Wikipedia:Requests for Comment/Request for de-adminship proposal

Request for de-adminship process[edit]

To comment on this proposal, click here.

This bulleted list is now proposed: Wikipedia:Requests for removal of adminship.

The following, is based upon these discussions.

Request for de-adminship process to specifically request "removal of adminship"
- the nomination/request to be framed like an WP:RFC/U
  - which requires at least 3 admins to initially certify (to prevent wasting the community's time and "pitchfork noms")
- and to limit harrassment, is limited in the number of times:
  - a particular editor may start one concerning any admin (6 months)
  - any editor may start one concerning a particular admin (12 months/1 year)
  - a particular editor may start one concerning a particular admin (24 months/2 years)
    - With it clear that the limitations above also apply to any certifier; and that WP:IAR may apply to all the various time limitations in cases of clearly egregious admin actions (but which do not quite qualify for "emergency de-sysopping"), though in that case requiring additional certification by at least any one bureaucrat (who then obviously would not be a closer of the discussion).
- is a community-wide discussion
  - which lasts 7+ days (length is at bureaucrat discretion, just like RfA)
  - uses the hybrid consensus/!voting model (just like RfA)
  - is closed by a bureaucrat (just like RfA, using RfA thresh-holds)
If the request for removal of adminship is successful, the admin has adminship immediately temporarily removed (presumably by the closing bureaucrat), pending review by the arbitration committee.
- Arbcom has 7 days in which to review the request. (Most common results likely to be to endorse or to overturn the request for de-adminship.)
  - Else, at it's option, to forgo the rest of the process and instead to have a full RfArb case opened (during which, adminship would remain temporarily removed).
- If Arbcom has not come to a decision by the end of the 7 days, the adminship is temporarily restored.
- Arbcom then has an additional 30 days to continue discussion. (To allow for full discussion)
- If Arbcom has not come to a decision by the end of the 30 days, then adminship is considered restored, and the request to remove adminship considered unsuccessful. (pocket veto).
If the admin in question voluntarily requests adminship be removed at any time during this process, such removal should be deemed to be "under-a-cloud", and fall under the relevant restrictions thereof.
If adminship is removed through this process, the editor is free to re-request adminship (following the standard RfA process) at any time after this request for removal process has concluded, at their discretion (unless under arbcom restriction to the contrary).
Nothing in this process should be considered to prevent or constrain Arbcom from taking action if deemed by them necessary. (That, of course, includes the situations of "emergency de-sysop" starting/declining cases, etc.)

The above addresses the suggestions and limitations noted in the discussions, while allowing for the community, the bureaucrats, and arbcom all to be involved. And there are enough steps and safety valves that it's (hopefully) unlikely that "pitchforks noms" will be successful.

With the time limit restrictions, remember that "it doesn't have to be you" to nominate or certify, just like our standard that "it doesn't have to be you" to close an XfD. And they are that lengthy because the process potentially takes a month and a half to fully resolve. (Though the process could be as short as 7 days, if arbcom is quick to come to consensus.)

Short version without the expanded explanations

Nomination requiring certification by 3 admins
Community-wide discussion in the style of RfA
Discussion closed and implemented by a bureaucrat similar to RfA
Review by Arbcom

I tried to make this all as concise as possible. If anything in the above is unclear, I would be happy to clarify. - jc37 19:23, 27 July 2012 (UTC)[reply]

Support[edit]

Oppose[edit]

Oppose. Process-heavy systems work poorly here. !votes, also, are unpredictable, especially where there are no limits on who can vote. Finally, I don't believe that bureaucrats, as a group, are well suited to referee such a divisive process. In addition, I oppose the underlying idea that we need such a process, at all. The Uninvited Co., Inc. 20:03, 6 August 2012 (UTC)[reply]

Discussion[edit]

After Wikipedia:Community de-adminship/RfC, I gave a lot of thought to whether we should propose something better, and the proposal here is pretty similar to where I would have gone. Basically, I like the idea of ArbCom having the final say, and I think that needs to be a feature of any method that will likely have community support. However, I think that RfC/U is an exercise in going nowhere. It never leads to consensus. There would have to be a process more like RfA in reverse, leading to either a decision to take no action, or a decision to forward the case to ArbCom. --Tryptofish (talk) 21:48, 27 July 2012 (UTC)[reply]
I agree. The RfC/U part of this is the format for the nomination only. The actual discussion would essentially be "RfA in reverse", as you note. - jc37 22:10, 27 July 2012 (UTC)[reply]
- Can it be less complex, and explained in fewer words? Please see my proposal at Worm's page, which avoids a lot of the paperwork by using trusted crats. For example, reading it backwards, aren't the last three bullets already a given? I'd make this proposal as simple as possible; it will have a greater chance of success. Tony (talk) 08:16, 28 July 2012 (UTC)[reply]
  - Hi. To answer your first point, people simply won't support something like this without the explanatory details which some might say could go without saying. But ignoring the details for a moment, let's say you were wanting to nominate an admin for de-adminship. The process would be: You write up a nomination similar to an RFC/U. Once 3 admins certify it, then a sort of "reverse RfA" (RfD-a) is held. If deemed successful, the closing bureaucrat temporarily removes adminship from the admin. Then arbcom reviews. If over-turned by arbcom, then the adminship is returned, if not, then they would need to go through another RfA to (re-)attain. Or even shorter: Nomination certified by 3 admins -> Community-wide discussion in the style of RfA -> Discussion closed and implemented by a bureaucrat -> Arbcom review. Is that clearer? - jc37 20:29, 29 July 2012 (UTC)[reply]

From a technical perspective, bureaucrats cannot currently remove adminship. Instead, a request must be made at meta:Steward requests. Does this proposal imply that bureaucrats should be given the right to remove the sysop flag? — This, that, and the other (talk) 11:44, 28 July 2012 (UTC)[reply]
- However they remove adminship now. The mechanics isn't important, is it? Tony (talk) 12:23, 28 July 2012 (UTC)[reply]
  - My point is that steward requests can sometimes take time to process. If the closing bureaucrat were able to do the honours themselves the "immediate" part of the proposal would be able to be put in place. — This, that, and the other (talk) 02:30, 29 July 2012 (UTC)[reply]
    - You may have missed the discussion last year which resulted in bureaucrats being able to remove adminship on en.wiki. And, having the closing bureaucrat do the removing is part of this proposal already. (Though it really doesn't matter who actually "pushes the button" as it were.) - jc37 20:39, 29 July 2012 (UTC)[reply]

I'm starting to really like the idea above, about crats, and maybe having at least one of the certifying parties being a crat is a good idea. Again, we need more crats. Dennis Brown - 2¢ © Join WER 14:02, 28 July 2012 (UTC)[reply]
Something else that I think is worth considering is not to have the 'crat remove adminship directly after the community discussion, at all. Instead, they would make a closing decision to forward the case to ArbCom, where the actual decision about whether or not to desysop would be made. Then a 'crat could remove the bit, on ArbCom's instructions. I make this suggestion for several reasons. First, my experience from the last time tells me that the community will come out in force to oppose anything that gives a whiff of "pitchforks", so it makes sense to me to give ArbCom, instead of the community at large, the actual decision. Second, I see this as an alternative, possibly more workable, path to request-for-arbitration. It's an alternative to the traditional RfC/U as a prerequisite for RfAr. It could be more attractive than going directly to RfAr, because it allows community discussion first. And it could, in effect, take the place of the evidence and workshop phases of a full arbitration case, allowing the Arbs to resolve the matter by motion most of the time. --Tryptofish (talk) 14:56, 28 July 2012 (UTC)[reply]
- I'm not sure, as technically we are trying to reduce the load at arbcom and limit their action here to one of oversight, not decision makers. They have to have the power to overrule (and continuing power to instate their own) but not sure they need to be involved in this process unless the chose to. Dennis Brown - 2¢ © Join WER 15:36, 28 July 2012 (UTC)[reply]
  - I'm all in favor of reducing ArbCom's workload, and I've been advocating it for quite some time. However, I think there are numerous other places besides de-adminning to do it. De-adminning is actually something that fits their core job description of resolving difficult conflicts. I used to be a big believer in making the community at large the decision maker, just as it is for RfA. But that approach was rejected by the community in the last poll. I don't know, maybe opinion has changed since then. But my past experience was to spend a huge amount of ultimately-wasted time developing the CDA proposal, only to get it slapped down. --Tryptofish (talk) 20:13, 28 July 2012 (UTC)[reply]