Wikipedia:Arbitration/Requests/Motions

Page semi-protected
From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Callanecc (talk | contribs) at 06:47, 11 March 2013 (→‎Motions with respect to functionaries: Both motions enacted). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

Motions


Motions with respect to functionaries

Preamble: In early January 2013, the Arbitration Committee reviewed several aspects of the appointment and review processes related to Checkusers, Oversighters and AUSC members, including the appointment extension of advanced permissions to former arbitrators. In preparation for this review, arbitrators retiring as of 31 December 2012 were permitted to retain Checkuser and Oversight permissions at their request on an interim basis until the completion of the review and decisions on next steps. Below are the motions that the Arbitration Committee will vote on; other motions may be proposed as well. All functionaries and community members are invited to participate in the discussion.


Motion on Audit Subcommittee

Audit Subcommittee (AUSC) members are provided with Checkuser and Oversight tools in order to carry out their responsibilities. Community appointees to the AUSC are discouraged from routine or regular use of either tool; however, they are permitted to use the tools in order to develop a sufficient skill level to adequately assess the actions of Checkusers and Oversighters, and may assist in addressing time-sensitive situations, or serious backlogs. Community AUSC appointees who held advanced permission(s) prior to their term will retain the permission(s) they held prior to their appointment. Community AUSC appointees who did not hold advanced permissions prior to their term may apply to retain Checkuser and/or Oversight during any Checkuser/Oversight appointment cycle that occurs during their term and, if successfully appointed, will assume their new role at the end of the AUSC term.

For this motion there are 13 active arbitrators, not counting 1 recused. With 0 arbitrators abstaining, 7 support or oppose votes are a majority.

Enacted - Callanecc (talkcontribslogs) 06:47, 11 March 2013 (UTC)[reply]

Support

  1. Risker (talk) 03:00, 20 February 2013 (UTC)[reply]
  2. I do not believe that members of the audit subcommittee should risk putting themselves in situations where their own use of the tools could be brought into question. Clearly uncontroversial circumstances, such as emergencies, should be fine, but in general I believe the less the tools are used, the better. The primary reason the tools are given to the members of AUSC (the community members, at least) is to allow them to review the logs; the way things are set up on our project, access to the logs comes hand-in-hand with access to the tools themselves. I also believe that arbitrator members should be subject to this as well, and so I would prefer that this motion be extended to include them, however this is at least (I believe) a step in the right direction. Hersfold (t/a/c) 03:04, 20 February 2013 (UTC)[reply]
  3. Kirill [talk] 04:29, 20 February 2013 (UTC)[reply]
  4. On the whole, "discouraged from routine or regular use of either tool" is tempered enough that I can live with it provided the exceptions listed are interpreted as exemplary and not exhaustive as is my current reading. — Coren (talk) 14:12, 20 February 2013 (UTC)[reply]
  5. Carcharoth (talk) 01:17, 21 February 2013 (UTC)[reply]
  6. WormTT(talk) 08:31, 21 February 2013 (UTC)[reply]
  7. Auditors need to remain aloof, yet we have not required them to be so; I think that needs to change. I don't accept the argument that this unjustifiably restricts the effectiveness of AUSC members; while it might restrict their ability to function as "regular" checkusers or oversighters, I don't think they should be performing such a function in the first place. AGK [•] 23:18, 21 February 2013 (UTC)[reply]
  8.  Roger Davies talk 03:59, 22 February 2013 (UTC)[reply]


Oppose

  1. Fixes a non existent problem, and reduces the discretion and judgment of the AUSC members. I would expect AUSC members to avoid controversial use of the tools, but this goes way too far in tying their hands. Courcelles 02:26, 20 February 2013 (UTC)[reply]
  2. Not particularly convinced that we need this. T. Canens (talk) 17:27, 21 February 2013 (UTC)[reply]
  3. NW (Talk) 21:24, 21 February 2013 (UTC)[reply]
  4. Substantially per Courcelles. I agree, as I think everyone does, that both arbitrator and non-arbitrator members of the Audit Subcommittee should best refrain from controversial uses of the Checkuser or the Oversight tools. But I don't see a need for them to refrain from routine uses (bearing in mind that the ability to distinguish between routine and potentially controversial use is a function of the cluefulness that would lead the community to elect someone as an arbitrator or an AUSC member in the first place). And the next time a serial vandal comes along and engages in massive on-wiki harassment that must quickly be addressed and undone, I see no benefit if six of the most knowledgeable checkusers and oversighters are virtually disqualified from helping with the problem. Newyorkbrad (talk) 20:40, 24 February 2013 (UTC)[reply]

Abstain

Comments from arbitrators

I'm of two minds about this. On the one hand, I believe this is a reasonable proposal given that the subcommittee members must also be seen as impartial regarding the use of those tools; but on the other hand, I feel that good recusal hygiene suffices to prevent us from loosing the help of three trusted contributors with those tools. I would be more inclined to support this if it counseled light usage of the tools (thus minimizing the risks) rather than discourage their use entirely. — Coren (talk) 02:19, 20 February 2013 (UTC)[reply]
There are problems with AUSC members being near the top of the activity table, as has happened on several occasions; it makes it difficult to assess whether there is too much work to be done, or whether AUSC members are carrying out tasks before those users who have been specifically vetted for those permissions have even had a chance to see them. Just as importantly, this formalizes the position that AUSC members must go through the community review process in order to retain CU or OS after their term.

This is an important separation between the functions. The community members are elected to oversee the issues, not to become active participants on a regular basis, which is what has happened on several occasions. We would be in a better position to assign them specific reviews if some of them weren't so busy using tools. Risker (talk) 03:00, 20 February 2013 (UTC)[reply]

Fluffernutter, just to be clear, I'm suggesting that all arbitrator members of AUSC be subject to this. You are very correct in that some of our most active functionaries have been members of the Committee; restricting the entire Committee from using their tools during their term would be a severe detriment to the project. Of course, we also have arbitrators who wouldn't know an IP address from a potato (the people I'm thinking of have more-or-less admitted this at one point or another), so I suppose for them it could be a moot point with regard to checkuser, and I myself tend not to use oversight very much mainly because the OTRS interface scares the willies out of me. Anyway, the point is, when it comes to managing who has access to the tools, there are enough members on the Arbitration Committee where one recused member wouldn't make a whole lot of difference in most cases. When it comes to simply reviewing them, there's only six people on the Audit Subcommittee; a recusal there can make a substantial difference when it comes to the final vote (or discussion, I've no idea how they do things). If we had substantially more active volunteers for both CU and OS, that may change things, but as things currently are, the loss of at least half a dozen or so additional active users of each tool (beyond that which we would lose from AUSC if the above passes) would be a fairly large blow to the project. Hersfold (t/a/c) 03:32, 20 February 2013 (UTC)[reply]
  • The problem with his motion is that it doesn't say something along the lines of "If you can pass it off to a normal functionary, or let it wait, do so" which would be reasonable guidance. Instead it tries to micro-manage their judgment, and doing it by using such meaningless terms as "time-sensitive". Most functionary tasks, are, by definition, time-sensitive to some degree (esp. the Oversight task); these aren't CFD closes that pretty much survive languishing until someone feels like dealing with them, if you're oversighting something, there's damned good reason, and that means sooner is better than later. Checkuser is often able to wait, through stopping persistent vandals often needs to be done now and not in three hours, but oversight tasks almost always should be done as soon as someone is there. This motion doesn't say pass it off if you can, don't wade into anything even remotely controversial, it says "see obvious problem, remember vague motion, walk away." Because are we really going to start judging not whether an action was proper under CU/OS policy, but whether it was or was not time-sensitive enough to justify acting? That's a bad road to go down. Courcelles 04:46, 20 February 2013 (UTC)[reply]
    • I hear your concerns, Courcelles. Let's look at Checkuser: I can't think of a good reason why an AUSC member would handle SPIs on a regular basis; they're generally not time-sensitive. About a third of the suppression requests are on old pages, or require second/third opinions; and those that have already been revision deleted (another good chunk) are generally not time sensitive and can wait for an hour or two. Indeed, I'd suggest that AUSC members shouldn't regularly be monitoring the Oversight OTRS queue, referring to it only when they are addressing a concern, or trying to work out some statistical information. I'd suggest that their use of the tools should be approximately equivalent to that of a steward: only when no other checkuser/oversighter can be located, and it's a suppression on a highly active page like ANI, that cannot be addressed with revision deletion. Risker (talk) 05:18, 20 February 2013 (UTC)[reply]

Discussion and comments from functionaries and community

  • In my two terms on AUSC, I've done this as a general best practice, so I don't think it's unreasonable to formalize it. However, I believe I am in the minority of current/former AUSC members in this practice. MBisanz talk 02:07, 20 February 2013 (UTC)[reply]
  • The trouble I see with this approach is that the caseload of AUSC (from what I understand) is extremely small. So in any case where an active functionary was appointed to AUSC (which in my mind is the ideal for at least some of that committee), the community would be trading hands that do work that needs doing in exchange for very occasional supervision. I understand the goal of discouraging AUSC members from being too heavily involved with tool use, but I can't escape the feeling that it nevertheless amounts to a net negative for a community which fairly commonly has backlogs of functionary-related tasks. A fluffernutter is a sandwich! (talk) 02:22, 20 February 2013 (UTC)[reply]
    Hersfold make a very good point in his vote. If the goal is that (as NE Ent phrases it) "those that audit should not do", it would make sense for all parties involved in auditing and supervising permissions use to be discouraged from using the tools. I still think that's probably a net negative (we rely on annual appointments of arbs and AUSC members to fill out the ranks of functionaries so work gets done), but if arbcom chooses to go the route of "auditors must remain neutral", it should apply to arbs as well, since they're the ones who ultimately give and take functionary permissions. A fluffernutter is a sandwich! (talk) 03:15, 20 February 2013 (UTC)[reply]
    Reply to Hersfold: Yeah, I realized after I saved that comment that given that the whole reason arbs are given CU/OS is to use them for arb work, it wouldn't make sense to flatly discourage/deny them use of the tools. However, I still think that if the goal is to keep those who supervise out of circumstances where their tool use might be one of the supervisory issues, it would be more consistent to discourage all arbs from using their tools in non-arbcom, non-emergency circumstances. This should go for AUSC arbs, at a minimum, but if the true concern is clean supervisory hands, it should also go for non-AUSC arbs. The fact that we lack the manpower to really take that approach sort of highlights the point that if we can't afford to do it with arb manpower, we probably can't afford to do it with AUSC manpower either, at least not without additional appointments of non-supervisory functionaries to take up that slack (which, come to think of it, would be a rather neat way to fix the entire "omg but manpower" issue regarding both arbs and non-arbs in supervisory positions). A fluffernutter is a sandwich! (talk) 03:54, 20 February 2013 (UTC)[reply]
  • Those that audit should not do, the solution to not enough doers is to get more doers, not blur the distinction between auditors and doers. NE Ent 02:56, 20 February 2013 (UTC)[reply]
    Agreed! As in "isn't that obvious?" (seems like it is not...) - Nabla (talk) 11:02, 20 February 2013 (UTC)[reply]
  • I personally feel that those that audit shouldn't be active functionaries for the duration of the terms. It not only is a question of neutrality, but also - perhaps I should say mainly - of an appearance of neutrality. Not being active users of the permissions they are entrusted to watch over would give a perception of a true super partes status and of the neutral court where one's grievances can be heard. One cannot truly be neutral when he's aware that a decision of his might have an impact on his own actions (if the Audit Subcommittee decides that an action taken by Checkuser X is disallowed, this will impact the Audit Subcommittee members' own practices and could even mean that some of their past actions would be disallowed and inappropriate). I would go so far as to suggest that the AUSC be given an ad-hoc userright that provides access to checkuser logs, checkuser and suppression logs (as I believe Checkuser might be needed in the course of their investigations) and that highlights their non-partisan and different role, much in the way as the Ombudsmen Commission does it (and yes, I am aware that now OC members can keep acting as local CU or stewards, but I believe that decision to have been taken in error, for again, the controllers should not be the controlled). Snowolf How can I help? 11:21, 20 February 2013 (UTC)[reply]
  • Technical question – if I understand Oversight correctly, those with the permission have two capabilities – the ability to oversight something, and the ability to see material that has been oversighted. I can imagine that the Audit role requires the second function, so they can review and confirm that the usages were correct, but it isn't obvious to me why they need the ability to oversight something. I suspect that giving them just the second half (i.e. read, but not edit) might not be technically trivial , but if it is possible should it be a consideration?--SPhilbrick(Talk) 14:24, 20 February 2013 (UTC)[reply]
    • This is the case with Checkuser as well - on the English Wikipedia, the "checkuser" and "oversight" usergroups contain both the ability to review the logs and use the tools (and in the case of oversight, view suppressed data). While I can imagine in some cases relating to checkuser it may be necessary to perform additional checks to verify that the checks in question were reasonable, that does not apply to oversight. It is possible to set up a usergroup that only has access to the logs and not the tools themselves, however to do so we'd need a developer's assistance and likely the Foundation's approval - it's a change that has to be made in the LocalSettings.php file used by Mediawiki, and the Foundation is (understandably) very sensitive about who can gain access to sensitive data. Hersfold non-admin(t/a/c) 19:17, 20 February 2013 (UTC) P.S. - Also, as the motion indicates, it can be helpful (especially with checkuser) to have some familiarity with how the tool is meant to be used so that an auditor has a better idea of what is a reasonable check and what is not. Granting them access only to logs prevents them from gaining this basic knowledge if they've never seen either tool before. Hersfold non-admin(t/a/c) 19:20, 20 February 2013 (UTC)[reply]
      • I did think about the possibility that there is some value in actually using the tool to gain more appreciation of how it is used. However, that can be accomplished by giving them the authority in the test wiki, if they want to try an oversight, so they can be sure they understand the process. While I understand it would take some work by a developer, I see value in letting the functionaries have the read and write capabilities in a test wiki, but only the read capabilities This isn't a major issue – if we have enough trust to choose someone as an auditor, it is highly unlikely they are going to abuse the tool, but it would be cleaner if we made the distinction. To use an analogy from the real world, we give financial auditors to ability to review checks issued by a corporation, but we would find it absurd to simultaneously give them the right to issue a check on behalf of the organization. Read and write access are very different, and in most systems, separable. We should separate them, and give them the read access.--SPhilbrick(Talk) 15:08, 21 February 2013 (UTC)[reply]
        • I don't think that the testwiki would really provide a good environment to give actual experience with the tool. That will help them see what the tool is capable of doing; however, it does not permit them to determine how a checkuser or oversighter would apply their judgment to determine when and how to use the tool. With checkuser in particular, there is a lot to be taken into account when looking for socks, and the checkuser must be able to correctly interpret the data to know if further checks are merited. Oversight does tend to be a bit more straight-forward, and that likely could be learned without actually using suppression at all. Hersfold non-admin(t/a/c) 18:43, 21 February 2013 (UTC)[reply]
          • I echo Hersfold's comments. Checkuser is very complicated and a bit of practice really helps figure out what is appropriate and what isn't. The way that I see it, letting AUSC members practice using the tools is a poor solution, but it's better than the only other solution, namely where AUSC would only recruit active checkusers. --(ʞɿɐʇ) ɐuɐʞsǝp 19:14, 21 February 2013 (UTC)[reply]
            • I grant that checkuser and oversight are different enough that the answer may be different. My comments were made solely thinking about oversight. I contend that oversight read access, possibility coupled with oversight write access in a test wiki is sufficient for an auditor. I'll defer to others about checkuser.--SPhilbrick(Talk) 19:23, 21 February 2013 (UTC)[reply]
  • Semi-related question: is there any way to remove a member of AUSC? --Rschen7754 18:02, 20 February 2013 (UTC)[reply]
    • Leaving reply down here, phooey on the separate sections. If there were reasonable cause to do so, I'd imagine the Committee could remove a member of AUSC using Level I or II procedures as appropriate, as we would with any other functionary who had seriously crossed some line. Hersfold non-admin(t/a/c) 19:12, 20 February 2013 (UTC)[reply]
  • @SilkTork: Noting here that per your comments I have marked you as recusing on the three motions. I draw your attention to the fact that this affects the majority calculations. Alexandr Dmitri (talk) 18:29, 20 February 2013 (UTC)[reply]
  • The Ombudsman Commission faced a similar situation recently; if a checkuser from a very quiet wiki (say, a wiki with only a small handful of checkusers) was appointed as an Ombudsman then that would be very detrimental to that project. To get around this, the size of the Commission was expanded and the requirement that Commission members not use their local checkuser rights was removed (although it should be noted that all of us, including myself, have either scaled back our usage significantly or stopped altogether). I'm not sure it's feasible to expand AUSC's membership (AUSC wouldn't appear very impartial if a significant number of the project's checkusers were in it), but that is something that could be considered as an alternative in the future. --(ʞɿɐʇ) ɐuɐʞsǝp 01:11, 21 February 2013 (UTC)[reply]
  • Personally, when it comes to the line "Community appointees to the AUSC are discouraged from routine or regular use of either tool," I think it may depend on whether or not the AUSCmember was an active OS/CU prior to AUSC service. I would think that someone who is already engaged in project-protection prior to AUSC service should not be forced to diminish/cease their activity because they are on AUSC, so long as they recuse from anything to which they are connected. -- Avi (talk) 18:42, 5 March 2013 (UTC)[reply]

Motion on CheckUser/Oversight and inactivity

The Arbitration Committee confirms the current procedures with respect to advanced permissions and inactivity as approved in March 2011, with the exception of retitling the provision "CheckUser/Oversight permissions and inactivity".

For this motion there are 11 active arbitrators, not counting 3 recused. With 0 arbitrators abstaining, 6 support or oppose votes are a majority.

Enacted - Callanecc (talkcontribslogs) 06:47, 11 March 2013 (UTC)[reply]

Support

  1. I'm not sure it's important that this be done – or that a motion to do so is worthwhile – but the change itself is innocuous and indeed a bit more accurate. — Coren (talk) 02:40, 20 February 2013 (UTC)[reply]
  2. It doesn't just change the title. It confirms that we reviewed this procedure, and that we still believe it is appropriate. Risker (talk) 03:01, 20 February 2013 (UTC)[reply]
  3. It basically just changes the title. The very presence of these motions confirm that we've reviewed our procedures, but whatever. Hersfold (t/a/c) 03:05, 20 February 2013 (UTC)[reply]
  4. I see Courcelles' point, but I think it's useful to affirm policy is still practice. Der Wohltemperierte Fuchs(talk) 03:22, 20 February 2013 (UTC)[reply]
  5. T. Canens (talk) 03:29, 20 February 2013 (UTC)[reply]
  6. Kirill [talk] 04:30, 20 February 2013 (UTC)[reply]
  7. Reaffirming the general provision after a period of time is not a completely useless gesture. Genuine housekeeping would be updating the wording per Philippe's comment below. I would support that as well, but such a wording change could be done without any formal motion. Carcharoth (talk) 01:19, 21 February 2013 (UTC)[reply]
  8. The community doesn't really see the activity standards being enforced (because that always happens off-site), so letting it know that the standards continue to apply is worthwhile. AGK [•] 23:18, 21 February 2013 (UTC)[reply]
  9.  Roger Davies talk 04:00, 22 February 2013 (UTC)[reply]
  10. Although as noted by many this motion is hardly essential, I see this as a reminder to everyone (including ourselves) of what the current policy is, as much as anything else. Newyorkbrad (talk) 20:42, 24 February 2013 (UTC)[reply]

Oppose

  1. Not in opposition to the change, per se, but opposed to the idea of motions that change nothing. Courcelles 02:37, 20 February 2013 (UTC)[reply]

Abstain

  1. For similar reasons as Courcelles. I'm not keen on the bureaucracy of ArbCom at the best of times and motions that change nothing embody this bureaucracy. However, I don't oppose the motion itself, so will abstain. WormTT(talk) 08:33, 21 February 2013 (UTC)[reply]
  2. Per WTT. If we can make the housekeeping changes to the point that Philippe and Avi bring up, then we don't need this. NW (Talk) 21:23, 21 February 2013 (UTC)[reply]

Comments from arbitrators

  • Retitling because our section on Removal of permissions refers to all types of advanced permissions, including administrator and bureaucrat. Risker (talk) 02:02, 20 February 2013 (UTC)[reply]
    • @MBisanz, it's housekeeping to be certain, but it's useful to publicly demonstrate that the change isn't just the whim of one arbitrator without discussion. Risker (talk) 02:22, 20 February 2013 (UTC)[reply]
    • @Philippe and others, some members of the Ombudsman Commission elect to not use their tools or to do so sparingly during their appointment. The wording used at present accommodates that, while not obligating them to refrain from tool use. I'd suggest an explanatory note as a "footnote" to the procedure identifying that refraining from use of tools is optional for Ombudsman Commission appointees, effective February 2013. Risker (talk) 03:56, 3 March 2013 (UTC)[reply]

Discussion and comments from functionaries and community

I'm not exactly sure why this clarification is needed, but sure, have fun continuing to do what you've been doing since 2011! MBisanz talk 02:10, 20 February 2013 (UTC)[reply]

  • I believe Ombudsmen may keep their bits now (at least the steward ones) so that last point may need to be reworded a tad. -- Avi (talk) 05:31, 20 February 2013 (UTC)[reply]
    • Avi is correct. The line that ends "or (c) holders who have temporarily relinquished access, including CheckUsers or Oversighters who accept appointment to the Ombudsman Commission" may be confusing. The Wikimedia Foundation currently does not require that those who hold the tools relinquish them for the duration of their time on the commission. Philippe Beaudette, Wikimedia Foundation (talk) 07:29, 20 February 2013 (UTC)[reply]
  • I am not sure why this is felt needed, or even felt that a full motion is necessary, but of course I trust the ArbCom's judgement that this deserves a vote. Snowolf How can I help?

A few points from me here.

  • I notice you're reconfirming this motion. However, before resigning my oversight rights on 29 December 2012, I had not used them for over eight months. I certainly qualified as inactive under this motion, so why were my rights not removed? Additionally, it didn't take me very long to find a person with checkuser rights who qualifies as inactive, and there may well be more. So my question is, are you actually enforcing this? Because, from my experience, it appears you are not, which makes me question why you reconfirmed this motion.
  • Five logged actions per three months does not qualify as "active" in my books, especially with checkuser where most SPI cases will have you make, as a minimum, four logged actions.
  • The section regarding the Ombudsman Commission should be reworded as it is not correct anymore. Ombudsman Commission members now retain their local rights as the requirement that they do not use them has been removed. Although the Ombudsman user group has global checkuser rights, we are obviously not allowed to use them unless it's on Commission business, so retaining local checkuser rights is the indicator that we are allowed to use them on that wiki. I would personally like for Ombudsman Commission members to be exempt from this provision for the duration of their appointment, but that is up to you to decide.

--(ʞɿɐʇ) ɐuɐʞsǝp 01:29, 21 February 2013 (UTC)[reply]

I did suggest that we adopt higher (and, as you argue, more accurate) standards of activity, but some of my colleagues disagreed—and in any case that suggestion does not appear to have made it into these proposals. I certainly agree that the current activity standards are a token gesture that are essentially useless in all but the most extreme cases of functionary inactivity. AGK [•] 23:18, 21 February 2013 (UTC)[reply]
I appreciate that that's difficult for everyone to agree on. My other point remains, however; to date this inactivity policy has not been enforced, so does this reconfirmation actually change anything? --(ʞɿɐʇ) ɐuɐʞsǝp 23:50, 24 February 2013 (UTC)[reply]

As for Ombudsmen, I think we should treat them the same as we do with AUSC members, where they are exempt from the activity requirements while they're in office like Deskana said above. This motion should be expanded to reword the sentence in part c so it's brought up to current practise. Techman224Talk 05:52, 23 February 2013 (UTC)[reply]

Motion on removal of CheckUser/Oversight for reasons other than inactivity (Level II procedures)

Motion 1

Holders of CheckUser and/or Oversight permissions are presumed to continue to have the confidence of the Arbitration Committee under normal circumstances. CheckUser and/or Oversight may be removed by the Arbitration Committee for reasons other than inactivity, including recommendation by the Audit Subcommittee, conduct inappropriate to a user trusted with CheckUser and/or Oversight permissions, or breach of the privacy policy or other related policy. Should the Arbitration Committee consider such removal, the holder of the permissions will be notified of this consideration, and will have the opportunity to respond to questions or concerns prior to a final decision by the Committee. Removal of advanced permissions may occur when:

(i) the holder of CheckUser and/or Oversight permissions has been notified of the concerns leading to consideration of removal of permissions and has had the opportunity to respond to those concerns, and
(ii) a majority of the committee supports the motion (irrespective of recusal or inactivity).

Note that Level I procedures for removal of advanced permissions applies equally to removal of CheckUser and Oversight permissions, and that this motion does not affect Level I procedures.

Appeal of removal of CheckUser or Oversight permission

Appeal may be made to the Arbitration Committee six months or more after the removal of CheckUser and/or Oversight permissions.

For this motion there are 13 active arbitrators, not counting 1 recused. With 0 arbitrators abstaining, 7 support or oppose votes are a majority.

Support

  1. Risker (talk) 03:30, 20 February 2013 (UTC)[reply]
  2. Kirill [talk] 11:28, 21 February 2013 (UTC)[reply]
  3.  Roger Davies talk 04:17, 22 February 2013 (UTC)[reply]

Oppose

  1. Holding a functionary position is a role that requires a great deal of trust. Someone untrusted by a significant part of the committee should not continue to hold such a flag, and to make a recusal, abstention, or even an arb we haven't heard from in a while a de facto no vote is just wrong. Courcelles 02:33, 20 February 2013 (UTC)[reply]
  2. No. This makes it harder to remove permissions for which trust of the committee at large is critical than it is for adminship. I could see a quorum being required (for this and other removals) but not something that effectively hobbles the committee after a handful of recusals or absences. — Coren (talk) 02:34, 20 February 2013 (UTC)[reply]
  3. Per Coren. An arbitrator recusing themselves from a vote is removing themselves from a vote. This procedure turns a recusal into an oppose vote in all but name, essentially making it impossible to remove oneself from a vote due to an actual or perceived conflict of interest. This also does not take into consideration the difference between a week's inactivity and an I've-fallen-off-the-face-of-the-planet-you-haven't-seen-or-heard-of-me-in-the-last-few-months inactivity, the latter of which has happened from time to time. The former should be treated as this motion describes; the latter should not. Hersfold (t/a/c) 03:16, 20 February 2013 (UTC)[reply]
  4. T. Canens (talk) 03:28, 20 February 2013 (UTC)[reply]
  5. WormTT(talk) 08:38, 21 February 2013 (UTC)[reply]
  6. I missed "irrespective of recusal or inactivity" when these motions were being word-smithed. I consider it to be quite problematic: in the event that one or more members are on long-term leave from the committee, this motion (if adopted and in a case where the balance of opinion is moderately divided) would prevent us from removing a functionary whose use of their permissions, while not abusive enough to warrant the use of Level I procedures, is nevertheless so subpar that removal under Level II is necessary. We should make CU/OS as easy (and difficult) to remove as is adminship. AGK [•] 23:18, 21 February 2013 (UTC)[reply]
  7. It seems silly that we could pass a motion to desysop someone but still not have the votes to remove checkuser or oversight from them. NW (Talk) 19:16, 23 February 2013 (UTC)[reply]
  8. In principle, recused or long-term inactive arbitrators should not count in determining the majority on any issue; thus I can't support this motion as written. However, where recusals or inactivity situations are inherently temporary (as occurred for example in the situation that arose during the election last year), the solution is simply to postpone permanently addressing the issue until afterwards. That is in essence what happened, and no harm done. If there were to be a significant number of instances in which the Committee is unable to act due to multiple recusals or the like, we would want to discuss and establish a backup mechanism (which would extend well beyond functionary appointments). However, in the six and one-half years in which I've been following the Committee's work, this has virtually never been an issue, and in the absence of real need, I am reluctant to add to what already can be described as elephantiasis of the procedures manual. Newyorkbrad (talk) 20:50, 24 February 2013 (UTC)[reply]
  9. Per NW and NYB. Carcharoth (talk) 01:01, 5 March 2013 (UTC)[reply]

Abstain

Comments from arbitrators - Motion 1

  • I note the comments of my colleagues and others. It should be *more* difficult to remove the tools from users who have already gone through three separate vetting procedures (acceptance of application, community review, final acceptance), including two by the Arbitration Committee, particularly when all of the evidence related to removal is by definition not available to the community for review and comment. It is very important that there is no possibility of removal for "political" reasons, which is very possible in situations where a significant number of arbitrators are unavailable for the short term. (Examples include travel to/from Wikimedia-related events like Wikimania, holidays such as Christmas or Thanksgiving weekends, where we frequently will have half the committee largely unavailable for several days in a row.) I agree with Fluffernutter that the Committee does need to address longterm inactivity on the part of arbitrators (my own inclination would be automatic removal of the arbitrator after x weeks of inactivity on Arbitration-related pages onwiki, or inactivity in three consecutive Arbcom cases); arbitrator inactivity has a deleterious effect on the Committee, especially if longtime inactive arbitrators suddenly appear (or disappear) in the middle of cases or other significant activities. The purpose of these motions is to address functionary-related matters, and I believe that the Committee should also address arbitrator inactivity, but arbitrator inactivity was not discussed as part of the functionary review and we do not currently have any consensus on how it should be addressed. Recusal is a trickier matter: while there have been improvements in the last term or so, recusal on matters has been used by some arbitrators to escape the fallout of having to address certain "hot potato" issues. There are very, very few situations where an arbitrator should recuse on a discussion of this import. At present, I can think of only one scenario where an arbitrator recusal would be necessary, and maybe one other where it might possibly be acceptable. Just as some see recusals as an oppose in all but name under this scenario, it can just as easily be seen as an automatic support if permitted as it reduces the number of real supports that are required to take an action. It is not a huge issue when the matter is public and the rest of the committee can appreciate the opinions and comments of the community; it is a much bigger issue when the discussion is entirely private and the community cannot evaluate and opine on the matters being considered.

    As to the point Fluffernutter makes about not "voting" on matters until there are enough people around to properly resolve them, all I can say is that in recent years we've had multiple arbitrators suddenly and unexpectedly make proposals without input from the rest of the committee, or publicly making different proposals than were made to the rest of the committee. It's bad practice, I'll be the first to admit, but it is essentially impossible to stop. It only takes one arbitrator posting a motion to set things on the path, even if the rest of the committee doesn't want to proceed that way, and premature motions do indeed often result in poor results. Risker (talk) 03:57, 20 February 2013 (UTC)[reply]

    I disagree. Those users have already gone through three separate vetting procedures because the tools are so sensitive. Checkuser isn't a badge of honor we should only remove gingerly once granted, it's a delicate tool that requires continuing utmost trust in its holders; it is – and must be – hard to get and easy to remove. — Coren (talk) 04:37, 20 February 2013 (UTC)[reply]
    Agree with Coren (again). I can think of a number of situations in which a recusal would be appropriate in this situation, and I'm not sure why a decision of this nature should be considered to be more important than (for example) voting on a case. While under normal procedures a recusal can affect the majority, the proposed motion above will cause it to affect the majority. See the table below, which assumes the Committee has 15 active members. Under current procedures, only odd-numbered recusals impact the majority (the first, third, fifth, etc.). Under the proposed motion, every recusal means one less person has to oppose the motion in order to make it mathematically impossible. If we have a situation such as occurred during the elections, where four arbitrators feel they must recuse from voting, that means a minority of four arbitrators - about a quarter of the Committee - can prevent the rest of the Committee from acting, even if all seven of the others support the motion. Under current procedures, the number of supports needed to pass a motion does decrease with recusals, but only half as quickly, and the voting is not unfairly weighted towards one outcome. Hersfold non-admin(t/a/c) 19:50, 20 February 2013 (UTC)[reply]
Table collapsed for readability
Number of recusals Number of opposes needed to fail Number of supports needed to pass
Under current procedures Under proposed motion Under current procedures Under proposed motion
0 8 8 8 8
1 7 7 7 8
2 7 6 7 8
3 6 5 6 8
4 6 4 6 8
5 5 3 5 8
6 5 2 5 8
7 4 1 4 8
  • I think the prospect that a small minority of arbs could take advantage of the absence/inactivity of a good number of arbs and do something bad is quite overblown and frankly inconsistent with AGF. In a number of our recent internal discussions, we've had little problem holding back proposals that are technically passing for votes from temporarily inactive arbs and potential vote changes. Moreover, if the lots-of-inactive-arbs problem requires a solution, the proper solution is simply to say that the committee would not act on non-emergency matters at all during those periods, rather than creating arbitrary heightened standards for some votes but not others. Surely, for example, banning a long-term editor is as serious a matter as removing a userright this committee granted in the first place, and which no one actually needs to edit? Does that mean that we need to pass a special procedure for remedies that bans someone as well? T. Canens (talk) 22:46, 20 February 2013 (UTC)[reply]
  • I recuse from voting on matters related to the advanced tools, and I have not taken part in the formal selection or removal process of functionaries, so I'm concerned regarding comments above that a recusal can equal a vote in some situations. If a Committee member is consistent in their recusal, rather than selective, then their recusal should not effect the vote. It would, surely, only have an impact if the member suddenly or inconsistently voted on a functionary issue. Following that line of thinking, my intention is not to take part in the voting on these motions, though I am open to being persuaded otherwise, if it is felt my vote is important. SilkTork ✔Tea time 09:29, 20 February 2013 (UTC)[reply]
    • NE Ent - Recuse is probably too strong a word. I historically haven't got involved. When I first joined the Committee we were actually quite busy, and few Committee members get involved in everything we do, and the thing that I decided not to get involved in was selecting the new functionaries as I was not familiar enough with the tools to make an informed judgement. Having not been involved in deciding functionary matters for over a year it seemed inappropriate to change that stance for an individual vote. And having not been previously involved and having recused on an individual vote, it seems appropriate to continue that stance, though I am open to the notion that I should get involved. I am simply interested in consistency - I have no personal objections to voting on the motions. SilkTork ✔Tea time 14:02, 20 February 2013 (UTC)[reply]
  • @snowolf: The ASUC reports are not strictly binding, but I can honestly not remember a single case of a report of that subcommittee that was not heeded verbatim by the committee at large. — Coren (talk) 14:16, 20 February 2013 (UTC)[reply]
  • There was one case where it turned out that AUSC based its report and recommendation on a factual mistake. T. Canens (talk) 20:01, 20 February 2013 (UTC)[reply]
  • Not sure about this yet. It does seem an imbalance that arbitrators only hold their positions for two years and have to stand for re-election after that period of time, but CU and OS holders don't have to sustain that level of scrutiny. I do think that functionaries should be rigorously reviewed at least every two years, more likely every year, and possibly some reconfirmation may be appropriate, both by the community and by the then-current ArbCom (who are able to assess the quality of CU and OS actions taken). I also think that former arbitrators, and those who have not gone through a vetting and community scrutiny process, should provisionally retain the rights if the then-ArbCom does not object, but should be required to apply formally through the existing process to retain the rights. Carcharoth (talk) 01:35, 21 February 2013 (UTC)[reply]

Motion 2

Holders of CheckUser and/or Oversight permissions are presumed to continue to have the confidence of the Arbitration Committee under normal circumstances. CheckUser and/or Oversight may be removed by the Arbitration Committee for reasons other than inactivity, including recommendation by the Audit Subcommittee, conduct inappropriate to a user trusted with CheckUser and/or Oversight permissions, or breach of the privacy policy or other related policy. Should the Arbitration Committee consider such removal, the holder of the permissions will be notified of this consideration, and will have the opportunity to respond to questions or concerns prior to a final decision by the Committee. Removal of advanced permissions may occur when:

(i) the holder of CheckUser and/or Oversight permissions has been notified of the concerns leading to consideration of removal of permissions and has had the opportunity to respond to those concerns, and
(ii) a majority of the committee, not including recused or inactive arbitrators, supports the motion.

Note that Level I procedures for removal of advanced permissions applies equally to removal of CheckUser and Oversight permissions, and that this motion does not affect Level I procedures.

Appeal of removal of CheckUser or Oversight permission

Appeal may be made to the Arbitration Committee six months or more after the removal of CheckUser and/or Oversight permissions.

For this motion there are 13 active arbitrators, not counting 1 recused. With 0 arbitrators abstaining, 7 support or oppose votes are a majority.

Support

  1. Dispels any uncertainty over the status quo. This also emphasizes the "opportunity to respond" part over current procedures, and creates a special appeal provision rather than "normal arbitration proceedings" in current procedures, which frankly can't usually be held for CU/OS related issues. T. Canens (talk) 14:26, 24 February 2013 (UTC)[reply]
  2. This works for me. Regarding the appeals procedure, it's probably not terribly ideal due to the echo chamber concerns Roger notes, but I think according to Foundation Policy we're the only body they can appeal to - the Arbitration Committee holds sole jurisdiction over who does and does not have CU and OS. Changing that would take a great deal of time and discussion with the Community and Foundation, and this will suffice in the meantime. For the latter half of the year, a delay of six months means that you're guaranteed to have a few new faces on the Committee anyway, so the appeal won't be heard entirely by those who voted for removal in the first place. Hersfold (t/a/c) 03:54, 25 February 2013 (UTC)[reply]
  3. This works, as does simply treating CheckUser and Oversight as no different than Adminship or Bureaucratship. NW (Talk) 18:52, 5 March 2013 (UTC)[reply]

Oppose

  1. I am concerned about the lack of effective appeal arrangements, which are greatly exacerbated by the lowered bar for removing permissions. As this is something which could be handled independently by the Ombudsman Commission, and as the commission's remit is currently under review with a view to expanding it, this motion is probably premature.  Roger Davies talk 06:53, 26 February 2013 (UTC)[reply]
  2. Current procedures are sufficient. We should leave this aspect for now and return to this later in the year. Carcharoth (talk) 01:04, 5 March 2013 (UTC)[reply]

Abstain

Comments from Arbitrators - Motion 2

  • Timotheus, I'm not certain what you mean by "special provisions". These are the current provisions, and indeed they're the current provisions for any case. The only difference is the description of reasons for contemplating removal of checkuser or oversight permissions. Risker (talk) 16:23, 24 February 2013 (UTC)[reply]
    • The current level 2 procedures says that "If the editor in question requests it, or if the Committee determines that a routine reinstatement of permissions is not appropriate, normal arbitration proceedings shall be opened to examine the removal of permissions and any surrounding circumstances." For obvious reasons, we can't open "normal arbitration proceedings", i.e., a case, for most CU/OS removals. The point is that this changes something over the status quo. T. Canens (talk) 16:39, 24 February 2013 (UTC)[reply]
      • Thanks, TC; that makes sense. Now, on the mailing list after the initial motion was posted, there was some talk of either (a) outlining a minimum period for such a discussion, to ensure that there was input from at least the overwhelming majority of the non-recused committee members (i.e., ensuring it extended long enough for those inactive for very short periods, such as holidays, to be able to participate) or alternately not initiating a motion until the end of periods where a significant number of arbitrators were temporarily inactive. (Examples of the latter were Christmas and Thanksgiving breaks when many are traveling and have limited access; and Wikimania, where multiple arbitrators usually participate and again have periods of travel and very limited availability.) Did you have any thoughts about including something along this line? For example, discussion before a motion to remove is posted and voted on? Risker (talk) 17:57, 24 February 2013 (UTC)[reply]
      • Interesting point there. Given that the committee can sometimes be a terrible echo chamber, is there much point in appealing to the committee? Might be not a different route, say to the community members of AUSC, or a panel of CU or OS peers, be better?  Roger Davies talk 18:28, 24 February 2013 (UTC)[reply]

Discussion and comments from functionaries and community

What is "conduct inappropriate to a user trusted with CheckUser and/or Oversight permissions"? It seems fairly open-ended. --Rschen7754 02:09, 20 February 2013 (UTC)[reply]

Specification of all possible examples is inappropriate. "Wrong stuff" will do. User:Fred Bauder Talk 02:17, 20 February 2013 (UTC)[reply]
Of course, per WP:BEANS, but this sort of ambiguity is part of what led to the stalemate before the elections last year, because arbitrators couldn't agree on what was "inappropriate". --Rschen7754 02:23, 20 February 2013 (UTC)[reply]
Rschen has a point here. When what constitutes "inappropriate" is the very crux of the issue, it doesn't do a lot of good to say "well, you can be removed for inappropriate activity." It would be more valuable for arbcom to draw up at least a rough guideline for what, excactly, is behavior unbecoming of a functionary. Leaking? Personal attacks? Using one's powers to one's own good? Having generally poor judgment? Any of these things could be perceived as inappropriate by some or all arbs, and as long as it's a matter of each arb's personal judgment where the line is, we're inevitably directed back to square one in any specific case. A fluffernutter is a sandwich! (talk) 02:29, 20 February 2013 (UTC)[reply]
  • To echo something I just said to Risker at another venue, it would make more sense to me to discount inactive/recused arbs in votes like this than to count them as pocket vetos (in the sense that if this motion is implemented, the mere existence of any recused or inactive arb numerically counts against the passing of a removal motion). In cases where, for whatever crazy reason, only a small proportion arbs remain unrecused and active, it would make more sense to say "Quorum is not available to hear this matter, so it will be deferred until quorum is available" (in normal cases) and "well, we're stuck with X arbs able to vote, so X arbs voting it has to be" (in emergencies) as opposed to what would happen with this motion, which would be to hear matters when not enough arbs are available to reach any decision, and then go "welp, not enough arbs are voting, so you're free to go, person who may or may not have done something terrible". This also raises the issue of why we're not just dealing with long-term arb inactivity in itself, to head off the need to do these case-law acrobatics. A fluffernutter is a sandwich! (talk) 02:52, 20 February 2013 (UTC)[reply]
  • I think we (functionaries) serve at the pleasure of the committee and are or should be removable even without just cause or have to provide a reason in public. We are appointed by the committee, we are not elected, the committee is. The community, at least thru the directly-elected Committee, should have the last word on who should and shouldn't be a functionary. Obviously, to remove a functionary is no small matter, having us being entrusted with advanced and sensitive tools and having us being vetted and checked before hand, and most likely us being trusted and respected community members for some time. However, the Committee should have a way to remove those who have not abused their tools but who do no longer retain the Committee's trust, be it for consistently poor judgement, personal attacks or other behavior unbecoming of a functionary and that casts a shadow on our work, conflict of interest violation, and so on. In fact, I strongly believe the committee should have an open-ended option to remove functionaries, provided it obviously notify the user in question as to what the reasons for the removal are, as that is only fair. I do not believe this should create major issues or that the Committee would start removing those "they don't like", that would be petty and silly, and the Committee members are, whatever our personal disagreements with them may be, entrusted with some aspects of the governance of the project among which is specifically managing the permissions of Checkuser and Oversighter. I much dislike the fact that the community has no real way of removing admins in such circumstances (that is, consistently poor judgement that does not raise to the level of abuse) as it might lift some of the pressure that some of us feel when commenting on Requests for Adminship, that it is, short of flat out abuse, a life tenured position and that a decision made there and then is not reversible by the community. This should also go for the Arbitration Committee and functionaries. They should feel that if, no matter their rigorous vetting, a functionary has shown that for whatever reason (s)he's not a good fit for the job, they can remove him. Also I suggest that an Audit Subcommittee's "recommendation" should be binding as it pertain to the exercise of their functions, that is, that the Committee should be delegating to the AUSC all matters relating to functionary abuse and violation of policies and as such that their decision would have the effect of removing the functionary's access. The neutral role of the AUSC and their decisions should not have to be validated by the full Arbitration Committee to be effective. In the specific of the voting system, I have to agree with the Arbitrators who opposed a qualified majority rule for such decisions to be made, the committee is fairly small and a few recuses could bring any such motion to a halt. I suggest a 2/3 majority of the voting arbitrators should be enough to both ensure that such decisions would be feasible and that extra-consensus would be required. Snowolf How can I help? 11:56, 20 February 2013 (UTC)[reply]

@SilkTork -- why do you recuse? NE Ent 12:01, 20 February 2013 (UTC)[reply]

  • I must say I respectfully disagree with Snowolf. The idea that the functionaries are mere minions of Arbcom, to be appointed or dismissed at will, is perfectly acceptable in terms of Mediaeval monarchs, but I find it quite disturbing in the context of Wikipedia. And the idea that Arbcom can remove a functionary essentially because some members don't like them - rather than because the rules relating to the use of the tools has been breached or threatened, or some other agreed definition of 'wrongdoing' has occurred - would surely create an open door to politicking and abuse, as Risker has pointed out above. And I am minded that if Risker is raising it as a concern, the possibility of it happening is not negligible. Elen of the Roads (talk) 13:33, 20 February 2013 (UTC)[reply]
@Carcharoth - admins are for life as well. Although if you could institute a review system for functionaries, it might encourage one for admins. Elen of the Roads (talk) 16:29, 21 February 2013 (UTC)[reply]
  • The problem with passing motions of any sort is, as that ArbCom's existing rules are biased against anything happening at all, and the pocket veto Fluffernutter refers to above is something regularly exercised by at least a couple of rather senior arbitrators. Consider that even in a regular vote, there are really only three sorts of responses:
  • 1) Support. Increments the number of supports by one.
  • 2) Abstain, Recuse, or Inactive. Decrements the number of voting arbitrators by one, and every other such vote decrements the number of supports needed by one.
  • 3) Oppose or no vote. Doesn't do anything to the number of supports or the number of supports needed.
I think to solve the more general problem, the committee should reorganize voting into two different classes of voting:
First, traditional pass-then-implemented-in-24-hours voting, where clerks are watching the events unfold like hawks.
Second, a "net four to close" motion, used traditionally under one umbrella in cases.
The first sort of vote, which is a "rush" vote (inasmuch as anything the committee ever does is a rush), could probably stand to stay like it is. However, when there's a "net four for closure" rule, there's a second layer of protections, in that arbitrators are rarely shy about opposing a closure because other arbs have yet to vote, even if their own votes have been set for days or weeks. I would suggest to the current committee that any vote that is to be closed by a "net four for closure" process treat a no-vote as an abstention, reducing the number of voting arbitrators and the passing threshold, rather than as an oppose as is currently done and would continue to be done in the first sort of vote. Jclemens (talk) 02:40, 21 February 2013 (UTC)[reply]

I read all the stuff above, and am fully unconvinced of the need thereof. ArbCom has the ability, through normal processes already existing, to remove admin privileges for cause - and checkuser and oversight privileges should surely be treated in the same manner -- that is, requiring the use of the currently existing processes. Creating a strange new process which is less restrictive on ArbCom than the extant processes is, IMO, improper, unnecessary, and a solution in search of a problem. As an aside, I find the "net vote" system to be arcane at best where use of "2/3" or the like is customary in the "real world". The current system may require "more than unanimity" if only 3 arbs are voting, and if only 7 are voting (as is not all that uncommon in the past) two arbs have an absolute veto power over any decisions. In the outside world, if only 7 people can vote, then 5 is generally the maximun vote required for decisions. Aside to Snowolf: the use of "without cause" would open up the committee to pure politicking - and is also a "solution without a problem." Noting here my agreement with points made by Elen and Jclemens as well. Collect (talk) 14:46, 24 February 2013 (UTC)[reply]

I don't think it would. I feel I am very independent of the arbitration committee in my duty as an oversighter, yet I would have no objection to being removed if the ArbCom didn't feel I was doing a good job. In fact, ArbCom and the other oversighters are the only users who can tell whether I'm doing a good job as an oversighter or I'm making a mess. So to me it seems natural that they should have the power to remove me if I'm not suited for the job... Nobody else can: the community at large can't really review my actions given they're privacy-related and suppressed, and I for one, if I am doing a disservice to the community as an oversighter (not talking about abuse, just saying what if I am not cut out for the job but this was found out after I got it), somebody ought to be able to remove me. That is the general point on why I feel the Arbitration Committee should have the power to remove me. On the specifics of "without just cause", it's precisely so that they have the freedom to act in the best interests of the English Wikipedia (in my experience with the Committee, the folks there try to act in the best interest of this project, tho of course different people have different ideas about what's best for the project, but none of them would remove a functionary on a whim or personal dislike) without having to go into details of why such a removal would take place. This doesn't mean that it should be taken lightly, just that they wouldn't have to announce the reasons behind it, tho I would of course expect them to disclose them to the individual in question. I believe trust must exist between elected ArbCom and the "life-tenured un-elected functionaries" and if that relationship was broken, one shouldn't remain a functionary. That is of course my personal view, and as you've seen above other functionaries have different views but I want to clarify that I do personally thing such a concept would lead to politicking or whatever :) Snowolf How can I help? 12:23, 5 March 2013 (UTC)[reply]

Note to clerks with respect to motions with respect to functionaries

These motions have now run their course and have received commentary and votes. The first two motions have passed, and the second two have not. Please enact the first two motions, adding the first to the procedures page, and making modifications as directed in the second. If you have any questions, please let me know. Risker (talk) 19:36, 10 March 2013 (UTC)[reply]