Why first resort?

edit

I'm not a fan of recall, partly because I think that Arbcom already does this fairly effectively and I'm not keen on a multi judicial system - not least because of the inherent risk of double jeopardy. I'm particularly not keen on a system that would allow a couple of dozen lulz seeking trolls to take out the admin corp so easily. This proposal doesn't even have a throttle whereby no more than 6 reconfirmations can take place at the same time.

I'm also very unimpressed with a system that doesn't require editors to make a good faith attempt to resolve their dispute with an admin before initiating this procedure. I.e. instead of "Editors may request recall for any reason, but are expected to do so civilly and in good faith." you really need something more like "Editors may only initiate the recall process after all normal forms of dispute resolution have been exhausted." ϢereSpielChequers 17:19, 30 June 2010 (UTC)Reply

I've been an admin for 2 years, and open to recall, on the same basis that Herostratus was, requiring 6 editors in good standing to start the recall process. I've dealt extensively with image deletions, and ArbCom Enforcement; in other words, areas which attract criticism. Nobody has even started a petition against me, so in all seriousness, I'm not remotely worried about this system which requires 20. On the other hand, I think some sort of note about considering dispute resolution is a good idea. PhilKnight (talk) 22:07, 30 June 2010 (UTC)Reply
Thanks for adding the dispute resolution bit, I'd prefer a bit more and to have it start with an attempt to resolve things on talk. But what you've added makes it a lot better. ϢereSpielChequers 23:09, 30 June 2010 (UTC)Reply

Troll magnet

edit
If 20 editors who you've never interacted before were to suddenly show up and ask you to resign, then I'm sure you'd get the tools back in 7 days. But if they were to do this simultaneously to all the admins who are ever active in the UK morning, then they would, with remarkably little effort, have created an admin free window of two or three hours per day for the seven days between those admins resigning and them getting the tools back. I'm sure that wasn't the proposers intent, and it would be easy to prevent this either by setting a throttle, or by letting the admin keep the tools for the duration of the RFA. But at the moment this is one of the implications of the system being proposed. ϢereSpielChequers 22:46, 30 June 2010 (UTC)Reply
In my opinion, there needs to be a balance between having a recall system that will remove incompetent admins, and a process that isn't gamed to remove competent admins. If this process needs to be adjusted, in order to discourage gaming, that could be achieved by increasing the minimum number of edits and tenure from the petitioners. At the moment, I don't believe such a change is necessary, however if this was very widely adopted, then perhaps some adjustment would be necessary. Regarding your hypothetical, I think common-sense would prevail, and the petitions would be closed without anyone resigning. PhilKnight (talk) 19:54, 2 July 2010 (UTC)Reply

I'm not too crazy about requiring 20 signers. That's a lot. If you can get 20 people to sign, you practically might just as well resign and skip the RfA. I think six is fine. Maybe ten then.

I do not like the idea of admins having to resign before the RfA. Why should they? I wouldn't do it, partly as a matter of principle and of pride. Why should I resign even for a week just because six people say so? If 48 people say so, that's very different.

I would suggest that closing might best be done by a bureaucrat. Bureaucrats as such aren't now tasked with closing recall RfA's. but a bureaucrat acting as a regular editor can do so. Bureaucrats are used to closing RfA's and presumably trusted to do so. This would add a de facto new power/task to bureaucrats toolkig; I don't see anything wrong with that. Bureaucrats can't remove admin rights, so the affected admin would ask the College of Stewards to do so.

I'm not at all convinced that 75% shouldn't be the threshhold for recall RfA's. The person closing can discount any votes based on personal vendattas etc. Herostratus (talk) Herostratus (talk) 03:04, 8 July 2010 (UTC)Reply

Regarding your comment "if you can get 20 people to sign, you practically might just as well resign and skip the RfA", I kind of agree, if 20 editors asked me for good reason, I would resign, and I wouldn't go through a RfA anytime soon. PhilKnight (talk) 17:20, 8 July 2010 (UTC)Reply

Admin recall -- moving the process along politically

edit

Not bothering with the technical details now, what I was thinking about was, how to move admin recall forward politically and make it become a de facto part of adminiship.

I was thinking about the Service Awards. There was a great deal of talk about having something like these, not having something like this, that it was a good idea, that it was a stupid idea, and so forth, the upshot being that it was obvious that no proposal to create these would ever pass. You need consensus for a proposal, and what is that? At least 75% I would think (granting that we don't actually count votes), and that was never going to happen, given that there was plenty of dug-in, intransigent opposition.

Instead, I just created them.

Sure there was an MfD on them, but it failed. After all, they already existed, and they looked reasonably nice and everything, so people said meh let 'em be. As the saying goes, it's easier to get forgiveness then permission.

Given that a proposal for a recall RfA policy will probably never succeed, maybe this same kind of approach could be taken here. We just do it.

Now, I'm not actually suggesting that the Four steps below actually be adopted as posted (more safeguards would be needed, and as given I guess it could lead to some fighting (at first anyway)). And obviously the actual category names would need to be much shorter, and so forth. But maybe some thinking along these lines could move things along. Four steps, no policy proposal/adoption process required. And the first two steps are easy.

STEP ONE: Remember that it's a wiki.

STEP TWO: Remember that, when you call a bunch of editors together to do something, you will get a lot more good people than bad people.

STEP THREE: You create a category "Editors who agree to start a recall RfA on any admin for whom [6/10/20/whatever] good editors have requested this, and who agree to work with other such editors to prevent these recall RfA's from being deleted or closed improperly (while at the same time making sure that recall RfA's that don't meet proper criteria are quickly deleted)."

Since no new policy has been adopted, admins failing their recall can be forced to do exactly nothing. They could resign if they want to but, but not if they don't. But...

STEP FOUR: You create another category "Editors who agree to place admins who have failed a recall but declined to resign in the category , and to work with other such editors to make sure that such admins stay in that category"

My guess is that Category:Admins who have declined to be recalled would not be heavily populated. Herostratus (talk) 03:13, 8 July 2010 (UTC)Reply

I am a firm supporter of taking unilateral action where necessary and dealing with the consequences afterwards, but I'm not convinced it would work in this case. Your service awards thing worked because once a page is created a consensus is required to delete it. It doesn't work like that for desysopings - if a steward unilaterally desysops someone there is no need for a consensus to resysop them, it would just happen. I know you don't actually propose any desysopings, but if an admin has declined recall they aren't going to resign just because you put them in a category. --Tango (talk) 19:25, 8 July 2010 (UTC)Reply
Well, you could call the category something else. Category:Admins who have dishonored themselves, say. I actually think the process outlined above might actually work. I'm not actually recommending it because it would probably be a huge destructive fight with probably people quitting and so forth, and no way would it be worth it. Herostratus (talk) 01:58, 9 July 2010 (UTC)Reply

"recall for any reason"

edit

Re: "Editors may request recall for any reason, but are expected to do so civilly and in good faith."

Disclaimer: I'm open to recall, but I'm currently using Lar's procedure. If this alternative process becomes widely accepted I'd strongly consider switching.

I'm unconvinced by the "for any reason" part. Editors have numerous ways of addressing my non-admin work. Surely the only reasons for recall should be perceived misuse of the tools, not a general unhappiness with an editor (who also happens to be an admin)?

TFOWR 10:38, 10 July 2010 (UTC)Reply

Hi TTFOWR, I think having a clause that a reluctant admin can argue over has the potential to cause unnecessary drama. Also, if the community felt I was 'out of step' on a policy issue, then I think a recall could be justified. PhilKnight (talk) 10:50, 10 July 2010 (UTC)Reply
Makes sense (I like less drama...), thanks! OK, that gets me thinking about the "20 editors..." question - but in a good way. Thanks again! TFOWR 11:01, 10 July 2010 (UTC)Reply

alternate way to have better oversight (oversight as in the English language, not Wikipedia jargon)

edit

The recall system is nearly a sham. The administrator makes up their own rules. Often, it's difficult to be hard to do.

Administrators should be responsive to concerns and accountable. To achieve that, I propose a standard set of conditions which administrators may agree to. Still, since they have to agree to it, it's not really fair to the community.

DRAFT 1

1. An administrator will be subject to a learning period of 30 days. During this 30 days, an administrator cannot be recalled except for the most severe misconduct.

2. An administrator can be recalled by one user if there is a violation of policy and there is not a correction to it within 7 days of a request for correction.

3. An administrator can be subject to a limited recall if requested by two users for serious violations of policies or repeated poor judgment or improper blocking. A limited recall would be suspension of administrative powers for 30 days. An administrator could be suspended for a maximum of 30 days per calender year. (In other words, even if ganged up upon, that administrator would retain powers for 11 months of the years.)

4. An administrator can be subject to a permanent recall if requested by 3 experienced users who have edited more than 500 edits and have edited for more than 6 months. If the complaining users are found to have started the recall in bad faith and without discussion and attempts of prior resolution, then they will be blocked for a period of 15 days.

5. Once an administrator accepts these conditions, they are binding for at least 2 years. They cannot drop the recall option whenever they want to.

Summary

Very simple. When you are first an admin, you get a break. If you are an admin and you violate policy, you can correct it but if you don't, you can be recalled. If you are not a good admin, you can be suspended for 30 days of the year. If you are really bad, you can be recalled permanently but it has to come from experienced users and the complainers get punished for frivolous recall attempts. EatingGlassIsBad (talk) 16:56, 4 November 2014 (UTC)Reply