Commons talk:Administrators/Adminship policy

From Wikimedia Commons, the free media repository
Jump to: navigation, search

Extracts from De-adminship talk page that are relevant[edit]

Voluntary deadminship[edit]

This needs to be mentioned if this wider policy is developed. If an admin requests removal of the bit voluntarily, under when can they be speedily reapproved and when can they have to go through the full process. I'd suggest if the former admin's activity is such that they would not have been forcibly desysopped at the time of the readminship request that it is granted speedily. Otherwise, they should go through the process like a forcibly desysopped user. Note: I bring this up as will apply to me.--Nilfanion 22:42, 15 May 2007 (UTC)

If an admin gives up his adminship voluntary due to inactivity, I see no problem with him getting it back speedily, upon his request (such as is the case with Nilfanion).
I think it is necessary to define under which circumstances this would apply. For example, Andre Engels recently resigned his sysop and b'crat bits for three different reasons. [1] Applies to him?
Furthermore, what if an admin was involved in serious disputes just before he announced that he was taking a break?
A solution might be that every request for readminship is to be posted at the COM:AN. It could then be evaluated independantly, and if there are no objections, a b'crat would restore the adminship. But if any other admin objects, then a standard RFA is required.
Fred Chess 23:55, 15 May 2007 (UTC)
Just a note, "the full process" is not exactly onerous, and I would expect any existing admin who was desysopped for inactivity or voluntarily requested a break, would easily pass another RfA. But the community has already once expressed their approval, so perhaps all that is needed is the person to confirm to a bureaucrat that they have checked what has happened to policy in their absence, so they are aware of the latest practices? --pfctdayelise (说什么?) 23:58, 15 May 2007 (UTC)
Agreeing fully with pfctdayelise & Fred - to me anyone who is a good contributor who voluntarily takes a break and hands back the tools (don't we wish others would do it) deserves respect. I would hope regaining the tools would be a simple & speedy process --Herby talk thyme 07:39, 16 May 2007 (UTC)
Agree with all above. Cary Bass demandez 12:17, 16 May 2007 (UTC)
I think there may be conflicts if you parse carefully but I'll agree with all above anyway. The key point here is that admins in good standing who voluntarily take a break ought to be able to get their bit back without too much fuss. I'd favour a request to a local 'crat and nothing more. ++Lar: t/c 13:37, 16 May 2007 (UTC)
  • reset & thinks

So are we trying to get through a deadminship policy and then tackle re-applying or can we do re-applying (based on the above section) afterwards (I'll happily work on it)? --Herby talk thyme 13:42, 16 May 2007 (UTC)

Seems to me the two are intertwined, for voluntary. Else what's the need of a voluntary deadminship policy  ??? ... anyone can deadmin themselves by turning up on Meta and saying "deadmin me please", with the proper crosslinks to show they are who they say they are... our policy ought to speak to what happens next, and that's about all it needs to say. If you see what I mean. Now, INvoluntary is a different kettle of fish I guess. Am I making any sense here??? My flight got me into the hotel at 1 AM last nite so it's very possible I am not. ++Lar: t/c 14:35, 16 May 2007 (UTC)
OK - what I thought we were do was coming up with a "de-sysoping through inactivity policy/process" (& I thought we might be nearly there - I sense Fred may want to deal with some).
I do think there should be guidance (policy?) for the re-sysopping of those of "good standing" who leave voluntarily (I kinda like to "no objections" route personally).
But do you (all - Lar after sleep!)want me to try and tie the two together? I'll have some time tomorrow I hope. --Herby talk thyme 14:45, 16 May 2007 (UTC)

Adminship is not an honorary title. It is merely an access level and is no big deal. It can be taken away and reinstated on request. -- Cat chi? 19:52, 16 May 2007 (UTC)

  • reset

Taking this forward. I see two different issues here

  1. Removing admins who are no longer active
  2. Dealing with previous admins who have relinquished the tools voluntarily and are seeking to return

Personally I do not see these as connected in a policy sense. So - I would like to complete the inactivity policy sooner rather than later? --Herby talk thyme 10:22, 17 May 2007 (UTC)

Possibly guidelines for dealing with returning users who desysopped at their own request[edit]

I think a hard and fast policy here would be wrong. "Guidelines" might be useful. If I were writing them they might go

Should a user who has been desysopped at their own request wish to become a sysop again they must first ensure that they are up to date with current policies. It would be expected that they would show some quantity of edits too. In the first instance they should approach an active 'crat to request a review of their position. The bureaucrat could decide based on the user's previous standing that the tools could be returned to them (possibly an exceptional case). Alternatively the bureaucrat could approach other active sysops and solicit their views before returning the tools. Finally the bureaucrat could direct the user to a conventional RfA

Maybe/Maybe not? --Herby talk thyme 10:22, 17 May 2007 (UTC)

I like it. Would be a good idea to make sure our bureaucrats are comfortable with potentially making such decisions too, though. --pfctdayelise (说什么?) 01:22, 18 May 2007 (UTC)
I'm pretty sure Fred is watching this page but you are right, it would need running past bureaucrats for sure. Personally I think it would be exceptional that a granting of rights again would take place in such a way but it would be an option. What I would like to do is clear up the deadmin thing first & then maybe put some more time/effort into this aspect of granting sysop rights --Herby talk thyme 11:37, 18 May 2007 (UTC)
I don't like it, I prefer to stick with my earlier suggestion: that all request for regained adminship are to be posted at the COM:AN. If there are no outstanding complaints against it, any bureaucrat can grant the request after 24 hours. It would: (a) give further input, (b) be better to not be forced to rely on a sole bureaucrat (who might not be active, or know the user beforehand, etc etc), (c) a safety valve against admins who have returned for wrong reasons, (d) it would inform the community about the admin's request. / Fred Chess 22:00, 18 May 2007 (UTC) edited by Fred Chess 16:37, 19 May 2007 (UTC))
As long as I have your attention, I would like to bring up Yet Another Issue. So far, when an admin is deadmined, he has been removed from commons:List of administrators (unless that is forgotten, which has happened). Sometimes users are also commented out, especially from the list of admins per date (and often deadmined admins aren't even removed from that list, nor the list of admins by language). It is probably useful to make a list of former admins, like they have on English Wikipedia, and also the reason why they were deadmined. It should be used for reference purpose. Is anyone up for starting the page Commons:Former administrators? / Fred Chess 22:15, 18 May 2007 (UTC)

From comments[edit]

  • Pictogram voting comment.svg Comment With this proposal, a sysop would be encouraged to carry out admin actions in order to keep his admin abilities. It's not a matter of play. If you don't check the appropriateness of the admin actions, any counting is useless. I wish a new proposal which endorses the current policy, in order to implement it. --Juiced lemon 10:15, 1 June 2007 (UTC)
  • So write one.   — Jeff G. (talk|contribs) 14:28, 1 June 2007 (UTC)
  • Of course administrators using their tools irresponsibly should have their privileges revoked. The risk that privileges are abused (by the administrator or by someone using the administrator's login details) increases with the number of users who have been granted adminship. The risk that an administrator's account is taken over and abused without detection increases as the administrator's activity decreases. Thus, the purpose of the proposal, as I understand it, is to reduce that risk before it is realized when this can be done without limiting active contributors. This proposal doesn't negate the ability to de-op users whose privileges have already been abused, but since inactivity is easier to detect automatically than questionable administrative actions, I think we have a good opportunity to prevent abuse proactively. LX (talk, contribs) 14:08, 3 June 2007 (UTC)
  • LX makes a good point - we need to include the standard clause for deadminship of an active admin, ie a consensus deadmin poll. The current wording suggests the only way anyone is ever desysopped is via inactivity. pfctdayelise (说什么?) 14:32, 3 June 2007 (UTC)

Comments by Fred[edit]

About the "Return to Adminship". Well Herby, you already know how I want that to be organized (at least I have expressed my opnion on it on my talk page and on the admin de-admin page) so I don't have to comment on it.

About the requirements -- I like them. It is important to note that the lower limit for edits should not be there as a formality -- it should just be there as a snowball clause -- so we don't have to vote for newbies all the time. It probably doesn't hurt if a minimum of 50 edits are required, but it is ok as it is -- if it turns out that lots of newbies request adminship it can be added afterwards.

Forceful de-adminship. Hmm, I don't know if you read the Jimbo interview where he said that one of the most difficult things all new Wikipedias go through is to expell experienced users. I suppose it might happen that a user is really not suitable as administrator but doesn't admit it himself. What to do then?

  • The simplest solution is to have a big polling thing for the user in question (e.g. at COM:A, calling it "requests for de-adminship"), and if 75% vote for deadminship, the user will be de-admined. Unfortunately these things take a lot of time and energy from too many users. Things tend to get very ugly soon. But, if nothing else, this works.
  • Another system is to have admins re-affirm their adminship on a regular basis, a-la meta. It is a friendly way to get rid of unsuitable admins, and a way for the community to re-affirm their trust in the admins. The disandvantage is that voting all the time takes time from useful stuff.
  • An ArbCom could also be established. The disadvantages are that it is complicated to build it up and also the voting process for arbitrators takes effort. But as a Wikipedia procject gets big, it is probably unaviodable.

Fred J 20:43, 25 June 2007 (UTC)

Response to Fred guessing you are watching this!
Snowball clause - below 100 edits would not be accepted? 50 seems a bit low to me. However is what the edits actually are that I think sould be considered (I could easily clock up over 100 edits in a day - I have done - by welcoming newcomers on Wikibooks) but I'd like to see a cross section of editing together with user interaction which, to me, is important.
As to the "return to admin" are you happy with 24 hours notice? Will people complain that they didn't get time to see it? And should there be an "acceptable" period that they can be away for - 6 months - fine, 2 years - probably not?
Of the de-admin options I am personally a fan of the meta system (it would also solve the inactivity one in possibly a better way too) but would it get support? Could we make it so simple and speedy that it didn't make for work? The little experience I have had of Arbcom suggests that it can be heavy handed so I thought a "process" might be best but I'd be happy to go with "confirmation" if you think it is worth a try. --Herby talk thyme 11:17, 26 June 2007 (UTC)
The current policy states at least 200 edits, and that is for me an absolute minimum.
Secondly, like I probably posted on other pages, I would prefer that somebody who has volunteerly laid down his adminship to just make a request on Commons:Administrators/Requests and votes. This makes the request much more visible than on the AN or VP, where such a thing could easily get lost. If the bureaucrats decide that handing back the admin tools does not require a full 7 days, they can decide to just close and promote. If they decide that more community input is required, they can just leave it open.
Thirdly, it often occurs that consensus on an RFA is just or just not reached, and that bureaucrats decide to have the RFA open for more than 7 days (geni, evrik). I think we should state in the policy that bureaucrats have this option. -- Bryan (talk to me) 18:45, 1 July 2007 (UTC)
OK I think I've addressed these now. The 100 edits is not a minimum but the fact they can be speedily closed at that level? Cheers --Herby talk thyme 12:00, 2 July 2007 (UTC)
Ah well, candidates with less than 500 edits are unlikely to succeed anyway, so I guess this is sufficient. Cheers, -- Bryan (talk to me) 12:46, 2 July 2007 (UTC)
Gosh, I hope we aren't getting too much tougher that way... when I ran I had well less than 500 edits and I only had one oppose. ++Lar: t/c 18:37, 9 July 2007 (UTC)
Well, the community evolves up to a point where you don't know everybody anymore... But I agree that there are certain ways our RFA should not go *points to some large wikis*. -- Bryan (talk to me) 18:54, 9 July 2007 (UTC)

Can bcrats remove user rights?[edit]

I thought it was the case that only stewards can do this, or am I mistaken? pfctdayelise (说什么?) 06:00, 4 July 2007 (UTC)

Nope - stewards only (or those even higher!). Did I get something wrong in the page (I'll go re-read but..)? --Herby talk thyme 06:51, 4 July 2007 (UTC)
Sorry, my understanding of the higher-level functions around here is just what I wrote, based on bits and pieces read and remembered here and there - please feel free to build upon it, and correct if/as necessary. Thanks!   — Jeff G. (talk|contribs) 04:59, 8 July 2007 (UTC)
I believe that only stewards can grant or remove Checkuser privs, even crats cannot grant them. See meta:Requests_for_permissions#CheckUser_access although I admit I may have misread it. Also 'crats cannot remove sysop permissions either, that too requires a steward. (although I don't know about removing bot flags, I was pretty sure crats can do that). Note that we don't have oversight here at all, right? At least I did not think we did. ++Lar: t/c 16:45, 9 July 2007 (UTC)
I think that basically stewards are the only ones that can remove rights BUT I know a bot flag was revoked by an en Books 'crat the other day. No one with oversight as far as I know (& stewards are not enthusiastic abut handing them out) - except in really legal cases selective restore of a delete by admins pretty much does the job --Herby talk thyme 18:12, 9 July 2007 (UTC)

Bot flags are the only ones that can be revoked by a bureaucrat. A bureaucrat can grant sysop, bureaucrat and bot status. Revoking of statuses and granting checkuser, oversight, boardvote, steward and other rights is something only stewards can do. -- Bryan (talk to me) 18:51, 9 July 2007 (UTC)

De-adminship other than through inactivity[edit]

Herby felt that this part was holding up adoption of the rest of this new stuff which, I feel, is good stuff... So he removed it, to try to unblock things... In a somewhat radical move for me, I reverted his removal, because I feel there is value in that bit too... howwever if it REALLY is holding up adoption, anyone should feel free to revert ME. That said, waht do we need to do to move this forward? Juiced lemon raised some objections... here: Commons:Administrators'_noticeboard#Inactive_admins and here: Commons:Village_pump#The_cunning_proposal... So how do we move forward? ++Lar: t/c 12:52, 18 July 2007 (UTC)

If issues of logistics could be solved I'd favour some form of "confirmation" meta style. Just my 0.02 (let's face it the logistics of involuntary de-admin are pretty bad!) --Herby talk thyme 13:02, 18 July 2007 (UTC)
What are the issues? Other than that people didn't know going in that they'd need to get reconfirmed? Or is it just the mechanics of when and how to hold the confirmation discussions? ++Lar: t/c 16:39, 18 July 2007 (UTC)
The mechanics to me (I have no problem with retrospective rules that are the will of the community). I still feel there is real validity in "having the tools while you need them". This would be a relatively harmless way of allowing those who weren't that interested to be relieved of the tools while allowing the community to keep those it wished to. I guess the fact that it only works at meta (?) may mean that others wikis find it less than useful but here we need admins - there are plenty of backlogs BUT the other side of that it they do (IMO) need to be reasonably active. The preceding unsigned comment was added by Herbythyme (talk • contribs) at 18:34, 18 July 2007 (UTC)

Okay, so this part badly needs fleshing out. User:Lar has suggested that de-adminship proceedings should run for at least a week, as for a regular adminship request. I agree with this in the general case, but exceptions should be possible in urgent cases so that we don't get a disgruntled administrator who can run amok unhindered for a week. Hopefully, such a provision should never have to be invoked, but formalising it would be a good precaution.

As I have also commented elsewhere, we should have a procedure that clearly requires the nominator to name specific deviations from the expected behaviour of administrators (listing examples and counter-examples) with reference to diffs or log entries to eliminate bad-faith nominations or at least give other administrators than the nominated one the mandate to speedily close them.

In my opinion, the requirements to overthrow a previous decision to grant adminship should be as high as the requirements placed on the original decision, meaning an administrator should not have his or her privileges revoked unless there is a three-quarter majority in favour of de-adminship. LX (talk, contribs) 23:04, 3 November 2007 (UTC)


You folks seem awful worked up about what people with admin tools should do, how often they should do it, and so on. Aside from the annoyance of having a bunch of people listed as admins who are MIA for whatever reason, why do we need this? --SB_Johnny | PA! 00:23, 16 August 2007 (UTC) (who is known to play devil's advocate from time to time)

Try some more![edit]

OK - let see if we can get this one finished and going soon.

To me there are two main aspects regarding admin policy that are not yet dealt with

  1. Activity levels from CheckUsers and 'crats
  2. Some methodology to deal with de-admining of those who do not fall into the inactive category

For the first aspect, as we are using 5 admin actions over 8 months to judge activity, we consider this a possible measure for both 'crats & CU?

For the rest I would still like to hear views on using a Meta style confirmation process?

Looking forward to a flood of views --Herby talk thyme 12:28, 5 October 2007 (UTC)

Should we rename this page?[edit]

It's not just about admins, so I was thinking Commons:User rights (or Commons:User rights policy) would be a suitable name. →Rocket°°° 17:37, 14 February 2008 (UTC)

Failed proposal[edit]

Proposal tagged with {{rejected}}, as no activity for four years. The present COM:ADMIN guideline seems to cover this adequately; attempts to revive parts of the proposal might be best as amendments to that. Rd232 (talk) 17:50, 26 March 2012 (UTC)