Wikipedia:Requests for comment/Impeachment of Functionaries

Statement of the question
In a recent arbitration case, a proposed remedy was to remove checkuser and oversight permissions from a user on the basis of other proposed findings that suggested he was guilty of "repeated and extensive edit-warring". In that discussion, one arbitrator commented "I have been thinking about this, especially given Wikipedia:Functionary's appearance, which tells me we're really moving into uncharted territory. As such, it is (possibly) not a given that the proposal is in line with community wishes (although likely). Maybe this is worth a separate RfC and is more apt for the community to decide than us."

The purpose of this RfC is NOT to comment on the user issues involved in that case, but to seek the community's views as to:
 * Under what circumstances should arbcom remove functionary rights (checkuser, oversight, functionaries mailing list, and perhaps also bureaucrat status)?
 * Should removal be restricted to instances where the specific tool has been abused?
 * Does the community agree that "Editors, such as arbitrators and operators of the CheckUser and Oversight tools, who hold positions of public trust or privileged access within the project are expected—more so than even other administrators—to serve as examples of good conduct for their fellow editors, to uphold the high trust placed in them by the community, and to avoid engaging in conduct unbecoming their positions."?

I repeat, this is not a userconduct RfC, but is rather to seek the community's views on what action arbcom should take, if any, when a functionary is held by them to have significantly transgressed editing norms.--Scott Mac (Doc) 17:13, 8 May 2009 (UTC)

Personal view by Scott Macdonald
Functionaries are human too, and need to be allowed to make mistakes. They tend to have high numbers of edits, and so will at times be involved in disputes where they don't handle themselves as well they might. Dismissing experienced and diligent functionaries because they make mistakes unrelated to their tools, or when are oft-times provoked, is unhelpful to the project.

However, there are sufficient nasty accusations that will inevitably be levelled at prominent people, that we need to know that when they are scrutinised they do enjoy the confidence of the community, and are among our best most level headed users, who understand neutrality, verifiability and proper editing. Thus where a functionary seriously or repeatedly falls below the standard of an experienced editor to the level that arbcom feels blocks, bans or editing restrictions are required, I do believe their continued possession of functionary tools becomes untenable.


 * Users who endorse this summary:
 * 1) --Scott Mac (Doc) 17:13, 8 May 2009 (UTC)
 * 2) Absolutely, and overdue. rootology  ( C )( T ) 17:35, 8 May 2009 (UTC)
 * 3) Agree that admonishments, blocks, topic bans and other type of sanctions given by the Community or ArbCom indicate that an user should not continue to possess functionaries tools.  FloNight&#9829;&#9829;&#9829; 17:43, 8 May 2009 (UTC)
 * 4) Indeed. Friday (talk) 18:16, 8 May 2009 (UTC)
 * 5) Fine. Stifle (talk) 18:38, 8 May 2009 (UTC)
 * 6) Seems reasonable. There is certainly a level of trust we expect from functionaries and if Arbcom has to resort to blocks and such, that trust no longer exists.  Icestorm815  •  Talk  18:51, 8 May 2009 (UTC)
 * 7) Definitely! -- Kim van der Linde at venus 19:26, 8 May 2009 (UTC)
 * 8) Yes. Something like repeated blocks would be grounds for removing functionary status.  Tiptoety  talk 19:41, 8 May 2009 (UTC)
 * 9) Tentatively yes. My musing is that do we think this is any more of the case than with anyone who holds an admin position on wikipedia? i.. the idea of WP being as egalitarian as possible - this retroactively seems to reinforce the idea of a cabal (??) Casliber (talk · contribs) 20:46, 8 May 2009 (UTC)
 * 10)  Majorly  talk  21:16, 8 May 2009 (UTC)
 * 11) Trust is everything for functionaries. LessHeard vanU (talk) 21:29, 8 May 2009 (UTC)
 * 12) Griffinofwales (talk) 22:15, 8 May 2009 (UTC)
 * Yes, even though I agree with Jimbo being put on the list if only for show (it'd be unlikely that Wikipedia revolts against the Sole Founder&trade;) Sceptre (talk) 00:57, 9 May 2009 (UTC)
 * Yes, though I would say any block or editing restriction explicitly supported by the community should also apply here. Mr.Z-man 06:37, 9 May 2009 (UTC)
 * 1) Endorse. Nothing is more intimidating than to get involved in a nascent edit conflict with an admin, checkuser or oversight (I don't even known what that is). If such a person has been argumentative to the point of being sanctioned, Wikipedia should withdraw all "officialdom" from them, lest their actions seem to have the sanction (using the opposite meaning of that word, heh) of the Wikipedia community. Resurr Section (talk) 08:37, 9 May 2009 (UTC)
 * 2) Yes, but... (see discussion)  Thatcher 11:57, 9 May 2009 (UTC)
 * 3) Agreed, although if a policy is created about this it needs to also make clear whether or not admins are "functionaries" or not; Functionary is kind of unclear right now. I support this statement either way. –Drilnoth (T • C • L) 13:21, 9 May 2009 (UTC)
 * 4) Yes. It is already within ArbCom scope that a user is desysopped - I assumed that extended to checkuser and other sysop functions. It would seem odd that ArbCom could ban and desysop someone who then can continue as a checkuser!  SilkTork  *YES! 01:27, 10 May 2009 (UTC)
 * 5) Yes. A reasonable person would already assume that they should behave the way the most senior members of the community behave. That is why it is important for admins and functionaries to act as role models. Clayoquot (talk | contribs)
 * 6) Cla68 (talk) 23:59, 14 May 2009 (UTC)
 * 7) — Ched :  ?  23:46, 16 May 2009 (UTC)
 * 8) Still would like clarification on what exactly a functionary is. Xymmax  So let it be written   So let it be done  14:59, 18 May 2009 (UTC)
 * 9)  Them  From  Space  04:42, 19 May 2009 (UTC)
 * 10)  hmwith  τ   10:01, 24 May 2009 (UTC)
 * 11) Functionairies are users of tools, plain and simple. Current ArbCom policy is that admin rights are only stripped if they were used improperly. The same yardstick should be applied here. It also is common practice to strip admin rights in cases of massive breeches of trust and integrity (which brings to mind a story of an admin who was de-admined for meatspace stalking of another editor) and that standard should be applied too. Those two standards should be the things to keep in mind.--Ipatrol (talk) 03:02, 25 May 2009 (UTC)
 * Yes, as well as most of the below being reasonable as well. It seems clear that if functionaries can't behave they shouldn't hold the extra tools any more. I don't believe a lot of handwringing is necessary to decide specifics of policy, since that time is better spent elsewhere. Just that arbcom knows they are empowered by the community to ask for the removal of rights when there are findings of improper conduct, etc. Use common sense. - Taxman Talk 17:08, 26 May 2009 (UTC)
 * 1) Sole Soul (talk) 19:45, 16 October 2009 (UTC)

Personal, simple proposal by User:Rootology
An even simpler, and more straightforward take on this. These positions are positions of widespread community trust. Trust is not permanent, nor eternal, nor are any of these positions 'rewards' intended to be kept for a lifetime.

Why: A Functionary that is not trusted by his peers--his peers, being all editors--cannot be trusted with the tools, position, or authority. Trust is earned, but must be retained with high standards of ongoing behavior in all your actions. Stray from the righteous path too many times, and removal should be as easy as obtaining them was to begin with. Will mistakes happen? Sure. Is the Functionary repeatedly being criticiized, investigated, and pulled before Arbitration, Dispute Resolution, or the soon-to-launch Audit group for their use of Checkuser/Oversight?

How: It's time for them to be made to step down. Earning the trust to get access to the higher level tools is one thing; you have to work to keep that trust. The simplest way to accomplish this is that if a Functionary repeatedly goes astray, then the community can bring them before WP:RFAR. If this happens, for this specific purpose, the Committee is required to either reject the request with reasoning why, or else 1) do a deciding public motion or 2) open a public case to review the removal of access, rights, and title. The cases must be public so each sitting AC member has to account. Private data can be private evidence, but the 'voting' and outcome must to be on the record.

Who: Any Functionary, up to and including sitting and former AC members, and Jimmy Wales (who is a Functiorary due to the fact we allow him to have some extra-special tools on en.wikipedia.org). None of us are above our own laws.


 * Users who endorse this summary:
 * 1) rootology  ( C )( T ) 17:46, 8 May 2009 (UTC)
 * 2)  Majorly  talk  21:23, 8 May 2009 (UTC)
 * 3) Trust (An interesting sidenote; Rules made by kings apply also to kings;which only works when the judges are not creatures of the king.) LessHeard vanU (talk) 21:34, 8 May 2009 (UTC)
 * History shows that a court need not be above a defendant, as long as there is support from below. Happy‑melon 22:04, 8 May 2009 (UTC)
 * you might want to check on what happened to those of his judges who were unfortunate enough to survive until his son restored the monarchy. DGG (talk) 23:23, 8 May 2009 (UTC)
 * Let's not forget the fate of his prosecutor (discussed in detail in Geoffrey Robertson's The Tyrannicide Brief) either. EdChem (talk) 07:44, 9 May 2009 (UTC)
 * And the story is the same in the French and Russian revolutions; those responsible for the overthrow usually didn't make it through the next decade. The issue is not that the court per se is vulnerable, but that whoever topples the existing power structure tends to try to kick it as much as possible while it's down; and a revolution is a great opportunity to get revenge for perceived wrongs.  It's not entirely applicable to Wikipedia, but the principle is a constant.  Fortunately, it's probably not something we need to worry about; no one's likely to get guillotined over here... Happy‑melon 07:57, 9 May 2009 (UTC)
 * 1)  hmwith  τ   12:22, 11 May 2009 (UTC)
 * 2)  Skomorokh   01:24, 12 May 2009 (UTC)
 * 3) Cla68 (talk) 23:59, 14 May 2009 (UTC)
 * 4)  Them  From  Space  04:44, 19 May 2009 (UTC)

Thoughts by User:Jtrainor
I don't see why this is complex. It should work as follows:

Step 1: Did the person in question abuse the power they have?

Step 2: If yes, remove them, and take further action as neccesary. If no, it's all good.


 * This is (of course) right in principle, but that doesn't make it magically workable. Who should be responsible for deciding the question in step 1? ArbCom? The body of Bureaucrats? The Admin body? The Community? Some random IP we grab off RecentChanges? When a complaint involves CheckUser or Oversight, the community in general may not be able to see or be given all the information needed to form a balanced decision.
 * If a decision is made that the rights have been abused, "removing" them is not trivial. Only Stewards have the ability to remove rights, and the Steward policy explicitly states that "Stewards do not make decisions... [they] implement valid community consensus". We can make local policies that govern Steward actions, but they must be explicit. Stewards will not listen to appeals to WP:IAR or WP:SENSE.
 * "Take further action"... this would consist of what? Blocks? Bans? Sanctions? ArbCom has authority to impose blocks and bans, as does Jimbo, as does the community. All three have roughly equal weight in the sense that a ban from one can be validly appealed to another.  Who is going to decide this aspect of the case?
 * In general, then, principles are fine, but will not do the job in a complicated and dynamic community like enwiki. So many different parties are involved in processing a complete complaint that expecting them to happily work together to the 'right' outcome without any guiding framework, is rather optimistic. Happy‑melon 19:14, 8 May 2009 (UTC)


 * Jtrainor, actually it is more complex. Traditionally, there has been no mechanism to remove special access tools unless the person violates the privacy policy, oversight or checkuser policy. But sometimes the user will not directly abuse a specific tool but over time they have received sanctions that would have almost certainly disqualified them from getting the tools in the first place. Since most use of the special access tool is done in private, then there is also suspicion that private actions of the user parallel the public problematic actions. This type of suspicion makes it hard for many users to accept the actions and reports made by the user. Since there is no good way to confirm that no abuse is happening, then conspiracy theories of cover ups occur. For these reasons, I think that users sanctioned for problematic conduct outside of abuse of their status still should have their access removed even if no direct abuse of a tool happens. FloNight&#9829;&#9829;&#9829; 19:20, 8 May 2009 (UTC)

Personal view by LessHeard vanU
The only criteria is "erosion of trust". Functionaries are typically long standing editors of good repute, often already administrators. Until recently functionaries did not have to undergo peer review to have the particular flags enabled, and therefore might not be considered as requiring the fairly stringent processes to be deflagged as is the current situation with sysops. Should there be any indication that the level of trust expected of a functionary be deprecated - in any form - that a deflagging is appropriate. Should subsequent events indicate that such mistrust was misplaced or manipulated then the flags may be restored - in any other case the former functionary may apply for return of the flags by the processes now existing. LessHeard vanU (talk) 21:26, 8 May 2009 (UTC)


 * Users who endorse this summary:
 * 1) LessHeard vanU (talk) 21:26, 8 May 2009 (UTC)
 * 2)  Majorly  talk  21:31, 8 May 2009 (UTC)
 * 3)  hmwith  τ   12:24, 11 May 2009 (UTC)

Personal view by AndrewRT

 * Under what circumstances should arbcom remove functionary rights (checkuser, oversight, functionaries mailing list, and perhaps also bureaucrat status)?


 * This has got to be a case by case decision. Ulitimately, just ask yourself - knowing what is known today, would these rights have been granted?


 * Should removal be restricted to instances where the specific tool has been abused?


 * No. If someone's behaviour breaches behavioural policies or abuses tools, that has an impact on the overall trustworthiness of that person. That may be relevant to whether or not they continue to hold fuctionary rights - even if those specific rights haven't been abused.

Does the community agree that "Editors, such as arbitrators and operators of the CheckUser and Oversight tools, who hold positions of public trust or privileged access within the project are expected—more so than even other administrators—to serve as examples of good conduct for their fellow editors, to uphold the high trust placed in them by the community, and to avoid engaging in conduct unbecoming their positions."?


 * Yes.


 * Users who endorse this summary:

View by Stifle
You need to work out what a functionary is first. Stifle (talk) 09:55, 11 May 2009 (UTC)


 * Users who endorse this summary:
 * 1) A functionary is a lot like John Kerry. Sometimes they lose, sometimes they're foreign relations chairmen. &mdash;harej 01:35, 12 May 2009 (UTC)
 * 2) Endorse. This is being argued elsewhere, but whether by fiat or consensus, this must be determined before this discussion can make any sense. KillerChihuahua?!? 09:59, 13 May 2009 (UTC)
 * 3) I probably would endorse something more, if I was certain of the definition of the terms being used.  MBisanz  talk 15:57, 13 May 2009 (UTC)
 * 4) I thought it was anything "higher" than a 'new' admin but below WMF-level. OTRS is not Functionary, WMF employees aren't functionaries (they have their own impeachment system, e.g. firing). Unblock-en access, IRC sysop/admin, Checkuser, Oversight, Arbs, ex-arbs. anything new or current that is "higher than normal" access. This is what I just posted about that. Basically, if you're not a Functionary, you don't have anything for access (including mail lists+rights) that any 'new' editor or 'new' admin wouldn't have on the day of their promotion or registration. Defrocking a Functionary in my mind is a demotion to either Level 1 Editor, or Level 1 Admin, to quantify it. rootology  ( C )( T ) 16:03, 13 May 2009 (UTC)
 * 5) Exactly. Xymmax  So let it be written   So let it be done  15:00, 18 May 2009 (UTC)
 * 6) I've been contributing to Wikipedia for ::mumble:: years now, & this is the first time I've heard this word "functionary" used. I guess I'm more out of the loop than I thought I was. -- llywrch (talk) 21:59, 19 May 2009 (UTC)
 * 7) The lack of common knowledge of an actual definition is frustrating.   hmwith  τ   10:03, 24 May 2009 (UTC)


 * 1) The issue of what is a functionary needs to be cleared up before further discussion. I also believe that the issue of what a functionary means should be sorted. Is a functionary a mere tool user? Or does it mean something more? Are functionaries just admins with more keys? Or does NOBIGDEAL not apply here?--Ipatrol (talk) 02:50, 25 May 2009 (UTC)

View by RayAYang
There is a distinction between a breach of faith with the Wikipedia community, and simply being outside the mainstream as a Wikipedia editor.

Breaches of faith come in two kinds: the first is simply sufficiently outrageous behavior that we can't trust them anymore (such as, say, engaging in blatant vandalism, outing another editor, etc). Here, we know what to do: remove all the tools and block them as soon as we can.

The second is more subtle: using the tools they have been granted in a manner contrary to their purpose. An admin who blocks people they've engaged in content disputes with, a bureaucrat who moves usernames without following the proper procedures, a Checkuser who regularly scans people who argue with them, etc. Somebody who can't be trusted to use particular tools for their intended purposes, even if they believe they are acting in the better interests of Wikipedia, should have their access to those tools revoked (WP:IAR exceptions noted, when ArbCom agrees).

However, many a Wikipedian with enhanced privileges may be outside the mainstream, and content disputes on Wikipedia are not always friendly discussions. We may have (these are all hypotheticals -- I don't have any users in mind) rollbackers who may hold what the rest of us consider terrifically strong points of view on highly controversial political subjects and engage in frequent edit-wars (often just short of, if not over, 3RR), admins who are sufficiently inclusionist that they don't believe in notability criteria and use AfD as a frequent battleground for their views, abuse filter editors who are often in blissful ignorance of the finer points of copyright policy, and oversighters who may occasionally be just a bit bitey of newcomers. So long as their holding these opinions don't lead them to misuse the rights they have been given, and they demonstrate a good ability to separate their personal positions as editors and their responsibilities as holders of privileges, we shouldn't revoke their access.

I think that Wikipedia benefits strongly from having a lively and intelligent group of highly involved people, and it is both unfair, and damaging to the greater good, to subject people who've invested a lot in Wikipedia to an all-inclusive litmus test just to hold any extra technical privileges. Ray Talk 03:54, 12 May 2009 (UTC)

Extra simple proposition by Ipatrol
I think that the simple solution is that since ArbCom appoints functionaries, they should have the right to remove that position, either ad hoc or as part of a larger case on any matter. How they should seek consensus on the matter should be left for ArbCom to decide.--Ipatrol (talk) 23:10, 14 May 2009 (UTC)
 * Actually we now have (limited) elections for functionary positions, which muddies the water a bit. ArbCom is answerable to the community as a whole - they aren't an executive branch. As such, ideally things like this should be influenced by the communities opinions on the matter. I note that this RfC was inspired by a comment by an arbitrator saying that an RfC/similar would be useful to check that the process is in line with the wishes of the community, which implies at least one member of ArbCom things that their method for seeking consensus should be decided by us, not them. Ironholds (talk) 00:17, 15 May 2009 (UTC)

Techncially speaking, ArbCom is answerable to Jimbo Wales. The ArbCom elections are just the way Jimbo decides. I basically propose the same thing; Arbcom can appoint or impeach functionaries by its own judgement and may hold elections at their descretion to help come to a decision as it chooses.--Ipatrol (talk) 01:56, 15 May 2009 (UTC)
 * Isn't that what they already do?  Syn  ergy 22:38, 17 May 2009 (UTC)

Exactly! I'm saying we don't need to change anything!--Ipatrol (talk) 02:18, 18 May 2009 (UTC)