User talk:Writ Keeper/Community desysop process

Question from possibly involved Demiurge1000

This seems a very interesting and carefully thought out proposal. It's also a solution to a major problem, of course. (There may be people who disagree that unlimited terms for admins, and immense difficulty in revoking admin status, and the things that result from these, constitute a major problem, but that's a separate discussion although I'd be happy to address that too.)

I am, in general, in support of it. I might have some tweakery at some point, in fact I'd be willing to consider making it less harsh (for the admin) just in order to increase its chances of acceptance, but I'll wait for others to opine before going there.

Here's another question. This question comes from my reading things very quickly and ignoring context; I got more than half way through before I realised that these were not intended to be solely your (WritKeeper's) recall/desysop requirements, but in fact were a proposal for all administrators including you.

So, here's a question. Would you (WritKeeper) be willing to adopt this as a recall/desysop procedure for yourself regardless of what traction it gains as being such a procedure for others? --Demiurge1000 (talk) 18:50, 13 August 2014 (UTC)

Sure, if it helps. My current recall promise is (arguably) significantly more severe than this: if three reasonably established editors ask me to resign, I'll do so. In the past, I've said that I wouldn't ask for the admin bits back for at least another 72 hours or something, and not ask for the 'crat bits back at all without another RfB, but at this point I probably wouldn't even bother. But yeah, if it makes it easier, I have no problem with being bound by this. Writ Keeper ⚇♔ 18:55, 13 August 2014 (UTC)

Parade-raining by am-I-involved?-what-does-involved-even-mean-in-this-case? Floq

So, you're going to tilt at this windmill again, eh Keep? Not to toot my own horn too much, but I've always liked my procedure here: User:Floquenbeam/Recall. Less bureaucracy (or, less apparent bureaucracy) than what you describe. Perhaps a good addition to my idea would be your bringing in the Crats earlier in the process, to keep things on the straight and narrow, but it's certainly less prone to abuse than "three reasonably established editors" (my god, how have you not pissed 3 people off simultaneously before?), while probably not significantly more prone to abuse than your suggestion. Simplicity is a big selling point, IMHO.

The problem of course, is that even among those who think a community desysop process is a good idea, everyone has their own idea of the perfect method to use, and a large enough percentage of them will vote against anything they consider imperfect that ultimately nothing will change. But, shame on me for raining on your parade... I'm rooting for you. I would certainly support my idea, or your idea, or pretty much anyone's non-insane idea. --Floquenbeam (talk) 20:53, 13 August 2014 (UTC)

Well, I like that too, and it's not unlike my own standard, in ways. My proposal here certainly trades simplicity for structure, but I also like its symmetry: you get the bits by an RfA, and you lose the bits by an RfA. The structure gives it some kind of definition of what will and will not result in a desysop; yours could be a little ambiguous (or could be made to be a little ambiguous), and I've never really liked the format of RfC/Us much, personally. But if people like yours better than mine, I'll champion (i.e. steal) it. Writ Keeper ⚇♔ 21:45, 13 August 2014 (UTC)

and yeah, I don't know what "involved" means here, either, and also, it's probably either timing or refuge in audacity: I've probably pissed three people off, but perhaps not at the same time, or perhaps they don't think that I'm serious about it and didn't bother to try. Writ Keeper ⚇♔ 21:50, 13 August 2014 (UTC)

Actually, I was just browsing through Wikipedia:Administrators open to recall/Admin criteria, and there are several more-or-less equally simple, equally fair methods on display. I'd probably settle for almost any of them. I'm not a big fan of Lar's method, which has quite a following (too complicated, too difficult to succeed), but most of the rest have minor weaknesses, but not fatal weaknesses. --Floquenbeam (talk) 21:56, 13 August 2014 (UTC)

Well, the obvious flaw with any voluntary recall system is that it's voluntary: it relies on the community to trust the admin in question (to stick to their word), but if the community is seeking to remove an admin, they no longer trust the admin pretty much by definition. Some of those processes, though, compound this flaw, and Lar's is an example. They talk about clerks of their own choosing (but they choose them to be impartial!) and judging petitioners to be ineligible and the like; well, in order for the process to work fairly, we have to trust Lar to make all those decisions fairly, but if we're seeking to remove them as admin, we presumably don't (otherwise, why are we seeking to remove them?) Writ Keeper ⚇♔ 22:02, 13 August 2014 (UTC)

First-read thoughts

Some stream-of-consciousness thoughts as I read this proposal in another tab:

• Generally, I think community desysop is an idea that's been a long time coming, and could do good for the community's collective health.
• if it's not certified by another user in 48 hours, it goes inactive and is subject to deletion - define "another user". Do they have to be uninvolved in the precipitating event? Do they have to be uninvolved with regard to the admin, in general? Or are they expected to be someone who has been involved in dispute with the admin, and tried and failed to resolve it? Do they have to be uninvolved with the person who put forth the desysop proposal, or can they be from the same side of the same dispute as that person?
• However, if it is certified, that's where the process diverges. Instead of proceeding to a general comments stage, a link to the request should then be forwarded to the 'crats via BN, who will then manage the procedural things - the gap here is a fail point. In between someone opening a RFDeA and the crats managing the certified version of it, there's a no-man's land of possible dramatic chaos. Can comments be made on the RFDeA during this time, or only a certification? Can there be multiple certifications? What if the whole thing devolves into a firefight?
• If this still isn't enough time, or if the subject doesn't reply at all within the original 48 hours, than the process may be delayed indefinitely, but the subject's admin bits will be removed until the desysop request can go forward - this is...surprisingly reasonable, generally speaking. I think it could work. If I were going to find a problem with it, I would say it's that an admin in these circumstances will be going into the RFDeA under a different status quo than every other admin, and as we have learned in the past, people's views of what it takes to change an extant status quo vary hugely. Will "well, they already don't have the bit anymore, so..." make a difference? Will the apparent assumption that the desysop proposal is "right" (why else would you default to desysopping, etc etc) make a difference to the admin in question or to the community's voting tendencies?
• but since this is probably the one place that will earn a reputation even worse than RfA's, such comments should be moderated (probably by, again, the 'crats). Again, 'crats should be extremely careful about what they choose to moderate here, and a person's actual !vote itself cannot be subject to moderation (though it can be indented for the usual reasons). - The potential for attacks and rage here is massive. Crats - or whoever ends up being in charge of moderating this - should be willing and prepared to moderate heavily. Making the guideline "only do this extremely carefully" for crats will amount to them practicing it as "don't do this" in most cases; instead, it should be practiced as "it's your job to use your judgment to keep things under control here, so do it."
• The minimum requirement for removal of the bits should certainly be between 50%+1 and 70% of people voting "Remove" (as 70% the lowest bound of the discretionary range for RfAs); it shouldn't be harder to remove an admin at this point than it was to approve them. - Makes sense to me. The numbers could certainly be fiddled with, but as a first approximation, "half to three-quarters of voters want you gone" seems like a reasonable place to draw the line. However, I think that crats should be much more consciously guided by argument strength in a desysop discussion than they are (seem to be?) in a typical +sysop one; desysopping someone on the basis of a number of people giving reasonless "no"s or general "I don't like him"s makes me uncomfortable. I'm almost inclined to think that all RFDeA's should be closed via crat chat rather than a single chat, to avoid this. A fluffernutter is a sandwich! (talk) 16:59, 14 August 2014 (UTC)
  • "Another user" is just another user who is most likely not a sockpuppet. These are the initial stages of the request; they shouldn't be that hard to pass. I thought about involved vs. uninvolved, but the more I thought about it, the less I was sure which group should be the one that shouldn't count, which sounds as good an argument as any for not restricting it to either. After all, this system includes a gatekeeper mechanism for the two-obvious-vandals-that-were-blocked scenario. This should be harder than just any random person with a grievance to start, but not *that* much harder.
  • Yes, the disjoint here is a problem. Perhaps it might help if the RfDA were to take place on the same page as the request itself. Basically the idea is that the first phase is no comments: the initiator lays out their case, someone seconds that (e)motion, and the whole thing is locked down until we can get the subject on board. Multiple certifications are unnecessary, and I really think that any hypothetical people who want to be another certifier on top of what is necessary for the thing to move on (presumably to get their scalp"I got an admin desysopped!" merit badge) should be discouraged. So no, no comment, just a single person to say, yes this is a good idea, and off we go. Likewise, the RfDA should not be commented on until it is officially opened. If more support is needed, like if the 'crats were considering to close the request for baselessness, these users can be much more useful in expressing their opinions to the 'crats directly--one of the reasons why the discussion would be forced to be a) a certain length and b) public.
  • Honestly, the community's voting tendencies are for it to decide. This should be seen as a purely temporary desysop simply to allow the subject the time they need; it shouldn't be held against them, is my idea. But if the community decides to hold it against them...well, then they decide to hold it against them. Is what it is.
  • Naturally, there's a tradeoff between freedom to speak out and criticize and the need for an at least tolerable atmosphere. We just have to do the best we can. Overagressive moderation can be as volatile, in its own way, as underaggressive moderation, or perhaps even more so. Care will need to be taken regardless. I take your point, though; we can hammer out the exact wording later. If the general idea is that of a moderated forum that still allows a reasonable freedom of speech and expression, well, that's what I'm going for.
  • Not unreasonable to mandate a crat chat. Of course, the rationale-less nos are likely to be balanced on the other side by rationale-less yesses, but again, point taken. Writ Keeper ⚇♔ 17:30, 14 August 2014 (UTC)

The time period between the beginning and the "automatic temporary desysop" is too short, and incredibly easy to game if someone knows that an admin they don't like (for whatever reason) is going to be away. I also strongly believe that there should be a requirement that anyone creating or certifying an RFC about anyone for anything to have made genuine efforts to discuss the issue with the person in question, and to be able to demonstrate such efforts.

I am also very, very concerned that any RFC-like process be considered binding: it stops being primarily a feedback process (the primary intention of user/admin RFCs) and turns into a disciplinary hearing that might just as well be on ANI as anywhere else. I expect that, as currently proposed, it will become a backdoor method of removing adminship from those who are largely inactive, instead of the community coming up with more enforceable rules for this. While I'm not particularly opposed to removing adminship from people who don't use the tools or only show up to do their annual edit or two to keep them, I think that more conscious thought is required. I also am certain that the admins who most need to be desysopped will be immune to this proceeding; they tend to be fairly popular with the anti-establishment segment of our community, and unless people are genuinely incensed at someone's behaviour, it's very difficult to work up the courage to make a public pronouncement that is extremely critical of someone else, particularly someone with whom they may be regularly interacting. Because we've made it so ridiculously hard to gain the sysop bits, it's pretty much the biggest insult one can make against an administrator to try to remove said bits. Regardless of what Jimmy says, at heart most editors on Wikipedia avoid being hurtful to people. Risker (talk) 17:05, 18 August 2014 (UTC)

• If an admin is going away, they have no need for the tools, and when they come back, they will be going through the exact same process that they would've if they had never left; that's not gaming, it's just acknowledging that, while we can't expect admins to devote every second to Wikipedia, we can't expect the community to wait on them to address admin issues either.
• Yes, they should have tried to resolve the dispute before. I actually explicitly stated this in the proposal: [...]including the attempts by users to resolve this previously.
• This is, of course, not an RfC, though the first part of it resembles the first part of an RfC/U: this is not a feedback mechanism, it is a desysop process. It is intended and designed to be as binding as an RfA, and the decision part of it--the important part--is modeled after RfA, not an RfC.
• As for the "backdoor" thing, I don't agree. Again, if a user can't be bothered to defend their own use of the bits, whether because of inactivity or anything else, I don't think they need them. One can base that in WP:ADMINACCT if one is the type to want current policy analogues to new policy proposals. Again, when they come back, they go through the process as if they had never left, and that process will not be short on conscious thought.
• I'm not going to comment about the immunity thing, since it seems to be rooted in your personal opinion of who should be desysopped rather than anything objective about the process, other than to say that that would be true of any community-based desysop process, and frankly, I think having some sort of community mechanism is too important to be derailed by such opinions.
• Saying something publicly certainly is hard, but that sure doesn't seem to stop people; regardless, I discussed a way to submit a statement anonymously in the proposal, which might ease that a bit. But there again, that's going to be true of any community desysop process; I don't think it's worth not having one at all just because people might be reluctant to use it.
• You say that passing RfA is ridiculously hard. I'm actually not sure I agree with that, either, but if it is, perhaps it has become that way because there is no community desysop process to recall an admin once they are elected. So maybe this will help. Writ Keeper ⚇♔ 17:37, 18 August 2014 (UTC)

Who is this for?

Are there any admins that the community has been clamoring to be desysoped that arbcom has not acted on?

I want to be convinced there is a problem before I look into solutions. Chillum 21:13, 21 August 2014 (UTC)

Chillum: This isn't for anyone. If it was, I'd be rightly laughed out of town--in fact, I myself have strongly discouraged people from crafting a process in response to a specific incident/person (more on that below), because hard cases make bad law, and it's not the appropriate way to go about things regardless. Simply put, this is about equity. The community is able to give out the admin bit but not take it away. That's uneven, and it should be fixed. Admins should be accountable to the community, not just Arbcom. (As an aside, the idea that Arbcom is the community is, well, untrue; look at the number of arbs that have resigned from Arbcom for evidence of that. They're separate entities.)

The thing is that, with Arbcom the sole purveyor of desysops, adminship appears to be an old boys' club: once you're in, you're in. If admins are only accountable to Arbcom (which is virtually always staffed by other admins), then admins appear to be only accountable to themselves, which basically means that admins are accountable to nobody (after all, admins will always close ranks to protect each other!) Now, that's not entirely true, but there is some truth to it, and there is the appearance of more. It heightens the appearance of a class division between admins and non-admins (since the lowly peasants have no recourse), and in this case, I'm of the opinion that appearance breeds reality.

Moreover, there's no way to predict future need from past demand here. It's quite possible that the community hasn't clamored to desysop some admins simply because there is no place for them to do so. Arbcom is seen by many as an insurmountable barrier, which probably has had the effect of quashing criticism before it's even expressed.

Nevertheless, there are a few examples. Two I've found out about through my role as 'crat (since such conversations always end up involving a 'crat for obvious reasons). One is here (perhaps not a great example), and another is here (probably better). The demand for a community desysop process is not imaginary. Writ Keeper ⚇♔ 04:34, 22 August 2014 (UTC)

I don't think there is a significant gap between who the community wants to be admins and who they have. You say the demand is not imaginary but in fact this has been proposed and rejected many times. The community has repeatedly refused this sort of idea consistently throughout the years.

I suppose it does not hurt to try again. Chillum 05:00, 22 August 2014 (UTC)