel that some content contribution is necessary. I agree with them. I do not think that requirement should be severe, not everyone's talent is in writing, but enough to point to and say "I helped do that, that the public will see and be informed by." I don't ask for a FA, FAs are labor-intensive enterprises that don't scale well on a community basis. Possibly you can get your way, but can you really change how people feel and how they will vote? I suspect the vote may not say the same thing, but it will be for the same reasons. Will anything actually change? With that, I have a watchlist full of things I must do.--Wehwalt (talk) 22:20, 7 August 2015 (UTC)
← For as long as I've been voting in RfAs, I've held that the process should default to granting the tools unless the candidate is proven inexperienced, untrustworthy, or generally at risk of chronically misusing the tools. It has always been my opinion that the onus is on the opposition to demonstrate why a candidate is unfit for the role, and I don't see why that should ever change; if an established editor in good standing volunteers to contribute in an expanded capacity, it would be nothing short of an insult to say "no, unless enough people can satisfactorily prove that you're a super-editor". *'''Support''', seen him around, does good work. ~~~~ is as valid a rationale as any. In fact, I don't have a problem with simply slapping your signature below the ===Support=== header. As Wehwalt notes, all of Wikipedia's vetting channels are constantly evolving to become more rigorous, and that's natural. For all the things that have changed, though, adminship itself is no more a big deal now than it ever was. Until someone shows me some evidence to suggest that we have too many admins overall, or not enough good ones, I think I'll continue to treat RfA as I did in 2008. I don't think I've ever seen an admin fail in his duties because he hasn't written enough GAs, made enough menial clerking edits at RFPP, or searched for a dispute to meddle in—requiring all those things of our admin candidates suggests strongly that we view a successful RfA as a reward for being the perfect editor, and not, more correctly, as a means of appointing a trusted custodian. – Juliancolton | Talk 17:18, 7 August 2015 (UTC)
I see that several bureaucrats and other editors still say that RfA is not a vote. I respectfully disagree. RfAs with 80%+ support pass, while those with less than 70% fail. (I suppose that someone might dispute these cut-off values.) In the middle ground, the closing bureaucrat makes a deciding vote. On rare occasions when an individual bureaucrat decides that the deciding vote is tricky to make, such as in Liz's RfA recently, the bureaucrat invites his colleagues to collaborate for the deciding vote.
Discussion during the RfA occurs too of course, and may sway some editors' votes. Axl ¤ [Talk] 11:50, 17 August 2015 (UTC)

Too many proposals going on at once[edit]

I can't keep it straight anymore. Right now, there's

To say my head is spinning is an understatement. There's so many threads going with so many proposals I can't even tie my shoes straight much less keep straight what is going on here. It's like the proposal world has suddenly exploded with supernova force. Uncle! Uncle! --Hammersoft (talk) 15:25, 6 August 2015 (UTC)

Here, a pair of slip-on shoes may help Typical Pair of Shoes (5375107159).jpg. -- Avi (talk) 15:38, 6 August 2015 (UTC)
I think the best way forward, which I addressed in User:BU Rob13/RfC for Administrator Re-election, is to evaluate each proposal first on their merits versus the current situation. Obviously, if something fails to gain consensus there, it's not worth moving forward with. We can then compare specific proposals aimed at similar processes/goals in later RFCs to construct an elimination bracket of sorts (maybe not the best turn of phrase to describe this, but it amuses me).
I think the issue, which certainly prompted me to begin writing, is that I feel that as soon as something like WP:BARC is accepted, we're locked in to that system even if a better proposal comes along. I'm very much feeling as if I have to get out my idea now or forever hold my peace, whether that's accurate or not. Given the difficulty with reform that I've seen others comment on, I don't think I'm too far off the mark. ~ RobTalk 15:52, 6 August 2015 (UTC)
  • Please understand I wasn't intending to cast any aspersions on the work you've done nor that of anyone else. I agree with you about BARC, and am fearful someone is going to take some mandate from that and attempt to move forward. It's fallen below 65% acceptance as of today, down from north of 70% a week ago. It continues to drop. The supports there are frequently equivocal as to what, exactly, they're supporting. I'm gravely concerned that RfC might send us down a pathway we can't easily recover from. --Hammersoft (talk) 16:19, 6 August 2015 (UTC)
Yes, RFA is broken, and has been broken, and will continue to be broken, so why are we still trying to ride it?

@Hammersoft: I agree; essentially the same discussion is going on everywhere, and really has been since at least 2006 when I first started editing, yet little in the RFA process has really changed much. The above "Is RFA broken" poll/RFC really is a great example of beating a dead horse.~ ONUnicorn(Talk|Contribs)problem solving 15:57, 6 August 2015 (UTC)

  • That's because the subtext of this isn't really that "there's something wrong with RfA's" – it's that there's something wrong with the way Wikipedia operates now. And that's a probably inevitable consequence of a fundamental change to the project that began c.2008, and majorly picked up speed in 2011. It basically boils down to something you have repeatedly said – the "big growth phase" of this project is over, and we're now into the "steady maintenance phase". And the kind of management structure you need for a "steady maintenance phase" of a project is very likely going to be different than the one that was needed during the previous "big growth phase". I just think it's taken ~5 years for that to start sinking in around here. But I get the impression that this year may be some kind of turning point where nearly everyone is starting to realize that this project can't go on the way it has in the past if it's going to continue to succeed. --IJBall (contribstalk) 15:59, 6 August 2015 (UTC)
  • Bingo, absolutely spot on. Unfortunately, getting it to change is attempting to have an afternoon tea and crumpets party with a hormone raged bull. It ain't gonna work. (not pointing fingers at anyone with this) The usual reaction to problems that arise here is to create ever thicker levels of bureaucracy. --Hammersoft (talk) 16:10, 6 August 2015 (UTC)
  • Well, that's the flipside of this – name a totally "volunteer"(-run) organization that isn't dysfunctional and poorly managed!... If anyone discovers one anywhere in the world, and studies it, I suspect it would lead the the Nobel Prize in Economics or something!! --IJBall (contribstalk) 16:20, 6 August 2015 (UTC)
    All kidding aside, I do not beleive it is the volunteer nature of Wikipedia as much as it is Wikipedia's size. In my opinion, Clay Shirky's essay A Group Is Its Own Worst Enemy should be required reading for anyone who wants to get involved with Wikimedia maintenance tasks. -- Avi (talk) 16:28, 6 August 2015 (UTC)
  • Well, here's a "proposal": Submit a request to WP:ANRFC that someone consensus-assess this entire page, and see if one or two points emerge from it that can be built upon. I'm not even kidding. I'd do it myself, but I've added enough things to that page lately someone else should do it.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  16:41, 6 August 2015 (UTC)
    • At best, such an assessment should be informal, as there have been a far greater number of participants outside of this page recently than there has been on this page, not to mention that those in opposition of RfA are more likely to come to talk in the absence of any RfC than those in support of it. ~ RobTalk 16:51, 6 August 2015 (UTC)
      • Well, of course it would be informal. The problem is that these discussions just archive away and recycle, instead of ever progressing anywhere.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  11:13, 9 August 2015 (UTC)

I understand why there is popular support behind Wikipedia's consensus decision-making tradition, but the enormous amount of discussion that occurs in a large group is one reason why it scales poorly. It would be hard enough to get a dozen people to read through all of the discussions listed at the start of this section; now try multiplying that by the dozens of editors that have commented in these threads. I've linked to Shirky's essay numerous times in the past and like Avi, I strongly urge anyone interested in managing online interpersonal interactions to read it. isaacl (talk) 17:22, 6 August 2015 (UTC)

I'll contribute a proposal of my own: everyone should stop typing out !vote unless it's drowning in irony. That stopped being cute in about 2009... – Juliancolton | Talk 19:12, 6 August 2015 (UTC)

!vote!vote!vote ~ ONUnicorn(Talk|Contribs)problem solving 19:34, 6 August 2015 (UTC)
!support --Stfg (talk) 19:47, 6 August 2015 (UTC)
  • !Julian for president! --Hammersoft (talk) 19:49, 6 August 2015 (UTC)
  • I am deeply shocked that we are proceeding without considering the 36 sub-proposals raised in the 2007 campaign for RFA reform. (I also remember an RFA reform argument from 2005ish but I can't find it). Face-smile.svg Manning (talk) 13:23, 7 August 2015 (UTC)
I'm not. Nobody generally thinks that far back. They don't even take into consderation the vast research that was done at WP:RFA2011 which some users loudly claim to be totally out of date even though apart from extending some tables into 2015 the actual trends are still exactly the same. The problem is that everyone is constantly complaining about the state of RfA and adminship but very few individuals are actually directly affected and even fewer are prepared to give it a go and propose at least something at RfC. And then all the participants do is try to shoot the messengers. It's rather disapointing that more people turn out to vote on one RfA than do for a major debate about policy. But then, policy discussions are not popularity contests or a reality show. Kudpung กุดผึ้ง (talk) 15:16, 8 August 2015 (UTC)
I know, I know, and I hope you realised I wasn't being particularly serious. But (as you obviously know) this has been one of the most intractable problems, going back a decade. My personal solution is drastically lower the bar to admin status, and then greatly enhance the ability of both the general community AND the admin community to desysop someone. That is pretty much the vision we had way, way back in the day. Easier to get and much easier to lose. Manning (talk) 19:55, 9 August 2015 (UTC)
I'm sure this one will stick...
  • This almost seems like a shotgun approach where there's the seeming hope something will stick. So you'll be there Hammersoft to give us a full report (joke)? Mkdwtalk 14:53, 10 August 2015 (UTC)
  • I'll issue a report as soon as I can swim to the surface of the paperwork. --Hammersoft (talk) 15:00, 10 August 2015 (UTC)

Too many proposals going on at once: current list[edit]

I believe that the following is the current list of proposals related in some way to to RfAs, in alphabetical order. Please feel free to update/correct as needed. --Guy Macon (talk) 09:29, 11 August 2015 (UTC)

  1. Wikipedia talk:Bureaucrats#RfC: Increasing the activity requirement for retaining bureaucrat rights, an RfC to potentially de-bureaucrat the inactive.
  2. Wikipedia talk:Reflections on RfX: a proposal to create an omnibus discussion on RfAs.
  3. Wikipedia talk:Requests for adminship#30/500 rule?: a proposal to limit who can !vote on RfAs.
  4. Wikipedia talk:Requests for adminship#Arbcom to take over RFA?: This time someone posted a comment with some random thoughts then someone else decided we needed Yet Another proposal.
  5. Wikipedia talk:Requests for adminship#Make recall criteria a mandatory question: a proposal to add a question regarding a candidate's recall criteria to the standard question set.
  6. Wikipedia talk:Requests for adminship#RfA: The Good, the Bad, and the Ugly Future: a discussion to highlight what's wrong with RfA.
  7. Wikipedia talk:Requests for adminship#RfC: Is RfA a broken process?: an RfC to determine if people think RfA is broken.
  8. Wikipedia talk:Requests for adminship#Split up the powers: another unbundling proposal. This is a perennial proposal that is being proposed once again.
  9. Wikipedia talk:Requests for adminship#Temporary Adminship before RFA: an idea to grant temporary adminship for a day before an RfA is run.
    1. Wikipedia talk:Requests for adminship#Temporary Adminship for one month: a similar idea to grant it for a month.
    2. Wikipedia talk:Requests for adminship#Temporary adminship during all RfAs with support from n current admins, a similar idea to grant it during the run of the RfA.
    3. Wikipedia talk:Requests for adminship#Wrong approach: Probationary period is preferable: another idea to have probationary adminship.
  10. Wikipedia talk:Requests for adminship#Totally different temporary adminship idea: Task-oriented: a proposal for temporary adminship based on certain tasks.
  11. Wikipedia:Administrators/RfC for BARC - a community desysopping process: an RfC about a supposedly lightweight process <cough> to desysop.
  12. Wikipedia:Administrators/RfC for binding administrator recall: an under-development proposal to re-elect administrators.
  13. Wikipedia:Requests for comment/Regulation Committee and alternatives to consensus: an idea to create an entirely new bureaucracy to determine consensus decisions.
  14. Wikipedia:Village pump (all)#RfC: Increasing the activity requirement for retaining administrator rights, an RfC to change WP:INACTIVITY.

Last checked/updated: 18:23, 11 August 2015 (UTC)


  • To create the above, I cut and pasted everything someone listed above, did a cursory search for more, fixed the bad links to, made sure I wasn't listing any redirects, replaced "_" with spaces, and alphabetized. --Guy Macon (talk) 09:29, 11 August 2015 (UTC)
I understand your concerns Guy, and it's an interesting list but most of those are just routine perennial threads on this talk page. This is probably also (a lot of the time) one of the busiest general talk pages on Wikipedia. The actual formal RfCs concerning adminship are ones about community desysopping. Other threads here and on other venues are probably people putting out feelers for what they might launch as RfCs at some time in the future, and as such , those discussions are necessary. They might possibly be ill timed though, as is the one RfC that has been started before BARC is anywhere near its closing date.Kudpung กุดผึ้ง (talk) 11:46, 11 August 2015 (UTC)
You probably want to address the above to Hammersoft in the section above this one. I don't have any particular concerns one way or the other; I just saw a list and made it more accurate. --Guy Macon (talk) 13:03, 11 August 2015 (UTC)
@Kudpung - Except that we are then necessitated to comment in each one, else someone will come along and decide something here has consensus (regardless of lack of community-wide participation).
I'm sorely tempted to start an RfC with the premise that all of these should be declared no consensus, and let's start over - centralising discussion a bit more. 1 or 2 or even 5, fine, but this, with a lot of varying time frames, all over various discussion pages is a bit much to expect full participation from all editors in a volunteer project. No matter how well-meant. And yes, I've held off refining and (re-)proposing WP:RRA for this very reason. - jc37 01:35, 14 August 2015 (UTC)
Why not close them all no consensus and take time to ponder the lessons of these? To paraphrase Heinlein, I wouldn't be sorry if you took a year or two to think things through and come up with a useful proposal, but I'd wonder at any proposal that came back in less than three months. We'll still be here.--Wehwalt (talk) 11:48, 14 August 2015 (UTC)
As I discussed elsewhere, if the community wishes to continue with consensus-based decision making, then it will have to accept that any changes can only be made with a lot of time and effort. One option is to empower a working group to hash through the issues and come up with a proposal designed to achieve maximum acceptance. This does require the community to be co-operative by empowering the group to perform the required analysis and investigation, and not be unduly skeptical of its efforts. isaacl (talk) 02:44, 15 August 2015 (UTC)
Consensus works fine for most things, and that includes reacting to new things. It has the huge advantage for a global site that it isn't sufficient to build a majority for something, you need to build a super majority. So when we do get solutions to things the vast majority are on board and generally they are better solutions. It works less well where there is a deep divide in opinion, and especially where two or more views are held by a blocking minority. As this site moves into its adolescence we've fixed most of the important things that can be easily fixed by consensus, and that has left us with a bunch of problems, most of which are difficult to resolve by consensus. When we come across something so important to resolve that we are forced to break the logjam one way or the other then with difficulty we can move to some sort of majority decision, but most of our difficult to resolve divides are things that the community has lived with for years, often over a decade, and which it is better to live with than to abandon our consensus model. So while I like your suggestion of having a group of individuals work on an idea that they can then present to the community, I don't think they should be elected, or have an expectation that they can determine things one way or another. Rather it needs people who will try to work across the divide and come up with a solution that resolves reasonable objections. That sort of approach can result in the sort of inelegant but working compromise that is WP:BLPprod, but to make it work well you need a group of people from very different sides of the debate who are willing to be frank as to what sorts of problems they are trying to fix, what situations they are trying to avoid, and most importantly, who are trying to come up with a solution that can get consensus, rather than bypass the consensus process by such means as multistage RFCs or closed RFCs that exclude alternative options and count as supports those who disagree with some of the details but think that can be amended later. ϢereSpielChequers 10:40, 17 August 2015 (UTC)
Sure, I didn't say anything about making the group elected or that they would determine matters; I explicitly said they should work on proposals designed to achieve maximum acceptance, meaning they should seek to satisfy as many persons as possible (or from the other point of view, result in the least dissatisfaction) and look for appropriate compromises. But this also means the community must be willing to work together towards a solution, living with some disadvantages and accepting incremental steps to improve rather than expecting everything to go one way at once. isaacl (talk) 18:48, 17 August 2015 (UTC)

History of RfA[edit]

It has been over a decade since we have been following the RfA. It has predominantly remained the same in structure (and not necessarily so in content). Has there been any documented history of RfA right from its origins (from what I understand, in earlier days, the discussions happened on mailing lists before the RfA became more formalized). I was just wondering if any stock-keeping has been done either quantitatively or qualitatively on the history of RfA. Guidance to any material such as wikilinks that at least refer to this would be greatly welcome. TIA. --Gurubrahma (talk) 12:23, 17 August 2015 (UTC)

Well that's just depressing. -- Ricky81682 (talk) 20:48, 17 August 2015 (UTC)
  • Well, I wrote a very brief summary of it here, but it's mostly centered around reform efforts and not the process itself. --Biblioworm (talk) 14:48, 17 August 2015 (UTC)

I am wondering if the questions for the candidates have always remained the same. Since when have additional questions come into vogue? Since when have co-nominations become prominent? Given the attention the RfA page generates, I am assuming that all this info must have been compiled somewhere, but I am not able to find it. Any pointers? --Gurubrahma (talk) 11:03, 18 August 2015 (UTC)

No, in the beginning there really weren't any questions. An editor would say "Hey, I'll do it", and unless anyone vehemently opposed the idea, they would be promoted. (Actually the very first group of admins in Feb 2002 didn't even volunteer, Jimbo just appointed about 20 of us). RFA only became a process when the project got too big (and Jimbo got too busy) to know everybody. Before 2004 we really didn't have any problems except for a admin named Isis who Jimbo desysopped when she threatened to sue Tarquin (an important early Mediawiki developer) for some reason. So RFA emerged really as a utility page. It began to develop into its current form by early 2005, at which time the first "We need to reform RFA" discussions also started. Manning (talk) 19:02, 19 August 2015 (UTC)
As an addendum to the above, here is a typical RFA from 2003. I do not believe there were any unsuccessful RFAs until 2004 (could be wrong on this point). Manning (talk) 19:26, 19 August 2015 (UTC)

An idea[edit]

(Now, I will probably be considered a disruptive editor for even mentioning another idea, especially one that has been proposed before, but oh well...) I've wondered for quite some time why we don't allow prospective RfA candidates to have a sort of "pre-RfA" (I'm also aware that the very same terminology has been used before...) to get an idea of what their chances are for passing the real thing. Currently, you really have no idea what will happen, and to find out you actually have to go through the one-week stressful process itself. For instance, I'm somewhat considering an RfA in the coming months, but I really have no idea what will happen, so in my case such a process would be quite helpful. I'm sure there's others who feel the same way. --Biblioworm (talk) 14:48, 17 August 2015 (UTC)

There used to be a process called Editor review that many used as sort of a pre-RFA. But it's marked historical. ~ ONUnicorn(Talk|Contribs)problem solving 14:56, 17 August 2015 (UTC)
In a similar style to Dennis Brown's pre-nomination RfA reviews? Sam Walton (talk) 14:57, 17 August 2015 (UTC)
I've seen such requests being posted on the talk pages of bureaucrats and other experienced users. Jo-Jo Eumerus (talk, contributions) 14:58, 17 August 2015 (UTC)
Of course the opinions of experienced users are very helpful (I've done that myself, as I mentioned below), but the issue is that you simply hear the opinions of one or two editors rather than the community as a whole, who in the end really has the final say as to who becomes an admin and who doesn't. --Biblioworm (talk) 15:14, 17 August 2015 (UTC)
The problem with this idea is that an RfA reflects the editors who show up and participate. You could have a successful pre-RfA and a miserable RfA because different editors choose to participate in each process. For example, editors who support you might come to the pre-RfA to encourage you while those who oppose you might only participate in the real RfA. Secondly, you don't know what aspects of your editing history people will take issue with. Meaning, RfAs are hard to predict. You could have an unsuccessful RfA, return in three months and have a successful RfA for a variety of reasons including more editors being familiar with you. Finally, an RfA can be a horrible experience. Having to go through both a pre-RfA and then an RfA is putting a candidate through a lot.
I think it is best to just ask an admin you respect (or two or three) to look over your contributions and assess your viability as a candidate. Preferably, these are admins who participate in RfAs so they have a notion of what general expectations are and where you could strengthen your editing background. I know a number of admins who are willing to do this service for potential candidates. Ultimately though, there are no guarantees of what you will face in an RfA. It's always a risk and you can only decide if you want to be a candidate and, if so, when you want to be a candidate...the voting and feedback you receive is out of your and your nominators' hands. But remember, even if you are unsuccessful, it's not fatal. You will learn a lot about yourself as well as what qualities other editors are looking for in future administrators should you decide to have a second RfA. Liz Read! Talk! 15:45, 17 August 2015 (UTC)
  • (edit conflict) Well, there used to be Wikipedia:Editor review, but that's been shut down. Slowly, bit by bit, areas of Wikipedia are shutting down. This is inexorable as the editing population continues to decline. I don't know there's been any sort of replacement of editor review. As for you, well you don't have 1x109 edits and you've been here less than 10 years. It's unlikely you'll pass. Also, your editing patterns have dramatically fallen off since the last quarter of last year. The picture of a latin dictionary on your userpage proves you really belong at Wiktionary, not on Wikipedia...much less the English Wikipedia. 42% of your edits have been done using automation, so your edit count is obviously way, way too low. One of your articles was deleted (Wikipedia:Articles for deletion/NutriBullet). Outside of that, you have less than 1x105 articles created. You've never been blocked, which means you haven't had to deal with controversy, making you as-yet inexperienced in conflict management on the project. Yes, even not being blocked now is a bad thing. You are quite obviously a disruptive influence on the project who can not be trusted with the tools. Maybe when it becomes easier to desysop someone we can consider passing a borderline case such as yourself. Until then, try to improve, maybe get blocked once or twice, start a few hundred articles and make sure at least 10% of them get to featured status without anyone having to help you. In case it's not obvious, everything from "As for you" forward is satire. Neverthless, it is at least partially true. --Hammersoft (talk) 15:01, 17 August 2015 (UTC)
    • ...and I also forgot to sign my post above. I'm clearly a newbie. Anyway, very entertaining post, Hammersoft, but seriously, I have discussed the possibilities with some experienced nominators, who (by my interpretation) gave tentatively positive responses if it were done several months from now. But this thread is not about me, it's about the idea; I just used myself as an example. --Biblioworm (talk) 15:14, 17 August 2015 (UTC)
  • Agreed, it's not about you (and I was happy to sign for you, you noob! ;) ). The satire was written to lambaste what are the all too frequent opposes around here. If such opposes didn't exist, we wouldn't have need of a "pre-RfA". Instead, we'd trust people who obviously have the bests interests of the project at heart (such as yourself). --Hammersoft (talk) 15:21, 17 August 2015 (UTC)
  • Samwalton9 mentions my review system above, which I created because there was no system in place and I felt a system that kind of mocked up RFA was best, where I could pick apart the candidate's weaknesses and allow them to defend each, or at least be prepared for the idea that others in the RFA would be asking. In short, I tried to really prepare the candidate, or give a good reason why they shouldn't run. The problem is that the candidate's tend to like this system, but few nominators have shown a willingness to do all this work. Sometimes, it seems that noms want to just slap their name on top of a "winning horse" and push them out the gate, hoping for the best, as the cost to them is minimal, but the cost to the candidate can be very high. I've only nom'ed 12 people, but rarely without a full review. My opinion is that often, nominators talk candidates into running when NO ONE has done a full background check. The check forces the candidate to be patient about when they get nominated, and to be open about previous problems. The same candidate that will fail with a bad nom could pass if he had a full background check, admitted his flaws up front, and was simply prepared for the type of criticism he will get at RFA. So who do we blame? In my opinion, a lack of due diligence by the nom(s) is often to blame, as there is no price for failure when noming. Dennis Brown - 16:12, 17 August 2015 (UTC)
Dennis Brown, is your review system one that candidates could be pointed at to run on themselves? valereee (talk) 16:16, 17 August 2015 (UTC)
I use a basic template that probably needs updating a bit, but no they can't run themselves. It requires that someone objective look at their history, do homework on it, review many of their edits, AFDs, CSDs and such, and "judge" them, so to speak. It requires the reviewer take the "supervoter" stance instead of the buddy stance and give an honest, hard review, in the same mind set as a skeptical voter at RFA. That is why it prepares them. Here are couple of examples of my reviews that are now admin: Go_Phightins! and Sam's, both of which I ended up noming for admin and both passed. Note that they used the same templates as RFA, same questions, plus extra examinations. This made it so they were familiar with it when they went to RFA. And no, I didn't coddle them, but I do give lots of advice. Dennis Brown - 16:31, 17 August 2015 (UTC)
Very nice, Dennis Brown. I can see you spent a ton of time on those; very cool of you to be willing to do that for candidates. Just looking at those reviews would probably be helpful to people considering RfA. valereee (talk) 16:51, 17 August 2015 (UTC)
  • I predict:
  • Oppose. Candidate had a pre-RfA review two months ago and was told to write more articles/get more AIV reports/comment in AfDs more often/make sure they have enough portal talk edits and didn't; oppose because candidate doesn't listen to community feedback.
  • Oppose This RfA comes too hard on the heels of the review five months ago at which the candidate was told to wait six months; obviously wants it too much.
  • Oppose Candidate has not had a pre-RfA review. We can't trust them if they're so uninterested in getting feedback.
  • Oppose Candidate commented negatively in my pre-RfA review.
All of these elaborate individual pre-nomination reviews are good feedback, but wow, those are exhausting. Not very "no big deal". Of course, the only thing stopping me from saying all candidacies should be self-noms is the fact that nobody would run that way. Opabinia regalis (talk) 17:25, 17 August 2015 (UTC)

RFA analysis[edit]

I am working on an analysis of RFAs from this year so far in my userspace here. I am done with getting the data but I still need to finish formatting the table. Help would be appreciated. Everymorning (talk) 16:58, 17 August 2015 (UTC)

Made it sortable, mostly because I wanted to see if I beat Ceradon for the lowest edit count :) Opabinia regalis (talk) 17:34, 17 August 2015 (UTC)
Hehe --ceradon (talkedits) 17:39, 17 August 2015 (UTC)
  • Added wikilinks to the rfas being referenced. --Hammersoft (talk) 17:48, 17 August 2015 (UTC)
  • Hey, it might be beneficial to add data since 2011 (since that was when the most recent data was compiled) I'll start working on that. --ceradon (talkedits) 17:59, 17 August 2015 (UTC)
  • May I suggest that months since last block would be a more useful variable than self nom, and manual edits a more useful statistic than the one you are using which I think is raw live edit count, (not counting deleted edits). There was at least one editor who opposed all self noms a few years back, but he is long gone, and while I doubt we can agree how much one should discount automated edits, and thankfully people have stopped opposing for the percentage of automated edits, there is an argument that five hours spend doing a thousand edits don't necessarily teach you more about this site than fiven hours doing fifty manual edits. ϢereSpielChequers 18:10, 17 August 2015 (UTC)
  • I have the set of RfAs since 2008 scraped from the tables in the successful/unsuccessful lists and reassembled into one big chronological list. I never did get around to putting together those tables for the previous years, but there's a pretty straightforward, albeit tedious, script job if anyone wants it. Other suggestions:
  • Keep track of as many features as you think might be interesting rather than trying to guess a priori which plausible features will matter most.
  • Keep the table entries formatted in a script-readable and -writable way; i.e. have a consistent set of responses for each column and format all free-text type comments the same way.
  • Keep the "time editing" list in months instead of fractions of a year, which will get very confusing if you go back far enough. Opabinia regalis (talk) 19:33, 17 August 2015 (UTC)
  • With respect to the time editing being in months not years, I think you're probably right. When I was working on this in Excel earlier today, I thought at one point that I should change it but I didn't know how to multiply everything in one column by 12. Everymorning (talk) 19:42, 17 August 2015 (UTC)
The element that seems more obvious than I imagined is how difficult it is to have a successful RfA if you've ever been blocked unless it was somehow an accident or mistake. Do editors who had a block or two during their early years just decide not to become a candidate? I would have thought that editors would be forgiving but it's hard to ignore how most successful candidates (and even some unsuccessful candidates) have clear block logs. It's not a requirement but it is definitely an obstacle that a candidate has to overcome. Liz Read! Talk! 19:54, 17 August 2015 (UTC)
It shouldn't be Liz. Admins who are blocked do not necessarily lose adminship status, so blocks pre and post RfA should be treated equally. Moriori (talk) 01:42, 18 August 2015 (UTC)
Most unsuccessful candidates have clean block logs, but that's also true of most core editors, there was a study a while back of the top 1000 editors and only 30% had ever been blocked, and the vast majority of those would be treated as clean at RFA - for starters a quarter of those 30% were accidental blocked by people who followed their block with a very quick unblock and apologys. It used to be that RFA candidates needed to be 12 months block free, now you might get some opposes for events even older especially if you don't explain why you've learned from the incident and it won't recur. ϢereSpielChequers 06:47, 18 August 2015 (UTC)
ONLY 30% have been blocked???!!!! - that’s a huge number. Unfortunately being blocked is still a stigma at other areas of Wikipedia, for example blocked users cannot be listed at wp:Missing Wikipedians. Ottawahitech (talk) 12:53, 18 August 2015 (UTC)
Very few of the 30% had a block that would have been seen as a problem at RFA, I think the third most common reason was edit warring, and I would be surprised if anyone opposed over that unless the block was less than a year ago. As for missing Wikipedians, I thought the policy there was not to list editors who were currently blocked, but I see exceptions are made even to that. ϢereSpielChequers 20:12, 18 August 2015 (UTC)
Would it be possible to have a column that contains the month of the year of the RfA? It would provide an indication of whether there were several RfAs going on during the same period of time, if it had been a long time since a successful RfA and if the time of the year has any influence. I don't think it does but data is more revealing than ones personal impression. Liz Read! Talk! 19:58, 17 August 2015 (UTC)
  • Another trend that I have only noticed recently is interest in content work. Many editors want to know if the candidate has enough content work/article creation experience. Ottawahitech (talk) 20:51, 17 August 2015 (UTC)
    • It's been around a while actually. It seems to me (at least from the cases I've seen) that it tends to be more persuasive as a "support" argument than as an "oppose" argument going by outcomes. Jo-Jo Eumerus (talk, contributions) 20:58, 17 August 2015 (UTC)
  • For some reason in the 2014 section on the page in my userspace, the edit count sort at the top of the column is not working. It would be great if someone could fix it--I thought it was the commas but apparently that's not it. Everymorning (talk) 02:06, 18 August 2015 (UTC)


Scottywong used to be our data magician for admin stuff but he has sadly retired along with a few other good admins from the 2011-12 harvest. With her renewed interest in this dark area of Wikipedia Opabinia regalis seems to be a whizz kid at producing charts and tables. I'm sure she would help if asked for anything but she would need to be convinced that the stats are really needed and will be used. Kudpung กุดผึ้ง (talk) 19:52, 19 August 2015 (UTC)

I made some suggestions above, but the enthusiasm of willing manual curators is a scarcer resource than the ability to fill in any gaps they might leave :) What surprises me here is that apparently there's nobody in San Francisco who wants to show off some data science chops on this. Opabinia regalis (talk) 23:23, 19 August 2015 (UTC)
A lot of work has also been done by NoSeptember (see his subpages here) and WereSpielChequers (see here for subpages). A lot of data has been collected over time but it hasn't been consistent over the past 14 years. When the admin corps were doubling in size (2005-2007) there was a great interest in collecting data and charting the increase. This also makes sense as many editors are just active for a few years and collecting longitudinal data is an investment of time. Liz Read! Talk! 23:39, 19 August 2015 (UTC)
This is one of those areas where a "Bot author" is needed. (Other places where Bot authors are needed is to find potential "Autopatrolled" candidates, and to count GA authors...) I really wish I knew code so I could write Bots, because I'd love to come up with a Bot for all of these jobs, including RfA analysis. But, alas - I'm not a "coder"; I'm just a schlub. Face-sad.svg --IJBall (contribstalk) 05:38, 20 August 2015 (UTC)

I hope...[edit]

...that the recent Ceradon incident won't be forever held up as an example by edit-count-crazy opposers, who might say something like: "This proves beyond any doubt that to prevent potential 'malicious socks' from becoming admins, we must require years and years of continuous editing and tens of thousands of edits." Certain voters seem to find a reason to oppose every time, using examples such as this to justify assuming bad faith. I just wanted to bring this up, so that any one who tries to use this in the future could be made aware of the community's opinion on such opposes. --Biblioworm (talk) 21:17, 20 August 2015 (UTC)

I fear that this is encouraging WP:BEANS, but strongly agree that anecdotal evidence based on a prejudice would not be enough to oppose a future candidate. Bilorv(talk)(c)(e) 21:19, 20 August 2015 (UTC)
It's hard to imagine a situation similar to Ceradon's occurring again. But any editor can vote for any reason in an RfA and I've seen some fairly strange reasons given for casting a vote one way or another. I don't think there is any way to prevent this case (or any other one) being used in the future to justify a vote. Liz Read! Talk! 22:29, 20 August 2015 (UTC)
This sort of thing has been happening for a long time. One of my RfAs (back when I was too green really to be a good admin, and before I learned I don't really want to be one) was totally derailed by false claims made by a sock puppet. The claims were dogpiled on the instant they were made, and it made no difference at all that the sock was rapidly outed as a sock (and my alleged WP:BITE of him disproven); the damage was already done and irreparable. This sort of thing is one of the reasons many of us want some proof of long-term editing before people are allowed to vote at RfA.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  01:15, 21 August 2015 (UTC)
There were at least two opose votes on my RfA that were based on deliberate lies and/or issues taken deliberately out of context to discredit me through my RfA for having been perfectly right (and civil) in discussions elsewhere. One has since been desysoped and the other has retired. There were a couple of others whose comments were less than civil but who later retracted their opposes. The damage however, was already done; those who simply piled on never came back to adjust their opinions. If ever clerking were to be introduced, or if ever the 'crats could see their way clear to extending their mandate just a tad, such pile ones, if not reverted by their users, should be indented just before final closure. Kudpung กุดผึ้ง (talk) 06:00, 21 August 2015 (UTC)
Yes. That issue of "piling on" is at the heart of it, I think. The trouble is that, while we all piously say that RfA is a discussion not a vote, the discretionary range is so narrow that it really is a vote in practice. If someone comes up with an oppose reason that's strong enough to make me believe that an editor (however good as an editor) is unsuitable for adminship, I can't sit it out on the basis that what needs to be said has already been said, because of this voting element. And I can't just say "per Bob", because those are disapproved of. So basically I have to repeat the link and the argument, or at least point to Bob's vote and say "I agree with Bob because ...". Is that piling on? How can I let my voice be heard without having to do that? Solve that conundrum, and I think RfA would instantly become nicer. --Stfg (talk) 12:04, 21 August 2015 (UTC)
I sometimes make votes that cite other users' votes - I did so on the Liz RfA for example. However, those other voters were echoing what I already knew. I agree that it can be difficult to be sure about pile-ons, but I think generally they don't check to see if the voter is right. I'm very wart of those who ay 'Per X' and nothing else, although the examples on my RfA were much more complex and spiteful, Kudpung กุดผึ้ง (talk) 00:28, 23 August 2015 (UTC)
"I just wanted to bring this up, so that any one who tries to use this in the future could be made aware of the community's opinion on such opposes." – Biblioworm. If the "community's opinion" is contrary to that oppose argument, then there should be plenty of support !voters to counteract the oppose !voter. Axl ¤ [Talk] 10:27, 25 August 2015 (UTC)
The problem of not having a set criteria is that you don't add additional criteria by consensus and a super majority wanting that additional criteria. In RFA if 30% of participants will oppose for a particular factor then whatever the other 70% think, the 30% have imposed an additional criteria onto the process. ϢereSpielChequers 09:21, 26 August 2015 (UTC)
Every individual !voter imposes their own criteria onto the process. Axl ¤ [Talk] 10:59, 26 August 2015 (UTC)
Yes, and once 30% of the community adopt a criteria they can impose that on the whole process. Other consensus based systems such as AFD take place in the context of a set of criteria that can only be changed by consensus. So an admin can close an AFD as delete if just one editor points out that the sources being offered are for the other footballer of the same name, or as keep even if the AFD is dominated by arguments that this is the English Wikipedia and we should ignore non-english sources. ϢereSpielChequers 15:57, 26 August 2015 (UTC)
"Other consensus based systems such as AFD." As I have previously mentioned, RfA is a vote. In my opinion, if 30% of the community have decided that a particular criterion is important, that is a legitimate reason to oppose. Even though 70% of the community may have decided that the criterion is not important, that is not justification to disregard (or "downgrade" as Dweller would say) the minority viewpoint. Axl ¤ [Talk] 14:24, 27 August 2015 (UTC)

From WP:ADMIN: "A discussion (not a vote) takes place for seven days about whether the candidate should become an administrator. This allows other editors to get to know the candidate, and explore the candidate's involvement and background as an editor, conduct in discussions, and understanding of the role they are requesting, and to state if they support or oppose the request, along with their reasons and impressions of the candidate. An uninvolved bureaucrat then determines if there is consensus to approve the request."

Note how it says it is a discussion and not a vote. It comes down to consensus, not a count. From WP:CONSENSUS: "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."

I would say that the crats' are very much correct to give less weight to any argument that has no basis in policy. The idea that people can just make up whatever criteria they want is flawed. Imagine if we gave equal weight to people who voted "Delete I don't like articles about <foo>" at AfD? I don't know where the idea came from that RfA is a free for all but the fact is that our current policy calls for filtering debates based on their basis in policy.

I think that if people want to make a criteria for being an admin then they should try to seek consensus for that criteria to be reflected in policy. Chillum 14:30, 27 August 2015 (UTC)

I know what the official policy says. Axl ¤ [Talk] 14:39, 27 August 2015 (UTC)
I guess the point I was trying to make was that RfA would be a lot less messed up if consensus was better viewed through the lens of policy. While it is essentially a free for all now, this is not what the community asked for when creating the policy. If RfA is indeed a vote, then we are doing it wrong. I think the crats' would do well to openly give little weight to arguments that lack a basis in policy, much as a closer at an AfD or any other debate would do. Chillum 14:50, 27 August 2015 (UTC)
As I alluded to earlier, the evaluation of consensus, generally speaking, isn't being done correctly in nearly all discussion threads involving a large number of participants. The structure of most of them—a list of people stating their preferred option, with threaded conversation underneath each—is not conducive to determining consensus. A participant needs to keep track of dozens of threads, often covering redundant territory. There's a reason why the typical bulletin board format remains popular on the web: there is one place to continue a conversation, and catching up is a simple matter of continuing after the last post you've read. Consensus decision-making should not be done via a straw poll and then trying to re-balance the votes to favour one option. It should decide what option will satisfy the most persons, including the large silent majority to whatever extent is possible, or, conversely, dissatisfy the fewest, since often the best choice is just the one that manages the negatives sufficiently. To that end, these discussions need to solicit opinions on what characteristics are desirable and undesirable in the option to be chosen, and their relative importance. That way, someone summarizing the discussion can determine that benefits A, B, and C are of highest priority, outweighing drawbacks D and E, and thus select the option that is most suitable (or in the case of a request for adminship, determine if the candidate is suitable based on the advantages versus the disadvantages). isaacl (talk) 16:39, 27 August 2015 (UTC)

"Discussion" section[edit]

There are concerns in Ceradon's 2nd RfA regarding the prominence of comments made in the Discussion section. This has been previously raised, particularly in Rich Farmbrough's recent RfA when comments I made in the discussion section were felt to be too prominent. Sometimes people wish to make comments, raise points, etc, that they feel are relevant, and should not be hidden away on the talkpage, but which are too general to be in the Neutral section. If the Discussion section being placed above the Support/Oppose/Neutral sections is felt to be too prominent, would it be felt more helpful to place it below those sections? SilkTork ✔Tea time 22:02, 20 August 2015 (UTC)

  • I think this would be a good idea. Everymorning (talk) 22:04, 20 August 2015 (UTC)
  • Personally I thought, and still think, that the current discussion should be moved to the Talk page. People comparing the candidate to their own experience, back-and-forth conversations, now discussions about the discussion - these belong at the talk page and not on the main RfA page IMO. --MelanieN (talk) 22:10, 20 August 2015 (UTC)
  • Hmm, I objected to SilkTork's discussion-section post but posted one of my own this time, because, ahem, I'd forgotten about that. Good idea. I'm not sure why there should be general 'discussion' section and a talk page often used for the same types of comments, though. Opabinia regalis (talk) 22:14, 20 August 2015 (UTC)
  • IMO, the relative ordering of the sections doesn't matter much (do people really read the RFA page linearly?). However I would suggest simply eliminating the discussion section as currently organized and replacing it by a (collapsed?) Table of Content of the RFA talkpage. That would simultaneously make the latter more visible overcoming the problem of relevant comments being hidden away. Abecedare (talk) 22:15, 20 August 2015 (UTC)
  • Restricting the main page to simply support/oppose statements really undercuts the idea that the Request for Adminship process isn't supposed to be a vote. If the process is supposed to involve discussion to establish a consensus of the pros and cons of a candidate, a space for consolidated discussion is needed. It should not be necessary to pursue a given point in response to multiple participants. isaacl (talk) 22:21, 20 August 2015 (UTC)
    • I agree that such a space is required. I just don't see the "Discussion" section on the RFA page being that place, since (1) it is inconvenient to have lengthy, sectioned discussion there, and (2) it is not clear how one is to decide whether a post belongs in that section or on the talk-page. Better to simply use the RFA talk-page for such discussions, and raise its visibility if desired. Abecedare (talk) 22:28, 20 August 2015 (UTC)
      • Frankly I think it is more inconvenient to have a long list of votes, for a process that isn't supposed to be a vote, rather than focusing on discussing the candidate's qualifications, without unnecessary forking of the conversation into redundant threads. isaacl (talk) 22:42, 20 August 2015 (UTC)
  • In my opinion it is a discussion and not a vote. The discussion should not just be among the votes, the "votes" should be the discussion. I think a better format for RfA would be 2 days of unstructured discussion follow by 5 days of "voting". Chillum 22:25, 20 August 2015 (UTC)
    • It's something I suggested above. At some point I plan to write a bit more about how the current process and how it's implemented is not conducive to determining consensus. isaacl (talk) 22:39, 20 August 2015 (UTC)
  • I know I posted a remark in the discussion section of my RfA, in part to respond to comments that had already been made in this area. But I think that the talk page is a better place for threaded discussion. I think I have seen old RfAs where the Support/Oppose sections were just a list of names and then at the bottom of the page was devoted to discussion. Liz Read! Talk! 22:26, 20 August 2015 (UTC)

I think it would be fine to have a "General discussion" below Neutral, but please consult with Cyberpower678 before implementing such a change to the RfA template to ensure we do not break Cyberbot678. –xenotalk 22:35, 20 August 2015 (UTC)

MelanieN's suggestion is interesting, but it would make it more difficult for bureaucrats who would need to flip back and forth between the main RfA page (which I assume would carry just the nom and q/a's?) and the talk page with the participant's commentary as they divined consensus. This concern is not a deal breaker. If we move the main discussion to the talk page, we have to find somewhere else to move the extended off-topic conversation that sometimes occurs and is shuffled off to the talk page. I suppose it would just get moved from after someone's comment down into one of the subsections of the "other discussion" below. In this way, I'd worry we'd have some RfAs devolving into long 'headerfests', though. So I'm not sold on the suggestion to move participant discussion to the talk page. –xenotalk 22:35, 20 August 2015 (UTC)

I believe MelanieN (and Opabinia and I ) are talking of only eliminating the discussion in the "lead section" that currently contains the {{RfA/RfB Toolbox}}. The Support, Oppose and Netral sections would remain unchanged. They can correct me if I misinterpreted them. Abecedare (talk) 22:40, 20 August 2015 (UTC)
It shouldn't break the bot if I recall correctly. Go ahead and move and see what happens. If you hear a nuclear explosion off in the distance, the bot may have broken after all.—cyberpowerChat:Online 22:42, 20 August 2015 (UTC)
Ok, I've gone ahead and reformatted the RfA to see if it breaks the bot (special:Diff/677068232). If anyone (including the candidate) feels this is problematic from a procedural standpoint, feel free to revert. –xenotalk 22:49, 20 August 2015 (UTC)
(edit conflict) Replying to Xeno and Abededare, I wasn't suggesting that we completely eliminate the section called "Discussion" from the RfA page. And I certainly wasn't suggesting that the "support" "oppose" "neutral" sections be removed. I was just saying that IMO the current discussion there should be moved to the talk page. That has been the usual procedure when such a discussion starts to get in the way. (Since RfA has no clerks, some passing admin or even non-admin usually just moves it.) As for the "discussion" heading, maybe we should move it to the bottom of the page. But IMO that section can be useful when used for one-off comments, particularly if they are not pro- or con- debating, but are more along the lines of providing information. --MelanieN (talk) 22:59, 20 August 2015 (UTC)
The issue is that the "current discussion" (if we're talking about Guy's comment and that which followed) was relevant commentary about the candidacy. The talk page is for discussion about the page overleaf, not necessarily the candidacy (so extended meta talk about a particular oppose that starts to be more about the participant than the candidate, etc.). Many people don't look at the talk page, so placing relevant commentary about the candidate/candidacy there probably isn't the best idea. — Preceding unsigned comment added by Xeno (talkcontribs) 10:42, 21 August 2015 (UTC)
I don't hear an explosion, so I'm assuming the bot's coping. :p—cyberpowerChat:Online 23:20, 20 August 2015 (UTC)
  • Support moving to bottom of page, not talk page. The talk page is usually ignored.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  01:10, 21 August 2015 (UTC)
  • I've 677147130|reformatted the RFA base template to put "General comments" as a fourth section below the three numbered list sections. –xenotalk 12:32, 21 August 2015 (UTC)
  • It seems past the point now but I would support moving the discussion section to the bottom. I've always felt it's been under utilized so maybe now it'll be out of the way but still warrant activity. Mkdwtalk 15:33, 21 August 2015 (UTC)
  • I support limiting comments to the appropriate support/oppose/neutral sections or the talk page. The general comments section should only be used for neutral housekeeping-related discussion, not discussion of the candidate's history or fitness for adminship. No objection to moving the general comments section to the bottom.- MrX 19:42, 22 August 2015 (UTC)
    • Typically for Wikipedia pages, though, housekeeping discussion is held on the corresponding talk page, and personally I can't think of a good reason to diverge from this practice for Requests for Adminship. isaacl (talk) 22:03, 22 August 2015 (UTC)
      • Myself, I'd like more to have discussion about the page (the aforementioned housekeeping discussion) on the talkpage and discussion about the candidacy, particular opposes/supports and the like on the actual RfX page itself - these are directly pertinent to the process and may be better off on the page rather than shuffled off to another tab. Jo-Jo Eumerus (talk, contributions) 22:21, 22 August 2015 (UTC)
    • (edit conflict)Since IP editors may not comment in the support/oppose/neutral sections, this would have the effect of banishing their comments from the page altogether. This was proposed recently, and the clear consensus was against such a change, if I recall correctly. In any case I think it is usually better for threaded on-point discussion to be in the General Discussion section, indeed I wouldn't mind an explicit rule against it in the support/oppose sections. DES (talk) 22:24, 22 August 2015 (UTC)
  • I've been saying discussion should be below voting section for ages but no one listens. That would make it conform to how we do other discussions, even at ANI for goodness sake. Dennis Brown - 23:26, 22 August 2015 (UTC)
  • I think the 'Discussion' section should be at the bottom. This is not to be confused with moving long vote threads to the talk page. That said, commentsare rare, let's not let this ceome a WP:BEANS. Kudpung กุดผึ้ง (talk) 00:33, 23 August 2015 (UTC)
  • I think having the discussion section below the support/oppose/neutral sections would be a good idea. Bilorv(talk)(c)(e) 08:58, 23 August 2015 (UTC)
  • I'm also supportive of the move. There are occasional important comments or clarifications that need to be made on the RfA page itself which are typically placed on the discussion section, so I don't think moving to the talk page exclusively is a good idea. Sam Walton (talk) 09:11, 23 August 2015 (UTC)

Define "failing"[edit]

I was reading over some things related to RfA and saw a comment that said RfA (in general, not a specific RfA) is "failing". So, if you would please: If you believe RfA is failing, please define 'Failure' in so far as it applies to the RfA system. I'd prefer if we didn't go skittering off into multi-layered discussions on this. Just a definition, thanks. --Hammersoft (talk) 18:37, 24 August 2015 (UTC)

  • "Failure" here describes a process which does not promote enough successful candidates and/or does not promote the right type of candidates. Some types of failure may be due to external influences (e.g. number of Wikipedia contributors dropping in general), while some may be the fault of the process itself (e.g. the stigma of RfA prevents good people from applying). Bilorv(talk)(c)(e) 18:44, 24 August 2015 (UTC)
Are we sure that RfA is 'failing'? Perhaps not everyone is following the right discussions. IMO it does what it says on the tin, i.e. Those who should pass generally do, while those who generally should not, don't (without prejudice to rare close calls). As for definitions, based purely on my own observations, it is possible that the risibly strict criteria insisted upon by some voters (not all them even regulars at RfA) may have somthing to do with it, but in fact the vast majority of RfA have a very clear consensus anyway. Perhaps some potential candidates are put off from coming forward due to RfA having a reputation of not being a very nice place. That reputation, in spite of some who insist otherwise, is no longer strictly true today and needs to be quashed.
Many members of the community suggest that if the bar were to be lowered there would be more candidates but at the same time, they insist that if the bar were indeed to be lowered it must be made easier and more effective to address issues of admins who violate the trust the community has put in them. Ironically, all efforts by the community to address these points appear to meet with resistance from the very same sectors of the community who call for them - those who complain loudest and longest do little to to come up with any concrete suggestios and present them to the community. If Hammersoft really wants to know the answer to his original 'fail' question, perhaps if he were to run for adminship himself he would find out out. --Kudpung กุดผึ้ง (talk) 00:34, 25 August 2015 (UTC)
I believe at some point someone got some hard numbers on the rate of admin promotion vs attrition. If you only look at active admins it is clear that we need to increase the rate at which we select them. Chillum 00:39, 25 August 2015 (UTC)
Those charts are produced and maintained by WereSpielChequers. They display raw numbers. They do not (and are not intended to) demonstrate that we actually have too few admins to share the work or that we are headed for such a situation. In fact the steep dip in RfAs since 2007 now appears to have bottomed out. A lot of the work of admins has been relieved by unbundling and bots. Among Wikipedia editors are some who suggest that some further form of unbundling might be appropriate. It would need a RfC to find out. Finding users willing to launch RfCs appears to be problematic - unproductive repeaed complaints about the same old things seems to be more the order of te day. Kudpung กุดผึ้ง (talk) 01:55, 25 August 2015 (UTC)
Chillum Kudpung There are seasonal variations and sizeable problems such as not knowing how many admins we need, and stats that treat admins who do a hundred admin actions a day the same as admins who do a few hundred edits a year but rarely if ever use the tools. However I do have various charts at User:WereSpielChequers/RFA stats. Three things I would highlight, we only appointed 22 admins last year and we ended last year with 48 fewer active admins than last year. Neither is good. And sadly, I can't agree that the rate of new RFAs has yet bottomed out. Yes there was a slight rally in 2013, but otherwise the number of new admins has fallen every year since 2008 and this year could still be a new low. But for the situation to stabilise we need RFA to appoint as many admins as we lose, and that means going back to the numbers we were appointing in 2011 if not before. ϢereSpielChequers 18:31, 26 August 2015 (UTC)
  • I believe firmly that RFA is failing to produce the number of administrators that are *useful* to maintain this website. In fact, we are producing so few administrators, and the attrition rate is so high and the activity rate so low, that the website is showing obvious signs of entropy. It does not matter how many "tools" are devolved. The core ones, blocking and deleting, are only ever going to be part of the admin toolkit. People used to actively participate in deletion discussions specifically to develop sufficient understanding to be eligible for adminship. Now it doesn't matter, because the chances that they'll even be considered at RFA are so minuscule. We have allowed the "standards" to creep up so ridiculously high for no good reason that hardly anyone is interested in putting themselves through the process. The community failed to fight back against the loudest voices, as is the usual pattern in any large online community, and now Wikipedia is failing too. Risker (talk) 02:37, 25 August 2015 (UTC)
  • First off, "broken" and "failing", when applied to RfA, are overused. Even after a decade of constant grievances and discussion, you would be hard-pressed to find any significant changes from the RfA of 2005, and you'd be even more hard-pressed to find any substantive RfA reform proposal that passed.
But to answer the question, here are the perennial complaints:
  • RfA is too hard to pass; people with fewer than 10,000 edits are cast into a lake of fire; one CSD mistake and you're not touching that mop.
  • RfA does not promote enough admins.
  • RfA sucks, without providing a reason.
  • RfA demoralizes candidates.
  • RfA is not fair.
  • And so on. Esquivalience t 03:01, 25 August 2015 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── There are other reasons too: Adminship is too big a job, with too many negatives, and not enough positives, and requires too much time. In short, the minuses for going for Adminship massively outweigh the pluses, under the current system. This gets back to the 'unbundling the bit' idea. Wbm1058's current RfA is ironically pointing up the folly of the present system: basically, Wbm1058 just wants "Page mover" rights. (Incidentally, that's right where I am too...) But people are opposing Wbm1058 because Wbm1058 doesn't have AfD experience, etc., even though Wbm1058 has clearly stated that they have no interest in article deletion, just page mover stuff. Unbundling the bit is the solution here – people like Wbm1058 would breeze through a request for "Page mover" rights, and you might actually get people like me who would probably never submit to an RfA (and I think there are increasing numbers of editors who feel the same way) to sign up for something like "Page mover". I'm surprised people are so whetted to the current system when it's clearly in the process of failing (both Adminship, and Bureaucrat). We have 6,000+ Pending Changes Reviews. Imagine if we could get ourselves 2,000 Page movers, 500 Article protectors, and so on... --IJBall (contribstalk) 04:27, 25 August 2015 (UTC)

  • The problem with unbundling, as I see it is that the tools often interact. Page Mover means the ability to delete target pages, and not infrequently to do a proper job one must be able to see deleted revisions. Moreover, It is often bound up with doing or considering history merges, and those require the ability to delete and undelete specific revisions. I would not be willing to trust that toolset to someone who had not passed a level of scrutiny comparable to an RfA.
Similarly a vandal fighter who wants to be able to block vandals but not to engage in deletion really needs to be able to see deleted pages to fully judge an editors past activities. Moreover, it would be helpful if not essential that such a person be able to speedy-delete at elast copyvios and attack pages.
It is, IMO, really hard to find a clear role that involves one of the admin tools that doesn't involve several, or require nearly the same level of trust as the full toolset. Template editor was a good spin-off, and just possibly Page Protector could be, but often someone doing page protections needs to be able to block or unblock as well. DES (talk) 15:51, 25 August 2015 (UTC)
This definitely comes up as a counter-point to the "unbundling" idea – that some of the User group rights might be redundantly handed out to different "groups" (e.g. Page mover, AfD moderator) – but I don't see that as a "killer" rebuttal to the concept. Incidentally, it looks like Jc37 has already done some of the work on the concept – at Wikipedia:Moderators/Proposal#List of user-rights included in the user group – and I'm not seeing much overlap there, though I'm uncertain if Jc37's list there is all inclusive... --IJBall (contribstalk) 15:46, 26 August 2015 (UTC)
IJBall, I worked a while honing that list. If you think it can get past the "we think blocking should be unbundled first" crowd, or the related: "If we implement this, I'm afraid I won't get consensus on some proposal I like, (or the "I don't trust the system (define "the system" however you like), and I fear what adding this to the mix will do"), etc., then I'd be happy to rewrite the proposal again. It's actually an easy implementation. due to an update awhile back, making user groups is not a difficult thing in the system now. - jc37 18:50, 27 August 2015 (UTC)
@Jc37: I definitely would like to see 'unbundling', with approx. four new user groups (Vandal fighter, Article mover, Article protector, and "AfD Admin" (or something like that...)), eventually proposed and implemented, because I do think the current concept of Adminship is actually slowly "failing" (RfA is just a part of that). But I personally suspect it's not quite the right time to propose this – I think there would be still too much opposition among the current Admin corps (as well as from some of the long-time regulars; and possibly some partial opposition from the WMF though I really don't know the details on the last one...). I think Wikipedia needs to "degenerate a little more", with growing backlogs and such (and possibly several consecutive months with no Admin promotions), before the community consensus will be wide enough to implement such a proposal because the "old way" of doing things is just plum no longer working for the project. So, on my end, I'm picking away at the idea, and maybe trying to collect some data along the way. (Your page will be invaluable on that, incidentally – I'm probably going to "steal" your good ideas there... Face-wink.svg ) But I think, right now, we're still in the "waiting game" phase before actually proposing this... --IJBall (contribstalk) 19:21, 27 August 2015 (UTC)
  • I rarely edit any more but I do check in here on a weekly basis and I think Risker's statement is spot on. We're promoting admins, but just at a far lower rate than needed. The relatively easy admin jobs are still being done, but the number of people willing to take on the hard ones has dwindled drastically. People get burnt out and no one steps in to replace them. I don't actually share the notion that RFA is any harder to pass now than when I ran (in 2006). It's just that we have far fewer people willing to go through the gauntlet. The current circle jerk is about content creation, but at other times in history it's been about ratio of article to project space edits, it's been about being too social, not social enough, even things as stupid as percentage use of edit summaries. It's always been a ridiculous exercise of jumping through hoops and humoring whoever wanders in with random questions and requests. It's failing because no one wants to go through it any more. --Spike Wilbury (talk) 01:44, 26 August 2015 (UTC)
You are right in that RfA goes through a cyclic pattern of voting trends. What we must not forget however is that tha significant number of those voters rarely if ever vote at RfA (proven by stats). What is wrong are the assumptions by some that RfA is still the bloodbath it was up until a couple of years ago. Those who are still loudly proclaiming that it is are doing us all a misfavour, and depending who they are, it's hardly surprising the potential candidates believe them. Kudpung กุดผึ้ง (talk) 02:37, 26 August 2015 (UTC)
Kudpung, I don't follow: "tha significant number of those voters rarely if ever vote at RfA" --Spike Wilbury (talk) 16:34, 26 August 2015 (UTC)
There are some aspects of RFA that are harder to pass than in 2006. The minimum edit count is much higher, tenure requirement is higher. Other things are easier - few actually look at your edits so it is easier for some candidates who would have had difficulty in 2006. On the other hand the use of stats and other superficial substitutes for properly checking edits has resulted in some very sloppy opposes. One RFA earlier this year had people opposing for the candidate recently having some articles deleted on grounds of notability, despite the fact that the articles were created several years earlier and the candidate hadn't objected to their deletion. Generally I think this has been an unhealthy development, there is more emphasis on things that don't matter less scrutiny of the actual edits and I fear RFA is getting less effective at differentiating between good and bad candidates.ϢereSpielChequers 06:15, 26 August 2015 (UTC)
WereSpielChequers, I just don't see it. I recall people asking for years and thousands of edits even back in the mid-2000s. And I thought of another "failure" looking at the RfA currently going. GregJackP is shitting all over the RfA in violation of WP:POINT and no one is lifting a finger other than moving comments to the talk page. Why does the candidate deserve to be caught in the middle of this? Not only are we failing to promote enough admins, but we're failing to protect the editors who do brave it. --Spike Wilbury (talk) 16:34, 26 August 2015 (UTC)
Spike Wilbury, I'll tell you what, when you convince all of your buddies to stop shitting all over people who oppose a nomination, perhaps you'll have room to criticize me. It is apparently OK for admins to call me a "troll" in violation of WP:NPA, to being treated uncivilly by other admins, etc. I don't notice you saying a single word to anyone that goes after those who oppose the nomination. Since telling an admin to STFU is not polite, I won't do that, but will merely thank you for your input. GregJackP Boomer! 16:46, 26 August 2015 (UTC)
Spike Wilbury, I agree, candidates don't deserve that sort of treatment, and I'm contemplating a proposal that if adopted would reduce some of that. But we are running low of admins, and this isn't the first RFA this year to get into serious troubles because non admins were opposing over deleted edits that they couldn't see and that admins wee looking at and saying were fine. But the WP:Editcountitis and tenure requirements have been inflated, when I first started nominating people it was possible to pass with 12 months tenure and 3,000 manual edits, now the edit count bit has definitely increased. ϢereSpielChequers 18:00, 26 August 2015 (UTC)
  • Not enough active admins, not enough competent admins, and not enough women admins. Also, the RfA process is, in itself, way too humiliating and stressful for its participants and too easily manipulated by groups with political agendas. This causes potential admin candidates to behave in a certain way that they feel will make it easier to pass RfA, but which is not necessarily the best way for helping keep WP viable in the long run. Cla68 (talk) 07:31, 26 August 2015 (UTC)
  • This is only marginally related, but has any discussion ever been had about a path for rebooting inactive admins? I haven't officially given up my bit, but the consequences of making even a trivial mistake can act as a strong disincentive to becoming active again. Manning (talk) 09:36, 26 August 2015 (UTC)
    I would suggest "watch and participate" in the relevant venues in non-admin role. If one feels comfortable with one's qualities after that, becoming active(r) again seems good to me. Jo-Jo Eumerus (talk, contributions) 09:54, 26 August 2015 (UTC)
  • We do have Wikipedia:New admin, OK it is designed for new admins, but I'm sure it would serve as a refresher as well. With RFA broken we are going to be more and more dependent on luring former admins back, and I'm sure we've discussed it several times, but not yet designed an admin refresher training course. A couple of times recently I've gone through some very old edits of an admin from the early era - one was from 03/04. I couldn't see much that would have been done differently today - vandalism is still vandalism. There are a number of additional admin functions, deleting per wp:BLPPROD, and allocating the various unbundled user rights such as Rollback, template editor and FileMover, but they all have clear criteria. My suggestion would be to pick an area that seems familiar, read the current rules and start out slow. Much as an admin would moving into a different area of adminship. Some areas like Speedy deletion are good in that you can start out just tagging stuff for other admins to delete and if they delete per your tags get back into deleting stuff yourself. ϢereSpielChequers 15:30, 26 August 2015 (UTC)
  • RfA is failing because it has become so difficult to pass, and therefore is not producing enough new admins. It seems to me that only the most experienced of the experienced can pass it now, and sometimes not even they make it through. For those who contend that RfA is not really broken, the nastiness of the current one simply reinforces my own opinion that the opposite is true. --Biblioworm 14:24, 26 August 2015 (UTC)
Which really goes to show, Biblioworm, that you are simply surmising and can't provide evidence that RfA has become more difficult to pass, and that you have not reviewed the RfAs since 2007 and don't have any idea what a nasty RfA really is. Opinions are fine, but non-admins, or users who have never gone through the process would help more on this talk page if they could be objective. Kudpung กุดผึ้ง (talk) 18:22, 26 August 2015 (UTC)
Evidence? The plunging decline in promotions quite plainly shows that RfA has indeed become more difficult to pass. (Coincidentally, although this may take somewhat longer, I will also be gathering statistics on the average edit count and experience of those who have passed RfA since 2014. In light of what you just mentioned, I may do the same for previous years and then compare the differences. That should settle the matter.) Also, I have indeed read some RfAs since 2007, although I don't have the time to read through all the hundreds of RfAs which have happened since then. And tell me how some of the comments made by certain users on the current RfA aren't becoming nasty. I don't think I must specify which ones they are; it should be quite plain. In regard to your last sentence, I believe I know through observation and primary sources (i.e., the comments of candidates themselves) that RfA is indeed quite difficult to pass and can indeed be very stressful. If you want proof, I'll be more than happy to post on the talk pages of random users who have gone through the process recently and ask how the experience was. --Biblioworm 18:33, 26 August 2015 (UTC)
That's not the case: declining numbers of promotions reflect declining numbers of candidates, not declining pass rates (which have been virtually constant since 2008). Whether the environment puts off viable candidates, or the community is producing fewer viable candidates, or the standards have evolved in a way that previously-viable candidates now no longer run, is not discoverable from those numbers alone.
There was some discussion here about gathering statistics for past candidates and I highlighted there the need to calibrate to a background distribution reflecting the core community at the time of a given RfA. This does not seem very amenable to gathering stats by hand or semi-automatically by looking at talk page stats posts. Someone who likes poking databases could do it, I'm sure. Opabinia regalis (talk) 18:49, 26 August 2015 (UTC)
Perhaps there is a decline in the number of candidacies because of the higher standards? It could very well be a chain reaction: Higher standards -> Potential candidates get discouraged -> Less candidacies -> Higher standards + Less candidates = Very few promotions. --Biblioworm 18:58, 26 August 2015 (UTC)
I wouldn't say the standards were higher overall, more arbitrary and more stat based yes, but not actually higher. The tenure and above all editcount requirements have gone up, but at the same time your edits are much less scrutinised than in the past. I've been watching RFAs, nominating candidates talking to possible nominees and keeping an eye on the mood of the RFA crowd for long enough to know that the editcount requirement is losing us some good potential admins. But of the people who I have tried to persuade to run, the drama and angst of RFA is the biggest deterrent. We have over 3,000 editors on the English language Wikipedia who edit over 100 times per month, hundreds of them could pass RFA if they were willing to run, but if we get ten more by Christmas I will be surprised. ϢereSpielChequers 19:28, 26 August 2015 (UTC)
Biblioworm, that's what I mean - the data we currently have is consistent with a "changing standards" explanation, but not sufficient to draw that conclusion.
Having sampled both the 2006 and the 2015 versions and ignored developments in between, the biggest deterrent for me is the fact that it's a big time suck now. When I first started editing again this winter and discovered the admin inactivity policy I took a look at RfA and thought "fuck that, too much work". Obviously I changed my mind eventually, but even for someone who doesn't care much about "angst and drama" and never says no to a perfectly good soapbox, the current setup is a much bigger investment of time and effort than it used to be. Considering that everyone seems to unironically believe that successive RfAs should be no fewer than six months apart, and opposes frequently add throwaway "would reconsider in six to twelve months" comments, the cost of a failed candidacy is much higher. Opabinia regalis (talk) 19:48, 26 August 2015 (UTC)

More bureaucrats[edit]

So far 2015 has gone by and we haven't had any new bureaucrats. Every other year for the past 10 or so we've had a few new crats. May I take this opportunity to encourage any admins with policy and discussion-closing experience to throw their hats into the ring. Andrevan@ 03:39, 25 August 2015 (UTC)

Do you have to be an admin to run? I've always wondered about that. Townlake (talk) 04:03, 25 August 2015 (UTC)
Realistically, I think the short answer is "yes", since bureaucrats have the ability to add and remove admins. AtHomeIn神戸 (talk) 04:11, 25 August 2015 (UTC)
I'm not sure it really is required. I suspect a non-Admin who showed that they were really good at discussion closures (e.g. RfC, AfD, RM, etc.) and parsing consensus could get promoted directly to Bureaucrat without being an Admin first. But, practically, I'm not sure there are all that many "awesome" closers who aren't already Admins... --IJBall (contribstalk) 04:30, 25 August 2015 (UTC)
It's not required per se but I think a non-admin bureaucrat would be significantly hobbled by the inability to view deleted revisions. Some years ago I tried to get view-deleted added to the bureaucrat package so that I could resign my adminship but continue 'cratting but was unsuccessful. Since then, however, I've found that bureaucrat tasks occasionally lean on other administrative permissions like un/protection and un/blocking as well. –xenotalk 09:19, 25 August 2015 (UTC)

I second Andrevan's call; additional bureaucrats - especially ones willing to work at improving RFA from the ground up - would be great. –xenotalk 09:19, 25 August 2015 (UTC)

Technically, admin status is not required to apply for bureaucrat status. Indeed there was such an application a few years ago. While the applicant did garner some support, many editors opposed because they were unable to adequately assess the applicant's decision-making skills without AfD closures and other admin activities. Axl ¤ [Talk] 10:33, 25 August 2015 (UTC)
It seems like a good place to recruit for 'Crats would be among the semi-active (i.e. relatively inactive) Admins corps. I see a lot of these around these days, and I would think the low-impact workload of the Bureaucrat position might be appealing to some of them. --IJBall (contribstalk) 15:50, 25 August 2015 (UTC)

More Crats means more opinions, which is always useful.

I'd especially welcome some candidates who've been here a bit less time than most of us hoary-headed Crats. Each wikigeneration brings its own perspectives. I love to hark back to the good old days when everything was just wikilovely, which never existed, but also sort of did.

Anyway, short version answer to this thread: Yes please. --Dweller (talk) 11:03, 25 August 2015 (UTC)

Agreed. We are all a bit long in the tooth! And as has been remarked in the past, we are also lacking in gender diversity. Suitably qualified candidates who self-identify as female would be particularly welcome. WJBscribe (talk) 11:10, 25 August 2015 (UTC)
  • The job just isn't interesting enough and as long as those who are long in the tooth resist any attempts to make it more attractive nothing will happen. Kudpung กุดผึ้ง (talk) 12:25, 25 August 2015 (UTC)
    • Do we really need more 'crats? There can't be enough work to go round the 32 that we have, especially now that local 'crats no longer handle renames. I know on some projects they're expected to take an active role in the leadership of the project, but on enwiki they just stick to the bureaucracy. HJ Mitchell | Penny for your thoughts? 13:08, 25 August 2015 (UTC)
      • Aaawww I dunno, I think it's good to have a few more voices when discussing tricky promotions. Cas Liber (talk · contribs) 13:38, 25 August 2015 (UTC)

I mentioned this earlier. It seems like people are having so much trouble at RFA, and I'm not just talking borderline candidates, but ones that I would have thought a few years ago would have been an easy pass. Therefore it doesn't surprise me that people aren't willing to run for RfB when arguably the expectations and scrutiny are much more emphasized. Mkdwtalk 14:11, 25 August 2015 (UTC)

Should we be taking a more proactive role in "shepherding" RfXs, for example, crack down more strongly on civility issues or refactor discussions that go off-tangent? If so, we could probably use more people. -- Avi (talk) 15:56, 25 August 2015 (UTC)
The problem is that without the extra crats to help start these changes, it's almost guaranteed that any RfB will only become an even hotter trainwreck than most RfAs. Kind of a vicious cycle, really. Parabolist (talk) 16:37, 25 August 2015 (UTC)
I would also caution against giving the impression that adding 'crats is part of some other effort. Samsara 00:40, 26 August 2015 (UTC)

There is one thing that editors overlook: You can withdraw from RfA and RfB at any time when you run. If you believe that you are up to the job, just run. If you decide that you don't like RfA/B, then withdraw. Esquivalience t 20:30, 25 August 2015 (UTC)

That's a bit too simplistic. Having candidates who run and then simply withdraw when the perfectionist attitude becomes too much to bear will not solve the main problem: Wikipedia is losing far more admins than it is gaining, and although most don't seem to realize it now this might become a real problem some day. I'm currently planning to organize data on this. --Biblioworm (talk) 21:32, 25 August 2015 (UTC)
Biblioworm, there is increasing evidence that the downward trend in RfAs has bottomed out. There is no evidence whatsoever that there is not, or will not be enough admins to do the work. This is a conjecture offten suggested by non admins. However, I welcome your data and look forward to it - I actually like being proven wrong. Kudpung กุดผึ้ง (talk) 01:31, 26 August 2015 (UTC)
I think anyone who has withdrawn will tell you even doing so was an unpleasant experience. It would seem now that many editors expect RFA to be an unpleasant experience. The larger issue is that it's keeping people away from running in the first place; and not necessarily we need solutions to address issues around editors who have run and figured out mid-process they didn't like it and didn't withdraw. Mkdwtalk 21:42, 25 August 2015 (UTC)
RfA withdrawals being demoralizing is a fair point; but they are not permanently demoralizing. There is heavy and occasionally blunt criticism, but one feature of a good admin is their good handling and well-taking of criticism. And of course, there's the WikiBreak, useful for stress. Esquivalience t 03:23, 26 August 2015 (UTC)
  • It seems odd to me that after years of suggesting that bureaucrats take on a more 'pro-active' role in RfA and being met with resistance on the lines that such work is not in their job description, that there is a sudden flurry from them to take an interest. All our Bureaucrats were around (as admins if not 'crats) in 2011 while WP:RFA2011 was going on yet with the exception of Worm who was one of the most active contributors to that project, they chose to do nothing and say nothing although Clerking RfA was the nearest we got to making a firm proposal before we all finally gave up in the face of increasing trolling.
Odder still is a fact that no one seems to have identified (or doesn't want to): among the 'crats, especially those promoted in 2011 and since, their participation on Wikipedia dropped dramatically after being endowed with the 'crat bit - or even retired completely shortly after being elected. Again, the exception is Worm whose Wikipedia work for two years was of a kind that is not reflected in edit counts. Anecdotally it looks as if 'crats think 'I've made it to the top, now I can sit back and watch the others work'.
Someone once said (I believe it's in the archives of this talk page) that one needs to be a pretty boring editor to become bureaucrat. Well, that kind of figures. Any active 'front line' admins would have accumulated too many enemies from just doing their job. RfB would degenerate into a re-confirmation RfA - pitchforks an' all. Kudpung กุดผึ้ง (talk) 02:29, 26 August 2015 (UTC)
Let's not derail this thread. I'm not proposing RFA reform or that crats take a more active role. I think the RFA/RFB system basically works. I just think I would like to hear more voices in crat chats and empower more folks to find a consensus without them. Andrevan@ 04:37, 26 August 2015 (UTC)
Well, RfA/RfB "works" if your goal is to promote about 2 Admins a month, and about 0 Bureaucrats a year! --IJBall (contribstalk) 04:46, 26 August 2015 (UTC)
One year and nearly seven months and counting since. Mkdwtalk 05:13, 26 August 2015 (UTC)
  • Let us not derail the efforts of those who are genuinely concerned and who are racking their brains for a solution to something the Bureaucrats themselves can't even find. Kudpung กุดผึ้ง (talk) 03:53, 27 August 2015 (UTC)

For what it is worth, I've already suggested to someone that she undergo RfB, and she demurred for the time being. I'll think over a couple of other people whom I think would provide good analysis and reasoning as regards complicated RfXs, which is our primary (and now probably sole) function. Don't forget, most any of you discussing this issue are more than welcome to volunteer for RfB Face-grin.svg. Who knows, maybe now that RfA is more difficult than it was in 2007, once someone vaults that hurdle, RfB will be less traumatic than it was in 2007. -- Avi (talk) 15:10, 27 August 2015 (UTC)


Querying statements or !votes is an accepted part of the RfA process. We value open discussion, and asking an editor to clarify their statement has value for everyone; however, this can at times be distracting to the point of disruptive, and occasionally lengthy interjections are moved to the talkpage. Increasingly, querying oppose votes is being seen as badgering, and this on occasion has impacted on the vote process and the outcome of the RfA, even when the nominator had no control over the situation. There is no guideline on when an interjection is inappropriate, nor when the resulting discussion goes so far off-topic or becomes so distracting it becomes appropriate to move it; moving interjections is a matter of individual judgement so becomes fairly random - though generally by the time an interjection is moved the damage has already been done. I don't think there is an ideal answer to this, but wondered if it might result in less disruptive and more peaceful RfAs if querying statements or !votes were done on user's talkpages, and not on the RfA page itself. And if people wanted to have open, detailed discussions about points raised in the RfA, this could be done in the discussion section, rather than scattered across individual user's !vote statements. To ensure this happens, we would set up a rule that interjections are not allowed; that queries should be made direct to the user in question on their talkpage, or - if it's a general point - that such queries are made in the discussion section only. SilkTork ✔Tea time 13:26, 26 August 2015 (UTC)

I think that tearing queries and especially clarifications or objections from the context of the !vote they are directed at is a problem. I agree with removing overly long discussions to the talk page (or, I would prefer but is not practice, to the discussion section of the RfA) but not as a general thing. Jo-Jo Eumerus (talk, contributions) 13:33, 26 August 2015 (UTC)
(ec) The problem is that completely disallowing answering of votes means that wrong statements about the candidate can be given undue prominence, without an accompanying explanation that they are wrong. Say, a silly oppose vote "oppose, 500 deleted edits means the candidate does not understand WP:N". Shouldn't that be answered by admins explaining what the deleted edits are ("all of them are correct deletion nominations")? Queries should only be disallowed if comments are disallowed (in a more pure system we would separate the discussion and the voting). —Kusma (t·c) 13:36, 26 August 2015 (UTC)
I'm with Kusma on this. If someone says something incorrect in an RFA !Vote then it is good to challenge that. I see less value in challenging per nom style !votes, especially as they are often the sorts of !votes that newish RFA !voters will cast. But the big area of contention is over the criteria for adminship, if we don't try to set those outside of the RFA then RFA will sometimes become a bunfight, especially if we are going through a phase when several editors are trying to exclude certain types of editor from becoming admins. Setting the criteria, or more realistically part of the criteria outside individual RFAs would have several advantages, RFA would become less contentious, and it would no longer be possible for 30% of the community to Blackball certain types of candidate. But it would mean empowering the crats to ignore out of scope !votes much as admins do at AFD. ϢereSpielChequers 15:15, 26 August 2015 (UTC)
I think only criteria on what community could realistically agree, would be some very minimum requirements below what RFA would be automatically closed as WP:NOTNOW.--Staberinde (talk) 15:26, 26 August 2015 (UTC)
I'm not sure. Obviously there won't be unanimity on any contentious criteria; But I suspect we would have a strong consensus that admins don't need an FA or GA, though I'd like to think that we could make it a requirement that admins needed to have added reliably sourced information to the pedia. ϢereSpielChequers 15:45, 26 August 2015 (UTC)
Now this makes sense. We have seen so many RfAs dominated by debates on what should be the general criteria, rather than the merits of the particular candidate. If we could converge on a common approach to a few issues about fitness for adminship, then the debate on each candidate could take place within that framework. Not detailed requirements like "must have written 350 featured articles", but broad issues such as:
  • Should the candidate have been significantly involved in content creation?
  • Should this creation include major contributions to good or featured articles?
  • Should this creation have involved the candidate in debates with other editors?
  • Should the candidate have been active on a range of non-content activities? - which are essential, which are desirable?
  • If they state that they just want adminship for one or a few special purposes, should this be accepted?
  • If they have either stated or implied by their actions that they are keen to become admins, is this a positive or a negative?
  • Are candidates assumed fit to become admins unless there is evidence to the contrary, or assumed not fit unless they have satisfed the !voters that they are?
If we need a heated and lengthy debate over such issues, it will be best to argue in general terms, rather than fighting the battles anew over each application: Noyster (talk), 17:14, 26 August 2015 (UTC)
I agree that it is better for everyone if debates can be had in general terms once, rather than on every candidate page dozens of times, although I doubt that's ever going to happen. One thing I think your bullet points miss is based on the way are they phrased: Should is not the right word. My stance, and I imagine the stance of at least a reasonable minority, is that GA/FA work is not required, but if a candidate has worked on getting articles to GA/FA status, then it should be counted as a point in their favor. So it's not that "the candidate must meet requirement x, y and z", it's that "the candidate could have done some of the following: [long list of various activities], or anything else which indicates they would be an overall asset with the mop". But I certainly agree that "can we promote people who only want the tools for a few specific purposes?" in particular is something that would be better discussed at length on WT:RFA, rather than on a case-by-case basis. Bilorv(talk)(c)(e) 17:34, 26 August 2015 (UTC)
  • The easiest thing to do would be to look at the list of personal voting criteria published by the regular voters at RfA and try to draw an aggregate of it all. Anyone who does not know where that list is is not really showing sufficient engagement to be commenting here, but for those who really don't know, one place to find it is athe bottim of WP:RFAADVICE. Of course, I realise that nobody likes plodding manually through stuff, but if we want change, it's sometimes the only way to be objective. Kudpung กุดผึ้ง (talk) 22:10, 26 August 2015 (UTC)

I've said that RfA is a discussion, and accordingly participants should be able to challenge positions of others, but I can see the other side of the coin; where someone feels they are 'under attack' for providing their honest opinion. Part of this feeling probably stems from the fact that to challenge someone's position, you need to put it right below their 'vote' as if you are challenging their very vote, rather than some or all of the position they advanced in the same. What about bare numbered lists of signatures with people providing their rationale(s) in a discussion section? –xenotalk 22:20, 26 August 2015 (UTC)

I don't completely hate the way we're (sort of) doing it at Wbm1058's nomination, with votes and rationales in their rightful sections, and discussions of each vote relegated to subsections of another super-thread somewhere else. I vote *'''Oppose''', user is ugly. ~~~~, and the resulting drama can erupt on the talk page under == Discussions of votes == === Opposes === ==== Juliancolton ====. People still have every opportunity to question my input (possibly with greater ease than trying to pick me out of the "rationales" area), but whatever case I've made isn't drowning in a sea of rebuttals, and I don't feel the need to viciously defend my one slice of the "oppose" section. I will say that it's nice to be able to immediately identify a user's justification for their vote, even if it's later proven to rest on shaky grounds. Honestly, though, with how few RfAs there are anymore (let alone contentious RfAs), it might just be best to take it on a by-case basis, as long as we establish which formats are acceptable. The current systems works fine for something like Wikipedia:Requests for adminship/Ritchie333, where discussions of individual votes were sparse and brief, but Wikipedia:Requests for adminship/Liz could have definitely used a more intuitive structure. – Juliancolton | Talk 22:38, 26 August 2015 (UTC)

2 days discussion, 5 days voting[edit]

I would like to propose a simple change to our current RfA format. I suggest that for the first 2 days we only have discussion. Then after the 2 days have passed the "voting" or as we call them "!voting" sections are put up for the remaining 5 days. Discussion would of course be allowed to continue.

This would accomplish a few things:

  • Instead of debate happening in between votes it will be more likely to happen out of the way.
  • Instead of people voting in the absence of information there will be an existing discussion that can be referred to.
  • It will re-enforce the ideal that RfA is supposed to be a discussion and not a vote.
  • It will give everyone time to get their concerns out and for those concerns to be audited by the community before people start making up their minds.

In my opinion 5 days is plenty long for voting, particularly if we spent 2 days before that doing our homework. I am looking forward to hearing suggestions, comments, criticisms, and any better ideas. Chillum 15:27, 26 August 2015 (UTC)

I wouldn't be adverse to this. As it is, I've come to think that 7 days of voting is too long. So 2 days of "discussion", followed by 5 days of voting, would be an improvement, to my mind. --IJBall (contribstalk) 15:50, 26 August 2015 (UTC)
I suggested a two-phase approach a couple of weeks ago. A concern was raised regarding the length of the phases, but regardless of the time devoted to each, I agree that giving interested parties more background information should improve the quality of the discussion. isaacl (talk) 16:34, 26 August 2015 (UTC)
  • I think this is worth considering. (At the moment, pretty much any idea to improve the process and reduce drama is worth considering.) But what sort of discussion is likely to arise other than "I will oppose this user because they don't meet my criteria", and "I will support because I trust this user"? I'm not sure what will actually change, other than in the first two days the comments will be in one place, and the next five days the comments will be spread across the RfA page. The process will be the same, and during the 5 day period there will still be interjections. SilkTork ✔Tea time 17:10, 26 August 2015 (UTC)
  • How about when a vote is to be discussed a link to a sub-section in the discussion area(=== Discussion of Chillum's !vote ===) is put below their vote? That way it is evident from a scan of the votes that discussion of the vote has taken place, however it can happen in a sub-section linked below. Chillum 17:16, 26 August 2015 (UTC)
  • I would be willing to support a 3 month trial period to see if it works. A community RFC at the end of the trial period to determine proceeding or reverting back. Mkdwtalk 17:19, 26 August 2015 (UTC)
  • I was thinking we could do trials with volunteer candidates. Chillum 17:45, 26 August 2015 (UTC)
  • This has been tried before on occasion, though I forget who or when. The advantage of the seven-day period is that people with a variety of different schedules can participate. This is one of a number of proposed "reforms" recently that are obviously devised by people who are on wikipedia on a daily basis, and probably are available regularly throughout the day. If anything, we should move toward models of community decision-making that are more inclusive of casual participation, rather than introducing "reforms" that privilege the active and available. Opabinia regalis (talk) 17:41, 26 August 2015 (UTC)
  • I did say the discussion did not stop, so there is still 7 days of discussion. This in no way effects someone who shows up into an RfA 2 or more days in. Right now you show up on day 3 and you can discuss and vote, same thing under my proposed idea. You would only miss the delay in voting. Chillum 17:45, 26 August 2015 (UTC)
  • I think the duration of each phase can be discussed. As I mentioned above, if the intent is to accommodate those who edit once a week, then each phase can be a week in length, and it may be helpful to start all RfAs on say Friday, so everyone can plan exactly when they will be able to spend time on the current week's requests. isaacl (talk) 18:31, 26 August 2015 (UTC)
  • I'm not going to express an opinion on this just yet, but I have a question: would RfAs be able to be closed as NOTNOW and/or SNOW before the end of the 2 day discussion period? I suppose it might be fairly obvious in most cases when to close NOTNOW, assuming people are allowed to before the voting. The idea of SNOW closed RfAs, at least in my eyes, is "you've been getting loads of opposes and few supports, so there's no need to let another hundred people pile on". If no-one has opposed yet, who decides when to SNOW close and what if somebody objects? Bilorv(talk)(c)(e) 17:45, 26 August 2015 (UTC)
  • Good question. I think the discussion would quickly reveal if a candidate is not yet ready. It would spare them a pile of bolded Opposes comments and they would instead get a discussion that will hopefully be full of good advice. I can tell you from experience that a long vertical stripe of Oppose prior to a "Not now" is a bit unpleasant, even if they are followed by kind words. Chillum 17:48, 26 August 2015 (UTC)
  • Ouch. To be quite honest, I'd rather get an Oppose than a "Sympathy Support" (at least "Moral support" doesn't sound as condescending). But back to the topic at hand, I think that's a good answer and can see it working quite well. Bilorv(talk)(c)(e) 18:10, 26 August 2015 (UTC)

@SilkTork, the way I envision it is that interested parties would list the results of their analyses of the pros and cons of the candidate. To make it easy for others to review, there could be one consolidated list of pros, one consolidated list of cons, and then an analysis section where each commenter can list an analysis. Discussion can ensue under each item so they can be challenged, broken down, affirmed, and so forth. By having a consolidated list, any newcomers to the page can easily come up to speed on all of the discussion related to one point, rather than having to read through dozens of threads from each commenter and mentally combining the information. At the end of the first phase, a new section would be opened where each person can list their support or opposition to the candidate, and they can describe it in terms of the relative weight they place on the various pros and cons. With this, closers of the discussion can actually weigh the consensus view of the interested parties on the qualities of the candidate, rather than looking for reasons to discount votes, which is not truly consensus at all. isaacl (talk) 18:16, 26 August 2015 (UTC)

  • As an FYI, this was tried with Ironholds 2. It didn't go over very well. You can come to other conclusions yourself. --Izno (talk) 19:50, 26 August 2015 (UTC)
    • The format of that request for adminship, though, is different that what I have proposed. isaacl (talk) 20:07, 26 August 2015 (UTC)
  • To be clear I don't think 4 days of question asking is any substitute for 2 days of discussion. For this reason I think my proposal is substantially different than the experiment done at Wikipedia:Requests_for_adminship/Ironholds_2. The problem with just asking a question and getting an answer is that it is a 2 person, 2 comment discussion. A proper discussion should involved several people and have as many comments as needed to explore the position presented. Chillum 22:07, 26 August 2015 (UTC)

This has come up before. And unlike most proposals, this is an easy implementation. As I recall it usually gets "stuck" over whether it should be 2/5; 3/4; 3/5; 1/6; etc. I prefer the 3/4 model personally. It gives everyone a fair amount of time to discuss, and yet 4 days for everyone to "vote". And the latter is another sticking point, those who say the whole thing should be a discussion, not a vote, and so the splitting legitimises "voting"... - jc37 18:40, 27 August 2015 (UTC)

I would not be adverse to 3/4. I think if this is tested it will likely be by a volunteer candidate, this will mean that they can decide on the fiddly details. If it goes well and is ever adopted then the community can decide then based off of real data. Chillum 18:57, 27 August 2015 (UTC)
I suggest calling the first phase a vetting phase, where participants are determining the abilities and characteristics of the candidate, and the second phase an analysis phase, where participants can weigh in on the relative advantages and disadvantages of the candidate. In theory, there is no inherent need for any voting in the second phase, as long as everyone expresses the relative weights they place on different candidate traits. isaacl (talk) 19:10, 27 August 2015 (UTC)
a vetting/evidence phase and an analysis/workshop phase... now what does that remind me of?... - jc37 00:44, 28 August 2015 (UTC)
With arbitration cases, the start of the phases tend to be flexible (people can start posting analyses right away) and the end fixed, whereas in this proposal, the start times would be fixed and the end flexible. People can continue to submit information about a candidate's abilities throughout the process; having a specific start time for the second phase is to allow for some minimal amount of data to be available to improve the quality of the expressed opinions. Submissions to an arbitration case tend to be very diverse, due to the lack of an specified scope for most of them (an improvement I suggested once was for arbitrators to delineate the case's scope when it is opened), whereas with requests for adminship, there is a definitive goal for all contributions: evaluating the candidate's suitability to perform administrative tasks. The proposed analysis phase would provide insight into the relative weight the community places on different characteristics of the candidate, so that the discussion can be closed based on the community's consensus view. The workshop phase of an arbitration case is not a process for determining consensus; it is more of a brainstorming session of ideas to feed into the final decision. So no, I don't see much similarity between this proposal and arbitration cases. isaacl (talk) 03:52, 28 August 2015 (UTC)
  • As proposed, this would mean that someone who edits only on a particular day of the week might be unable to 'vote" if the initial discussion phase fell on that day. The "voting" phase should not be less than a full week, if this is to be implemented. In any case, i suspect that most comments in the 'discussion" phase will effectively be supports or opposes, so this won't actually have much efect, but I could be wrong about that. DES (talk) 12:39, 28 August 2015 (UTC)
    • As I mentioned, if the desire is to include those who edit once a week, then both phases should be at least a week long. It's up to those who are seeking a change to set the tone: if you want the initial phase to focus on discussing the pros and cons of a candidate without expressing an opinion on the relative balance amongst them, then set an example for everyone else to follow in your comments. (Some gentle reminders from a moderator may be helpful, too.) isaacl (talk) 15:24, 28 August 2015 (UTC)

Require endorsement[edit]

  • (ec) I've been thinking it might be a good idea to require an administrator's endorsement before throwing your hat into the ring. This would eliminate most (all?) of the NOTNOW-type nominations, which often serve only to demoralize the candidate. An "endorsement" would by no means equal a "support", but instead indicate that the endorsing admin believes the candidacy has the potential to be constructive. – Juliancolton | Talk 17:50, 26 August 2015 (UTC)
  • Can't say I'm thrilled with that suggestion, as it would force all candidates to have to begging and pestering their Favorite Admin!(tm) just in order to run in an RfA (and failed "endorsements" would still surely be held against the Admin in question...), and worse give even more power to the current Admin class (as Admins would now have 100% "control" over who could even join their ranks). So I'd oppose that idea... --IJBall (contribstalk) 17:59, 26 August 2015 (UTC)
  • (edit conflict) It could probably be any auto-confirmed editor. It's possible an individual might resort to sock puppetry to accomplish meeting this criteria but I feel like the majority of NOTNOW candidates simply haven't bothered to read the requirements -- and don't have the intention to willfully disrupt the system. Mkdwtalk 17:59, 26 August 2015 (UTC)
  • Now that's a more interesting suggestion, that's worth pondering... --IJBall (contribstalk) 18:03, 26 August 2015 (UTC)
  • Again, though, an "endorsement" doesn't mean you think the candidate will pass, only that they're at a position where the comments won't just be self-repeating Opposes per NOTNOW. I'm not married to the idea of it being restricted to admins, but when was the last time somebody passed RfA without the support of at least one admin? I guess maybe it could be a bureaucrat thing, as they can probably be trusted to endorse or deny candidates fairly and with impartiality. Just thinking aloud... – Juliancolton | Talk 18:14, 26 August 2015 (UTC)
  • If 'crats wanted to do it, that's great. If only admins can do it, okay. (I don't feel it really gives them any more power because they're a diverse group of over a thousand people, not a collection of zombies who all think alike. They have no combined intention to destroy non-sysops and even if some would refuse to nominate serious candidates, there'd be plenty that would.) I'm not sure it would work with autoconfirmed; you might still get SNOWs with people who happen to have a real life friend with an old autoconfirmed account lying around, or from people who are beyond the level of complete newbies (and therefore able to get an autoconfirmed to nominate them), but still far short of the community's requirements. I suppose it would be a help to cut out most NOTNOWs, but it would also be a burden (albeit a small one) to anyone who wants to self-nominate. Bilorv(talk)(c)(e) 18:30, 26 August 2015 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── I hope you do not mind but I moved this to its own section. It is an interesting idea that deserves discussion, but it is a bit off topic in the above thread. Chillum 18:20, 26 August 2015 (UTC)

In my opinion there may be some merit in requiring a nominator. I do not however think we should require this nominator hold special status, even auto-confirm. While not requiring auto-confirm invites gaming, if someone wants to game their way into a "Not now" situation then they can have it. If a new user noms another new user we can still tell them not now, but there is a very solid chance they will learn that when seeking a nominator. Chillum 18:22, 26 August 2015 (UTC)

My first RFA was a self nom, I admire those who self nom and wouldn't want to rule that out. It would be good to fix a "NotNow" criteria, and set an edit filter on RFA such that only editors who met that could transclude. I suspect we can all agree that editors with <2,000 edits and 6 months tenure don't currently stand a chance at RFA; as a pragmatic suggestion, how about anyone with less than 12 months tenure or less than 6,000 edits can only run if an admin nominates them? That would still have allowed all of the recent successful self noms to run, and if we get an exceptional candidate who is ready with less than that they can still run if an admin nominates them. ϢereSpielChequers 18:41, 26 August 2015 (UTC)

You make a good point about the value of allowing self noms. Though I am not sure that we should put hard limits on the edit count or tenure. While my self nom with a mere 734 edits did not turn out well, I think I would have been even more insulted if an automated process told me I was not ready. Perhaps there is value in a future admin tried to get the mop too early to experience the resulting advice from the community. While the whole thing was unpleasant I was given very solid advice and don't feel anyone acting in a mean spirit. I think it was productive and helped me recognize when the correct time for me to run was. Chillum 18:46, 26 August 2015 (UTC)

Successful self-noms since 2011[edit]

Year RfAs p/f User RfA Score
2015 14/27 Cyclonebiskit 2 121/12/9
2015 14/27 Opabinia regalis 2 118/22/5
2015 14/27 Jakec 79/11/5
2015 14/27 SarekOfVulcan 4 130/30/5
2014 22/40 0 0
2013 34/40 Acather96 74/9/1
2013 34/40 Trevj 84/5/0
2013 34/40 Cerebellum 81/0/3
2013 34/40 Trappist the monk 85/35/8
2013 34/40 Grandiose 84/11/6
2013 34/40 Jason Quinn 2 138/29/16
2012 28/68 Scottywong 89/4/11
2012 28/68 Dpmuk 64/3/1
2011 52/88 Guerillero 59/14/6
2011 52/88 CharlieEchoTango 85/2/1
2011 52/88 GB fan 88/1/0
2011 52/88 Rannpháirtí anaithnid 71/10/4
2011 52/88 Tyrol5 65/0/0
2011 52/88 Worm That Turned 121/3/2
2011 52/88 TParis 2 48/4/4
2011 52/88 Jimp 75/4/0
2011 52/88 Sadads 97/2/3
2011 52/88 SarekOfVulcan 2 166/63/10
2011 52/88 Catfish Jim and the soapdish 109/2/1
2011 52/88 RHM22 78/16/8
2011 52/88 Feezo 47/1/16
2011 52/88 JaGa 83/3/0
2011 52/88 Neelix 69/14/12
2011 52/88 Boing! said Zebedee 160/1/0
2011 52/88 ErrantX 85/2/4
2011 52/88 Rami R 2 66/12/8
2011 52/88 ErikHaugen 82/27/1
2011 52/88 Smartse 129/0/1
2011 52/88 Gimme danger 77/11/6

Kudpung กุดผึ้ง (talk) 23:22, 26 August 2015 (UTC)

  • I would be against giving admin more power than they (we) already have. I'm for making RFA easier to do and easier to pass, not harder, so I think this would not only make it harder, but would give the impression that admins are a "special" group...even more than it seems now. Admins aren't and shouldn't be the gatekeepers for RFA, the community is and so far, seems to be doing a fine job of filtering the wheat from the chaff. We just need more wheat to apply for the job. Self-noms, by themselves, have not been shown to be a problem that needs "fixing", so not sure of the purpose of the proposal. Dennis Brown - 17:37, 27 August 2015 (UTC)
  • The table above shows that out of 150 passes, 33 were self noms. That's 22%. I think that's a significant number. It also probably means that around a quarter of successful candidates are not so scared of the process that they need a friend to hold their hand.
Like Dennis, I believe expecting admins to 'authorise' RfAs would probably not be a viable solution. It would also be to play into the hands of the users who complain that admins have too much power already. Full circle, back to the drawing board: it's not the candidates or the system that are the problem, it's the voters. Fix the voters and RfA will fix itself. Kudpung กุดผึ้ง (talk) 01:25, 28 August 2015 (UTC)
  • I'd disagree. I suspect a portion of the discussion will go into the administrators themselves and how they know the nominee. -- Ricky81682 (talk) 08:58, 28 August 2015 (UTC)

Hatting long comments[edit]

Wikipedia:Requests for adminship/Wbm1058 has been the source of a lot of messing about with the format of RfAs. Context is available at Wikipedia talk:Requests for adminship/Wbm1058#Location of comments. The current situation is this: comments in response to !votes are made using #: beneath the !vote, as usual; responses to responses get an extra colon etc. When someone deems the discussion to be too long for people to comfortably navigate and/or irrelevant to the user whose RfA it is (e.g. when things descend into personal attacks, questioning motives, meta-discussion), they can hat the comments (everything below the original !vote) and use an li value thing to fix the numbering system below. Look at the page to see how this works. Tangent: we might find a problem when the most recent !vote in a section has a hatted discussion below it, and the next user to come along doesn't understand the li value things, but this can't really be a bigger issue than the one we have at present with people forgetting to use # signs instead of just colons, which was complained about here.

Does everyone agree with this hatting method currently implemented? Does everyone disagree? Anyone have any thoughts, comments, or just want to administer a trout to me or someone else? Here's the place to do it. To minimise disruption to the current RfA, please try to discuss formatting changes someone on this page and/or in this section, rather than at the RfA pages. Bilorv(talk)(c)(e) 22:07, 26 August 2015 (UTC)

Bilorv, you did a great job there. I like the hatting idea, but I just wonder whether there's a simpler way than the #<li value="3"> trick. I suspect most editors will find that difficult, and if people indent their votes, don't we need them to go through adjusting all subsequent li values? I can't think of anything right now -- can you? --Stfg (talk) 22:27, 26 August 2015 (UTC)
  • May I suggest that instead of using collapse templates which traditionally are used to end a discussion that we instead place a link as the only content below a !vote to a sub-section below dedicated to discussing a particular editors !vote? Like:
  • This would allow for extensive discussion without constantly making sure we don't break the numbering. It will avoid the need for special markup(#<li value="3">) to avoid the hats breaking numbering. And it will give a short clear indication that a particular vote has more discussion without that discussion making it hard to read the votes. At most a !vote will have only 1 extra line if discussed so the length of the discussion will not distort the size of the sections. Chillum 22:31, 26 August 2015 (UTC)
  • I will also point out that having an active discussion under a hat is no good because you cannot search with the browser "find" function for anything under a hat. This is bloody annoying. Chillum 22:52, 26 August 2015 (UTC)
    I like your idea; I'd suggest that the original !vote is duplicated in the discussion sub-section to save people scrolling back up or using the back arrow. The hatting has caused some issues I didn't think of before I implemented it (but I reckon I wouldn't find it too annoying personally; just view things in the edit window - you'll need to click edit at some point anyway if you're planning to reply). Maybe we should say the conversation should have been dead for x amount of time before it's hatted? (As is standard all around Wikipedia, I think, anyone can still boldly hat stuff where it's clear people are going off-topic, getting worked up, not saying anything constructive etc.) I also wonder which format the closing 'crat would find easier to read through – the standard talk movements only when things get off-topic, the everything-other-than-votes-on-talk idea, the hatting idea, your sub-section idea or something else. Bilorv(talk)(c)(e) 23:13, 26 August 2015 (UTC)
  • Thanks Bilorv for all that work; this was a good idea to sort out this particular RfA. In future cases without so much prior messing around, I agree with Chillum that the "discussion closed" templates should be replaced with something less finger-waggy. Is the idea to continue discussion within the collapsed sections in order to make the page easier to navigate, or to hat threads on the main page in order to shoo people over to the talk page? I find that I dislike both ideas for different reasons - the collapsed sections can't be seen at all in mobile view, so again you have the problem of challenged votes appearing as if no one has commented on them. Moving things to the talk page seems likely to discourage actually reading the follow-up discussions if you have to keep going back and forth. Expecting people to copy votes they're replying to is unlikely to work. Hmmm. I guess this is the sort of pointlessly disagreeable thinking-out-loud that ought to be hatted :) Opabinia regalis (talk) 23:17, 26 August 2015 (UTC)
  • That hatting method is good. In fact, I would suggest to use that for all cases where long conversation disrupts the RfA flow; this way pertinent info still stays in the listing. Of course, when searching for stuff you need to un-collapse the discussions. Jo-Jo Eumerus (talk, contributions) 23:33, 26 August 2015 (UTC)
  • Instead of using Template:Hat, it might perhaps be better to use Template:Cot. The difference is that the former says that there should be no further discussion, whereas the latter allows continuing discussion within the collapsed area. It seems to me that the goal here should be to keep the RfA page navigable, rather than to tell editors who want to discuss something that they must stop discussing it. --Tryptofish (talk) 00:43, 27 August 2015 (UTC)
    • Nicely done Bilorv. I was thinking of {{collapse top}} as Tryptofish mentioned (having just used it on an Arbcom evidence case :). -- Avi (talk) 01:02, 27 August 2015 (UTC)
    • Yep, that's a good idea. I've never seen {{cot}} and {{cob}} before (although I probably should have noticed it listed here), but I was actually considering making my own version of {{hat}} without the "archived" text (and with a less boring color). Bilorv(talk)(c)(e) 08:07, 27 August 2015 (UTC)
  • Good change per this. Samsara 02:08, 28 August 2015 (UTC)
  • I do not think this is a very good change. RfA doesn't know whether it is a vote or a discussion, and collapsing some parts of discussion just applies a band-aid to the problem instead of solving it. Given how broken WT:RFA is (hint: more broken than WP:RFA), more than a band-aid is probably asking too much though. —Kusma (t·c) 09:08, 28 August 2015 (UTC)
  • Bad idea: Have to agree with Kusma. I'd voted on this RfA and came back to it. On seeing the collapsed discussions, it felt to me like discussion was intentionally shut down. I strongly dislike this. If we're to do this, we might as well turn RfAs into a straight up/down vote, and do away with bureaucrats entirely. --Hammersoft (talk) 12:54, 28 August 2015 (UTC)
  • Good faith concerns have been raised about the collapsing, so I've modified the RfA to instead include a '#Replies' section, where the hatted content has been placed with markers from the comment in question. Please double-check my work. With thanks to Wbm1058 for understanding. –xenotalk 15:46, 28 August 2015 (UTC)
  • @Xeno: Cyberpower678 edited some of the content to replace some hash signs with colons here—that was probably my fault for leaving the hash signs when hatting the comments in the first place, but (I think) worth noting anyway. And Cyberpower678 messed up some signatures and other things in that edit which I cleaned up here. Bilorv(talk)(c)(e) 19:18, 28 August 2015 (UTC)
    Whoops. Sorry.—cyberpowerChat:Online 19:59, 28 August 2015 (UTC)
  • Thanks xeno, that was my original proposal. Here's a template that allows for a bit of shorthand, if desired: {{RfA-replies}}. Now, do we need to place back anchors as well, or do all mobile browsers provide a functional "back" button that works with anchors? Samsara 02:59, 30 August 2015 (UTC)

Please come back after addressing concerns (or not)[edit]

I was about to post another rant to a currently open RfA that is already on the verge of collapsing under various fundamental discussions that are outside the candidate's power and not connected to the question whether that candidate should be trusted with the tools. So instead I will put some general thoughts here.

I do not like "RfA criteria", and often vote support in order to cancel out a third of such a vote. What I like even less is asking good candidates to change the way they work in order to gain another possible support vote in a future RfA. If a candidate is good in some field, has some clue and treats other people with respect, then it is not a problem that they do not have experience with images, do not have a certain namespace balance, or have not been involved in new page patrol or the Articles for Creation process. What matters is whether they are here to help work and improve Wikipedia, and manage to do so while getting along with other people, helping to solve problems and conflicts rather than creating them.

Now some RfA voters are asking people who are good in some field (may it be copyrights, images, adding inline citations, copyediting or writing brilliant prose) and volunteer their hard work there to additionally take on some other task (say, do new page patrol, work with categories or take photographs or write a Featured Article) before they will consider them as a possible administrator. That is not a good idea. Adminship is not a reward, and we should not encourage people to take on various tasks they do not enjoy just so they can be an administrator some day. Volunteers should do what they enjoy and what they do best, and all those who volunteer for extra duty and that can be trusted (keeping in mind that we aways assume good faith) should be allowed to do that extra duty (adminship). They should be allowed to do so without having to do an internship in the Wikipedia department of some random opposer's choice first. I am more happy to support a failed candidate in a future RfA if they continue to do what they do best in order to improve Wikipedia instead of attempting to look like other people's ideal candidate. So to all people who failed RfA earlier: please continue to work here and do not try to address arbitrary concerns. —Kusma (t·c) 10:10, 28 August 2015 (UTC)

Those same people will oppose the second time because the editor "only" did what they suggested they do. You can always vote oppose and if things go wrong, hey not your problem. The default will always, always be the safer option. I'm coming on ten years based on these fools but the key then was "he's lacking experience here" was responded with "...I assume he'd review the relevant policies if he intended to take admin actions outside of his experience" and not an oppose vote. I'm still constantly overturned but we have mountains of processes to fix things (DRV for a wrongly-decided AFD, draftspace for new ideas, bots, templates). The problem is the other extreme in that ARBCOM has given way more discretionary to admins in certain topic areas which is why it's a total guessing game. The truth is you can't know how someone will act with the tools because often they won't know until the situation arises and we don't know what situations are arising. Reminds me, have we even had a recent admin who came here being an admin somewhere else? -- Ricky81682 (talk) 10:57, 28 August 2015 (UTC)
Totally agree. I really wonder if this is one of the main factors scaring people away from RfA. I think a lot of long-time editors think to themselves, "Why bother with an RfA, when I don't care about AfD, and I know that my RfA will fail because so many voters basically demand vast AfD experience?" Or, "Why bother with RfA because I don't have a GA or FA under my belt, really have no idea how to go about that, and really don't have the time to tackle doing one?" Sub in CSD or AIV experience, etc. for either of these examples and the result is the same. There is basically an effective minority in all RfA's that will sink any candidacy unless the candidate has pretty literally "dabbled" in every possible permutation of Admin action. And this is true even for candidates like Wbm1058 who have specifically expressed a desire to only work on one facet of Adminship. --IJBall (contribstalk) 16:39, 28 August 2015 (UTC)
  • I agree; we're not aiming for an archetype with RfA candidates. For example, would it be sane for a company to require their janitors to have bookkeeping experience? Or how about requiring volunteers in the community to learn and use vector calculus? Frankly, the areas that are often demanded of candidates are boring to most: AfD is boring and repetitive to most people (delete, fails WP:N, sign, repeat), content creation (as opposed to discussion) is just Jenga with words, and new pages patrol is really only interesting to people who like to mash buttons. We have many admin backlogs because people who have the skill don't fit into the archetype of the "perfect RfA candidate". The archetype is just getting ridiculous. Esquivalience t 18:47, 28 August 2015 (UTC)
    • One issue I can see with candidates that are good at one particular aspect of the wiki administratorship but not at others may make issues in the "other" parts. A statement "I will only work in part X of Wikipedia" in the Q/A1 is not a magic spell that binds the person for the future. On the other hand, I am not familiar with such a problem actually occurring - most narrow-scope adminship requests I know of worked out well. And AfD is a sort of prominent administrator workspace. Jo-Jo Eumerus (talk, contributions) 19:03, 28 August 2015 (UTC)
      • More to the point, if a successful Admin candidate "reneges" on a pledge to "go slow" in areas that they aren't familiar with, there are proper forums to deal with that: WP:AN (or WP:ANI), or in really bad cases WP:ArbCom. It's not like there's no recourse for situations like that... But insisting that every Admin candidate know ~100% of the toolset and job before going in is a recipe for passing almost no one out of RfA. --IJBall (contribstalk) 19:17, 28 August 2015 (UTC)
  • The problem with the issues discussed in this thread is that it's perennial talk. I've seen the same comments dozens of times over the last 6 years. People like to talk, but what's stopping them actually doing anything about the problems? Someone could, for example, take the initiative to contact those voters who apply ridiculously high criteria and politely suggest they refrain from disrupting RfA with such such nonsense. Better still, those who post long and loud in this talk page ought to throw their own hats in the ring and find out for themselves what RfA is really like. I did, and frankly I have little sympathy for those who constantly whine but who are scared stiff of the process. Kudpung กุดผึ้ง (talk) 19:49, 28 August 2015 (UTC)
    I am doing something about the problem, I go to RfAs and vote "support" if bad arguments are being made in the oppose section. Also, this thread is in direct response to your vote. Should I have put that on your talk page instead of starting this general discussion? Anyway, I am now asking you politely to refrain from asking people to do an internship in your favourite areas of Wikipedia to gain your possible future support. If you do not think a candidate can't be trusted with the tools, just do not oppose. It is as simple as that. —Kusma (t·c) 05:53, 29 August 2015 (UTC)
Kusma, that's a strawman argument and a borderline PA. You are perfectly aware that with 'do something about it' I meant start an RfC with some concrete proposals for reform. Kudpung กุดผึ้ง (talk) 07:51, 29 August 2015 (UTC)
I honestly believe that your oppose votes are hurting RfA more than your reform attempts may have been helping it. Reforming RFA by RFC after RFC as you suggest is something that has been tried with a complete lack of success (probably that is because our general governance model is broken), so my current method of improving RFA is to work within the system and support as many candidates as I can and to call out oppose votes that I find unhelpful. (I do find many of your votes unhelpful, but I do not think that is even close to a personal attack). —Kusma (t·c) 08:43, 29 August 2015 (UTC)
YEs, our governance model is far from optimal but we don't have another. I don't believe tactical voting is the cleanest approach but I also don't think you'll curry much favour with the community if you're just going to tell them how they should vote rather than perhaps politely suggesting that their criteria are not only absurd, but in many cases their way of expressing themselves is often beyond the pale. Kudpung กุดผึ้ง (talk) 08:59, 29 August 2015 (UTC)
  • If an RfA fails due to low content contribution, I would expect the candidate to at least give it a shot. Maybe not an FA but something. There is no rule that good vandal fighters cant be good content contributors. Even with nearly 5 million articles, there is still much to be added to the project. I understand that not everyone is a good writer, but we're not always looking for perfect prose on some articles, just an improvement. Obviously an editor can do what he wishes after a failed RFA, but I think he should definitely work on content before reapplying (if that was one of the main reasons for failure). A little AfD participation never hurt anyone, and I dont think it is always boring. Sometimes it can be just as much fun saving an article than writing if from scratch. ~EDDY (talk/contribs)~ 20:16, 28 August 2015 (UTC)
  • I concur with you, Kusma, as far as asking people to "jump through hoops", as I've called it in the past, is concerned. Nobody should be asked to jump through hoops, dignity should be respected and retained. We're all volunteers. However, if a candidate has mis-stepped in an area, it's only fair to tell them what kind of experience may give them a better understanding of why their response to a situation was inappropriate - in fact, our discourse culture requires that the reasoning for !votes be laid out (sometimes excessively). Without knowing the details of the particular case that is causing your complaint, I will add that areas of WP may not be as easy to isolate as you suggest. At our core, there is content creation, copyright law and some other issues that are probably impossible to avoid. There is a notion that something may have received "administrator attention", and an expectation that any obvious problem would have been addressed. If a candidate has incomplete experience, and misses something crucial, it can create subsequent drama. Ultimately, I think this all feeds into discussions on further unbundling. Samsara 03:51, 30 August 2015 (UTC)