Wikipedia talk:Arbitration/Policy/Update and ratification

Points

Looks good, the only 2 points that I can see are:

1. The one noted by Sandstein - should probably read "serious and primarily conduct-related disputes", or "serious disputes, primarily related to user conduct". My bad, I didn't spot earlier the difference between scope being "primarily, serious conduct-related disputes" vs. "serious, primarily conduct related, disputes".
2. Arbitrators are expected to make high quality, project beneficial, and impartial decisions. I'd have liked to see that in there as a statement of the fundamental duty and spirit of the role, and general guidance in future cases.

In both cases, probably too late at this point. FT2 ^{(Talk | email)} 19:48, 31 May 2011 (UTC)

Jurisdiction question

For clarification, am I correct in understanding that ArbCom does not have jurisdiction over an EnWiki editor acting in the capacity of steward, which is a Foundation-level role? Granted, stewards are not supposed to perform actions on their home wikis, but there is an exemption for emergencies, specific cross-wiki violations, and clearcut rights self-requests (see Wikipedia:Global_rights_policy#Stewards and m:Steward_policies#Avoid_conflicts_of_interest). -- Avi (talk) 20:16, 31 May 2011 (UTC)

Yes, that's right. Steward actions are excluded.  Roger Davies ^talk 20:26, 31 May 2011 (UTC)

Well, that's true up to a point; we certainly don't have the authority to "destewardize" someone. But if a steward who was also an admin on English Wikipedia took 30 blatantly rogue steward actions and appeared to be out of control, I think we would feel empowered to desysop him on this project, even though he or she hadn't done anything wrong yet with that particular hat on. Newyorkbrad (talk) 02:50, 4 June 2011 (UTC)

Curses, foiled again! Brad has seen through my dastardly plan . -- Avi (talk) 21:48, 10 June 2011 (UTC)

4.To resolve matters unsuitable for public discussion for privacy, legal, or similar reasons;

This is a joke right? Arbcom resolve stuff? Pedro :  Chat  21:42, 31 May 2011 (UTC)

Arbitrator participation

I've seen cases in which arbitrators did not recuse, but then did not vote in any of the findings or measures posted for the proposed decision. Could verbiage be added in which it is stated that non-recused arbitrators are expected to participate in voting in the proposed decision? This is important because of the rules in place to determine how binding motions are passed. Cla68 (talk) 00:55, 1 June 2011 (UTC)

I think this is already covered by "Conduct of arbitrators", point #3 ("Participate conscientiously in the Committee's activities and deliberations..."). Anything more specific than that would be better placed as procedures than as part of the core policy, in my opinion. Kirill ^{[talk] [prof]} 01:00, 1 June 2011 (UTC)

Sounds good. Cla68 (talk) 01:03, 1 June 2011 (UTC)

In practical terms, a formal "if you're not recused you must participate" clause wouldn't work. This is something volunteers do in their spare time, not a paid job; it's not at all unusual for someone to have an intermittent presence on Wikipedia (and thus not formally inactive) but not have the time to keep up to speed with a complicated case. (This is my situation vis-a-vis the treeshaping case, for instance.) Particularly in cases where there's a clear majority pointing in one direction and the action of any one arbitrator will make no difference, I don't think it's a particularly good use of time reading shedloads of commentary to reach a decision which would have been made anyway; I'd predict that a formal "must participate" clause would just result in a lot of "flip a coin" and "back the majority" voting. – iridescent 14:45, 1 June 2011 (UTC)

The expectation that an arbitrator will participate in everything is certainly unreasonable, because they are volunteers, but is also undesirable, because the associated fatigue would lead to a higher rate of burnout. In my view, it would be better for each arbitrator to continue to not vote on a small number of cases, than try (and inevitably fail) to keep on top of everything. AGK [•] 19:04, 1 June 2011 (UTC)

I'm afraid I don't agree 100% with your reasoning here. Case decisions are among the, if not the, most important of the deliverables that the Committee produces. Sometimes, important proposed decision motions don't pass by a hair's-breadth because one or two arbitrators don't vote. To me, and I'm not trying to push this analogy too far (because there are valid arguments made that ArbCom cases should not resemble legal proceedings), it's as if not all the justices on any given nation's supreme court decided not to vote in a decision. It's what they are there for, and if they disagree with the decision, their contrary or dissenting opinion is still important. Cla68 (talk) 00:05, 2 June 2011 (UTC)

The general rule of thumb is if you are going to only have time to be active on a few cases, to be organised and state this (publicly) upfront at the start of a case and to not go active halfway through the case (unless there is a real dearth of available arbitrators). The other thing is to not have activity levels for arbitrators for a case change during voting. That always messes things up. Work out who is going to be active on a particular case before a proposed decision is posted and before voting starts. And then don't change that until voting finishes. Sometimes pressure has been put on inactive arbitrators to go active to "break a deadlock" - that is something that shouldn't really be done either. Regardless, all this is something that each iteration of the committee has to work out for itself, and comes under procedures more than policy. Carcharoth (talk) 00:39, 2 June 2011 (UTC)

Ratification level

Gimmetoo's point about the low bar of 100 supports to "ratify" does seem valid actually. It looks to me as if 100 supports is a rather arbitrary number. Admittedly, setting that bar too high would be embarrassing. Can anyone track down precisely when the "majority support, with at least one hundred editors voting in favour" language entered the proposed policy? My concern is that the best value for this number may vary over time if the size of the active core editing community at en-wiki (that could be persuaded to vote on this) varies. I suspect this value has been set at a level so that we never end up with a situation where it is not possible to get future changes passed. But maybe for the next set of changes (hopefully not for a few years), this could be tweaked. The metric I would use for this number would be related to the area of Wikipedia that currently sees most active participation. Is that still the ArbCom elections, or are there RFA and bureaucrat nominations that get more participation? Anyway, maybe link this ratification number to the number voting in the ArbCom election previous to the vote on any changes? That would be an elegant way of having the minimum level adjusted to fit the size of the community that takes an interest in these matters. Though hopefully there will be far more than 100 editors giving their opinions here. Carcharoth (talk) 01:25, 1 June 2011 (UTC)

I don't think the 100 figure is an intent at quorum so much as a safeguard against trying to slip in a change "under the radar" as it were. Anything 100 people have commented on has been visible. — Coren ^(talk) 01:41, 1 June 2011 (UTC)

The 100 figure has been in the draft since 9 June 2009.  Roger Davies ^talk 10:06, 1 June 2011 (UTC)

As a further safeguard to "prove" that it was as visible as reasonably possible, you could add links to where this has been publicized (which would also ensure a wide cross-section of the community were notified). Have all the places listed in the preamble been notified about this? I would presume so, but the preamble seems to only say that the drafts were announced there. FWIW, the current listing of where this page is linked from is here. That may change over the next few days, of course. I can't make head or tail of it at the moment - presumably some template is in use on many of those pages accounts for the links from many AfD log pages - the WP:CENT template maybe? This is why an independent listing of where this has been announced (maintained by those who made the announcements) would help. Carcharoth (talk) 02:01, 1 June 2011 (UTC)

I was intending to spread the announcements - which will appear in all the venues used to publicise the drafts - out over a couple of days, rather than just have a single blitz of announcements everywhere. Yesterday was WP:AC/N and Signpost. This will have the effect of the later ads acting as reminders for earlier ones in other venues and keep it visible for longer.  Roger Davies ^talk 10:06, 1 June 2011 (UTC)

I interpreted this the same as Coren - we don't want the bar so high that bad practices can be entrenched due to change being seen as discouragingly impossible. We do want to ensure the policy isn't changed frivolously, or by small groups with agendas, or in a way that hasn't involved high visibility community discussion. If 100 editors agree on seeking consensus for some change, that seems neither so high as to discourage genuine improvement nor so low as to be the result of non-significant discussion. So I'm relatively happy.

My only (and slight) concern in this area is the possibility of drama driven proposals in the wake of some notorious future case, leading to a bad policy change. For that reason requiring a 2/3 majority rather than a 50% majority, or endorsement by 2 arbitrators might have been safer (if not even 2 arbs would agree then it's probably a very poorly conceived idea). But then again people do generally understand the importance and seriousness of Arbcom's remit, and if there were good reasons for the current practice, hopefully they would prevail anyway. We do want the policy to be capable of steady improvement for significant matters so there's a balance. FT2 ^{(Talk | email)} 10:15, 1 June 2011 (UTC)

Authority

Commenting on Gimmetoo's other point ("One could view AC's authority as delegated by the community, but that view appears contrary to the current wording of the "jurisdiction" section"), some historical information and background:

Arbcom's authority is part within, and part without, the community. The originating concept (this was back in the days of 2003-04!) was that if the community failed to resolve a serious dispute, then Jimbo Wales had absolute authority to make a unilateral decision, which role and authority he delegated to a panel of users chosen from within and by the community - which was the arbitration committee. So the committee's authority stems partly from within the community as volunteer editors in good standing, and part from Jimbo's original role (which has become more ceremonial and less executive over time) as project leader of English Wikipedia. That might be "constitutionally outdated" or an anachronism, but as history, that's how it came about. (Another way to describe it is that the community elects the editors it wishes to see operating Jimbo's original and delegated residual powers as project leader, on Jimbo's behalf.) The "part and part" style has kind of stayed with the committee ever since - it's answerable in a sense to the project itself, and responsible for making decisions in the best interest of the project (ie the goals and aims of Wikipedia), rather than the purely to the court of public opinion, and has a slight independence from–and authority deriving from outside–the community, as well as being drawn from and answerable to it. Like with most things wikipedia, it's a technical muddle that oddly works if you don't analyze it too closely :) FT2 ^{(Talk | email)} 09:50, 1 June 2011 (UTC)

I'm aware of the history. Wikipedia has experienced a process of discovery and development analogous to many young countries' development of government and a legal system, from the roles of various ruling bodies to the role of precedent. Someone could write a law review article about this. Practically, the AC exists because some disputes are not easily settled by the community as a whole; if the community could solve all disputes, then even Jimbo's role would likely have related to issues external to the community, such as legal and business issues now handled by the WMF. If wiki-government appears to be evolving from monarchy to elected democracy, then policies should avoid enshrining language which appears contrary to the latter. Gimmetoo (talk) 11:49, 2 June 2011 (UTC)

As the comment purely references the legislative source of authority, it's of the nature of a recital rather than an actionable matter. Let's wait another iteration, until it's much more visibly redundant (if it is). Another 5 or 7 years (passes like a flash!) will allow old traditions to be more clearly anachronisms, the community's competence to self-govern the worst and most private of matters clearer, and so on. No harm done by deferring the reconsidering of a recital. At this point it's accurate and including it eases the transition from the slightly ambiguous situation and the policy of 2004 to this one. FT2 ^{(Talk | email)} 12:08, 2 June 2011 (UTC)

Ratifying changes

Probably some of this text should have been in the procedural header. As it stands, it can be read two ways:

• Proposed amendments may be presented to the community for a decision if they obtain 1/ Committee assent, or 2/ a petition of 100+ endorsing users.
• Proposed amendments must undergo the "identical ratification process" - adoption by the committee and formal referendum. They can be presented to the community if they obtain 1/ Committee assent, or 2/ a petition of 100+ endorsing users.

With hindsight this section should simply say "Future amendments require <process>" rather than referring to other texts or using unnecessary terms like "an identical process". I think this is what's actually meant:

"This proposed policy requires formal ratification through a community referendum and enters into force upon receiving majority support with at least one hundred editors voting in favor of adoption. Until and unless it passes, the existing arbitration policy remains in effect."

"Amendments to this policy [likewise] require formal ratification through a community referendum and enter into force upon receiving majority support with at least one hundred editors voting in favor of adoption. Proposed amendments may be submitted to the community for ratification only after being approved by a majority vote of the Committee, or having been requested by a petition signed by at least one hundred editors in good standing. Until any changes are ratified, the existing arbitration policy remains in effect."

FT2 ^{(Talk | email)} 13:48, 1 June 2011 (UTC)

I do not think that the present version of the Ratification and amendment section is unclear or ambiguous. AGK [•] 19:07, 1 June 2011 (UTC)

Non-public matters

In response to Protonk's question, this is the kind of issue that comes to mind:

• Previously good standing user W (600 edits, rollback) is facing a long term block/ban for misuse of rollback and heated threats followed by evasion socking in a dispute which has already seen them evade a 1 month block to respond to an accuser. They have in fact just lost their job and had a breakdown while that was going on, and want it taken into account for leniency, but don't want any of the personal matters publicly known.

• User X is being sent love notes daily by a user, she doesn't want to say so publicly because of the Streisand effect, but it's resumed despite a a past low profile and minimally described block done by a privately contacted admin and the Foundation is clearly not going to get involved on this low level and leaves it to the community. Another admin has explained that community consensus won't allow them to make a block unless they are prepared to publicly provide evidence beyond vague claims to back it.

• User Y, in a Christian family and having a strong sense of social equality, edits on religion and LGBT topics among others with a commendably neutral viewpoint. He stopped editing a topic after getting a wiki-email from an opposing editor saying they know who he is and hinting they will "out" his LGBT-positive contributions to his (religious) social circle via Facebook. He wants to ask advice and feels someone ought to judge if any action is needed, but doesn't want to push the guy towards outing him now it's all gone calm, so a public comment or evidence is out of the question.

• User Z is on an external mailing list where 3 users have been engaged in avid discussion and 2 of them are canvassing on a topic area with instructions how to sway consensus. The emails provide some personal information that appears to identify one of the abusive accounts. He wants someone to have a heads-up and deal with it but is inexperienced himself.

Which users handle these and decide on a possible block or unblock? That's the kind of thing that "matters unsuitable for public discussion" covers - all the miscellaneous "not bad enough for law enforcement or courts, not going to get WMF attention, but needs users trusted by the community to handle matters off-wiki and in private, to deal with them" cases. FT2 ^{(Talk | email)} 04:12, 2 June 2011 (UTC)

In case W, I don't see why this information should be considered anywhere, in public or private. We have no verification about his job. More fundamentally, the block applied is supposed to be protective, not punitive: it doesn't matter what the editor's damage is, only what he does. The proper remedy for such situations is to make the blocking process less humiliating for all blocked editors, e.g. by not having visible Scarlet Letters on blocked user pages; the block log and ANI archive is more than enough.

In case X, someone would have to be able to intercept the email at Wikipedia to know who is actually telling the truth. Can ArbCom do this? Otherwise you can only choose to take one person's word that the messages are being sent over the other's that they are not, based on some prejudice.

In case Y, it is clear that a fresh account is needed - to begin with, to ask this general advice privately without having to trust that the private ArbCom proceedings won't end up on Wikileaks. Multiple accounts, properly used, are Wikipedia's sovereign solution for privacy issues.

In case Z, Wikipedia is becoming imperialistic. We're supposed to know about communications on private e-mail lists? What sanction can be taken against users here for what might be posted on the list? Are you going to punish people only if they use the same name (probably their real name)? It's overreaching.

In general I think that the more such secret consideration is curtailed, the more honest Wikipedia will be, and the less it will be swayed in improper ways. Wnt (talk) 20:21, 2 June 2011 (UTC)

I agree with Wnt that none of these are actionable, which is not to say that somebody shouldn't empathize with the editors in question and give them advice. Which of these couldn't have been handled by OTRS? TotientDragooned (talk) 21:31, 6 June 2011 (UTC)

• In fairness to WNT and TotientDragooned, the examples FT2 laid out aren't particularly realistic or helpful. The first example mixes a standard unblock request (which is an appropriate Arbcom role, as the "court of last appeal" for blocks) with the overstressed user issue. In the latter situation, the usual position of the committee has been "it's time for you to take a bit of a break" and generally speaking we've had a fair number of successes in persuading users to do so before their behaviour adversely affects more than a limited area of the project (or we've had to formally remove permissions/block the user). In some cases, this has included encouraging people to voluntarily resign administrator permissions. (On a rough guess, over the 2.5 years I've been on the Committee, abou 70% of these situations have never really been obvious to the community as a whole, though a few users will be aware of them.)

  The second situation is something we have dealt with occasionally. Many users (including administrators) may not really have recognized that using the Wikimedia interface ("Email this user") to send harassing emails to another editor is absolutely grounds for blocking. However, as the emails being sent will often contain nonpublic personal information about the harassed editor, they are not suitable for onwiki discussion. (Generally speaking, it is also inappropriate to insist that the harassed person publicly reveal the full extent of the harassment.) Arbcom has executed blocks and removed email access from those who have used the project's technology to directly harass another editor.

  As I indicated, I've been around for a while on the Committee, and I'm also pretty familiar with the mailing list archives, and I can't recall any situations all that similar to the last two that FT2 has proposed. We did deal with the Eastern European mailing list case a while back, but in that situation we were essentially handed the entire archives of the mailing list and the case itself was managed onwiki, so the situation isn't really comparable.

  Sorry this has been so wordy. Essentially, the issues that are dealt with in a non-public way are (a) serious personal issues where the editor needs to take a break (b) outing and/or harassment. It's very rare that there won't be at least some public discussion of anything else. Risker (talk) 23:07, 6 June 2011 (UTC)

Admissibility of evidence

The "Admissibility of evidence" section makes no sense. The way I am reading this, checkuser evidence must either be disclosed publicly (a horrific invasion of privacy) or the committee must individually on a case-by-case basis make a determination to allow it. That's a pointless addition of process. For that matter, you have essentially given a blank check to a stalker to repost things that have been oversighted. The policy needs stronger language demanding that revisions not available to the public only be permitted to be entered into evidence publicly if doing so does not violate a user's privacy. --B (talk) 23:00, 2 June 2011 (UTC)

I'm not sure I read that the way you do. The section enumerates three things which are always admissible, and two which requires a determination. The last paragraph (which I suppose is the one you are worried about) simply states that any evidence can be submitted privately, but anything that can be public should: i.e.: things should only be private if there is a compelling reason to do so. Checkuser data and supressed revisions are just a good example of something for which there is a compelling reason. — Coren ^(talk) 01:01, 3 June 2011 (UTC)

Others (Comment by TheBlueWizard)

I don't want to create a new subsection, but I don't see any suitable place for my comments, as I am not so knowledgeable about the policy, and I thus consider myself neutral regarding the amendment(s). Nonetheless, I want to say something: Suppose Jimmy Wales dies today. Then the clause(s) involving his name will obviously have to be changed, and of course these would have to go through the approval process again. I would suggest changing "Jimmy Wales" to "President of Wikimedia (currently Jimmy Wales)" or something like that. In the past, specific names would be OK for an organization akin to a treehouse club, but Wikipedia is obviously too big for that now. Also, I notice that it doesn't seem to mention the minimal number (I believe that would be called a quorum...right?) of arbitrators required in order to function. I could be missing something, or maybe that's actually not a part of this process??? Anyway, that's my two cents. --TheBlueWizard (talk) 02:49, 3 June 2011 (UTC) Moved from ratification page. Risker (talk) 02:58, 3 June 2011 (UTC)

Jimmy's role in the English Wikipedia is a unique one as its (co-)founder. He isn't "President of Wikimedia" or anything like that. If, heaven forbid, he were to pass away any time soon, I doubt any single person would take on his position in dispute resolution. The role would most likely just be retired. Jimmy has stated that in the event of his death or incapacity the ArbCom may decide how to amend policies that rely on him, subject to ratification by the community. the wub "?!" 19:37, 3 June 2011 (UTC)

Regarding Jimmy's role, I think the wub is probably right. If he were to become unavailable for any reason, it's unlikely we would select a particular person to take his place; I am sure there would be some discussion of whether the community (not the Committee!) should create some other mechanism to perform some of his functions.

Regarding a quorum, the smallest number of arbitrators who can be active to finish a full-fledged arbitration case, under the current policies and procedures, is four, because it requires "net four" votes both to accept a case and to close the case. At the end of 2006 (when I was a newish editor and well before I joined the Committee), I remember a few weeks when this was actually an obstacle to the committee's functioning, because there had been a few resignations during the year and there were only a handful of active arbitrator, some of whom were busy with real life and inactive on-wiki. At present, though, there are 18 arbitrators and all but two or three of us have active at any given time this year, so I don't imagine this will be an issue anytime soon. Newyorkbrad (talk) 02:46, 4 June 2011 (UTC)

The only bit about Jimbo I can see says "Remedies may be appealed to, and amended by, Jimbo Wales, unless the case involves Jimbo Wales' own actions." If he should die, that will not be a problem - people can still appeal to, and hope for an amendment by, a corpse if they so wish ;-) -- Boing! said Zebedee (talk) 18:10, 4 June 2011 (UTC)

Missing: greater purpose

The "Scope and responsibilities" section nonwithstanding, I am disappointed that there is no section about greater principles, be it "justice" or "good of the project", or "protection of editors", or something like that. As it is, it is just another building block for the letter above spirit attitude... Sure, it is a technical improvement over the past policy, but is it a step in the right direction? No, it is just some nicer quality paving... --_{Piotr Konieczny aka Prokonsul Piotrus| talk} 22:33, 3 June 2011 (UTC)

A mission or statement of purpose? Sounds useful.. John Vandenberg ^(chat) 03:04, 5 June 2011 (UTC)

Conduct of arbitrators

If the "Conduct of arbitrators" section had been in force in 2009, would FT2, as being in repeated and gross violation of points 1 and 2, have been suspended/removed by the Committee? Without necessitating any input from the community? Or would a frustrated community member such as myself still have had to block him to get any action? FT2, reply if you like (you seem to be replying to most of the questions on both talk and project page here), but please be aware that it's preferably not you I'm asking. I'd like to hear from an arbitrator, ideally one who was active both then and now, so they have some knowledge of the case and the RFAR proposal (of me, not FT2) which followed.[1]

If the answer to my first sentence is "yes", I approve of the "Conduct" section, and it's high time we had one. Integrity is the first — nay, in a sense, it's the only — thing we must demand from an arbitrator. Bishonen | talk 15:32, 4 June 2011 (UTC).

Yes, it would have made that particular episode much easier to handle.  Roger Davies ^talk 17:38, 4 June 2011 (UTC)

I concur with Roger Davies. Risker (talk) 18:14, 4 June 2011 (UTC)

I don't know what was going on with that, but I'm more concerned that FT2 has done extensive policy rewrites including things I don't agree with, which I don't think were really discussed at the time of the change - e.g. the "Scarlet Letter" policy for userpages and overthrowing the GNG, allowing a special notability guideline to throw away articles. I haven't even tried to change these back because I feel like it would be a hopeless battle to "change consensus". I hope that the statement against making policy by fiat will tend to encourage more equality in the amount of resistance people encounter when trying to change policy in the future. Wnt (talk) 16:03, 4 June 2011 (UTC)

FT2 is neither an arbitrator nor speaks on behalf of the committee. Even if he were, individual arbitrators have no more authority when proposing draft policy than any other editor, though the nature of the arbitrator role means they may sometimes be more familiar with aspects of particular policies and how those aspects intersect with other policies.  Roger Davies ^talk 17:38, 4 June 2011 (UTC)

Thank you for the replies, Roger and Risker (and would any arb like to apologize to Bishzilla for the inappropriate admonishment she got way back then?) I too was surprised to see Wnt going off at a tangent here. I hope nobody is under the impression that FT2 is an arbitrator. Bishonen | talk 20:14, 4 June 2011 (UTC).

On the issue of apologies, I'll give the same reply that I gave to Peter Damian (in a personal capacity yadda yadda): I'll certainly concede that a lot of former Arbcom members have acted badly in the past, but I don't think it's the place for the present committee to be issuing apologies. There isn't a monolithic beast called "The Committee" which just happens to suck in various people from time to time; there's a collection of individuals, which changes radically each year. You can have a Queen-in-Ireland style "personal expression of regret" if you want, but a formal apology for something in which I wasn't involved in the slightest would be as insincere—and as pointless—as when Tony Blair took it on himself to apologise for the slave trade or the potato famine. – iridescent 21:39, 4 June 2011 (UTC)

No big deal in any case. I just thought I'd open an opportunity for a few people (you know who you are) who were arbs then and are still arbs now to give a thought to their actions at the time. Bishonen | talk 23:03, 4 June 2011 (UTC).

I never got my apology. But watching that FT2 being slapped about is good enough for me. Besides, Bradspeak apologies bore the living fuck out of me, and have the value of what I think of FT2's moral fiber. OrangeMarlin ^{Talk• Contributions} 03:38, 5 June 2011 (UTC)

OM, from what I remember, based on the diffs that were offered as evidence, it wasn't FT2's case decision which was flawed, it was the manner in which he arrived at it. Cla68 (talk) 01:19, 6 June 2011 (UTC)

Bit more complicated than that Cla68, context and importance of process at the endpoint (arbitration) need(ed) also to be considered. Casliber (talk · contribs) 01:38, 6 June 2011 (UTC)

Wnt, any editor is welcome to discuss changes to wording on policy pages, and make changes after discussion and consensus is reached on talk pages. All editors are equal in this respect, so if you are concerned a significant change has been made without consensus then you have every right to query that change on the policy talk page or other venue. Casliber (talk · contribs) 22:02, 4 June 2011 (UTC)

But sometimes changes are pushed through with a large number of edits, tweaks, and overly-optimistic claims of consensus, effectively overwhelming any mere editor who tries to question them. Jehochman ^Talk 03:48, 5 June 2011 (UTC)

Interim elections

In exceptional circumstances, the Committee may call interim elections, in a format similar to that of the regular annual elections, if it determines that arbitrator resignations or inactivity have created an immediate need for additional arbitrators.

This is more a question than anything (the proposal, over all, seems unremarkable), but does this part envision the creation of new seats or the election of new people to existing seats? If there's been a resignation then that's obvious, but "inactivity" typically doesn't result in an actual empty seat, just the appearance of one ;). In the past of course new seats were created whenever, which was fine, but there seems to be a movement toward codifying practices so some clarification on this point might not be amiss (just so there's no accusation of misunderstandings later). Mackensen (talk) 18:30, 4 June 2011 (UTC)

I think the worrisome scenario is the combination of both: there have been enough resignations that inactivity of (some of) those who are left becomes truly problematic; in that case filling the vacant seats should suffice to shore things over. Honestly, if there are no vacant seats that could be filled and there is so much inactivity that the committee cannot work then the problem is a bigger one than a simple interim election could fix. — Coren ^(talk) 18:39, 4 June 2011 (UTC)

The committee is the best judge of its own capabilities to meet workload, and inactive arbitrators don't have voting rights, so the majority vote decision on interim elections seems like an effective mechanism to me. The issue of what constitutes an empty seat is a thornier one; the elections have treated resigned seats as vacant and inactive seats as filled, but we have also seen resigned arbitrators re-assume their seats without re-election. An open question left unresolved by this iteration of the policy, and one which is likely to come back to bite us in the ass in future. Skomorokh 19:31, 4 June 2011 (UTC)

I was going to ask which arbitrators have resigned in the past and then resumed their seats, and then I remembered who those arbitrators were/are. FWIW, I think arbitrators needing a break should go inactive, and arbitrators that resign in a huff or on a point of principle should not be able to resume their position, regardless of what happened in the past. There is an exception to this, however, to which I don't have an answer. I agree with Skomorokh that this needs looking at, but as something to be put on the agenda for the next round of changes. Incidentally, while so many people are being drawn to this topic and discussion, would this time of year be a good one to have the discussions about ArbCom elections that the community always plans to have just after the preceding one, then postpones to later, and then has a few weeks or days before the actual elections? Carcharoth (talk) 23:50, 4 June 2011 (UTC)

Aye, it's on the agenda and an RfC is planned for sometime in the summer once the ratification is out of the way. Suggestions are very welcome at last year's review or by email. Skomorokh 21:22, 5 June 2011 (UTC)

Interim elections are unlikely ever to be needed again, now that the committee has been expanded to 18. Newyorkbrad (talk) 21:23, 5 June 2011 (UTC)

That doesn't address the point that Skomorokh raised above: "the elections have treated resigned seats as vacant and inactive seats as filled, but we have also seen resigned arbitrators re-assume their seats without re-election". May I ask if the current committee have had the chance to consider whether arbitrators that resign can later resume their seats and under what circumstances? This has at times been a potential problem when resignations occur just before or after an election. (My view is that arbitrators who resign should have to go through some formal and public process to gain both committee and public approval to be reinstated.) If the current committee has a view on this, it should at least be formalised on the 'procedures' page, even if it is not a matter for arbitration policy. Carcharoth (talk) 23:44, 5 June 2011 (UTC)

There's been no specific discussion on this point to my recollection. If there were, my position would be "handle it on a case-by-case" basis, which probably wouldn't help much anyway. Hopefully we won't have many resignations any time soon. (The shortened term length is one reason the number of resignations lately is less.) Newyorkbrad (talk) 23:54, 5 June 2011 (UTC)

Thanks. This year's committee has (touch wood) seen less resignations than in past years. I think you are right that shortened term lengths help there. Hopefully issues surrounding activity levels (the other perennial issue I remember from my time on the committee) will similarly not arise when this year's elections get closer. Carcharoth (talk) 00:03, 6 June 2011 (UTC)

Wouldn't it be easier to do what the government does, and simply accept the peson who got the next largest number of votes at the previous election? Hawkeye7 (talk) 02:56, 6 June 2011 (UTC)

There is the realistic chance that the "next largest number of votes" may fall below the minimum cutoff of 50% +1 support. Given the number of seats open at each election, and the number of candidates who receive sufficient community support, that's not a lot of wiggle room. I think only one candidate last year who met the support criteria was not appointed. Risker (talk) 03:51, 6 June 2011 (UTC)

If there are insufficient numbers of candidates who meet this requirement, is it still the case that Jimbo Wales appoints the remainder? Hawkeye7 (talk) 03:15, 7 June 2011 (UTC)

Both the community and Jimbo Wales have agreed in the last three elections that only candidates with greater than 50% support are eligible for appointment. You'll note that the number of arbitrators, the minimal support level, and other criteria for appointment have not been included in the policy. These things have largely been left to the community to determine, just as the community determined term length. I believe that in the most recent election, Jimbo undertook not to appoint anyone who had not run in the election or any candidate who did not receive greater than 50% support, but someone else will have to find the link to that discussion. Risker (talk) 03:44, 7 June 2011 (UTC)

I agree with Risker's analysis here. But it is worth noting that Jimmy has already stated
Here, let me by decree in this very instant make the following binding pledge upon myself: In the event that the ArbCom makes a ruling against me, overturning any decision I have made in my traditional capacity within Wikipedia, the ArbCom's decision shall be final.
So theoretically, at least, if a case were brought against Jimmy for appointing arbitrators out-of-process, ArbCom could rule. One option open to the committee would to be to suspend the contentious appointments pending an RFC, thus remitting the core issue to the community. On past experience - and in the spirit of the proposed new policy - that is a reasonably likely outcome.  Roger Davies ^talk 05:21, 7 June 2011 (UTC)

Scope and responsibilities comment

Just my 2 cents worth since many ppl are making brief comments on this while they vote. I think ArbCom is a necessary safety valve for when our attempts at community consensus decisionmaking fail. This happens primarily in the cases described in the scope and responsibilities section, in particular intractable conduct disputes and issues where non-public information is at issue. However, it can also occur in intractable policy disputes, as has been the case around BLPs, etc. So, I would welcome ArbCom to have within its scope policy making in cases of last resort, since I value the endproduct (Wikipedia, the encyclopedia [nearly] anyone can edit) more than the underlying experiment in the scalability of consensus-driven decisionmaking processes. I recognize that the bar for policy involvement needs to be higher than for conduct involvement, and also that others will disagree... Martinp (talk) 23:40, 4 June 2011 (UTC)

Governance

This is intended to centralise discussion about governance concerns.

• First, the mere inclusion of the word "primarily" does not give unambiguously ArbCom authority over any issue. It can only be read to do so by extrapolation: many arbitrators - myself included - do not agree with the extrapolated interpretation and would therefore oppose any proposed decision based on it. You need only to look back through the failed requests for arbitration to see just how often we do turn down content cases to see how strongly this ethos of division of responsibilities is embedded in the committee.
• Second, the committee has no enthusiasm whatsoever for ruling on content issues (Which RS is best? X or Y? Is X really a RS? Does the new edition of Y trump X now that it's been substantially revised? etc etc etc).
• Third, as with all policy, this proposed one would be interpreted by both its letter and more importantly its spirit. There is nothing in the spirit of this proposal that suggests even remotely that governance is ArbCom's role.
• Fourth, what are the practicalities of an attempted coup d'état? Any move into governance would require the active support of the community. Past experience suggests this would not be forthcoming. And what use is a decision without implementation? Not use at all, it's just empty words. So, the scenario that you suggest would require: (a) an ultra vires decision to be passed by ArbCom; (b) complicity and collusion by adminstrators and others in enforcing that decision; (c) complicity and collusion by adminstrators and others in not undoing the actions of those enforcing it; (d) complicity and collusion by Jimbo and the stewards in letting it stand.

In other words, it would require very broad consensus indeed to survive and without that consensus it would simply not be implemented. On the other hand, if there were broad consensus supporting the ArbCom "decision", even if it were outside of Arbcom's remit, then there wouldn't be a crisis at all.  Roger Davies ^talk 09:07, 5 June 2011 (UTC)

And yet the scope is still worded so broadly that I bet someone could fit the act of harming babies into it somehow. Just because yourself and other arbitrators do not agree with with extrapolating things out of it doesn't mean that it won't happen, nor does it excuse how broad the interpretation of this policy is. Also, I'm glad that we're finally dropping all of the pretending barriers and referring to ArbCom actions as "governance" now, which really is suitable (especially with the new policy). Ajraddatz (Talk) 00:25, 6 June 2011 (UTC)

Roger, I don't believe that the statement, "Any move into governance would require the active support of the community" is totally accurate. The arbitration committee does have the ability to impose governance by fiat, and the reason is that you each are appointed by Jimbo, and only he and the other arbitrators have, if I understand right, the authority to remove arbitrators. If the Committee is promising to stay out of any governance role in Wikipedia, it's because you are choosing not to do so, not because you aren't able to. Cla68 (talk) 01:25, 6 June 2011 (UTC)

In 2009 the committee attempted to establish a governance review committee, WP:ACPD, and even that was scrapped. I agree that the arbitration committee can impose governance, and it has done this occasionally (and usually not directly by writing policy), but the committee's ability to do this depends on the community desire to support, implement and enforce the committee's decisions. Often the committee writes a decision for a specific case and the relevant policy is improved by the community in due course. Whether or not the committee decision was a catalyst or not is debatable. John Vandenberg ^(chat) 23:35, 6 June 2011 (UTC)

Roger, your argument here belies your actions at this motion. That motion is a perfect example of the Committee imposing a new policy -- "no unsourced BLPs" -- by fiat. Your comments on the motion voting page were policy opinions -- and since you held a seat on the committee, your opinion counted, the rest of the community be damned. And it took neither a grand conspiracy nor community consensus to implement. Just a minority of administrators willing to uphold the motion, and a majority unwilling to disrupt the project to oppose them. TotientDragooned (talk) 20:51, 6 June 2011 (UTC)

Verifiability of BLPs was already part of our policy, and the wmf:Resolution:Biographies of living people in April 2009 heavily underscored the need for this. The way I read it, the ArbCom at the time signaled to the community that it would not sanction admins for enforcing the enWP policy and WMF resolution, even if those admins were heavy handed. I know that many decent stubs were deleted in this process, and the eventualist in me squirmed but, like the majority of admins at the time, I did nothing because I knew that the firehose was producing more indecent stubs that the community could fact check, let alone working through the backlog in a careful manner. --John Vandenberg ^(chat) 23:49, 6 June 2011 (UTC)

TD, thanks for that example. I view that as exactly the *type* of policy engagement I want for the AC to have, and that I actually wish were more strongly enshrined in the new policy. The line between defining, refining, and interpreting policy (as it pertains to issues of conduct) is a blurry one. This is an instance where community consensus building processes about an important issue were failing (i.e. the community model of Wikipedia reached its limits of effectiveness). Arbcom stepped in and facilitated a partial resolution, which led to conventional WP processes now largely being able to work effectively in this area again. Such cases are rare, and shouldn't (I would hope) happen more than once a year or so. But when they do, I want ArbCom to do something rather than just watch things going around in circles. I may disagree with *what* they do, but it will be better than nothing. Martinp (talk) 04:12, 8 June 2011 (UTC)

• Noting that alternate governance concerns also occur: some people are in favour of judicial supremacy and the concept of the common law. Fifelfoo (talk) 07:16, 9 June 2011 (UTC)

Privacy and cases

Copied from: Wikipedia:Arbitration/Policy/Update and ratification#No, retain the old policy

Time to write out the "private hearings" provision that allows for closed-door cases without onwiki discussion. This is entirely contradictory to our open-model of governance and ArbCom's high position is no excuse for a run-around of this important feature of our model. If for legal reasons some cases cannot be held in public, then at least ArbCom needs to publicize on-wiki: 1. that an offwiki case is taking place 2. the parties involved and 3. any editor restrictions or policies developed from the case. I cannot support this unless I am certain that the final results of all arbcom decisions, including those decided off-wiki, are logged publicly on Wikipedia (preferably in the same place, such as the ArbCom noticeboard). ThemFromSpace 17:01, 9 June 2011 (UTC)

TFS, I believe your suggestion does nothing to protect the victims of stalking, harrassment, and outing. You have focussed entirely on the principle of openness at the expense of that other great unwritten Wikipedian principle "do no harm". These are some of my reasons for saying this:

1. Unlike real life situations, anyone can turn up and comment at an arbitration case. This produces an atmosphere totally unlike any real world situation. In some instances, it's called transparency; in sensitive issues, it's simply horrifically invasive. It is impossible to discuss private stuff publicly, no matter how cautiously, without the dots getting joined up or speculation/allegations running riot.
2. How many people will complaint about harassment, stalking, or outing, if they know they were going to have to face the public gallery? How does naming the victim and insisting on a public case help build an encyclopedia?
3. BLP applies everywhere on Wiki, as do libel laws. How does it help build an encyclopedia to denounce someone publicly on one of the world's most visited websites as a stalker or a pedophile, remembering that a libel judge might disagree?
4. The community has long accepted that some stuff needs to be dealt with in private. This mirrors the real life situation, where in civilised jurisdictions, the law provides protection for certain classes of victim and some defendants. The current update directly addresses this and builds in some procedural safeguards.

I appreciate that you are coming from this in complete good faith on the basis of sincerely held beliefs but please don't vote against this update just on that basis. It has too many other improvements in it to be anything other than a net benefit.  Roger Davies ^talk 21:56, 9 June 2011 (UTC)

Your points are very well taken, and I think you raise some important things to consider. I guess I should be a little more clear with my opinion. I agree that these are all serious issues however I think that arbcom should be public and that if there are serious issues that should not be public then they should be handled via office actions or by the foundation itself the advice of the wikimedia foundation legal council, not by arbcom.

It is possible to separate out what should be private from the public facets of arbitration. A simple summary of the dispute like "user BadWikizen outed InnocentUser, and the arbcomm has been emailed the offending edits" is all I'd ask for. If this is not possible I don't think it's a matter for ArbCom. The reason for this is ultimately that we elect arbiters expecting they will engage in public arbitration and we elect trustees with the expectation that trustees will be involved in office actions that are not always public. I think vesting both of these important abilities (oversight-type bans as well as public arbitration of disputes) in the same hands leads to a conflict of interest.

What is important here is that no one could accuse the arbcom (rightly or wrongly) of making wikipedians disappear into night and fog. That perception, while false, would create way more drama than keeping things public and open and redacting and removing as little information as is possible from the public eye. HominidMachinae (talk) 22:15, 9 June 2011 (UTC)

It is indeed often possible to separate out private matters from public ones. Public cases do sometimes contain findings that allude to oversighted and otherwise private material. However, it is simply not possible to have a meaningful public case which is essentially about something very private. We often can't even conceal the identity of the victim because it can be guessed from edit histories and so on. Nor, if the editor has a bunch of friends, is it possible to have one without offering any proof, particularly if the editor denies doing anything wrong or claims what he says was misrepresented, without the whole thing becoming a huge drama-fest.

Actually, I don't believe that arbitrators are elected to deal with everything publicly. My take is that the community elects arbitrators because they trust them to handle all their various roles.

As a matter of policy, the WMF is easing out of office actions because it jeopardises their public carrier status. (The legal principle here is that as a public carrier they are not liable for defamation or copyright infringement because they do not control content. With 491 wikis, it would of course be impossible for them to do so. Office actions that do impact on content or conduct undermine this position.)

We have done a great deal over the last couple of years to make things more transparent: the noticeboard was an ARbCom initiative; publishing our internals procedures was another; releasing regular statistics is a third. In my experience, though, people can and will make all kinds of allegations and it would utterly impossible to do anything if they were taken seriously.  Roger Davies ^talk 23:08, 9 June 2011 (UTC)

PS: Administrators, by the way, can and do make editors disappear in the night and fog all the time, often with no more than a couple of words on a block log. Why aren't they required to do everything transparently and publicly, diligently setting out a skeleton case for each indefinite block at AN/I?

I see, so because admins aren't required to painstakingly record all of their actions, ArbCom doesn't need to either! What terrible logic. It is impractical for admins to record every block, considering some do upwards of a hundred a day. ArbCom hardly does that many, though, so it is definitely practical (and good practice) for them to publicly announce outcomes of even private investigations. With all due respect to people's privacy, which I believe should be maintained, ArbCom should also be operating at least partially in the open, not with the possibility of operating entirely in the dark. Ajraddatz (Talk) 23:39, 9 June 2011 (UTC)

There's terrible logic at play all round here :) It arises out of the difficulty in reconciling to everyone's satisfaction the very real tensions between two important principles: openness to the community versus harm to people.

First, ArbCom places very few blocks and is only circumspect about a subset of those.

Second, it is circumspect about these blocks because it wishes to protect the privacy of individuals concerned.

What might be good practice for blocks is not good practice for privacy so it comes down to a simple moral choice. Is it morally acceptable to sacrifice the vulnerable victims of abuse on the altar of openness as a public relations exercise? Is it acceptable to cave in to baying for prurient detail? No. I'd rather take the flak.  Roger Davies ^talk

Roger, I'm assuming that this "abuse" is off-wiki abuse, since on-wiki hounding, harassment, etc is already fully public. (Some concrete examples of situations requiring secret trials and blocks would be extremely helpful. FT2's above were not convincing.) While I find it admirable that the Committee feels morally obligated to investigate and intervene in what it feels is inappropriate off-wiki activity, my (perhaps callous) view is that a) policing off-wiki activity is simply not Wikipedia's problem; b) it is impossible for the larger community to tell whether Arbcom's intervention is doing more harm than good, whether due to deception and gaming by "victims" or plain old mistakes, incompetence, and unintended consequences; and c) allowing Arbcom to become embroiled in off-wiki incidents exposes the project to severe legal and PR liability. TotientDragooned (talk) 06:20, 11 June 2011 (UTC)

It is perfectly ridiculous to call blocks, which would and could have been placed by an admin completely in compliance with policy had an admin been known the facts, "secret trials". Contrary to your view, ArbCom actions do not embroil either the project or the Foundation in any legal consequences. But, conversely, in many situations, there may be appalling consequences if we don't act. Your assumption that this is all to do with off-wiki conduct is incorrect: oversighted death threats or posting home addresses being obvious examples. The most serious cases often involve a mixture: for example, heavily veiled on-wiki threats plus creepy phone calls; off-colour talk page remarks plus wildly inappropriate emails. (I'm not going to say more on WP:BEANS grounds.) The perpetrator frequently admits the facts when confronted.  Roger Davies ^talk 07:08, 11 June 2011 (UTC)

On the contrary, if you, in your role as an administrator, want to block a user, I see nothing wrong with that and didn't intend to imply otherwise. But I'd expect the block log to hold up on its own, without needing to reference "per private Arbcom decision" or anything to that effect.

• That's normally precisely what happens. If someone emails the list, and a Committee member acts on that email, are they acting as an an admin or an arb or an oversighter?  RD

I'm disappointed that we can't discuss a detailed concrete hypothetical scenario; if someone wants to commit crimes on the internet they hardly need to work as hard as trawling arbitration policy archives to find inspiration. I'll comment the best I can on the sketches you've given above.

• Well I could do but it probably wouldn't be very good and we'd only end up arguing it infinitely anyway :) Despite your scepticism, I'm also reluctant to go into detail on this kind of stuff for WP:BEANS reasons.  RD

"oversighted death threats": I'm appalled that Arbcom would even consider doing anything official here, other than possibly categorically forwarding the threats to the Foundation. What do you think would happen if a Wikipedia editor came to harm after an on-wiki death threat, and it was later leaked that the threat was officially discussed on arbcom-l, with the Committee deciding to do nothing?

• What makes you think we wouldn't notify the foundation and also advise the victim to contact law enforcement?  RD

"posting home addresses": The oversighter blocks the editor with block log, "outing a private home address." Why is there any need to involve the Committee at all?

• Because editors do bring these things, of their own volition, by email to the list and they need urgent attention. Nearly all arbitrators are also oversighters.  RD

"heavily veiled on-wiki threats plus creepy phone calls" What evidence is there the calls actually took place? Even if there were some way for Arbcom to perform real-life investigations of this nature, these calls are a serious matter for the victim and his local police department: Arbcom investigating the calls or confronting the alleged caller in any way is exactly what I mean by "embroiled" and is a terrible idea.

• Because the perpetrator admitted making them, claiming they were a joke? But if these are tied into on-wiki activity then, per policy, it's a clear conduct issue. Again, what makes you think we wouldn't advise the victim to contact law enforcement?  RD

"off-colour talk page remarks plus wildly inappropriate emails." Again, how can Arbcom possibly determine the emails were actually sent by the editor? If the emails are criminal, then it is a matter for the police, and wildly inappropriate for an Internet play court to try to intervene. If the emails are merely annoying, then they are none of Arbcom's business. As an independent administrator aware of the problem, you might consider giving the victim advice on how to set up an email filter.

• The "Email this user" feature come squarely under admin jurisdiction and as I mentioned off-wiki stuff is explicitly included in the harassment policy.  RD

TotientDragooned (talk) 08:13, 11 June 2011 (UTC)

• Replied inline and signed "RD".  Roger Davies ^talk 10:30, 11 June 2011 (UTC)

That's kind of what I'm getting at. I am uneasy with anything being entirely private in a community that operates by consensus. The Arbcom is by its essential nature different than anything else on Wikipedia. They are the ultimate "final say" in conduct matters. There is no actual equivalent for content disputes, but conduct disputes are different, they have to have someone capable of sorting everyone out and imposing rules in order to keep the encyclopedia from degenerating into anarchy. The fact that they operate so extraordinarily out of what is normal means they must be open and accessible to avoid eroding trust. HominidMachinae (talk) 00:41, 10 June 2011 (UTC)

And that where we differ, I think. Arbcom is the deus ex machina when our community consensus-driven model fails and we need to temporarily sacrifice our principles to Make Nasty Intractable Problems Go Away. And that sometimes means violating consensus, sometimes being unfair to someone, and sometimes needs discussions in private. Perhaps as a (former) mathematician familiar with Godel's Theorem I am less disturbed than most by operating in a system where I know there are conditions on the boundary where the system fails, but I do think that's true of a lot of systems, including Wikipedia. Now, we don't want Arbcom to intervene in these contra-wiki-philosophical ways more often than absolutely necessary, and that's why we elect only a handful of people we trust to take on the role (and while we've had a few snafus, generally they have repaid our trust), and we also need an Arbcom policy that constrains reasonably when these deus ex machina approaches are to be considered. But I don't want to tie their (and our) hands by saying "Never!" when there are clear cases where it's necessary. Martinp (talk) 03:24, 11 June 2011 (UTC)

But, at the same time, do you want to widen their scope so much that they end up slowly being able to do far more than the community intended them to do, and especially for things like appointment of CU/OS rights, which works just fine in the hands of the community? You're right, ArbCom exists to sort out the issues that consensus can't... so then why are they expanding so far beyond that that some people can reasonably call them GovCom instead, due to their seemingly endless scope? Ajraddatz (Talk) 04:44, 11 June 2011 (UTC)

Hold on a minute. The Arbitration Committee appoints CheckUsers and Oversighters under remit from the Wikimedia Foundation; it is not an English Wikipedia rule but one from the WMF that applies to all projects with Arbitration Committees, with very limited exceptions. Originally, only selected arbitrators were permitted access to these tools, and it is the Arbitration Committee that has brought the community into the selection discussions and expanded it to include community- and Arbcom-trusted users. The community has not been able to develop a mechanism that will successfully select checkusers/oversighters, monitor their activities, and provide a mechanism for removal of these tools: the Arbitration Committee has done all of these things, and has included the community in the development of the processes where applicable, even including three community members in the Audit Subcommittee. Arbcom is NOT GovCom, much as some community members wish it was, and the updating of this policy accrately reflects that. Risker (talk) 05:15, 11 June 2011 (UTC)

I'm striking my oppose vote, this issue is a relatively minor one compared to the scope of the changes needed and made. It can be dealt with through separate means. HominidMachinae (talk) 19:57, 11 June 2011 (UTC)

% of support

As of 22:17, Sunday June 12, 2011 (UTC), there is 70% of support for the new policy.
Just a note
~~EBE123~~ ^talk_Contribs 22:17, 12 June 2011 (UTC)

133 support votes, 19 oppose votes = 152 total votes. That is an 87.5% support rate. Please check your math. Risker (talk) 22:24, 12 June 2011 (UTC)

Just a estimate. ~~EBE123~~ ^talk_Contribs 22:37, 12 June 2011 (UTC)