Open main menu
Wikipedia Help Project (Rated NA-class, Top-importance)
This page is within the scope of the Wikipedia Help Project, a collaborative effort to improve Wikipedia's help documentation for readers and contributors. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks. To browse help related resources see the Help Menu or Help Directory. Or ask for help on your talk page and a volunteer will visit you there.
 NA  This page does not require a rating on the project's quality scale.
 Top  This page has been rated as Top-importance on the project's importance scale.


Resysop criteria: RfC on principlesEdit

One of the biggest issues with any RfC about activity requirements is that they fail because no one likes specific proposals. To try to deal with that issue, I'm starting the first of what is a potential two part RfC with the following options:

  1. Option 1: The status quo – the resysop criteria for inactivity should remain the same.
  2. Option 2: Stricter – the resysop criteria for inactivity should be made stricter in some way.
  3. Option 3: Looser – the resysop criteria for inactivity should be made less strict.

Ideally, if option 2 or 3 gain consensus, a second RfC focusing on specific proposals based on the principle that achieves consensus would take place. The goal here is to see what the community consensus on the status of the current policy is, not to discuss specific proposals. TonyBallioni (talk) 03:25, 20 July 2019 (UTC)

Option 1Edit

  1. As an example, El C returned in 2017 after being mostly away for 8 years.[1] His work is just as valuable and competent as it ever was. Jehochman Talk 03:43, 20 July 2019 (UTC)
    El_C is indeed very good, but even in that case, I think it took some time for them to get back up to speed. In any case, one specific example doesn't really prove the general case, and (at this point) we have no idea if the re-sysopping criteria are tightened up, if El_C would have qualified or not, mostly because we don't know what they are yet. Beyond My Ken (talk) 05:21, 20 July 2019 (UTC)
    Just pointing out that the last time El C was desysopped for inactivity was in 2013, and he was resysopped within the month at his request. To my knowledge, there has been no serious discussion of concerns about his adminship since that time. Therefore, I return to my key point: resysop criteria aren't the issue here, it's just code for "change the retention criteria". Risker (talk) 00:25, 23 July 2019 (UTC)
    El_C is an outstanding administrator; he has both the courage to stand up for what he believes is right, and the patience to engage with others who may be having a rough time finding their place in our community. Wikipedia is much better off with him having the tools. Stricter resysop criteria might have prevented his return, and that would have been a clear net negative. Kurtis (talk) 07:22, 23 July 2019 (UTC)
  2. Adminship is really about the person behind the account. And as long as the person is the same, I'm happy for people to regain access. I also think we're going up the wrong tree here. The reason people are concerned about people who might not be perfect or are outdated regaining admin status is because removing that status when problems do occur is insanely hard. I think fixing that is a much more worthwhile (albeit more difficult) problem to solve. Legoktm (talk) 03:55, 20 July 2019 (UTC)
    Not fixing Problem A because the vaguely-related Problem B also exists is not good logic, especially when Problem B has been a spectacularly hard one to solve after many years of attempts to do so. Beyond My Ken (talk) 05:21, 20 July 2019 (UTC)
    • Moved to Option 2. Wug·a·po·des​ 05:32, 23 July 2019 (UTC)The bit is a set of tools. If someone still finds use for them, even limited use, it's a net negative to the encyclopedia to remove them as long as the editor isn't causing problems. Recent changes to the software which prevent admins unblocking themselves limits the potential damage from compromised accounts, so security arguments hold less sway for me than they would have a few months ago. That said, I do appreciate the concerns about security and could live with some minor tightening to something like "make one logged administrator action every so often" to show that the editor actually has a use for the tools. In general though, if an editor was competent enough 3 years ago to use the mop, I trust that they're competent enough to read the fucking manual before doing things they know could cause problems. Wug·a·po·des​ 05:24, 20 July 2019 (UTC)
  3. Admins have been selected as they are responsible and will not do things beyond their competence. Some of use that have been admins for a long time have probably found a niche for actions, and returning to that area should not have issues. when going to unfamiliar territory, I would expect them to find out the rules for that. There was no evidence supplied of problems, so the rules have no reason to be amended. Graeme Bartlett (talk) 05:32, 20 July 2019 (UTC)
  4. If it ain't broke, don't fix it. We may regain a useful, productive administrator. We may regain an administrator who uses their tools appropriately but infrequently. Both of those options are good for the project, although obviously the first is the best. Or, we may welcome back a rogue administrator. We have plenty of tools to deal with that, and the scanty record of that specific adminstrator would be a major factor in any discussion about their conduct. In conclusion, we need more administrators rather than fewer. Do not brush off our early administrators who did not resign under a cloud. Welcome them back, while encouraging them to familiarize themselves with our current norms, and to be cautious. Cullen328 Let's discuss it 06:13, 20 July 2019 (UTC)
    If it ain't broke, don't fix it. It's broke. Fix it. —/Mendaliv//Δ's/ 06:18, 20 July 2019 (UTC)
    Mendaliv, please present convincing evidence here that the current process is broken. I readily admit that I am an old guy who does not follow the recent trends closely. I want to see rock solid evidence that this is an ongoing problem. Thanks. Cullen328 Let's discuss it 06:25, 20 July 2019 (UTC)
    Cullen328, personally I think any policy that winds up with someone saying "I'm uncomfortable with this but my hands are tied" is a policy that at the very least needs a review. I'm definitely not advocating that we force all inactive admins through RfA, but I think there is a middle ground between that and a full, immediate, automatic re-sysop. CThomas3 (talk) 20:58, 20 July 2019 (UTC)
  5. I'm going to plant my flag here for a couple of reasons - firstly, I think the activity criteria are absurd, easily gamed and even for active administrators who do a couple of logged actions every few months, they can pick the easy, non controversial, can't-go-wrong actions to carry out. They can close an obvious XfD as 'delete' and then delete the page/file, they can CSD a very obviously spammy/promotional page, they can CSD a clear cut copyright violation, or they can block a slam-dunk vandal. It's largely what I do, to be honest, because I lack the time to review and potentially place sanctions on users involved in complex situations. It means that getting back up to speed with current trends and thinking on complex situations, on things like current discretionary sanctions and the like takes some work and time. In short, despite being a permanent presence here since passing my RfA 12+ years ago, I feel I could easily make the same problematic administrator actions as are being discussed in the section below (though I would like to think, being self-aware in this respect will stop me from doing this). I don't think it's a unique enough problem to pick on the old, gone away for several year types of administrators. My second reason is much more simple - we need different ideas, we need some (more) of the old-guard to come back and give us the benefit of their experiences with Wikipedia, particularly useful, I think, are those who haven't really seen the site and culture much in a lengthy period and can give us a more disconnected, dispassionate assessment of what we're doing right and wrong. Nick (talk) 06:51, 20 July 2019 (UTC)
  6. Agree with Nick. in general, but do not see the harm in what has been described as gaming the system, which does indicate that the person is still alive and paying some attention. · · · Peter Southwood (talk): 07:11, 20 July 2019 (UTC)
    I am going to stick with this option on the grounds that the question is wrong. I do not see this as a place where "stricter" or "looser" requirements for resysopping are meaningful options. What we need to identify are the problems, and the causes of the problems, and then consider actions which adress the problems and the causes of the problems. Changing the criteria without an understanding of the actual problem is poking it with a stick in the dark. · · · Peter Southwood (talk): 09:33, 25 July 2019 (UTC)
  7. Another, "it ain't broke, so don't fix it" !vote. We get is approximately right: if someone's inactive for some time, we suspend their bit but give it back, with a gentle reminder (direct or indirect, just by having to ask) to reflect on how the environment may have changed. If someone is truly completely absent for a Really Long Time, we pull the plug. If someone comes back and mops poorly, we can help them come up to speed, or if there's a deeper problem (unwillingness to adapt), we'll deal with the problem (up to possible desysop) then. But overall, we have much more of a challenge with a) admin retention, b) admin tunnel vision from what they go through in the trenches, c) proliferation of rules, and rules-based thinking as opposed to clue, than with admins breaking the wiki after returning from inactivity. Martinp (talk) 12:10, 20 July 2019 (UTC)
  8. Per Cullen328. GABgab 14:48, 20 July 2019 (UTC)
  9. Unless there's a way to de-sysop some of the very old-time admins who haven't been active for a decade, I don't think changing the current re-sysop rule solves any problems. power~enwiki (π, ν) 15:56, 20 July 2019 (UTC)
    Erm, they're automatically desysopped after a year of inactivity? What's being discussed here is under whether they should be allowed to come back automatically, and if not then under what terms. ‑ Iridescent 16:11, 20 July 2019 (UTC)
    I mean, the problem I see isn't the lack of recent activity, it's the comparatively lax standards for adminship historically, combined with the 2-edit-per-year activity for ten years. I'm not opposed to a token amount of re-engagement with the project for re-adminship (maybe 10 edits to article space). power~enwiki (π, ν) 16:52, 20 July 2019 (UTC)
    power~enwiki, if you're not opposed to token amount of re-engagement with the project for re-adminship, I'm curious why you're here. There are no proposals for specific requirements now. All this RfC is asking is whether or not the community wants any change at all to the resysop policy. If you look at the comments below, most of them would likely be fine with a requirement for token re-engagement as something better than what we have now. TonyBallioni (talk) 17:01, 20 July 2019 (UTC)
    I do see a difference, though you may not. An activity guideline doesn't change who is eligible for re-adminship, it just means they have to jump through some hoops first. I'm fine with hoops, I oppose a change to who is eligible to go through those hoops. power~enwiki (π, ν) 21:11, 20 July 2019 (UTC)
  10. I'm with Cullen328 on this. People have been routinely complaining when old admins request their bits back, citing fears of out-of-touchness and the like but so far, I cannot remember any instance of such a resysop actually causing problems like that. Section 2 is full of comments like "Far too many controversial resysops", "obviously" and "aren't fit to return immediately" but can anyone actually provide examples of that? We don't block before a user has done something wrong. We don't protect articles before something happened. Why should we require returning admins to go through hoops (RFA or similar) if there is no evidence whatsoever to support it? This sounds like process for the sake of process... Regards SoWhy 16:55, 20 July 2019 (UTC)
  11. I personally think that the current criteria is appropriate. For me, the main concern with resysop after a long period of inactivity is that the risk that the account may have been compromised and is no longer under the control of the original user. As long as that issue, which I believe is rare, has been protected against then I don't see the real need to make the process any tougher on returning admins. Michepman (talk) 18:21, 20 July 2019 (UTC)
  12. If we want to end "adminship is for life", we should end that, not fiddle with the re-sysop procedures. Making RFA expire after 10 years would make more sense. —Kusma (t·c) 18:46, 20 July 2019 (UTC)
  13. If I recall correctly, the initial impetus behind introducing activity criteria for admins was one of security of admin accounts, there had, I think, been a hijacking of inactive admin accounts. This has since been countered by the introduction of activity criteria for admins, the requirement for stronger passwords for those with advanced permissions and the introduction of 2FA.
    This RFC makes no mention of securing admin accounts, rather it seems to seem to be about ‘I don’t like it’ when admins request their bits back after a lengthy period of inactivity. No evidence has been provided that the current activity criteria are problematic in any way. For these two reasons - that this RFC is not concerned with the security of admin accounts and no evidence has been provided that there is a problem with the current activity criteria - any proposal for changes to the current activity criteria should fail. Malcolmxl5 (talk) 00:56, 21 July 2019 (UTC)
  14. This would just create more instruction creep for what is already an extremely difficult to comprehend policy. Remember when being a sysop was no big deal? —pythoncoder (talk | contribs) 16:38, 21 July 2019 (UTC)
    I’m curious how an RfC with no actual instructions or proposal can lead to instruction creep? If anything, any change would simplify the complex timelines. TonyBallioni (talk) 17:39, 21 July 2019 (UTC)
    TonyBallioni, option 2 says "the resysop criteria for inactivity should be made stricter in some way". That includes silly things like "must have at least 5 edits to WP:ANI in the last 5 years, at least 12 blocks during the last 1500 days, and at least 50 edits and 3 other log entries in the past 327 days". I don't see anything in the proposal that makes me expect it will become less complex than it is now. —Kusma (t·c) 18:15, 21 July 2019 (UTC)
    That’s a strawman argument about proposals that have never been made, would never be made, and if made would get voted down in a heartbeat. The two lines that are getting the most traction below are adding some discretion or lowering the time to request resysop. Both of those would likely be easier than the current 3/5 calculation, but, again, this isn’t even about specifics, it’s about if the community is satisfied with the current language of the policy. TonyBallioni (talk) 18:22, 21 July 2019 (UTC)
    (edit conflict)Why would that necessarily be “instruction creep”? It’s a simple rule as it is. Refinement and improvement of a rule isn’t instruction creep. —/Mendaliv//Δ's/ 18:23, 21 July 2019 (UTC)
    If we're concerned about the RESYSOP policy being too complicated, we could always do away with it entirely. Then every resysopping would have to be by consensus at RfA. Perhaps that should be one of the options in the upcoming RfC (if Option 2 wins) for those concerned about instruction creep. – bradv🍁 18:34, 21 July 2019 (UTC)
  15. I am in the "not broke/don't fix" camp and endorse the comments of User:Malcolmxl5. --kingboyk (talk) 00:14, 22 July 2019 (UTC)
    LOL @Kingboyk: [2] ——SerialNumber54129 14:10, 22 July 2019 (UTC)
    Do you have a point to make or are you just having a laugh at my expense? --kingboyk (talk) 17:02, 22 July 2019 (UTC)
  16. Essentially, I'm with Martinp here. What use is it to make returning a tad more difficult, when the real problem might be a compromised account? Lectonar (talk) 07:44, 22 July 2019 (UTC)
  17. Show me some examples of long-term absent admins returning where they screwed everything up and/or went on an insane rampage - enough to outweigh the many examples of long-term absent admins returning and doing so successfully - and I'll change my mind. I am tired of saying we need to make the climb to admin status less steep, not steeper. Fish+Karate 12:57, 22 July 2019 (UTC)
    Perhaps the unfortunate Andrevan case would sway your opinion? That heap of WP:RGW drama wasted a lot of editor and admin time, starting at AN and ending at ArbCom. Granted, Andrevan had not lost the tools due to global inactivity, but he was making little use of them until he went ballistic on editors he perceived as "paid Russian/GOP/NRA advocacy agents". The recent desysop of Rama is also a case of tool abuse to right great wrongs by an otherwise rather passive admin. — JFG talk 13:01, 23 July 2019 (UTC)
    @JFG: Thanks for the examples, I agree these were serious issues but I don't think either Andrevan or Rama were resysopped before their escapades? Fish+Karate 15:16, 23 July 2019 (UTC)
    That is correct; however they are representative examples of the "I'm infallible" mentality of certain early-days admins who have made no significant use of their tools in several years, and suddenly wield them in disruptive ways. A lack of use of admin tools should in my opinion lead to automatic removal after a couple years, under the "no big deal" philosophy; I do realize that's probably for a different discussion about the inactivity threshold. — JFG talk 15:28, 23 July 2019 (UTC)
    As someone who strongly argued in favor of Rama's desysop, I certainly understand where you are coming from. However, those cases also demonstrate that the established processes can handle such admins if and when they decide to start a RGW-crusade, To be fair, I will admit that my view on this is probably influenced by the fact that I might have been one of the "victims" of such a stricter rule during my real life induced 2012-2016 stretch of low activity. Regards SoWhy 15:43, 23 July 2019 (UTC)
  18. Per SoWhy. Useight's Public Sock (talk) 14:34, 22 July 2019 (UTC)
  19. The request misses the point. It's not the resysop requirements that are at issue. It's the retention requirements that are problematic. I have yet to see any evidence from anyone that there has been a problem with inappropriate admin actions of resysopped admins and, absent that evidence, this is a solution looking for a problem. Now, if you were going to propose that we change the retention criteria, I'd look at things a lot differently. Risker (talk) 00:20, 23 July 2019 (UTC)
  20. I largely agree with Cullen328 and SoWhy. There's no evidence that resysopped administrators have caused any problems, so why prevent old hands from returning? Most administrators are competent enough to get themselves up to speed before doing anything reckless. Kurtis (talk) 07:34, 23 July 2019 (UTC)
    I should add, however, that I like the suggestions of Mendaliv and Cryptic in the Option 2 column. Kurtis (talk) 07:41, 23 July 2019 (UTC)
  21. Although it certainly can be gamed, I think the most important criteria for restoring admin status isn't edit counts but Over five years since administrative tools were last used. If a former admin hasn't performed any admin activity in over 5 years, they should have to go through an RfA. So much can change over 5+ years. As this is the current policy, I guess I'm supporting keeping things the way things are. Liz Read! Talk! 00:24, 24 July 2019 (UTC)
  22. I am honestly failing to see the point of this? The main complaint about the RfA process, as far as I ever recall is the that fact that it is (typically) an awful process, one where anyone you’ve ever pissed off (and their mates) will scrutinize minute details of edits to make you look like the worst candidate ever. That issue cannot be resolved by making the process “easier” or more difficult. Further, in terms of inactive admins — who cares? I was effectively MIA between 2010-2015 and even then not really doing a huge amount of admin tasks again until May of this year. I was busy. Life happens. That’s the point of volunteer work. Unless you prove yourself incapable why the hell are people so worked up to ensure admins are “in the loop” with all the Wiki-drama they may have missed? Hell, many admins stay out of the drama and stick to doing “admin work” and effectively are active but out of the loop. It’s a ridiculous concept. When I returned from a long period of inactivity Wikipedia didn’t burn down (and let’s be honest the basics around admin tasks have largely unchanged over the years). Anyhow, none of these proposals, in my opinion, addresses how to make the RfA process “better”. There are fundamental issues to address, which, in my opinion, behaviour of groups of people against others is a major one that feeds into the RfA issue as well. This is a deeper issue and not something that can be fixed by amending a procedural process. N.J.A. | talk 01:51, 24 July 2019 (UTC)
    As someone who, in 105 consecutive months had exactly five months with more than 20 edits (July 2010 to April 2019), I'm really not surprised you don't see the problem. —/Mendaliv//Δ's/ 01:59, 24 July 2019 (UTC)
    Do tell. What do you believe the issue is that I have missed due to a random number you’ve cited? Specific examples are ideal, e.g. something I missed during the period of inactivity along with examples of how, upon my return, I clearly demonstrated an ineptitude due to said inactivity and further what you propose would resolve such issues moving forward? N.J.A. | talk 10:02, 24 July 2019 (UTC)
  23. Being one that was away for a while at times, then coming back I'll say this: People are elected to RfA because the community trusts them to use common sense in how they act. That doesn't change over time. If someone comes back and makes a mistake, that's fine, and we show them what's new. If someone comes back and screws up a lot - then they shouldn't have been an admin. in the first place. (and they get desysoped) If there were a flood of returning admins being brought to wp:rfar then I could see changes being made. Trying to fix a problem that doesn't exist usually creates tangent problems. (edited 8/3/2019) — Ched :  ?  — 12:39, 26 July 2019 (UTC)
  24. I've seen that most admins have enough sense to get themselves up to speed on what's changed after an extended absence. No sense in attempting to reinvent the wheel IMHO. OhKayeSierra (talk) 01:16, 29 July 2019 (UTC)
  25. Largely a solution in search of a problem. Neutralitytalk 18:37, 3 August 2019 (UTC)
  26. I worry about minimum requirements exacerbating burnout / stress in existing admins. I'd rather an admin that was in a period of stress took then time they needed without the added stress of worrying about hitting some arbitrary limit of activity. Such policies have to be consider in terms of who they might impact. I worry making the requirements any stricter will slightly increase the chances that admins facing burnout may suffer a breakdown which is bad for the project as well. Without stronger evidence of the harm of allowing admins to return I can't support strengthening the requirements. PaleAqua (talk) 21:03, 3 August 2019 (UTC)
  27. We need more good admins. Keeping it relatively easy for former admins to return may help. Jonathunder (talk) 04:27, 4 August 2019 (UTC)
  28. If the user has not done anything to show that s/he can't be trusted with the tools, then why not? --rogerd (talk) 04:46, 4 August 2019 (UTC)
  29. We're doing fine. No change is needed here. -- œ 05:20, 4 August 2019 (UTC)
  30. Three reasons. One, Per u:Risker, the problem is not in being to easy to (re)become an admin, the problem is that it is too hard to remove one that is causing trouble; Two, per u:Nick, low activity admins are not a problem (I think, I do only a few easy actions when I have the time, so that the ones that are more involved may focus on harder tasks, am I hurting anything?); Three, all mine, I know it may be hard to get consensus, but I rather dislike a RfC process of slicing out opposition, this RfC may be approved by 60% then some stricter rule (out of a set of only stricter rules) be approved by 60% and bingo! a "consensus" of 36% is born. - Nabla (talk) 10:48, 4 August 2019 (UTC)
  31. The system works fine as it is. Mjroots (talk) 11:50, 5 August 2019 (UTC)
  32. Solution looking for a problem. Stifle (talk) 15:26, 6 August 2019 (UTC)
  33. Per Fish+Karate. I think the drama and stress created by extra RfAs would surpass the drama and stress caused by extra desysoppings. DaßWölf 03:26, 7 August 2019 (UTC)
  34. Solution looking for a problem, if it ain't broke don't fix it . Note: I had trouble doing admin tasks from an Iphone when my computer died.. Got another computer and now it becomes easier. I just want an automatic way to avoid someone finding the username and password of an inactive, incapacitated or dead admin and coming to do mischief, or even selling it on the dark web.. A long period of time with no activity leading to deadmin satisfies my concerns. If an admin spends some time in military service, recovering from illness, doing a dissertation, or launching a startup and wants to come back I would welcome them. Edison (talk) 21:33, 7 August 2019 (UTC)
  35. I've read all the Option 2 votes and discussions below, and did not find a single example of a (real, not potential) problem arising from the current policy. Vanjagenije (talk) 14:52, 23 August 2019 (UTC)

Option 2Edit

  1. There have been multiple controversial resysops within the past year. While I do not personally think having an extremely strict resysop criteria is ideal, 1 logged action 5 years before an inactivity desysop and a logged in edit or action within 3 years is a bit ridiculous. Even in the 3 years since I have become active on the project, the culture and policy applications have changed. I can't imagine coming back after 2 years and 364 days and being familiar with everything that was going on. Having some higher standard is needed. I don't know what that standard is yet, but moving in that direction and recognizing we need to move there is important. TonyBallioni (talk) 03:25, 20 July 2019 (UTC)
    But the wiki is pretty big, so you (and any other admin) is not familiar with everything going on whether your last action/edit was 5 secs ago or 2 y + 364 days. If we decided years ago we trusted someone to have enough Clue at some point to not break stuff, why can't we by default trust them to get informed before acting in either case? Not trying to pick on you specifically, but I'd wish we'd harness the energy behind these periodic community outbursts of "Must Man the Ramparts Against Dangerous Returning Admins" into "let's help admins at any level of activity stay on top of changes in policy and its application." Martinp (talk) 12:27, 20 July 2019 (UTC)
    Admins are expected to participate in the project if they're supposed to be admins for it, and given that it's a dynamic project, one should be expected to at least try to keep up with changes. This proposal is in fact exactly "let's help admins at any level of activity stay on top of changes in policy and its application". No one who doesn't participate on the project can be helped so that is the first step.--Jasper Deng (talk) 17:42, 20 July 2019 (UTC)
    As far as I can tell, the "controversy" was pretty much at BN, caused by people who don't like the current system but are unwilling or unable to provide any evidence that it's not working; the rules were followed, generally speaking, and I have yet to see any evidence that any of the resysops resulted in any problematic administrator activities by the resysopped admins. Risker (talk) 00:40, 23 July 2019 (UTC)
  2. While I'm pleased that we have an automatic desysop criteria, which we didn't have before, I agree that the resysopping criteria is much too lenient. While many people say that we need more admins, in actuality what we need is more active admins, not hat collectors. The criteria should help ensure that if resysopped, the admin wants to be, and will be, more active. Beyond My Ken (talk) 03:35, 20 July 2019 (UTC)
  3. Support: Honestly this is the opportunity to begin trying out some continuing education mechanism. Conditioning restoration of admin rights for people who have been out of the loop for more than a year on taking some kind of continuing education gives us a great testbed for this, frankly, which should be a requirement for all admins. —/Mendaliv//Δ's/ 03:39, 20 July 2019 (UTC)
    I agree. Beyond My Ken (talk) 05:15, 20 July 2019 (UTC)
    In principle, an interesting and possibly useful suggestion. Take it further. · · · Peter Southwood (talk): 07:17, 20 July 2019 (UTC)
  4. Long term we need to draw back in formerly active project members. But that doesn't mean an easy reentry to sysop is the right way to do so. There are several ways I can think of that I would support as drawing a better balance between competing priorities. Best, Barkeep49 (talk) 03:44, 20 July 2019 (UTC)
  5. Sadly probably symbolic because long-term inactive admins will suddenly reappear and oppose this proposal. But I support this. --Rschen7754 03:49, 20 July 2019 (UTC)
  6. As an admin who returned after many years away. If nothing else, I'd expect a reasonable amount of activity before being resysopped, the same way I'd expect an inactive but not-actually-desysopped user to spend some time reacquainting themselves with the project before diving right back into administrative tasks. —Cryptic 03:56, 20 July 2019 (UTC)
    I've put myself in Option 1 above (since I really don't think there's a problem worth fixing by more Rules here). But I wouldn't oppose a modest editing activity requirement before re-adminning, like "edit (outside own talk page) for at least [5] days in the last month", something that encourages ongoing re-engagement with the community for a while, not something that just encourages going and doing a mass of mindless edits in one evening. Martinp (talk) 12:33, 20 July 2019 (UTC)
  7. Support per above. Far too many controversial resysops for my liking. If people aren't going to use the mop and bucket, then they don't need access to them. Adminship is not a trophy. -FASTILY 05:27, 20 July 2019 (UTC)
  8. I think Cryptic has an excellent idea here. It’s not that I oppose re-sysopping of long-inactive admins, because I don’t, but why are we doing so immediately upon their return? There is plenty to do as a normal user, and we should be requiring at least some demonstration of sustained activity and ongoing commitment to the project prior to a re-sysop. We expect no less of our new admins. A review of this activity wouldn’t require a RFA, but could essentially skip straight to the ‘crat chat. CThomas3 (talk) 05:31, 20 July 2019 (UTC)
    A very reasonable condition. · · · Peter Southwood (talk): 07:17, 20 July 2019 (UTC)
  9. Support I've never been a fan of those who play the system and make a token edit once every 12 months to hold onto the tools. I guess though, by default, they're not admins, as they're not here to do any admin-esque work. Maybe a cull of the most repeat offenders would be the solution. If they feel strongly about retaining the tools, then an RfA would suffice post-desyopping. Lugnuts Fire Walk with Me 05:43, 20 July 2019 (UTC)
    I guess though, by default, they're not admins, as they're not here to do any admin-esque work. One of the differences, though, is that these folks still have access to deleted materials. People who are not current on the standards for administrator conduct shouldn't have access to deleted materials, which are often sensitive for reasons that don't meet the oversight standards (or which have simply escaped notice by oversighters). —/Mendaliv//Δ's/ 06:13, 20 July 2019 (UTC)
  10. Obviously, yes, the current requirements are far too loose. Boing! said Zebedee (talk) 05:47, 20 July 2019 (UTC)
  11. Obviously. We don't need a complicated set of timescales; all we'd need is the addition of a single line along the limes of "if the applicant was desysopped more than (six months? twelve months?) before requesting the restoration of permissions, the restoration will not be automatic but instead will be subject to a vote among the current bureaucrats". Since this discussion has already been personalised I'll take the same example; if El_C is such a fantastic asset to the project, he presumably would have had no problem demonstrating that to the 'crats and convincing them that giving him the admin bit would be a positive. ‑ Iridescent 06:22, 20 July 2019 (UTC)
    So you would add a rule "crats should restore adminship if they believe this is a positive thing"? That sounds like a quasi-RFA in which only crats participate and that basically overwrites the RFA consensus that granted the sysop in the first place. What if crats believe someone should not be resysoped because they were a controversial admin back when they were active even if they were never formally admonished? Crats are tasked with interpreting consensus and this would give them the ability to not only establish consensus but replace community consensus with their own. That seems problematic. And yes, I would probably trust the current roster of crats to make the right decisions in such cases. Regards SoWhy 17:07, 20 July 2019 (UTC)
    If I were Dictator of Wikipedia I'd either make adminship auto-expire after three (maybe five) years of activity or one year of inactivity, with anyone wanting to continue as an admin having to re-sit RFA, or go with the de-wiki model of cumulative complaints forcing a reconfirmation once they reach a certain level—either would simultaneously purge Wikipedia of those who'd lost public trust, and bring back "no big deal" as a concept since the RFA voters couldn't be too picky unless they wanted to wreck the project altogether. However, I don't believe there's sufficient appetite for such a radical change, so the best we can do at present is prune away the most egregious elements of "adminship is a lifetime position" and then see where we stand once the "I will give up my admin bit when you pry it from my cold dead fingers" contingent have realized that the change has occurred and the sky has not fallen in. ‑ Iridescent 18:58, 20 July 2019 (UTC)
    On the basis of this comment alone, Iridescent, you have my !vote whenever you decide to open your RfDoW. Happy days, LindsayHello 22:26, 20 July 2019 (UTC)
  12. Support automatic reysop for historic users is an unnecessarily loose position to hold, the tools and trust levels are very different in modern times and as such are given less freely than they were in the early days. Can I have my advanced permissions back please when there is no evidence at all that they will or have the experiance to actually use them for the benefit of the project seems much less beneficial than offering the tools to really currently active experianced contributors that would actually use them. Govindaharihari (talk) 07:01, 20 July 2019 (UTC)
  13. Support per supporters; suggest that!voting on this RfC, parts one and two, is restricted to editors with over e.g. 25 edits over the last calender month. ——SerialNumber54129 07:09, 20 July 2019 (UTC)
  14. Support If we are discussing ideas, the idea of voting before a resysop sounds okay to me, there's the popular notion of people coming who have an axe to grind but there are literally tons of wikis with proper measures in place and none of them have been set on fire yet. I suggest voting is restricted to editors with more than 100 edits in the last year (i.e. date of voting). --qedk (tc) 07:44, 20 July 2019 (UTC)
  15. Support. Stricter resysop criteria would encourage participation from administrators, and help ensure that admins are up-to-date with changes to policies and guidelines. Tightening the criteria would be a net positive if the resulting increased participation outweighs the few edits/actions a year we would lose from admins who wouldn't meet (and wouldn't be motivated by) the new requirements. The only way to determine the effects of these changes is to conduct a trial, which I recommend. — Newslinger talk 08:33, 20 July 2019 (UTC)
    Additionally, changes in policy don't necessarily need a perceived "problem" (although others have made the case that the status quo is problematic). Changes can be justified if they encourage positive behavior, including participation. — Newslinger talk 21:43, 21 July 2019 (UTC)
  16. Support. Maybe give them all the other tools and ask them to work on tough areas for a month; and crats take a vote, maybe admins as well, or some other criterion of inclusion for very active very experienced editors. Maybe ask them to edit actively for three months and go through an RfA with strict criteria for eligibility of the cast vote. Anything but the current norm of getting it immediately by asking without anything to show that they mean business and not just trophy collecting. Usedtobecool ✉️  08:57, 20 July 2019 (UTC)
  17. Support. I'm concerned whenever someone who earned the role a decade ago comes back with a few sporadic edits during the interim and is returned the role with effectively no questions asked. I'd be hard-pressed to give someone rollback if their last major spree of editing was during the sites infancy. Make it two years and a required amount of edits and maybe it'll be more understandable. Taking a break for a few months is fine. Taking ten years off is fine too, but one shouldn't expect to come back to the site's most powerful buttons. Anarchyte (talk | work) 09:22, 20 July 2019 (UTC)
    @Anarchyte: this, but it also includes abusefilter, template editor, interface editing, etc etc etc. — xaosflux Talk 13:56, 20 July 2019 (UTC)
    @Xaosflux and Anarchyte: I agree with this. I don't see why other advanced permissions shouldn't be treated similarly. CThomas3 (talk) 20:53, 20 July 2019 (UTC)
  18. Support people who were active a very long time ago but haven't been very active since often do not make good administrators if they return to activity, because they aren't familiar with accepted norms. If they do attempt to perform significant admin actions they often do inappropriate things. The current standard is ridiculously generous. Hut 8.5 10:34, 20 July 2019 (UTC)
  19. Support since wp:competence is required and things have changed fast. Not just admining, but everything starting from how content is created etc. --Pudeo (talk) 10:57, 20 July 2019 (UTC)
  20. Support But mainly only if it is a subjective criteria rather than an objective one. Something like "legitimate issues are raised by the community during the course of the waiting period, and there is a consensus among crats that the user's level of activity is well below that which would normally qualify a candidate to stand for RfA". (Note: stand for RfA, not pass an RfA.) Different objective criteria will only lead to different gaming, and we can't anticipate every situation. We give crats limited discretion in RfA and I don't see why they should have zero equivalent discretion in resysops. We definitely shouldn't rope ourselves into a situation like this where we have crats saying essentially "I'm not comfortable with this but our hands our tied", and people like me trying to carve out the most niche possible crevice in policy to try to justify the concerns of the community (in this case, "are you active enough to give literally any answer whatsoever to issues raised under ADMINACT"). GMGtalk 11:31, 20 July 2019 (UTC)
  21. Support Unequivocally - as I have always maintained at every opportunity. Leaky caldron (talk) 11:47, 20 July 2019 (UTC)
  22. Support . I think this is really a no brainer. The policy needs tightening up but we should wait for the outcome of this RfC before discussing specifics. Kudpung กุดผึ้ง (talk) 11:51, 20 July 2019 (UTC)
  23. Support - The policy does need tightening up, I've never agreed with resysops for those who make 1 logged action every 5 years or a logged in edit or action within 3 years but never opposed them as technically policy currently allows it... so I certainly support tightening the policy up. –Davey2010Talk 12:20, 20 July 2019 (UTC)
  24. Support - it definitely is insanely low. (Some of) the status quo/lower !voters seem to consider that we'd require project wide knowledge to regain the mop, which is clearly insanely high, but the current level would let a returning admin be out of touch with everything since they'd last handled it. Nosebagbear (talk) 12:38, 20 July 2019 (UTC)
  25. Support for all of the "it's no big deal" champions - then make RfA no big deal if you use it again. I'm not sure what I think the best "activity" threshold is, but as far as resysoping I'd like to see a "1 year rule" - If you stop sysoping and it's been a year, go to RfA - leaves plenty of time for temporary leaves of absence. — xaosflux Talk 12:56, 20 July 2019 (UTC)
  26. Support per Fastily. Admins away for years, especially those that hadn't been using tools, aren't fit to return immediately to adminny work. Most just wield the banhammer as a status symbol and for convenience. Chris Troutman (talk) 13:12, 20 July 2019 (UTC)
  27. Support I think there should be some level of renewed editing activity before resysop. The ideas below for continuing/refresher ed below sound promising as well. Schazjmd (talk) 14:04, 20 July 2019 (UTC)
  28. Id actually be in favor of desysopping a number of legacy admins (the ones promoted either without an RFA or when RFA was literally oh he or she has not vandalized in the 100 edits he or she has made so +sysop plz), and these bright line rules lead to things like an admin making a token edit every year for a decade to retain the bit. If RFA is making it too difficult to promote new, qualified, admins then RFA needs to be fixed. That should not mean resyssoping should be as trivial as it is. nableezy - 16:09, 20 July 2019 (UTC)
  29. Support. Having recently returned from an extended wikibreak myself as a non-admin I found myself tripping up in wiki processes more than a few times (things like leaving out bits of expected formatting in article creation). Obviously, for an admin, the problem areas can be much more numerous and potentially more serious. Since people tend to agree that there's a need to get back up to speed after long periods of inactivity, I think it's fine to expect that some evidence be shown that this has happened before handing back the mop. —Nizolan (talk · c.) 16:27, 20 July 2019 (UTC)
  30. Support This wiki has one of the most lenient policies of all Wikimedia projects on this subject and it appears from above that there is a problem caused by this (i.e. "if it ain't broke, don't fix it" does not apply).Jasper Deng (talk) 17:39, 20 July 2019 (UTC)
    Could you maybe point out which problem that is exactly? I see a lot of people claiming there is a problem but none actually specifying what it is or providing examples. I'm happy to be convinced if there demonstrably exists a problem and not just a feeling of a problem. Regards SoWhy 18:40, 20 July 2019 (UTC)
    dormant antique accounts with advanced permissions that they would likely not be granted today with continued access to restricted content? Govindaharihari (talk) 19:24, 20 July 2019 (UTC)
    How is that different from semi-active antique accounts with advanced permissions that they would likely not be granted today? Regards SoWhy 20:01, 20 July 2019 (UTC)
    I would class an account only completing the edits so as to qualify for the continued advanced permissions as basically dormant, there is little difference in benefit to the project for either user to continue to hold access to restricted content. Govindaharihari (talk) 20:32, 20 July 2019 (UTC)
    They're different problems. This RfC is not intended to address the "semi-active antique accounts" problem. —/Mendaliv//Δ's/ 18:30, 21 July 2019 (UTC)
  31. Support My views are similar to Xaosflux above. -- Dolotta (talk) 18:34, 20 July 2019 (UTC)
  32. Support – our resysop policy is supposed to give the crats a very clear line on which users are likely to retain the trust of the community. I am as excited as anyone else when someone returns to the project, but the level of access that the crats give these returning users often does not match the level of trust that the community has in them. It's difficult to give specific examples without appearing to criticize individual admins, but I am convinced that we need to tighten our resysop criteria. – bradv🍁 19:05, 20 July 2019 (UTC)
  33. Support It's all been said above as far as the direction of travel is concerned. Johnbod (talk) 02:42, 21 July 2019 (UTC)
  34. Per Xaosflux and TonyBallioni above. Vanamonde (Talk) 04:55, 21 July 2019 (UTC)
  35. Support Removing the rights is just security theatre if they are returned without any reconfirmation. Andrew D. (talk) 09:12, 21 July 2019 (UTC)
  36. Support As between the three choices this is the sensible place to be: status quo could only mean there is no reason for further discussion in hopes of reaching consensus, which is not the case; looser, is where we've been before and there is no reason for going back. So, here am I, waiting for the details, where they say the devils lay. (Please when you do go to the next step, remind everyone what admins have access to and what they could and can do with it, and how our systems/processes/people know they are who they say are.) Alanscottwalker (talk) 09:32, 21 July 2019 (UTC)
  37. Support. Administrators should be currently engaged with the community and familiar with current policies. The present criteria do not reliably guarantee this is the case. Thryduulf (talk) 09:59, 21 July 2019 (UTC)
    Achieving that would probably also require desysopping at least half of users on the Wikipedia:List of administrators/Semi-active and Wikipedia:List of administrators/Inactive lists, wouldn't it? Regards SoWhy 17:41, 21 July 2019 (UTC)
  38. Support. I hesitated before deciding to support, because there certainly is a very good case to be made that we need to be getting as many active admins as we can. But I end up feeling that there is something essentially illogical in saying that, on the one hand, present-day community norms for new RfAs (whether too high or not) are what they are, but on the other hand, criteria for re-sysopping "legacy" admins should reflect "legacy" standards. And this RfC does not commit us to anything, just to have further discussion and further examination of options if this gets consensus – and I see a clear net positive to opening the door to further consideration instead of shutting the door. --Tryptofish (talk) 16:20, 21 July 2019 (UTC)
  39. Support per Xaosflux and TonyBallioni above .Note most of those who are Wikipedia:Former administrators/reason/inactive have had there RFA between 2003 to 2009 .That is over 10 years ago Pharaoh of the Wizards (talk) 20:06, 21 July 2019 (UTC)
    Over 80% of successful RFAs were in that era, so it should not surprise us that most of the inactives are from then. The bigger problem is that over 95% of successful RFAs were more than five years ago - hence our unhealthy reliance on keeping as many as possible of our long established admins. ϢereSpielChequers 15:56, 27 August 2019 (UTC)
  40. Support 100% - The status quo that was meant to abolish lifetime appointments is still easily gamed. For an example, see Pakaran, an admin and bureaucrat with <10k edits, who was appointed by 13 people with no discussion, has not made a single edit since February 2018, has no 'crat "activity" since December 2016 (that was a single comment at BN), has no logged 'crat actions since February 2015, and has not been active since 2006 (save for a few months in 2010). Another one, Yelyos, a user with 1700 edits, no edits since September 2016, no admin actions since February 2015 (a single deletion), has also not been active since 2005 (2006 if we're being generous). Not meaning to attack these users personally, but it's hard not to see the problem. We need to be able to prevent these situations. ~Swarm~ {sting} 20:33, 21 July 2019 (UTC)
    Swarm, it is easy to see potential problems, but it is hard to see serious actual problems (who exactly is doing something bad with their admin tools?). It isn't a "problem" if Yelyos is an admin, it is just amazingly unfair that RFA was so much more gentle 15 years ago. Probably nothing other than asking all of us old (ex-)sysops to go through RFA again is going to create inter-wikigenerational fairness. —Kusma (t·c) 21:08, 21 July 2019 (UTC)
    If you don't think it is a problem that users who objectively don't meet current standards for these positions are able to retain them by virtue of gaming the low activity requirements, as opposed to actually remaining active, then we'll have to agree to disagree, Kusma. I really don't even feel like it's worth my time arguing whether a user with 1k edits and 15 years of inactivity should be an admin. ~Swarm~ {sting} 00:54, 22 July 2019 (UTC)
    Swarm, as long as no real activity is needed to stay an admin, I just don't see what a change to the resysop rules accomplishes (other than annoying people who have a real life for a while). People who became sysops in the old days were trusted to have clue and not break the wiki. People who became sysops a bit later had jumped through a few hoops and were trusted to have clue and not break the wiki. Only a couple handful of people were made sysops in the last couple of years. People in all three groups have been desysopped for showing a lack of clue and/or breaking the wiki. I just don't see the evidence that our new standards have accomplished much so I don't see not meeting the standards as an obvious problem. But if it is such a problem, we should go away from "adminship is for life if you never become inactive" instead of worrying about what the precise amount of inactivity is that we wish to allow. —Kusma (t·c) 06:07, 22 July 2019 (UTC)
    Can you point me to a discussion where the consensus was that the intent of the status quo is to abolish lifetime appointments? As far as I recall, we are at where we are now for reasons of security, and security alone. --kingboyk (talk) 00:12, 22 July 2019 (UTC)
    I'm not sure what you're even getting at. "Lifetime appointment" is not and never has been a formalized concept. It was simply the reality of how things used to be. At some point the community abolished this reality when it implemented the "5-year rule" which required a re-RfA due to inactivity. This replaced the "lifetime appointment" status quo with an activity requirement. In other words, the community mandate from an RfA is no longer indefinite, but is presumed to expire after 5 years of inactivity. I'm not interpreting anything in terms of "intent". This is literally the current rule. If you're curious what the intent was, you can dig up that discussion yourself, I have no idea where it is. ~Swarm~ {sting} 00:54, 22 July 2019 (UTC)
    The RfC establishing 5 year rule is here. The current activity policy is based on the fact that adminship was basically something thrown together when this was a hobby website of Friends of Jimmy® who never thought it would be where it was today, and who knew if it would reach 10 years much less 20. The de facto appointment for life thing is something we inherited because we never bothered to set a policy around it until it was pretty late in the game. Most other projects have significantly stricter resysop policies. While I do not want to go as strict as some of the others, I do think that our current policy on restoration of rights may as well not exist because it is so lenient, and to be honest, if I weren't in "Option 2" I'd be in "Option 3" because the status quo is a joke. TonyBallioni (talk) 01:14, 22 July 2019 (UTC)
  41. I echo Xaosflux and TonyBallioni here. This needs to be done. Nihlus 21:32, 21 July 2019 (UTC)
  42. Somewhat preliminary, as I may or may not support any specific measure. But asking some questions to admins seeking return on WP:BN seems like a reasonable move. Jo-Jo Eumerus (talk, contributions) 08:37, 22 July 2019 (UTC)
  43. Xaosflux's "no big deal" point is excellent. ~ ToBeFree (talk) 01:52, 23 July 2019 (UTC)
  44. I've seen some WP:BN resysops where the ex-admin could not even acknowledge that they needed to do some work to update themselves before using the tools, so I support a mild strengthening of requirements. Johnuniq (talk) 02:08, 23 July 2019 (UTC)
  45. Per xaosflux mostly. I wasn't too thrilled with option 1 to start with, but my main reason was that even if the returning admins don't do much, what little they do would be better than nothing. The gain from that would probably not be significant, so whether that's delayed 7 days for an RfA wouldn't make much of a difference. And at least this way we know the community supports them. So I don't see much reason to oppose tightening them. Wug·a·po·des​ 05:43, 23 July 2019 (UTC)
  46. Wikipedia standards have evolved, so that long-absent admins should both demonstrate a current need for the tools and an awareness of what is expected of them today. Going through a full-blown RfA again may not be the appropriate process, because there would in most cases not be enough recent editing history to pass judgment, so that would be mostly a popularity contest. In the spirit of "tools are not a big deal", I would rather advocate for a probation period of 6 months, within which the returning admin should be able to demonstrate their competence. What to do after 6 months should be up to a hopefully lightweight community discussion, format to be discussed in followup to this RfC. — JFG talk 11:58, 23 July 2019 (UTC)
  47. SUPPORT - All you have to do is observe the absolute reaming that even pretty good editors receive at WP:RFA (it's a dragon's den - I wouldn't pass - but that's another topic) whilst trying to become admins to realise just how unfair it is that people can waltz back into adminship with relative ease. No, you go back through WP:RFA along with everyone else, and have to show there that you can do the job. FOARP (talk) 15:20, 23 July 2019 (UTC)
    EDIT: I would also like to see a time-out period (3-6 months) before re-sysoping where the de-sysoping was for cause or for non-activity. The editor in question should have to show they are a good citizen, and are up to speed on current policy, before reapplying. FOARP (talk) 08:51, 29 July 2019 (UTC)
  48. Support, mostly per Iridescent. Note that we've had a couple of particularly egregious examples of dubious resysop'ing lately that really do seem to be in the "it's a trophy" category - if they want a trophy, edit some articles or whatever and perhaps someone will present a barnstar. - Sitush (talk) 13:14, 26 July 2019 (UTC)
  49. Support: tightening the activity requirements makes sense at this point. --K.e.coffman (talk) 00:02, 27 July 2019 (UTC)
  50. Support Our resysop criteria are really a joke. Returning admins should have to demonstrate at least some engagement with the community. Pawnkingthree (talk) 12:09, 27 July 2019 (UTC)
  51. Support Honestly, I can't think of any reason why all resysops shouldn't just go to RFA. It would put an end to puerile protest resignations and to the question of trophy holding, whilst also ensuring that the WP:RFA maintains a level of realism about adminship candidates. Xaosflux and Iridescent have good ideas that are worth serious discussion at a later stage. Promethean (talk) 09:30, 1 August 2019 (UTC)
    @Promethean, I personally wouldn't support all resysops going through RFA. There are some short-term desysop/resysop cycles that genuinely do fall under routine maintenance ("I'm spending the next three months at sea and know I won't have internet access, please temporarily remove the admin bit in case my account is hacked while I'm away and unable to do anything about it", "I am working in North Korea for the next month and am uncomfortable potentially giving the authorities there access to my checkuser permissions", "I'm on medication which has documented psychoactive side-effects, please temporarily remove the admin bit until the course of treatment is complete so I don't end up blocking or deleting something during a paranoid episode"…). Making all resysops have to go through a full week-long RFA every time would likely have a counter-productive effect by discouraging people from temporarily giving up advanced permissions in circumstance when they really should. ‑ Iridescent 08:44, 2 August 2019 (UTC)
    Right, this is reasonable. Not all resysops should go through RfA, but I think most if not all that are properly characterized as resignations and retirements should. But as you say, there are legitimate reasons to ask for the advanced permissions to be temporarily removed. —/Mendaliv//Δ's/ 08:52, 2 August 2019 (UTC)
  52. Support {{u|waddie96}} {talk} 06:51, 4 August 2019 (UTC)
  53. Support I'm awfully late to this discussion, and I don't know if we're going to have that second RfC. But in case we do, I'd like to note that, according to my rough count, approximately 80% of the editors favoring Option 1 are admins. While admins certainly have as much right to weigh-in as non-admins, it seems that admins are a little more likely to weigh in if the discussion is held on this page. If broader input was sought from the entire community, perhaps via a watchlist notice, it would be very informative to see how many non-admins would support Option 1. As for myself, I support tightening the requirements. Lepricavark (talk) 14:53, 13 August 2019 (UTC)
  54. Support I would recommend having a requirement for a certain number of edits, say 25 or 50, during the 30 days prior to their resysop request. This would indicate that they are coming back to the project in earnest. -- Dolotta (talk) 21:32, 13 August 2019 (UTC)
  55. Support People will always game the system, whatever it is, but a stricter system would have better results. Enigmamsg 23:27, 24 August 2019 (UTC)
  56. Support Long overdue. There is an observable trend of a small minority of admins who clearly are gaming the current requirements despite clearly having no real engagement with the project. That's exactly the mentality we don't need in an administrator and I support anything aimed at curbing it. Beeblebrox (talk) 19:56, 26 August 2019 (UTC)
  57. Support. Long overdue, in fact, and that is probably because no one likes specific proposals, just as the OP (TonyBallioni) says above. This way of setting up the RfC will hopefully overcome that problem, and I want to compliment Tony for coming up with it. Bishonen | talk 20:09, 26 August 2019 (UTC).
  58. I basically agree with Risker (who favors Option 1) and am likewise largely unconvinced that the resysop requirements... are at issue; it is indeed mostly the retention requirements that are problematic. Still, I don't think it'd be bad to tighten them up a bit (especially as concerns longer-term absences) so I'm here weakly. ~ Amory (utc) 11:51, 2 September 2019 (UTC)
  59. Support. Since this is still open, I might as well add my support for stricter resysop requirements, for multiple reasons stated multiple times above. (I also support stricter desysop rules as well, but one thing at a time.) --RL0919 (talk) 23:04, 6 September 2019 (UTC)

Option 3Edit

  1. Getting the mop is supposed to be no big deal, with the same level of power and prestige as being a janitor or garbage collector. It should be available to anyone that wouldn't misuse it. Of course, if returning admins were forced to work on all admin-related roles without being allowed to re-read recent policy first, something could then go wrong. But (unless there's some recent policy I'm missing) noone is (yet?) forced to work in any specific areas they don't know about. I don't really think there should be any limit to how long someone can be inactive and still get the mop back.
    However, as I see it, the main problem is that RFAs have ballooned from a sanity check that the user warrants basic trust to a marathon of checking that the user fulfills 101 mutually exclusive subjective criteria, has refrained from contributing to any controvertial topics and can recite all areas of Wikipedia policy in Latin by heart. Even if all old mop-holders got the mop back, there still wouldn't be enough mop-holders, with almost no way of squeezing new ones through RFAs, let alone getting them to even apply.
    A random idea (which unfortunately probably won't get implemented) would be to add a "This person should be an admin" button next to the "Thank" button — and if some threshold of people press it, the person is notified that they should perhaps apply (with a note that if they're not interested in power or prestige, they should definitely apply). Or actually, the "Thank" button itself could do that — no need for an extra button. Κσυπ Cyp   10:05, 20 July 2019 (UTC)
    Says the admin whose last 50 non-minor mainspace edits go back to 2008, and hasn't made a talkpage comment of any kind since 2017… ‑ Iridescent 13:21, 20 July 2019 (UTC)
    Except for the Wikipedia talk and User talk kinds of talkpage. Κσυπ Cyp   14:08, 20 July 2019 (UTC)
    Support Option 1 now (unless Option 3 comes back), since Option 3 turned into a comment section. Option 3 came back. Κσυπ Cyp   11:49, 20 July 2019 (UTC)
    @Cyp: I do think yours was an interesting case. I recently read about it, and if I recall correctly, there was a great deal of pushback. Are you of the opinion that if the activity policy were stricter that you would not be an admin right now? –MJLTalk 00:17, 21 July 2019 (UTC)
    @MJL: Yes, if it were stricter by any non-trivial amount, then I probably wouldn't be. I was astonished at the amount of discussion, given that it was supposed to be possible to come back after any amount of time. I'm not sure what happened. Maybe some people felt RFA was a bit of a deal, and gave some people had slightly hard RFAs, leading them to find it a bit bigger deal, making the next RFAs even scarier, eventually leading from NOBIGDEAL to REALLYHUGEDEAL, and then mainly just people who like really huge deals applying, which might not be the only set of people suitable for adminship. Κσυπ Cyp   05:13, 21 July 2019 (UTC)
    @Cyp: [Thank you for the ping] Well, you are an admin, so why not go nominate some experienced users for RFA? If RFAs weren't so rare, then you'd see that attitudes on adminship change imo. ¯\_(ツ)_/¯MJLTalk 05:53, 21 July 2019 (UTC)
    Ladies and gentlemen, an example of the problem. ~Swarm~ {sting} 04:20, 22 July 2019 (UTC)
    Indeed, good example. For those in the Option 1 group who have troubles seeing the problem, would you please summarize a few of the problems that this particular example has caused? Κσυπ Cyp   08:03, 22 July 2019 (UTC)
    @Swarm: Ah, but what sort of adverse effects have come about as a result of resysopping him? I've not seen any controversial actions on Cyp's part since he came back. Kurtis (talk) 07:51, 23 July 2019 (UTC)
    The fact that a very small number of users continue to be "relic administrators" with no community mandate and no notable involvement with the project, by virtue of meeting extremely low activity requirements, as opposed to actually contributing to the project, even though we've had fairly consistent and high standards for adminship for about 15 years, is the problem. How many untold thousands of hours have people contributed to this project, only for it to not be good enough at RfA? We have standards, users like Cyp do not meet those standards. Like I said above, it's not a personal attack, it's a statement of fact. It's a problem in itself, Cyp, without making any commentary about your integrity, and you should be able to understand that. But the fact that you're condescendingly asking for examples of a problem, which I can find in 2 seconds of looking at your logs, just shows how hilariously out of touch you actually are. Your block log is fucking atrocious. Apparently you never got the memo that we notify users we block? Hard blocking users with no edits? Revoking email and tpa without cause? Issuing blocks without log entries? Blocking users who have not been sufficiently warned? It's literally deja vu from the recent Enigmaman desysop, but at least he was an active admin who seems to have just stopped caring over the years. You're an actual admin with no RfA, requested the mop because you were "curious", have no notable activity since 2004, have less than 4k edits, you clearly are painfully out of touch with our norms and standards, and on top of that you're so arrogant that you don't even know you're out of touch, and deign to dare others to find faults with your adminship, even though such faults are consistent and obvious. You'd think an admin who never actually had to be vetted by the community would be a little more humble, but I guess your attitude is just another thing we can chalk up as a problem that would have been screened out by any modern RfA. I'll put to you a simple proposal: if you're really so confident that you're a good admin who deserves the mop, just run an RfA. If you don't want the community to actually vet you, as 100% of us have been vetted for the past 15 years, then you shouldn't be an admin. ~Swarm~ {sting} 04:56, 26 July 2019 (UTC)
    Yes, looking at my block logs, I do see some examples of that, for example, I blocked Wikipedians have a below average penjs size and If you community ban me i will cruise more without TPA despite them not having edited, without any warning or notifying them, and without even giving a reason in the block log. Unfortunately, it was some time ago, and I since I didn't write a reason in the block log, I don't remember anymore why I blocked them. And I'm afraid there are many other comparable examples. This doesn't apply to all my blocks, at least. I have though given warnings and/or block notices for some users, especially for the users who could plausibly both be human and not already have accumulated 400+ warnings and 100+ block notices. I don't think I've yet gotten any complaints about specific actions, except for one about having deleted a page as non-notable instead of as vandalism (or something like that), and some complaints about having forgotten to revoke TPA when blocking an LTA. I certainly may be out of touch, and tend to stay away from what I feel most out of touch with. As for RFAs, I certainly see a difference between how RFAs were back then, compared to now. Rather than me running a reRFA by recent standards, I think a better solution would be adjusting new RFAs so they match the older standards better. I get the impression that the RFA process now screens out most non-extroverts before they even get to the stage of applying, and that might not neccessarily be ideal. Κσυπ Cyp   11:56, 26 July 2019 (UTC)
    With regards to those two editors: though I cannot discern what you were thinking at the time with any more certainty than you can, I would hazard a guess that it had something to do with their usernames... Kurtis (talk) 12:14, 26 July 2019 (UTC)
    Your bad blocking practices are a consistent and unwavering pattern, not "two users a long time ago". The fact that you can breach norms and standard practices and engage in abusive practices just because no one screening your logs, and then blame your incompetence on the fact that no one was supervising you, is a sign that you should not be an admin. No one should have to supervise an admin on the assumption that they don't know what they're doing. We simply shouldn't have to be worried about this. That's why we have RfA, and that's why admins without RfAs should subject themselves to the community for vetting. Regarding your position on RfA, there's another problem: what you describe as "older standards" really means "no standards", and "recent standards" is actually just "standards", standards that have been fairly stable for nearly 15 years. If you think it's realistic to call for the community to drop their longstanding standards for adminship, and run a NOBIGDEAL RfA on your own merits, then go for it. Put that theory to the test. I will argue that your logs are evidence that there's a reason we should have standards, but it is the community that can weigh our opposing arguments. The community governs this project, not your opinion that you deserve to be an admin because there should be no standards for adminship. It's a pretty bad argument, IMO, especially coupled with actual incompetence. ~Swarm~ {sting} 03:58, 28 July 2019 (UTC)
    (edit conflict) @Swarm: I stand corrected. To tell you the honest truth, I based my comment more on the fact that I haven't seen any drama arising from his resumed adminship. Despite my sporadic activity, I do keep myself updated on Wikipedia's latest developments, though it's possible I missed something. If what you say is true – chiefly, his heavy-handed approach to blocking – then there is a serious question as to whether Cyp is currently fit to remain an administrator. I've seen problems with what you describe as "relic administrators" before, and I think that they should have their tools revoked if they're repeatedly misusing them. On the other hand, there are plenty of other relic administrators who continue to display the competence necessary to at least get themselves up to speed before taking action. I believe it is best handled on a case-by-case basis; administrators who are demonstrably out of touch should lose their sysop tools, whereas administrators who have not caused any issues thus far should be allowed to keep them. As far as potential future problems are concerned, we can cross those bridges when we get to them. Kurtis (talk) 12:06, 26 July 2019 (UTC)
    May I please add that it is very important that administrators who are not using the tools correctly get proper feedback about it in a timely fashion, otherwise they end up in a situation which we recently have seen with Enigmaman.--Ymblanter (talk) 14:29, 26 July 2019 (UTC)
    Agreed. I was under the impression that after my re-mopping, that there were going to be 100 people carefully going through each action I do, and immediately lynching and/or giving proper feedback the second I did anything at all questionable. And that no feedback was therefore positive feedback. Κσυπ Cyp   17:30, 26 July 2019 (UTC)
    I agree wholeheartedly. It was not my intention to imply that misuse should result in a quick desysop. A demonstrated inability to learn is more what I had in mind. Kurtis (talk) 22:18, 26 July 2019 (UTC)
    Uh, I don't know what the hell you're talking about, but Enigma was not desysopped because he made good faith mistakes that no one corrected him on, he was literally desysopped after abusively blocking an inactive account that had insulted him a decade prior. He knew it was abusive, he just did not think that he would be caught, and he only was caught because that user was still around under a different account. I subsequently investigated his edits and found that there was an extensive pattern of willfully abusive conduct. It's a bit bizarre that you're blaming "lack of feedback" as being responsible for straightforward admin abuse. ~Swarm~ {sting} 04:08, 28 July 2019 (UTC)
    Without speculating on whether he did or did not think he would be caught, if this block were an isolated accident it would probably not result in a desysop. The thing was that when this block was uncovered at ANI, users started to look at his administrative actions and found a number of actions which were really not ok but nobody discovered them before. Assuming good faith (and the ArbCom proceedings did not show any evidence which would push me in this case to stop assuming good faith), it was just a behavioral creep - he started with something which was marginally ok, and then did something which was not almost ok but not exactly ok, and assumed it is ok, and then did something which we would think clearly not ok, and so on. If anybody would talk to him in the middle of the process and tell him that what he was doing was not ok he would likely stop doing it. We all know that there are many greyish areas in the admin activity (what is involved? what is pure vandalism listed as an exception for 3RR? etc) which can lead to the behavioral creep pushing an admin into a zone where the actions are clearly not ok. Feedback in the middle of this creep would certainly help.--Ymblanter (talk) 11:27, 28 July 2019 (UTC)
    I disagree with your first point. It was a willfully abusive block, out of spite. There was no "oops, here's my excuse", nor was there an "I fucked up, I'm sorry, it won't happen again". I can think of no more egregious act of abuse that would result in a desysopping if I tried. I actually think it's a bit bizarre that you'd even suggest that we wouldn't desysop admins for malicious revenge blocks just because it was a first offense. Your argument strikes me as akin to someone being convicted of murder, and saying "if only someone had warned him against being a jerk". ~Swarm~ {sting} 02:10, 4 August 2019 (UTC)
    I unblocked the account myself and apologized. There's nothing else I can do. I would also note that while it was a bad block, no one was harmed because of it. Enigmamsg 23:25, 24 August 2019 (UTC)
  2. Inactivity isn't usually a bid deal unless there's concerns about the user's conduct which is far more likely to be an issue with an active admin than an inactive one. Crouch, Swale (talk) 21:23, 3 August 2019 (UTC)


A few points:

  • Performing lots of sysop actions does not necessarily make one competent to be a sysop. If a person is perceptive, logical and cautious they will do a good job now or five years from now. Good judgement means they will look around and familiarize themselves with community norms when they need to.
  • Presumably people were made sysops because they have enough clue to read the manual when they need to.
  • It's often not evident if somebody has been lurking or IP editing or working on a sister project. There are also sysop actions that aren't logged, such as reviewing deleted diffs. Edit counts and log counts aren't everything.

Before I'd support any tightening of the criteria, I'd like to see evidence that the lenient criteria currently in effect was causing a problem. To my mind we need to avoid solving non-existent problems. Jehochman Talk 03:40, 20 July 2019 (UTC)

Jehochman, you used the example of El C, who is one of my favourite colleagues. He has not been desysoped for inactivity in 6 years, and there is nothing here that would suggest he should have been desysoped sooner. In fact, this RfC has nothing to do with the deysop criteria. This is only about whether or not our current 3/5 resysop criteria (3 years inactive/5 years logged actions when desysoped) should be made stricter. Nothing more. There are not even specific proposals here, just the question of if what the community thinks of the current resysop policy. TonyBallioni (talk) 03:48, 20 July 2019 (UTC)
Continuing with that example, he logged in once in a while and avoid being desysoped. Had he not done that, I'd still want to be sure we would have welcomed him back without a new RfA. Jehochman Talk 03:52, 20 July 2019 (UTC)
I understand it's not nice to talk about people. Can you indicate, perhaps anonymously, whether we are resysoping people who proceed to cause disruptions? Lately it seems like that role has been taken on by WMF. Jehochman Talk 03:54, 20 July 2019 (UTC)
  • As I said above, there really need to be continuing education requirements for adminship in general, and I think this is the perfect vehicle to start requiring it as a condition of reentry. For those unfamiliar, continuing education is a requirement in many if not most licensed professions. This not only includes doctors and lawyers, but also barbers and interior designers. The way this is structured varies a lot between professions. For lawyers, you usually have to take courses or lectures, but you can also get credits for teaching courses or writing articles. For doctors, a portion is satisfied just by doing research of some kind on patients' cases, but you also need to submit to some form of accredited continuing education (which may include providing such education). And, commonly, when someone has been inactive or is disciplined, the licensing authority may require some form of continuing education as a condition for reentry. I have long thought that administrators on Wikipedia should be brought up to speed periodically on what it means to be an admin, on changing professionalism requirements, etc. For instance, the case that was discussed on WP:BN that led to this RfC was for an admin who passed RfA before we even had rules against paid editing.
    I admit that there needs to be continuing education content for this sort of thing to work, and at present there is none. That said, I think there's a way to achieve this. Like I said, if this can be implemented fully for all admins, we'll have our active admins creating content for new and returning admins. Similarly, we can work with the WMF and chapters to develop training materials. Imagine the following lectures: "Identifying and Responding to Incivility", "Calculating Rangeblocks for IPv6 Addresses", "The RfA Process in the 2010s", "Speedy Deletion in the 2010s", "Arbitration Enforcement: Practice and Procedure", "Determining Consensus and Closing Discussions at Noticeboards", "Revision Deletion", "Copyrighted Text Detection and Deletion", "Fair Use on Wikipedia", "Writing Proper Block Log Entries", "Including Permalinks in Edit Summaries and Log Entries". These are all things admins should know about and should keep current on even if their areas of interest aren't all-encompassing.
    Anyway, my point is that this is something we should be moving towards in the next couple of years, and the problem of low-activity admins is as good an opportunity as any to start addressing it. —/Mendaliv//Δ's/ 05:17, 20 July 2019 (UTC)
    I like the idea, and would like to see such training being available for all Wikipedians. There could be a suggestion of which would be most useful for particular applications, and requiring a disruptive editor to participate and pass a relevant course could help with some of the recurring dramaboard participants. If you go further developing this idea, please ping me as I would like to contribute where I can. I have a little experience with Moodle, which could be a suitable platform. · · · Peter Southwood (talk): 07:28, 20 July 2019 (UTC)
    Yep, that's a very good point, that continuing education could be used as an alternative to bans, and could also provide a way for sanctioned editors to demonstrate that they're ready to move on from their sanctions. I can think of a case that controversially closed on AN just the other day that would probably have been avoided if there were materials available on things like "avoiding close paraphrasing" and similar. —/Mendaliv//Δ's/ 07:59, 20 July 2019 (UTC)
    Sufficient potential for solving long standing problems that it is worth going further. One way to start is to list the skills that should or could be trained, then for each one list the questions that should be answered correctly to indicate understanding of the policy. Then write a set of multiple choice questions with a good set of correct answers and incorrect answers, complete with instructive feedback where possible for incorrect responses. Seting up the assessments on Moodle is the easy part. · · · Peter Southwood (talk): 10:47, 20 July 2019 (UTC)
    It could also help with RfAs as a person who had done all the training could reasonably be assumed to know most of the important stuff. · · · Peter Southwood (talk): 14:45, 20 July 2019 (UTC)
    Continuing education is really hard work. I know of at least one effort to track the evolution of guideline and policy that ended because the user doing the work came to the (correct?) conclusion that it wasn't worth it (that was Tony1 and the MOS). Now, it's possible that the guidelines in question were under much more-rapid evolution at the time than those we have today, and maybe e.g. the admin newsletter will help give us a monthly overview of deltas, but even then someone needs to put the training together, etc. And they will need to do it over both a longer timeframe (years) while simultaneously needing to support different lengths of inactivity (2012 to 2016, or 2007 to 2017, etc.). None of this leads to much in the way of productivity (which might generously include keeping admins up on the Joneses). I'm not trying to say this is impossible (I'd love to do a year over year diff against each of our PAGs, as a research project I know I don't have time for), but the person (or people) who contribute into this will have quite a bit of work cut out for them, and we all know this is a volunteer project with enough backlogs as it stands.... --Izno (talk) 06:07, 26 July 2019 (UTC)
  • Iridescent, I don't want this to get too into specific proposals since I think that's part of the reason we haven't been able to solve this yet, but yes, even something as simple as before restoring the administrator flag a bureaucrat should be reasonably convinced that the user intends to return to activity would be an improvement, and a line like that would not preclude someone from coming back, getting turned down, editing for 3 months and showing they aren't an idiot, and requesting at BN again without an RfA. Something like that likely wouldn't be as strict as some want, but it would certainly be an improvement from what we have now. TonyBallioni (talk) 06:32, 20 July 2019 (UTC)
@TonyBallioni: "One of [your] favorite colleagues" has just all but ceased communication with anonymous contributors...indefinitely. ——SerialNumber54129 07:18, 20 July 2019 (UTC)
Indefinite doesn't mean infinite. It just means I'm not telling the abuser the exact length of the protection. El_C 07:31, 20 July 2019 (UTC)
In a world where "persistent sockpuppetry" means one edit, yeah. ——SerialNumber54129 08:24, 20 July 2019 (UTC)
"B.S." is not a particularly civil edit summary, nor does it assume good faith. Any reason for that? Anyway, that sock had iterations in the double digits today: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14. El_C 08:33, 20 July 2019 (UTC)
Apologies, El C, I should've found a more polite expression to convey the meaning! ——SerialNumber54129 08:48, 20 July 2019 (UTC)
  • Serial Number 54129, would that not require an RFC0 to establish? The criteria to decide who gets to vote in these RFCs, I mean. Usedtobecool ✉️  09:04, 20 July 2019 (UTC)
  • While I don't think there's much point expelling old admins, I think the main problem is getting new ones. One random idea is to piggy-back on the Thank button — if someone is getting a decent quota of Thanks from enough different users, then they could perhaps be encouraged to apply, especially if they're the type of person that wouldn't normally feel qualified for adminship. (Summary of comment in option-3 vote.) Κσυπ Cyp   12:27, 20 July 2019 (UTC)
That's an excellent idea Cyp, although it will probably lead someone to demand a concomitant "Do not adminate" button! :) glad you got your posts all sorted out now, sorry for any confusion. ——SerialNumber54129 12:30, 20 July 2019 (UTC)
That's fine, I get confused easily, anyway. Κσυπ Cyp   18:13, 20 July 2019 (UTC)
  • Sorry if this is a bit confusing — my viewpoint on a lot of the discussions on this page related to these 3 items, which are problems if true:
A) In the early days, getting the admin bit was too hard. (currently false)
B) The difficulty of getting the admin bit in the early days is not the same as getting the admin bit now. Or A ≠ C. (currently true)
C) Getting the admin bit is too hard now. (currently true)
Current situation is ¬A ∧ B ∧ C, and the goal should be ¬A ∧ ¬B ∧ ¬C. Commonly proposed solution: Fix B by retroactively making it harder to get the admin bit in the early days (with the side effect of retroactively making A true), and worry about fixing C later, resulting in A ∧ ¬B ∧ C. What I'd prefer: Fix C (with the side effect of fixing B, and letting A remain in the fixed state), resulting in ¬A ∧ ¬B ∧ ¬C. Κσυπ Cyp   18:13, 20 July 2019 (UTC)
Have you any actual evidence for Getting the admin bit is too hard now, or are you just making shit up to try to support your case? The two RFAs this month got 231 supports and 167 supports, a level of support which back in the old days was so unusual we had a special page to list the occasions when it occurred. The "RFA is impossible to pass nowadays" thing is something a lot of people like to say, but they tend to fall silent when anyone suggests they might want to provide some evidence. ‑ Iridescent 20:41, 20 July 2019 (UTC)
It's my impression that almost noone applies. At least, 2 in a month sounds low to me, for a site that's now this size. I'd guess there are lots of qualified people who wouldn't dare apply (and may or may not pass if they applied anyway), but I don't really have any specific evidence for it myself, sorry. Κσυπ Cyp   22:01, 20 July 2019 (UTC)
Looking at this, it seems as if the mean number of successful RfAs per month has been fairly flat since 2012. Certainly the current situation is nothing like the heady days of 2003 to 2009, but one does have to realize that since adminship is a more-or-less permanent appointment, all of those hundreds of newly-minted admins from that period -- less those who have retired or been desysopped for various reasons -- remain on duty. It would be unreasonable to expect that level of successful RfAs to have continued forever: more editors means more people interested in who is and isn't an admin, means fewer admins getting the bit with minimal input from the community-at-large. Some further statistical investigation would be nice, such as how many participants there were in each RfA (successful and unsuccessful) over the years, compared to how many active Wikipedia editors there were at the time; how the margin of success and failure has changed, etc. In short, I don't think that the evidence is there yet to make such broad generalizations, beyond saying that (obviously) fewer admins are minted these days than 10-15 years ago. Beyond My Ken (talk) 23:51, 20 July 2019 (UTC)
  • To be quite frank, I doubt any of this is going to have much impact on the project. Short of a complete overhaul of the whole add subtract abilities I don't think adding a month or year to various wording is going change much. Those that want to disrupt will find a way. Those that accidentally disrupt will quickly be brought into some acceptable situation. (we do have provisions for emergency desysop.) I suspect that you're going to find more change coming from whatever the WMF/T&S have in mind moving toward 2030, how Arbcom are going to liaison, and how much the community will change due to different circumstances in the future. I just can't imagine that anything here is going to have any substantial affect to what we have presently. — Ched :  ?  — 02:46, 21 July 2019 (UTC)
    To be quite frank, I doubt any of this is going to have much impact on the project. I agree. It will probably be an uncommon situation, but in my view that's actually a factor in favor of implementing some change as a prelude to an overall revision to the adminship process. As I've discussed above, the idea of continuing education as a general requirement (that is, not merely as a special requirement for returning admins) has a lot of merit. If such a system is implemented, it should lead to a softening of the unreasonable standards at RfA. I suspect that you're going to find more change coming from whatever the WMF/T&S have in mind moving toward 2030, how Arbcom are going to liaison, and how much the community will change due to different circumstances in the future. Again, when seeking to make changes and improvements, especially to core community processes, we should seek to make incremental changes, rather than revolutionary changes. There is not only nothing inherently wrong with baby steps, but I would argue that such steps should be preferred to sweeping changes. —/Mendaliv//Δ's/ 12:02, 21 July 2019 (UTC)
    With baby steps there is less chance to break things, and you can change direction in mid stride if it becomes necessary. · · · Peter Southwood (talk): 15:45, 23 July 2019 (UTC)
Probably a bad idea, never mind.
The following discussion has been closed. Please do not modify it.

Another random wild idea for trying to get more new admins through RFA — give anyone who could ever plausibly become admin someday the apprentice bit, which should let them perform admin actions which are carried out as soon as being reviewed by an admin (or not carried out, if disapproved). If someone does too many useful admin actions with no recent serious mistakes and the admins get tired of reviewing everything, change their apprentice bit to an admin bit or send them to a RFA where people can already see the track record of a candidate admin as an admin. This of course has the downside that it needs to be coded, although even an incomplete implementation (allowing some admin activities but not others) might be better than nothing. Κσυπ Cyp   18:13, 20 July 2019 (UTC)

  • Wikipedia:Perennial proposals#Hierarchical structures. Regards SoWhy 20:22, 20 July 2019 (UTC)
    Also, Legal requires anyone getting access to the admin tools in any way to have gone through some kind of RFA-like process (cf. WP:PEREN#Automatically grant adminship to users with a certain number of edits or time editing) and if we had to choose "apprentices" through such a process, we can just make them full admins as well and save us the bother. Regards SoWhy 20:25, 20 July 2019 (UTC)
    What I mean is slightly different, I think. The "apprentices" should have a safe toy version of the complete admin tools (except Legal stuff like viewing deleted revisions, I guess), except any use of them would require case-by-case approval. I think this is different than a hierarchical structure, since it's the whole set of admin tools (once the implementation is complete). Since they wouldn't actually be able to do anything at all with the admin tools without each action being approved by an actual admin, I'd expect that to pass by Legal. And an apprentice having a track record of proposing sensible admin actions via their toy admin interface might then pass though a RFA easier without having to explain hypothetical examples of how they might be as an admin, since everyone can see non-hypothetical examples in their log. Basically, almost anyone should be given the "apprentice" bit, which would let them use-by-proxy the admin tools without having the admin tools, and (hopefully) make it easier to get the real admin tools later. Κσυπ Cyp   21:06, 20 July 2019 (UTC)
    I mean, that kinda just sounds like building a "Thank"-like tool that you use instead of tagging pages for CSD/AfD, or reporting users to AIV, or doing NACs of discussions. I think it just kind of kicks the problem down the road. —/Mendaliv//Δ's/ 21:11, 20 July 2019 (UTC)
    Cyp, there isn't any point in pretend-blocking a vandal or pretend-deleting an attack page. Most other admin actions are slightly less time-critical, but can't stand on their own: the admin action is usually accompanied by some edits. You can't wait for approval of the admin action without time-delaying the accompanying edits as well, which isn't going to work for blocks and posting on ANI. If you want, we already have "deletion apprentices" in the form of WP:NAC of debates, some of them even leading to speedy deletions based on their NACs. —Kusma (t·c) 21:15, 20 July 2019 (UTC)
    Pretend-blocking a vandal might get them real-blocked sooner, if admins are looking at a list of actions to approve. But I can't think of a solution to actions that don't stand on their own, so I guess my idea wouldn't really work well for many (most?) things. Oh well, never mind. Κσυπ Cyp   22:01, 20 July 2019 (UTC)
  • Just throwing out a few thoughts.
    I might be what some call a "legacy admin" -- I got the bit back in September 2003 thru one of the first RfA votes, & all it took was 3 yes votes to do it. And I've rarely pushed the mop, so many might think I'm not qualified to be one now. Yet I've followed most of the conflicts & developments on en.wikipedia over the years, so if I had the chance to be more involved I could easy ramp up & assume more of the burden. (Work, commuting, & the demands of a family are the reasons I haven't been very active over the last 6-8 months. And the more recent Foundation misdeeds are not making me eager to try to squeeze out time for Wikipedia as I have done in the past.) On the other hand, an admin who lost the bit due to inactivity, & who moreover was not that active to begin with, is someone who never had an idea what the job & duties truly are. Are we that desperate for janitors to accept people who either aren't going to do the work, or not do the work properly? -- llywrch (talk) 07:17, 21 July 2019 (UTC)
  • If we want to change sysop activity criteria, they should apply to "active" and inactive admins alike. A sysop who doesn't do anything except sending a "happy birthday" post to another user once per year for five years is no more likely to be engaged with the community than a user who has been away for three years, possibly even less. If we want to make sure make sure someone is engaged with the community and keeping up with what is going on we should expect "regular editing" for at least a couple of weeks per year, but I don't have a good way to measure that. Anyway, if we want to do something about the regular angst when a long-term inactive admin returns, finds the message on their talk page that says they can go to WP:BN and request resysop, then asks for resysop, my suggestion would be to either encourage people to edit for a few weeks before requesting resysop or to require people to edit for a few weeks before they are allowed to request resysop. All that is now needed is a better way of measuring "regular editing for a couple of weeks" than "I know it when I see it". Or we need to decline to measure it and leave it up to a bureaucrat vote. (That might work better, actually). —Kusma (t·c) 18:37, 21 July 2019 (UTC)
That's literally the whole damn point. One edit per year should not make you "active". So, why does it? ~Swarm~ {sting} 04:32, 22 July 2019 (UTC)
A not unreasonable opinion, so what should make one "active"? · · · Peter Southwood (talk): 05:44, 22 July 2019 (UTC)
Many projecys require more than one edit per year for an admin to keep the flag. Out of my memory, Wikidata requires five actions in 180 days, Commons as well. When I was still active in the Russian Wikipedia, the was a requirement smth like 30 admin actions and 30 edits per year, otherwise the desysop case goes to ArbCom. None of these projects have resysopping without RfA.--Ymblanter (talk) 19:36, 22 July 2019 (UTC)
Swarm, that is a damn good point, but not the point of this RfC. —Kusma (t·c) 12:13, 23 July 2019 (UTC)
It does seem relevant to the problem this RfC is intended to address. · · · Peter Southwood (talk): 15:38, 23 July 2019 (UTC)
  • Perhaps stricter and looser are not the best choice of words here. There seems a fair amount of support for having more meaningful resysop criteria, but whether they would usefully be described as stricter is not clear and may be putting people at cross purposes. I don't think there is a magic bullet solution, but there may be some common ground in resysop criteria, RfA problems, The T&S fiasco and the problems of maintaining quality while providing an acceptable working environment. · · · Peter Southwood (talk): 07:54, 22 July 2019 (UTC)
  • I think you might be missing a key criterion. As long as the criteria for retention of sysop permissions are so low, it really doesn't matter what the resysop criteria are. In fact, what seems to happen is that there is a big kerfuffle periodically at a resysop request, usually with whining that so-and-so shouldn't be resysopped because they haven't been active enough....and then...what? Do we have any evidence that resysopped admins are performing admin tasks inappropriately? That they're showing evidence of taking action based on antiquated understanding of policy? Frankly, our problems are more often seen in administrators who have never been desysopped. I understand where this RFC is coming from, but it's missing the point. You don't have to worry too much about resysop requirements if you have adequate retention requirements. Risker (talk) 00:18, 23 July 2019 (UTC)
    +1. See my discussion above with Fish and Karate about Andrevan and Rama. — JFG talk 15:31, 23 July 2019 (UTC)
    I agree it makes more sense to keep the criteria for regaining administrative privileges in synch with the criteria for retaining administrative privileges, and if desired, tighten both together. isaacl (talk) 15:34, 23 July 2019 (UTC)
    @Risker, there are problems. We had a recent arbitration case where an inactive admin showed up and unilaterally undeleted a page. We had a less-recent arbitration case where an largely inactive administrator deliberately pushed the limits of appropriate engagement during an arbcom election to make a point. So far, the statistics show that around 15% of all administrators ultimately end up resigning under a cloud or losing their tools involuntarily. I have gone through the contribution history of some long-term marginally active administrators and have found some really questionable actions, admins using undeletion and page protection to defend articles they wrote and so on. The psychosocial problem is that these individuals are no longer afraid of the community and are willing to "cash in" their adminship to prove a point or push their POV. I have tried a strategy of engaging one-on-one with some of these admins, and except in rare (5%) cases they don't respond in any way to a friendly note left on their talk page and emailed. They're disengaged, and have moved on, but still want their bit for the trophy case. UninvitedCompany 18:14, 23 July 2019 (UTC)
    UninvitedCompany, neither of the administrators to whom you refer above were desysopped admins who requested readminship. Both were semi- or fully active editors (also a low standard) and had not been desysopped for inactivity prior to their encounters with Arbcom. This RFC is about standards for re-sysopping, and would not be applicable in any case to either of those administrators. They were never resysopped so any standard that is developed for resysopping would not apply to them. I believe what you want (and in fact what many people are mis-using this RFC to discuss) is to change the standard for retention of sysop permissions. I'm going to be undiplomatic here and note that you yourself barely remained active enough to retain your own sysop and 'crat bits for many years, and it's ironic that now that you've returned, you have repeatedly denigrated the low retention standards that allowed you to keep your bits through all those years. If people want an RFC on retention of bits, then we should have that RFC. This is not that RFC. Risker (talk) 19:56, 23 July 2019 (UTC)
     ::shrug:: I'm back, I try to contribute, others are also. It would not have been the end of the world if I had gone through a sensible level of review before returning to active administrative work. We have this automatic re-adminship policy because it was what we had to adopt to get consensus on the inactivity policy back in the day. I've looked through the history of people who have asked for their bit back and with a few rare exceptions they don't contribute. At a minimum I'd like to see us require a return to active editing and ongoing participation for a month or two before we return the tools. And yes, I believe that the overall handling of inactive contributors should be tightened, including the re-adminship portion of the policy we're discussing now. UninvitedCompany 21:26, 23 July 2019 (UTC)
    UninvitedCompany, you say the statistics show that around 15% of all administrators ultimately end up resigning under a cloud or losing their tools involuntarily. What types of "losing their tools involuntarily" are you including here? On what basis have you arrived at that statistic? I'd like to see some details here, because that's a pretty audacious statement to make. I think you're also applying some pop psychology to adminship that isn't really validated by statistical information, either. Do you have anything to illustrate your statements? I really have a problem with people throwing around statistics without linking to their data and analysis. Risker (talk) 06:02, 26 July 2019 (UTC)
  • While this says "for inactivity" - we have basically the same situation for "resign then become virtual inactive" that shouldn't have a "loophole" either if this goes forward - can be explored if there is a part 2. — xaosflux Talk 20:32, 23 July 2019 (UTC)
  • A pointer to the repository of continuing education for Admins might be helpful to some legacy Admins. Otherwise someone might give it a shot, then be told they missed the IMPORTANT new things Admins should know. Edison (talk) 21:48, 7 August 2019 (UTC)
  • Note I have requested closure it is over 32 days now.Pharaoh of the Wizards (talk) 10:16, 21 August 2019 (UTC)

Option 4?Edit

Per Boing! "Let's stick to addressing resysop requirements, and save inactive desysop requirements for a separate discussion at another time". ——SerialNumber54129 08:03, 25 July 2019 (UTC)
The following discussion has been closed. Please do not modify it.

It seems to me that a lot of the discussion above is not about "the criteria you need to fulfil when you request resysop should be stricter" but about "the criteria for when you are classified as an active admin who retains their bit should be stricter". There is evidence of tool misuse among people who have never been desysopped, but not so much among people who have been resysopped under our lax criteria. Should we add something like the following? —Kusma (t·c) 21:17, 23 July 2019 (UTC)

  • OPTION 4. The criteria for retaining adminship should be made stricter (currently, one edit or log entry per year is sufficient to retain the bit indefinitely).
  1. Support. This seems very reasonable to me. --Tryptofish (talk) 21:56, 23 July 2019 (UTC)
  2. Support in addition to option 2. —/Mendaliv//Δ's/ 21:59, 23 July 2019 (UTC)
  3. Support alongside option 2. Anarchyte (talk | work) 04:09, 24 July 2019 (UTC)
  4. Oppose at this point. One problem with addressing things like this is that we get multiple proposals all suggesting different things (and I'm sure that's why the original proposal is crafted the way it is), and that pulls us in all directions and we never get a consensus for anything. I think we need to focus and try to achieve one thing rather than diversify and possibly achieve nothing. Please, let's stick to addressing resysop requirements, and save inactive desysop requirements for a separate discussion at another time. Boing! said Zebedee (talk) 08:02, 24 July 2019 (UTC)
    Boing! said Zebedee, the whole point of potentially having this as a separate option is that all of the examples in the discussion above are about people who should have been desysopped, not about people who should not have been resysopped, so the discussion is already happening. Do you have a better suggestion how to focus discussion above on resysop criteria? —Kusma (t·c) 09:24, 24 July 2019 (UTC)
    Yes, I think we should just focus on choosing option 1, 2 or 3 and resolve resysop requirements first, because that's the original proposal and is what the consensus was intended to be judged on. Sure, the discussion section is covering inactive desysop requirements too, and that's fine - but it doesn't mean we need to resolve that in the same RfC right now. A new RfC for inactive desysop requirements can come later, as an offshoot from the current discussion. My real fear is based on my experience of seeing discussions like this expanding in scope, time and time again, to try to fix everything at once rather than picking off one specific thing at a time. (And please be assured that I would support tightening of inactive desysop requirements too, but as a separate RfC later.) Boing! said Zebedee (talk) 09:56, 24 July 2019 (UTC)
    Boing! said Zebedee, as I am one of the people who don't see the problem (I see a perceived problem and the controversy it generates, but I have been unable to identify any problems with the actual outcome of the current policy), I don't think resolving resysop requirements is something that should be done independently from the desysopping/inactivity criteria. (You may change my mind if you can explain to me what the problem is). —Kusma (t·c) 10:13, 24 July 2019 (UTC)
    Well, I'm not actually trying to change your mind ;-) I just think we'd have better chance of success addressing one thing at a time, while you think the two aspects should be addressed together. Such is the way consensus is built. Boing! said Zebedee (talk) 12:09, 24 July 2019 (UTC)
  • Support (along with option 2) I agree with Barkeep49 below that this is really a different question. Schazjmd (talk) 14:45, 24 July 2019 (UTC)
  • Oppose for now per Boing! This is a different discussion. Most of those who know me and have followed the RfCs I have started in the past know that it is my belief that RfCs are only good for documenting pre-existing consensus, and that otherwise they are set up to fail miserably. I do think there is a pre-existing consensus that the resysop criteria are a joke, and even more of a joke than the desysop criteria.
    While this option is getting support now, it also isn't the focus of the RfC, and is a subheader at the bottom that is likely viewed as a side-issue by most participants. If we had a full RfC on this question, I don't think it would achieve consensus (heck, we basically had this a few months ago.) We should focus on the question at hand, and document consensus on that question. If there exists consensus that the resysop criteria should change, then we can go about changing them, and after we set them, we can take another look at the resysop criteria. One step at a time is a good model for change, and this is putting the more difficult policy step to take first. Let's focus on the question at hand. TonyBallioni (talk) 23:07, 24 July 2019 (UTC)

Option 4 DiscussionEdit

While I support this option this feels related and yet distinct in that both this and an option 1-3 could gain consensus. Given the number of editors who had already participated above, and might not revisit this RfC, perhaps it should be its own RfC or at least some real effort made to invite those editors back to weigh in on this, and perhaps notifcations at the noticeboards Tony did in case someone has an opinion about this but not the previous issue? Best, Barkeep49 (talk) 22:56, 23 July 2019 (UTC)

  • I added this option because so many people above seemed to be arguing for it, although it was not one of the questions originally asked in the RfC. It also seems closer to any actual problem (other than generational injustice). —Kusma (t·c) 06:21, 24 July 2019 (UTC)
  • [ec] This seems like a bit of a non-sequitur. I am not against changes to bit-retention or bit-recovery criteria, but I am missing the part of this suggestion that addresses tool misuse by people who have never been desysopped or by people who have been resysopped. In fact I seem to be missing what the point is altogether. What is the problem that Option 4 is intended to fix?
    I think that the terms strict and lax are red herrings in this context. They distract one from the point which is (or should be) quality assurance.
    I assume we would all agree that we need competent, reliable and trusted people to do the admin work. We would probably also accept that these people are volunteers and are not obliged to do any specific parts of the work. Although sysops get all the tools, they may use those they choose to use and not use those they don't choose to use, so great depth of knowledge and skills in using the tools should not be required for all the tools the person has no intention of using. We should analyse the problem, identify places where improvement is needed, and propose actions which could fix those problems. If someone has all the tools and does not use any of them, there is no useful point in keeping them. Is there any significant risk of harm in keeping the tools? If so, what harm? We should be considering ways of mitigating that risk, not expanding rules which do nothing useful. · · · Peter Southwood (talk): 06:38, 24 July 2019 (UTC)
    Pbsouthwood, to be perfectly honest, I think the main problem with the resysop policy as it stands is the amount of controversy that some resysoppings generate (and a more peaceful WP:BN would be nicer). We do not seem to have any examples of a tool misuse problem that would have been prevented by a change to the resysop criteria, but we may have a more general quality assurance problem with legacy sysops. But maybe we just need to make sure we desysop those quickly enough when problems do arise instead of worrying what are the signs that problems might arise. It may be that our processes work fine, but people still feel they don't work fine. If the problem is that RfA is this terrible ordeal, maybe instead of forcing more people to go through it we should make it less of an ordeal? But as we all have tried to change RfA and given up, we tend to go and fiddle with other parts of the process. I think there is more evidence that the retention criteria are flawed than that the resysop criteria are flawed. But you are absolutely right that we should define the problem we're trying to solve so we can then decide whether any proposed solution solves that problem. —Kusma (t·c) 09:39, 24 July 2019 (UTC)
    I think part of the problem is that we have a culture of adversarial debate, where a dispute is generally "won" by the side that can wear their opponents down with walls of shortcut links and diffs, and where the concept of seeking a reasonable compromise or a better solution takes a back seat.
    Would RfA be less adversarial and ad hominem if there were objective measures of competence in the necessary skills for the work that the candidate offers to take on? It is hard to tell, as we don't have any such tests yet. We could try setting up something for one or two of the skills which are most controversial, and see if it makes a difference. · · · Peter Southwood (talk): 15:23, 24 July 2019 (UTC)

The apparently arbitrary closure of the above discussion ignores the point that they are parts of the same problem, and ignoring the causes does not give one a good handle on fixing the effects. · · · Peter Southwood (talk): 09:14, 25 July 2019 (UTC)

I too do not see why this had be overwhelming consensus either way...Lectonar (talk) 09:29, 25 July 2019 (UTC)

Mandatory recall votes in other WikipediasEdit

This is mostly WP:OTHERCONTENT trivia, but I saw a recent administrator-related RfC that I found very interesting in the Finnish Wikipedia. They already have a policy in place, in which any editor can call a vote for binding recall. Now they almost unanimously lowered the required treshold for desysopping via recall from 60 % to 51 %. A neat thought experiment imagining such a policy in the English Wikipedia, that administrators would be required to keep community trust after passing their RfA. --Pudeo (talk) 06:20, 16 August 2019 (UTC)

  • The classic objection to binding/mandatory recall seems to mirror the same debate between elected and appointed/lifetime seats on courts, specifically that if admins are exposed to possible election/recall, the desire to avoid that may invade their discretion and cause admins to avoid making tough, unpopular (but correct) decisions. I’m sure there are other objections that aren’t just variations on this theme, but I’m really not aware of them.
    My thought for an alternative recently was to explore the possibility of mandatory continuing education for admins, where in order to maintain the bit, they’d need to complete a certain number of credits every once in awhile (perhaps annually), which is pretty standard in licensed professions. I really think much of adminship is closer (or should be closer) to a licensed profession than a judicial posting or elected office. Even the RfA process being roughly consensus based rather than a straight vote is reminiscent of how bar admission used to work when it was an unabashedly protectionist system (insofar as someone being “not one of us” would face serious character and fitness challenges to admission). Anyway, I think continuing education modules, perhaps developed on Wikiversity or another platform, could provide a way of ensuring admins’ knowledge is still in line with community expectations, and when admins deviate from those expectations, we would know with confidence that it wasn’t a minor lapse. These modules could cover a wide variety of tasks as well, ranging from introductions to various on-wiki processes (e.g., filing SPIs or closing AfDs), to handling technical issues (e.g., how to calculate IPv6 rangeblocks), to strategies for dispute resolution. Of course, getting people to create and update content may be a challenge, but I think it’s worth exploring as an alternative or as a companion to a recall system. —/Mendaliv//Δ's/ 07:07, 16 August 2019 (UTC)
  • A system based on any editor can call a vote for binding recall would be unworkable here; English Wikipedia has around 1100 admins, while Finnish Wikipedia has 35 admins and (as of December 2018 when the WMF stopped publishing the data) only had 86 active editors so it genuinely is possible for everyone to be familiar with everyone else. Given not only that every admin who's done something remotely controversial would regularly have someone calling for their head, but also that all the inactive and semi-active admins would regularly have people calling for them to resign, on en-wiki we'd realistically be looking at between 200 and 500 recall votes active at any given time. German Wikipedia has piloted a system of cumulative complaints, in which an admin who accumulates a certain number of complaints over time has a recall vote triggered, which might be more workable. ‑ Iridescent 07:18, 16 August 2019 (UTC)
I can't see that working here; it would still discourage admins from doing their jobs, since it actively penalises the more active ones. Let's say your average admin has an error rate of 1%, generating a complaint each time they make a cock-up. An admin who logs 10 actions per year - well above the minimum necessary to retain the bit - can expect to deal with one complaining user, hardly enough to trigger a recall or even cause anyone to bat an eyelid. An admin who logs 1000 actions per year has 100 strikes against them, and will find themselves dragged to ANI for a recall vote every year. That's got to be rather dispiriting... Yunshui  07:33, 16 August 2019 (UTC)
Your math seems to imply an error rate of 100 out of 1000 or 10%. Regards SoWhy 08:58, 16 August 2019 (UTC)
  • @Mendaliv:A minor correction, but another frequently raised objection against admin recall/reelection systems is that they are redundant to Arbcom. Personally, I wonder if a Commons-like "recall only if there is prior discussion leading to some consensus for removal, otherwise the request may be speedily closed by a bureaucrat as inadmissible" process may assuage some of the other issues mentioned above. Jo-Jo Eumerus (talk, contributions) 07:57, 16 August 2019 (UTC)
  • I used to think that a mandatory recall process would be a good idea; but in the last couple of years I've realized that, for whatever reason, editors on Wikipedia who bear a grudge are willing to go to unbelievable lengths to get back at somebody. If, as an admin, I have dealt with a large POV dispute, it is not unlikely that one entire "side" will show up to a recall !vote; and a couple of dozen !voters are enough to upset any reasonable community-based judgement. I still sort of believe that recall shouldn't have to involve a two-month ARBCOM case: but I'm no longer certain about this either. A number of admins have created recall criteria; 140, last I checked. How many have had recall proceedings initiated against them? I cannot think of any, which makes me think the enthusiasm for a community recall process might be misplaced. Vanamonde (Talk) 15:25, 16 August 2019 (UTC)
  • Interesting, thanks for that link. They do indeed seem to have gone out of fashion. Not that that's necessarily a bad thing; the few links I skimmed rarely seemed to show actual tool misuse or conduct unbecoming...Vanamonde (Talk) 15:43, 16 August 2019 (UTC)
  • All those "recall criteria" are not binding and not recognised by any formal policy. And any admin can refuse to step down even after they failed "reconfirmation" and in fact, it did happened. So not initiating them is not only because there's no interest. People tend to shun what's essentially meaningless. – Ammarpad (talk) 16:19, 16 August 2019 (UTC)
  • That wasn't a formal recall petition; that was a bunch of people turning up kvetching about him until he resigned in disgust. (As was explicitly stated there, there is no official process for a Bureaucrat Recall discussion, nor is there policy support to enforce results.) That kind of harassment-into-resignation happens relatively frequently, but it's not what's being discussed here; we're talking about the formal process. ‑ Iridescent 17:37, 16 August 2019 (UTC)

"*No He was formally requested in the crat page by a user in the course of a discussion over a particular discussion. "u have the option to recall yourself on your userpage and I'd like to take you up on that now." That is he did not enjoy I no longer enjoy the confidence of a majority of the community (i.e. 50% +1) in a representative discussion that is User:WJBscribe#Recall. Pharaoh of the Wizards (talk) 17:51, 16 August 2019 (UTC)

  • @Pharaoh of the Wizards; And in that resignation you do notice he put "recall" in a scare quotes. He knew it's not binding. But his case is outlier here for three things: 1. This is a bureaucratic action, a level held in higher level of trust than adminship. 2. His actions were clearly over-the-top and unprecedented. 3. There was an active Arbcom case at the time which was closed the way it was, largely because of his resignation. So I wouldn't really count that as a "successful recall". Not mention I did not see whether the recall votes follow his own "recall criteria" if he had any.Ammarpad (talk) 17:53, 16 August 2019 (UTC)
Return to the project page "Administrators".