A request for comment to discuss changes to the policy on adminship term length WormTT(talk) 13:39, 9 March 2021 (UTC)[reply]

Wikipedia has recently reached its 20th year and has changed significantly since its inception. As a consequence, many administrators were elected to a very different community, with very different norms. With a significant majority of administrators being elected over 10 years ago, I propose revisiting a perennial proposal - that of fixed adminship term length - by adding the following to the Wikipedia:Administrators policy:

Poll after 10 years

Adminship on Wikipedia is not a lifetime status and is reviewed at a fixed interval. After a period of 10 years following the closing of their last successful request, administrators must either submit a "Poll to retain adminship" to confirm they retain the confidence of the community, or request removal of their administrator user-right. The administrator should receive more than 60-65% support in the poll to retain their user-right, although some bureaucrat discretion is acceptable. Periods where the administrator have relinquished their tools will not be discounted from the admin's term length.

The poll will remain open to votes for 1 week and will be linked to a separate section of the Wikipedia:Requests for adminship page. Polls will be clerked by the bureaucrats and should be considered more lightweight than Request for Adminship with votes only including a very short statement and questions / discussion limited to the talk page. Editors who are not "extended confirmed" status or have been sanctioned by the administrator in the 3 years prior to the poll will not be eligible to vote, but may comment on the talk page, where they should declare their ineligibility.

Those individuals who do not put forward a poll by their allocated date will be considered as giving up their user-rights voluntarily, but are no longer eligible to request reinstatement of adminship at Wikipedia:Bureaucrats' Noticeboard. Instead, they may subsequently use the poll process until/unless they reach inactivity requiring a new Request for Adminship. If an individual is not successful in their poll at any point, they will be involuntarily desysoped and they are not eligible for another poll, however they may submit a fresh Request for Adminship at any time.

Initial implementation

It is noted that there is a minimum of 876 administrators eligible for poll.[note 1] Acknowledging that one of the largest concerns leading to the perennial nature of term length is how to manage the volume of polls, the initial implementation will be staggered between 1 July 2021 and 31 December 2022.

At the close of this RfC, eligible administrators will be allocated a week for their poll to be put forward and will be notified of this date. Allocation will be at a rate of 15 administrators per week ordered by the oldest valid[note 2] request for adminship to the newest. The administrators will be contacted for the second time to put forward their poll one week before their allocated week or their user-rights will be removed. Administrators may put in their poll earlier than their allocated date at their discretion. Wikipedia:Requests for adminship has previously managed high number of requests, for example throughout 2007 there was an average of 76 requests per month.

An hypothetical example poll can be viewed at Wikipedia:Request for comment/Adminship term length/poll example.

A rough list of admins with approximate date of their most recent RfA is found at Wikipedia:Request for comment/Adminship term length/list. Someone better with database queries / bots will be able to make a much better list if this proposal is successful.

Notes
  1. ^ As of 1 March 2021, there are currently 1,109 and there have been 233 successful requests for adminship since February 2011, meaning that at least 876 administrators passed their Request For Adminship over 10 years ago.
  2. ^ A valid request for adminship is the most recent successful request for adminship for an administrator, including "Reconfirmation RfAs".

Survey

edit

Support

edit
  1. As proposer. I believe fundamentally that adminship should not be a lifetime role and that it should be down to the community to decide what the activity requirements or confidence level should be. We have a great many administrators whose RfA's occurred on a very different project. Whilst the vast majority have adapted with the times or disappeared to inactivity, there are still some who remain doing the bare minimum or are not familiar with our norms. This poses a number of problems, from giving the impression that we have more administrators than we actually do to risking those coming out of the woodwork to take "cowboy" actions that are not appropriate.
    Whilst we could, say, increase minimum activity levels, or admin tool usage levels, this system is far more flexible and will allow the community to give feedback to those administrators. I have attempted to put in safeguards to stop railroading and explained how an initial implementation could work (since this has been the biggest sticking point in the past).
    I will also say that I believe this proposal should be considered the "minimum", so that we are able to implement it. Once past the initial implementation we, as a community, can make changes to it, such as reducing the term limit, altering the poll requirements and so on. What's more, whilst this is not a community desysop proposal, it forms a structure that could be used for one in the future.
    Finally, I'd like to thank those who participated in the workshopping of this proposal (available on the talk page) - I fully expect some of those individuals to oppose this solution, but I really do appreciate their time giving me their thoughts. WormTT(talk) 10:53, 9 March 2021 (UTC)[reply]
  2. There is the issue of how to deal with the initial surge of 10+ year admins, but I'm sure the details can be worked out based on the suggestions above. I'm very firmly in support of this. Boing! said Zebedee (talk) 10:46, 9 March 2021 (UTC)[reply]
    (Oh, and wouldn't you know it - my 10th admin anniversary was just over a week ago ;-) Boing! said Zebedee (talk) 10:48, 9 March 2021 (UTC)[reply]
  3. This is very well thought out, it seems, and a good, reasonable way to ensure admins retain the confidence of the community. ProcrastinatingReader (talk) 10:57, 9 March 2021 (UTC)[reply]
  4. Shouldn't be an onerous task as most will go through with little effort. Surely just like an Annual general meeting election of officers or the annual election of company directors. Adminship shouldn't be a job for life. Nigej (talk) 11:13, 9 March 2021 (UTC)[reply]
  5. Support Almost everything in life where someone is in a position, that is part of some office; it is usually fixed-term, why not here? The majority of the polls would sail through. Only those who were contentious would get some flak, and that alone is a good reason for it. Only those tired and dodgy admins who needed to go would get focus. It would open up new space for new folk to come in. scope_creepTalk 11:18, 9 March 2021 (UTC)[reply]
  6. Reconfirming with the community once every ten years is not too much to ask. Alternatively, ditch the reconfirmation and just make it a ten year limit, period. We don't need lifetime admins; fresh blood and regular turnover in volunteer corps are essential; admins are no exception. There's nothing about the block, protect, or delete buttons that benefits from such lengthy tenure, whereas there's quite a bit of risk in lifetime appointments. Levivich harass/hound 15:11, 9 March 2021 (UTC)[reply]
    @Levivich: you'd propose we should lose c.80% of the admin corps? Nosebagbear (talk)
    Yes! As radical as that sounds, yes. We don't just need terms (reconfirmation every X years), we need term limits (a cap on how long an editor can be an admin). But for now I'll be happy to support terms without limits. Levivich harass/hound 15:53, 9 March 2021 (UTC)[reply]
    It took me 3-4 months just to become properly adept in my original admin fields, and a year to broaden out to my current main areas and pick up sufficient competence in them. Assuming a proportional loss of the active admins, do you have 240 admins to step up the second we lose them, and what about that intervening period of inexperience? Nosebagbear (talk) 17:45, 9 March 2021 (UTC)[reply]
    I mean there's so much wrong with your set-up I don't know where to begin :-) We wouldn't toss the 240 overnight; we'd cycle through over time. Also, admin tasks are not evenly distributed among admins: it's not true that for every admin we lose, we need to gain one admin, in order to maintain backlog levels. The 80/20 rule applies here as everywhere. Even still, we could probably lose some set of 240/500 active admins and barely even notice it; whereas if we lost 10 or 20 certain specific admins, entire areas of the website would totally shut down. It is not a numbers game at all. Meaning, it doesn't really matter how many admins we have, it matters much, much more who they are. I think it's a bad thing when one admin is blocking 1,000 users or deleting 10,000 pages, or handling 75% of whatever backlog or process or noticeboard. And I think it's even worse when that one person continues doing that for years and years. It's not healthy for the community to develop these long-term dependencies on a few individuals. I think, as a general comment, that the whole "we need more admins!" or "we're going to lose admins!" line of thinking is flawed. I think a lot of people don't run for RFA simply because they don't think they're really needed, and volunteering to press the block/protect/delete buttons isn't that interesting to a lot of people. I think we should consider that what's stopping people from stepping up is the lack of people stepping aside. When backlogs get too long is when we'll see an increase in editors volunteering to be admins, IMO. Levivich harass/hound 18:19, 9 March 2021 (UTC)[reply]
    My ideal "life cycle of an editor" would look like this: get here, spend a year or two editing, run for RFA, spend a few years pushing the admin buttons, then if you're still around you can try other hats like CU, OS, crat, arbcom, steward, whatever, but you wouldn't spend more than a few years in any one position. Each of these roles would have regular turnover, and an editor who'd been here for 10 years will have, at some point, possibly worn many hats (possibly in multiple rounds), but will not have worn any one hat for 10 years straight. A system like that prevents both burnout and stagnation. And "admin" will be a role fulfilled, and not a rank achieved. Levivich harass/hound 18:27, 9 March 2021 (UTC)[reply]
    I never said we'd lose 240 at once, but we would lose them far quicker than they'd be replaced. Given how much the current mop corps goes around hunting for candidates, I'm not sure how strong a "I don't apply as there isn't a need" group there is amongst the suitable candidates. If you have genuine evidence to think not just a few but lots of potential admins are in that group please, please, share - even a few more would be useful under whatever we do. Yes, certainly you'd expect a weighting towards the less active, absolutely, but the specific admin point to me supports concern about it, not support, unless everything goes right. This is particularly the case when the proposal is likely to cause more disruption in certain categories (rather than a straightforward proportion split, or even adjusted for inactive/active admin split). "When backlogs get too long is when we'll see an increase in editors volunteering to be admins" is a position I find too negative to me - it would come with major unpleasantness in the interim, plus the dislocation time as experience has to be gained after. It's the same reasoning that I get concerned with "if ArbCom gets more aggressive about desysops, the Community will become less stressed at RfAs and re-RfAs". Eventually, maybe, but in the meantime we get some squashed individuals. Nosebagbear (talk) 02:10, 10 March 2021 (UTC)[reply]
    This pretty much flies in the face of WP:NOBIGDEAL. Admins have the tools because they're trusted to use them, and because they have a use for them. As long as both of those remain true, there isn't a reason to remove the bit from someone. Ideally, every competent, good-faith editor would be an admin, after all (though this obviously can't be true in practice for a multitude of reasons). Elli (talk | contribs) 09:30, 10 March 2021 (UTC)[reply]
    I think it's lifetime appointments that fly in the face of "no big deal," and we have no idea if most admins have the trust of the community because the community of today was never asked about most admins. (And never will be.) The concept of adminship as a rank flies in the face of "no big deal." Levivich harass/hound 15:26, 10 March 2021 (UTC)[reply]
  7. I find the idea that an admin is a lifetime appointment (subject to some fairly minimal activity levels) to be antithetical to our project's ethos as more authoritarian than democratic. 10 years is a long time so there is plenty of cover for admin to work in troublesome areas without fear of looking over their shoulders and the process some built in protections. But for all those who say that there is no problem here, I fundamentally disagree: lifetime appointment and the privilege that comes with it is a problem as the people in power get farther away from being a new editor and being understanding of what that experience is like. Best, Barkeep49 (talk) 15:44, 9 March 2021 (UTC)[reply]
    At the risk of being pedantic, lifetime appointments are a key feature of the judicial branch in some democracies, notably in the US. The rationale is that lifetime appointments free judges to rule in accordance with the law without fear of momentary popular whim, and to make unpopular but legally correct decisions. In other words, the US Constitution, which is generally considered a central democratic text, incorporates lifetime appointments as essential to an independent judiciary. Regardless of whether one agrees with that rationale or its current practical application, it is clearly not the case that lifetime appointments are inherently "authoritarian or "anti-democratic".

    In regard to your final point, I agree that the gulf between admins and in-the-trenches editors is a problem. I've always, from my early days here, both edited and adminned in difficult and controversial areas, but that is an extremely difficult path and not one that I would recommend in our current community climate. I'm intimately familiar with what it feels like to deal with tendentious, disruptive, or obstructionist editing in the trenches, and that experience has always informed my approach to adminship. But for better or worse, avoidance of difficult topic areas or discussions is interpreted positively, as evidence of "neutrality", and the easiest path to gaining and retaining adminship is to be as anodyne as possible and to avoid difficult topic areas, conflicts, and other hard stuff. I wish it were easier to function as both an editor and admin, to avoid the out-of-touch problem, but I'm not sure that this proposal solves that problem. MastCell Talk 18:10, 9 March 2021 (UTC)[reply]

    Administrators aren't judges. And let's be clear lifetime appointments are only at the federal level - the states I'm aware of generally have terms. I think in this analogy you could make a better argument for lifetime crat appointment than admin. The people who ostensibly make the hardest decisions, ArbCom, are on two year terms and basically every arb who wants gets re-elected. People who hold the highest level of permissions globally, stewards, are on one year terms and that too works. I'm loath to use my personal experience as universal, but I don't think I'm anodyne - if I was would we even be having this conversation? - and have indeed stepped into heated areas and somehow come out ok. Do I work those intently as some? No but I also don't avoid them. And again I return to arbs and stewards as reasons why my personal experience might not be off base with the experience on the whole. Best, Barkeep49 (talk) 18:19, 9 March 2021 (UTC)[reply]
    I should add: at heart I trust the community as a whole to be reasonable. This is why I think people are disinclined to oppose people at RfA even when they had a one off maybe not great experience. I think there are plenty of ways for people to disagree without being disagreeable and I see that outcome respected on Wikipedia all the time. Best, Barkeep49 (talk) 18:42, 9 March 2021 (UTC)[reply]
    One reason I trust the community as a whole to be reasonable is that there are 157 recall-able administrators, all of whom have made controversial decisions, all of whom have accumulated disgruntled editors, but none of whom have ever been recalled, or even had a petition started (AFAIK; the last recall was 10 years ago I think). Some of the admins listed have been brought to arbcom, some even desysopped, but not by recall. To me, that strongly suggests that the danger of vindictiveness is low, perhaps even miniscule. Levivich harass/hound 18:48, 9 March 2021 (UTC)[reply]
    I don't think we can extrapolate from the recall process. I wouldn't expect you to know this, because it happened before your time (I think), but there was a particularly high-profile case in which an admin passed RfA by the skin of her teeth, in large part on a promise to be open to recall. After a series of questionable judgments, her recall process was invoked, and she simply refused to honor it. The case went to ArbCom, which basically decided, meh, recall is not enforceable and basically just a campaign promise which an admin can choose to disregard without consequence.

    After that, I think the wind went out of the sails as far as recall and people gave up on it as a meaningful process. So, in my view, the lack of successful recalls doesn't speak to any absence of vindictiveness or what-have-you, but rather to the low credibility of the recall process and a lack of faith or interest in it. MastCell Talk 20:06, 9 March 2021 (UTC)[reply]

    According to the stats I've seen, something like 90% of active editors have been here less than five years. So I don't believe that this one incident you're referring to, which (correct me if I'm wrong) happened like 10+ years ago, has any effect on anyone today (or for the past five years at least). Levivich harass/hound 20:15, 9 March 2021 (UTC)[reply]
    Eh, I think MastCell's point definitely stands. The fact that the community has not invoked any recall process for those 157 administrators is not because the community is just that virtuous, but because the recall process itself is a largely toothless procedure and can generally be ignored. Mz7 (talk) 00:52, 10 March 2021 (UTC)[reply]
    Barkeep49, admins most certainly are judges. Many use it to penalize individual editors , relying entirely on their judgment alone; some use it to judge articles and other miscellaneous functions.
    Most judgeships in US states are term limited. It make them liable to party politics as it changes. DGG ( talk ) 02:13, 12 March 2021 (UTC)[reply]
    It's clear most of the community disagrees with me on this topic and since I'd rather not be told to read the room again I'm hesitant to continue discussion about this topic. Best, Barkeep49 (talk) 02:26, 12 March 2021 (UTC)[reply]
  8. Support as a very reasonable restriction. Initial implementation is an issue, but I'm sure we can figure out a way around it (either a trickle over time, as WTT suggests, or just grandfather in those currently at 10+ years). I don't share the concern that administrators who work in controversial areas will be removed by this process. Clearly that's a widely held concern in the community, and is being voiced here and at the desysop-for-cause proposal. The same community members who fear such removals will be around to oppose them at the proposed re-RfAs. I highly doubt they'll be overwhelmed by a sudden majority of the slighted. If I'm wrong about that, we can always revisit the process to tweak the details or cancel it altogether. Ajpolino (talk) 15:45, 9 March 2021 (UTC)[reply]
  9. Ten years is a very long time. The standards of adminship have changed wildly in the time since. I think it is reasonable and useful to ask for a "booster shot" of adminship. Admins ought not be appointed for a lifetime. I think this should help with a variety of issues, including out of touch legacy admins, and admins who make only token actions to keep the bit. The staggering of reviews solves the problem of overwhelming the community. I also hold out hope that this will encourage our older admins to be more active and engaged. The issue of our first admin cohorts has been simmering and tackling it is long overdue. This should not be interpreted as being anti-legacy admin, I think if well executed it will actually benefit legacy admins. CaptainEek Edits Ho Cap'n! 16:05, 9 March 2021 (UTC)[reply]
  10. Support - This is something that I've been in favor of for many years, and I'm grateful to WTT for making the proposal. Let's start with something, and if there are problems int he implementation, it can be tweaked, however, when I've played with this concept myself, I recognized that admins can collect "enemies" (for lack of a better word) because of their use of necessary blocks and other sanctions, so in my own version I thought that a bare majority (i.e. over 50%) would be sufficient to re-confirm, rather than the supermajority of 60-65% of this proposal. Beyond My Ken (talk) 16:56, 9 March 2021 (UTC)[reply]
  11. Support After ten years the expectations for admins from the community, and the consensus on what admins should do will have changed. However, I am concerned about having 15 !votes per week, as this could prove to be too much at once. βӪᑸᙥӴTalkContribs 17:07, 9 March 2021 (UTC)[reply]
  12. Support A needed step in my view. If adminship is not a big deal, asking them to be reconfirmed every ten years (a generous timeframe in my view) should likewise not be a big deal. Intothatdarkness 18:12, 9 March 2021 (UTC)[reply]
  13. Strong support The herd of legacy admins is long overdue for a culling. Iaritmioawp (talk) 19:54, 9 March 2021 (UTC)[reply]
  14. Support as someone coming up for 16 years of adminship I have no problem with this proposal. I might extend the implementation period a bit - certainly I'd start it off with fewer than 15/week while the process is finding its feet; and I would be a bit more flexible about timing (a ~2-3 week window around your anniversary) but these tweaks do not eliminate my support. Thryduulf (talk) 22:46, 9 March 2021 (UTC)[reply]
  15. Support. I have been convinced of the urgency of this proposal by reading some of the oppose votes. Maybe some divas will step down and leave at the mere thought of a reconfirmation. This would be a good thing by itself and by streamlining the process. But this proposal has no chance to fly: the sleeping admins wake up at the simple evocation of their sleeping status. Pldx1 (talk) 12:00, 10 March 2021 (UTC)[reply]
  16. Strong support 100% - and here are 2 links to prior discussions: User_talk:Jimbo_Wales/Archive_230#Term_Limits_for_Admins, User_talk:Atsme/Archive_27#Thanks_on_term_limits. I think the current proposal is a welcome and much needed change. Over time, it is human nature to form alliances and harbor prejudices. When admins have too much "control" over discussions - which is what DS/AE created - it is not good for the community, or this Project because of Wikipedia:POV creep. Besides, why would any admin want to maintain their position knowing the majority in the community have lost trust. It should not be a 50-50 proposition because admins have blocking power and that requires a clear majority support to keep the tools. Not getting a clear majority may well be sending a strong message of caution. In the past, I have proposed a trial period for new admins, and a 10 year evaluation with 2-year break, and then coming back to RfA. Also, I'm a little concerned over having administrators vote their own fate for an elected lifetime position. It's a bit much to ask of them. What we've been dealing with over the past decade is no longer about vandalism or real disruption -- it's about micromanaging articles, POV pushing with "authority" (closing RfC, and sole discretion to impose DS/AE), and unfortunately, a measure of WP:POV railroad to eliminate editors they don't like. Nothing about that is good for the project, or for maintaining a NPOV in the pedia. Mainstream has taken notice of WP's NPOV issues, and that is not good thing. Another concern re:this survey - admin iVotes are certainly welcome as is their wise input, but their input should be treated as INVOLVED for obvious reasons. Just sayin'... Atsme 💬 📧 12:35, 10 March 2021 (UTC)[reply]
  17. Strong Support per all above, no need to rehash what is a good step. Sir Joseph (talk) 14:42, 10 March 2021 (UTC)[reply]
  18. This would obviously apply to me (and I've whacked my hand up to be involved with any trial that goes ahead). There is no doubt the community has changed massively since I got my tools, and even since before my extended break to now. What I think is important is the way I have chosen to "reintegrate" - slowly, with some level of trepidation, being careful not to wade into things with the institutional knowledge I think I have, but might not actually have. Some may define this as "gutless", and you could be right - but it's my way of ensuring that I gently ease back in. I can't be alone here in this situation, and to be honest, a 10 year reconfirmation may allow for a better, more in-tune admin corps. I will say that it needs to be very well-moderated - no new accounts (autoconfirmed), the sanction requirement is critical to protect those working in controversial areas. That being said, I don't mind this idea as a matter of principle. Daniel (talk) 01:09, 11 March 2021 (UTC)[reply]
  19. Support The only oversight available to reign in poor admin behavior is arbcom and while this has been effective in removing some there remains a number of admins who are abusive and misuse their position. If all is well then the reconfirmation should pass.--MONGO (talk) 21:00, 11 March 2021 (UTC)[reply]
  20. Strong Support More liability is required. Admins have too much power for this not to be the case. J.Turner99 (talk) 17:44, 12 March 2021 (UTC)[reply]
    Note that this account was indefinitely blocked for disruptive editing. Doug Weller talk 13:33, 13 March 2021 (UTC)[reply]
  21. Support - I was thinking of this one for a little while now. I think it is worth perusing, even if it is not perfect I think it is a good basis to work from. We should keep in mind that we don't want perfection to be the enemy of good enough. If we are going with the argument that becoming an admin is no big deal then neither should this be. Also in recent months there has been issues with legacy admins coming back to the project unprepared for the current norms. So it is not a solution in search of a problem, the problem keeps happening. I am also not really seeing any downside here. If a good admin runs again their reconfirmation should be a breeze. This is especially true given bar for getting the bit is lower than when they got theirs. One of the oppose issues I have seen brought up is stating RFA is horribly broke, which I disagree with but, is not really a reason to oppose this. That is a reason for them to try and fix the "broken" process. I am also not sure I agree with the argument that we need more admins, more would be okay I suppose, but more good and active ones is really the issue here. Just having more in general would probably be a net negative. Finally this should also lead to a larger discussion on activity requirements to maintain the tools. While we are all volunteers here and no one is required to participate, we are also not required to let them keep the tools after long periods of inactivity. Which can cause major harm to the project in the wrong situations. PackMecEng (talk) 15:44, 13 March 2021 (UTC)[reply]
  22. Support Many admins appointed during the early days of Wikipedia were apparently just given Adminship without any serious scrutiny, unlike today, and frankly aren't competent for the role. Like any holder in a position of power all admins should be subject to periodic review and scrutiny. Mztourist (talk) 04:10, 15 March 2021 (UTC)[reply]
  23. Support. The timing of this RfC is rather bad. It would have been far better to wait for the desysop RfC proposal to close first. However, fundamentally, I believe that this proposal provides the right approach. Desysop for cause should be left to ArbCom as an extraordinary procedure based on a fact-finding process. I see this proposal as a reconfirmation rather than as a desysop proposal. As WTT says above, adminship should not be a lifetime appointment. Every admin, whether they are doing a great job, a so-so job or a lousy job, should periodically stand for re-confirmation to see if the community still trusts that admin with the tools. I expect that most admins would be reconfirmed, and rather easily. But the process would allow for genuine accountability to the community, substantive feedback and critical review of the admin's performance. I believe the existence of this process will cause many so-so admins to shape up. It will also provide the incentive for the admins who don't perform many admin tasks to start using the mop. A reconfirmation format provides a much more positive setting than a desysop format, and will be much less acrimonious. There is no need for a tortorous certification process involving ANI, etc. The ArbCom desysop process would not be affected. In real life accountability of elected officials to the voters is ensured through having them run for re-election, not through making their appointments permanent and introducing elaborate impeachment type procedures. We should use the same principle here on Wikipedia. Nsk92 (talk) 19:18, 15 March 2021 (UTC)[reply]
  24. Support In any kind of power structure, clear terms can never hurt. ~ HAL333 16:34, 16 March 2021 (UTC)[reply]
  25. Support that any advanced permission on wikidedia should not be a lifetime role, use it or lose it is my general position. A small caveat, there are people on wikipedia, editors that have stated that they will oppose anything I support. Govindaharihari (talk) 09:23, 19 March 2021 (UTC)[reply]
  26. Don't let the perfect go in the way of the good. feminist (talk) 16:55, 21 March 2021 (UTC)[reply]
  27. Hell yes. We have needed this from the start, and it's been stark obvious that we've needed it. We even have it already in one form (inactive admins are auto-desysopped, though they can get the bit back by simple request if they get active again and are not "under a cloud"). "Adminship is no big deal" has stopped being true for a combination of three reasons: effectively lifetime appointment, ArbCom instituting WP:AC/DS, and it being nearly impossible to get someone desysopped. The third of these is already gradually becoming less true as ArbCom's resistance to taking desysop cases wears down; the second is slated for some reform; and the first can be fixed right here. We need to get back to the situation we had a decade ago, where anyone competent, sane, and civil could become an admin without much difficulty, and we had admins to spare. The only way to get there is undoing the problem that if the community approves someone who later turns out to be a bad choice that we're stuck with them forever unless the uck fup so badly that even the most desysop-resistance Arbs have to cave. This should not be like impeaching a Supreme Court justice. Proposals like this fail because of generalized fears and anxieties; just stop. The WP community decides what the WP community is and does; if some detail needs to be worked out or adjusted later, then we can do that.  — SMcCandlish ¢ 😼  21:19, 21 March 2021 (UTC)[reply]
  28. Support, this is a needed and well-thought-out way to further involve the community in the adminship process.

Oppose

edit
  1. Oppose Waste of community time to trawl through these polls that are bent on simply proving a point that adminship is not for a lifetime (says' who?). To be clear, I am mostly in the WTT camp. Just this time....Lourdes 10:58, 9 March 2021 (UTC)[reply]
  2. Oppose Agreed, waste of time. If I became an admin, I wouldn't want to have to be looking over my shoulder to see if any actions I took with the admin tools were popular or not. The C of E God Save the Queen! (talk) 11:05, 9 March 2021 (UTC)[reply]
  3. I hate to be "that guy" but I think this proposal has several flaws, any of which is sufficient to make me oppose it:
    • Incorrect basic assumption: The basic assumption seems to be that "old" administrators (which comprise ~80% of all admins at this time) might not be familiar with changes ("were elected to a very different community, with very different norms"). This disregards that most of these admins have in fact kept up with those changes and actively worked to familiarize themselves with the current norms.
    • 2007 is not today: Saying RfA has in the past handled dozens of requests per week ignores the fact that back then, the active user base was much larger than it is today. The reason we have so many admins from 2006 to 2009 is exactly because there were so many more active users in that time. There is no reason to believe that today's RfA can handle the same amount as 2007's could.
    • RfA fatigue: But even if we assumed that current RfA could handle 60+ requests per month, there is no reason to assume that this will be sustainable over 15+ months. Participation will decline sharply as fatigue sets in, leaving these confirmation polls to those editors who hold a grudge against the admin in question, skewing the numbers in their disfavor. Any admin would be wise to start their poll early to get as many non-partisan editors as possible to chime in, which means we might end up with hundreds of requests in the first month.
    • Admin fatigue: It's likely that scores of admins will decide to retire instead of facing such a poll, especially those, whose RfA was an ordeal they don't want to repeat. This will include admins who have nothing to "fear" from reconfirmation.
    • Lack of need: The idea that admins should demonstrate that they "retain the confidence of the community" ignores that there are already ways to remove admins that no longer retain that confidence (i.e. ArbCom).
    • Admins in controversial areas: Like Wikipedia:Requests for comment/Desysop Policy (2021), this proposal does not address the problem that admins who make necessary decisions that are not popular, such as sanctioning editors with large support bases, will likely face large opposition based on this alone and will likely lose their bit for just doing their job. In the long run, this will lead to many admins avoiding making any controversial decisions, thus enabling abusers to run wild. The reason lifetime appointment exist in real life is to allow people to act without fear of reprisals just for following the rules when it's unpopular. It's the same with admins on Wikipedia.
    • Popularity contest: Similar to the previous point, these polls will favor admins who are "popular", regardless of the actual skills and record.
    (TL;DR)  This is a proposal in search of a problem that will most likely lead to far more devastating consequences for this project in the long run by losing us a lot of admins. For these reasons and more, I strongly oppose this proposal. Regards SoWhy 11:07, 9 March 2021 (UTC)[reply]
    Thank so much for the feedback. I don't agree that I've disregarded the idea of admins keeping up with our norms, these individuals would sail through a poll. What's more - I believe you are not giving sufficient credit to our community, who I believe would be able to show confidence in admins who are willing to accept feedback. 60% is not an onerous task - and actually, I'd be concerned about any that do not reach that level of "popularity". As for the fatigue that might be associated with my initial implementation, I accept this may be a problem, but it would be something that could be changed on the fly. WormTT(talk) 16:11, 9 March 2021 (UTC)[reply]
    I disagree. Those who kept up with the norms are those who are also most active because if they hadn't, they would have quickly created problems for ArbCom or another venue to solve. But many of those will also fall into one of the other categories, especially those who made difficult decisions. There is no reason to assume that they would "sail through a poll". And yes, I do realize that I might have a personal stake in this as well, considering how at my second RfB I have received a number of comments from people who would vote to desysop me if they had the chance.
    Speaking of RfB, the huge workload for few crats is another reason to oppose. At this point of time, we have 19 active crats, many of whom have multiple roles that are time-consuming (three, including you, are Arbs; one is a steward, CheckUser and Oversighter, three others are Oversighters and three are BAG members). Adding closing ~60 polls per month for 15+ months, a number of which will be controversial and require crat chats, is a significant workload increase for which we probably would first have to double the number of crats. Regards SoWhy 18:04, 9 March 2021 (UTC)[reply]
    While maybe we don't have enough editors to replenish the admin ranks we definitely have enough admin who could be crats. It's just that there's not a ton of need right now so people figure why bother. If the desysop proposal passes we'll need more crats then too. But that is definitely a problem we can solve. Best, Barkeep49 (talk) 18:12, 9 March 2021 (UTC)[reply]
    We have very little data on how admins would fare at this sort of poll, though I believe many are concerned that administrators who have been removed "for cause" have fared poorly at RfA. However, two cases where we have genuine reconformation adminship passed at 90% and 73%. These were both highly active admins, one of which was given strong feedback in doing so. However, despite the strong feedback, it was a clear pass under this proposal. So yes, I believe we do have evidence that admins who make difficult decisions would pass. WormTT(talk) 18:59, 9 March 2021 (UTC)[reply]
    Both your examples are from ten years ago. If you are arguing that ten years ago, things were radically different to the point that we need this proposal, then logically you cannot use examples from back then to make your point now. Both those RfAs also had a significant number of people complaining that it was a waste of time, including a number of opposes. I don't think we have any evidence how these admins would fare today nor how others would. As for crats, RfB is even more of an ordeal than RfA (speaking from two experiences) and most admins, even if they could, probably don't want to go through something like that, especially if it means they were then on the hook for a huge pile of work that no one else could do. After all, with 1000+ admins, no admin is really "forced" to do any job if they don't feel like it but with only few crats, no crat can really decide to sit these out without attracting attention. Regards SoWhy 07:35, 10 March 2021 (UTC)[reply]
  4. The devil is in the detail, I so want to support something like this because we do have a problem with some older admins who have not kept in touch, and I like what Worm has done, and I suspect it will get a lot of support. However, this appears to a solution looking down the wrong road. It is not the amount of time a person has been an admin that is the problem (we tend to prefer experienced admins for roles such as bureaucrat and ArbCom) it is the level of engagement with the project. Asking already trusted and active admins to submit to a poll is a waste of everyone's time. We need to be focusing on those admins who simply check in once a year to keep their admin status. I would support a scheme that tackles inactive admins rather than one that tackles experienced admins. SilkTork (talk) 11:17, 9 March 2021 (UTC)[reply]
  5. per SoWhy. I freely admit that I'm likely near the front of the line for desysop under this 10+ year removal plan, and if so, that's fine. I have no interest in returning to any RfA gathering. On the face of it all it seems to have a "those of you with the most experience must leave now" vibe. Given how that past few years have trended here, maybe that fits best with the current group of editors. — Ched (talk) 11:33, 9 March 2021 (UTC)[reply]
    Ched The last thing I want is "those of you with the most experience must leave now" vibe. I personally believe that feedback is good - I've been an admin 10 years and in that time I have gone through 4 arbcom elections, a RfB and an admin review. Each, I learned from. I have no doubt you would pass a Poll to Retain - but that's not the point, it would of course be your choice. WormTT(talk) 11:45, 9 March 2021 (UTC)[reply]
  6. Oppose, not just because I'm coming up on 15 years of adminship ;) We have in the last couple of years tightened the activity criteria, and out of touch sysops who return to do an inappropriate IAR have been quickly desysopped, either by ArbCom or voluntarily after an ANI. There's a community desysop proposal going on right now that could also help get rid of problem cases. Could we do something more constructive instead of a yes/no poll on continued adminship? Some scrutiny of admin actions and some feedback on admin performance would be helpful (at least to me). Having that combined with a vote isn't so helpful, as it may be too late to fix some of the things that people are unhappy about. —Kusma (t·c) 11:46, 9 March 2021 (UTC)[reply]
  7. Oppose mostly per SoWhy. I agree we've had problems with people who became admins a long time ago and haven't kept up with changes in norms since. However it doesn't make much sense to make everybody go through RfA to deal with those. RfA is generally acknowledged to be a horrible broken process, it isn't a great way to provide feedback to people. We have about 300 admins who use the tools reasonably frequently [1], I strongly suspect that most of them would have to go through this process (we haven't had 300 successful RfAs in the last 10 years, and many of them won't be active admins any more). Disclaimer: this will affect me, but I'd have a better time than some. Hut 8.5 12:33, 9 March 2021 (UTC)[reply]
  8. Oppose SoWhy did a good job of spelling out my reasoning, but in general, I find the concept of adminship term limits and re-RfAs to be deeply unappealing in any context that doesn't simultaneously come with adminship going far more WP:NOBIGDEAL than it has been since about 2005 (think something like 'automatically giving the bit to everyone with a certain edit count/tenure combo -- and not a very high one'). I don't expect any proposition that makes adminship genuinely no big deal to pass; accordingly, I would oppose any move along these lines. There are certainly...serious considerations...regarding legacy admins, but IMO the solutions to this are much more complicated than term limits would be, and relate to things like our notoriously awful ability to not strangle new users in the crib. Vaticidalprophet (talk) 13:11, 9 March 2021 (UTC)[reply]
  9. Oppose, per SoWhy, SilkTork, and Lourdes. If I hadn't stepped on two or three toes and been stripped of the bit almost a year ago to the day, last week I would have also reached the end of a 10-year term and I wouldn't have minded in the slightest, and like others who reach pensionable age like I did a few years ago, would have rejoined the 'rank-and-file' with the dignity of knowing that I had not been lazy and that not all I did over the years was 'very, very bad'. WTT and I presented an idea for a 'community' desysoping a few years ago, and while it was in fact generally very well received, it failed on a couple of (minor) technical points, like this will and the other RfC by Tony that is running at the moment. I don't think a radical stampede to cull the ranks of admins either by a programme of reconfirmation or an alternative method to ArbCom for ripping admins apart is needed. Better would simply be to greatly, much greatly, tighten the requirements for activity and put an end to the nonsense of those who make one edit a year to retain what they consider to be the 'prestige' of being an admin. The number of admins who misbehave is not so great that they need to be smoked out by a system of procedural, popularity poll re-RfAs. Throwing out the sysops with the most experience and institutional memory to damnatio ad bestias is not the kind of diversion that is needed on the Foundation's flagship project. I have enormous respect for WTT for giving this thing a try - like BARC it's well thought out and like BARC it's a case of throwing ideas at the wall and seeing if any of them stick, but this proposal isn't it. And no, Ritchie333, I don't believe for a moment that there are many people around who would want your guts for garters. Kudpung กุดผึ้ง (talk) 13:40, 9 March 2021 (UTC)[reply]
  10. Oppose - Opens many cans of worms, perhaps even those which may not be conceived at this time. Also the amount of time seems abritrary, leaning toward the modern favoritzation (for many good reasons, but none that likely apply here except ease of tracking) of a ten-base system; a period of time, if necessary at all, would need to be much better defended. — Godsy (TALKCONT) 14:14, 9 March 2021 (UTC)[reply]
  11. Oppose per SoWhy. Length of tenure is not entirely equivalent to what I guess could be termed falling behind the curve. Rather, inactivity is more likely to cause that, and we already have a process for desysoping inactive administrators. If you look through RFAs when we were processing that kind of throughput, they were much more relaxed, while it is not a fun place for a candidate today. Additionally, I am not convinced the kind of necessary throughput could be done by the RFA system. Wikipedia:RFA by month shows that we haven't averaged more than 2 RFAs per month since 2013. Hog Farm Talk 15:23, 9 March 2021 (UTC)[reply]
    Hog Farm, not to split hairs but I think you meant successful RfAs: we had more than 24 RfAs in 2020 (please   mention me on reply). Sdrqaz (talk) 16:35, 9 March 2021 (UTC)[reply]
    Thanks for pointing that out, I misread the table. The point that we haven't consistently had anywhere the amount of RFAs we would have under this system for many years, and I don't think the system would hold up under the strain. I'm worried that trying to run a lot of candidates through in a short time will just make it even more toxic than it already is. Hog Farm Talk 16:50, 9 March 2021 (UTC)[reply]
  12. Oppose - This seems to be aimed at reducing the numbers of admins, when what is needed is more admins. Mjroots (talk) 15:31, 9 March 2021 (UTC)[reply]
  13. Oppose - I am one of those “legacy” admins, having held the mop for more than 14 years. On one hand, it would not greatly affect me if I did not have access to the admin tools. I have performed just 34 admin actions in the past year. On the other hand, I would have to spend time requesting admin actions in such cases, and the remaining admins would have to deal with them. Not a big deal, but putting just a little bit more burden on others. I see this proposal as aimed at winnowing out “legacy” admins. If such admins are removed, what admin work they currently do will be shifted to the remaining admins (I don’t see this system encouraging new applicants for adminship). So, a lot of bother to reduce the number of admins and increase the workload on those remaining. - Donald Albury 16:19, 9 March 2021 (UTC)[reply]
  14. Oppose. While a good idea in theory, what stands out the most to me is the sheer number of reconfirmations that will have to take place: 15 per week for 78 weeks when the last time the mean number of "full" RfAs was more than one a week was nine years ago (see User:Ivanvector/RFA statistics § Table 1: trends in RFA) would be a definite shock to the system. This number does not take into consideration the number of new RfAs either, and we may have an 18-month period where new candidates are unwilling to put themselves forward for fear of adding to that burden. Moreover, the problem of "legacy" administrators taking actions contrary to accepted practice is overblown: this example shows that the pressure of the community is effective without the need for reconfirmations. The right course of action, in my opinion, is overhauling the inactivity procedure instead, such as basing it solely on administrative actions (please   mention me on reply). Sdrqaz (talk) 16:31, 9 March 2021 (UTC)[reply]
  15. Oppose as written. An active admin will close many contentious XfD discussions, RfCs, and other disputes. In some of those instances, it will inevitably be the case that some number of disputants will feel wronged by the admin. The more dedicated an admin is to the job, the more of those situations are likely to crop up. "Blocked within the past three years" is not a strong enough deterrent for spite votes. Layered on top of that is the potential for a mass-discussion event to be gamed and dominated by an observant few with an agenda to disrupt. BD2412 T 16:33, 9 March 2021 (UTC)[reply]
  16. Oppose. No showing of need, and would divert community energies from the productive work of improving the encyclopedia. Neutralitytalk 16:50, 9 March 2021 (UTC)[reply]
  17. Maybe if we were starting from scratch. But a proposal to start 876 reconfirmations is plain irresponsible. I'd rather see that effort go into content - especially when we have stuff like WP:URFA. --Rschen7754 19:06, 9 March 2021 (UTC)[reply]
    I will add to this because it irks me to no end. Quite frankly - were this to pass I would not be retaining my adminship and would seriously consider retiring entirely. Would I pass? Most likely, since I was a steward and nothing was brought up about anything I did then or have done since - but my own mental health would not stand up to 200+ voters voting up or down and some casting aspersions and ad hominems (some even offwiki) that basically go unchecked. I am not going to put myself or my family through that. The silver lining would be that it would at least give me an excuse for retiring besides "WMF sucks" and "life got busy". Even the fact that this proposal exists sends an unhealthy "you're not wanted" vibe and I suggest that @Worm That Turned: withdraw it for that very reason. --Rschen7754 01:55, 10 March 2021 (UTC)[reply]
  18. Oppose. COI: Admin for over 11 years. We need more new admins, but us oldtimers are not in their way. The proposal is likely unworkable because of the volume of reconfirmations required and it holds a risk of discouraging and losing admins and making admins risk averse in case we upset the wrong popular editor. I think the fear of inactivity is overblown: I didn't make an edit per day on average during 2013-14 and I had gaps of a few months, but I returned to editing and adminning. By 2009 most of the major policies and guidelines were in place and RfA was hell week rather than a "no big deal" shoe in. The edge case of admins given the role by fiat in the early days who suddenly return and start blocking people doesn't justify the disruption this will cause. Fences&Windows 19:24, 9 March 2021 (UTC)[reply]
  19. Oppose. As an editor who has been here 15 years (but never an admin, and not very active I admit), I recognize sometimes we have admins who become out of touch, or get battle fatigue. Let's get rid of those when there is a problem, whether via arcom or via a community deadminship process. But even more, we have the challenge of ever-growing instruction creep and bureaucracy. Sending 10-year tenured admins automatically to a reconfirmation RFA will invariably become a game of gotcha, forcing them to prove they've kept up on regulation 312.6(b) or criterion A42.1(rev. c). But if an admin's been wielding the mop uncontroversially for 10 years, frequently or sporadically, and sensibly staying out of areas where they're not up to date, that's great and let them continue mopping with minimum fuss. Conversely, if an admin's behaving like a bull in a china shop, and can't seem to get their act together after reasonable feedback, then we need to remove the bit now, not wait for their 10 year anniversary. Martinp (talk) 19:39, 9 March 2021 (UTC)[reply]
    Adding: I'm not opposed to moving to term-limits on +sysop in general. It would be a big shift, but I could see benefits for e.g. 3 year (+-) appointments requiring reconfirmation (and we'd have to figure out the transition plan). I actually have a lot of faith in the community's ability to support admins who have done a good job but accumulated grudge-bearing enemies. But however complex it would be to get there, a term-limited adminship approach to be effective should have much shorter terms, to position it as "is the admin doing good mopping effectively?", not as a gotcha for tenured admins to have to "re-prove" they know everything. Martinp (talk) 22:55, 9 March 2021 (UTC)[reply]
  20. Oppose but would be happy to change and strengthen inactivity rules. The worry of a mostly-inactive admin doing something crazy is valid, but can be fixed that way. A bad-but-active admin, however, should just have their adminship bit removed by the normal procedures rather than hoping they'll be shamed into not re-applying or something, which is likely to catch just as many okay-but-controversial admins. SnowFire (talk) 20:35, 9 March 2021 (UTC)[reply]
  21. Oppose, we need a policy that singles out those legacy admins from a bygone era which have actually gone past their use-by date, not one that just throws everyone at the wall and hopes the good ones stick. Devonian Wombat (talk) 20:54, 9 March 2021 (UTC)[reply]
  22. Oppose on privacy concerns. I was forced to rename after my RfA due to harassment that spilled off wiki. While I believe the person and their socks are gone and know the link between the two isn't particularly hard to find, would not want to have to advertise it and have the old user name inevitably brought back up in a week long show and tell. After a few quieter years, I returned to being a mostly active admin in this last year and wouldn't want to have to give up the mop to preserve my privacy. StarM 22:01, 9 March 2021 (UTC)[reply]
  23. Oppose: Sorry, but to me, this is one of the worst recall procedure proposals I've seen so far. Implementing it would lead to an unmanageable amount of unnecessary reconfirmations, a lack of thoughtful responses to each of the requests, an unhealthy amount of pile-on opposing or support, desysopping of completely unproblematic administrators and an unduly strong influence on the process by the few users who take the time to "review" all of the requests and manage to write the first comments. ~ ToBeFree (talk) 22:20, 9 March 2021 (UTC)[reply]
  24. Oppose – I'm deeply sympathetic to the desire for increased community control over admins, and but this surely is not the way to go about doing it. In brief: 1) Fifteen RfAs a week is clearly not feasible. It will lead to fewer and lower-quality !votes. 2) There's no way to desysop bad apples (aside, of course, from ArbCom) without waiting for up to a decade. This hardly promotes accountability, and the root problem (the community can't desysop admins who have lost community trust) remains largely unresolved. 3) Good admins would be disinclined to make hard calls (particularly at the end of the decade), and many would simply decline to go through another RfA ordeal. The admin corp would shrink substantially, which is the last thing we need right now. I could go on - there are many clear issues with such a proposal, and the arguments listed at WP:PERENNIAL are in my view very convincing. Something more carefully calibrated than this proposal is required. Extraordinary Writ (talk) 22:43, 9 March 2021 (UTC)[reply]
  25. Oppose This would be selection bias on steroids—someone willing to research and carefully consider several confirmation RfAs per week is very unlikely to be helpful in that role which should be filled by editors engaged in article building and who have a dispassionate view about administrators. An admin who did useful work by blocking time-wasters and POV pushers would attract a wave of "good faith" opposition while those who benefit from the admin's efforts wouldn't even notice the RfA. Johnuniq (talk) 23:00, 9 March 2021 (UTC)[reply]
  26. Oppose this will cause semi-active admins to lose their tools because they are not around enough to pass people's RFA criteria. Furthermore, the large number of polls will take up community time that would be better spent improving the site. Instead of these polls, we should increase the number of admin actions required per year that are necessary to keep the tools. Z1720 (talk) 23:04, 9 March 2021 (UTC)[reply]
  27. Oppose I'd like to support the main proposal but I can't. Having 1-2 of these per week as a steady state is fine, but 15 of these per week, every week, for a year is an overwhelming burden on the community and will cause editors to burn out. The initial batch has to be done some other way. Possibly a lighter-weight poll so the non-controversial admins can skip this round, possibly smoothing it over 5 or 10 years, I don't know. I do support allowing a trial so the next proposal will be better informed; perhaps this RFC can determine that admins are allowed to stand for re-confirmation. power~enwiki (π, ν) 23:07, 9 March 2021 (UTC)[reply]
  28. Oppose Absolutely not, there is not such need for this proposal. I don't think we should be moving the goalposts because there have been some administrators lately that are out of touch with the community and administrator expectations. When they have realized their mistakes they have appropriately resigned or corrected their actions. Plus, there are plenty of appropriate venues to handle such concerns regarding admin activity and behavior. This proposal would offer no help to Wikipedia and further discourage future RfA's. Looking at Desysoppings by month we have a clear problem on Wikipedia with a slow but steady decline of admins and this proposal would exacerbate that problem. JayJayWhat did I do? 23:44, 9 March 2021 (UTC)[reply]
  29. Oppose. I agree with my fellow editors above who point out that the "initial implementation" stage of having 15 reconfirmation RfAs per week for half a year presents an excessive burden on the community for very little gain. Furthermore, even if we leave out the initial implementation, I am even more skeptical of the idea of reconfirmation RfAs than I am to the idea of a community-based desysop process. It is no secret that a lot of us believe the RfA process is "broken" because of a variety of issues: for example, it overemphasizes trivial issues such as edit counts and raw lengths of tenure, fails to require corroboration for any allegations or claims, suffers from a "pile-on" effect that can irrationally magnify the impact of recent controversies, and is often simply an unpleasant and intense experience that can and has led to good editors leaving the project for lack of enjoyment. These issues will continue to exist, if not worsen, with a reconfirmation RfA process, as an important administrative task involves resolving controversial disputes. The administrator may experience a sort of "chilling effect" in which they are unlikely to make the right call if the decision is likely to be controversial, as it may negatively affect their reconfirmation RfA.
    In essence, my view is that the status quo is not so bad that we need this proposal. If an administrator is truly engaging in misconduct that causes us to lose trust in them, we already have an evidence-based process that can result in that administrator's removal. If an administrator is inactive for a long time and fails to keep in touch with the community, we have inactivity provisions that can result in that administrator's removal—if those activity requirements are too low, then we should first attempt to raise them before we attempt a radical change like term limits/reconfirmation RfAs. Mz7 (talk) 00:15, 10 March 2021 (UTC)[reply]
  30. Oppose This wouldn't affect me for quite a while since I have been an administrator for less than four years, but I believe that this would create more problems than it would solve. I have spent a lot of time for over a decade studying and thinking about each RfA that comes along, but it would be very difficult for me or most editors to pay sufficient attention to a lengthy flood of reconfirmation RfAs at such a high volume. I truly fear that many competent and fair administrators would retire rather than run the gauntlet. Recently, a few barely active administrators have done clearly inappropriate things, failed to self-correct, and have been criticized so pointedly that they have resigned. I favor significantly higher participation standards, specifically requiring X number of administrative actions per year, with X being much higher than a handful. I think that each of us should let our ArbCom members know (without badgering) that we want administrators held to a high standard and that those who refuse to self-correct should have their mops removed. I also favor a community desysop procedure, which I believe would help build trust between administrators and the editors who aren't. I know that some vocal editors disagree, but I think that the rapid and unnecessary retirement or removal of many administrators in a short time frame would be a real blow to the project. Cullen328 Let's discuss it 01:44, 10 March 2021 (UTC)[reply]
  31. Oppose. I agree completely with SoWhy, and also with Rschen7754. In fact, the vast majority of 10+ year admins have not only kept themselves informed of current standards, they are usually the ones helping to develop them and to enforce them. It's a slap in the face to those of us who have, in fact, worked hard to keep ourselves informed and educated. It sounds like a make-work project for bored bureaucrats, because it doesn't help the community (who has time for all those polls?), it doesn't help the admin corps, it doesn't ensure that there are sufficient admins to do the work. In reality, it will be highly destructive, without any upside. Anyone who has time to take part in 800 "retention of sysop polls" over 15 months...frankly isn't here to build or support an encyclopedia. Risker (talk) 02:23, 10 March 2021 (UTC)[reply]
  32. Oppose Per @Risker. This will hurt long-standing admins that are valuable to the project, as they will have to go through a process that real-life may prohibit them from having the time to go through, and may cause them to leave rather than do this. There are former daily-user admins on this site who are now active 1-3 times a year, yet do admin actions properly and are trustworthy and knowledgeable. I also fail to see an advantage - admins that engage in untrustworthy behaviour are warned, dragged to ANI, dragged to ArbCom, and/or desysopped under Level I/II as appropriate. There is already an inactivity policy; I believe that is sufficient. Steve M (talk) 02:54, 10 March 2021 (UTC)[reply]
  33. Oppose this is solving a non-problem. We have an inactivity policy already - I'm not convinced that there is a scourge of 2009 admins who are messing things up. Elli (talk | contribs) 04:02, 10 March 2021 (UTC)[reply]
  34. I believe the way to correct admin's behavior is to give them rational feedback (which we are extremely poor in), not to establish fixed term.--Ymblanter (talk) 07:44, 10 March 2021 (UTC)[reply]
  35. Oppose My first thought was to support this, but I think we have a robust enough process to get rid of the bad admins, regardless of their length of service. I can think of several admins with 10+ years service who don't really contribute to WP in any meaningful way (couple of edits a year, if that). So they might be admins by their RfA, but not in reality. Anyone not keeping up with the times either gives up the tools, or gets the ANI treatment. Lugnuts Fire Walk with Me 08:19, 10 March 2021 (UTC)[reply]
  36. Oppose. Length of tenure bears little relationship to whether an admin is a bad admin, inactive admins do no harm (and are no more inherently susceptible to being hacked than any other admins) and being out of touch is the most easily remedied of problems. Term limits will be counterproductive in insulting good admins with long service—this is a volunteer position and often a thankless one—and, like removing tenure protection elsewhere, will encourage the ambitious to make decisions based on popularity. We need admins who do not view adminship as a political position and who are not afraid to do what they believe to be right. Provide a path to the community withdrawing adminship from those who have lost its confidence without putting all concerned, not least the admin, through the torture of an Arbcom case, and those admins who no longer care that our trust is how they became admins will, I hope, get enough of a shock to shape up, and failing that, can be removed. That's what we need. This is aimed wrongly and will cause more damage than it helps. Yngvadottir (talk) 08:54, 10 March 2021 (UTC)[reply]
  37. Oppose The user who proposed this obviously didn't read WP:IDONTLIKEIT. This "reconfirmation" RfA is, to me, a bad idea. The only problem I have is users like User:Thunderboltz who has not edited in two years but retains the bit by deleting April Fools vandalism. 🐔 Chicdat  Bawk to me! 11:19, 10 March 2021 (UTC)[reply]
  38. Oppose: waste of time and energy that can be used for better things. Unless there is evidence that admins from 2009 or earlier are causing problems, I don't think I can support, for this then would be a solution for a non-problem. We do have an ArbCom to deal with any bad apples, and this decidedly tedious process is unnecessary: adminship is WP:NOBIGDEAL and this is not a position as dangerous as, say, an arb or steward. Few gains from a tedious process that itself could be a source for many problems. JavaHurricane 12:54, 10 March 2021 (UTC)[reply]
  39. Oppose per many of the reasons above. The identified problems will not be fixed by term limits. Admins are more like cops or social workers then elected officials. They are selected by appointment for having requisite qualifications. Besides, in the real world, term limits are generally a disaster and result in a loss of institutional memory and then a lot of reinventing the wheel. Admin tenure really isn’t a problem that I see; no issue of old admins mostly hanging on and not making room for new people. And it’s a solution in search of a problem—Folks get tired, folks quit. I know any number of talented Wikipedia and administrators who have left the project, and we have a shortage of new people who want to come on board partly because the RFA process has been so toxic. I do think that a desysop for inactivity should be kept, and perhaps tweaked a little. Maybe there needs to be the equivalent of “continuing education” where admins have to stay up on things in order to keep their credential. I also think that there should be a system to consider general desysops that is separate from ArbCom. Too often it has become a place for the disgruntled to flip the board over and scatter the pieces. At its best, it is a place to deal with long-term systemic problems with certain users. It is not a good venue to assess whether an administrator is making proper use of the toolkit. Montanabw(talk) 16:43, 10 March 2021 (UTC)[reply]
  40. Oppose as a measure that would waste the community's time, and doesn't really deal with a valid problem. As an admin myself for a little over 10 years, I don't feel things have changed all that much in that time. Maybe from 15 years ago they have, but if there's an issue to be addressed, it's the removal of admin privileges from a small minority whose admin rights are causing a problem to the project. That's quite a small nut, and this is a sledgehammer that might crack it, but would result in even fewer good admins. --Michig (talk) 19:09, 10 March 2021 (UTC)[reply]
  41. Oppose While I understand that many things change over time and have sometimes witnessed some long-time admins who would correspond to the criteria, the community lacks active admins and there's an existing process that can lead to the loss of the privilege when it's blatantly misused. There also are established long term admins who should not be forced into review or RFA again. —PaleoNeonate19:24, 10 March 2021 (UTC)[reply]
  42. Oppose Announcements on major policy or guideline changes are made monthly at WP:ANEWS, which is posted at WP:AN and on subscribers' talk pages. Admins elected in Wikipedia's early days do not have noticeable behavior problems, so there is no benefit in getting rid of them. –LaundryPizza03 (d) 19:37, 10 March 2021 (UTC)[reply]
  43. Oppose We would create a lot of work, and probably lose quite a few experienced admins, for little if any discernible benefit.-- P-K3 (talk) 19:39, 10 March 2021 (UTC)[reply]
  44. Oppose. This is a perennial proposal, and I believe the reasons given there for not doing it are still valid. I'm also not sure what the problem we are trying to solve is other than "things are different now". 331dot (talk) 22:20, 10 March 2021 (UTC)[reply]
  45. Mild oppose. Part of me wants to support, simply because everything remotely like this (ideas for community desysop, RfA reforms) always fails when it gets proposed, and I'm tempted to be tetchy because this looks like it will be rejected too. But (takes a deep breath, decides to try to be grown-up), I don't believe that we should do this automatically to admins who have been doing good work for a long time. Agreeing with SilkTork's oppose, and with Wugapodes' neutral comment, I could see doing this for admins who have been inactive or minimally active for a long time. Put another way, long-time admins who continue to do good work should be left alone; long-term admins who get out of touch and misuse the tools should be desysopped for cause; and we should do more to require ongoing activity for someone to retain the tools – reasonable, not excessive activity requirements, but requirements a bit higher than what we have now, in order to be sure that they have not lost touch. --Tryptofish (talk) 00:08, 11 March 2021 (UTC)[reply]
  46. Oppose. Existing RFA's are exhausting already. With more RFA people's scrutiny will be less severe, and this is not good. If there are tons of RFA people could just vote carelessly, not vote at all, or even vote randomly. This will be bad for good admins, and the 10-year limit would just be another popularity contest. With the community de-sysop getting more traction, we as the community already have the limited ability to recall/desysop bad administrators, and we don't need this. Let bad admins be desysop-ed by the community, and let the good admins enjoy their appointment without fear.SunDawn (talk) 06:30, 11 March 2021 (UTC)[reply]
  47. Oppose. Massive waste of time and a solution looking for a problem. (Conflict declaration: I have recently ticked over 15 years as a sysop.) Stifle (talk) 14:14, 11 March 2021 (UTC)[reply]
  48. Oppose. RfA is pretty well broken at the moment, so I would not support any more RfA-like processes unless there is a dire need. --Jules (Mrjulesd) 16:07, 11 March 2021 (UTC)[reply]
  49. Oppose. Administrator misconduct can be remedied by going to ArbCom. Inactivity is remedied through existing policy. It's unclear what issue(s) this proposal is meant to remedy. SuperMarioMan (Talk) 18:59, 11 March 2021 (UTC)[reply]
  50. Oppose. While I appreciate the idea, it's unnecessary. If an administrator behaves inappropriately, their behavior will be raised by the community and they will be desysopped at the appropriate venue. There's no need to periodically re-evaluate a person's suitability as an administrator through voting as they are indirectly evaluated during their time as a sysop. Periodic votes assume administrators are behaving erradically without their behavior being raised by anyone. Inactive administrators are already desysopped. This is merely a waste of our time. Heymid (contribs) 19:46, 11 March 2021 (UTC)[reply]
  51. Strong Oppose. I appreciate this idea, but it seems like it'll waste our time. It vaguely reminds me of a retention election like we have for judges here in Florida. That system is broken, too; as everyone just votes "yes". -- Rockstone[Send me a message!] 00:51, 12 March 2021 (UTC)[reply]
  52. Oppose. My wife would be ecstatic if this passed, because I would then spend less time on Wikipedia. The proposer says ”whilst the vast majority (of admins) have adapted with the times or disappeared to inactivity, there are still some who remain doing the bare minimum ....." That’s me, firmly entrenched in the bare minimum camp. I visit daily, but briefly, check my watchlist and make little edits which I am sure have helped the project a great deal. I need the tools for many of those edits. I am not harming anyone, or the project. I edit things that require attention, not to a numerical requirement. The proposer says the "community should decide what the activity requirements or confidence level should be". That will be the day. I am one of the old fart admins and if anyone tells me they have created “activity requirements” for a volunteer like me, I will wave goodbye to them as I resign the bit and from Wiki too. Moriori (talk) 01:22, 12 March 2021 (UTC)[reply]
    @Moriori: There are activity requirements for admins: Since June 2011 there has been a requirement to make at least one edit or logged action every 12 months - see Wikipedia:Inactive administrators. However this proposal is completely independent of activity requirements - you would need to reconfirm your adminship every 10 years whether you make 1 edit a year or 1000 edits a month. Thryduulf (talk) 04:12, 12 March 2021 (UTC)[reply]
    @Thryduulf:, give me credit for some clue -- I know we have the requirement for one edit a year. Did you actually read the No 1 support for this proposal? It’s from the proposer and it states the "community should decide what the activity requirements or confidence level should be" (for admins). That thin edge of the wedge rings alarm bells for me. In case anyone forgets, our current WP:ADMIN policy says admins "…are never required to use their tools". If we overturned that policy to compell volunteers to make X edits a year I would be outta here pronto. Moriori (talk) 22:58, 12 March 2021 (UTC)[reply]
  53. Oppose. Lots of work for little gain. What is needed isn't a blanket process to force older admins to re-run RFA, but a process to better weed out bad admins. This proposal would only mean that admins with less activity, but who still contribute good work will be phased out if they don't want to devote a week to a RFA process on their record. ---- Patar knight - chat/contributions 01:58, 12 March 2021 (UTC)[reply]
    Is 1 week every 10 years really too much to ask? Thryduulf (talk) 04:13, 12 March 2021 (UTC)[reply]
  54. Oppose per SoWhy. We need more admins, not fewer. GABgab 06:11, 12 March 2021 (UTC)[reply]
  55. Oppose: this just seems like make-work. Let's tighten the activity requirements instead. --K.e.coffman (talk) 06:45, 12 March 2021 (UTC)[reply]
  56. Oppose When an admin is shown to be mis-using their extra tools, we can handle that on a case by case basis. I don't see what benefit there is to making admins beg for their tools ever so many years. Pointless idea. --Jayron32 12:09, 12 March 2021 (UTC)[reply]
  57. Oppose I don't think we have an issue with admin accountability. We have ways in place to deal with problematic admins. My arguments are the same as many above: solution in search of a problem, we really have too few admins, would be a ton of bureaucratic work, etc. Jip Orlando (talk) 14:01, 12 March 2021 (UTC)[reply]
  58. Oppose of course. Why eliminate capable admins on a time basis. We're all dying off faster than we can be replaced anyway. We have effective methods for weeding out those who need weeding. It is, once again, a solution in search of a problem. Per Patar knight, "What is needed isn't a blanket process to force older admins to re-run RFA, but a process to better weed out bad admins." And we have that. --Deepfriedokra (talk) 21:48, 12 March 2021 (UTC)[reply]
    Oh good grief. I'm a "legacy admin". One cannot judge an admin based on when they RFA'd. The few bad ones can be removed by existing processes. The many good ones get lumped in with the bad for no good reason. --Deepfriedokra (talk) 21:53, 12 March 2021 (UTC)[reply]
  59. Opppose as a solution in search of a problem. We don't really have very many problems with "legacy admins", certainly not enough to warrant something like this. Perhaps a tightening of the activity requirement is in order, but nothing anywhere near this scale. — csc-1 23:06, 12 March 2021 (UTC)[reply]
  60. Oppose per what's already been said in previous discussions such as at Wikipedia:Perennial proposals#Reconfirm administrators namely (1) there would be many uncontroversial admins that would clutter it (2) admins who are busy in real life/a bit inactive now may be lost for the future with this since they may not want to go through the poll at the time (3) dealing with those that are problematic as opposed to someone who became a bit unpopular due to a small conflict say would be difficult due to the clutter of uncontroversial ones and (4) if someone is problematic we should deal with it now rather than a person on their 8th or 9th year being left for a year or 2 people not bothering since they know the poll will be coming up soon. So yes I agree this would help to get rid of a few bad admins it would also likely get rid of some good ones to as well as being an expensive process. Crouch, Swale (talk) 18:17, 13 March 2021 (UTC)[reply]
  61. Oppose. I'm not the best person to ask, I'm not up to speed, but I haven't seen a good response to SoWhy's objections. (Which kind of supports his points ... do you really want to be getting rid of people who have managed for years to do effective work that requires the admin bit, and have managed to stay out of trouble, just because they haven't kept up on everything that might be asked at an RfA? Have you considered the cost?) - Dank (push to talk) 22:33, 13 March 2021 (UTC)[reply]
  62. Oppose. The fact that people invest way too much significance in adminship is a problem. This is not the solution. The sysop bit should be more widely granted, not less. Curtrent RfA criteria are "requests for sainthood". Fix that, not a non-problem of long-standing admins with no evidence of issues. Guy (help! - typo?) 23:10, 13 March 2021 (UTC)[reply]
  63. Strongly oppose: This is a "solution" for an imaginary problem, and the proposed "solution" itself will create real problems. You don't get rid of good admins, doing good work because they fail what is already a popularity contest; this will only lead them to being replaced by inexperienced editors who will do work of unknown quality. The experience someone builds up working in an area is valuable.
    I'd agree that inactive or minimally active admins that are not contributing as admins, should be removed and there is already a process for dealing with bad admins. This proposal would only create new problems, allow existing problems to fester, and generally add to the "Lord of the Flies" atmosphere that is all to common in some areas, leading to a deterioration of the environment and the coresponding loss of good editors that don't want to work in chaos. It will lessen the quality of the encyclopedia not increase it.
    If the concern is a concentration of power in an old guard, the answer is to broaden the admin pool with fresh blood, not narrow it by replacing experience with inexperience.  // Timothy :: talk  23:52, 13 March 2021 (UTC)[reply]
  64. Oppose: No compelling evidence has been presented as to why this is necessary, and would therefore be instruction creep and overbureaucracy. Regards, User:TheDragonFire300. (Contact me | Contributions). 01:48, 15 March 2021 (UTC)[reply]
  65. Oppose: Per Risker and TimothyBlue. There is a problem with editors and admins who don't know what they don't know. Add: To me this is illogical. Why assume long-term admins are not aware of change? So the new admin coming in is? How is it that a new admin is more knowledgeable than one with more experience. Implicit in this idea is that a more experienced admin is somehow not able to keep up. Keep up with what? Wikipedia is a big place and it takes a fair amount of time to understand it's workings. IMO change needs to be implemented in other places. One at the RfA level. Not every admin has the same capabilities. I would propose categories of admins. And I would require that each of those categories require different qualifications. Some might be technically driven. Others people driven. possibly with a category for highly experienced admins with multiple abilities. Some RfAs would then be simpler while others more like what we have now. I am actually concerned about the lack of experience in some of our " bodies". Wikipedia has changed since it began. the old rules don't work very well work in some places. Littleolive oil (talk) 22:53, 15 March 2021 (UTC)[reply]
  66. Oppose. When we get to a point when there are so many admins that there is nothing to do, then we can happily solve the problem of too many admins. But somehow I don't think this is a problem that needs to be solved right now. --B (talk) 03:09, 16 March 2021 (UTC)[reply]
  67. Oppose Solution in search of a problem. Will just lead to wasted time and drama. -DJSasso (talk) 15:01, 16 March 2021 (UTC)[reply]
  68. A problem in search of a solution --Guerillero Parlez Moi 21:00, 16 March 2021 (UTC)[reply]
  69. Strong Oppose. Full disclosure: I am approaching ten years of adminship (wow, time flies). My concern is fourfold.
    (1) I worry about what admins, such as myself, who use the tools only sporadically, but enjoy logging-in here as a hobby and pitching-in when, and as, able (while being certain to stay abreast of current policy before jumping in anywhere) might be subject to; what are the standards? Am I helpful enough? Do I still "need" the tools? Folks may oppose for a variety of reasons (and, by the way, even if frivolous opposition were discounted, the subjecting of long-time editors to it at all will prove a grave danger to editor morale, I suspect), and it seems antithetical to the spirit of Wikipedia volunteerism to subject long-time admins to what seems to me could well become a de facto timed obligation to contribute a certain amount and in a certain way. I enjoy contributing, and working the background admin functions, but this is a casual hobby for me — and it may not ultimately be worth doing RFA again (and my experience with it was positive the first go — I passed unanimously). The message it sends, when there already exist processes in place to sack abusive admins (and are others actively under consideration), is that if one isn't willing to participate in a mandatory drama board, her or his Gnomish help is unwelcome. I sincerely do not believe this is Worm's (or anyone else's intent).
    (2) is related to (1). As with any consensus-building endeavor on Wikipedia, it will produce drama, in a way that may well prove uninviting to editors, such as myself, who avoid it at all cost. Perhaps that means I'd have an easier go of it, but the focus of discussion on longtime admins who've involved themselves deliberately in controversial areas — not in search of drama, but in an endeavor to mediate the building of the content (in many cases, and genuinely problematic admins on the other hand as well, to be sure) — will become a forum for axe grinding for those most dedicated to participating in such a large volume of discussions that will be more off-putting than the role (and the process for stepping into it) might already be. To me, some level of drama is probably the unavoidable, and foreseeable, result of managing such an extensive resource for content in a manner that balances the competing perspectives of a large number of parties — but, in this case, I fear it is simply a contributing factor in a process which would cause implacable harm to the community. The inevitable drama calls for a positive payout at the other end to warrant it, and I do not see it in this instance. The best-case scenario, it seems to me (perhaps pessimistically), is a "wash" — the positive aspects of subjecting sysops to a vote are nullified on the other end by the negative (rational decisions and actions, including the shying away from the more trying and involved admin areas, which I think are already deprived of reasonable minds in many instances, made with an eye towards securing a drama-free "re-election"), with a net negative outcome driven by the tangles and snags of the process itself.
    (3) The proposed process, I think, diverts resources at RFA (which, by the way, is far, far different than what the process was in 2007, when we did see such a high volume) from the recruitment and thorough, thoughtful evaluation of new candidates, to an incessant conveyer of bureaucratic, task-laden drama. The former, I think, should be the focus of energy and attention.
    (4) The intended simplicity of these decennial polls seems to me to belie the investigation and thought which should probably be put into voting in them, on account of the variety of roles sysops (and editors generally) fill here, and the lack of familiarity and appreciation many of us might have for the work others spend their time doing — this drive-by concept strikes me as deeply problematic, and threatens to reproduce that fundamental flaw of RFA on a very broad scale.
    In spite of all this, I thank Worm for the thought he's clearly put into the proposal, and for stepping out and putting his idea forward. That's not an easy thing to do. Tyrol5 [Talk] 00:02, 17 March 2021 (UTC)[reply]
  70. Oppose Would only waste time and add more bureaucracy. A (hypothetical) disruptive admin shouldn't have 10 years to cause problems before being removed, so if the Arbitration Committee isn't up for the task in your view then a more precise mechanism would be better. Uses x (talk) 10:35, 18 March 2021 (UTC)[reply]
  71. Oppose: sorry, but I don't see what this would solve other than to reduce the number of admins. RFA is toxic enough the first time, why should someone be subjected to it again when the process will just bring out all those with a grudge for some mistake (minor, real or perceived) the admin made years ago? If an admin has failed to remain current and isn't editing, they will lose the tools for inactivity. If an admin is abusing the tools, well there are mechanisms for them to be removed already. Regards, AustralianRupert (talk) 10:08, 19 March 2021 (UTC)[reply]
  72. Oppose With no disrespect whatsoever toward the proposer, this is a terrible idea. People get desysopped all the time; there are procedures for that, and WP doesn't need to lose nondisruptive admins otherwise. This would be an incredible waste of time and energy, and would cause far more harm than good- it tries to solve a "problem" that doesn't need solving, and instead creates more problems. -- Kicking222 (talk) 10:18, 19 March 2021 (UTC)[reply]

Neutral

edit
  1. The sheer number of these that would be required at first is rather concerning (hence neutral), but beyond that I've got no real problems with the idea of term limits. I would prefer that we just fix the inactivity criteria directly though. I don't have time to think through the following idea, but I've been thinking that after N consecutive inactivity notifications we require a reconfirmation RfA (idk, maybe N=2 or N=3 since if I remember right that would put it around 4 to 6 years of 1 admin action per year). Having watched BN for a while, this seems like it would result in a handful of discussion per month (spreading out the start-up cost) and makes the criteria hard to game since you've essentially got to return after a half-decade absence or give the community a good reason not to yank the tools. Just spitballing here though. Wug·a·po·des 01:50, 10 March 2021 (UTC)[reply]
    I like this idea Wug·. Draw it up as a proposal and I'd support it. SilkTork (talk) 10:08, 10 March 2021 (UTC)[reply]
    But I'm lazy I threw my comment alongside a link to the apparently similar Commons policy onto Wikipedia:Requests for comment/Change to sysop activity requirements for drafting. Anyone's free to go to town on it (no, seriously, just make changes) Wug·a·po·des 00:52, 11 March 2021 (UTC)[reply]
    I feel like whenever I've been subject to a quota outside Wikipedia I've sometimes preferred to get the quota out of the way so I don't have that looming over my head. For this reason, I think quotas are usually bad. It forces people to make a set amount of actions, which can lead to bad actions, too. More regular inactivity durations seem a bit more like a quota? ProcrastinatingReader (talk) 12:17, 10 March 2021 (UTC)[reply]
    Sounds like a good idea.--Vulphere 15:07, 10 March 2021 (UTC)[reply]
    Agree that tweaking the inactivity desysop is useful. Probably a combo of time and a reasonable quota that doesn’t encourage gaming the system but doesn’t reward the driveby making two vandal blocks on April Fool’s Day. I also was thinking about a “continuing education” concept as well. Stuff does change and admins need to be current. Montanabw(talk) 16:47, 10 March 2021 (UTC)[reply]
    This is similar to how it is done on Commons. You get the first warning and get a chance to make 5 needed admin actions in 180 days. If you do this, you keep the mop. In after 180 days you have a second warning, you lose the mop. This works to some extent, though of course if someone is determined to game the system they can easily game the system.--Ymblanter (talk) 18:38, 10 March 2021 (UTC)[reply]
    This is a palatable option. --Izno (talk) 19:00, 11 March 2021 (UTC)[reply]
  2. Support the principle. That adminship is a lifetime appointment is largely an historical quirk. Wikipedia was in its very early days when the concept of "administrators" came about. Nobody could have predicted that Wikipedia would reach the age of 20, contain millions of articles, and come to be read and relied on by millions of people, so there was no need to consider how long somebody would serve in that function. If it were invented today, we would probably impose a term length, because our modern concept of administrators are simply regular editors who have access to certain functions to help with the maintenance of the project, and that it is not a "good editor" award. Oppose this specific proposal solely because it would create an enormous bureaucratic overhead which has no gain for the mainspace or the readers, and because (as WSC points out above), we risk losing the "long tail" of hundreds of semi-active admins who do as much of the work as the few dozen highly active admins (have a look at WP:ADMINSTATS for an idea of what that looks like). HJ Mitchell | Penny for your thoughts? 20:32, 11 March 2021 (UTC)[reply]
  3. Ideas like this and the community desysop that is also going on, they had more of my support years ago. I'm not against the concept of limiting adminship or requiring a vote of confidence, although I'm not sure this is the only way to do it. Basically, I don't have the answer here but I do have some reservations. My main concern is the same as with many admin proposals, which is that as admin, we ruffle a few feathers from time to time, and even when we were 100% in the right and the community 100% supports the actions we took, those people are much more likely to show up to vote than people who haven't really noticed you before (ie: good editors that don't visit the drama boards). So in the end, the vote gets skewed. This is no different than the average waitress, who has mainly satisfied customers, but gets more negative feedback than positive. It is just human nature. We say nothing when all is well, we speak out when something isn't. Worse, what would end up happening is the same group of admin (and some established regulars) would end up voting in all these in support of the admin who is being polled, to offset the "he blocked me once" votes. Good intentions, but it would start to look like a cabal, and make people lose faith in the process, which is worse than not having a process. So this does sound like an Oppose, and in a way it is, but I'm open to other ideas as they develop here. I can't support unless we can magically find a way to have the votes actually reflect community consensus, and not just an angry mob + some admin defenders. (ie: a popularity contest) Dennis Brown - 21:06, 12 March 2021 (UTC)[reply]

Discussion

edit
  • 15 a week sounds like a huge amount. I would only (!)vote support for those candidates I knew well (and respected) or did my research for - same as I do in regular RfAs. But I would oppose any that I thought were unsuitable. I'm pretty active and wouldn't have time for the better part of 15 detailed reviews a week, so likely would skip many. I imagine many others will be the same. This means there will be a natural weighting towards opposes not seen when there are few enough that vetting each is not a major drain. While it will heavily increase the timescale, I'd propose, say, 7/week, and individuals who want to get it over and done with are also free to do so. I would also propose that individuals are actually allocated a month, and just required to start it some time within then. That will avoid a veritable deluge of requests to "working that week/no internet access, please let me move" - without them being temporarily losing the tools until they can process it. I would also question if we have the clerking resources to actively monitor 15 - not just for a few weeks but years, without a pause. Nosebagbear (talk) 10:46, 9 March 2021 (UTC)[reply]
    Nosebagbear I appreciate that, and we could slow it down if there is poll fatigue, without too much difficulty. The reason I went for 15 per week is that the community has managed that in the past and it gives a clear psychological deadline of "end of next year". WormTT(talk) 10:58, 9 March 2021 (UTC)[reply]
    @Worm That Turned:, what would you expect as the rough participation in these? Back when the Community was handling 15/week (obviously before my time), the number of participating editors on average was much lower. If we get a drop of, say, 1/3 in active participation between month 1 and month 3, would you say that is too much poll fatigue? Nosebagbear (talk) 11:18, 9 March 2021 (UTC)[reply]
    I've mentioned that it should be considered a light version of RfA, so I certainly wouldn't be expecting the 200 editors you get there. I'd be expecting something closer to 60 or so editors for most cases, similar to 2007 levels WormTT(talk) 11:20, 9 March 2021 (UTC)[reply]
    I think the reality is that with so many administrators being reviewed a week, most reviewers will only be able to perform a superficial review (other than those who have interacted extensively with the administrator in question). isaacl (talk) 17:01, 9 March 2021 (UTC)[reply]
    A "light" version of RFA. That's...unhelpful. There's going to be 10 years of admin actions. Seems kind of pointless to see if someone's a decent admin by doing a "light review". Otherwise, it's just a popularity contest. Risker (talk) 04:11, 10 March 2021 (UTC)[reply]
  • I'm also somewhat concerned with this being fully open that we're going to get caught between changes that people think are good and having to, repeatedly, contact everyone who has !voted. Nosebagbear (talk) 10:48, 9 March 2021 (UTC)[reply]
    I don't intend the proposal to be "fully open", the proposal above is the proposal. If you think there is a better way for me to lock it down to !votes and discussion, let me know. WormTT(talk) 10:57, 9 March 2021 (UTC)[reply]
  • I might as put my cards on the table. I know a lot of people like what I do and think I'm a good editor / admin, and conversely there are a bunch of people who think I've done some utterly boneheaded things and should be desysopped in the same way as Kudpung and RHaworth. So I don't think I've got a hope in hell of passing a reconfirmation. However, I think general idea of reconfirmations is a good one. Ritchie333 (talk) (cont) 11:31, 9 March 2021 (UTC)[reply]
    I feel like the 60% threshold is sufficient to allow leeway for the few folks with long-term grievances not to affect the outcome. I also feel like most admins who show at least some degree of competency and temperament would/should pass reconfirmation. (and if they can't, then should they even be admining?) For comparison/thought, I noticed some editors in the steward reconfirmations explicitly say they wouldn't support if it were a SE, but that 'the bar is lower for reconfirmation so keep'. ProcrastinatingReader (talk) 12:34, 9 March 2021 (UTC)[reply]
  • (Resolved) Seems to have a workflow missing: As current administrators may put in their poll earlier than their allocated date, should they do so and fail to pass the poll - they have a showing that they are no longer supported by the community, however this workflow leaves them in limbo: it doesn't actually result in a desysop - but they can no longer use the poll option - leaving them to serve out their 10 years and then have to do a new RFA? Suggested remedy would specifically including that any time that a poll is not successful results in an involuntary desysop, requiring a new RfA. — xaosflux Talk 12:08, 9 March 2021 (UTC)[reply]
    • Perhaps modify: If an individual is not successful in their poll at any point, they are not eligible for another poll, however they may submit a fresh Request for Adminship at any time. to If an individual is not successful in their poll at any point they will be involuntarily desysoped and will not be eligible for another poll, however they may submit a fresh Request for Adminship at any time. or the like. — xaosflux Talk 12:18, 9 March 2021 (UTC)[reply]
  • @SoWhy: as I understood it, ArbCom tests the evidence against the standards of WP:ADMIN, not tests if there remains community confidence in the admin? These are two distinct things, in my opinion. For example, various admins restricted in their usage of tools probably don't have community confidence (in the sense that they'd probably fail RfA)? Additionally, I suspect some things ArbCom doesn't think are a big deal/desysoppable would be a ruin if tested at RfA. Also, some behavioural aspects can probably lead to loss of community trust even though they're not WP:ADMIN violations per se. ProcrastinatingReader (talk) 12:37, 9 March 2021 (UTC)[reply]
    • @ProcrastinatingReader: Strictly speaking, yes, however, it's hard to see how any admin who has violated WP:ADMIN to the point where ArbCom considers desysopping can still retain the confidence of a community which (rightfully) expects admins to adhere to the rules set out in that policy. My very point was that at a reconfirmation RfA, people can (and will) fail their reconfirmation even when they have not violated WP:ADMIN if they have made correct decisions that angered a large enough segment of the community. However, even if you were correct, the solution would not be to require all admins to periodically be reconfirmed but instead to have a separate process to handle these admins when there is cause (which Wikipedia:Requests for comment/Desysop Policy (2021) is all about). Regards SoWhy 13:39, 9 March 2021 (UTC)[reply]
      @SoWhy: well, I meant the opposite way around. I suspect all admins who violate WP:ADMIN have lost the confidence of the community, but I think there can also be cases where admins have lost community confidence but haven't violated WP:ADMIN to the point of an ArbCom desysop. I think the desysop proposal addresses the same for-cause cases as ArbCom, perhaps a couple more (like the BLP stuff at ANI atm), but not all cases of loss of community confidence. Worm's proposal seems to be a strict test of whether the community wants someone to be admining, a bit like steward confirmations (eg the events of meta:Stewards/Confirm/2021#Rxy are not strictly a violation of any policy afaik (?), but obviously led to a loss of community confidence). I guess overall I just feel like there's a notable difference between WP:ADMIN failures and loss of community confidence. and imo I don't think an admin without community confidence should be admining. ProcrastinatingReader (talk) 13:53, 9 March 2021 (UTC)[reply]
@ProcrastinatingReader:, there's a very big difference between losing community confidence and actually doing something drastically wrong. I think SoWhy is on the right track. He might have the enviable (or unenviable) advantage of being a sitting Arbitrator, but like me, he is critically aware of the (hopefully temporary) climate of not quite such a general consensus as one might be led to believe, that a lot of admins need to be got rid of by any method or reason the community can think of and as quickly as possible. Kudpung กุดผึ้ง (talk) 14:22, 9 March 2021 (UTC)[reply]
Maybe, but who gets to decide whether something is "drastically wrong"? ArbCom only evaluates against WP:ADMIN, not against the subjective "drastically wrong". The Rxy stuff is an example of something the xwiki community thought was "drastically wrong" but (again, afaik) didn't actually violate policy. fwiw, I don't actually believe in the idea that there's rampant admin abuse or that there needs to be a great purge. I also don't think Worm's proposal will result in a great purge of good admins. I do think there are a few, though, that are sufficiently damaging and toe the line sufficiently to not reach an ArbCom case yet still erode community trust over time and are just net negatives. Sometimes they crop up at ANI, but much of the time it's not worth bringing up because what's the point?
Again, I firmly believe in the idea that any admin (as with any other permissions holder, really) must retain community confidence, and an admin whose status is in sufficient question should do the right thing and relinquish their tools or subject themselves to re-RfA. Regardless, why is +sysop so sacred? Any normal editor can already be evaluated at AN/ANI. Any now-unbundled permission can be evaluated in the same way. The vast majority of the time those venues work just fine. Why are admins and their holy bit off-limits? Between this and the desysop proposal (which I opposed but for other reasons) I've struggled to see someone present a valid, evidence-based answer to that (pure speculation doesn't count). ProcrastinatingReader (talk) 15:58, 9 March 2021 (UTC)[reply]

Why should administrators only be held to account when they have done something egregious (or wholly inactive)? Surely it would be better to have a system where feedback can be given - rather than having a long term malaise associated with a significant portion of the community. I understand that this would be a seismic shift - however, I really do think it's needed. It would encourage a re-invigoration of our community and encourage fresh blood to the admin corps. WormTT(talk) 16:04, 9 March 2021 (UTC)[reply]

There is a difference between giving feedback when feedback is needed and forcing everyone to go through a feedback process regardless of whether feedback is actually needed. Your argument would be wellplaced in support of reviving WP:RFC/U or WP:ADREV and I would probably support a mechanism that allows feedback for admins where needed but this is not it. As for "fresh blood to the admin corps", I fail to see how the lack of term limits prevents that. There is no data I know of that supports the idea that more "newer" editors would be willing to stand for RfA if only some old admins were made to go through RfA again. If anything, the months-long process would more likely bring RfA to a hold in that time, making it harder for new editors to run when the community is already lacking sufficient resources to review the old admins. Regards SoWhy 16:14, 9 March 2021 (UTC)[reply]
If no feedback is needed, the poll would be extremely light-touch. On meta, stewards are reconfirmed every year. I'm suggesting we reconfirm our admins every 10. I'm not suggesting a complex process, just a light poll - with the option of discussion on the talk page - a different venue. Too many polls? Well, I may have got my numbers wrong, it would be easy to re-assess the community input. I am far more concerned about a system that encourages issues to boil up unchecked until they reach a point where there is little or no chance of meaningful feedback (a desysop). WormTT(talk) 16:22, 9 March 2021 (UTC)[reply]
It would still be a burden for the admin who will have to set aside a week of activity to monitor the poll, wouldn't it? I also don't think the steward comparison works. There are 38 stewards for all Wikimedia projects. Numbers-wise, your proposal would be as if we reconfirmed all stewards every 2-3 weeks. Steward reconfirmations work because the number of stewards is low and the pool of editors eligible to vote is high. In this case, the number of candidates is high and the pool of likely voters is low (based on my experience, there are probably less regular RfA participants than there are admins) and, with fatigue setting in likely to get lower. Regards SoWhy 18:18, 9 March 2021 (UTC)[reply]
Comparing admins with stewards is apples and oranges. It's a red herring and there are only 38 of them. There might possibly be bad apples in every barrel but there's still no reason to systematically smoke them out (the steward reconf system is a joke anyway); we don't register every human being's DNA at birth in case they slip up later in life (although that may become a dystopian reality one of the days). This mass reconfirmation would make every admin and adminship candidate already guilty by association with the the tools - no one would want to run for the bit under such a Sword of Damocles. A fixed 10-year term is one thing, but forced reconf puts mustard not custard on the admin trifle. Kudpung กุดผึ้ง (talk) 00:02, 10 March 2021 (UTC)[reply]
  • Perhaps an initial no discussion polling round where 50 to 100 number of admins are listed each week and only those who register greater than 20% (or x number of editors) opposition to keeping their bit are put through the more in-depth process requiring 60% support to remain an admin. This should allow relatively uncontroversial admins to sail through as well as minimize the discussion phase (and all the ill will & general unpleasantness that RfA discussions generate)Slywriter (talk) 13:33, 9 March 2021 (UTC)[reply]
  • Not sure to what extent this is an aside, but upon my return to active editing and admin work after a 2008-2016 hiatus, I made some laughably-inept mistakes. From bizarre custom sanctions, to generally misunderstanding policy developments and/or exceeding my purview. I'm thankful to the community for its patience with me — for being tolerant by giving me some time to update myself, which in hindsight, I ought to have done less clumsily and with greater caution. Seems rather unlikely that I would have passed a reconfirmation during those early months, though. Of course, I still make mistakes (occasionally even spectacular ones), but, again, nothing like those first few months. For whatever that's worth. El_C 14:31, 9 March 2021 (UTC)[reply]
  • Once the backlog is cleared, will there be a plan to spread out future re-confirmation reviews so that there isn't another surge another ten years after that? This would require scheduling reviews earlier (or later) than they would be held otherwise. (Of course, the effects of a second surge will likely be reduced through natural attrition.) isaacl (talk) 16:52, 9 March 2021 (UTC)[reply]
    I did some contemplation about the surge in 10 years, and between natural attrition (we lose about half our admins in a 10 year period), those admins who decide not to stand for reconfirmation and those admins who wish to run their reconfirmation again within 10 years - I believe the surge would be significantly decreased. WormTT(talk) 19:10, 9 March 2021 (UTC)[reply]
    Hmm, well if the numbers drop to, say about 500 admins in ten years (from the current total of 1,109 according to Wikipedia:Administrators), then with an average of 15 new admins per year (somewhat conservatively estimated based on the last five years), up to 150 or 30% of the admin population will have been added during that decade. (It would be less, as some of those will have departed from attrition.) It might be simpler to just have term limits apply only to new admins. isaacl (talk) 23:22, 9 March 2021 (UTC)[reply]
    Grandfathering is definitely an interesting proposal, but I'm cautious that it may just result in a net nothing change for at least a few years. This proposal exists to try to weed out mid 2000s admins that make a few actions a year and letting them keep their tools because they became admin in the mid 2000s is inherently counterintuitive. And just to support your numbers, if we take a five year period from 2016-2020 (inclusive), we've resulted in a net loss of 224 admins. This is an average of ~45 per year lost. So, taking that, if we forced everyone to have a reconfirmation RfA tomorrow, by the time the next one would roll around we'd have 450 less admins simply through inactivity. Anarchyte (talkwork) 06:37, 10 March 2021 (UTC)[reply]
    There are various positives that having fixed terms could bring in itself, including the following: More editors may be willing to help out for a pre-arranged fixed period (yes, with today's setup, they could relinquish administrative privileges, but I feel the psychology is different when you know your term has an expiration date). Encouraging turnover helps ensure admins don't get burned out at their tasks. And it provides a checkpoint to re-assess community trust in an administrator. There are of course negatives, which others have described. If we determine that support for having fixed terms is conditional on it being phased in, that's OK. The benefits will take longer to achieve, but there will also be more opportunity to re-evaluate and adjust. isaacl (talk) 07:16, 10 March 2021 (UTC)[reply]
  • One of the reasons often cited for proposals such as this is the hypothesis that if we get rid of lots of admins lots of new ones will come forward as the community will be much more willing to trust new people with the tools and RFA will suddenly become a less fractious and more zenlike space. I have long thought the opposite, especially as the reasons for RFA being fractious are more about our disparate attitudes to both deletionism and civility enforcement. But given that various sister sites such as Commons have adopted various ways to get rid of more admins, can those who want such change now point to a sister site that has managed to prune their admin cadre in ways that have brought forth the flowering of a new generation of admins? ϢereSpielChequers 17:18, 9 March 2021 (UTC)[reply]
    I would suggest the annual confirmation of stewards run in parallel of the election of new stewards shows that such a confirmation process can work and be low key for most stewards most years. Best, Barkeep49 (talk) 17:51, 9 March 2021 (UTC)[reply]
    There are 38 stewards for the entire global community; reviewing that number of individuals in a year might be reasonable. However, I will point out that none of the recently elected or reconfirmed stewards had anywhere near the amount of support (in a voting pool that numbers in the hundreds of thousands) than an average enwiki RFA. Further, the majority of steward actions are largely invisible to those voters (checkusers, reasons for global blocks and locks, and the majority of admin actions on languages the individual voter doesn't understand) and thus the reviews are hardly comparable. The only stewards who tend to be removed are those who are comparatively very inactive and those who have conducted themselves in a controversial way. We already have ways to deal with admins like that. Risker (talk) 03:46, 10 March 2021 (UTC)[reply]
    Indeed, and there are significant differences in how that process is run. In the end, all that matters is a supervote by the stewards based on that confirmation, see m:Talk:Stewards/Confirm/2021. Opposes that are ax-grinding or are completely irrelevant - or even have no explanation - are explicitly disregarded by stewards, if other voters don't challenge them. --Rschen7754 06:57, 10 March 2021 (UTC)[reply]
  • Questions: (Q1) If I'm a long-standing admin (I'm not) and put myself up for a poll, and inside of a week only 10 (or 5?) people respond, is that good enough to decide my adminship? If not, what then? (a) extend the week, (b) default to failed, (c) default to success, or (d) simply tally the responses and retain me if three respondants approve me? (Q2) If I do get 10 responses, and 4 are "he's a swell enough chap, as I recall", 3 are "the guy's a menace to Wikipedia", and 3 more are "why not?", do I retain my adminship? — JohnFromPinckney (talk) 17:50, 9 March 2021 (UTC)[reply]
    I would be unconcerned if there is a small number who respond as we have some past RfAs which are equally small but still considered valid. However, I don't expect that we will have very small turn-out - I've explained I believe we'd end up around the 50-60 editor mark. Due to the "light" nature of the procedure, we'd be expecting firm votes - see the example - so at 7 vs 3, you'd pass at 70%. WormTT(talk) 19:10, 9 March 2021 (UTC)[reply]
  • I am unhappy with the threshold as written - "should," "60-65%," and "some bureaucrat discretion" aren't exactly...uh...exact. Frankly, given the recent trends of how the community sees 'crats, I'd rather set a hard limit and take discretion off the table. SubjectiveNotability a GN franchise (talk to the boss) 18:00, 9 March 2021 (UTC)[reply]
  • The idea of feedback is great irrespective of specific problems. Rather than !voting, which could result in possible desysoping, the process could be used to elicit feedback from the community. Vikram Vincent 19:53, 9 March 2021 (UTC)[reply]
  • I've tried to come up with some numbers on what would do to the active admin corps. There are 291 non-bot admins who have at least 100 logged admin actions in the last year [2]. 91 of those admins have passed RfA in the period 2012-present and wouldn't have to stand for reconfirmation in the near future. In other words about 2/3 of active admins would have to go through a recall RfA if this was to pass. Hut 8.5 21:23, 9 March 2021 (UTC)[reply]
    Out of interest, how many inactive admins (by your metric of < 100 logged admins actions in the last year) are pre-2012? ProcrastinatingReader (talk) 21:29, 9 March 2021 (UTC)[reply]
    Well the figure I got is 411 of 445, but I would take that with a pinch of salt as it is possible for non-admins to get small numbers of admin actions in that table (I think if you move a protected page that counts, for example) and the list doesn't include people who made no admin actions at all. But I think it's safe to say the vast majority are pre-2012. Obligatory caveat: comparing admins by number of logged actions is like comparing editors by edit count and has all the same flaws, with the added issue that not all admin activity generates logged actions. Hut 8.5 08:48, 10 March 2021 (UTC)[reply]
  • Comment - just to note that while I do oppose this, I also greatly appreciate WTT putting themselves out there in his continued efforts to bridge any gaps between editors and editors with extra buttons. A lot of people spend a lot of time complaining about admins, RfA, and any procedures to fix broken things. Discussions on how to improve things is never a bad thing - and I just want to take a moment to thank Worm for taking the time to open this up. — Ched (talk) 06:54, 10 March 2021 (UTC)[reply]

Trial

edit

Since this is still early enough in the RfC, I want to throw out the idea of maybe doing a trial with volunteers. Ideally we would get 15-30 or enough to do a couple weeks of this. The evidence wouldn't be perfect since the composition of volunteers might not reflect the community as a whole but it would give us some data about how this would work in reality which might help convince some skeptics. Under the 10 year term I'm not due for a bit so I am not eligible to volunteer but perhaps some others would be, including I suspect WTT. Best, Barkeep49 (talk) 19:48, 9 March 2021 (UTC)[reply]

  1. Of course. I'll put my money where my mouth is WormTT(talk) 19:53, 9 March 2021 (UTC)[reply]
  2. If we get at least five volunteers (i.e. at least three more than Worm That Turned and me) then I am happy to participate. If we can't get at least that many volunteers then I don't think it's worth pursuing even as a trial. Thryduulf (talk) 22:48, 9 March 2021 (UTC)[reply]
  3. Absolutely no doubt the community has changed a heck of a lot since I got my tools in Dec07, so if a trial does go ahead (per above comments and below reservations), I'm happy to be involved. Daniel (talk) 01:05, 11 March 2021 (UTC)[reply]
  4. Wikipedia:Requests for adminship/HJ Mitchell 2 was elven years ago in May, and Wikipedia:Requests for adminship/HJ Mitchell 3 (which was mildly controversial at the time) will be ten years ago on the same date. I support the principle behind ideas like this, even if I think there are details to work out in the execution, so I would be willing to be part of a trial. HJ Mitchell | Penny for your thoughts? 16:21, 11 March 2021 (UTC)[reply]

Discussion of trial

edit

I think one thing to be wary of is extrapolating how much interest will persist. There are very few ongoing initiatives unrelated to content that I can think of, and even fewer that require multiple volunteers. It's proven hard to get people to work on projects, even ones that they endorsed enthusiastically at creation time. isaacl (talk) 22:25, 9 March 2021 (UTC)[reply]

For a trial to really be useful, I'd want to see it involve some admins who have been subject to significant complaints and for whom there isn't a guarantee they'd pass. Unfortunately, I don't think it's likely such admins will sign up (and I can't really blame them). {{u|Sdkb}}talk 00:00, 10 March 2021 (UTC)[reply]

To be honest, I think how smoothly the process can handle a large number of simultaneous reviews for uncontentious admins might be more important. People raising concerns will likely attract attention and closer examination. But how much net value will we get from pro forma votes by a handful of participants for most admin reviews? isaacl (talk) 00:11, 10 March 2021 (UTC)[reply]
I'm totally against the whole idea (and I'm not alone), but I for one obviously have nothing to lose - nor to gain - by letting a trial go ahead. That said, per Sdkb for a trial to be really useful, I'd also want to see it additionally involve some admins who know full well that they are bullies and bandy around with baseless threats and/or intimidate newbies and oldies alike, but who are careful to stay under the radar. If the trial is made up only of goody-goody-two-shoes, it's worthless from the start. Kudpung กุดผึ้ง (talk) 00:31, 10 March 2021 (UTC)[reply]
Sdkb hits the nail on the head. The results of this kind of trial will not be meaningful because of voluntary response bias. Mz7 (talk) 01:06, 10 March 2021 (UTC)[reply]
  • There are two major facets that cause issues with the trial helping us see how the whole project would work. One is non-typical respondents. Actually, that's not the case, as most admins would pass with ease. Nor do I care about bullying admins not feeling represented. It's the "make themselves unpopular doing good work" group that I'd like to see represented. Non-Arb/CUOS/Crat AE admins etc. The second is that I am positive the Community will do a sterling job participating in a month (60, under this scheme) reviews. It's whether we have an ability to keep doing so month after month (tbh, I'd expect a major drop off in just the second). Assuming we don't have at least, say, 80 admins (which would be a huge trial!) I'm not sure how good a check we can do. There would be one big ancillary benefit - in the past there seems to have been reticence about admins wanting to do Confirmation RfAs. To me, if they could become more socially acceptable, which some big names running them would do, that could be useful in a range of ways going forwards. Nosebagbear (talk) 02:02, 10 March 2021 (UTC)[reply]
  • A trial would not be representative of how the procedure might run. A handful of well-known admins would agree to a trial (those who aren't well known would not be aware of the trial or would be otherwise engaged). Then a large group of mostly reasonable editors would join in and it would all look good. The question is what would happen after many weeks when reasonable participants would give up. It's quite easy for socks to reach extended-confirmed with some advance planning, and haters never forget. Johnuniq (talk) 02:56, 10 March 2021 (UTC)[reply]
I agree with Sdkb on this point that the way to have the most value in this is if we get a range of admins including some who work hot button areas. But the risk of good admins working hard areas is not the only reason I see people opposing and so not the only value I think a trial could provide. Best, Barkeep49 (talk) 02:58, 10 March 2021 (UTC)[reply]
One of my fears about such schemes is the loss of the long tail. Admins who may once have been very active power users but who nowadays we see here for the occasional evening. I don't see a trial resolving how far down the long tail this would go, any more than a trial would tell us how meaningful the 300th review would be after the initial enthusiasm had abated. ϢereSpielChequers 12:30, 10 March 2021 (UTC)[reply]
  • What's to trial? I'd rather see a proposal permitting reconfirmation RFAs than this. It won't be representative. And I cannot believe, honestly, that the community as a whole has any interest whatsoever in participating in 60 admin reviews a month. Seriously? With everyone under review having more than 10 years of admin and editorial actions, even the most proficient reviewer will be devoting a couple hundred hours a month to a process that, at best, will maintain status quo. I have no issue with people handing in the bit and running for adminship again, I have no problems with reconfirmation RFAs if people want them. But we're already weeding out dozens of admins a year who aren't active, or who aren't maintaining standards. If I wanted to join a social media site, I wouldn't have come here in the first place; there are much better social media sites. Risker (talk) 03:34, 10 March 2021 (UTC)[reply]
    If this fails then a proposal to explicitly permit voluntary reconfirmation RFAs (maybe limit them to a maximum of 1 per year per person to avoid anyone being silly with them; explicitly no minimum) would be a good idea. Thryduulf (talk) 03:57, 10 March 2021 (UTC)[reply]
    Not sure this helps with the issue this proposal aims to tackle imo. Those that voluntarily go for it are never going to be the ones that will actually fail the reconfirmation (though they might still get helpful feedback nevertheless, if they want that). ProcrastinatingReader (talk) 12:16, 10 March 2021 (UTC)[reply]
  • Read the room: this proposal is <50% support right now. A trial is not going to change things. --Rschen7754 03:53, 10 March 2021 (UTC)[reply]
    It been open only a little over 12 hours, things may (or of course may not) change. There was significantly less opposition when a trial was first proposed, so the suggestion that the idea of a trial is a failure to read the room is rather disingenuous imo. Thryduulf (talk) 04:00, 10 March 2021 (UTC)[reply]
    Agreed with Thryduulf. It was not unreasonable to suggest a trial at the time it was proposed. Mz7 (talk) 08:22, 10 March 2021 (UTC)[reply]
    My comment was more intended as a suggestion as to what to do now rather than a statement about what happened in the hours before. But given that this proposal is approaching ~33% support, a trial would be a disaster. --Rschen7754 20:02, 10 March 2021 (UTC)[reply]
    @Rschen7754: I can see why a trial might be pointless (there is little benefit in trialling something when the outcome will have no impact on the thing being trialed's chance of implementation), but I don't understand why it would be a "disaster"? What harm will come from a trial? Thryduulf (talk) 18:47, 11 March 2021 (UTC)[reply]
    My suggesting a trial was reading the room. I saw that people had concerns about something I think a good idea. So I thought to myself, what could I do to build support? Knowing that some people with reservations proposed a trial for desysop, I thought such a proposal here might be effective (especially knowing there would be at least one volunteer in WTT). And as pointed out by others it came 9 hours after proposing or early enough that I hoped a new idea might change the flow of comments and consensus. Best, Barkeep49 (talk) 15:44, 10 March 2021 (UTC)[reply]
  • Comment I am curious as to what would happen if someone with CU or OS (such as me) lost their Admin status this way - I presume that would be meant to remove those rights. Doug Weller talk 11:14, 10 March 2021 (UTC)[reply]
    • Almost certainly. To me, intuitively, one shouldn't hold enhanced admin permissions without also being an admin. That wouldn't make sense. I remember when Xeno requested to have their sysop flag revoked but still retained the bureaucrat flag (link) due to, erm, meteorological reasons, but with climate change being what it is that only lasted for a couple of months before the inevitable re- +sysop'ing again. (I don't believe the bureaucrat flag alone gave him access to, say, view deleted pages or revisions, for example...) //Opens umbrella! El_C 13:01, 10 March 2021 (UTC)[reply]
    @Doug Weller: If I were to fail at a reRFA I would resign my OS privs as without the ability to delete or see deleted revisions I couldn't make much use of it. I would suggest that such resignation would not automatically be a barrier to my requesting it back should I ever regain adminship (the committee would be under no obligation to agree, but neither would they be obliged to say no). Thryduulf (talk) 13:04, 10 March 2021 (UTC)[reply]
    Thryduulf: A non-administrator oversighter can still view deleted revisions, and suppress revisions. (For a brief time, we did have a non-administrator holding OS/CU privileges as part of AUSC.) That said, a CU or OS that cannot maintain a mandate from the community would have a tough time retaining the confidence of the committee. El_C: Absent admin (or cu/os), bureaucrats cannot view deleted revisions (and cannot remove certain redundant user rights when adding administrator). –xenotalk 14:02, 10 March 2021 (UTC)[reply]
    On the other hand, there is no actual rule that prevents a non-admin from being elected as an Arbitrator, which would lead to them getting CU/OS as part of their duties despite not holding +sysop. Regards SoWhy 16:30, 10 March 2021 (UTC)[reply]
    This sub thread puts me in mind of previous discussions about unbundling the tool kit. There are a number of admins who primarily do gnoming, they remove copyright violations, they delete articles, they revdel offensive material until an oversighter can get to it, and so on. It seems like everyone here talks about the vandal fighting aspects of being an admin, and the ability to block other users, but the truth is there is a ton of admin tools that have nothing to do with user behavior directly, and I think we need to consider that there really are two different groups of people that have the toolkit. — Preceding unsigned comment added by Montanabw (talkcontribs) 23:23, 10 March 2021 (UTC)[reply]
    I’m with Johnuniq, socks can get to EC quickly and haters never forget. In a random trial, the admins who do the tough work and make the hard calls —the ones we need the most—will be the ones who will be dogpiled by trolls. The whole “Behavioral management” component of Wikipedia is already too Lord of the Flies as it is. Montanabw(talk) 16:57, 10 March 2021 (UTC)[reply]

One-off RfA review

edit

I think a one-off RfA review to deal with the issue of the RfA from the early days could be a way to deal with those who admin status from the early days but are not what the community would consider as admin material today. Emir of Wikipedia (talk) 21:54, 10 March 2021 (UTC)[reply]

I am concerned about this whole idea that people from the “old days” are too out of it to be admins. Some of the best we have are those have been around a long time. I think that it’s more question of bringing people up to speed where there are certain deficiencies. Montanabw(talk) 23:36, 10 March 2021 (UTC)[reply]
The issue is not with people being out of it, but rather that it was easier to become an admin in the very early days. If some of the best became admins in those days then they should have no problem passing a review. I am not even suggesting the review should be binding, it could just be advisory. Emir of Wikipedia (talk) 19:29, 11 March 2021 (UTC)[reply]
Montanabw, Emir of Wikipedia, the problems I've seen tend to occur when an admin stops being/thinking about building the encyclopedia and feels obligated to protect it, which is a good thing if there are controls in place to prevent potential abuse. The abuse issues become magnified by DS/AE which allows admins to use sole discretion and take unilateral actions that cannot simply be overturned by another admin. Over time, some admins tend to become thought police, and therein another problem lies. It leads to their being "involved" because they form preconceived notions about certain editors whose POV don't align with their own. It is the creation of "absolute power" coupled with a decade of "command" and we have created the perfect storm. Prejudice against an editor is unambiguous involvement, especially if both are long term and work in controversial topic areas. Opposing views are the reason WP is neutral, but that concept gets buried under prejudice. Add to that, alliances with others of like mind, and over time it grows into a feeling of bold obligation to rid the project of editors they consider disruptive per their prejudiced POV. I see this proposal as a step in the right direction; something needs to change. Atsme 💬 📧 12:48, 15 March 2021 (UTC) Added note: I was reading down the page in proper sequence, and had not yet read what Montanabw wrote in the next subsection. My comment probably belongs there, and not here. Feel free to move it. 12:51, 15 March 2021 (UTC)[reply]
Atsme, I think you are describing a different issue here which I think is an issue but is out of the scope of this pages proposal. Emir of Wikipedia (talk) 15:21, 20 March 2021 (UTC)[reply]
Hi, Emir - I was thinking along the lines of reasons that justify polling admins after 10 years in a one-off RfA review; i.e. potential changes in behavior from when they were elected and why the community would not consider them admin material today. Otherwise, why would we need to poll them? Also see the example RfC poll and diff examples. If that is out of scope, please accept my apologies and simply ignore my comment, or hat it. Atsme 💬 📧 15:38, 20 March 2021 (UTC)[reply]
When you say it like that I can see what you mean now, but personally I think that what you are on about is a difficult situation to solve. We need to stand up to the problematic admins who you talk about, and make sure we uphold the vision of Larry Sanger and Jimmy Wales. Emir of Wikipedia (talk) 15:54, 20 March 2021 (UTC)[reply]

Recusal rules?

edit

One of the complaints that I think is worth looking at has to do with admins who over time may lose their neutrality on certain people, issues, or topics. While expertise is required, I think there are some people who don’t recognize that they’ve gotten hooked and for that reason they have to put the mop away and just interact as an ordinary editor in those areas. Rather than desysop or even just AN drama, Maybe there just needs to be a place where somebody can go to say, “Gee this admin is opposing me, and I don’t think they are being fair — could somebody else look at this?” It might need more than just other admins to reach consensus, perhaps some folks with checkuser or other enhanced tools would be the ones to weigh in on it so that it doesn’t just lead to potential wheel wars. It also could be a good thing for everyone in the current admin core to perhaps create a some page where they disclose areas where they acknowledge they they will not use the tools, at least without a second opinion (except for blatant vandalism or other time-sensitive actions) because they either have very strong feelings about a topic, a conflict of interest, or something else that may lead to loss of the objectivity. I really don’t have this idea fleshed out well enough to put up an RfC, just curious about what others think. Montanabw(talk) 23:36, 10 March 2021 (UTC)[reply]

So, @Montanabw: I don't think it would need anyone with special tools (if no need for admins or special tools, default to community decision-making), but someplace less aggressive than ANI where an Admin who disagrees with an individual that they are too linked to a topic should not exercise their tools in it could serve usefully. I'd go so far as to say that findings there are specifically not of fault, to encourage not hammering an admin (obviously, ignoring its decision would then be a major infringement). Whether or not that is possible, I would be a really strong supporter of a general page where admins could just declare areas they felt that they could not admin 100% fairly in. I have a few, but they just sit in an offline word doc to remind me. There are issues - there are a number of admins (amongst others) who get major following by an LTA, and this would basically be a "ah, I can try to provoke the admin into blocking this account while in this area, and get them sanctioned" honeypot. Nosebagbear (talk) 01:51, 11 March 2021 (UTC)[reply]
I am doing a lot of Eastern European conflict administration, and this is also my primary editing area. I consider myself objective in the sense that I am here to implement policies (and not selectively) and not to push POV, however, I see that the users in the area, and sometimes even neutral users claim that I am supporting one of the sides (always different sides). Thinking that may be in the end of the day I indeed became not impartial, I tried several times to stop administering this topic area - but I seem to be the only active admin speaking Russian, and one of the very few willing to do anything in the area, and when I stop doing anything (including stopping responding to requests at my talk page) then the disputes tend to go forever, and in any way significantly increase the workload for the few admins in the area. I do not have any good solution to this. May be I should stop doing it anyway and not care what happens.--Ymblanter (talk) 06:52, 11 March 2021 (UTC)[reply]
Yeah, there are areas (e.g. British politics) where I know I am not neutral and basically do not use my admin tools there other than for absolutely clear situations. There are editors I know I'm not neutral towards for various reasons (for example HJ Mitchell is a good friend and I very strongly criticised BrownHairedGirl's behaviour related to portals during the arb case) so I would not use my admin tools with respect to them as I am not confident that I would be both neutral and, importantly, seen to be neutral. These are the easy biases though - the ones I am consciously aware of. Some sort of venue where civil, no fault, discussion of biases I might not be aware of would be good. It would need to be structured and moderated such that it wasn't flooded with either friends or opponents (or both) circling as often happens at ANI. It's power would need to be limited to providing non-binding recommendations though to prevent it gaining negative associations or overlapping with arbcom. Thryduulf (talk) 13:19, 11 March 2021 (UTC)[reply]
Thryduulf, Yes, it would solve part of a number of problems to have a more strict interpretation of "involved" -- so far arb com has not been willing to do this, but the community probably could. I see this not as much avoiding doing admin tasks on your friends or opponents, as not trying to dominate admin actions in any one specific subject field,-- especially when dealing with ANI or AE. . DGG ( talk ) 02:33, 12 March 2021 (UTC)[reply]
On that, though, DGG, I should note that while avoiding single admin domination in fields is an appreciable positive, that can force flipside issues where an admin is stuck between: doing those things (and causing that domination), not doing it (and those things not being done), or going and finding an alternate admin (with major time loss if at any frequency) Nosebagbear (talk) 11:21, 12 March 2021 (UTC)[reply]
I was not thinking of out of the way areas, but of such areas as AP, where there is fdrequent attention and hundreds of non-subject involved admins. (which, in this particular case, probably means ones from outside the us) DGG ( talk ) 06:01, 21 March 2021 (UTC)[reply]
Perhaps something like a call for other opinions ping would be useful? Maybe if an admin feels that input into a discussion or page would be beneficial they could add something like {{more admins|page}} or {{more admins|section}} and a bot would deliver a talk page message to ~5 randomly selected admins who have made edits within the last ~2-3 days (admins would have the ability to opt out of receiving such messages and would receive a maximum of N messages ever $periodoftime). This idea has literally just occurred to me so I have no idea whether it will work, or is even technically possible, and I don't know if there is scope for abuse and if so how to avoid it. Thryduulf (talk) 12:07, 12 March 2021 (UTC)[reply]
Pinging random admins to opine on areas they are almost certainly not active in does not sound very productive to me. I occasionally look at one or another area that could benefit from increased admin activity, but the easy cases generally get resolved before I can act, and I find myself spending a lot of time researching policies and guidelines and trying to understand how to act on the harder cases. If I kept at it, I figure I could become proficient in a particular area, but I tend to get distracted. - Donald Albury 12:56, 12 March 2021 (UTC)[reply]
Not sure if this will help, but see the following: conduct and involved from the 2020 BHG case. The ruling clearly states: 5) With few exceptions, editors are expected to not act as administrators in disputes in which they have been involved. Involvement is generally construed very broadly by the community, to include current or past conflicts with an editor (or editors), and disputes on topics, regardless of the nature, age, or outcome of the dispute.
While there will always be borderline cases, best practices suggest that, whenever in doubt, an administrator should draw the situation to the attention of fellow sysops, such as by posting on an appropriate noticeboard, so that other sysops can provide help.
Admins can argue against being "involved" with plausible deniability, and that they were acting solely in their admin capacity despite a pattern of targeting/HOUNDING a particular editor (or editors), especially in controversial topic areas where we expect to see conflicting views. Admins who have managed to keep their hands clean in a specific topic area but have otherwise made known their opposing views known should not be automatically excused. "Current or past conflicts with an editor (or editors)" is rather broad, so perhaps we could make it a bit more specific as relative to HOUNDING/targeting editors they are prejudiced against, provided there is evidence (diffs) demonstrating a persistent pattern of same? Atsme 💬 📧 16:17, 20 March 2021 (UTC)[reply]

Time to close?

edit

I know it's very early in the discussion, but with only 22 supports compared to 64 opposes as of this comment, it does appear that the current proposal in its current state has little chance of passing at this point. Perhaps it could be time to close the main discussion per WP:SNOW and start a new discussion refining the proposal per comments and concerns raised from both sides? Narutolovehinata5 tccsdnew 11:04, 15 March 2021 (UTC)[reply]

We don't count iVotes. We look at the strength of the arguments to form consensus. You may find this article enlightening. Atsme 💬 📧 17:24, 15 March 2021 (UTC)[reply]
I'm aware of that, but based on the current trajectory of the discussion, there seems to be little chance of the proposal passing. Of course, actual discussion could still continue, perhaps just not the !vote part. Narutolovehinata5 tccsdnew 01:22, 16 March 2021 (UTC)[reply]
Early closures are only appropriate when there is near-unanimous support for one view. Otherwise, it's better to let discussion continue. isaacl (talk) 17:35, 15 March 2021 (UTC)[reply]
I'll second the motion to close as-of tomorrow (March 16 1200GMT), as it has will have been open for 7 days. We're down to 2 or 3 !votes per day (although maybe that's down because of the weekend), and there's no chance there will be consensus to pass the proposal. There's no need to shut down discussion, but also no need to continue to pretend that the proposal might pass. power~enwiki (π, ν) 17:50, 15 March 2021 (UTC)[reply]
As long as discussion remains constructive, there's just not much benefit to stopping it. It can often lead to later discussion about whether or not the request for comments was cut off early, and then more re-discussion. We may as well let it proceed until there is a natural conclusion point. (I don't think anyone is pretending; Worm That Turned has already expressed their expectations of the outcome on the talk page.) isaacl (talk) 18:01, 15 March 2021 (UTC)[reply]
Let it run. As has been said. No need to rush. --Deepfriedokra (talk) 18:06, 15 March 2021 (UTC)[reply]
This is a Request for Comment and as such I don't see that whatever is discussed here is binding in terms of changes in any way. A change in our adminship status is a huge step and needs a community wide notice and the discussion that results from that, seems. If this is a first step then it seems there isn't a lot of support to go on. Littleolive oil (talk) 04:39, 16 March 2021 (UTC)[reply]
  • I don't anticipate it, but moods can swing. Since the primary purpose of the RfC is discussion, not to implement a specific change just yet, it is best to let it run it's course and see what new ideas or rationales come to light. Dennis Brown - 10:40, 16 March 2021 (UTC)[reply]
  • Not every RfC needs a close. To the extent that this one does the consensus is obvious (RfCs need not run 30 days if consensus is clear). If there were an eligible editor who wished to close it they could do so now. Or it can fade into the past with no formal close needed. Either of these outcomes would be appropriate for this RfC, in my opinion. Best, Barkeep49 (talk) 14:49, 17 March 2021 (UTC)[reply]

A different suggestion: instead of "closing" the discussion, we just remove it from WP:CENT. (power~enwiki, π, ν) 17:52, 18 March 2021 (UTC)[reply]

Going once, going twice ... I will remove this discussion from CENT in around 24 hours unless there is an objection here (or someone else removes it first). (power~enwiki, π, ν) 16:26, 21 March 2021 (UTC)[reply]

Almost 3:1 oppose to support, time to tag {{failed}} and move on I think. Stifle (talk) 10:30, 19 March 2021 (UTC)[reply]

Excluding sanctioned editors

edit

This may be something to consider if the proposal is ever brought back with some revisions. The provision about making editors sanctioned by the admin ineligible to vote is coming from a good place, and I see some positive comments on it here. But I'm reminded of a point I've seen made about American politics recently, that goes something like, "A government that disenfranchises felons is one with a vested interest in jailing its opposition." I don't think this would lead to an epidemic of tyrant-admins that hand out sanctions like candy just to ward off potential oppose voters on their reconfirmation vote... but I also wouldn't want to create a perverse incentive, where an admin defaults to sanctioning just to save themselves potential trouble down the road. Maybe the 3-year window is enough to address this, since I didn't see anyone else raise the objection (I skimmed). I don't have an answer to this myself. --BDD (talk) 16:15, 15 March 2021 (UTC)[reply]

Thank you for bringing it up because the points you've made are valid. Atsme 💬 📧 17:26, 15 March 2021 (UTC)[reply]

Oldest mods who did not go through standard RFA

edit

Some of the very oldest mods did not go through a standard RFA, but were instead appointed by Jimbo Wales or through a mailing list discussion. Based on your list, there are at most 18 such mods who still hold the mop from these processes. (These individuals can be cross-checked against this list but it's incomplete). How would you ensure that these mods don't slip through the cracks if this RFC passes? Spirit of Eagle (talk) 20:06, 16 March 2021 (UTC)[reply]

Spirit of Eagle, I would take my list with a pinch of salt. WormTT(talk) 15:48, 18 March 2021 (UTC)[reply]
While the mailing list discussions were a long time ago, I wouldn't assume they were less rigorous than early RFAs on wiki. ϢereSpielChequers 11:28, 19 March 2021 (UTC)[reply]
The earliest RfAs were indeed very informal, with some users accepted as admins with just four or five supports (see [3]). That does not necessarily mean those admins have been a problem for the project. I looked at the admins selected from the creation of the RfA page on 14 June 2003 to the end of that month, 17 or 18 new admins in 17 days (see the table here). (The last user to apparently be accepted is a problem; I have not been able to find a record of the user ever being an admin.) Nine of those first admins are still around, although some are barely active. Six Seven quit editing, and have been desysoped for inactivity, and one surrendered the bit. I will leave it to others to decide whether that early informal process was any worse at vetting potential admins than the current incarnation of RfA is. - Donald Albury 18:24, 20 March 2021 (UTC)[reply]
Very informal, but they do look like people who have encountered that editor chipping in with a comment, usually a support for someone they have seen do good edits. Remember the community was much much smaller in those days. Before the RFAs became cluttered with the question section and hundreds of people !voting on unfamiliar candidates who they haven't personally cecked out beyond reading the RFA. ϢereSpielChequers 08:05, 21 March 2021 (UTC)[reply]
In one of those earliest RfAs, after a user said they had no opinion because they were not familiar with the candidate's editing, another user suggested that users who are not already familiar with the candidate shouldn't comment on the RfA. I haven't bothered to comment on an RfA in years because I have not seen any candidates I felt I knew well enough to judge their fitness for the mop. - Donald Albury 11:12, 21 March 2021 (UTC)[reply]
Agree with WereSpielChequers. Just because there weren't all that many supports for either the earliest onwiki RFAs or the mailing list ones doesn't mean that the candidate was appropriately vetted by the community as it existed at the time. Dependent on the time, there could have been a larger part of the active editing community participating in those RFAs than happens now. Risker (talk) 05:27, 22 March 2021 (UTC)[reply]

The assumption of good faith

edit

I've continued to reflect on this proposal, and the assumptions that underpin it. Aside from the patent absurdity of carrying out a "light" poll on the past 10 years of someone's work - if you're going to do a review, it should be a thorough one - it occurs to me that this proposal seems to assume that administrators have not been following the Administrators policy. Alternatively, it assumes that almost every admin is going to sail through this process with nothing but kisses and roses, and maybe one or two comments saying "you could do X a bit better". Both of these assumptions are unfounded on any kind of research or factual data.

I think I speak for an awful lot of administrators when I say that I assume that every single administrator action that I take is subject to review, and I act accordingly. In my opinion, administrators are being constantly assessed in their abilities, strengths, weaknesses, errors and useful actions. One doesn't need an every-10-years review to determine if someone's messing up on a regular enough basis to question whether they should still be administrators. There are a myriad of other reasons, but this is the bottom line for me. If I've messed up, I expect to be told about it forthwith so I have the opportunity to correct the error. (And I *do* sometimes mess up, and I *have* been told about it and have taken steps to fix my errors.) If someone disagrees with something I've done, I expect that they'll let me know so that we can discuss it civilly.

There really aren't that many administrators whose actions are egregiously inappropriate enough to have the community encourage changes at a noticeboard, let alone at Arbcom. The things we assume to be most likely to result in negative "reviews" are limited use of admin tools (and we've had lots of discussions about how certain actions like editing through protection don't leave any logged uses), limited general participation in the project, and active administrative and editorial work in contentious topic areas. The first two are actually related to the existing inactivity policies, and should be discussed there. The third is administrative actions that have always been subject to immediate review, both by the community and by peers, and often by Arbcom, too. In a separate and concurrent RfC, there is discussion of a potential process for community-based desysop. In other words, there are plenty of existing, ongoing, or potential processes that would all achieve a better result than a mandatory 10-year review. Maybe we can improve on the use of positive feedback on an ongoing basis. Maybe some admins may want to voluntarily participate in some kind of review process. But this isn't the way. Risker (talk) 02:59, 17 March 2021 (UTC)[reply]

I think that underlying this process proposal is a rather brutal assumption of bad faith. It's like saying "hey they've been around for ten years, surely they're starting to lose their marbles?". Its the underlying assumption that is so very wrong, that you can only last in adminship for a limited period of time and remain competent. --Jules (Mrjulesd) 10:06, 17 March 2021 (UTC)[reply]
I disagree that Worm That Turned had an assumption of bad faith in making the proposal. As I wrote previously, there are advantages to having fixed terms. If we were designing the process from scratch, I think it may have been a reasonable choice. I believe it's fair to have a discussion on whether or not transitioning to fixed terms now is a desirable goal. I also agree with the significance of many of the raised concerns about the current proposal. isaacl (talk) 14:41, 17 March 2021 (UTC)[reply]
I also disagree that there was an assumption of bad faith here. The intention was clearly a process that admins who are not causing any problems would sail through. "The community put their faith in $admin 10 years ago, do they still have the community support?" is a reasonable question to ask and does not presuppose an answer in either the positive or the negative. Thryduulf (talk) 16:24, 17 March 2021 (UTC)[reply]
Asking the question itself requires the assumption that some of those admins will have done something negative in bad faith or we wouldn't have to ask the question. I don't think the proposal itself is in bad faith, but the idea that we have to prove that someone has not lost their marbles is exactly the opposite of assuming good faith. We are in fact assuming they have lost their marbles and using a poll to prove they haven't so to speak. -DJSasso (talk) 16:28, 17 March 2021 (UTC)[reply]
@DJSasso Yes that's basically it. I'm not saying the proposal is in bad faith, but rather the presumption of the proposal (i.e. to check admins after 10 years) is. I'm sure the proposal is in good faith, but will lead if passed to a presumption of bad faith. There is a subtle difference there. --Jules (Mrjulesd) 22:34, 17 March 2021 (UTC)[reply]
Our encyclopedia and our community has grown enormously over the past 20 years. Gone are the days of fast creation and firefighting. We've codified standards and had a long time for them to bed in. Indeed, it is now very rare for a significant policy change to happen. I'm not criticising our administrators from those days - I am sure every one had the encyclopedia's best interests at heart and are part of what makes it the world leading resource it is today.
There are, however, questions to our current system. We have administrators appointed directly by Jimbo, or ones with less than 10 editors showing confidence in them. I look through my own RfA, 10 years ago, and about half the editors are no longer here. Now, since values change, and since our administrators largely have been working hard for over a decade, it's worth giving them the opportunity to receive feedback on their work. Doing so should allow them to reflect and improve. What's more, with a fresh mandate beneath them from the current community, they should be able to act with a new found confidence.
There was no presumption of bad faith of individuals, but an acceptance that things change. I accept that the community isn't ready for this proposal, but I don't believe it's wrong. I don't believe our community is as hostile as some of the oppose voters expect it to be. Our administrators enjoy a lot of confidence from the silent majority in the community and I would love that to be brought to the fore. I will keep thinking on how. WormTT(talk) 15:46, 18 March 2021 (UTC)[reply]
Just as a datapoint supporting WTT's comment. My RFA in May 2005 finished with a final tally of 57/2/1 and came shortly before the first RFA to ever get more than 100 votes (and I think before the first time 100 Wikipedians agreed on anything, but I'd need to double check that). I went through and looked at all the editors who cast a vote, of the 57 supporters 31 made their most recent edit in 2020 or 2021, 25 made their last edit in 2019 or earlier (7 between 2015-19, 9 2010-14, 9 2005-09) that only totals 56 so I must have missed someone but it's not worth the effort of going back to count again), one of the opposers last edited in 2017 (the other was indeffed for "trolling and vandalism" a few days after my RFA concluded) and the neutral voter has not edited since 2019. The last RFA (that was not withdrawn) to conclude with fewer than 100 votes total and the last with less than 100 supporters seems to be Wikipedia:Requests for adminship/Ceradon in 2015 (72/14/3). The most recent successful RFA with fewer than 60 supports was Wikipedia:Requests for adminship/Mentoz86 in 2013 (58/13/3). All of this shows that my RFA is not evidence that I do (or do not) have the support of the present-day community Thryduulf (talk) 18:11, 18 March 2021 (UTC)[reply]
But that doesn't actually matter because per "Assume good faith" as long as the editor had the trust of the community at one point we assume they still do until proven otherwise through their actions. Forcing all admins good or bad through essentially a second Rfa is against the very core of assuming good faith. -DJSasso (talk) 18:16, 18 March 2021 (UTC)[reply]
That's a novel interpretation of AGF. Levivich harass/hound 12:00, 19 March 2021 (UTC)[reply]
Le me reiterate what I said in my vote above. Providing feedback for administrators is a good thing, and it is a pity that we almost do not have mechanisms for it (btw in your case you were elected to ArbCom, so what happened at your RfA is absolutely irrelevant). But finding a way to provide feedback is one thing, and sending everybody to a compulsory reconfirmation RfA is a completely different thing. There are many arguments in the oppose section explaining why this is not a good solution.--Ymblanter (talk) 19:04, 18 March 2021 (UTC)[reply]
I can remember in the past people quoting theories from I suspect online gaming that the typical time someone spends in an online community was 18 months. Clearly the longevity of our admins and some other groups of editors shows that we are in a different part of the internet to online gaming. Perhaps the internet has grown up, or the early adopters have long been eclipsed by others. Thryduulf's RFA 16 years ago was in the fifth year of Wikipedia, years before my first edit. Who could have expected in 2005 that the project would still be going in 2021, or that the majority of those who participated would still be around l5 years later? I think we have morphed into something more akin to a voluntary society such as an allotment committee, where people may be around for decades. What I hope is that the newer people who look to clearout a lot of dead wood don't see those of us who opposed this particular change as trying to stop any change let alone the inevitable generational one. It is more that while I don't fancy my chances of still being around in 15 years time, I want the Wikipedia of 15 years time to be something that the majority of the people who participated in this RFC are still part of. ϢereSpielChequers 07:41, 19 March 2021 (UTC)[reply]
WTT, certainly appreciate your desire to bring to the fore the support of the silent majority in cases in which it's merited; I, for one, do not take (and never took) your proposal as anything but an assumption of good faith (and I hope my lengthy oppose is not construed otherwise). But I think you touch on, intentionally or perhaps not, one of my concerns: the silence of the silent majority. It strikes me as exceedingly difficult to gauge that majority effectively in such an iterative process of high volume. And if an ancillary purpose of the proposal is to give the good ones the benefit of the community support they do have, my sense is that sentiment won't be reflected in a representative way in a process that requires a consistent and representative cross-section of involvement across time. I'm not envisioning high participation in many of these, which I suspect would give undue weight to incidents and anecdotes that might otherwise be useful discussion points if they weren't so disproportionately featured in a thread. To me, if that feedback is the goal, some form of meaningful, narrative feedback is much preferable to a high volume of approval polls. Perhaps the concern is one of conducing an open dialog and line of communication between sysops and the community at large at any given time, rather than compulsory reconfirmation, in that case. Tyrol5 [Talk] 03:04, 20 March 2021 (UTC)[reply]

Real world analogies - Next steps

edit

One of the problems with RFA reform is that we disagree as a community as to what the real world analogy should be. Is it like a UK driving licence, subject to periodic review after your 70th birthday? When I first passed my test I had a licence that wasn't going to expire for 51 years.... I have relatives born since I took that test who have failed on sections that didn't exist in my day. Or is it more of a political office and only a limited number of mops should exist? What should be the minimum usage before it is deemed to have lapsed? Under what circumstances should it be taken away? We have Arbcom to deal with people who need their bits taken away for cause, and for all the muttering and dissatisfaction at particular cases it has been many years since the community voted out all sitting arbs on grounds of dissatisfaction with their performance. Or since the community has voted that a new offence be deemed so serious that perpetrators merited desysopping by Arbcom - those who are saddened by the way this RFC is going might do well to try and define the sort of offence that they don't think Arbcom would currently take action on, but should, and launch an RFC on making that sort of (mis)behaviour desysopworthy. We have a system for removing the mop from people who have simply stopped editing, and by and large that operates uncontentiously. I suspect there would be consensus to broaden the inactivity criteria to compulsorarily retire admins who aren't around sufficiently to continue to be familiar with the tools. 600 edits or logged admin actions in the last six years would seem to me to be a sensible additional requirement - it would lose us a few admins who are still around and who are probably occasionally useful, but for whom there is a risk that they may not be keeping up with changes. This seems to be the group who get accused of "gaming the system" so there would be some who are OK with deadminning them. The other or alternate change that I suspect could get consensus would be for some sort of periodic admin training module. You could make it multiple choice and entirely automated and allow admins multiple times to pass. But it would filter out those of us who are slipping into dementia, and if you kept it up to date with new speedy deletion criteria etc it would serve as a way to ensure that all admins at least knew of new rules during their three yearly electronic exam. Another advantage of such an exam is that potential RFA applicants could sit it, and it wouldn't matter if 200 admins sat it in the first week it was available. ϢereSpielChequers 17:36, 17 March 2021 (UTC)[reply]

The abstract notion of some form of time-based or activity-based (or both) continuing education process that is more exacting than the present activity-based desysop policy (and which would supplement such policy) without compulsory forfeiture of the tools absent showing of wrongdoing (and assuming compliance with the learning requirements) strikes me as a potentially good, and useful, one. Tyrol5 [Talk] 18:15, 17 March 2021 (UTC)[reply]
If we were starting from scratch (so let's say with a process that only applies to new administrators), I'd consider flipping the question around: have editors who are approved to receive administrative privileges sign up for a fixed term to help out (they can, of course, choose to leave the role early). Once the term is complete, they could decide if they wanted to continue for another term, and an opportunity for providing feedback would occur. Of course, nothing prevents anyone from giving feedback earlier, as desired or deemed necessary. Having an explicit checkpoint, though, often helps prod people into action. (The same goes for recruitment: it can still happen any time, but having someone choose not to continue in an admin role can help trigger a search for replacements.) This would foster the idea of sharing the workload across the community and encourage new editors to pitch in. isaacl (talk) 20:19, 17 March 2021 (UTC)[reply]
A bad driver can easily kill or maim people. I have never anything suggesting dementia in a bad admin decision and it's certain that an admin who misused tools would be caught up in a discussion which would quickly escalate to Arbcom if needed. Johnuniq (talk) 23:48, 17 March 2021 (UTC)[reply]
I doubt we have had many admins reach that stage of life thus far, the community skewed quite young in the era that most admins were elected. But one implication of lifetime appointments is that an awful lot of us will be here until we lose our marbles. As for the idea that a misuse of tools will quickly reach Arbcom. I'm not convinced this is true unless regulars are impacted. ϢereSpielChequers 16:54, 18 March 2021 (UTC)[reply]
How quickly an issue gets to arbcom is hard to predict. Complaints about RHaworth's use of speedy deletion began at least a couple of years before the arbcom case. Obviously arbcom's involvement would have been premature at that stage, but the timescale was still not "quick", now consider someone who makes fewer edits/actions in an area of the project less watched than speedy deletion. An issue with say dealing with new editors in an obscure topic area (the history of New Caledonia for example) will likely take much longer to come to the notice of the wider community and longer still to reach a level that requires arbcom involvement, especially if they have a coupe of vocal defenders who are ANI regulars. Thryduulf (talk) 17:26, 18 March 2021 (UTC)[reply]
I think RHaworth is an example of a more general problem with behavioural issues. We are very good at dealing with some types of bright-line misbehaviour, such as edit warring or sockpuppetry. Break one of those and you won't last long. We're not nearly as good at dealing with other types of behavioural problems, such as persistent incivility. Persistently making bad deletions and responding badly to complaints about them doesn't breach one of the bright lines we've established, so it was tolerated. The ArbCom case on RHaworth was opened after he reversed a checkuser block without consulting a checkuser, which is a bright-line offence, even though objectively speaking it isn't nearly as problematic. Hut 8.5 19:22, 18 March 2021 (UTC)[reply]
I'd describe overturning a checkuser block as an action that impacted a regular. It is also the sort of thing you could remind admins of in periodic online retraining. Obviously 600 edits or logged actions in 6 years would have had no effect on Rhaworth, even after he limited himself to 100 deletions a day. Also if we decided to change "should inform" people whose articles you were tagging for deletion to "must inform" then that would have been something to put into the retraining. ϢereSpielChequers 07:08, 19 March 2021 (UTC)[reply]