Wikipedia talk:Arbitration Committee

Mailing list

The following was un-archived by NE Ent. --Tryptofish (talk) 23:00, 28 February 2013 (UTC)

Back at Wikipedia talk:Arbitration Committee/Archive 13#Thinking about the mailing list, I started a discussion about mailing list policy, and I'd like to re-start it. At the time, some of the Arbs wanted to wait until January, more or less, and I think we are clearly there (also, from more than a week ago, User talk:Risker#Checking if the time is OK and User talk:Tryptofish#Mailing lists). Really, I hope this doesn't come across as pushing anyone before they are ready. But I think the nature of institutions like this is that it can be helpful to make sure that things don't slip through the cracks, and that's the spirit in which I bring it up again. No pressure, just trying to help!

Here's what I said then:

... I'm not proposing anything that is fully thought-out here, but I hope to start some useful discussion. Perhaps we could revise policy regarding the mailing list so that all material on the list would, in effect, be in one or the other of two categories:

• Category A would contain:
  • anything that would identify or otherwise violate the meta:Privacy of users (people contacting the list, parties to cases, arbitrators themselves).
  • sensitive material, such as stalking situations, users who are minors, etc.
  • What else? We should discuss what else should be included here.
• Category B would contain:
  • everything else.

Category A would continue to be treated as it is now, per WP:AP#Transparency and confidentiality: fully confidential, full stop, violations regarded as very serious problems.

Category B would be treated in a new way. The "default" assumption – the way everything on the list would be thought of, until explicitly determined to be otherwise – would, again, be full confidentiality. But Category B material could be released from full confidentiality by a simple majority vote of active, non-recused arbitrators. Such a vote would have to determine: (1) that there is nothing in Category A (or anything Category A is to be redacted before release), and (2) that there is a valid reason to release the material, either on-Wiki or through a more restricted release to persons not on the mailing list. Anyone could make a request to the Committee for such a release, or any arbitrator could propose it. The majority vote would mean that no individual arbitrator would be allowed to release anything without majority consent, and also that no individual could veto majority consent. (I'm not sure whether the vote itself would need to be recorded on-Wiki. It might be enough to conduct it on the list itself.)

I think recent experience has shown that Category B might include material that most members of the Committee have found to be unhelpful, as well as communications that are more of a gossipy nature, as well as things that are directly related to the Committee's business, but that do not reveal confidential material under the privacy policy. I don't see any good reason for the majority to vote to routinely release wholesale slabs of material, but my hope is that this would give the Committee some flexibility to deal with situations where there may be some value in making something more transparent to the community. --Tryptofish (talk) 21:22, 18 December 2012 (UTC)

The response back then seemed to be positive, so let's look into this further. Thanks. --Tryptofish (talk) 23:39, 11 February 2013 (UTC)

• Support per proposer. NE Ent 23:59, 11 February 2013 (UTC)
• Comment: Thanks Tryptofish for raising the issue of mailing lists again. I think we can now honestly say that there's no such thing as a good time to raise these sorts of issues; with three open cases and a few other serious tidbits to review, we have a pretty full plate. Having said that, after your prod on my talk page, I have initiated a discussion amongst arbitrators to examine the mailing list "alternatives" that we saw in early November 2012 that the WMF is able to support. Many of the issues that we have with incoming mail are stymied by the use of the inflexible Mailman software (for example, we can't delete anything from archives without adversely affecting the posts that remain), and we have too many mailing lists for everyone to keep track of effectively. I won't say that response to my discussion has been enthusiastic, but I'm not giving up hope!

  I think it might be helpful to try to figure out some examples of information that might be considered "suitable for release", both from the arbitrator side and from the community side. I think you might be surprised at how much actually shows up onwiki that is simply a reiteration of the positions of individual arbitrators; the most recent example would have been the motion on withdrawing cases, where there was pretty solid correlation between what people said onwiki and what they said on the mailing list, and indeed some were more expansive here onwiki. Risker (talk) 00:41, 12 February 2013 (UTC)

  Thanks Risker, and I'm certainly very sympathetic to how much you and the others have to deal with. Two suggestions: (1) what I'm talking about here is really unrelated to the technology of how the list works, instead focusing on how the humans deal with the material, so I recommend looking at it that way; (2) maybe the best way to figure out what is suitable for release is to start by determining the opposite: what must be kept confidential, because what does not need to be kept confidential becomes suitable for consideration for release. --Tryptofish (talk) 01:30, 12 February 2013 (UTC)

  Once my colleagues with more knowledge of this issue have had the opportunity to weigh in, I might offer some comments; but for now, I'll keep out of this discussion. Nevertheless, I'd like to thank Tryptofish for how thoughtfully and generously he has framed these issues, and to assure him that I do not consider him to be badgering us in any way. AGK [•] 02:14, 12 February 2013 (UTC)

  Thanks, AGK, that's very nice of you. --Tryptofish (talk) 21:59, 12 February 2013 (UTC)

I've been thinking about something along these lines myself. I'd say that Tryptofish's proposal is heading in roughly the right direction, but without an explicit, compulsory separation between ArbCom business and ArbCom chatting that is applied to every message, the Committee is going to get itself into trouble again when its members use a 'business' list for 'personal' activities.

I would be inclined to propose two lists, along the lines of

1. Arbcom-en-L The current list, to be used exclusively for material directly related to carrying out ArbCom's responsibilities. Primarily for material related to ongoing case deliberations. -en-L would continue to enjoy its completely privileged status; its messages and contents cannot be released or discussed outside the list (or the other areas similarly sealed to ArbCom) without the clearly expressed, explicit, obtained-in-advance permission of message authors—or as released by the mechanism described below. Arbitrators are expected not to 'game' the protections of this list by, for example, appending significant 'non-business' content on to 'core business' messages. Arbitrators who repeatedly or egregiously abuse the protected, privileged list should be sanctioned: public censure, temporary loss of list privileges, or even removal from the ArbCom.
2. Arbcom-chat A mailing list to which all of the Arbitrators are subscribed, to be used to broadcast messages to the entire Committee that aren't directly related to carrying out the current business of ArbCom. Message content on this list isn't restricted. Arbs are welcome to use it as they see fit, though they are expected to confine official business to Arbcom-en-L as much as possible. This is the list where Arbs can carry out general conversations, discuss changes to policies that aren't directly under ArbCom's control, engage in electioneering, and talk about who is going to bring the potato salad to the ArbCom Picnic. Messages sent to Arbcom-chat are treated exactly the same way as messages sent between Wikipedia editors using the 'Email this user' function, or sent to a fellow editor's private email address. Arbitrators who repeatedly post private, case-related matter to Arbcom-chat should be sanctioned: public censure, temporary loss of list privileges, or even removal from the ArbCom.

What are the practical ramifications?

• Arbitrators who are using the existing Arbcom-en-L list appropriately and solely for its intended purpose now will face essentially no changes.
• It should go without saying that Wikipedia's privacy policy continues to apply to material on both lists—and to private material whenever and wherever Arbs may handle it on- or off-wiki.
• Material on the Arbcom-chat list enjoys the same protections – and limits to those protections – as regular email between editors. Under the current policy, if Coren mentions on Arbcom-en-L that he will be bringing the potato salad to the ArbCom picnic, and then Newyorkbrad mentions Coren's potato salad on-wiki, it is a technical violation of the list rules for which Newyorkbrad could face sanctions. Under my proposed policy, the picnic and its potato salad would be discussed on Arbcom-chat, and Newyorkbrad could mention it on-wiki without precipitating a crisis.
• Arbitrators who repeat or describe, in whole or in part, content from Arbcom-chat to individuals outside of ArbCom remain fully and personally responsible for the content of what they post, as they would for information (privacy-policy-governed or not) or messages they receive by email. In other words, they are expected to use discretion and good judgement, and not do things like repost or forward full messages without a damn good reason. As part of that discretion and good judgement, Arbs are expected to recognize when their fellows have used the wrong list, and treat the occasional bit of ArbCom business on Arbcom-chat accordingly.

Finally, we have the situations where an Arbitrator has used the Arbcom-en-L list inappropriately, for one or more messages which ought to be on Arbcom-chat. Messages, or portions thereof, can be copied from Arbcom-en-L to Arbcom-chat under the following circumstances (numbers and timing are suggestions, subject to tweaking):

1. With the clearly expressed, explicit permission of the message's author(s);
2. Upon receiving the support of a majority of sitting, active Arbitrators; or
3. If the Arbitrator requesting the transfer identifies his request as 'urgent', with the support of at least two-thirds of the Arbitrators voting within 48 hours (or such longer time as the Arb may request) of the request for a vote.

In all but the most time-sensitive situations, this provides a mechanism by which abuse of the Arbcom-en-L can be moved somewhere where it can be mentioned publicly if necessary. (For example, this would have permitted members of ArbCom to publicly discuss, or at least acknowledge, Jclemens' attempt to use the privacy of Arbcom-en-L to manipulate the 2012 ArbCom elections.) Arbitrators are encouraged to use method #1 informally wherever appropriate and possible.

It's still rough around the edges, but I think establishing a clear demarcation between ArbCom business and Arbitrator chatting (and the appropriate level of secrecy or privilege for each) requires that material be confined to separate lists. TenOfAllTrades(talk) 03:33, 12 February 2013 (UTC)

I get where you're going. A couple of issues: So if I post to the list saying that I will be on very reduced activity because I'm going on an exotic vacation, is that business or chat? [Hint: arbitrator absence is business, personal information like holiday plans is confidential, and it will inevitably evolve into chat when a group of people who are friendly with each other have the chance to say "pics or it didn't happen", for example.] Secondly....oh god not another mailing list. Please not another mailing list. Especially not another Mailman mailing list. Risker (talk) 04:14, 12 February 2013 (UTC)

Appending: It is not physically possible to "move" messages from one mailing list to another. We are physically unable to delete "wrong list" messages from the archives without irreparably damaging other posts. At best, one can redirect the *response* to a "wrong list" email to the right list (which we do if an unblock request comes to the main mailing list), but one has to be very very careful to strip off previous messages that are in the right place, which is darn hard to do from one's smartphone. And if anyone thinks that ANY arbcom list is suitable for electioneering, then I think it's time to pack up the committee and go home. That's possibly the saddest thing I've read on this page, ever. Risker (talk) 04:21, 12 February 2013 (UTC)

Obviously 'move' is a figure of speech, rather than a literal, technical expectation. To 'move' a message to the other list simply means to 'copy without violating policy'. I'm sure that that interpretation is understood. TenOfAllTrades(talk) 04:46, 12 February 2013 (UTC)

(ec) Perhaps -chat isn't the best name for the list; perhaps the two lists should be Arbcom-private and Arbcom-mail or something of that sort. Go ahead and hash out a reasonable division. Maybe declarations of absence should be cross-posted. (Sure, arbitrator absence is business, but it is business that should be shared with the community; the mere fact that you're unavailable doesn't trigger the need for secrecy.) Maybe warnings of an upcoming absence should be kept private on en-L at the discretion of the Arb involved, pending their chosen time to announce it.

I'm sure you noticed I explicitly acknowledge that the 'non-private' list still maintains the usual protections afforded by the privacy policy—your fellow Arbs would be in trouble if they told the community where and when to go to rob your house while you were away, regardless of where you posted your messages. Your fellow Arbs should also be able to demonstrate the responsibility and discretion to confine their "pics or it didn't happen" requests to a 'chat' list. Honestly, this new level of care may seem onerous at first, but I expect that you'll quickly get used to it. Further, there's no obligation to participate in, mail to, or even subscribe to the 'chat' list. If important business is appearing only on the -chat list, then there's something wrong.

In all seriousness, if you don't want to get stuck with another mailing list, make an alternative suggestion. When Jclemens abused your current list (in an attempt) to manipulate an important public Wikipoedia process – nominations for the ArbCom election – it took weeks for the ArbCom to come up with any sort of public statement, that came too late to be much use (inasmuch as it was released well after the close of the nomination period), and even then it appears that the ArbCom's principle concern was the leak and not the abuse of their private list. (In techincal terms, this is what we call 'not getting it'.) You're facing these proposals now because it's clear that having an absolutely-inviolably-secret mailing list is too much responsibility. TenOfAllTrades(talk) 04:46, 12 February 2013 (UTC)

Right now we have three all-Arb mailing lists, plus the functionaries list to which all arbs are subscribed, plus the appeals mailing list to which almost all arbs are subscribed. Some arbs are also on the AUSC, oversight, and global checkuser list on top of that. What you're proposing is sending messages to a mailing list that...nobody even subscribes to? Nobody can ever make a joke with their colleagues in this brave new world, without risk of being publicly censured? Would you also ban introductory emails to the list? I am relatively certain that if we had done the "right thing" about Jclemens' inappropriate use of the mailing list (and there are multiple interpretations of what that "right thing" would have been, ranging from telling him not to be a jerk to banning him from the project, both of which were suggested by respected members of the community, with several points in between), a significant and equally loud portion of the community would have been just as ticked off at us for making a "political" decision about that, too. We "get" a lot more than we're given credit for. A separate mailing list isn't the solution, and wouldn't have been a solution even if it had been in existence at the time. It seems to me that the problem you're "solving" is the development of any sense of internal community within the committee, rather than how to deal with arbitrators who abuse their position wherever they may do so. Risker (talk) 05:39, 12 February 2013 (UTC)

I understand your frustration, but I trust that you also understand that there's as much frustration outside the Committee. Based on Jclemens' comment in Wikipedia talk:Arbitration Committee/Noticeboard/Archive 19#Where were the rest of the Arbitrators during all this, and what do they see as their responsibility to the community?, the response of the Committee to his use of the mailing list for his own electioneering – attempting to intimidate potential competitors in the then-upcoming election – was a stern finger-wagging from "a few" other Arbs, followed by utter silence. The ArbCom (according to Jclemens) considered the matter "a resolved issue". Things apparently rested that way for five-plus days, until the ArbCom became aware of the leak by Elen—even then it took two more weeks for the ArbCom to release any statement. A sitting Arbitrator had made a credible attempt to use his privileges to manipulate an election for his own benefit, but the ArbCom was content to keep the community entirely in the dark on the matter.

As someone without access to ArbCom mailing lists, I don't know if this type of abuse is a regular occurrence or not. I doubt that it is, and I fervently hope that it isn't. But right now, the ArbCom is batting 0 for 1 in the sole case that the Wikipedia community knows about. The only thing that the community can see is that the ArbCom didn't take any visible action until weeks after they were aware the matter had been leaked.

You've said that if the ArbCom had handled the matter differently, there would have been a different pool of editors crying for your blood, and I certainly am not naive enough to dispute that. (Sometimes you really are damned if you do and damned if you don't. Being an Arbitrator can really suck, and I wouldn't stand up and volunteer for that misery even in the unlikely event that the community offered it to me.) The unfortunate impression that those of us on the 'outside' get, however, is that the ArbCom's initial choice in the matter wasn't between pissing off Group A and pissing off Group B. Instead, it was between pissing off Group A, and not doing much of anything while relying on the secrecy of the mailing list to protect everyone on the ArbCom from any fallout. To be clear, I doubt that anyone on the Committee was consciously engaging in that sort of deliberate political calculation; rather, I suspect that the ArbCom just plain failed to recognize that they needed to make any decision at all. We are told that "a few" Arbitrators took it on themselves to criticize Jclemens, but did any Arbitrator take any positive step to discuss or direct an appropriate response by the ArbCom as a whole? Forget bringing anything to a vote, did anyone even sketch a motion, or suggest in any more than the vaguest way that ArbCom might be obliged to make an announcement or take other publicly-visible, collective action? TenOfAllTrades(talk) 15:33, 13 February 2013 (UTC)

• I encourage the committee to do as much discussion and deliberation onwiki as feasible per the study referenced in the SignPost last October: ""the reported use of interaction channels outside the Wikipedia platform (e.g., e-mail) is a cause for concern." That said, arbcom-l must exist to handle information not suitable for public dissemination.
• With regard to the list itself: too much overthinking. Not supposed to a bureaucracy here. Existing policy, to wit,

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. The Committee treats as private all communications sent to it, or sent by a Committee member in the performance of their duties. is fine. What's lacking is simply a protocol to determine whether a arbitrator sent communication is "in the performance of their duties." Logically, only the committee can decide that because the rest of us shouldn't see it until a determination is made. So simply add a protocol that if one of ya'll think an email is outside the bounds of "performance of their duties" a majority vote can determine it can be revealed.

• I do not care about, and would actually prefer not to know, details of the "potato salad" nature. TMI. NE Ent 14:09, 12 February 2013 (UTC)

• As I compare and contrast the approaches in my suggestion and in TenOfAllTrades', a lot of it comes down to technology versus human judgment, and that's actually something that also arises in Risker's first reply to me. I kind of think that anything that relies on technology, whether that means two separate lists as TenOfAllTrades proposed, or various technological upgrades to the list system, as have come up at various times before, won't really get at the key issue. What matters is letting a consensus of Arbitrators determine whether something on the list should be made public or be kept confidential. I trust collective judgment about that, in a way that I don't trust the judgment of any single individual. If you look back at the archived discussion, you'll see an exchange between me and SandyGeorgia, in which we both agreed that: (1) a single Arb should not decide unilaterally to release something, and (2) a single Arb should not be able to unilaterally prevent the majority of the Committee from releasing something. If there's a decision about which of two lists to post to, that decision will always be made by a single individual. I'd rather have something like a majority vote of Arbs, to decide whether to lift confidentiality for any particular message. --Tryptofish (talk) 22:11, 12 February 2013 (UTC)

Timestamp. --Tryptofish (talk) 22:16, 19 February 2013 (UTC)

I think this might have run its course. There are a number of cases etc taking up the attention. I think the solution is much simpler.

1. There is a list.
   
2. No-one should send personally identifying information to it, because any mechanism that fires messages into the private mailboxes of individuals is by definition not secure. Send it to an Arb you trust, and tell them not to share it, only to summarise it
   
3. The list should only be used for official business. Any use for other business should not expect the level of privacy afforded to official business, and should be showered with sardines.
   
4. If someone announces that they are unavailable due to surgery/visit to Venice etc, and other Arbs want to know more information, that's what personal email is for.
   
5. my personal view is that it's bloody ludicrous that Arbcom should expect to keep its own deliberations secret after the fact, and I don't know how it ever got from the base premises of the Wikipedia founders to that position. But that's my personal view, and not relevant here. Elen of the Roads (talk) 22:57, 19 February 2013 (UTC)

Hi, Elen – as always, it's good to hear from you. I certainly didn't expect that to be the reaction to my timestamp. In my opinion, what I've proposed has not run its course. --Tryptofish (talk) 23:17, 19 February 2013 (UTC)

Although we don't run a justice system per nojustice, we can choose to be informed by concepts taken from justice system. In the context of the US Constitution 6th Admendment, a Connecticut public defender notes:

"The right to an open court in criminal proceedings is “an effective restraint on possible abuse of judicial power,” In re Oliver, 333 U.S. 257, 270 (1948), which functions for “the benefit of the accused; that the public may see he is fairly dealt with and not unjustly condemned.” Waller v. Georgia, 467 U.S. 39, 46 (1984).

and Wikipedia specific research has found "ensuring participation and transparency is crucial for maintaining the stability of self-governing communities." NE Ent 23:44, 19 February 2013 (UTC)

This is the end of the content that had been archived. The following has been added new, after the restoration of the thread. --Tryptofish (talk) 23:00, 28 February 2013 (UTC)

Mailing list section break

In a justice system, you get to see the evidence presented against you. You don't get to read the judges' notes during the case or listen in on the conversations they have with their clerks. NW (Talk) 02:38, 28 February 2013 (UTC)

True, but what should be done when one judge is using the tools of his office to secretly manipulate the selection of his peers? TenOfAllTrades(talk) 02:46, 28 February 2013 (UTC)

I'd venture to say whatever it is, it's not "each of the other judges gets to personally select the excerpts they think make that first judge look the worst, and submit them anonymously to one reporter of their choice, under the condition the reporter not write about it" or "all the other judges publish the entirety of the case's notes, discussions, and private evidence, with no eye to which items are relevant to misuse." That is to say, any system of handling questionable mailing list (or just routine "FOA"-type disclosure about what goes on on the mailing list) use needs to be a) concerned with retaining privacy in any issues that require it b) determined by the committee as a whole, not presented as a fait accompli by the first arb to decide a message is "misuse" and release it.

I do happen to agree, however, with those who are urging Arbcom to focus more on onwiki deliberations. Private matters need to be handled privately, but everything else - and I'd venture that that "everything else" is probably a vast majority of the discussions that go on - should be happening onwiki. That includes proposing draft/working motions; fighting with each other over which remedies, FoFs, etc should pass; hashing out the wording of things; etc. There is no benefit to keeping those things socked away on a mailing list just because there's a tradition of posting a mostly-cohesive PD onwiki at a later date instead. The community knows you all disagree a lot; for most disagreements among arbs, it is of benefit to the community to see who thinks what and what is going into the committee's convergence on a given PD/motion/etc.A fluffernutter is a sandwich! (talk) 15:07, 28 February 2013 (UTC)

I don't disagree. Your point about leaking as a respone to misconduct is one that I touched on way back in November—"I hope I don't...see this mailing list dispute turn into Team Jclemens versus Team Elen. In practice, both of them screwed up in different ways, and neither deserved to be excused for their conduct nor held up as a 'victim'." Pretending that either of them handled themselves particularly well, or that either's conduct should be a template for future policy, is obviously ridiculous.

As a matter of the ArbCom's response to the situation, misuse of the mailing list in an attempt to manipulate the candidate pool for a public election was insufficiently serious to prompt a public statement, motion of censure, or any other address to the community. (As far as I can tell from the mostly-stony-silence we've been getting from the ArbCom, nobody even considered making any sort of public statement.) Release of material from the mailing list – material which wasn't related to ArbCom business and contained no personal information – was considered sufficiently serious to prompt a public ArbCom response. I don't know if that's because the ArbCom thinks leaking is more serious than misuse of the ArbCom tools, or if the ArbCom only wants to publicly acknowledge misconduct that's already public knowledge, or if the ArbCom was just plain asleep at the switch and hasn't got a clue how to properly respond to Arbitrator misconduct. So far, though, I'm still waiting to see any Arbitrator offer a suggestion about how to approach any of those possibilities, or offer any alternative explanation for the way they handled things. TenOfAllTrades(talk) 15:57, 28 February 2013 (UTC)

I'm willing to let explanations of what happened in the past go as water under the bridge, but I strongly agree with TenOfAllTrades that it really behooves the Arbitrators to start commenting publicly (here would be ideal) on what they will do in the future. That's really what I'm trying to get to by pushing this discussion.

It seems to me that the kinds of problems discussed just above arise when a single person on the Committee decides to leak something, and the other members are left having to figure out what to do after it happens, or when a single member of the Committee believes that the community would be served by making something public, but cannot. We know that some things on the mailing list need to be kept private, and some other things do not.

I think the way forward does not lie in anything technological, but rather, in human judgment. But the judgment should be collective, and I trust the Committee, as a group, to make those judgments (perhaps other editors disagree). What I want to avoid is: (1) one Committee member decides unilaterally to release something, and (2) one Committee member can veto the release of something that the Committee as a whole believes ought to be made public. My pre-proposal in the yellowish box near the top seeks to avoid both (1) and (2). --Tryptofish (talk) 23:12, 28 February 2013 (UTC)

• I hope that the members of the Committee will comment here on ways to move forward, because I think that it would be regrettable if you simply wait until the next crisis. --Tryptofish (talk) 22:41, 4 March 2013 (UTC)
• I realize that the Arbitrators have a lot of people talking at you, and consequently find it hard to attend to this discussion thread. But I'll keep on timestamping here until we get some progress. --Tryptofish (talk) 21:31, 8 March 2013 (UTC)
• Another timestamp, another ping. --Tryptofish (talk) 19:13, 14 March 2013 (UTC)
  • Hey Trypto, could you perhaps spell out what steps you're looking for? Insofar as current procedures, I don't necessarily see any obstacle to arbs releasing email threads with the consent of all parties (including whoever else is on the thread, assuming it's not just intra-arb stuff.) Doing it by a majority gets a bit more grey, since that's essentially the email participants relinquishing their rights to privileged correspondence sight unseen, and would definitely require a motion, I suppose. That really can't be asked of non-arbs, so I guess your proposal would have to be restricted to solely arb discussion.

The other issue I'm seeing with the proposal is the line that "anyone" could make such a request. I'm not sure how we could honor those requests, however. Editors don't have citation information for an email, after all (or I suppose if they did, that would mean they were either a participant in the email, and thus the above proposal is irrelevant in that we could release it by agreement as mentioned above, or else it was leaked to them and thus a violation of confidentiality already), and "I want to know everything that happened on December 20-January 15" just enables fishing expeditions or wastes a lot of time trying to figure out if it's acceptable to publish, what needs to be redacted, what has or hasn't been redacted (and should be) before it's posted, etc.) If you have any suggestions on how that would work, I'd love to hear them.

(TL,DR version: it seems like you're suggesting essentially a Freedom of Information Act-type setup, I'm concerned about some pitfalls that may not be initially forseen.) Der Wohltemperierte Fuchs^(talk) 20:07, 14 March 2013 (UTC)

The TL,DR version of my reply is that the next step I want to see is the kind of reply that you just provided! (Per my talk page, Risker is also planning to discuss it.)

My understanding of the ArbCom policy as it exists now is that everything on the mailing list is considered private, so making it possible to release anything on the list will require some revision of policy. Therefore, such a revision is what we should ultimately be aiming for. I agree with you that something can be released if everyone involved in the discussion agrees – meaning the entire Committee and everyone else, outside the Committee, who was involved in that discussion. As a practical matter, that is something that happens infrequently.

Although none of us wants to re-litigate the mess that happened just before the last election, it's an example of where existing policy doesn't work, and I hope that this is painfully obvious without going into the gory details. There needs to be a way to agree that something can be made public when it's in the community's interest to do so, even when one member of the Committee is standing in the way. At the same time, there needs to be a way for one or more Committee members who, arguably, want to blow the whistle on something to be able to make it public, but without doing so unilaterally and in potential violation of privacy. I've suggested a majority vote. I see from your comment that you are concerned about what that kind of vote would mean for the minority who dissent. That's a worthwhile question. The Committee is composed of people who have been vetted by community vote; perhaps we can trust a majority of them to do the right thing in such instances? Or perhaps you can come up with a better method – but the status quo isn't it!

Yes, I can see that a freedom-of-information type process would be a troll's delight. That's an example of how my posting is an attempt to generate a discussion, as opposed to being a completed proposal. But there's a TL,DR reply to the question of how the Committee can deal with requests that are just fishing expeditions: say no to them. There should not be anything in any policy change that would obligate the Committee to release anything on demand! People can ask, but they can get a negative reply. --Tryptofish (talk) 20:35, 14 March 2013 (UTC)

Ent version: Arbcom-l (or it's technical successor) is to be used for official arbcom business and the stuff on it is private. If an arb attempts to use it for other purposes, it's not private. Whether a particular post is official biz or not is decided by a majority vote of the committee. NE Ent 20:40, 14 March 2013 (UTC)

I agree fully with that. --Tryptofish (talk) 20:43, 14 March 2013 (UTC)

With the additional observation that, when the majority decides that something is not official biz, they don't have to release it publicly, but they should be able to do so when there is community interest in doing so. --Tryptofish (talk) 20:46, 14 March 2013 (UTC)

Yes. Not private = releasable upon an arb's judgement, not "must be released." I'd greatly prefer the committee not release the who's bringing the potato salad stuff. NE Ent 20:50, 14 March 2013 (UTC)

That's fine by me, I hate mayo. I wasn't particularly concerned for the minority of theoretical arbs in this release scenario, assuming that the policy has been changed, but more how the actual process would work. I can see it working fine from the arb POV (in that someone wants to bring the discussion to a wider fora, so we all vote to publish a transcript of what's discussed so far), it's that opening it up to community requests (which would in my opinion from list discussions be a more valuable aspect) brings up the aforementioned cans o' worms. I'm not sure "just say no!" works in this instance, because then we're left with a lot of annoyed people wondering why we denied their requests, and we're not in the position to say why beyond very general (and unsatisfying) reasons.

I suppose the ultimate issue I have with the proposal, the more I think about it, is that I'm not sure it's really solving many problems. From my estimation the lists aren't used for gossiping and chatter about who's bringing what to the picnic that much. I feel like the bigger issue that this proposal was obliquely trying to tackle is the matter of making sure deliberations made on-list are appropriately and transparently explained on-wiki. While your proposed policy change could enable arbs to expedite some of that via outright publishing of the emails, I think it requires something a little less legislative in that we (the arbs) should all be more proactive about pushing each other to explain our reasonings beyond the (useful, if dry) list of supports and abstains. Unless I'm completely misinterpreting the impetus of the initial thread, which is always possible. Der Wohltemperierte Fuchs^(talk) 21:08, 14 March 2013 (UTC)

It's about JClemens attempting to run an anti-Malleus campaign on arbcom-l and Elen getting her hand caught in the cookie jar trying to figure what the heck to do about it. It shouldn't be something that comes very often, if ever again. While I've commented elsewhere about the desire for AC to be more transparent in certain circumstances, I don't think overly complicated procedures / protocols for the discussion list is the way to go. I want ya to be able to do you jobs efficiently, which includes being able to speak frankly without wondering "oh will this end up be posted onwiki someday?" We just don't want a member to be to abuse the privilege cause there's no safety valve in the system. NE Ent 21:22, 14 March 2013 (UTC)

Fair enough. While I've been responding to Trypto's comments above, are there any particular actions you think the Committee should take beyond stridently calling out the behavior you described? Does Trypto's proposal count as your safety valve or are you thinking of something else? Der Wohltemperierte Fuchs^(talk) 21:27, 14 March 2013 (UTC)

(edit conflict) For me, from outside the Committee, I very much disagree with the suggestion that it doesn't solve any problems. I'm pretty sure that you will find that a lot of us on the outside came away from what happened last fall believing that there is a problem that ought to lead to policy improvements. Working harder to provide better explanations – sure, no one would say that that's a bad thing! But we're talking about human beings, and it's easy to say that, but difficult to make it happen. I'm uninterested in the potato salad, too, so that's not the issue. And it really isn't simply a matter of greater transparency about how decisions were arrived at. After all, that does belong on proposed decision pages. It's more about clearing things up after something discussed on the list appears to have gone wrong. On the other hand, you've offered your own solution to those people who are annoyed by having their requests turned down: give them better-written explanations.

But maybe it would be a mistake to say that anyone can request. Maybe it should be limited to people who took part in the communications and members of the Committee (the latter, of course, potentially responding to community interest). What would be the arguments for and against that? --Tryptofish (talk) 21:32, 14 March 2013 (UTC)

(edit conflict) Tryptofish's proposal would suffice, although I think it goes farther than necessary. Don't see a need to support external requests for release; unless there was a leak they'd essentially be fishing expeditions and the committee is already tasked with many responsibilities; don't see any need to add more. NE Ent 21:37, 14 March 2013 (UTC)

Well thanks for your thoughts. I'm going to be busy this weekend with drafting but adapting what you've got into something the committee can vote on will be on my plate for the near future. Der Wohltemperierte Fuchs^(talk) 02:50, 15 March 2013 (UTC)

Thanks. I'll be looking forward to it. (And no need for potato salad on that plate.) --Tryptofish (talk) 23:07, 15 March 2013 (UTC)

I have to make another observation. Potato salad notwithstanding, is there something in the water? I just read Coren's resignation, coming just after Hersfold's. I get the distinct feeling that the Committee needs some more transparency, for its own good. --Tryptofish (talk) 23:45, 15 March 2013 (UTC)

(Disclaimer: Strong criticism ahead. Those with no interest in wikipolitics may not wish to read ahead, but I think this needs to be said.) My colleagues may beg to differ, but I consider Coren's resignation to be the rankest drama-mongering. What you saw when he resigned was an editor, whose judgement was otherwise satisfactory, throwing the toys from the pram because the community has moved on and the arbitrators who now staff the committee were not quite willing to go along with his extreme views on running ArbCom. I wouldn't pay it too much attention. As for Hersfold's resignation, I think it should be clear that his primary reason for resigning has very little to do with the committee in general, a little to do with a vocal minority of the community who will remain unbranded, a little to do with the people at WO, and a lot to do with the fact that he simply doesn't want to spend his nights doing what it is arbitrators do for the project. OMMV. AGK [•] 00:12, 16 March 2013 (UTC)

I'm not surprised by the language in Coren's statement, seen it before; but I am surprised he's actually walked away this time. Elen of the Roads (talk) 02:24, 18 March 2013 (UTC)

Considering how he was dragged through the mud on Jimbo's talk page...I'm not.--Amadscientist (talk) 03:39, 18 March 2013 (UTC)

Wikipedia:Arbs are people too

I have written the above essay as a reminder to all concerned in ArbCom disputes that arbitrators are people too. Perhaps this should be considered more often. AutomaticStrikeout (T • C) 23:50, 14 March 2013 (UTC)

Thank you very much. I for one appreciate the sentiment. Regards, AGK [•] 00:13, 16 March 2013 (UTC)

	The Kobayashi Maru Award
	Awarded jointly to all members of the 2013 arbcom, for earnest, diligent and responsible efforts to find the best possible solution to a no win situation. --Demiurge1000 (talk) 19:11, 17 March 2013 (UTC)

I have to hope that conflict resolution is not a "No win scenario" on Wikipedia, but good call Demiurge1000, AutomaticStrikeout and AGK!--Amadscientist (talk) 03:34, 18 March 2013 (UTC)

I will endorse that so far 2013 has been a vast improvement on 2012. I still have major reservations about this year's case actions, mainly sanctioning users on historical problems, rather than being forward looking. However while negative outcomes continue to be the norm, at least they are not in the extreme vein of many (indeed, possibly all) 2012 outcomes, and it is good to see positive steps to re-integrate valuable editors into the community, and to remove some of the really old and outdated remedies from years ago. Rich Farmbrough, 03:13, 20 March 2013 (UTC).

OK I have reviewed some of the 2013 actions I wasn't aware of. I can now say that we are closer than I thought to the old model where the community would be best served by implementing the direct opposite of what ArbCom first decides. Rich Farmbrough, 08:01, 20 March 2013 (UTC).

I am actually not sure what 'Arbs are people too' actually means. Sure, Arbs are people, but 'Wikipedia:Admins are people too'; 'Wikipedia:Bot-owners are people too'; 'Wikipedia:Spammers are people too' (at least, most of the time); 'Wikipedia:Vandals are people too' ... maybe that is also something that should be considered more often. --Dirk Beetstra ^{T C} 11:19, 20 March 2013 (UTC)

"It's easier to be an asshole to words than people" Worm^TT(talk) 11:45, 20 March 2013 (UTC)

Exactly, WTT. --Dirk Beetstra ^{T C} 11:54, 20 March 2013 (UTC)

• I have been and remain heavily critical of ArbCom as a body (as opposed to its members). I grant they are people too. So is everyone else on this project; something that is lost on many people, not just the members of ArbCom and the people who are critical of ArbCom members. Where people tread on criticizing individual members of ArbCom, them being people too applies. Where such criticism is towards ArbCom as a body, it does not. ArbCom as a body has for years now been woefully inadequate to the tasks laid before it. The structure that is in place to deal with final conflict resolution on this project is in abstract, in my opinion, a serious detriment to the furtherance of the project. ArbCom has been brought to the well many times over the last many years to fix the inherent problems in their structure and conduct. To date, ArbCom has refused to drink on almost every occasion. A serious, deliberative, and professional overhaul of the final dispute resolution steps on this project is extremely overdue. As is, this latest series of dust ups will be swept under the rug, and nothing will be done to learn from it, to improve the process. These problems will continue to arise. --Hammersoft (talk) 14:19, 20 March 2013 (UTC)