Wikipedia talk:Arbitration/Policy/Update and ratification/Archive 3

From WikiProjectMed
Jump to navigation Jump to search
Archive 1 Archive 2 Archive 3

Reviewed by the 2011 ArbCom

The 2011 committee has now had an opportunity to review the draft and has made a number of changes. Most of these are for concision or clarity or style though many sections haven't been touched at all. The major significant addition is the "Transparency and confidentiality" section, the thrust of which fell through the cracks from the current policy.  Roger Davies talk 08:01, 13 April 2011 (UTC)

So when will the Committee be adopting the new policy? It seems as near to perfect now as it ever can be. AGK [•] 23:01, 15 April 2011 (UTC)
We're currently discussing the exact timeline for the ratification process; but I would guess that we'll move forward within the next few weeks. Kirill [talk] [prof] 03:03, 16 April 2011 (UTC)
Looks good. A few remarks:
  • "Disclose any alternate accounts ..."—I don't see the need for the two commas; they make the meaning a bit wobbly.
  • "In certain circumstances, the Committee may take notice of off-wiki conduct where such conduct impacts adversely upon the project or its editors." When it was "In exceptional circumstances," the first phrase was worth keeping. Now I can't see the purpose, so can it be removed? "take notice of" might be better as "consider".
  • "While the Committee will typically take into account its earlier decisions when deciding new cases, previous decisions do not create binding precedent. Additionally, as community policies, guidelines and norms evolve over time, previous decisions will inform the Committee only to the extent that they remain relevant in the current context."—This seems to contradict itself: "do not create binding precent", but then "will inform" (I'd say "may inform", or something else that is not so black and white). Tony (talk) 00:05, 16 April 2011 (UTC)
The intent of the last sentence is to set out conditions for exclusion rather than inclusion ("...will inform the Committee only to the extent..."); perhaps changing it to "...will be taken into account...", to mirror the preceding sentence, would clarify the meaning? Kirill [talk] [prof] 03:03, 16 April 2011 (UTC)
Thanks for the input, I've pulled these about a bit, to implement the first point, and tweak the second and third ones.  Roger Davies talk 07:45, 16 April 2011 (UTC)
On further thought, I edited #3 to read "The Committee will sometimes consider off-wiki conduct but only when such conduct impacts adversely upon the project or its editors".  Roger Davies talk 06:27, 21 April 2011 (UTC)
Yup, this is good (#3). Also, you might consider "The Committee may consider off-wiki conduct if believes such conduct may impact adversely on the project or its editors", which removes doubt that might have to consider from time to time to adhere to this requirement, and places the call firmly with the Committee (which is entirely appropriate, IMO). I'm being fussy. Tony (talk) 14:09, 21 April 2011 (UTC)

Supplementary documents

The text previously in the supplementary documents section has been now been copy-edited and adopted by the Committee]. It will be merged into applicable documents (primarily Arbitration Committee/Procedures, Committee sub-pages, and page templates) in due course.  Roger Davies talk 07:59, 17 April 2011 (UTC)

Final draft

For comments on the announcement,  Roger Davies talk 10:46, 21 April 2011 (UTC)

Suggestion re: transparency of rationales

First of all, congratulations and thanks to the drafters on work done so far; the current draft is a significant improvement over the existing policy. The Transparency and confidentiality subsection currently reads (emphasis mine):

Committee deliberations are often held privately though the Committee will make public detailed rationales for decisions related to cases unless the matter is unsuitable for public discussion for privacy, legal or similar reasons. All communications sent to the committee—or from a committee member in the performance of their duties—are private.

I realise this is largely a holdover from the existing policy, but would strongly suggest this clause be expanded to all decisions significantly effecting editors, – not just those related to cases. Lack of substantive rationales in the announcement of decisions, particularly where cases are not concerned, is the one area I find the Committee consistently underperforming in. I am speaking of instances like the involuntary removal of rights, significant unilateral changes in processes.

Decisions often seem to be announced absent sufficient explanation, leading in many cases to prolonged discussions requiring multiple arbitrators to justify and elaborate on the decision (i.e. a lot more effort than had adequate rationale been initially provided). As well as the unnecessary drama these situations involve, they tend to sow mistrust of the Committee in the community (given insufficient info about arbitrators' motivations, editors are left to make assumptions, and being Wikipedians don't often make use of much good faith in the process). Underexplained decisions can be unfair to those directly affected (tarred with a black mark without having anything public to defend themselves against).

While arbitrators are rightly constrained in a minority of instances by privacy/legal etc. concerns, and I certainly wouldn't want that weakened in the policy, these instances too can be better handled, and the policy ought to offer encouragement. It may be an overreaction to the fallout over the removal of advanced permissions from a former functionary, but there is often no explanation provided as to why no explanation is provided, merely a "please contact the Arbitration Committee" or similar. If you're indefinitely blocking someone w/o talk or email access in response to threats of violence made in private, say, it's understandable not to write "Said she would ritually disembowel editor Susie Smith of 56 Sharptown Road, Gloucester" as the blocking reason, but it's a far distance from that to "Please contact the Arbitration Committee regarding this block". Reviewing administrators ought at least be able to tell if the offence was public or private, sockpuppetting-related or not.

Some recent instances of discord over this problem:

Some examples of good pre-emptive exposition:

So to reiterate, I would petition the Committee to adopt substantive rationales as a matter of best practice in all significant matters, and to reflect this in the policy by dropping "related to cases" from the relevant clause. Appreciate your reading, Skomorokh 12:48, 21 April 2011 (UTC)

Oddly enough, I started to write a response to this and stopped dead in my tracks, because you've identified specific situations and individuals, and I don't want my comments to be linked with the individuals involved in these situations. I'll think for a while on how to address these issues. But I'll tell you honestly that I'm not about to start blurting out that Admin PRQST was desysopped because of personal health issues that were creating disruption in the project. The disruption part, maybe; but certainly not the non-public personal information of the individuals involved.

I'm not really seeing a lot of connection between several of the examples you've given. Why are you grouping the advanced permissions issue (a policy matter) with decisions related to individuals? Risker (talk) 13:44, 21 April 2011 (UTC)

And interesting question, Skomorokh, and some general observations in response.
  • The expression "case" really refers to proceedings generally, so I suppose that ought to be tweaked.
  • More generally, we do deal with (and therefore I suppose technically deliberate on) all sorts of very varied stuff not directly related to cases. These include COI name block appeals; persistent appeals from long-term banned users; requests for advice in all sorts of circumstances; enquiries from the general public; lots of stuff sent to us instead of OTRS; complaints about admins; lengthy messages on all kinds of topic from conspiracy theorists; tip-offs about all sorts of things (rarely acted on); and general well-poisoning about other editors. With our limited resources, we have enough trouble getting timely replies back let alone publishing details for each of them, with rationales: it is exceedingly unlikely anyone would read them anyway.
  • In the fictional (disembowelling) example you give, the block log probably would probably refer to threats. Non-narrative blocks are normally reserved for issues which, because of other policies, cannot be discussed on-wiki. Examples include child protection; mental health problems; suicide threats; and so forth. In these instances, even providing a reason why no reason is given is enough to defeat the purpose of confidentiality.
  • I agree that we are sometimes not good at getting rationales out speedily but a great deal of this is down to the logistical difficilty of getting 18 volunteers scattered in different zones all around the world, and doing things in their spare time, to agree a joint statement. If the text changes during the process and needs re-approval, even getting something simple out can take two working days to get majority support. This is something we are working on though.
I hope that answers everything,  Roger Davies talk 14:39, 21 April 2011 (UTC)

Jurisdiction over advanced permissions

The current draft addresses the Committee's role in the management of advanced permissions (e.g. CheckUser, Oversight) in Scope and responsibilities as follows:

The Arbitration Committee of the English Wikipedia has the following duties and responsibilities:...
To approve and remove access to (i) CheckUser and Oversight tools and (ii) mailing lists maintained by the Arbitration Committee.

By my reading, this means that as far as this area is concerned, Arbcom's only role is to select/deselect personnel. Advanced permissions are not mentioned in Jurisdiction as an area over which the Committee claims jurisdiction, although historically individual arbitrators have made this claim (arbitration pages are another example, though they are alluded to in "The Committee retains jurisdiction over all matters heard by it"). A recent example of this presumed jurisdiction beyond personnel selection would be the request to unilaterally alter the technical abilities associated with CU/OS, but it's come up in other situations before, and language like "ArbCom oversee/have authority over CU/OS matters" is often deployed. Some other hypothetical moves the draft policy would seem not to allow would be ArbCom altering the requirements for CU/OS (e.g. to make adminship a requirement) or to dissolve the functionaries mailing list if it became compromised.

So while I personally favour the restricted role, I wonder does this interpretation have the support of the full Committee, have I misread the historical stance of broad authority or am I reading the draft too narrowly? Thanks, Skomorokh 13:11, 21 April 2011 (UTC)

I think you're misreading this. By having the jurisdiction to approve and remove access to CheckUser and Oversight tools, that also means we have the jurisdiction to gather information in which to inform our approval/removal decisions, such as monitoring tool usage, determining standards by which we will base our approval/removal decisions, outlining performance expectations, reviewing complaints about usage, establishing criteria for candidacy for these roles, and so on.

With respect to the mailing lists, I don't think we'd have very much pushback from the WMF to shut down a compromised mailing list within our jurisdiction, given that all of the mailing lists that are within our umbrella deal with non-public information to some extent or other. Risker (talk) 13:34, 21 April 2011 (UTC)

Comments from Carrite

  • Comment - Everything was beautiful, clear, and perfect down to the very end, which stated that decisions of the ArbCom could be appealed (and potentially annulled) by the unelected personage of Jimbo Wales. That is absolutely undemocratic and unacceptable. Carrite (talk) 14:36, 21 April 2011 (UTC) Moved from the draft policy page. Risker (talk) 14:41, 21 April 2011 (UTC)
That's certainly a lot of power for a person who professes to be just another editor, eh? Final authority over all arbitration... That 2004-era anachronism seriously needs to go in favor of some formal procedure for appeal to the Wikimedia Foundation or a body of its chosen representatives. Carrite (talk) 15:06, 21 April 2011 (UTC)
It is really a lot of power? Jimbo's actions are as open to criticism as anyone else's and I imagine that if he abused the position (which he hasn't and I'm not aware of any suggestions that he has in the seven years the policy has been operational) the community would react very strongly. Conversely, in earlier discussions (see archives) on the same theme, some people expressed the view that Jimbo's role was a useful safety valve in the event of a maverick ArbCom.  Roger Davies talk 15:10, 21 April 2011 (UTC)
I, for one, am far more worried about a maverick Jimbo than a maverick ArbCom. Personally I think it should be the other way around, decisions made by Jimbo should be appealed to ArbCom. Sven Manguard Wha? 20:21, 21 April 2011 (UTC)
The closest "ArbCom" came to being a maverick in recent years was the ill-fated WP:ACPD. More usually an arbitrator acts on their own as a maverick, and either ends up worse for it, ArbCom backs them, or .. umm .. ArbCom closes ranks .. ;-)
A Jimbo decision to block admin user:Bishonen was appealed to ArbCom but the case didn't proceed[2] and WP:JIMBO records that Jimmy has said that "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."[3] John Vandenberg (chat) 12:35, 22 April 2011 (UTC)

I rather agree that providing for an appeal to the WMF would be better. But do note the rider "Remedies may be appealed to, and amended by, Jimbo Wales, unless the case involves Jimbo Wales' own actions." In theory, might that not mean that a successful appeal to Jimbo could be overridden by Arbcom taking into account Jimbo's actions in relation to the appeal? Rd232 talk 02:15, 22 April 2011 (UTC)

For legal reasons, appeals to the WMF are not really an option.  Roger Davies talk 03:10, 22 April 2011 (UTC)
Roger, you might consider cutting unless the case involves Jimbo Wales' own actions down to simply unless the case involves Jimbo Wales to be in line with WP:COI/WP:UNINVOLVED – broad-based definitions to discourage wikilawyering. It would be inappropriate for Jimbo to rule on any issues closely involving him, not just those where he himself has made an action. Actions by those he has a close relationship to, or controversies regarding accusations against his person would be some hypothetical situations where it would be improper. Skomorokh 11:38, 22 April 2011 (UTC)
"for legal reasons"? Do explain, please. Rd232 talk 07:29, 23 April 2011 (UTC)
The Arbitration Committee is a creature of the English Wikipedia and is responsible only to this project; the Wikimedia Foundation has no authority over it, nor does it accept such a responsibility for this project or any other. The activities of the project are the purview of the individual project, provided that it operates within the (very limited) guidelines set forth by the WMF. This reduces the likelihood of a successful claim that the WMF is responsible for all of the content and activities on the hundreds of projects it hosts. Risker (talk) 07:41, 23 April 2011 (UTC)
Mmm, makes sense. Though the though occurs that a similar logic applies to Jimbo... Rd232 talk 03:39, 27 April 2011 (UTC)
I'd like to see the Jimbo bit go, myself. I have no antagonism against him, or anything like that, but there was a time and place when such language was needed, but I think that need has passed now. Heck, he's said himself that he's stepping back from governance over the last few years (or something to that effect...). Maybe we should create a DRV like process in which bureaucrats could review decisions, or something? (I like the appeal to the Foundation myself, but I don't see it happening.) You know, there should be an arbitrator recall procedure, as well (but that is a whole different conversation).
— V = IR (Talk • Contribs) 18:45, 23 April 2011 (UTC)
I don't have anything against Mr. Wales either, but democracy doesn't have any place for a king. If there is a real need for a check upon the actions of ArbCom (I'm not convinced of that), then it should be via a mechanism of elected delegates, not by decree of an individual. I'd prefer the whole section be stricken, myself. Carrite (talk) 15:46, 2 May 2011 (UTC)
I'd love to see the appeal ad Jimbonem removed as well. It is a longstanding feature of the Arb policy and one that made sense when the ArbCom was considered a creature of Jimbo and derived its legitimacy from him rather than from the community directly (e.g. Jimbo appointed the first Arbitors rather than holding an election). However, in the current environment I view it as an unnecessary anachronism. Wikipedia is not necessarily a democracy but I don't see the need for a King either (or for Jimbo to be the monarch if one is necessary). Eluchil404 (talk) 09:34, 3 May 2011 (UTC)

Question from Fetchcomms

Sorry if I missed it somewhere, but is there a list of all the differences between this draft and the current policy? /ƒETCHCOMMS/ 14:48, 21 April 2011 (UTC)

I'm afraid not. This one was written from scratch, based on the old one, rather than edited from it. After two years in gestation, and a huge number of revisions, a single diff would be impossible.  Roger Davies talk 14:55, 21 April 2011 (UTC)
(edit conflict) I remember asking that very question a while back. It might be better to read and absorb both to get an idea, because this version is substantially different from and the old version, as might be expected in nearly six years of wiki-evolution. A shopping list of individual changes might end up being more difficult to grasp than a simple reading of the actual versions. Just my suggestion. Tony (talk) 14:57, 21 April 2011 (UTC)

Question from Hasteur

Perhaps I'm missing it, but I do not see a explicit requirement for candidates for Arbitrator be Administrators. I know that "admin is no big deal" yet the discussions about unbundling the CU/OS/RevDelete permissions from Admin role are of sufficient contention that I would like to know if the policy document should be revised to add this requirement or if the committee is intending to lobby for the policies elsewhere in the community to have the necessary elevated permissions added to the Arbitrator Role. Hasteur (talk) 15:31, 21 April 2011 (UTC)

The decision on who is a suitable candidate to be an arbitrator is probably one better left to the community as a whole, reflecting the beliefs and philosophies of the community at the time of any given appointment cycle. I'd be very hesitant to include anything more than the bare minimum. You'll note there is also no minimum edit requirement, no tenure requirement, and no requirement that the candidate have any history of certain activities. Aside from the identification requirement (which comes from WMF policies) and the disclosure of prior accounts (which comes from the community particularly, although both the Committee and Jimmy Wales are strongly supportive of it), the community has complete responsibility for analysing candidates and determining their suitability. Risker (talk) 15:42, 21 April 2011 (UTC)

Proposed changes by Sven Manguard

Cross-WMF participation

Where it says: "The Committee will sometimes consider off-wiki conduct but only when such conduct impacts adversely upon the project or its editors."

Consider: "The Committee will sometimes consider cross-project or off-wiki conduct but only when such conduct impacts adversely upon the project or its editors."

Why: The Committee, in the past, has weighed the behavior on and contributions to other WMF projects when deciding on unblock requests by users banned on this project.

Cross-wiki stuff is already covered to some (probably a sufficient) extent in the evidence section, allowing, for example, block logs to be raised.  Roger Davies talk 03:36, 22 April 2011 (UTC)
Works for me. I'm a fan of explicit documents, but this is implicit enough already. Sven Manguard Wha? 05:41, 22 April 2011 (UTC)
Jimbo and elections

Where it says: "Members of the Committee are appointed following annual elections organized and run by the community. Candidates must:"

Consider: "Members of the Committee are appointed following annual elections organized and run by the community. The results of the elections are formally announced by Jimbo Wales once the votes have been certified, however Jimbo Wales cannot overrule the community vote. Candidates must:"

Why: This became a topic of discussion last election. Technically Jimbo has the right to appoint anyone he wants, however that would be problematic. If memory serves, the Signpost ran the winners list before Jimbo announced them anyways, since technically his was a rubber stamp role anyways.

As Risker points out above, this really is something for the community to determine separately on a case-by-case basis.  Roger Davies talk 03:36, 22 April 2011 (UTC)
I disagree here. The election process is decided in one of four ways, or a combination of the four. 1) Volunteer coordinators decide things and by virtue of being the only ones volunteering, it becomes an enforced mandate unless... 2) An RfC is held to get community input (note that historically these are always held too late in the cycle to make any big changes) which dictates details of the election, which are enforced unless... 3) existing policy documents (like this one) dictate certain things must or must not happen. An RfC would have to specifcly overrule said documents, and as mentioned above, that won't happen because RfCs don't get held until right before the elections start. Policy documents trump all unless... 4) Jimbo overrules something he does not like, and dictates that something is to be done or not done. My point is that that fourth level needs to be thrown out; the full weight of ArbCom and the community needs to say "No, in ArbCom elections, Jimbo cannot be involved." Sven Manguard Wha? 05:41, 22 April 2011 (UTC)
Equally, this policy is broadstroke and should avoid getting sucked into the minutiae of elections and appointments. It is also descriptive of de facto policy rather than a prescriptive wish list ;)  Roger Davies talk 06:04, 22 April 2011 (UTC)
Identification and elections

At the bottom of the section: "Selection and appointment"

Consider: "Anyone who wins a position on the Committee will not receive permissions or mailing list access, and will not be allowed to serve on cases, until self identification to the foundation is confirmed. If a candidate states during his or her election that he will not self identify, and that candidate is elected for a position, this stipulation overrules their campaign statement.

Why: This also became an issue last election. Policy is conflicting, and this needs to be set in stone.

Per WMF policy, candidates who don't identify don't get access to privacy-related material: that's crystal-clear and trumps anything here.  Roger Davies talk 03:36, 22 April 2011 (UTC)
That policy needs to be carved into a metal plate and hung at the doors then. In other words, I want that to be specifically stated, on it's own page, with the policy banner attached, and I want that before the election starts. This was a pain to deal with last year. Mind you I'm not asking/demanding you do it, just saying what I want to see. Sven Manguard Wha? 05:41, 22 April 2011 (UTC)
It probably already has sufficient emphasis for all practical purposes :)  Roger Davies talk 06:04, 22 April 2011 (UTC)
The Police

Where it says: "Preserve in appropriate confidence the contents of private correspondence sent to the Committee and the Committee's internal discussions and deliberations."

Consider: "Preserve in appropriate confidence the contents of private correspondence sent to the Committee and the Committee's internal discussions and deliberations. The Committee, in cases where it is deemed necessary, can pass information to relevant legal authorities. Unless restricted from doing so by law, the Committee must notify in private any user whose information was passed to law enforcement.

Why: We all know that the committee deals with it's fair share of pedophiles and other psychopaths. However it is never specified as to what the committee does, short of a ban.

Again, there are various en:wiki and WMF policies that cover these kinds of situation (child protection, self-harm, harrasment etc) and ArbCom policy needs to be read in conjunction with them.  Roger Davies talk 03:36, 22 April 2011 (UTC)
Okay. As I said above, I love the explicit documents, so as long as they are there, and can be found and linked to, I guess I'm fine. Sven Manguard Wha? 05:41, 22 April 2011 (UTC)
Cool :)  Roger Davies talk 06:04, 22 April 2011 (UTC)

Potential topics for clarification from Serpent's Choice

  • On the "binding" nature of arbitration: Section 1.2.1 identifies the Arbitration Committee as "a final binding decision-maker", but 1.3.8 states that decisions of the Committee "do not create binding precedent" (emphasis mine). While I know what is intended by both of these, the verbiage is in conflict. At the very least, there should be explanatory text somewhere.
    • As I'm sure you realise, what is means is that the Committee is not bound by its own past decisions. I'll see if the language can be improved to make this clearer. Roger Davies talk 04:32, 22 April 2011 (UTC)
      • Right. I don't have any personal confusion about the situation, but it might be unclear to people not familiar with the established structure. Maybe reword the precedent section to avoid having a parallel wording? Nothing specific to suggest at the moment. Serpent's Choice (talk) 13:42, 22 April 2011 (UTC)
        • Maybe remove the "binding precedent" wording and state that the committee does not follow stare decisis, which is the Latin legal term for the concept behind binding precedent. Imzadi 1979 → 20:47, 27 April 2011 (UTC)
  • On removal of sitting arbitrators: Section 1.2.3 provides conduct mandates, and a mechanism for suspension or removal of Arbitrators who fail to uphold them. Although "motions" are traditionally proposed solely by the Committee, there is no formal statement of that in the policy draft, leaving open the potential misconception that such motions could be offered by other parties.
  • On recusal: It is probably worth clarifying that an arbitrator's routine work in their capacity as an editor also does not typically mandate recusal. Suggest the sentence regarding routine actions in section 1.2.4 instead read, "An arbitrator's previous routine interactions as editor, administrator, or arbitrator are not usually grounds for recusal."
    • Yes, this is probably worth including.  Roger Davies talk 04:32, 22 April 2011 (UTC)
      • Glad to help. I seemed to remember that having actually come up at some point. Serpent's Choice (talk) 13:42, 22 April 2011 (UTC)
  • On transparency: The clause in section 1.2.5 about "communications ... from a committee member in the performance of their duties" is unclear. This could be read to imply that private communication originating from the committee carry the weight of a gag order preventing their recipient from repeating or discussing them elsewhere. I am uncertain if that is intended and offer no suggestion as to a clarification; communications related to the private hearings in 1.3.3 may have their own expectations of privacy.
    • No, that's not the intended meaning.  Roger Davies talk 04:32, 22 April 2011 (UTC)
      • I didn't seriously believe that was the intent, but I'm having some difficulty determining what that clause does intend. Are you able to provide some insight? Serpent's Choice (talk) 13:42, 22 April 2011 (UTC)
  • On jurisdiction: The statement about "off-wiki conduct" in section 1.3.1 is in conflict with the statement about evidence from Wikimedia products other than the English Wikipedia, in 1.3.5. Additionally, that statement in 1.3.1 is not, strictly speaking, a matter of jurisdiction and so may be misplaced in that section to begin with.
    • I'm not so sure that they do conflict as the first is purely about conduct, whereas the second is about probative material. Roger Davies talk 04:32, 22 April 2011 (UTC)
      • I think the problem is really that the fourth statement in the Jurisdiction section isn't, as currently worded, about jurisdiction. "Considering conduct" is an admissibility of evidence topic. Also, I know at one point there was talk of giving ArbCom jurisdiction over the Wikipedia IRC; I have no idea what the final resolution of that situation was, and have no horse in the race, so to speak, but it might need to be added to either the first or second statement in the jurisdiction section (as appropriate). Setting aside the IRC topic, might I suggest refactoring the jurisdiction section somewhat, as below? Serpent's Choice (talk) 13:42, 22 April 2011 (UTC)
The Committee has jurisdiction over user conduct on the English Wikipedia. The Committee has no jurisdiction over official actions of the Wikimedia Foundation or its staff, or over Wikimedia projects other than the English Wikipedia. The Committee has no jurisdiction over conduct occurring outside the projects of the Wikimedia Foundation.
The Committee may consider conduct outside its jurisdiction in making decisions about conduct on the English Wikipedia when such outside conduct impacts adversely upon the project or its editors.
The Committee retains jurisdiction over all matters heard by it, including associated enforcement processes, and may, at its sole discretion, revisit any proceeding at any time.
With a bit of further tweaking, it comes out like this:
The Committee has jurisdiction over user conduct on the English Wikipedia.
The Committee has no jurisdiction over: (i) official actions of the Wikimedia Foundation or its staff; (ii) Wikimedia projects other than the English Wikipedia; or (iii) conduct taking place outside the English Wikipedia.
The Committee may take notice of conduct outside its jurisdiction when making decisions about conduct on the English Wikipedia if such outside conduct impacts adversely upon the project or its editors.
The Committee retains jurisdiction over all matters heard by it, including associated enforcement processes, and may, at its sole discretion, revisit any proceeding at any time.
I'm okay with this as it's substantially clearer, with no change in meaning.  Roger Davies talk 01:23, 24 April 2011 (UTC)
Agreed, that substantially satisfies my concerns. Serpent's Choice (talk) 17:21, 26 April 2011 (UTC)
Tweaked the draft,  Roger Davies talk 04:25, 27 April 2011 (UTC)

From AGK

  • Note from me: in #Selection and appointment, the first point reads: "Meet the Wikimedia Foundation's criteria for access to non-public data and confirm they will fully comply with it in their election statement". That should be "Meet the Wikimedia Foundation's criteria for access to non-public data and confirm in their election statement they will fully comply with it" (word order changed and underlined for emphasis). The less capacity for gaming, the better. AGK [•] 22:13, 21 April 2011 (UTC)
    • Yes, that is better. I'll tweak it.  Roger Davies talk 04:33, 22 April 2011 (UTC)
      • I haven't looked at the draft to see what ended up there, but now "it" is unclear; in the upper version the pronoun refers to data as singular, I presume (which is fine ... or is it meant to refer back to the Foundation's criteria? If so, you might consider ending the new version with "... comply with the criteria", or something like that. Tony (talk) 08:36, 26 April 2011 (UTC)
        • Done,  Roger Davies talk 04:25, 27 April 2011 (UTC)
          • Yeah, that's better still :). AGK [•] 12:42, 4 May 2011 (UTC)
  • Another one for you Roger: #Forms of proceeding should be Forms of proceedings (emphasis added), should it not? Or am I missing some obscure grammatical tidbit there :P? AGK [•] 12:42, 4 May 2011 (UTC)
I suppose the grammar is that "proceeding" is used non-count (ie generically) per "Types of coffee stocked" as against "Types of coffees stocked" but it's six of one and half a dozen of another ... and I've no strong feelings either way.  Roger Davies talk 18:29, 4 May 2011 (UTC)
Fair point. And the pedant in me does not like the double 's' of "Forms of proceeedings", so I'm with you on this one :). AGK [•] 20:23, 6 May 2011 (UTC)

"unsuitable for public discussion"

"To resolve matters unsuitable for public discussion for privacy, legal, or similar reasons;" Examples of such matters? And is there an equivalent clause in the current policy/practices? --Cybercobra (talk) 09:38, 22 April 2011 (UTC)

As mentioned above, these include but are not limited to: child protection; mental health problems; suicide threats etc. Other examples could include: defamation; and certain kinds of RL harrassment. The discretion aspect is mentioned in other en:wiki and WMF policies.  Roger Davies talk 12:08, 22 April 2011 (UTC)
Seems fine then. I was just concerned what exactly this meant due to how generally it's phrased. I would perhaps suggest including the list in the policy, but I don't see a way to do that without making the bullet point in question ungainly. --Cybercobra (talk) 12:39, 22 April 2011 (UTC)
I agree, shorter is better and the proposal is very well done in its brief presentation. MBisanz talk 06:37, 23 April 2011 (UTC)

Handling private evidence

I think that in the section about private evidence, something along the lines of the following should be added: "The Committee will make a public summary of the private evidence, provided this can be done without defeating the purpose of that evidence being submitted as private". We should have the right to know as much as possible about all ArbCom cases. עוד מישהו Od Mishehu 07:47, 26 April 2011 (UTC)

I wonder whether that would defeat the purpose of the privacy? Providing advice and other activities to editors and others is a vital role for ArbCom, and this serves the project and us as individuals in a way that no other process could. We elect people we can trust to do this. Tony (talk) 08:29, 26 April 2011 (UTC)
I agree with Od Mishehu, in principle, that more transparency is better. But when it comes to private evidence that may not always be true. For example, there was a matter a while back in which a respected user turned out to have been abusing sock accounts. The method of finding the socks relied on his repeated sloppiness in some actions which clearly tied the accounts together. Finding this clue was very time consuming. If too much information about the nature of the evidence, even its broad outlines, had been shared with the user he might have deduced his errors and his socks would have become much harder to find. If the ArbCom were a court of law it would be unacceptable to use private evidence that isn't shared with the accused. But it isn't. It's purpose is to resolve otherwise intractable behavioral and administrative problems which disrupt the writing of the encyclopedia. That didn't come up in this case because the user resigned himself to having been caught. If he had insisted on more access to the evidence it would have placed the ArbCom in a difficult position. And that's on top of the numerous cases where disclosure would compromise the personal privacy of editors. The ArbCom has to follow a fine line. Too many policy instructions could make that harder.   Will Beback  talk  08:49, 26 April 2011 (UTC)
Whether or not it is a good idea, it definitely isn't happening right now, nor is it achievable in the near future as and we don't have the infrastructure or spare resources to produce summaries like this. Remember that we are also volunteers, however unlike volunteers in the community we have nobody else who can assist us with the tasks within the committee.
IMO the first priority would be to push out another six-monthly report like January to June 2009. We need to gradually get better at reporting the internals. As you can see at Wikipedia:Arbitration_Committee/Noticeboard#Internal_teams, we've started to break up the workload a bit.
Could you provide some specifics of what information you want in that summary. John Vandenberg (chat) 09:09, 26 April 2011 (UTC)
No, this doesn't work well, especially because some purported private evidence is (1) not evidence whatsoever, and/or (2) only given to us with the expectation that it remains private. I would be willing to require that evidence relied upon may be disclosed to the accused party, because fairness is owed to the accused party, although even this practice has limits—Will Beback identifies such an exception above. More fundamentally, every passerby on the internet is not owed a summary of evidence submitted in confidence. Our function is to resolve disputes, not spread gossip. Cool Hand Luke 16:28, 26 April 2011 (UTC)
I agree with Cool Hand Luke and Will Beback. Socking findings in particular are highly susceptible to gaming, and bluntly put there is no benefit to the project in revealing certain methodologies for identifying alternate accounts. Even if the targeted sockmaster takes nothing from such revelations, others might well "learn" how to avoid detection. As well, the majority of "private" communications to the committee wind up playing no significant part in the outcome of a decision. Risker (talk) 17:51, 26 April 2011 (UTC)

Serious concerns

There have been a lot of changes in this version. Some have serious concerns. In particular two of them are flat oppposes and one is a serious omission.

  1. Conduct issues - Remit has been seriously reduced. The previous wording reflected that the Committee's DR role has always been primarily behavioral issues. It's now been modified to formally exclude all but conduct issues. Even if AC never handles other disputes I would oppose a change to formalize that into policy as one cannot tell what may happen in future. This removes all final routes of resolution for every non-conduct dispute.
  2. "Similar restriction"? - "To hear appeals from blocked, banned, or similarly restricted users". But it isn't clear which "restrictions" are "similar" to bans and which are not, so this is meaningless. It also misses the point. Suggest "To hear appeals of blocks, bans and restrictions that have exhausted all community routes of appeal or which were imposed by Arbcom".
  3. "Harm" - the new wording "that may harm editors or harm or seriously disrupt the project" is overbroad (ordinary vandalism may "harm" the project!) Suggest moving a word: "...or seriously harm or disrupt the project..."
  4. Conduct' - "Respond promptly to questions" should be "promptly and fully". As drafted nothing is said anywhere about the quality of reply. Fully, completely, with full disclosure - something of that nature.
  5. Recusal (1) - "Previous routine editor...interactions". Routine editor interaction can include matters that convey a perception of non-neutrality. Perception matters. "Unremarkable" or similar may be better than "routine".
  6. Recusal (2) - Section should be explicit on the spirit of recusal, and this isn't. Also perception of bias has been dropped somewhere along the way. Reword and expand: "Perception of arbitrator non-partiality to parties is paramount. Arbitrators must recuse from any case, or any aspect of a case, in which they have, or may be widely perceived to have, a significant conflict of interest."
  7. Transparency - "All communications...from a Committee member in the performance of their duties are private. " - plainly unacceptable. If a user receives a message from Arbcom or from an arbitrator "in their duties" this effectively forbids the user to show it to ANYONE else, for ANY reason, including concerns, consultation or the like. (Even if it were supposed to just apply to arbitrators receiving a communication and not all users as drafted, arbs need to consult too).

The bold points are the ones I consider to be serious and misdrafted or misconceived. FT2 (Talk | email) 01:15, 5 May 2011 (UTC)

Regarding #7, the recipient of an email from arbcom is not covered by this policy, and arbitrators have the ability to consult the committee, and can still discuss issues with people outside the committee without sharing emails or giving specifics contained in them. If arbitrators need to provide specifics to people outside the committee, the committee should be aware of this. John Vandenberg (chat) 03:25, 5 May 2011 (UTC)
@FT2. Taking your points in order and using your numbering:
  1. 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) In any event, I suppose the key point here is that a content dispute isn't a problem unless and until it becomes disruptive and disruption, per se, is conduct based, at whch point the committee can and does get involved. The current wording reflects this.
  2. "Similarly restricted" applies to interaction restrictions, 1RR restrictions and whatever other similar restrictions evolve in the fullness of time. The current wording reflects this, as well as the wide range of appeals we do hear: /Amendments and /Clarifications; community bans; de facto bans etc. We are also the sole forum for checkuser block and child protection appeals, and - apart from unblock-l - the only effective appeal route for users indeffed for COI etc.
  3. The "serious harm" qualification seems to me to open the door to wiki-lawyering. "Yes, I harmed him, but not seriously".
  4. "Respond promptly to questions": the "fully" was removed because (a) sometimes it will not be possible on privacy legal etc grounds to respond fully and (b) because it could be used to tie arbitrators up with huge lists of questions, all of which require "full" responses.
  5. Recusal (1) - There's nothing that can be realistically or sensibly done about perceptions of non-neutrality as these are subjective and in the mind of the beholder. If an editor perceives, for example, that all blond people are biased against him that cannot be grounds of requiring all blonds/blondes to recuse. Perhaps the most common example of this is the frequent perception that because arbitrators are (so far, at least) also administrators they will automatically close ranks to defend other members of the administrator cabal: in fact, nothing could be further from the truth.
  6. See previous response.
  7. Per John.
Thanks for reading it so closely :)  Roger Davies talk 04:06, 5 May 2011 (UTC)
@FT2, ah, Roger beat me to it. I was going to answer along the lines of above - I suppose the only thing to add was differentiating "routine" vs. "unremarkable" - prefer the former really, but not by much (two syllables mainly). I figure "mundane" is another adjective we could have used. Casliber (talk · contribs) 04:11, 5 May 2011 (UTC)
  1. Arbcom has no enthusiasm for it. But it may be needed in some future issue anyway. "Having no enthusiasm for" is not the same as "excluding all possibility of in future". This ties the hands of future committees. It is not foreseeable what kinds of dispute or issue may arise in future. I believe the expression is "reserve power" - one not expected to be needed but held just in case. The ability to finally close down all types of dispute is potentially important at some point, whether enthusiastically or not. Add back "primarily" would cover it.
  2. Then clarify the point where appeals will be heard rather than use a vague word like "similarly". Arbcom's norm is it hears urgent appeals, or those which have exhausted community routes. This characterization is much clearer - and possibly also saves casework going forward as it makes the hurdle for AC appeal explicit.
  3. The point was of "serious harm to the project" which probably needs to go back in. You're going to have trouble justifying "Arbcom exclusively handles all pedophilia cases" under this rule if there's ever on-wiki drama, because deciding whether to block a user expressing support of a pro-pedophilia viewpoint does not usually involve "privacy" or "legal" reasons or anything close or "similar" to them. Other matters affected similarly. However much Committee members might believe the draft wording covers all things you think are "not suitable for public", they do not. They are actually very limiting. The Committee should retain the right to take jurisdiction of a slightly broader class of sensitive or potentially impactful issues.
  4. Then add better wording. Right now there is nothing there about quality of disclosure. Are you saying that it is not possible to frame wording that covers this omission? "Promptly and with candor ... If full disclosure to the Committee is impractical the user shall wherever possible provide details or reasons to Jimmy Wales". (Sometimes having him on the list is useful!)
  5. That some (eg the one person involved in the case) will misconceive is never a good reason to avoid the best wording we can, if the wording can be improved.
  6. Ditto. You are essentially arguing here that because some users will never be satisfied with the case's fairness, we should not bother even to express the crucially important spirit of recusal for the many who would take note of it, nor acknowledge the importance of "wide" perceptions by the community at large. I don't agree. It becomes important to say what the spirit is here, and to respect "wide perception" by uninvolved users that there is a COI.
  7. Regarding #7, the wording is what the wording is, and right now it has the effect stated. Nor does it allow (as John's answer assumes) any provision at all to provide specifics under any circumstances, to anyone. It says all communications from Arbitrators (in their Arbitrator role) are private in the hands of the recipient - and nothing else. That's how it has to be read, because that's what it says.
It was never going to be the case that a major redraft that changed so much, would get it right in all ways and not need any further editing. I think this redraft went backwards in some points - seriously backwards for a few. Please consider them more carefully. FT2 (Talk | email) 10:28, 5 May 2011 (UTC)
Again, following your numbering:
1. I'll add "primarily" back in.
2. Actually, ArbCom doesn't hear "urgent appeals" and the summary version you've suggested isn't particularly accurate (cf. the earlier examples I gave). If someone writes to ArbCom appealing mistakenly, it's trivially easy to re-direct them (we have boilerplate responses for just this purpose). I still can't see the advantages of going into detail when this policy is deliberately intended to be broadstroke and so avoid the need for referenda every few months to update unimportant details. I will though change "similarly restricted" to "otherwise restricted" as it will broadly cover any interesting variants that evolve in the fullness of time.
3. Pedophilia blocks have great potential for (i) libelling the individual and (ii) real life harm.
4. Again, let's keep this broadstroke. I'll add "and appropriately" to "promptly", which should cover your concerns.
5. & 6. As far as I can tell, dramas about recusal are few and far between. On the odd occasions when they arise, they are usually one editor waving a homemade torch and a borrowed pitchfork. I suppose the key point here is your proposal seeks to introduce something which is without precedent: admins, for instance, are not involved because of a "perception of bias" but because of actual disqualifying actions which might give rise to a perception of bias. Your suggestion seeks to place the "perception of bias" before "conflict of interest" and in a small unit, this is untenable. In any case, these things have a habit of being self-regulating: if the request gets traction, it will automatically attract the critical attention of fellow arbitrators, who will individually take a view.
7. The concerns here are trivially easy to fix. Instead of: "All communications sent to the Committee – or from a Committee member in the performance of their duties – are private", we can express it instead as: "The Committee treats all communications sent to it – or sent by a Committee member in the performance of their duties – as private". I'll tweak the text accordingly.
Thanks again for your input.  Roger Davies talk 06:57, 8 May 2011 (UTC)
#1, #4, #7 -- resolved, thank you. Those were big ones. #2 okay, can take your point on this. Final explanation on the other two:
  1. Scope - The practical issue is the scope of "things not suited for handling on-wiki". Examples: 1/ certain sock or abuse investigations, 2/ certain matters that are not privacy or legal infringing but would be inappropriate to handle on-wiki anyway (most pedo cases could be legal/privacy related but most in practice are not; nonetheless those that are not, we still want to have authority to make "AC handled"). So there are matters that we all agree should at times be handled by AC jurisdiction and handled off-wiki, even though the case may not be privacy, legal or "similar". Maybe the scope could have "primarily" added (ie "To resolve matters unsuitable for public discussion primarily due to privacy, legal, or similar reasons")? Essentially the same fix as you applied elsewhere.
  2. Recusal - Recusal is like you say, rarely a source of dispute these days, and when it is, commonsense prevails. It was more of an issue years ago. Nonetheless it would be good to formally state somewhere that non-partiality to parties is core to an arb's role, because that is 1/ the essence and spirit of Arbitratorship, 2/ worth saying explicitly, and 3/ provides broad guidance in other cases. It's not needed for case management but would be good to say because it captures so much of an Arbitrator's Wikipedia role.
One lesson of the past is that proposed wordings need to be considered in light of what they appear to say, not what the authors many years ago thought everyone would accept they were intended to achieve. While we're not a bureaucracy, it's best that the intended scope and aim of core dispute resolution policies are not easily misinterpreted by our core editorial community (some of whom have been very divided and polarized on some issues in the past and a few prone to wikilawyer with the best of them).
Thanks again to you and the rest of the Committee, for the edits made, which solve the big ones. FT2 (Talk | email) 16:44, 8 May 2011 (UTC)

Trivial concerns

I must say, the document is looking rather swank after all of this nit-picking. Just a few comments on the surface language:

  • "Preserve in appropriate confidence the contents of private correspondence sent to the Committee and the Committee's internal discussions and deliberations."—con ... con ... corr ... Comm ... Comm. And there could be a leaking-over on first reading: "... correspondence sent to the Committee's internal discussions ...". Does this still express the intended meaning? "Preserve in appropriate confidence the Committee's internal discussions and deliberations, and the contents of private correspondence sent to the Committee."? Unsure about "appropriate" ... I guess there needs to be leeway, but what is appropriate is unknown.
    • Sure. But we have the choice here of spelling out a lengthy but probably not exhaustive list of possibilities or leaving appropriateness to be determined on a case by case basis.  Roger Davies talk 03:28, 16 May 2011 (UTC)
  • Suggestion for shorter wording: "An arbitrator may recuse from any case, or from any aspect of a case, with or without explanation. Arbitrators are expected to recuse from any case, or any aspect of a case, in which they have a significant conflict of interest. Typically, this includes ..." -> "An arbitrator may recuse from any case, or from any aspect of a case, with or without explanation, and is expected to do so where they have a significant conflict of interest. Typically, a conflict of interest includes ..." It would be much neater, but only you can tell whether it's watertight now. The existing text doesn't actually clarify whether they need to provide an explanation in significant instances of CoI: I'm guessing not, so it's exactly the same deal for CoI and any other reason: up to the arb whether to specify it or not. I think CoI needs to be spelt out in place of "this".
    • I've tweaked this to take these points into account.  Roger Davies talk 03:28, 16 May 2011 (UTC)
    • Further slight tweak: "they have" to "he or she has" to avoid awkward verb change from singular to plural in mid-sentence.  Roger Davies talk 03:06, 17 May 2011 (UTC)
  • "... should recuse shall first post a message on the arbitrator's talk page asking the arbitrator to recuse and giving reasons"—"shall" sticks out a bit; "will" is used in every other instance on the page. Could there be a comma after "recuse"?
  • "Committee deliberations are often held privately though the Committee will make public detailed rationales for decisions related to cases unless the matter is unsuitable for public discussion for ..."—Could there be a comma after "privately"; it's a long enough sentence to need it, I think.
  • "The Committee treats all communications sent to it – or sent by a Committee member in the performance of their duties – as private." Possibly simpler to remove the dashes? "The Committee treats as private all communications sent to it, or sent by a Committee member in the performance of their duties."
    • Knocked out the dashes, and on reflection recast per above with the comma,  Roger Davies talk 03:28, 16 May 2011 (UTC)
  • "The Committee may create or modify its procedures, provided they are consistent with its scope; and form subcommittees or designate individuals for particular tasks or roles."—Perhaps "The Committee may create or modify its procedures, provided they are consistent with its scope, and may form subcommittees or designate individuals for particular tasks or roles."
  • Yep.  Roger Davies talk 03:28, 16 May 2011 (UTC)
  • Since FT2's suggested "primarily" at the top ("To act as a final binding decision-maker primarily for serious conduct disputes the community has been unable to resolve;"), there's just slight dissonance with the first statement in "Jurisdiction" ("The Committee has jurisdiction over user conduct on the English Wikipedia."). Just something to consider.
    • Good catch. Probably the easiest is to replace "over user conduct on the English Wikipedia" with "within the English Wikipedia" as the types of issue are, as you say, already defined above and also "within" works better with "outside" in the next clause.  Roger Davies talk 03:28, 16 May 2011 (UTC)
  • "conduct taking place outside the English Wikipedia"
  • Temporary injunctions: so they can't be issued before the acceptance of a case (e.g., to give the Committee flexibility in encouraging community efforts to resolve a matter, thus forestalling the need to accept a case)?
    • No, that was just sloppily worded. I've changed it to read: "any time between the request for a case being made and the closure of the case",  Roger Davies talk 03:28, 16 May 2011 (UTC)
  • "or create procedures through which policy and guidelines may be enforced"?
  • Is "Additionally" necessary?

Thanks. Tony (talk) 08:53, 8 May 2011 (UTC)

Thanks for the close read: my comments are inline above. That was very helpful,  Roger Davies talk 03:28, 16 May 2011 (UTC)

Why do we need an arbitration committee?

Isn`t it just a waste of time? --Müdigkeit (talk) 20:04, 9 May 2011 (UTC)

I suggest replacing the current policy text with "The english Wikipedia does not have an arbitration committee anymore".--Müdigkeit (talk) 20:22, 9 May 2011 (UTC)
That would be somewhat untrue, however appealing the concept sounds... But to avoid endless arguments, we need some explicit process for reaching definite final decisions on matters that require final decisions, and a democractially elected committee with a defined procedure for adopting binding motions seems as good a solution as anyone is likely to come up with. (Though admittedly some language Wikipedias and other projects manage to get by without.)--Kotniski (talk) 08:34, 10 May 2011 (UTC)
Ok. Your arguments seem to be good...the community is probalby too large for having no such instance.--Müdigkeit (talk) 08:48, 10 May 2011 (UTC)

When. Will. This. Be. Pushed. Forward?

The delay in this is becoming rather funny now. I seem to remember being told a few weeks ago that the "final push" on putting this to the community was a couple of days away. We're now halfway through May 2011; this draft was created in April 2009. Seriously - why the delay? AGK [•] 12:01, 17 May 2011 (UTC)

We've taken note of the last round of comments and the draft is currently being adopted by the committee. The voting has been open for a couple of days but I expect it will take a couple more before everyone has a chance to go through it and the associated diffs. The present delay is mostly my fault - I've been unexpectedly very tied with real life stuff for the last couple of weeks - hence the time drift. Anyhow, apologies all round and we'll hopefully move on the final stage real soon now.  Roger Davies talk 20:41, 19 May 2011 (UTC)

Temporary injunctions

I noticed today as I was officially passing a temporary injunction that much of the old language about temporary injunctions (that they take effect immediately after a majority has voted to support, or 24 hours after net four has) has been taken out of the draft policy. Was that for a specific reason? NW (Talk) 15:49, 22 May 2011 (UTC)

Yes, it reflects the way the updated policy has been organised, with broad stroke principles going into policy and the detail going into the Procedures pages. I see this is missing on the /Procedures (good catch!) so I've added it there now. The language will probably need updating slightly after the updated policy is ratified.  Roger Davies talk 22:50, 22 May 2011 (UTC)
Sounds good. NW (Talk) 22:51, 22 May 2011 (UTC)
Retrieved from "https://mdwiki.org/wiki/Wikipedia_talk:Arbitration/Policy/Update_and_ratification/Archive_3"