Wikipedia talk:Arbitration/Policy/Update and ratification/Archive 1

Last resort
"To act as a decision-maker of last resort for Wikipedia disputes," - this needs to make clear that it is disputes about user behaviour and NOT content. The committee is not the final arbiter of content.--Scott Mac (Doc) 01:06, 9 April 2009 (UTC)
 * We do assist with disputes about policy and content by assisting the Community in finding ways to settle the dispute and on some occasions we say that the community decision needs to stick. So, we need to be careful with our wording so to not mislead about what falls in our scope. FloNight&#9829;&#9829;&#9829; 12:53, 10 April 2009 (UTC)
 * Actually it's only by custom that ArbCom don't resolve content issues, since it is technically within their formal remit:
 * "The Committee reserve the right to hear or not hear any dispute, at their discretion" - policy
 * "will primarily investigate interpersonal disputes" - policy
 * (Also note that until very recent years, most Requests for Arbitration were resolvable in principle, by addressing the conduct issue alone. A lot of the 2004-07 cases for example were "please help with this disruptive user" problems.)
 * FT2 (Talk 12:00, 11 April 2009 (UTC)

Normal appointments
I propose changing "The members of the Committee shall be appointed by Jimbo Wales, in his role as project leader, following advisory elections whose format shall be decided by the community" to read "The members of the committee shall be appointed by Jimbo Wales, in his rôle as project leader, in accordance with the results of elections whose format shall be decided by the community". (changes boldes for clarity) DuncanHill (talk) 05:17, 9 April 2009 (UTC)
 * Up to now, Jimbo has used the results of the election to advice him about who to choose for the Committee. Since he has decided to expand the Committee, we've not had elections where people were running for specific seats. I think the original wording reflects what has happened in the past. FloNight&#9829;&#9829;&#9829; 13:01, 10 April 2009 (UTC)
 * It does reflect what has happened in the past, but it should reflect what should happen in the future. I see his rôle as developing into a more constitutional one, rather than an arbitrary one as it is now. DuncanHill (talk) 12:27, 20 April 2009 (UTC)

General Comments
I think this draft looks pretty good: it seems to follow the current practices very well. However, I am unclear exactly what changes are to be made... or is this just a clarification of existing policy? If that is not the case, perhaps a summary of significant changes would be useful to readers who aren't extremely intimate with Arbcom.  bahamut0013  words deeds   05:49, 9 April 2009 (UTC)


 * As you say, the draft follows current practices and tries to cover all current issues following the norms in place. The Arbitration Committee is in fact undergoing some structural adaptations which are set to be leveled with the the needs of a rapidly evolving encyclopedia that anyone can edit. This draft comes to lay down the general foundation of the adapted Arbitration system.


 * The adaptation process follows an agenda. In January 2009, the Committee has set up a noticeboard to keep the community informed of all its activities. The noticeboard serves for announcements of and statements made by the Arbitration Committee. Some of the newly implemented procedures have already been put in place. Starting April 2009, The Committee established a new new ban appeals subcommittee and procedure. Also, the Committee is actually implementing a new Arbitration case management system. From now on, the drafting of new opened cases will be assinged to in individual arbitrators while teamed with one or two more arbitrators to work proactively in cases. In the same area, the organization of Arbitration case pages will be shifted to a more flexible system of organization based on a new process and a newly formatted layout of the Workshop pages where importance is given to the proactivity of the Arbitrator(s), the active involvment of the participants through questions and answers and the empowerement of the Clercks to protect against disruption of the process. A deadline will also be fixed in agreement with the parties at the start of the process; though may be changed depending on the flow of the case. Note that this change is only in its experimental stage and a flexible guideline will be produced after the closure of a few cases. Also note that, individual or subgroup Arbitrators Arbitrators may still opt for they own preferred guidelines. So keep reviewing the ArbCom noticeboard to keep you informed. Thanks. -- FayssalF  - Wiki me up® 01:45, 10 April 2009 (UTC)
 * Ok, that summary helped a great deal. It makes me feel warm and fuzzy to see the Committee willing to be transparent.  bahamut0013  words deeds   13:43, 10 April 2009 (UTC)

Soooo many headings
The number of headings in this document is utterly excessive. Unlike legal documents, which this is somewhat simulating, we don't need to be able to reference each and every individual clause separately. Many of the headings can be combined and reduced; I'll do a diff. Happy‑melon 09:56, 9 April 2009 (UTC) This would be a good start. Happy‑melon 10:06, 9 April 2009 (UTC)


 * We're aware of that, and many of the headings will likely be removed in a subsequent draft. At the moment, they're in place more for ease of commenting than anything else. Kirill [pf] 12:39, 9 April 2009 (UTC)

re; Expedited decisions and motions
I suggest that simple majority is not the benchmark for the "substitution" for a full hearing (although it may suffice where the alternative to motion is a denied request), that a 51/49% split between Arbs is insufficient to satisfy the community that there isn't evidence that needs placing and reviewing in a full case. Perhaps a wording similar to "majority + 2" or "Nett 3 support" gives sufficient split without likelihood of opposes hamstringing this option? LessHeard vanU (talk) 13:11, 9 April 2009 (UTC)
 * How about "majority plus two or consent of those named in the request."? --Barberio (talk) 13:20, 9 April 2009 (UTC)
 * Named in the request or affected by the motion(s) in question? Since people are free to add themselves as parties, do we want to give everyone doing so the ability to affect this? Kirill [pf] 13:26, 9 April 2009 (UTC)
 * Since RfAR's are not brought with the intent of being superceded by motion (at least, I hope not) then the decision is best regarded as an internal one, but not one that should balance on one or two votes - I am looking for something between "overwhelming" and "simple" majority to make clear that a sizable percentage of the ArbCom believes motion/expedited is the best solution. LessHeard vanU (talk) 15:31, 9 April 2009 (UTC)

Things Missing
Here are a few things from the past discussions that I think had consensus which are missing from this draft,


 * 1) Identifying the Arbcom's ability to make and vet appointments to various functions of en.wikipedia
 * 2) Requirement to communicate clearly with the community.
 * 3) Switching phrasing 'Jimbo Wales' to 'Wikipedia Project Leader'.
 * 4) 'Recommend changes to policy' functions.
 * 5) Automatic acceptance of cases referred from the foundation.

There may be other functional differences with the version at Arbitration_policy_proposed_updating that should be reconciled. --Barberio (talk) 13:16, 9 April 2009 (UTC)


 * I think some of those are present, albeit not necessarily in the same form as in earlier drafts:
 * Present as point 4 under "Duties and responsibilities"; or are you looking to put more detail on the appointment process itself into the policy?
 * "Arbitrator conduct" and "Language of decisions" might be relevant here; or does that not cover it explicitly enough?
 * This is a constitutional issue somewhat beyond the arbitration policy; it's not really appropriate to do that unless some provision exists to have a leader other than Jimbo, which I don't believe is the case at the moment.
 * The second clause under "Scope of arbitration" addresses that to an extent. I'm not sure that mere recommendations outside of dispute scenarios need to be covered here; we can recommend policy, sure, but so can any other editor.  Or did you mean something other than the normal proposal process?
 * That's probably a pretty moot point, since the WMF has never referred a case to us, and almost certainly wouldn't do so for legal reasons in any case.
 * Kirill [pf] 13:24, 9 April 2009 (UTC)
 * Generally, more detail on the appointment process would be good. For instance, can the community over-rule/veto an arbcom appointment?
 * Yes, there needs to be more explicit coverage of what behavior is expected, even if that's not strictly enforceable, it establishes a 'spirit of the law' for the rest of the policy.
 * The "recommended policy changes" text from the previous proposal is less a 'power' than it is an explicit restriction of how Arbcom or Arbcom Clerks should behave when editing policies, to prevent even the perception of policy by fiat. --Barberio (talk) 19:58, 9 April 2009 (UTC)
 * edit conflict with Kirill -- some points are duplicated
 * For #1, I think this is found in point #4 here.
 * For #2, is this what is needed? Or do you mean a more general need to communicate clearly with the community?  It would be good, I think, to have such a commitment from the Committee.
 * For #3, I personally think that would be misleading. It is far from clear that, in Jimmy's absence, anyone would fill that role.  There are several references to "in his capacity as project leader"; perhaps this could be cleared up with a section that explicitly defined the relationship between the Committee and Jimmy.
 * I don't think I've seen references to #4 or #5 before. To be honest, I can't imagine a circumstance where the Foundation would refer a case to the Committee (beyond their saying "no, we can't sort this for you; try these guys").  Having the Committee draft and corporately endorse changes to policy would, I think be an excellent idea (I have advocated it privately in the past).
 * [[Sam Korn ]] (smoddy) 13:38, 9 April 2009 (UTC)

I am also unsatisfied with the detail in discussing arbitrator conduct (I'd like some more adjectives: diplomatic, serious, diligent, mature maybe) and overall responsibilities (Block/ban reviews, OS/CU, Administrator conduct)--Tznkai (talk) 13:57, 9 April 2009 (UTC)

Jimbo
Has the comittee considered removing the Jimbo function from its charter? Hipocrite (talk) 13:35, 9 April 2009 (UTC)
 * While I'm somewhat sympathetic to this idea (build for a future without Jimbo, it will come eventually, if only due to natural causes) I think this would have the effect of provoking a constitutional crisis.--Tznkai (talk) 13:44, 9 April 2009 (UTC)
 * Only if the Jimbo figure opposed removing the Jimbo figure from the policy. Is that the case? Hipocrite (talk) 13:46, 9 April 2009 (UTC)
 * Considering he fights to keep titles on smaller projects he never edits, I think it can be assumed that he's not relinquishing the GodKing title without a fight. لenna  vecia  15:19, 9 April 2009 (UTC)
 * Hey, Jimbo's the guy who got this thing up and running, and stopped it going down really bad pathways. He has not allowed advertising. He's a good PR person. These are good reasons for not changing this aspect—not yet, anyway. There are far more urgent aspects to get right before his relationship with ArbCom and the project were ever challenged. Tony   (talk)  15:39, 9 April 2009 (UTC)
 * I don't mind the policy referring to Jimbo, or to him having certain functions within the policy, but I do think that this should only be as formally given to him by the community. He will not be with us forever, and I think it is time that we learnt to stand on our own several million feet. DuncanHill (talk) 12:24, 20 April 2009 (UTC)

Duties and responsibilities
I would like to offer my thanks to whoever wrote this for their efforts. I do have one question. Duties and responsibilities #1 says "To act as a decision-maker of last resort for Wikipedia disputes, to determine which of these disputes are suitable for arbitration, and to carefully review and render binding decisions in those disputes;..." I hope I'm not starting anything here, but it seems to me as if by the phrasing it is being indicated that the ArbCom becomes the final authority short of Jimbo and the Foundation for all matters related to wikipedia. Personally, I would welcome having someone identified as the final authority, and there have been several discussions that basically didn't result in anything regarding the matter of "final authority" being vested somewhere, so I don't mind seeing that, but would like some clarification as to whether that is what it is intended to say. John Carter (talk) 13:38, 9 April 2009 (UTC)
 * Indeed, this is a major undertaking, and arbitrators and clerks who have contributed deserve thanks. Tony   (talk)  15:59, 9 April 2009 (UTC)

Clerks
A narcissistic moment here: but the the clerks are referred to 4 different times, without an explanation once of what it is we do. It might be useful to put clerks within the overall category of "Delegation" which would also cover subcommittees and any other panels or groups ArbCom wants to put together - and mention those things in some detail. Much of this policy is vague and broad, and there are some odd pieces of specification.--Tznkai (talk) 14:03, 9 April 2009 (UTC)
 * The clerks' role does need to be spelled out in this policy, I believe. It's easier to comprehend, and the right thing to do for all parties. Tony   (talk)  15:42, 9 April 2009 (UTC)
 * I just found it odd we were mentioned at all in the main policy.--Tznkai (talk) 20:21, 9 April 2009 (UTC)
 * We could, I suppose, avoid mentioning the clerks specifically and leave them as an unnamed group to which authority has been delegated; but, given the central role you guys play in the case procedures and the firmly established nature of the role, it seems appropriate to mention you explicitly at certain points. Kirill [pf] 00:14, 10 April 2009 (UTC)
 * It might be useful to put clerks within the overall category of "Delegation" which would also cover subcommittees and any other panels or groups ArbCom wants to put together... I agree. -- FayssalF  - Wiki me up® 04:20, 10 April 2009 (UTC)
 * The Clerks' role needs to be explicated in a very public place. At least, certain things about their role. The ArbCom policy can't properly be revised without revising the policy that governs the roles and responsibilities of the clerks. Probably one separate section is require; it doesn't have to be long. Tony   (talk)  16:20, 11 April 2009 (UTC)
 * I agree that a section detailing the role of the clerks would help. But there are other groups that ArbCom (or at least this current ArbCom) co-ordinate with. Off the top of my head, we have the functionaries (former arbs, checkusers and oversighters), various foundation people (mainly Jimmy and Cary Bass), and sometimes checkuser and oversight people on other WMF wikis (though not that often). If the role of the clerks is being explicitly laid out, maybe some of these other relationships could also be made more explicit? Oh, sometimes stewards as well (when emergency desysops take place) and sometimes bureaucrats. That last one is interesting: the relationship between ArbCom and bureaucrats has never really been clearly laid out in my view - maybe because the roles shouldn't really ever interact that much - but it has come up in relation to bureaucrats declining requests for resysopping. The relationship between ArbCom and community processes like RFA also came up recently, and is often discussed in relation to desysopped admins and where resysop applications should be made: ArbCom, bureaucrats or RFA? Carcharoth (talk) 16:48, 12 April 2009 (UTC)
 * The clerks perform a close, continual and essential role in relation to ArbCom. They are different from all of the other groups you mentioned. They can speak for ArbCom. On their actions often rests the balance and justice of a case. Tony   (talk)  09:02, 25 April 2009 (UTC)

Doubts
I don't suppose I'll have much effect by writing this, but I'll take the opportunity to point out what I think is fundamentally wrong with the arbitration setup.
 * 1) If you're primarily dealing with conduct (not content) disputes, then (to use a legal metaphor) you're a criminal court, not a civil court. In fact, since bad conduct is dealt with in the first instance by admins, you're a criminal appeal court. The parties in your cases should be appellant and appellee - the editor who feels he's been wrongly treated, and the admin who so treated him. No need for any more drama or legal posturing than just talking to those parties, deciding whether the admin's action was right, and taking appropriate corrective action. You certainly shouldn't be encouraging editors to debate at length about each others' conduct - this is just the sort of discussion that (deliberately or otherwise) derails the important process of reaching consensus on content matters. Of course, editors should be very concerned about incivility and other bad behaviour, as this will destroy our community if unchecked, but that doesn't mean we should all be encouraged to take on the role of prosecutor.
 * 2) You would have much more value as an institution, however, if you did take on content disputes in some meaningful way. These are the ones that matter - we're writing an encyclopedia, remember, and it's the content of that encyclopedia (and, secondarily, of the guidance given to those writing it) that ultimately matters to all of us. Of course, if we had an effective means of resolving such disputes, then we would need no help from ArbCom. But of course we don't - OK, most disputes go away, usually with the right result, but not always - and the serious conduct disputes arise precisely because we don't have a means of settling content disputes properly. Bad conduct would happen much less if it weren't rewarded in terms of getting what you're arguing for. But under the current system, it very often is rewarded, because in practice, it's the most effective edit-warriors and bullies who get their way if they're determined enough. ArbCom should be part of, or at least encourage the institution of, a system for determining, based on rational and civilized discussion, what the community's view is on a given content matter and for enforcing that view on those occasions when there are still people who disagree strongly with it.
 * 3) All right, I could go on, but I'll stop. Please think about it.--Kotniski (talk) 14:12, 9 April 2009 (UTC)
 * Erm, Criminal courts deal with crimes against the People (or the Crown, or the State) and civil courts deal with controversies between individuals, with enforced resolution by an outside entity. Arbcom covers both. It has universal original and appellate jurisdictions - and there is very often more than two groups of adversaries. I agree that our current "evidence" proceedings can, and do descend into a free-for-all attack though. As for this thing on conduct versus content, thats part of a larger argument best served elsewhere, but I will make the claim that a normalized, civil editing environment is the most efficient context for good article writing.--Tznkai (talk) 14:17, 9 April 2009 (UTC)
 * I agree strongly with your last statement, and for that reason, bad conduct is a crime against the People (possibly with a specific victim as well). But with real crimes, that victim (or other bystander) doesn't generally arrest or prosecute the perpetrator. On WP, if editors see bad behaviour, and gentle warnings don't help, they should be encouraged to report it to the admins and expect action to be taken. It's only if an administrator's action is questioned that we need ArbCom - the fact that you get cases that are more complicated is probably due to failure to require that complaints about conduct be specific and be addressed to administrators in the first instance. And I don't understand why the "larger" argument should take place elsewhere - where did you have in mind? --Kotniski (talk) 15:07, 9 April 2009 (UTC)
 * On this language issue, I want us to be very very very clear... ArbCom is not in any way a 'civil court' or a 'criminal court', Wikipedia is in no position to impose bailiff enforced fines or custodial sentencing. It is Wikipedia's internal Tribunal, who's powers against a person are limited at most to removing them from the project.
 * I'd probably suggest that we should consider renaming at some point from 'Arbitration' to 'Tribunal', to try and correct the terminology issue. --Barberio (talk) 20:19, 9 April 2009 (UTC)

I believe that Kotniski is raising issues that need to be talked through.

Adding content disputes would be a major change. I have specific questions on this. Can arbs/clerks explain here the original/continuing reasons for the restricting of the scope to behavioural issues alone? Is it a perception that the community should sort content out, always? Is it a fear that ArbCom might be swamped with work, or with cases that are highly political and difficult to rule on?

One major problem in the current scope is that it's often difficult to separate behavioural from content issues. The latter can keep festering as the source of the tension between editors. Tony  (talk)  15:56, 9 April 2009 (UTC)
 * I will give my best apologia for the standing restriction of scope:
 * A fundamental part of the wiki process, is that in content matters all editors are equal. As rational creatures, we should be able to come to an agreement (that is, consensus) through reasoned argument between editors. The best argument, the best content, wins on its own.Not only is this a philosophy we abide by, but it is as a practical matter, one of the few ways we have to quickly generate neutral writing on a wide variety of topics - with the input of as many voices as possible. When this process fails, it is because of behavioral issues. Civility issues clouding editorial judgment, users unwilling to participate in reasoned debate, bad faith users, trolls, vandals, and others who do not wish to act like rational creatures benefiting the project. That is, those who disrupt the editing environment. By eliminating those users from the editing environment, the editing environment should return to normal.
 * In addition there is no group of editors, certainly not the Arbitration Committee which is qualified to make judgements on all content matters - perhaps not any content matters. Human knowledge, which Wikipedia tries to capture, is far larger than any one, or any small group of humans could hope to manage. It is in fact, such a large topic, that it takes a project like Wikipedia to generate that much data that quickly. Our own success limits the ability of any one body to successfully control it. For example, there is no way, however laudable it would be, to create a unified "voice" for Wikipedia, a distinctive writing style, because there is far to much writing for any group to unify. Creating multiple content bodies would create either a useless and overwhelmed bureaucracy, or a high functioning but extremely slow bureaucracy.
 * Behavior problems, by comparison, are pretty simple. Most of us, most of the time, recognize when someone is being a jerk, especially when we're not involved. The Arbitration Committee then, is a collection of people who are responsible, mature, sensible and most importantly distant from the problems at hand. They are able to view it as an outsider, and sort it out.
 * Thus, the wiki process is best served, and perhaps can only be served, by a group focused on behavioral problems. While we have seen complexity rise that puts the model I've described in question (the fuzzy line between bad behavior and advocating for "wrong" content) certain fundamental truths have not changed. The amount of legitimate disagreement has not gone down, (perhaps our patience to admit the legitimacy of the opposition has though), the sheer breadth of content has not decreased, and the value of a normalized editing environment is still paramount. What has arisen then, are not examples that invalidate the Committee's underlying theory, but exceptions that need to be adjusted for.--Tznkai (talk) 20:20, 9 April 2009 (UTC)
 * I think we agree in principle with all that, but sometimes the facts on the ground fail to live up to the ideal. For me, it breaks down when you say "the best argument, the best content wins on its own". Yes, in 99.9% of cases, Just as in 99.9% of cases of behavioural problems the right result is achieved through the normal processes of peer pressure and admin action. It's only because of the exceptional cases that we need ArbCom at all. And in practice, the way we do things at the moment, the exceptional cases tend to be mixtures of content and conduct disputes all rolled in together, feeding off each other and disenfranchising the reasonable editors who aren't interested in fighting.


 * Rather than go on complaining, I'll just set down my proposed model in a new thread (below).--Kotniski (talk) 07:20, 10 April 2009 (UTC)
 * Tznkai: thanks for your detailed explanation, which does present a reasonable case for the behavioural-only practice. However, a few things remain: if all behavioural cases were "easy to recognise", why is ArbCom necessary? And what of behavioural issues that arise from unresolved content issues that are seemingly intractable without authoritative decision-making? And ArbCom can, as pointed out above, deal with content issues at its discretion (but this needs to be defined a little more closely). Tony   (talk)  16:30, 11 April 2009 (UTC)
 * To butcher a U.S Supreme court Justice's statement: "ArbCom is not final because they are necessary, they are necessary because they are final." What it really comes down to is that sometimes you need a body at the end of the road to take the responsibility to make a bold and definitive decision, because people get unreasonable and stubborn. As for content-originated and other intractable disputes? They're a serious problem (as I have commented elsewhere) and tend to be genuinely difficult to solve. I'm not convinced however, that unshackling ArbCom to make content decisions is a good solution. The current structure of Arbitration cases invite more bickering in these disputes, not less, and in no meaningful way illuminate on the underlying content issue, and we still have the problems I pointed out above, namely that I'm not sure I trust any group of users to dictate a content decision, let alone a body that was elected to do something else. Of note, the body that does deal with content decisions, WP:MEDCOM, is often avoided or actively sabotaged by participants. I agree there is a serious problem, but I don't think the solution is within arbcom expanding its scope. I have made some fairly radical suggestions elsewhere, but I think there are also some very simple ones. All of these however, involve encouraging arbcom to be more aggressive, and that is usually met with significant resistance.--Tznkai (talk) 18:34, 19 April 2009 (UTC)

A model for dispute resolution
This is my basic model, which I suspect is far too simple and rational ever to be accepted at WP, but one never knows... The principles would be something like this ("admins" may be extended to include other neutral and respected editors):
 * 1) Disagreements over content are decided by good-natured reasoned discussion.
 * 2) Disagreements over what has been decided in reasoned discussion are resolved by neutral admins. (This isn't a fundamentally new concept - it just extends what already happens at AfD etc.; and doesn't require the admins to have specialist knowledge, just to recognize a cogent argument.)
 * 3) Complaints about conduct (if a polite peer-to-peer remark isn't sufficient) are made to the admins, who take appropriate action. Unacceptable conduct includes disrupting the process of good-natured reasoned discussion (per 1) and editing contrary to what has been decided in reasoned discussion (per 2).
 * 4) Complaints about admins' acts or omissions are made to the wider body of admins, and if still unresolved, to ArbCom.
 * 5) Administrators' and arbitrators' rulings (where not obvious) should be preceded by focused and mutually respectful discussion with those involved, to ensure that all arguments have been noted and understood.

Details to be filled in of course, but honestly, wouldn't this be a far better and clearer basis to start from?--Kotniski (talk) 07:17, 10 April 2009 (UTC)


 * I'm not sure where to reply, because I have issues with what you say in the section above this one, but you've sort of closed that down to move here. My first issue is that arb-com isn't a court, it is an arbitration committee.  It's meant to arbitrate. Where I think arbitration has failed is that it doesn't tend to arbitrate anymore. If it worked along the lines of an actual arbitration body, limiting itself to issues of behaviour, and provided a final, binding decision, then maybe things would move a little better on Wikipedia.  For starters, if arb-com better enforced our policies and guidance on civility, we wouldn't be in half the mess we are, because everyone would have to speak on a level playing field.  Similarly, if it upheld the idea of consensus, half the issues we have would quickly be resolved. How do I mean?  Well for me, if I have a dispute with someone, I'll try and settle it by building consensus or discussion.  And when that fails, I try and get a wider input, to get a decision made, one way or the other. Just so we can have it done and dusted for the time being and all go do something less boring instead. That's how Wikipedia is supposed to work.  It doesn't always work like that though.  Sometimes things get a little awry, and I'm as guilty of that as the next person.  But that's where WP:DR and arbitration come in.  If you've got a behavioural dispute, where two sides are breaching 3RR or edit warring, or whatever the behaviour may be, that's where arbitration has got to step in.  And we've limited arbitration to behaviour, so for me, it's quite simple.  Arbitration should just call out bad behaviour, and deal with those committing it.  And if it is at arbitration, then that call out is the final call out. So "kick them out of the game". If they can't build consensus period, bye bye. If they can't build consensus on a specific issue, then they don't get a say on that issue. For me, it's that simple. Our consensus making policy should actually resolve most of the disputes we ever have here, because it ultimately means that the side with the most people dedicated to seeing the issue to a finish will get the "final say".  The trouble starts when the side in the minority refuses to recognise that. And that's when arbitration kicks in, because part of consensus policy is that you have to compromise. Someone not compromising is not building a consensus. So you can see people who refuse to compromise will end up on the smaller side at the end of a dispute, because people will have eventually compromised by dropping out of the dispute.  That's how it should work. That's the framework of decision-making on Wikipedia. There's a handy flow-chart at File:Consensus_Flowchart.svg that delineates it. That's my take on it, anyway. Hiding T 12:05, 10 April 2009 (UTC)
 * The one question that comes to mind for me is when we have intractable, entrenched, partisan arguments, when basically the only people consistently involved are those who are least rational and reasonable. I too think that in such cases policies or guidelines could be written to deal with them, but then what will probably happen is that the same intractable partisans who were arguing before will continue to argue, saying the other side is at fault. I agree that ArbCom should be primarily an arbitrating body. However, there are, admittedly rare, situations in which more is needed. Individual admins could conceivably step in and act according to rules in such cases, but in the more contentious fights there will already be admins probably on both sides, making it harder to know what to do. For situations such as these, when the community makes it clear that for whatever reason it cannot or will not act to resolve a matter which seemingly needs to be resolved somehow, I can see ArbCom perhaps not dealing with it directly, but maybe helping create a body of experienced, reputable editors who can review it and arrive at a fair solution. Like I said, these situations aren't that common. However, the longer we continue to not have an effective way to deal with them, the more common they will become, and the more difficulty there will be in resolving them. John Carter (talk) 13:17, 10 April 2009 (UTC)
 * When we have intractable, entrenched, partisan arguments, we have people in breach of WP:NPOV. What more do we need to know? It really is that simple. Hiding T 13:49, 10 April 2009 (UTC)
 * Agreed. So, in those cases, everybody involved is in violation of POV. What do we do, block or ban the lot of them from given content? What does that do the content? Particularly, what if they're the ones developing the content in the first place? This happens, particularly with things like the 9/11 conspiracy theories and the like, or when they're the ones most familiar with the content so that they would have the best of idea of WP:DUE and the like. Yes, in an ideal world, we would have enough community-minded individuals willing to undergo virtual torture to deal with raving loons on both sides. That don't happen that often though. And, of course, if an outsider were to consistently oppose one side or the other, possibly because they're, well, wrong, that side will now consider the outsider a partisan for the other side and begin to ignore any input from them. We could virtually permanently lock the content in dispute, but that's generally something we try to avoid. And, in some cases, even if we did, the only thing that would happen would be that, as soon as the block is lifted, the edit wars start over again, from the same people doing the same things.
 * I myself don't see it as an ideal solution, but I'd rather have a way to bring these arguments to a clear conclusion or cessation pending definitive outside input than have edit wars going on for a number of years, potentially costing us some of our best editors when they find themselves in a minority position or such. And, yes, if we don't like the "interim measures", that gives us all the more reason to try to resolve them some other way more quickly. John Carter (talk) 14:07, 10 April 2009 (UTC)
 * You've lost me. At what point are partisan people the right people to adjudge WP:UNDUE? Only people who do not take a partisan approach cen determine what is due weight. Yes, people should be blocked and topic banned, that's the only way you'll get a consensus on the issue, that's how consensus is supposed to work, how Wikipedia is supposed to work. That's the well from which everything springs.  Hiding T 14:26, 10 April 2009 (UTC)
 * I was actually thinking of undue weight within a given article, like say how much of a central article to devote to a given conspiracy theory, for instance. But, if we did the blocks or topic bans like things should work, particularly on the more contentious subjects, we run the risk of basically losing that content, either through vandalism which takes place after all the regulars have been "kicked off" the page or simple lack of updating. I agree it's the way wikipedia is supposed to work. But with almost 3 million articles now, most of us, including maybe you, are really just barely able to keep up with the things we are most interested in, and unfortunately sometimes willing to leave the other stuff to the hypothetical others who care about that. But if those others have all been blocked or banned from that content, there's nothing they can do either. And we can't force people to take an interest in and be willing to work on stuff they're not really interested in. I wish, but we can't. How often does it happen that important content becomes damaged through lack of interest? I honestly don't know. But I can see how it easily could happen fairly frequently. And maybe to prevent that, in those rare cases when it's the only way to do so, allowing ArbCom to maybe step over the line of arbritration for a while and make "interim declarations" or whatever you want to call it to preserve the work that's already been done on some subject might be acceptable. But, yes, we would want it as rarely and for as short a time as possible in each instance. John Carter (talk) 14:46, 10 April 2009 (UTC)
 * I'm not following you at all. I reject your model, but if I do follow it, then I'd opt for the lesser evil, which is a poor article, than having a poor article people argue over. But where I reject your model is at the start. Someone who can adjudge how much of an article to give to a given conspiracy theory is not someone likely to get blocked or content banned, are they? Basically, if everyone followed the policies, we'd have no problem. And if everyone who wanted what was best for Wikipedia followed the processes, we'd have no problem either, because at that point anyone not following the processes is easily identified as not wanting what is best for Wikipedia. So if we start acting like that, we actually start supporting those editors you think would get topic banned. Hiding T 15:00, 10 April 2009 (UTC)
 * I understand your position. But there will be cases when an article has been good, maybe even recognized, and then the circumstances of the world or whatever change and new content has to be added. I'm thinking of the Ireland naming mentioned earlier, very inflammatory new published allegations regarding a person or subject, or any case where there is a hotly contested new issue regarding a given subject. That issue may not be all-important to the article, but it might be all-important to those editors at that time, for however long the issue hangs around. The rest of the article might be in good, maybe very good, shape. And the new controversy might only affect a piece of it, but be almost constant. If there are "sides" to those controversies, and there often are, they might be comparatively unbending. I'm told some of the Israel-Palestine articles, and other material on contentious issues, is like this, where there are editors who want to make the content fair and balanced on both sides, but both sides reject the sources of the other as unreliable. And, to an extent, in some cases, they're both somewhat right about that, but they might be the only sources we've got. And in these issues in particular, when personal issues also often figure in, it can be very hard to be really objective about the material submitted from the "other side". Yes, we've got a noticeboard for issues like this, but I know of a few very good editors who have been blocked or banned from helping the content anyway. Personally, I'd rather have some less than perfect interim rules with a few more very good, but less than perfect, editors still able to edit, than better rules and fewer good editors. But I can see how others might think differently. John Carter (talk) 15:14, 10 April 2009 (UTC)
 * I can't see how this problem would not be solved through correct application of WP:BRD and WP:DR. Hiding T 15:35, 10 April 2009 (UTC)
 * I maybe have less experience of these major conflicts than most people here, but I really don't see how it's better to let something go on and on spiralling out of control, and have a committee attempt to "rule" on it some long way down the line, than have admins take control of the situation early on, as soon as it becomes clear that there are enough belligerents around to thwart civil discussion, or where it's clear that those involved in the discussion are not going to be able to agree for themselves what conclusions that discussion has led to (this is perhaps the key point that people usually don't get - I've been saying it all through the date linking thing - if we're going to say that consensus rules, then we must have a convergent process for deciding what consensus is - just like we already do reasonably successfully at AfD). If we laid down clear principles that would be applied on an ongoing basis, like the ones I've tried to formulate above, and the existing (theoretically) ones others have referred to, then ArbCom's role would be much clearer and the cases it hears much more focused - basically they would be reviewing specific administrative actions (or possibly inactions). The first place anyone (particularly bystanders) should take any complaint about behaviour should be the appropriate admin noticeboard, and admins must act decisively (I don't mean block without warning, but talk tough) to rein it in straight away. I may be missing something, but I see no other role for ArbCom than the important one of a final reviewer of admins' and other officials' actions. It doesn't need to create ("interpret") new policy - that could be done the community without any help if we could be sure that a consensus-forming and -enforcement process existed and would not be allowed to be derailed. Happy Easter everyone, --Kotniski (talk) 17:08, 10 April 2009 (UTC)
 * It's not better to let something spiral out of control, and that's been my point to John Carter all along. I've always maintained if everyone stayed within the bounds of behavioural policies, it would be easier to work out who is causing problems, because they'd break cover immediately. Hiding T 18:57, 10 April 2009 (UTC)
 * And, for the most part, I agree with you. My points of disagreement are (1) while I agree with you regarding future problems, your solution goes no way toward addressing current ones that somehow have come into existence despite the best efforts of many people, (2) even when many people are behaving well and some are not, there can be a situation where the person(s) misbehaving may in some way be immune from any sort of sanction or those who could impose one, for whatever reason, choose not to do so, and (3) unfortunately, in several cases, once it gets to outsider intervention, the outsider might face the problem of sanctioning most or all the parties involved, for the occasional "slips" they may have made since the disagreement began. In such cases, the only people who might be able to intervene fairly are the ArbCom, but, as we know, sometimes it can take a while for a decision to come down, which can exacerbate the situation itself.
 * In short, while I agree that I would prefer seeing what Hiding proposes, I think that realistically if we chose that option as the sole tactic we would, regretably, still see many of the same problems we see today. And those sorts of problems, which develop over time and involve several editors, are generally the most difficult to resolve. Maybe, in those cases, it might be better to have a bandage placed on the problem quickly, and actively seek to create a more permanent solution immediately. Of course, even then, when new policies and guidelines are proposed to deal with specific problems that have arisen, it often happens that they are not approved because people either aren't interested or think that even a limited new guideline would be misused. Then we're right back where we started again. John Carter (talk) 19:32, 10 April 2009 (UTC)
 * I'm not really sure what you're saying. We don;t really need any new behavioural policy, we have all we need.  What we need is to actually apply it.  Too many times we kind of let people off the hook because, well, we just do.  If we stuck to our guns and simply held to the fact that you don;t disrupt Wikipedia to prove a point, and recognised that consensus means that ultimately, the majority view does decide, there are no issues. Any dispute, active right now, can be settled with recourse to appropriate policy right now.  There's just very little will to do so, and there's very little authority left on Wikipedia to be able to do so. It's very hard to make a block against an established user stick now, and it's very hard to get a consensus established now.  There's just no will to recognise anti-consensus actions for what they are any more. Even this discussion proves that. Hiding T 16:14, 13 April 2009 (UTC)

Executive summary of the major changes?
That would make community input so much easier and more effective. Did anyone think of doing this, or are the changes so great as to make that not useful? Tony  (talk)  15:58, 9 April 2009 (UTC)
 * I second the motion. Looie496 (talk) 19:14, 9 April 2009 (UTC)
 * I can try to put a list of the most salient changes; but, overall, the changes from the current policy are so substantial as to make a comprehensive list infeasible. (A list of changes from the earlier drafts would be much easier to construct; but I don't think that's what you're asking for, is it?) Kirill [pf] 00:17, 10 April 2009 (UTC)
 * For my part, a list of the most salient changes would help, so I don't have to do a point-by-point comparison to figure out what the important changes are. Looie496 (talk) 00:21, 10 April 2009 (UTC)
 * Okay, I'll work on that once I finish some of the other paperwork I have this evening. :-) Kirill [pf] 00:24, 10 April 2009 (UTC)

Here's a list of some of the major items in the draft that aren't present in the current policy:


 * Arbitrator appointment and removal
 * Establishing interim measures until consensus forms
 * Not ruling on content disputes
 * Expedited hearings
 * Private hearings
 * Right to respond to cases
 * Departure and resignation during cases
 * Evidence from outside sources
 * Private submission of evidence
 * Amendments to the policy

This certainly isn't an exhaustive list, and I'll probably add more detail if/when I have the time to go through and do a clause-by-clause comparison; but it should be enough for people to comment on. Kirill [pf] 03:15, 10 April 2009 (UTC)
 * Most of those appear to be what ArbCom is presently doing. Are there any changes from present practice? Septentrionalis PMAnderson 18:42, 10 April 2009 (UTC)


 * The only significant changes from nominal current practice that I can think of are the restrictions on the purposes for which evidence from outside sources can be used, and the explicit recognition of a right to respond. The other stuff is mostly either formalization of existing practice (e.g. departure and resignation rules) or contingency cases that haven't actually occurred (e.g. arbitrator suspension and removal). Kirill [pf] 02:11, 11 April 2009 (UTC)

Timeliness
The previous discussions also had support for 'timeliness' requirements. As I've stated elsewhere tho, I think it might be unwise to put those in at the same time as making substantial other changes, since we don't and can't know how the work flow will settle out till after the new system has been running long enough to be 'normal'.

So I'm going to recommended again that we hold off on 'timeliness' and 'fast track' alterations till at least the end of the year. --Barberio (talk) 20:11, 9 April 2009 (UTC)


 * I'm not convinced that the precise details of case scheduling need to be spelled out at the policy level in any case (beyond the nominal minimum period for responses); and, given some of the other reforms to case handling that we're moving on, this may become somewhat of a moot point. Kirill [pf] 00:19, 10 April 2009 (UTC)
 * Policies reflect current practice; they aren't a vehicle for reform. If we're looking to insert a clause into the policy resolving to process cases in a timely fashion, then it's because that's what's currently being done—and because it's what is expected. On that basis, Kirill, declining to insert the clause into the policy on the basis that it's a "moot point" and is now present practice seems wrong to me. AGK 19:14, 20 May 2009 (UTC)

Confidentiality
I may be missing this somewhere else, but I think the arbitrator conduct section should outline how arbs are expected to treat confidential information. And maybe elsewhere outline what kind of correspondence with ArbCom or arbtrators should or should not be assumed confidential.-- Birgitte SB  20:35, 9 April 2009 (UTC)
 * Agreed. Some common sense is needed, but I would support moving to a system where material sent to arb-l is not considered automatically confidential, but that people sending in confidential information should mark it as such. The current system, where people send in stuff and we have to say "sorry, that's not confidential, please post it on the case evidence pages" does work to an extent, but I would prefer it if people e-mailing arb-l said explicitly "this is confidential". One obvious case is people e-mailing us under their real names. People do rightly assume that this is confidential, but in some cases we have to pass e-mails to others to help us resolve particular cases, and things get held up as we ask for permission to do this. Sometimes, people say "no, please don't pass this e-mail to XYZ", which is fair enough. But if XYZ has had allegations made against them, then we have to summarise what has been said. Those are just two examples. Checkuser cases are even more complex in terms if privacy and confidentiality. Carcharoth (talk) 16:40, 12 April 2009 (UTC)


 * I changed the summary a bit. You need not focus on arbitrator conduct if you instead label certain lines of communication exclusively confidential so that everyone is aware that other lines of communication are not confidential.-- Birgitte  SB  16:07, 22 May 2009 (UTC)

Exclusivity
It might be worth noting that ArbCom (currently) has nearly exclusive jurisdiction over certain things: CU/OS, non-emergency, admin desysop, I'm certain I am forgetting many things.--Tznkai (talk) 20:50, 9 April 2009 (UTC)

Scope of arbitration
In exceptional cases of communal division or exceptional risk of harm to the project, the Committee may determine a interim means to achieve the aims of communal policies or to curtail dispute in the area, until the community has developed a better consensus.

I'm having a dense moment, but for the life of me I have no idea what this means. I can grope towards something like if the crap hits the fan, we have the right to try and pull a rabbit out of a hat, but that's no basis for scope, and that's not really within arb scope anyway. Or is it? Hiding T 11:31, 10 April 2009 (UTC)
 * Doesn't it mean things like temporary injuctions, like I understand has been imposed in the date linking saga?--Kotniski (talk) 11:44, 10 April 2009 (UTC)
 * I don't know, that's why I am asking. I haven't really followed the date linking saga. It's one of those tiresome issues where a decision is needed but is being filibustered. The way I look at it, if you allowed everyone to act within policy over the issue, auto-formatting would be deprecated, because if no-one breaches WP:3RR then they'd be removed from articles based on weight of numbers. Sometimes it is that simple, once you reduce policy down to their actual intent.  That's why policy cruft has been the biggest barrier to getting things done on Wikipedia. And my god, am I sorry for the part I've played in crufting them up. Hiding T 12:11, 10 April 2009 (UTC)


 * This is actually more spurred by things like the Ireland naming dispute and so forth, but date delinking is another example. Basically, if the community can't agree on a policy, we're not going to impose one by fiat; but we will, essentially, pick an approach to be used in the interim until a policy emerges, to try and limit the spread of the conflict.
 * For example, if there's a fight over the meaning of a deletion criterion, and the community is having trouble coming to a consensus, we'd say something like "until the community decides otherwise, administrators should interpret the criterion as meaning 'X'".
 * Currently, we try to do this by using temporary injunctions, but those only last for the duration of the case—meaning that we have to keep the case open for them to have any effect—and are really supposed to apply only to the parties to the case itself. The idea here is to make it explicitly permissible for us to issue a temporary ruling in an active dispute without having to leave a case sitting open in the meantime. Kirill [pf] 12:40, 10 April 2009 (UTC)
 * I'm not sure I agree with this then. I'd hate for arb-com to start deciding how policy should be interpreted. I'm not sure I see why this is necessary either. There are plenty of other tools within the arbitration box to take this forwards. Surely there would be a distinction between good faith efforts to resolve the issue and bad faith efforts, that would be evinced through evidence. I've looked at Requests for arbitration/Ireland article names but no temporary injunction is listed.  What have I missed? Hiding T 13:11, 10 April 2009 (UTC)
 * The "No moves pending discussion" remedy is the most obvious; but the "Back-up procedure" and "Binding resolution" remedies are also applicable here. Basically, the entire outcome of the case was a temporary ruling pending the community coming up with a permanent one. Kirill [pf] 13:17, 10 April 2009 (UTC)
 * "No moves pending discussion" would already lead back to WP:CONSENSUS, WP:BRD and move policy, wouldn't it? "Back-up procedure" is simply implementing WP:DR. I'm staggered at "Binding resolution". I need time to build a suitable response to that. My initial reaction is one of horror. Hiding T 13:38, 10 April 2009 (UTC)
 * I can't work out a way in which I could support a solution such as "Binding resolution". My apologies. That's crossed the line for me. Hiding T 13:54, 10 April 2009 (UTC)
 * It certainly is something that we would not use except as an absolute last resort. This ruling is an example of why that we need to have the ability to do when there is no other way to end a dispute that the Committee and the Community agree needs to end. FloNight&#9829;&#9829;&#9829; 14:14, 10 April 2009 (UTC)
 * Then I guess I oppose it as vehemently as I can, for whatever that is worth. My sincere apologies. I can't support something that allows arb-com the right to make such decisions. I appreciate what you are trying to do, but I don't think Arb-com has or should be able to have the right to limit the actions of the entire community. I think you agree with me, judging by your comments when the decision was proposed. Hiding T 14:33, 10 April 2009 (UTC)
 * Hiding, in some ways what the Committee does is similar to an admin protecting an article. An admin looks at the article closely enough to make sure that we are not protecting it with a major policy violation like a blatant copyright violation, vandalism, or obvious defamation. But other than that, an admin does not choose which wording of the article to use, instead, they protect in the form that they see without making a decision about the content dispute. When the Committee acts in policy or content dispute we are interested in halting the active reverting or other conduct that is making the situation absolutely unacceptable, and assist in the ways that be can to help the Community find a solution to the dispute, and if necessary use our ruling to make it stick. So, we dig into the dispute slightly so we understand the nature of the conflict and what needs to happen to get a stable result. But we are not interested in making a decision on the actual dispute. We are interested in finding a process that will bring an end to the conflict. For small issue we will not get involved at all. But if the conflict covers a large number of articles, then something needs to happen to assist in bringing stability to the articles. FloNight&#9829;&#9829;&#9829; 13:34, 10 April 2009 (UTC)
 * I'm not sure what you are responding to here. Hiding T 13:47, 10 April 2009 (UTC)
 * A general response to all your comments in this thread. You didn't seem to understand the reasoning behind our decision to get involved, our method of intervening to solve the dispute. FloNight&#9829;&#9829;&#9829; 14:11, 10 April 2009 (UTC)
 * I think we are talking at cross purposes. I don't understand the language in this injunction, nor what it means.  But I don't disagree with what you have written, so maybe I should try another tack.  Why is this addition necessary?  What will it allow that isn't currently allowed?  Hiding T 14:16, 10 April 2009 (UTC)
 * Your questions and concerns are understandable. I don't have time at the moment to give as complete a reply as this needs, so I'll get back to it later. I do want to explain as best I can, but need to do a few other things related to case work first. FloNight&#9829;&#9829;&#9829; 17:39, 10 April 2009 (UTC)
 * No problem. Hiding T 18:59, 10 April 2009 (UTC)

Possible "full disclosure" clause
Part of the problem with arbitrations is that editors tend to give very partisan evidence, and actively conceal important evidence that may hurt their case. It would be helpful if they were required to provide full facts, even stuff fatal to their case, as a way of moving things along. Roger Davies talk 14:04, 10 April 2009 (UTC)
 * Exactly how are we going to enforce this? In the U.S the analogous structure is that the prosecution must reveal all pertinent facts to the defendant (in the interests of justice) and both sides have certain obligations to announce and prepare each opposing side with certain information. We don't have prosecutors and lawyers, we don't have the power to hold people in contempt of court, and we can't disbar people for misconduct. I mean, it'd be awesome if people were expected to treat arbitration with the same attitude they are supposed to approach articles ( and even better if they did so), but specific behavioral injunctions/requirements such as this beg to be enforced - and I don't see a good way to do it. --Tznkai (talk) 15:38, 10 April 2009 (UTC)


 * This is a difficult question. The Committee has often found itself having to dig up a significant amount of the relevant evidence on its own in order to make a fair decision that reflects the needs of the encyclopedia and the community. Ideally, that evidence would best be developed by neutral third parties; when this happens, I believe the decision is often better balanced and can be more timely as well. Risker (talk) 16:14, 10 April 2009 (UTC)
 * That would be delightful - but how to incentivize these 3rd parties to actually show up?--Tznkai (talk) 16:30, 10 April 2009 (UTC)
 * Incentivize? I presume you mean "what incentive can be provided to neutral third parties to develop evidence?" I don't have a good answer to this question. I know that several current members of the committee have done this before their tenure; from what I recall, their motivation was genuine curiosity at what the results would be. Evidence is a lot more useful in crafting good decisions than, say, extensive discussion of a fairly subjective evidence submission, or proposing findings based on such limited evidence. Risker (talk) 16:41, 10 April 2009 (UTC)
 * For whatever reason, there are usually more statements than people providing evidence, so interest in commenting on the case doesn't correlate well with providing evidence. Who knows? Maybe simply establishing that neutral evidence is welcomed will help.--Tznkai (talk) 17:13, 10 April 2009 (UTC)
 * It would suffice if editors approached articles with the attitude that they are supposed to use in approaching articles. But they don't. Septentrionalis PMAnderson 18:45, 10 April 2009 (UTC)
 * Given that we are often dealing with angry people that dislike each other and hold strong points of view, I think that the involved parties think that they are giving a fair telling of the facts. At least that has always been my impression. FloNight&#9829;&#9829;&#9829; 16:45, 10 April 2009 (UTC)
 * Anyone has the right to say 'I forgot that', 'I am sorry, I missed those evidences' or 'I didn't believe that would be considered as an evidence of something', etc... That'd be moot. -- FayssalF  - <font style="background: gold">Wiki me up® 17:02, 10 April 2009 (UTC)
 * I think that the previous attempts at dispute resolution will be the only venues where there is a hope of third parties having looked neutrally at all the evidence (RfC, Third Opinion and the Noticeboards) - by the time it has got to RfAR the positions are entrenched. I think this is adequately covered by the requirement for there to have been previous attempts, since these histories can be reviewed. Further, by the time a dispute gets to ArbCom it has usually been distilled into the pertinent policy violations and there is no need to see what collateral disruption existed and had been previously commented upon. Where there is an arguable dispute both sides will present their cases along with evidence, for which it is the ArbCom's duty to sift and test along with the community at the workshop and other pages, and where one side is lacking in reasonable points it is indicative of them being in error. Why have the necessity of trying to find "missed" problems? LessHeard vanU (talk) 19:31, 10 April 2009 (UTC)
 * Why have the necessity of trying to find "missed" problems? Because our mission is to restore a collaborative editing environment for the Community so that stable NPOV articles can be written. If one or both sides in a dispute do a lousy job of presenting the evidence of how contributors are really violating policy, we still need to know it so we can find the appropriate remedies to end the dispute. FloNight&#9829;&#9829;&#9829; 16:35, 11 April 2009 (UTC)

May I suggest that if the Committee has good grounds to believe that a party has engaged in deliberate deception on the Evidence page, whether by omission, distortion or straight-out lying, that this should be addressed in the judgement. Perhaps there are grounds for stating at the top of the Evidence page that all participants are expected to express themselves in good faith, without knowingly distorting their account of the matter or engaging in deceptive behaviour. Beyond this, the suggestion is impractical and unenforceable. Like it or not, ArbCom hearings are adversarial and, as pointed out above, parties are likely to put forward evidence that favours themselves and evidence that disadvantages their opponents; whether they fully realise they are doing this is another matter. The RL legal system, and many other forums at WP, work in the expectation that parties' exaggerate their cases (lawyers make a profession out of such exaggeration, although they are bound to serve the court by not consciously lying). Risker's idea of having third parties dig up other evidence may occasionally help proceedings (usually not, I suspect), but it is hard to see how this could be made a systemic part of the process; it would bring with it the potential for further complexity and subjective viewpoint, as evidenced in RL by the practice by both sides of bringing in their own "expert" witnesses. You can never be sure of their motivations. However, the more telling point in this discussion is that the expansionary model is still at play. The key problem in ArbCom hearings is excessive, not insufficient text. At least if arbitrators didn't have to make their way through such a huge jungle of angry irrelevance to scope, they'd have more time to visit the pages and see for themselves. A tight, trim structure is urgently required to make all parties take the matter of relevance more seriously. The current advice and structure opens the gates to fluff and tongue poking. There is a lack of decorum in the hearings. Tony  (talk)  15:02, 24 April 2009 (UTC)


 * I couldn’t agree more with everything Tony wrote above. Well put. Greg L (talk) 15:19, 24 April 2009 (UTC)

FT2 comments
Comments are split into two sections - those where the effect is substantively unchanged but needing loopholes fixed or other copyediting, and those where important omissions or issues require some checking by the Committee and community.

Copyediting
There are several places where wording could be simplified or is too formalized, as well as some other points where some clarification may help:
 * With suggested copyedits annotated (link)
 * The same but without annotation (link).

Summary of main suggested edits (kept to a minimum):


 * 1) Subsection fever - Too many subsections tends to encourage a legalistic view, or be tiring to readers. A plethora of one sentence sections isn't needed - kill most of the third level headers, merge and cleanup.
 * 2) Arbitration proceedings section header - Needs to have a "one sentence version" of the actual goal of arbitration cases (as opposed to the duties of the arbitration committee). Added "Arbitration exists to determine an appropriate and binding way forward for intractable disputes (etc)"
 * 3) Normal hearings and dismissal of cases - both better merged elsewhere.
 * 4) Explanation of conduct vs. content disputes - Repeated contentious issue at RFAR, well worth clarification. Possible reword and footnote added.
 * 5) Conduct - Clerks manage conduct on most arbitration pages, not just the "case pages". Conduct that the clerks have to manage can routinely spill over onto WT:AC and the like, which aren't "case pages" as such. Fix.
 * 6) Private hearing participation - 1/ the right to participate is moved from "private hearings" to "participation", and 2/ sentence added per Kirill's point below to clarify that in case of genuine emergency or temporary action there will be opportunity to participate and respond after that action and before the case is finally decided or affirmed.
 * 7) Evidence - Considerable scope for improvement here. Many issues, some unnnecessarily complicated wording. Changes:
 * Easy to simplify the wording of "acceptable evidence"
 * Mediation policy, itself an exception to a general evidence rule, itself makes clear exceptions; important to note this and clarify better than it does at present.
 * A common case where external evidence routinely may be salient, is prima facie omitted; make it explicit.
 * Simplify the paragraph about posting such evidence on the wiki. Much easier to say "if you wish to post evidence like this, submit to ArbCom who will say whether its relevant and if so, whether it should be private or public".
 * 1) Follow up cases and clarifications - drafted a better explanation.

Hopefully the copyedits shown improve the draft.

FT2 (Talk 19:28, 10 April 2009 (UTC)


 * Brief comments on some of the points:
 * 1. Yes, this has already been noted above; the section fracturing will be reduced in the next draft.
 * 2. Isn't this redundant to the first section of the policy? I'm not in favor of repeating things merely because some people don't want to read the entire document.
 * 5. We are dealing specifically with proceedings here. General talk pages are adequately governed by the normal policies; we can simply add an overall "ArbCom can set conduct standards on its own pages" clause if it's really necessary.
 * 6.2. We're not forced to follow the minutae of the mediation policy; if there's a legitimate exception, MedCom will approve the use, and we'd rather not bother with arguments over gray areas.
 * 6.3. Merely because something is commonly used doesn't mean we must permit it. We don't need or want to rule on intent, in any case.
 * Kirill [pf] 02:25, 11 April 2009 (UTC)


 * 2. The first section states AC is "resolver of last resort" (as one of its roles), but nothing in the policy says what "resolving" means in a Wikipedia context. Not a trivial exclusion - we get a lot of people confused innocently or deliberately over that point, so it's worth adding explicitly, the purpose of dispute arbitration, which isn't the same as the role of the arbitration committee. Resolving doesn't mean "court-like figuring out who did what", it means "determining the best way forward for the community", and that's sufficiently different from the everyday meaning that somewhere, it should say so.
 * 5. An "ArbCom can set and enforce conduct standards on its own pages" clause would do the job just fine. Point here is that arbitration pages may have (or need) tighter behavioral controls and the right to set and enforce standards that are felt appropriate, more than the normal wiki page, since any kind of gaming, disputing, dirty-washing-hanging-out, grudges, accusations (true or untrue), untramelled multi-party arguments, etc may take place on these pages far more than on an average page elsewhere on the wiki.
 * 6.2. People taking a bad actor to arbitration will read AC policy where they may well not have read mediation policy (or forgotten the detail). Saying point blank with no exclusions "may not be quoted" misrepresents mediation policy and may encourage misconceptions (and worse, can encourage misleading quotation or 'lawyering).
 * 6.3 Then the case of "conspiracy to ...." needs wider scope. One person cannot be a "conspiracy", and in cases of serious disruption, POV pushing etc we might not name the real world connections, but editors need to be able to allude to the "bad actor"s real life affiliation with some business, group, or whatever. Since solo dedicated hardened POV pushers are a routine matter at arbitration, it seems odd to say that evidence from the outside is permissible when there are 2 or more individual people in a "conspiracy" to violate policy, but not when it is one person alone. Since external evidence should never be used unless directly salient, a wording to the effect "direct evidence" of a "seriously disruptive" whatever... makes sense. You don't want such evidence unless it's both serious, and very salient indeed, but to strike it out in all cases unless 2 or more users can be identified in a "conspiracy" seems to provide good odds of an issue best avoided. A user is entitled to say "this seems to me good evidence of a serious disruptive intention" (in private if required), even if in fact ArbCom reviewing the case ultimately form their own view of the case.
 * Thoughts. FT2 (Talk 10:59, 11 April 2009 (UTC)

Issues oustanding in the draft
There are also about 5 important points where community or ArbCom review is important, and more input may be needed:


 * {|class=wikitable width="95%" style="font-size:90%"

! Item ! Possible rewording ! Original ! Concern
 * Arb conduct
 * "Arbitrators are users endorsed and trusted by the community to undertake some of the most sensitive and wide ranging issues within the community. The trust of the community, the integrity and good faith of arbitrators, and the high standards expected, are to be respected in private wiki-related discussions, as in public."
 * "Arbitrators are users endorsed and trusted by the community to undertake some of the most sensitive and wide ranging issues within the community. The trust of the community, the integrity and good faith of arbitrators, and the high standards expected, are to be respected in private wiki-related discussions, as in public."

"Good communication and high standards are expected of arbitrators. More so than administrators, they are expected to enact high standards even under reasonable pressure. Arbitrators will explain their Wiki conduct and matters relevant to their trusted roles, to their colleagues, if their actions cause concern, and (privacy permitting) to the community."
 * Arbitrators shall, at all times, act with integrity and good faith to uphold the trust of the community and the high standards expected of them. Should their actions cause concern, arbitrators shall explain their conduct in matters relevant to their trusted roles to their colleagues and to the community.
 * style="background-color:#ffd8d8" | Not nearly strongly enough worded. Commented same on the previous draft of the policy. Needs to be explicit about the standard expected.
 * Recusal
 * "Arbitrators are not expected to be free of "history" of interaction with each user and topic, but their ability to reach a decision neutrally and fairly is paramount . Arbitrators are expected to assess their neutrality for each case and to take note of the concerns of others. An Arbitrator who may be perceived to have significant past interactions should either recuse, or else should disclose any prior issues (privately or on-wiki as appropriate), and explain the basis of their belief they are able to assess the case neutrally, if necessary."
 * An arbitrator shall recuse himself or herself in any case in which he or she has a conflict of interest, such as prior involvement in the substance of the dispute or a history of disagreements with one of the parties, such that his or her impartiality in the case could reasonably be questioned.
 * style="background-color:#ffd8d8" | As above.
 * Private hearings
 * "Regardless of whether an editor's conduct is considered by the Committee in private or public, the editor concerned will usually be given due notice and opportunity to respond before any decision. In very serious cases, or those where for privacy or other reasons this is not practical, a provisional decision and actions will be announced to prevent or reduce harm, and the editor concerned will have a right of appeal and discussion (in private or public as the Committee may decide) before any final decision is made or affirmed."
 * "... will be given due notice and an opportunity to respond, as provided below."
 * style="background-color:#ffd8d8" | Recheck by Arbitrators whether this is tenable as drafted. Issues:
 * "... will be given due notice and an opportunity to respond, as provided below."
 * style="background-color:#ffd8d8" | Recheck by Arbitrators whether this is tenable as drafted. Issues:

Although prior warning and discussion is often desirable and very common, a number of actions (including many admin actions every day) in fact get enacted first, with the right of appeal instead. The important feature is the user has the right to respond and to address the case, either before or afterwards. Is every case involving the Committee guaranteed to bear "due notice", prior warning, opportunities to discuss the case etc, before any decision or action? That needs considering before writing into a policy.

As examples, a number of cases involving very serious issues have not involved prior posts to the user, lengthy discussion etc; rather, a summary decision is made by the Committee with any discussion following. Runcorn, ArchTransit, certain serious harassers, are examples. So a commitment of this kind would be a change, and it needs considering whether it will always be viable, or sometimes not.

The wikilinked text "as below" does not in fact link to anything that answers the issues.

Private cases are a difficult area. Suggested rewrite fits current practice, and may help.
 * Community kept "in the loop"
 * "If any stage of a case extends (or seems likely to extend) beyond a usual or reasonable time, then any Arbitrator shall advise the community of the progress of the case and the matters causing the delay, no less than once every seven days."
 * Although lengthy cases are the norm, there should be a positive requirement for communication by the Committee, when there's a long delay, why it's so and what's actually going on. Experience suggests if this duty is laid on the committee as a whole it may never be done (or very watered down) hence given to any individual arbitrator to ensure it's explained, so that there isn't a hold-up in order to get a formal statement by the whole committee first. The time period ensures time is kept firmly in focus.
 * style="background-color:#ffd8d8" | No provision for informing community.
 * Public explanation for decision
 * "The arbitrators may discuss and deliberate on cases between themselves, particularly where warranted by privacy or similar factors, but the substantive basis for the final decision should be apparent from the decision itself or from arbitrators' comments on it."
 * The substantive reason for a decision should be given to the community, even if no details can be disclosed. This isn't said clearly. It should be.
 * style="background-color:#ffd8d8" | Clear public accountability and awareness.
 * }
 * style="background-color:#ffd8d8" | Clear public accountability and awareness.
 * }

FT2 (Talk 19:28, 10 April 2009 (UTC)


 * A few (brief) comments on the five points:
 * I don't see any functional improvement here; the proposed version is longer, but doesn't add any concrete rules for conduct. The substance of the clause is the requirement to explain one's conduct; everything else is just background, really.
 * The only functional difference I see here is a requirement to disclose potential conflicts of interest, yes?
 * We're not going to conduct private hearings without giving everyone involved the opportunity to respond; and genuine emergency actions aren't covered by that portion of the policy.
 * This seems more like a question of internal procedure. I don't think it ought to be in the policy to begin with; and its relevance is questionable, given the other procedural reforms we're working on.
 * This could probably be handled by some minor additions to the description of the decision components; really, the basis for the remedies should be apparent from the findings of fact.
 * Kirill [pf] 02:18, 11 April 2009 (UTC)


 * 1+2: You're missing the point I think. The point here is not procedural or functional in the sense that it doesn't lay down a proposed change of process. The aim of 1+2 are to emphasize the overriding need in these areas and emphasize the standard expected. The key expressions are not the disclosure and such, they are these two points:
 * "Arbitrators are users endorsed and trusted by the community to undertake some of the most sensitive and wide ranging issues within the community... More so than administrators, they are expected to enact high standards even under reasonable pressure"
 * "Their ability to reach a decision neutrally and fairly is paramount."
 * Those strongly worded points are what is missing. Recusal and conduct should not be a generic "high standard expected"; the wording should be rather more forceful and forthright than that. Saying that these are "not making a functional change" suggests the focus of these has not been appreciated.
 * 3. That's fine, then it needs to add that "This does not cover genuine emergency and temporary actions, which will be followed up by opportunity to respond." Updated: I have amended the suggested draft to cover this.
 * 4. No. Explaining to the community and participants what's holding up a case for an (apparently) unreasonable period is not merely "a question of internal procedure", nor in light of track record is reliance upon "other procedural reforms we're working on" sufficient. If a case is stressing users by taking forever, there should be a positive requirement that an arbitrator is to directly post to the talk pages, "what's going on to cause the delay, and where it's at", and to keep that updated by regular follow-up posts if the delay persists.
 * 5. It should be made clear. This adds a specific requirement that however the matter is addressed and presented, whatever the case or issue, the end result should be the community has some understanding of its substantive basis. Usually of course that will be via the FoF or other statement, but this guarantees the end result for all cases regardless of presentation, whereas "an FoF will be published" does not.
 * FT2 (Talk 11:18, 11 April 2009 (UTC)

FT2, I strongly disagree with your rationale about recusing. Many arbitrators have deeply personal reasons for recusing from certain cases, and publicly identifying some of those reasons can lead to revelation of private information about the arbitrator. The reason that an arbitrator recuses is not usually all that important; the fact that an arbitrator is willing to recuse if they perceive a conflict of interest is what is key. In my own case, there are very few on-wiki reasons for me to recuse (I don't anticipate many arbitration cases involving the James Blunt article, for example), but several personal reasons that I may recuse that will not be obvious from anything I have ever done on this site. In many cases, if my reason is linked to something that can be personally identifying, I am probably not going to provide details even to my fellow arbitrators. That is as much for their protection as mine; I do not wish to put them in the position of unintentionally revealing personal information about me, something that is very easy to do, especially when the primary means of interpersonal communication is via mailing list. Risker (talk) 03:59, 11 April 2009 (UTC)
 * The question then is, "Should an arbitrator who feels unable to give an explaination why they consider themselves neutral in the dispute, expect to (or feel obligated to) recuse"? It's not that common a problem, really. If an arbitrator wants to recuse (without explaining) they can, but if they are challenged on it, and feel themselves neutral, but do not wish to say why... that's a problem. The rest, see comment to Kirill above. FT2 (Talk 11:18, 11 April 2009 (UTC)
 * If an arbitrator recuses without giving a reason, no harm comes. There is no conflict of interest, and no personal agenda affecting the outcome. This is *not* a problem. Problems come when people *fail* to recuse, not when they do so without explanation. Risker (talk) 12:54, 11 April 2009 (UTC)
 * Then we're on the same page. The suggestion I've made is that "An Arbitrator who may be perceived to have significant past interactions should either recuse, or else should disclose any prior issues (privately or on-wiki as appropriate) and explain the basis of their belief they are able to assess the case neutrally if necessary". Presumably an arbitrator with no neutrality issues doesn't have a problem; nor does an arbitrator who will recuse. But an arbitrator with a possible or perceived neutrality issue should disclose it and their reason for believing they can be neutral if (and only if) they do not plan to recuse. Does that clarify? FT2 (Talk 13:49, 11 April 2009 (UTC)

right to not participate
A matter seems to have come up on an outstanding case, discussed at WP:ANB which bears on the proposed section 2.7.3 "Editors named as parties to an arbitration case and given due notice of the case are expected to participate in the proceeding" which in an apparently contradictory way continues to say, in 2.7.4, "Should a party to a case fail to respond ... or explictly refuse to participate in the case, the Committee may nonetheless rule on that party's conduct in his or her absence."--which second part seems to be the current rule. DGG (talk) 04:29, 12 April 2009 (UTC)


 * I don't see the contradiction here. A party has the right to respond, and they're expected to do so; if they fail to do so, we'll go on without them. Kirill [pf] 05:14, 12 April 2009 (UTC)
 * ...and it'll affect poorly on the party who doesn't. If you're "expected" to do something, and you don't, you can have no complaints when your failure to do that thing results in further sanction. For me, 2.7.3 and 2.7.4 make sense. Daniel (talk) 07:44, 12 April 2009 (UTC)
 * I think it's OK - parties are expected to respond, but failure to respond should not prevent a case proceeding. DuncanHill (talk) 12:28, 20 April 2009 (UTC)

Jimbo Wales's powers in relation to ArbCom: clash of texts?
The current text says: "In April 2007, Wales confirmed that the Committee could overturn any decision he makes in his traditional capacity within Wikipedia."

At that link, Wales said: Here, let me by decree in this very instant make the following binding pledge upon myself:

In the event that the ArbCom makes a ruling against me, overturning any decision I have made in my traditional capacity within Wikipedia, the ArbCom's decision shall be final.

*This* is a significant change to our policies.

Does this "traditional capacity within Wikipedia" include his relationship with the Committee itself? It appears to. This is confusing. Will it be retained in the updated text? If not, are we to understand that the current draft is a "climb down" from Jimbo's statement, in that he will now be given several explicit powers that are, ultimately, in conflict with his "decree" in 2007. If Jimbo sacks an arb. tomorrow, he has explicitly said that the Committee can say "get f...., we're overruling you". Will that be the case under the rewritten text? Tony  (talk)  15:42, 12 April 2009 (UTC)


 * I don't believe that the new wording represents a change in procedure, although I think both the old and new wordings are fairly unclear and would prompt something of a constitutional crisis if they were ever to be tested. I would view the "traditional capacity"/"role as project leader" clauses to effectively cover every action not taken in the role of a Foundation Board member; that is, office actions remain outside jurisdiction, everything else remains within, including actions related to the Arbitration Committee.  That is, I would expect that in the event that Jimbo sacks an arb, and the arb takes Jimbo to ArbCom, ArbCom would have jurisdiction to hear the case and potentially reverse Jimbo's decision.  The question of whether the arb should sit on the committee is mute because they would have to recuse anyway.  Ditto for a contested term alteration or controversial appointment.  There is no case I can think of where it would be impossible to find a neutral ArbCom to hear a case against Jimbo unless he tried to sack the entire ArbCom, which would be such a messy situation as to render any policy wording we can make here mute (since he would be effectively tearing up the entire ArbPol anyway).  I don't think it's a particular issue. <b style="color:forestgreen;">Happy</b>‑<b style="color:darkorange;">melon</b> 16:00, 12 April 2009 (UTC)

It's debatable whether this is a useful question. A lot of time has passed and ArbCom is no longer new, Jimbo is no longer in the same day to day "hands-on" role he was in 2003 when the idea of an Arbitration Committee was suggested. But if one is referring back historically, then this seems to be the rough position:


 * Originally, a user who was seriously disruptive, could only be banned by Jimbo, effectively as being the project leader. Disputes were handled so far as possible within the community and the community used consensus as its decision process. When the community gave up, the case was referred outside the community to Jimbo, for an executive decision as he saw fit.
 * The Arbitration Committee was very clearly created as a delegated vehicle to Jimbo. That is, instead of poring through the evidence himself, he could ask the committee to hear a case, and broadly he'd be happy to trust their decision. (It was only later that it was proposed cases could be submitted to Arbcom itself directly by the community.) The position was still, in effect, that all disputes were to be resolved by the community, collaboratively, using consensus seeking, and if a serious dispute could not be resolved this way it was submitted to a body outside the community, for a decision (ie Arbcom acting as Jimbo's stand-in).
 * This places Arbcom in an unusual position in which it is partly in and of the community, as a panel of trusted users, but also partly outside the community as a vehicle for Jimbo's day-to-day powers as project leader (ie to hear and decide certain cases) to be managed by a committee rather than by him personally.
 * So my interpretation of the relationship is roughly, that when it comes to actions within the community, such as blocks, bans, changes or access, editing decisions etc made by Jimbo as a normal contributor, Jimbo has said "if there is doubt, I will accept Arbcom's view on it". But when it comes to the Jimbo-Arbcom relationship generally, Arbcom is still in part constituted as a vehicle outside the community to handle Jimbo's executive powers on a day-to-day basis, and Jimbo still "owns" that relationship (even if by inclination he doesn't wield it dictatorially but as a benign partner and participant). And if Jimbo says "this is my decision as project leader", then that's probably basically it. Right to leave and right to fork.

If one is after any meaning of "traditional role", that's my own understanding on the point.

However the reality as of 2009 is, Jimbo consults and discusses, but almost never these days directs; I cant remember any occasion in 2008 when Jimbo "ordered" ArbCom to do anything as such or even came close to it. His preference has been to gradually allow ArbCom to take over the role it now has, and that is roughly what the policy draft reflects. FT2 (Talk 20:52, 12 April 2009 (UTC)
 * re : arbcom being partly outside the community as a vehicle for Jimbo's day-to-day powers as project leader - I'm not sure either that this is the case, or that it should be. Presumably other arbcoms (for example the german) are entirely 'of' their communities? I think our dear arbcom probably should be / is too :-) Privatemusings (talk) 05:11, 13 April 2009 (UTC)

Delegation
"The Committee may, at its discretion, delegate any authority granted to it by this policy to another group of its choosing."

I've never liked clauses like this. Arbitrators are subject to a long process of community evaluations and approval. It stands to reason therefore that the community expects to be involved in deciding who is trusted with your authority. While handing off some functions to persons who were not so publicly vetted is likely to be fine, other functions may be controversial. You weren't elected to "pass the buck" as they say.

I'd like to see this expanded to either be more explicit in its scope (e.g. targeted to which powers and under what circumstances), or include explicit community consultation on hand-offs of power, or both. I'm assuming that the existing Committee has ideas for how they might use delegation, and perhaps it would be better to elaborate on those. Personally, I consider this unrestrained clause unreasonably broad. Dragons flight (talk) 23:12, 12 April 2009 (UTC)
 * Might be an idea to start by listing the possibilities for delegation. Currently, the committee has subdelegated within itself the ban appeals process, in an attempt to have most appeals (where the full committee is not needed) delegated to a subcommittee of three (in theory freeing up the time of the other arbitrators to concentrate on other things, such as cases). It has been mooted that this could be delegated even further at some point - but how far should this be delegated? The other delegation going on at the moment is the CU/OS audit function, where ArbCom are expected to review allegations of misuse of the CU (checkuser) or OS (oversight) tools. It was recently announced that there will be a subcommittee of three arbs and three editors to carry out this CU/OS audit function (three initially appointed to get things set-up, and then three elected once it is clear how many elections people will put up with and what timings would be best) - see here. I think those two functions are all that is or was on the agenda to delegate. It should be noted that in both cases ArbCom retain the authority to some degree, because they still receive a final report and can do a full review if needed. As for other delegations, when the update to arbitration enforcement procedures comes along (the recent RfC on that is being analysed at the moment), some delegation might possibly happen there, or, depending on what people want, arbs may get more closely involved with arbitration enforcement. The other thing being delegated, in this case to to the community, though still with some ArbCom input, is the selection of new OS and CU people (previously done by ArbCom following a period of consultation). There were elections on that recently, and depending on the review of how successful that was, similar elections will probably be held when more CU/OS people are needed. There is even some subdelegation going on with caseload, as we are now designating an arbitrator to be a point-of-contact on each case, along with the clerk(s) for the case (see Template:ArbComOpenTasks). There is only one other thing I can think of that might be covered by the wording you have mentioned, and that is passing on enquiries, and other matters that come to us, to other people to deal with. That is not so much delegating our authority, but saying "this isn't something we can or should deal with - go and see these people instead" (e.g. go and ask checkusers, oversighters, the admin who blocked you, etc.). There are also times when issues that date back many years come up again and we ask former arbs for advice (they may have been the ones around for the original case). I think we normally ask them to advise us, but there are cases where it can be easier to say "can you deal with this?". Is that an acceptable "delegation of authority", or is that just simple delegation of a task? In general, most of the time, delegation happens either because we don't have time to deal with something, or because we aren't the best people to deal with something. I guess the question is: do people want a consultation period before any major formal role is delegated, asking whether ArbCom should retain the role instead of delegating it? Carcharoth (talk) 01:18, 13 April 2009 (UTC)

.
 * I think almost no one would complain about internal delegation. In fact, until your reply, it hadn't crossed my mind that the delegation clause was even meant to encompass internal matters.  As long as the Committee of the whole can be called upon to appeal subcommittee actions when appropriate, I don't think anyone would be bothered by the Committee finding ways to better organize the workload internally.  It should probably be mentioned in the new policy, but I see no problems there.


 * The other extreme is delegating power to agents outside of the committee. AE is sort of an example of this, though not a very strong one since ArbCom hasn't historically ever done much enforcing.  One might imagine more blatant examples, such as giving people who aren't current Arbs final authority over ban appeals etc.  I would say that if one intends to delegate power to people outside of the committee (which is how I initially read the clause), then a process of community consultation ought to be expected and necessary.  Dragons flight (talk) 03:52, 13 April 2009 (UTC)
 * A long while ago, certain matters were delegated to the clerks. We were controversial then, but not so much now. I think the solution to the democracy/vetting problem is to elect arbs who are good judges of character. --Tznkai (talk) 04:20, 13 April 2009 (UTC)


 * May I suggest,
 * "The Committee may, at its discretion and with general consent of the community, delegate any authority granted to it by this policy to another group of its choosing."
 * as an alternative. --Barberio (talk) 23:31, 14 April 2009 (UTC)


 * How would that "general consent" be obtained? It is a fact of life that constitutions, rubrics, laws, cannot be infinitely fine-grained. It is a matter of practicality and workability that ArbCom be able to delegate powers to others it trusts. ArbCom itself is ultimately responsible if such delegates mess up; this is what happens in real life. However, in the case of the clerks, I believe their role is central to hearings and that it needs to be more closely defined. Tony   (talk)  15:29, 17 April 2009 (UTC)
 * How about
 * "The Committee may, at its discretion and with assumed consent of the community, delegate any authority granted to it by this policy to another group of its choosing."
 * ie, unless it's demonstrated that the community doesn't consent to delegation of an authority.
 * I'd also recommend the following also be added after it.
 * "The Committee will supervise the activities of these groups, and have final power to overrule or alter any delegated decision or action."
 * --Barberio (talk) 13:25, 19 April 2009 (UTC)


 * That may be usable, although it doesn't define what would demonstrate a lack of consent. We probably need to add some more material on ratification processes and so forth in the next draft. Kirill [pf] 16:09, 19 April 2009 (UTC)
 * Or just let ArbCom delegate as it sees fit. It is impractical to force ArbCom to jump through hoops every time it needs expert or skilled assistance. I do believe that (i) the avoidance of conflict of interest should be explicitly mentioned in ArbCom's general power to delegate, and (ii) that the clause should specify "person or group" (as a matter of practicality). There's also this wretched fluffy language: ""The Committee may, at its discretion, delegate any authority granted to it by this policy to another person or group of its choosing , provided there is no apparent or actual conflict of interest in doing so." Tony   (talk)  15:12, 24 April 2009 (UTC)

Expedited decisions and motions
Starting this section to fulfil a promise (from January 2009) to make sure this issue got discussed. Am contacting the three editors involved at the time, to invite them here to discuss the section of this draft policy describing expedited motions and decisions. Please could others also comment, so that there is an indication of what the feelings are about whether expedited motions and decisions are good or not. Some links to help get up to speed: It is that last discussion where I promised to bring this up again when it was being discussed. There are several examples of previous instances of expedited motions instead of a full case. Does anyone want a list? Carcharoth (talk) 02:00, 13 April 2009 (UTC)
 * Relevant section in draft policy
 * Previous discussion (1)
 * Previous discussion (2)
 * Previous discussion (3)
 * It seems to me that the third paragraph of the relevant section in the draft policy is all that is needed here. Whether a case would benefit from a speedy resolution is something that the members of the committee can make an informed judgment on, whereas whether they have all the pertinent facts is veering into Donald Rumsfeld's "unknown unknowns" territory. When the facts of a situation are genuinely "substantially undisputed" the case shouldn't be at Arbcom in the first place, so the request should be rejected. In cases where user conduct is called into question there certainly shouldn't be a facility to circumvent the editors' ability to put their case. Yomangani talk 08:57, 14 April 2009 (UTC)

Penalties for arbitrator violations
There are no penalties listed for violations. Arbitrators already don't follow the rules without penalty, so why bother tinkering with the rules? -- SEWilco (talk) 03:00, 13 April 2009 (UTC)
 * The penalty is found in the ballot box.--Tznkai (talk) 04:20, 13 April 2009 (UTC)


 * This made me chuckle. In all of history I think there has only been one Arb with more than 6 months active service to be re-elected.  Should we conclude therefore that nearly all Arbs do a terrible job?  One would have to better manage the burn-out and/or shorten the terms before the ballot box would be much of a deterrent to anything.  Dragons flight (talk) 17:35, 13 April 2009 (UTC)
 * I guess the job is the penalty then.--Tznkai (talk) 04:06, 14 April 2009 (UTC)
 * So there is authority with no responsibility. Why the pretense of the rules, then?  -- SEWilco (talk) 18:47, 20 April 2009 (UTC)


 * FWIW, arbitrator recall is on our agenda at the moment. Kirill [pf] 05:32, 14 April 2009 (UTC)
 * So the penalty would be losing the job sooner rather than at the next election? Getting less work doesn't seem like much of a penalty.  -- SEWilco (talk) 18:47, 20 April 2009 (UTC)

Private submissions of evidence, and the role of the Community
Surely all evidence should be made public unless there is good reason not to. That doesn't seem to be the internal logic of this section, which notes evidence that could be made public won't automatically be made public. Also, why isn't discussion of evidence done in public? There is still no attempt to link decisions to evidence, making decisions sometimes hard to fathom. Also, the arbitration committee does recognise it is answerable to the community, doesn't it? I can't really see that principle enshrined. For example, the committee is free to modify its internal procedures, and free to delegate any authority granted to it by this policy, but it appears the community does not have that right. Surely the Committee serves the community, and serves at the community's request and tolerance? I would have thought that was a part of the unwritten constitution, muchg as Jimbo's power is limited by what the community will tolerate, as has been evinced on other Wikipedias. I think if a new draft is being settled upon, there really should be some thought to how the power is structured. Currently arb-policy apparently needs "consensus among Committee members". Will we be reflecting that truly, it requires consensus in both houses as it were, the Committee and the community? Hiding T 14:28, 13 April 2009 (UTC)


 * I think more accurately, the committee is answerable to the project - that is, to the aim of creating a neutral free encyclopedia. "Wikipedia" per se is a project whereby a community writes a neutral encyclopedia of free knowledge, and the community is one part of the equation. If the community were to drift away from those goals (and many well-meaning communities do over decades drift away from their founding ideals) then the committee's obligation is to the goals of the project, rather than to "whatever whims might sweep the community". Thus in RFAR cases you'll always see the committee referring to principles of the project, but rarely or never to "this years" current moods and trends of the community, in guiding its decisions, unless those are seen to also serve the project's goals.


 * In that sense, the committee is not just there to serve the community without limit. It's there very definitely to serve the community in the community's goal of furthering the project -- and part of that service is to say when the community (in its view) needs reminding of those goals and principles, or they need enforcing beyond what the community is currently doing, or it's (in the committee's view) off track. And it's also there to serve Jimbo Wales as project leader, in its role of delegatee of some of his traditional role and powers. FT2 (Talk 16:13, 13 April 2009 (UTC)


 * <edit conflict, above text may have altered since this reply was initially composed>I'm not sure we're talking about the same things. I'm a bit baffled by the statement that the Committee can remind the community of its goals and principles. You seem to have taken things to a confrontational extreme, and it does beggar the question; what would be the outcome if the community told the Committee to get stuffed? Within the unwritten constitution, not a lot, unless the Committee has powers I am currently unaware of. I think you've conflated the Committee with the Board. The Committee's pronouncements and remedies are only effective as long as there is a consensus amongst the community to respect and implement them. That needs to be recognised in policy in some sense, surely? Also, the Committee is not there to serve Jimbo Wales as project leader. It's there to arbitrate on disputes. That's what Jimmy delegated to it. Jimmy delegated the responsibility for Wikipedias aims to the board, and responsibility for content was delegated way back when to the community. Hiding T 16:38, 13 April 2009 (UTC)


 * What would happen if the community tried to defy ArbCom? The community would lose. Any editor who defies ArbCom can be blocked: all arbs are admins.  Any admin who wheel wars with ArbCom can be desysopped: ArbCom can command the stewards. It is not possible for a community to exist in defiance of ArbCom, because ArbCom is, as FT2 says, ultimately answerable to the project above the community.  If ArbCom decides that an element of the community is acting counter to the goals of the project (and can convince the members of a suitably high-level group that they are right) then any necessary action can be taken.
 * Of course, that is not the point. The community might 'lose' such a war, but there is no way ArbCom or the Foundation could be said to have 'won'.  A project without a community is not going anywhere, and the need to retain an active and happy community must be weighed as one of the goals of the project which ArbCom is defending.  If the situation deteriorates to the point where ArbCom and the community have such radically different views that there is active fighting, something has gone very badly wrong.  Part of ArbCom's purpose is to nip such digressions in the bud, and keep the community aligned with the goals of the project.  As long as those interests remain aligned, any discussion of such "confrontational extremes" is entirely academic.  <b style="color:forestgreen;">Happy</b>‑<b style="color:darkorange;">melon</b> 16:52, 13 April 2009 (UTC)


 * Not sure I'm comfortable with either of these completely.
 * Hiding - the WMF board operates the project but not the content of Wikipedia; its role is the provision of an operational charitable framework, server and software, on which the community can exist, and set a handful of the very top level policies (privacy, NPOV, etc), etc. Most large systems have checks and balances; ArbCom is a check and balance as well as an operational part of the whole. As a check and balance, it can't simply reflect "whatever happens to be the trend"; sometimes it may have to act in opposition to a trend.
 * Happy-Melon - Arbcom isn't some kind of top level executive board in the sense that post might suggest. Its role is broadly to resolve problems that impact the project's editors, not to go round wholesale blocking and banning. But your general point is correct that extreme divergences are highly undesirable; what is a desirable or undesirable trend is very much case by case though.
 * Mostly though, the idea is to aim for the happy middle road, where we're all solidly behind the aims of the project and attempt to develop these in the best way we can, and with good collaboration skills. If that fails, ArbCom's "dispute resolution" role is basically, to fix it if the wider community can't. FT2 (Talk 17:03, 13 April 2009 (UTC)


 * Concentrating, for the moment, on the relationship between the Committee and the community: there's a distinction that should be drawn between the constitutional basis of the Committee's authority and the practical aspects of the Committee's work. It's one thing to say that, in principle, the Committee is subject to the community; it's another to try and have the community as a whole micro-manage what the arbitrators do on a day-to-day basis.  The driving force behind §1.3 and §1.4 is the need for us to be able to make routine changes to our internal roles and procedures so that we can actually get things done; if we must to conduct a community ratification vote every time we adopt a mail sorting process, or if the community starts making changes to our workflow on a whim, things will simply grind to a halt. Kirill [pf] 04:38, 14 April 2009 (UTC)
 * I've got no argument with that. But surely there's got to be a better way of framing that.  Perhaps something like "To make routine changes to our internal roles and procedures as necessary to ensure the smooth running of arbitration." That wouldn't suffer from degradation. But I agree with the comment at the bottom of the page, this is far to legalised and technical.  We've always tried to avoid that approach on Wikipedia if possible, because it makes things too complicated. But I also wanted to focus on the issue of evidence here too.  Is there any reason evidence submitted privately should be kept private without good reason? Surely cases are held as publicly as possible, rather than as privately as possible. Which one would be the arbitration committee's preference? Hiding T 09:18, 14 April 2009 (UTC)


 * Maybe we could combine 1.3 and 1.4 into something like "The Committee may, at its discretion, modify its internal procedures, form subcommittees, or designate individual arbitrators for particular tasks or roles, as necessary to ensure effective operation. Where appropriate, the Committee shall allow a period of community consultation for proposed changes before adopting them."?
 * As far as the evidence is concerned, we do prefer things public unless there's a need to have them private, hence the "The Committee may request that privately submitted evidence be posted publicly if there is no reasonable cause for it to be private" clause in 2.10.4. Perhaps simply stronger wording there would be sufficient? Kirill [pf] 12:29, 14 April 2009 (UTC)


 * Decisions are already made without evidence nor discussion. In what useful way is formally hiding evidence different from no evidence, no known evidence, or no discussion?  -- SEWilco (talk) 18:50, 20 April 2009 (UTC)

necessary jimbofication?
The jimbo count on the proposed draft is in double figures, with ten mentions! - I'd very much like to encourage a sensible process for measuring wide community input into the merits and wisdoms of this level of jimbofication. Personally, I'd like to see Jimbo as a constitutional monarch - the role he's also identified himself as the best fit. This draft as written doesn't really match - with the appearance of rather wide-ranging discretionary powers. I feel sure the english wiki arbcom can function along similar lines to the german arbcom, without a single person holding any special power. I would edit the draft to reflect this, but can't (that's another discussion for another day ;-) Privatemusings (talk) 22:47, 13 April 2009 (UTC)
 * You could add a footnote here. How many mentions of Jimbo are in the current policy? Carcharoth (talk) 23:41, 13 April 2009 (UTC)

Cut it up
This is creepy. It's far too long, it's couched in language that's too legalistic, and it's at once obsessively focused on procedural minutiae, yet way too broad in scope. A policy document should be just that: a statement of the irreducible policies that govern ArbCom, unencumbered by anything else.

Chop it up. I think you need several short documents, not one long one — with each document serving a clearly-defined function and addressed at a clearly-defined group of users.

For example:


 * 1) Overview of ArbCom (introductory and historical material that gives background, intended audience: someone who doesn't know what ArbCom is)
 * 2) Policies (the things ArbCom couldn't change, or would never change, intended audience: participants at ArbCom)
 * 3) Custom & Practice (the way ArbCom does things at the moment, but could change if it saw fit, intended audience: participants at ArbCom)
 * 4) Procedure (step-by-step breakdown of a typical case, with timescales, intended audience: participants at ArbCom)
 * 5) Scope (i.e. What ArbCom Is Not, intended audience: someone who's thinking of bringing a case)
 * — S Marshall  Talk / Cont  00:54, 14 April 2009 (UTC)


 * Well, I think that it's a fairly safe assumption that anything which ends up in the final "arbitration policy" document will essentially be set in stone for the foreseeable future, while anything that's considered merely a Committee procedure will remain significantly more open to change. This has, to some degree, guided the choice of contents here; we've tried to put in those items which we thought should be fixed.


 * There's certainly a reasonable argument to be made for omitting some of the more technical and procedural items in the text (e.g. the acceptance criteria in "Voting on acceptance", the activation procedures in "Active arbitrators", and so forth) from the formal policy and handling them as procedures, and concentrating the core policy on the duties and authority of the Committee, and the rights and responsibilities of editors in arbitration; but I'm not sure whether the community would prefer that approach.


 * What do people think? Should we reduce the amount of procedural detail in the policy, putting things in terms of rights and responsibilities instead?  Or stick with the current approach? Kirill [pf] 05:16, 14 April 2009 (UTC)
 * I'm no fan of the current approach, I think it's fair to say. Hiding T 09:19, 14 April 2009 (UTC)
 * Nor me, though splitting it up isn't necessarily the answer either. I certainly think we need to discuss carefully the fundamental principles behind all this and decide what we want to achieve. Above all, people should know what sort of complaint is appropriate to take to ArbCom. My suggestion above - which AFAICS no-one addressed either positively or negatively - was that it should be only complaints about specific actions (or failures to act) of admins and other "officials". Ordinarily, if you've got a complaint about someone's behaviour, then your first port of call (preceding ArbCom, anyway) should be the admins' noticeboard. Others probably disagree with that would-be principle, but surely we can make some effort to answer this question. This seems a far more essential sort of thing to get right than the minutiae of procedure which we seem to be concentrating on at the moment.--Kotniski (talk) 11:28, 14 April 2009 (UTC)
 * Well, that's the point of having a "What ArbCom is not" document.— S Marshall  Talk / Cont  12:56, 14 April 2009 (UTC)
 * OK, one thing we have to decide is what ArbCom is not; another (less important) thing is whether that information is best presented on a separate page from other information. I agree with your proposed list of topics that need to be addressed, though I think the presentational question of how many separate pages are needed is not one we need worry about yet (we need to know what the information is before deciding on how to present it). I would like to make a formal proposal about Scope and about the dispute resolution process in general, along the lines set out above at, but since I have relatively little experience in this area, I'd like to get some concrete feedback about these ideas from those who do.--Kotniski (talk) 13:12, 14 April 2009 (UTC)


 * Sorry, I do not at all agree with the idea of "cutting this up", nor that its tone or register are over-legalistic. The community and ArbCom itself needs to know where it stands, and a whole-of-picture document that sets things out clearly, simply, and precisely, will serve this purpose best. The term "legalistic" is a pejorative term for these qualities. The argument has not been made as to why the document should not embrace them. Tony   (talk)  15:12, 17 April 2009 (UTC)
 * It should not embrace confusing terminology because it then becomes unclear to readers what is being said. Try and parse "In exceptional cases of communal division or exceptional risk of harm to the project, the Committee may determine a interim means to achieve the aims of communal policies or to curtail dispute in the area, until the community has developed a better consensus." I'm pretty sure words four to fourteen are irrelevant. From word fifteen on, the intent become muddy. If they really want to say "In exceptional cases, exceptional action may be required", fine, but don't dress it up in jargon. I'm not going to attack the odd typo, but you've got aspirational grammar with no definitive point to it. "may determine"? "achieve the aims"?  You are absolutley right, we need a whole-of-picture document that sets things out clearly, simply, and precisely.  We deserve so much better than, and I apologise here, corporate management speak.  Could we get much closer to "the arbitration committee will provide blue sky thinking and forward thinking methods for thinking out of the box while ensuring longstanding goals are delivered within a reasonable time frame, allowing for input from the wider community where their opinions impact in the day to day implentation of consensus." Hiding T 17:10, 17 April 2009 (UTC)
 * Agreed: the whole text needs a plain-English copy-edit. See the footnotes appended by several users to the draft. Tony   (talk)  16:01, 19 April 2009 (UTC)

"Jurisdiction over actions of Jimbo Wales"
Why is this limited to just his admin/steward actions? I strongly recommend this extend to any of his actions involving the Arbcom itself, so that the seated Arbcom firmly state that they can override his actions in regards to them--which they already have the power to do, as Jimbo cannot supercede any Arbcom decision anymore. For example, in the current wording of this document, Jimbo could remove an Arbcom member for "reasonable" cause, but without the approval of the Arbcom. Or, he could place someone on the Arbcom without the approval of the community or Arbcom itself. Jimbo is one person, and a person is never an institution. The Arbcom and Wikipedia are separate institutions; Jimbo is neither Wikipedia nor the Arbcom. This subtle wording change is the perfect opportunity to start defining boundaries for long-term success. What do you say, Arbcom? <span style="color:#0D670D; font-family:Georgia, Helvetica;">rootology ( C )( T ) 21:56, 14 April 2009 (UTC)
 * Jimbo can get up one morning and abolish ArbCom in its entirety, right? As I understand it, ArbCom only exists because Jimbo delegated certain powers to it, and he retains the right to reverse that. In any case, if we're going to discuss what ArbCom might or might not do in respect of Jimbo, we ought to ensure the participation of the great man himself.--Kotniski (talk) 13:54, 15 April 2009 (UTC)
 * I addressed a similar question a while ago, but I'll bottom line it. We don't know who has the ultimate wiki-legal power. We don't want to know. Wikipedia is not a bureaucracy, nor a government. We gave up no natural rights on Wikipedia, and we took no civil rights in return. If the "Jimbo question" is ever provoked and subsequently contested, it will be moment of complete and total failure of Wikipedia governance.--Tznkai (talk) 14:21, 15 April 2009 (UTC)
 * Well said Tznkai. I should make one more point about Jimbo. In a TV interview, he answered many of a reporter’s questions with “whatever the community decides”, and “whatever the community consensus might be on that matter is best”, and “I have complete confidence in ‘community consensus’&thinsp;”. Jimbo’s faith in this bedrock principle is absolute . Tznkai’s last words, above, are powerful and worth repeating. If there was ever a confrontation between Jimbo and Wikipedia’s governance, it would, in my opinion, be due only to a colossal failure of our governance that had somehow undermined Wikipedia’s capacity to arrive at a community consensus, to interpret that consensus, and/or to implement it. Conjecturing about Wikipedia needing jurisdiction over Jimbo—besides (frankly) striking me as being simultaneously somewhat brash and naive—springs principally from a lack of understanding of Jimbo’s values. If he ever had to step in, it would be to restore power to its citizens. Greg L (talk) 23:15, 24 April 2009 (UTC)

I second this endorsement of Tznkai's statement (mostly). It's an odd situation—to use Jimbo's own analogy, rather like the Queen's relationship to Parliament. She could in principle take dramatic action (sack the government; refuse to sign a bill into law) or less dramatic action (pull strings behind closed doors). I suspect that Jimbo does do the latter, but until I know otherwise, I'm willing to trust that this is OK. His trumpeting of his "complete confidence in 'community consensus'&thinsp;" is the kind of line any politician would take; it is rhetoric we should not mind as the public face of a system that is order within chaos.

We should be eternally thankful for his:
 * setting up of WP's pillars the way he did;
 * resisting of the poison chalice of advertising, and
 * being a pretty good PR front guy.

Without these gifts, he might not have retained the powers he has, even though there is community discontent with the dysfunctional admin system and ArbCom hearings process he has allowed to grow (we need to grapple with these soon). The relationship may evolve into one in which he has less power over ArbCom—in particular, a winding back of his power to appoint arbitrators outside the electoral process—but give it time. The truth he knows is that significant community discontent with his role would probably force him to accept change. I don't see that at the moment. Tony  (talk)  04:01, 25 April 2009 (UTC)

Consistency
Will the Committee be updating the other pages related to the policy when it puts forward the final policy version to ensure consistency? Pages such as:
 * Arbitration rationale
 * Arbitration Committee
 * Arbitration policy/Case handling
 * Requests for arbitration/How to present a case
 * Arbitration guide
 * Arbitration Committee/Procedures
 * Arbitration policy/Procedure for changing this policy

Put otherwise, is someone making sure we don't have contradictory guidance floating around at the end of this process? Thanks.  MBisanz  talk 06:44, 18 April 2009 (UTC)
 * Noted, thanks. Before any final version goes live (and the next iteration will probably be put out for comment again - though I'm not certain about what comes next here), those pages should indeed be updated, and hopefully this will not be neglected or forgotten. Carcharoth (talk) 10:14, 18 April 2009 (UTC)


 * 1) Why is this list not centrally located so that everyone can easily gain access to all text related to ArbCom?
 * 2) The Clerks policy page is missing from this list.
 * 3) Looks like a case of bloat to me. Can't some of them be rationalised?
 * Yes, they all need to be updated, rewritten, copy-edited, and put here for community comment. The language is pretty bad throughout. Tony   (talk)  16:48, 19 April 2009 (UTC)


 * We're working on it; see WP:AC/N. ;-) Kirill [pf] 17:38, 19 April 2009 (UTC)

Continuation of the expansive model for hearings is untenable
The model for ArbCom hearings was established as expansive and all-inclusive. It appears to be implicitly premised on two notions:
 * (i) that the body of evidence and opinion on which arbitrators form their judgements should be subject to minimal restrictions in scope and size; and
 * (ii) that participation in the process should be open not only to the named parties, but to the whole town square.

While I have no doubt these notions were advanced in good faith, in practice, bloat and irrelevant material have become major barriers to the good management of cases and to the delivery of fair and prompt judgements. Any case beyond the very simple can easily get out of control—witness the dates case and its humungous volume of hogwash, and worse, of its encouragement of parties to poke their tongues out at each other, to bicker, to accuse, to carry on their dispute with renewed vigour on ArbCom’s pages themselves. This damages the reputation of the hearings process and of WP itself, and makes the task of the arbitrators unreasonably difficult. It leads to lengthy cases and a backlog for ArbCom that could easily affect the morale of the Committee.

The expansive, inclusive model is consistent with some of the statements in “important features”; not only this, the all-important section shows an uncertain, internally contradictory understanding, and a vagueness that is embarrassing if the hearings process is to be taken as the peak instrument for resolving disputes at the world’s most important information site. This garbled advice apparently underpins the structure and management of ArbCom hearings:
 * “Arbitrators focus on the risk and benefits for future, not on past historic issues.” [The grammar is so mangled that I have to think hard about what this actually means.]
 * But then we learn that the past is important: a few lines later, “old incidents may not be actionable but can sometimes show a persistent history of problems.” [Many people would question the focus on “old” incidents, and this statement seems to contradict that focus on the present and the future.]
 * Then, “A person's general manner is probably evidence of their likely behavior”, and “insightful impressions by reasonable people may be valuable, even if just "impressions"”. [The use of vague terms such as “general”, “likely”, “impressions”, and “may”) adds nothing to anyone’s understanding of the process, and is in conflict with the task of carefully weighing up of evidence. This sounds like an unwise elevation of hearsay and unevidenced impressions.]
 * “Arbitration is more likely to ask if a user can change or what restrictions would have an effect, than "who said exactly what"” [But hang on ... up to 100 diffs per participant are allowed of who said what to whom, when].

The oversized structure and laissez faire approach to management are an invitation for cases to spiral out of control, or at least to be far bigger deals than they should be. Arbitrators seem to be swamped by text. It is unruly, undisciplined, often chaotic. This is the opposite of what is required for an efficient, fair process.

Below, the four elephants in the living room are outlined. They need to be addressed in this re-write of the policy.

It's time to dispense with the Workshop page
The Workshop idea appears to have been the idea of Fred Bauder; I don't know the details, but its inclusion was possibly a decision made in haste. The page is nothing more than an invitation to admit bloat and irrelevance, and to attack. A senior admin recently informed me in writing that: "'several arbs [have said] not to waste time on workshops, nobody's gonna read them anyway ... not once they're long and messy'." Having four separate pages—Request, Evidence, Workshop, and Judgement—is excessive and needs to be reduced. Why is the Evidence page insufficient to provide the evidence required to make a sound judgement? Just why anyone but the arbitrators should be proposing "general principles", "remedies", and "enforcement provisions" is unfathomable. There should be an application, evidence, and a judgement—not brouhaha in the village square. Take a look at the Workshop page for the "dates" case.


 * Well, I've always found the workshop to be invaluable on cases I draft; where else would I do the actual drafting? It's a fair point that participation there should be more carefully managed—you might want to take a look at what Fayssal is doing on the Ryulong case for one approach we're considering—but getting rid of the page entirely would inconvenience the arbitrators more than it would help the parties. Kirill [pf] 16:02, 19 April 2009 (UTC)
 * Why do arbitrators need to be told by the parties and whoever wants to fly by what principles are stake, and what enforcements and remedies are appropriate? Don't we elect them to do that? Don't we put our trust in them? What kind of convenience is the Workshop page for the Dates case? Tony   (talk)  16:42, 19 April 2009 (UTC)


 * You're talking about who does drafting; I'm talking about where drafting gets done. Even if non-arbitrators were barred from commenting entirely, the arbitrators themselves would still need the workshop (or some equivalent page with a different name) as a place to draft decisions before proposing them for voting. Kirill [pf] 17:40, 19 April 2009 (UTC)
 * That would be fine—parties present the evidence; arbitrators draft the decisions. Arbitrators might ask for clarification on the evidence page, but not the other way around. These parts of the process really need to be kept separate. At the moment, much of the drafting is done privately by the arbitrators, anyway. That is not a problem to me. Tony   (talk)  18:02, 19 April 2009 (UTC)
 * Alternatively, if the current Workshop system is maintained, I think clerks/Arbs should do a better job of explaining and regulating what and what not would be appropriate for Proposed principles, FoFs, and proposed remedies. For example (taking it from the dates case), Requests for arbitration/Date delinking/Workshop is just plain silly. A better and more unclear example is Requests for arbitration/Date delinking/Workshop—this seems like a content ruling, and is it something that Arbcom could rule on regardless? Dabomb87 (talk) 03:41, 20 April 2009 (UTC)
 * The dates case has clearly shown that the workshop page is a useless duplication at best (where parties are unsure about where to post something, they do it in both places) and at worst, it is an open forum for a disruptive slanging match between parties where they can and do throw the kitchen sink at each other. How parties' submissions here can help Arbcom arrive at a resolution when one or both sides baying for blood of the other side (ie exacting revenge) by proposing tit-for-tat sanction motions is yet to be demonstrated. Sure, having been through one case doesn't make me an expert, but this case has been particularly acrimonious, and has brought out the worst in everybody that nobody (including Arbcom and its clerks) comes out looking good, so I think we could use this as a benchmark for a "doomsday scenario". I take note of Kirill's point, and would simply state that this can be the working page of Arbcom should it wish. I would encourage/welcome that additional transparency in the proceedings. However, if it is Arbcom's continued wish to work on various elements of judgement off-line, WORKSHOP's only rightful place is the waste receptacle. Ohconfucius (talk) 14:37, 20 April 2009 (UTC)

It's time to drop the peanut gallery
It is eccentric that a quasi-judicial process should allow anyone who wishes to throw paper planes into the room. It has the capacity to make the proceedings undignified and threatens any focus that might have been achieved if participation had been restricted to the parties, the clerks and the arbitrators. The current dates case is full of irrelevant comments from users who want to have their two cents' worth, with apparently little or no control over relevance, standard of proof, hidden and not-so-hidden agendas, or potential libel. This is not only a matter of retaining focus, but of procedural fairness and dignity, and making arbitrators’ lives liveable, without outsiders' licence to fling in drive-by comments without constraint.

Possible exception. Occasionally, a user who has not been named as a party in the application may have cause to make a useful contribution. I believe this should be allowed only by open application to the clerk, who should be convinced that that the user will make a bona fide and relevant contribution before allowing outsider comments or questions.


 * See my comment above about managing participation. Kirill [pf] 16:02, 19 April 2009 (UTC)
 * Which comment in particular? Can you lay out specifically what you mean in relation to the peanut gallery? Tony   (talk)  03:09, 20 April 2009 (UTC)
 * The one about Fayssal; see Requests for arbitration/Ryulong/Workshop (and related pages) for what he's working on. Personally, I don't believe that we'll be making any major changes to how cases are conducted without trialling them first; this happens to be our first attempt in that regard. Kirill [pf] 03:19, 20 April 2009 (UTC)


 * Comment - I've found in more than a few cases that the most impartial, factual and/or complete evidence has been presented by people who are not directly involved in the dispute. Timelines, documentation of edit/action patterns, identification of impacts into other areas of the project that may not be immediately apparent to the involved parties, are all things that have been brought to the table by non-involved parties, and all have been useful to the present and previous iterations of the Committee in coming up with the best solution for the encyclopedia. One must keep in mind that, when a dispute gets as far as an Arbitration Committee case, it's almost never just an issue between a few editors, it is something that impacts the project itself. Risker (talk) 13:25, 20 April 2009 (UTC)
 * This is useful, and right at the opening? Not that I want to single out one third party, or to imply that the parties themselves didn't write a huge amount of tripe.
 * I'd like you to consider the following points in response to your post:
 * The Arbitrators are elected to be impartial judges, and a great number of them typically contribute to the collective wisdom of the Committee.
 * I have proposed (below) that third-party contributions be allowed if accepted after open application to the Clerk, on the basis that the third party has a relevant and useful contribution to make. I believe this is an efficient method for admitting (indeed, even encouraging) the valuable, impartial, factual evidence you refer to from non-involved parties. Of course, my assumption is that the scope has already been stated (and explicitly modified if necessary); without that, all is lost. Tony   (talk)  13:45, 20 April 2009 (UTC)


 * Whilst those with demonstrably valuable input may be welcome (per Tony), admitting tricoteuses is an invitation to irrelevance and bloat. In the dates case, there were enough parties alone shouting "off with his head", tricoteuses should thus be banished. Ohconfucius (talk) 14:46, 20 April 2009 (UTC)


 * A neutrally developed timeline is helpful in almost every single case in which it appears, and in fact I would like to see them in more cases. I do not want to put any obstacles in the way of neutral parties developing and then submitting such information, and I cannot see a requirement that clerks approve evidence submissions to be anything but a problem.
 * Non-party submissions tend to fall into two categories: personal opinion without useful information, and personally gathered useful information. The first category is unlikely to have any effect on the outcome of the decision; the second is amongst the most useful information the Committee receives. I explicitly want to encourage the latter type of evidence submission; otherwise, committee members wind up having to do that work ourselves, when others are often in a better position (time-wise and skill-wise) to do the task. Risker (talk) 15:02, 20 April 2009 (UTC)
 * Risker, this is helpful information. However, most third-party contributions are of the clearly not helpful kind, and can easily be dismissed by a Clerk. On the occasions where useful information or detached opinion born of much experience might be presented for inclusion, there is no practical impediment to the Clerk's passing this by an arb. before agreeing to its admission. Clerks are perfectly able to distinguish between the obviously no, the obviously yes, and the grey areas where they need to pass it by the Committee; that is partly what Clerks are for—to filter for you, the arbs. Putting in place a benchmark for third-party inclusion will inevitably tend to act as a partial filter in the first place. The expansive model is hampering ArbCom from achieving its potential; it needs to hear more cases in a leaner, more focused and efficient process, more promptly. I am proposing just a few basic rules that will help the personnel to achieve this. Tony  (talk)  04:55, 21 April 2009 (UTC)
 * Speaking as somebody who is experiencing the "peanut gallery" in an ugly case right now, I would at a minimum ask for a rule that enforces a basic piece of common sense: that only people should be allowed to comment on arb pages who can be expected to have some relevant input to give. We are seeing a situation right now (at WP:ARBMAC2) where a large proportion of activity comes from newbie editors who only registered their account after the case started. At a minimum, I think that a legitimate participant in arb discussions should be either (a) a genuine involved party, i.e. somebody who has substantial prior history of taking part in the dispute, or (b) an experienced well-established editor in good standing, with some demonstrable prior experience with Wikipedia dispute resolution. Of course, that would still not exclude a lot of unproductive noise from established users, and I don't really know how to distinguish procedurally between the positive and the negative types of input from among those, but the newbie meddling really should stop. – To Risker: are you sure the irrelevant noise types of contribution are really "unlikely to affect the outcome"? Of course, I'm sure you regularly ignore them, but are you certain you aren't also missing out on noticing contributions that would actually be valuable, but which are drowned and hidden in the noise from the others? Fut.Perf. ☼ 09:49, 1 May 2009 (UTC)

Need for explicit scope and the active control of relevance
A key part of the expansive, inclusive model of ArbCom hearings is the avoidance of stating the scope at the start. It is unacceptable that neither parties, clerks nor arbitrators know the boundaries of the scope until the arbitrators finally sort out a judgement. This precludes proper action by the clerks to enforce relevance on the Evidence page (probably by liaising with a presiding arbitrator for each case). Why is there no requirement for the original filer to state their preferred scope, and the explicit expectation that this might then need to be negotiated with the clerk / presiding arbitrator. Better it be done at the start than after chaos ensues. The fact that ArbCom can, at its discretion, decide on matters beyond the behavioural makes it yet more urgent to be transparent from the time of accepting a Request. It would be much less work for the arbitrators to decide on a statement of the scope at the start than deal with tons of sludge later. I believe ArbCom routinely deals with four or more cases at once: that is crazy.


 * I'm not sure how feasible fixing the scope at the start of the case would be; in many cases, new issues emerge during the investigation. Possibly something like a regularly updated list of matters being investigated would be more useful? Kirill [pf] 16:05, 19 April 2009 (UTC)
 * The "new issues emerge" during the investigation precisely because the scope is not fixed at the start, surely? Frankly (though these comments are based only on the one ArbCom case I've had any significant contact with), the whole thing is really the most appalling mess, where no-one has any idea what's going on or whether it's worth spending their time on trying to influence anything. Unless you actually like extreme drama (which I suppose to some extent you must do if you agree to sit on ArbCom as it currently works), you must surely see that significant changes are necessary, not just tinkering with the wording of the policy. If someone "brings a case" (makes a complaint, in more normal language) to ArbCom, they should say against whom they are making it and what the complaint is, and why they feel the need to bring it to ArbCom and not (say) to the admins' noticeboard. As I've already said several times, I can't see any reason why complaints against ordinary editors should be brought to ArbCom rather than to the admins - but even if you think there are such situations, the complaint should still be specific in terms of who and what is being complained about. If "other issues emerge" - i.e. people start complaining about other things - then they should be told to make separate complaints through the appropriate channels. --Kotniski (talk) 16:30, 19 April 2009 (UTC)
 * In response, I don't think that new issues emerge solely because the scope isn't set. It can be and several times is the case that there have been other actions meriting ArbCom comment which are not included in the original matter, but which are related. These could include misconduct by the filers which are not included in the original comments, evidence of involvement by parties not mentioned in the original filing, etc. While I can see how it might be useful to limit the scope up front, and how it would certainly make things easier, it might also obstruct justice regarding other matters. The members of ArbCom aren't really in a position to open requests themselves, so with a strictly limited scope they would be basically unable to do anything about misconduct which becomes visible during the examination of a case but which is not directly included in that case. As the bereaved parties may in some cases have left wikipedia since the incidents, there is potentially also a lack of parties to file such a complaint. Having a undetermined scope allows such actions to be addressed, while a tighter scope might not. John Carter (talk) 19:16, 19 April 2009 (UTC)
 * "The behaviour of X and Y in relation to [matter] since [diff]."? "Since" keeps going until the judgement. This is a red herring, John. All serious judicial/judging processes have a scope. This idea that "things might pop up during the case" that need to be covered by the scope is sheer nonsense as an argument for not stating the scope at the start. Tony   (talk)  03:08, 20 April 2009 (UTC)
 * Well, it's not quite that simple. Suppose that we have a case brought concerning the conduct of user A, and so we open it with a scope like "behavior of A since [time]".  It's quite common for A, when presented with findings against them, to plead extenuating circumstances—for example, that user B, whom they were attacking, was harassing them elsewhere.  We would be negligent if we did not investigate this claim; but B's conduct is not covered by the scope, so we would be left with, as John proposes, modifying the scope during the case based on what comes up.
 * I'm not certain that there's really a good solution here; fundamentally, if we don't have a flexible scope, we'll be forced to turn a blind eye to misbehavior on the technicality that it was not mentioned in the original complaint. Personally, I don't think a fixed scope is worth it in this scenario; obviously, your opinion may differ. Kirill [pf] 03:27, 20 April 2009 (UTC)
 * The solution is to require A (the filer) to state a preferred scope upon application to hear the case. That is what happens in the real world. In the unusual case where significant other aspects need to be judged, the scope can always be changed, but it should be done explicitly and in writing, so all parties and the community can be in no doubt. That is the whole point: swimming in porridge, where no one, not least the arbitrators, has a clear idea of the scope, is nonsense. If the scope changes, we all deserve to know. One would hope that usually the scope doesn't change, but I understand why it might. Unfortunately, it's hard to know what proportion of cases have and have not involved a change of scope, to guide us in this respect, because ... um ... no one knows what the original scope was. The hearings need to be open, transparent and clear. Tony   (talk)  03:37, 20 April 2009 (UTC)
 * I think Tony is spot on: without a tightly defined scope, we will get into situations where we might ask questions like "how bad is bad", or "how long is a piece of string". It's those typically open-ended situations which inspire the filibusters and malingerers (and guarantee proliferation of drama). The dates case, now in its third month, has not yet set a record. However, it probably would have been over by now had the three elephants above not been inhabiting our living room during this time. The obvious advantage of banishing these pachyderms is that it would cut the case time and hence workload of Arbcom members. Ohconfucius (talk) 04:46, 20 April 2009 (UTC)
 * I don't see how pleading extenuating circumstances would cause a breach of the defined scope. If, following a complaint about A's behaviour, A seeks to justify it with reference to B or C's behaviour, then that's perfectly within the scope of the complaint. But if the "case" becomes a focal point for any D, E and F to start complaining about any X, Y and Z's behaviour in vague relation to a particular topic, then something's clearly broken down. (And once again, admins are perfectly capable of dealing with these complaints between editors. I'm still waiting for someone to tell me why ArbCom needs to accept complaints about other editors' behaviour, as opposed to complaints about admins' handling of such complaints.)--Kotniski (talk) 06:03, 20 April 2009 (UTC)


 * The topic of the case should be stated. "Reopen" and "actions" are hardly descriptive.  See |here, where Arbs refused to follow the requirement that they explain why they voted, and participants (including Arb) didn't know what was involved.  -- SEWilco (talk) 19:12, 20 April 2009 (UTC)

Better organisation
The Clerks need to be more active, which is a good reason for them to be highly sensitive to issues of conflict of interest, and for each case to be allocated a presiding arbitrator, who can liaise between the Committee and the Clerk, if necessary. There needs to be a time limit for the submission of evidence, regardless of the perceived complexity of the case, and lower word and diff limits for each party's submission. Being able to predict when the Evidence page is complete would allow basic organisation of ArbCom's workflow. Arbitrators are busy people, and should be given the courtesy of being able to plan a few weeks ahead.


 * This is already being done as a procedural matter. I don't think how the Committee allocates cases among arbitrators internally should be covered in the policy, beyond the mention in §1.3. Kirill [pf] 16:03, 19 April 2009 (UTC)
 * If it's already a procedural norm, can it be written on the application page and elsewhere? Tony   (talk)  16:50, 19 April 2009 (UTC)

Summary of recommendations
Tony  (talk)  15:46, 19 April 2009 (UTC)


 * I've commented in the individual subsections above. Kirill [pf] 16:05, 19 April 2009 (UTC)
 * I completely agree with Tony's suggestions, at least as a minimum as regards what needs to be done.--Kotniski (talk) 16:35, 19 April 2009 (UTC)


 * Tony, I would encourage you (and other interested parties) to read Wikitruth through Wikiorder which is an academic paper reviewing the history of ArbCom through 250 decisions. You appear to be trying to make ArbCom "better" by making it more closed, rigid, and closer to real world judicial proceedings.  I would argue, in line with the central thesis of the cited paper, that much of ArbCom's success stems from the fact it is fundamentally not a process of law.  Given the nature of the wiki-environment, I believe the community is better served by the open and flexible process than it would be by one that was more rigid and closed.  Some cases do get a bit out of hand, but I think the answer to that is better management to address specific problem cases and not broad global restrictions.  Dragons flight (talk) 03:46, 20 April 2009 (UTC)
 * So how do you propose ensuring this better management? If ArbCom itself has failed to provide it, then the community has to guide it in some way.--Kotniski (talk) 06:17, 20 April 2009 (UTC)
 * "A bit out of hand", Dragons flight? The dates case is a farcical explosion of chaos, confusion, un-wiki angst, and time-consumptive crap. Several others I've seen also approach this level of festering uselessness, e.g., the repeated one - I think it's been here three times now - about TV show episode/character articles and related WikiProject controlfreakism and "microconsensus" that tries to carve out special niches immune to WP-wide consensus. I have to agree with Tony1 on virtually all points raised here. The only one I recall feeling a need to concede on is that ArbCom itself would still benefit from a workshop page, provided that outsiders were not allowed to edit it; the workshop as currently operating certainly should be "binned". And especially, without scope being explicitly limited from the start (with the proviso that the ArbCom can expand it as genuinely necessary) ArbCom cases are simply an exercise in forum shopping for new venues to tumidly re-air the same dirty laundry, practically endlessly. —  SMcCandlish  &#91;talk&#93; &#91;cont&#93;  ‹(-¿-)› 10:34, 20 April 2009 (UTC)
 * As a general principle, making global policy based on the worst possible examples leads to bad policy. In a majority of arbitration cases I've followed I'd say that the things Tony is against (i.e. open evidence pages, open workshops, and flexible scope) made the cases better not worse.  The Dates case is a monster, and yes it would be nice to do something to improve that.  However, I look at it something like the 2000 US presidential election.  The Bush v. Gore situtation and confusion in Florida got way out of hand, but the solution to that is not to curtail / restrict all elections.  I'd encourage people to look at less nuclear and more targeted solutions.  For example, rather than banning all people from editing the workshop or all non-parties from providing evidence, give Clerks the ability to ban disruptive individuals from editing those pages  Dragons flight (talk) 15:30, 20 April 2009 (UTC)
 * Clerks and Arbs already have that power, and I have argued that it wasn't properly exercised. One could take one step further back and argue Arbcom should have dismissed the case as 'forum-shopping'. The case has damaged WP and brought disrepute to Arbcom. However, I hope we can learn from the 'monster' case, deal with the structural problem and not get sidetracked about problems which were related to the workings of a flawed structure. Ohconfucius (talk) 03:39, 21 April 2009 (UTC)

Dragons, I believe your post is in good faith; however, here are my responses. Chaos and time sink. Is "a bit out of hand" code for "chaotic"? The article you cite (Hoffman & Mehra 2009) points out in its first paragraph that the Darwin's/Lincoln's birthday case "generated over 30,000 words in evidentiary submissions and thousands of volunteer man-hours". They seem to be making the point that there's an awful lot of huff and puff for a trivial matter. (And sorry to be a logic-snob, but how does an ArbCom case "generate" man-hours?) But wait, we can beat that hands-down with the Dates case: A legal process or not? Now, you say "much of ArbCom's success stems from the fact it is fundamentally not a process of law". That's odd; the opening page of the article you cite states that the "Arbitration Committee ... hears evidence presented by the parties6 before issuing findings of fact and conclusions of policy and law". (My italics in both quotations.) Within seconds, the authors state: "[ArbCom's] decisions, organized by volunteer arbitration clerks and disseminated by volunteer reporters, have created a virtual Wiki-common law." Oops. Sounds pretty much like a process of law to me. Flawed article. Problems leap out at the reader all over the place in that article. For example, Footnote 6 says "And sometimes the parties’ advocates". Tell me, to what extent were PMAnderson, MBisanz and many other self-appointed commenters "advocates" of the parties in the Dates case? But more fundamentally, ArbCom hearings are clearly described in the article (three times by the end of the second para) as a "dispute resolution" process, yet—oh dear—in the abstract, we are told "In effect, the system functions not so much to resolve disputes and make peace between conflicting users, but to weed out problematic users while weeding potentially productive users back in to participate." My head is spinning; and ... weed back in? Pardon me. False binary: masses of flab or the truth cannot be arrived at. What bothers me about your post is the assumption that big and baggy and loose and unfocused is inherently good for the pursuit of ArbCom's goals—that any restriction on evidentiary freedom amounts to a curtailment of "tha truthe", a cloudy concept at the best of times. The limitation of this line of argument is that clarity, sound judgement, justice, and—let's be honest—the reasonable efficiency of the process, are egregiously damaged by acres of fluff, anger, irrelevance and the time sink. Your binary suppositions that these are a necessary part of an "open and flexible" process, and that any attempt to bring the system under proper, professional management would render it "closed, rigid", have no basis in evidence. This article you refer to is certainly not reliable evidence; it is seriously flawed from the outset. Tony  (talk)  11:25, 20 April 2009 (UTC)


 * To use a Parkinson's Law parallel, if you plan to accept chaos and bloat, chances are that's what you will get. To ensure efficiency of the process, we need to redesign the structure so as to deliver the desired result, and not close our eyes and hope for the best. Ohconfucius (talk) 14:59, 20 April 2009 (UTC)

Suggested changes summaries
I'm going to try and go through the above, and highlight out significant suggested changes, to make it easier for drafting purposes. --Barberio (talk) 12:04, 20 April 2009 (UTC)
 * This turns out to be a bit more difficult and time consuming on freeform discussion than I thought, so if anyone else would like to chip in to put up summaries of the above discussions? --Barberio (talk) 12:53, 20 April 2009 (UTC)
 * Yes, that would be great; anything that makes it easier for us to keep track of the changes being proposed will speed up further drafting. Kirill [pf] 04:00, 21 April 2009 (UTC)

Wording of main duty
Under "duties and responsibilities", the first (and probably most important) item is:
 * "To act as a decision-maker of last resort for Wikipedia disputes, to determine which of these disputes are suitable for arbitration, and to carefully review and render binding decisions on those disputes;"

The problem I have with this is that "dispute" is left completely undefined; indeed, the second clause implies that the committee itself is to decide on each occasion, not necessarily on any consistent basis, whether a given complaint represents a "dispute" for the purposes of the first clause. Can we not do better than this? I'm sure that if we can define more accurately what we mean by a "dispute suitable for arbitration" (or perhaps use different language altogether), then a lot of the problems that have been discussed above about scope, forum-shopping etc. would be placed in much clearer perspective.

My own (fairly radical) proposal, moving away from the concept of a "dispute" altogether, would go something like this:
 * In the capacity of a final decision-making body, to consider complaints about specified actions of administrators or other officials on English Wikipedia, or about the individual or collective failure of such officials to ensure the proper functioning of community processes in specific cases; and consequently to issue binding instructions or other recommendations in order to remedy identified errors or failures and to prevent their recurrence.

OK, I know it's wordy, but at least it says something. Other suggestions?--Kotniski (talk) 11:28, 21 April 2009 (UTC)
 * Kotniski ... I'm afraid I don't like your suggestion one bit, as much as I respect your other input here. Yes, ArbCom needs to deal more with admin transgressions, but this wording seems to imply that it's all they would do. I don't like the assumption of failure. The existing sentence, by the way, doesn't make any sense: "to carefully review" what? It doesn't say. "Render" is la-de-dah code for "make" or "issue". And "carefully" is fluff ... as if ArbCom wouldn't be careful (I'm serious, it's a bad look to append "careful" to just one phrase.) Tony   (talk)  12:06, 24 April 2009 (UTC)
 * Agreed it's not particularly clearly worded (my proposed version, I mean), but what I had in mind was that ArbCom would look at (1) administrator transgressions; AND (2) administator failures (failure to ensure that processes ran smoothly). That would mean that somone with a complaint would take it to the admins first, and only on failing to get satisfaction there would have the option of going to ArbCom. (Or else the admin community might itself refer the matter to ArbCom.) Like I've said several times, I can't personally think of any situation where this isn't the appropriate order in which things should be done.--Kotniski (talk) 14:35, 24 April 2009 (UTC)

Time as well as space limitations
I agree with your summary. The talk pages are ridiculous. There should also be a time limit on these cases. They shouldn't be allowed to drag on for months, with people's lives in wiki-limbo. The "scope" of the case is terribly important, since people need to know for what exactly they are providing evidence, and arbitrators should have a (reasonable) time limit to decide and if they can't meet it, the case dismissed without prejudice and would have to be re-filed. Tundrabuggy (talk) 07:37, 23 April 2009 (UTC)
 * In fact this business about a predetermined scope is a particular issue in an ongoing arb case Requests for arbitration/West Bank - Judea and Samaria. Although the dispute was over naming conventions and editing behavior within the area of West Bank--Judea--Samaria editing, it has clearly become something else altogether.  Had a scope been delineated, we would not be seeing ArbCom apparently on fishing expeditions & suddenly dragging in year-old irrelevant material on which to hang certain editors.  This strikes me as highly unfair and clearly "arbitrary!"  Tundrabuggy (talk) 03:28, 24 April 2009 (UTC)
 * I don't know that case, but it sounds very much like the situation with the one case I have had the misfortune to become involved with. Can we get some response from arbitrators about this problem? It seems to me that if you allow "cases" with generalized names and no clearly defined scope, then they're always going to get out of control in this way.--Kotniski (talk) 08:44, 24 April 2009 (UTC)
 * That is an interesting point Kotniski. Back at WP:ACE2008 I brought up the issue of case names in my questions to the candidates.  At one point in time in the past case names were formulated as Party X v. Party Y wherein Party X alleged Party Y had done something that violated policy.  Party X presented its allegations and Party Y responded, and Arbcom ruled on the issue.  The subsequent style of case was named as Party Y in which a person was accused of doing something wrong and evidence was presented on that person and Arbcom issued a ruling.  More recently the dominant style of cases seems to be centered around Topic Z, where all of the events in Topic Z are presented and arbcom comes in and sanctions and thanks editors and defines sanctions over the topic area.  While the original X v. Y and Y style cases had a very low recidivism rate and a very fast resolution time, the Z style cases seems to continue to come back to arbcom time and time again and take much longer to decide without satisfactory results.  One reason I've often seen given is that Arbcom does not wish to foster an adversarial environment, so case names are given in a manner that does not pre-suppose a guilty side of a particular conflict, but I'm not sure that trying to reduce the adversarial nature of things at arbcom (when both sides already hate each other), necessarily results in better findings.  MBisanz  talk 08:54, 24 April 2009 (UTC)
 * Thanks for that explanation, it makes a lot of sense. I suppose it's less important what a case is called if its scope is precisely defined from the start. The adoption of a neutral name to avoid personalizing shouldn't be an excuse for the case to develop in different ways. (I mean, if a case concerns the behaviour of Smith over owl articles, then we might call the case "Owls" rather than "Smith" to reduce the personal stigma, but that choice of name shouldn't entail any wider scope to the case than it would have had if it were called "Smith", or even better "Smith and owls".)--Kotniski (talk) 09:16, 24 April 2009 (UTC)
 * Yes, thanks MBisanz (Michael? Matthew? It would be so nice to know people's first name.) Perhaps an entirely neutral number should be used as a case name, which would be good for locating in archives and would avoid unintended constraints or bias, and would save the trouble of carefully inventing a safe descriptive name. Why not go simple: AC0915 for the 15th ArbCom Case of 2009? On the scope issue, what is presumably a leak about the upcoming judgement of the "dates" case suggests that one matter from the past is going to be pinned on a particular party without that party's knowing that it would be part of the scope; thus, the user may have been denied the natural justice of knowing that he would need to defend himself on this matter during the case. If this leak is born out in the judgement next week, it will be a pertinent demonstration of why it is egregiously unfair not to define the scope at the beginning and—if the Committee wants to expand the scope during a case—to announce this on the Evidence page. This is the only way for everyone to know where they stand and to focus on presenting evidence appropriate to that scope. At the moment, it's like throwing dice to participate in an ArbCom case. Tony   (talk)  12:00, 24 April 2009 (UTC)
 * His userpage indicates his first name is Matthew. Ohconfucius (talk) 12:32, 24 April 2009 (UTC)
 * Hehe, I am pretty well documented at User:MBisanz/Infobox.  MBisanz  talk 09:20, 25 April 2009 (UTC)
 * There are many individual pieces of evidence presented by various parties, and a number of evidence submissions in the dates case which were highly questionable to me even using a broad definition of scope. I have specifically in mind the submissions made by Omegatron and Thunderbird2. I will leave it to the reader to judge whether it a scope issue, or a clerking issue. Ohconfucius (talk) 12:47, 24 April 2009 (UTC)

[od]To me scope is key. From the beginning it was unclear whether the West Bank/ Samaria case was a content dispute or a behavior issue. So some folks put up diffs related to the content, while others put up diffs related to edit-warring. Then suddenly, as the decision begins to come down, new supposedly damning (although even that is a matter of opinion) material from well over a year ago, has nothing to do with the I/P area in general, or the West Bank, Judea and Samaria area in particular, gets introduced and suggested for sanctions. How can one defend oneself from such a thing? One could never put the past behind one, whether resolved or not. Anything can be opened up again at any time, and used in unrelated areas. Scope must be issue-related and time-related and clearly delineated. It needs to be announced and clearly explained at the beginning of each case. As Tony  says above, it is imperative to have scope in order to determine pertinent evidence, and if the Committee wants to expand it, that too needs to be announced (in a timely fashion) on the Evidence page. I realise that this gives the Committee more work, but it's only fair to all the parties involved. Tundrabuggy (talk) 03:52, 25 April 2009 (UTC)
 * Less work in the end, actually. The process is seriously disorganised. Tightening up the timeline and scope will make the arbitrators' job easier, even though a few key decisions would need to be made early in a case. This is apart from the manifestly unfair burden on the parties to provide every conceivable type of evidence as a defence against what might turn out to be the scope of the decision, and the potential for miscarriage of justice where a party does not second-guess that scope in their evidence. Tony   (talk)  04:09, 25 April 2009 (UTC)
 * You have that exactly right. I read where they recently implemented an appeal system for those who have been long-term banned. That's something, at least. Tundrabuggy (talk) 05:40, 25 April 2009 (UTC)


 * I think this is a beneficial discussion. I had concerns over the scope of a recent case after it seemed to expand to include a tangential related issue which had seen prior discussion between the involeved part and an arbitrator.  I'd certainly like to see scopes established, and also clarify how we deal with conflict between arbitrational duties and editorial issues.  Should an unrelated discussion between an arbitrator and a party to arbitration become a part of the case to the point that the arbitrator involved in that discussion makes proposals regarding that discussion and votes on them? I'm unclear on the separation of roles with the current arb-com, and how closely arbitration findings should be used to settle disputes and how far they should be used to attempt to move the community in an arbitrator's desired direction. Hiding T 09:45, 27 April 2009 (UTC)

mediation
Evidence from mediation: No evidence from official mediation shall be presented or considered in an arbitration proceeding except with the prior written consent of the Mediation Committee.

I think this is a terrible idea. In general courts tend to use:
 * discussion -> mediation -> arbitration -> full trial quite frequently

for civil issues. That's very similar to the wikipedia idea of:
 * discussion -> informal mediation -> formal mediation -> arbitration.

The steps should feed into one another in a cooperative way. Further damaging or undermining a mediation process should be a serious offense regarding to disruptive editing. Editors should not be allowed to misbehave in mediation and mediators should be able to be called. In fact I would even argue that an entirely opposite policy like, "In the case of a failed mediation resulting in a dispute advancing to arbitration a mediator can be asked by the arbitration committee for a finding of fact". It might be reasonable to say something like, "mediators will not be asked for any private correspondence". jbolden1517<sup style="color:darkgreen;">Talk 20:47, 13 May 2009 (UTC)
 * Absolutely not. That undermines the inherent freedom parties have in-Mediation. See Mediation. —AGK 19:23, 20 May 2009 (UTC)

Status of draft #2
Just to give anyone waiting here some status: I'll be wrapping up work on a preliminary version pf draft #2 in the next few days, and I expect that we'll post the draft publicly a week or so after that, once we've had time to go over it internally. Kirill [talk] [pf] 03:46, 31 May 2009 (UTC)

Private hearings
It should be a requirement that the existence and scope of a private hearing be made public before it starts and that the conclusions of it (explicitly redacted if necessary) be made public when it is completed. --Tango (talk) 19:41, 25 June 2009 (UTC)
 * Why? I'm not saying that I totally disagree with making some on site notation of the hearing. But if all involved parties want a private hearing to discuss something that they feel is best discussed behind closed doors, then why should it be made public? FloNight[[User talk:FloNight|&#9829;&#9829;&#9r it couldn't be done.
 * For one thing, you can't be sure all the involved parties know about it unless you make it public. There may be someone involved that has either been forgotten or is being intentionally kept out of it. Secondly, any action taken as a result of the case is likely to be difficult to keep private - for example, blocks are a matter of public record. If you have a block log that says "indefinitely blocked following a private ArbCom case" and nobody has any idea what that ArbCom case was, you can guarantee there is going to be a lot of drama. Thirdly, transparency is a good thing in itself. Unless there is a good reason to keep the existence of the case secret (and the involved parties wanting it to be isn't a good reason), it shouldn't be. --Tango (talk) 00:46, 26 June 2009 (UTC)
 * I tend to agree with Tango here. It does not disclose any private information to post a notice of the case; and could very well forestall considerable drama. I prefer transparency unless contra-indicated; unless there is a "why it should not be mentioned that there is a case" then I see no need for much argument for "why should it be made public". Transparency should be the default; closed or private, the exception. While I can certainly accept the need for some hearings or evidence to be private, I don't see any rationale for the mere existence of said hearings to be private. KillerChihuahua?!? 00:53, 26 June 2009 (UTC)
 * Of course public cases with transparency is the main way that cases are handled. My comment was related entirely to Tango's comment that it should be a requirement that a private case be made public and the scope placed on site.
 * If a posted notice keeps a party from filing a request? Or keeps the Committee from resolving an sensitive issue with expediency? I don't think that are good resolutions to a problem. I think that we need to be careful that we think this through. If the only people affected will be the involved parties, I don't think that we need to make a request public before the start of a case that could be handled best with complete discretion. Similar to how all OTRS situations are not made public, sometime the party needs to have a private venue to be heard. Only if an action results in some change that will be noticed in public would the matter need to have public disclose. And then it may be limited. Issues such as complaints of sexual harassment, pedophile related bans, and people with severe mental health problems that need editing restrictions are some of the issues that are often best dealt with behind closed doors with out any notice to the community necessary. FloNight&#9829;&#9829;&#9829; 01:18, 26 June 2009 (UTC)
 * If the only people that are going to be affected are the involved parties then it isn't a matter of ArbCom. ArbCom exists to facilitate the production of an encyclopaedia. If the encyclopaedia isn't going to be affected, ArbCom should stay out of it. If the encyclopaedia is affected, then the community as a whole has a right to know about it. I don't see where the problem is with sexual harassment cases - the names can be kept out of the public notification, the only time a name needs to be mentioned is if some action is taken against the perpetrator, in which case the perpetrator's name needs to be made public but that seems perfectly acceptable to me (their name should, perhaps, be kept secret until it has been confirmed that they are guilty). I don't see what ArbCom would have to do with paedophile related bans - the only way we know someone is a paedophile is if they tell us, in which case we just block them, there is no ArbCom case. Editors with mental health problems is an interesting one. I'm not sure it is in the encyclopaedia's best interests to have editors with secret editing restrictions, and that is what we should be basing our decisions on. If we can be nice to people without significant harm to the encyclopaedia, then we should obviously do so, but I think secret editing restrictions do have the potential to be harmful. I'm also not sure how you would keep something like that secret - surely it would only come to ArbCom's attention if somebody complained and the person complaining isn't likely to know there is any reason to keep it secret so would do so publicly? Unless you are planning on oversighting RFAr, I don't see what you can do. --Tango (talk) 02:52, 26 June 2009 (UTC)
 * Maybe we are talking past each other. And that is understandable since you do not know the volume and type of private email complaints that get directly referred to the Committee, or come by way of OTRS agent, an experienced admin or other functionary.


 * The Committee is aware that we are working in an organization where transparency is the norm. The ArbCom policy shows that the Committee agrees that transparency is the default way to handle cases. My point is that there are valid reasons for privacy in some instances, and practical process issues that need to be addressed, too.


 * In any large organization there are issues that need to be addressed outside the eye of the public either totally, or partially. Some of them need swift action that would be hindered by the official requirement that the Community be notified prior to the Committee taking any action. As I stated already, I have no problem with the Committee making on site acknowledgment of each email complaint received. But I don't think that it needs to happen prior to us taking action, or the at the level of detail suggested in your first remark.


 * If there is going to be a full case heard mostly in private because of the type of evidence, but it is requested in public, then the case will be already be on site. So, the only cases that we are talking about would be the type of cases that are totally heard in private. I think the Audit Subcommittee's and the Ban Appeal subcommittee's approaches to dealing with the cases referred to them is good. I don't think that we need to massively change those process now. The other cases that we deal with in private are unusual for some reason, so having a rigid format to handle them doesn't work well. FloNight&#9829;&#9829;&#9829; 11:26, 26 June 2009 (UTC)
 * I am mostly talking about full cases. If ArbCom decides that there isn't really a need to hear evidence and that they can just take immeadiate action (a decision that should only be made when there is no doubt that it is the right decision) then there is no need to notify people. --Tango (talk) 13:33, 26 June 2009 (UTC)
 * Ok. I think that we are in agreement then. It is important to acknowledge the significance of the issues that are addressed in private and the various reasons that we do it. All the issues that come to the Arbitration Committee have some element about them that makes them best handled by the Committee. Sometimes it will be due to factors that make privacy a key factor. In these situations I want to reserve the ability to deal the issue first and then later address the best way to notify the Community. It's possible that public notification would come later because of the way that the situation evolved. As long as we all are in agreement that transparency is the customary approach, and trust that when divert from it, the reason will be well substantiated by the particular circumstances of the incident. Does that make sense?FloNight&#9829;&#9829;&#9829; 14:30, 26 June 2009 (UTC)
 * Yes, that's ok. My main concern is that the results of private cases should be made public (with appropriate redaction). The case not being made public in advance is unfortunate and should only happen when absolutely necessary, but if it turns out to be a problem there can always be an appeal after the results are published, so it isn't the end of the world. --Tango (talk) 15:29, 26 June 2009 (UTC)

Part 1 of the new draft: copy-editing and other suggestions
This is classy compared with the drafting standard of the existing policy text and the first draft of the proposed new policy. It must have been a great deal of work, and whoever did it should be thanked.

General point: Have you thought of numbering everything for easy reference?


 * Duties and responsibilities
 * "3. To deal summarily with urgent or emergency matters—for example, blatant abuse of administrator or other privileges, and threatening or malicious conduct—that presents a danger to the project or its contributors;"
 * Perhaps "urgent matters and emergencies"?
 * Grammar—the type of mistake I make: "present".
 * BLP issues can involve people who are not contributors. I'd have thought this would also come under ArbCom's responsibility. But maybe that's a hornet's nest for possible legal action; or a buck that can be passed to the WMF. <Pulls ignorant face>
 * "5. To appoint those functionaries granted access to privileged information, including the holders of the CheckUser and Oversight privileges."
 * Remove "those".


 * Selection and appointment
 * "The members of the Committee are appointed by Jimbo Wales, in his role as project leader, following annual advisory elections, whose format is decided by the community."
 * Remove commas after "Wales" and after "elections".


 * Conduct
 * "Arbitrators must act with integrity and good faith at all times to retain the trust of the community and maintain the high standards expected of them. They must preserve the confidentiality of (i) private correspondence sent to the Committee and (ii) the Committee's internal discussions and deliberations."
 * The two sentences don't appear to be sufficiently linked to belong together. Should be two different paragraphs/points?
 * "Arbitrators are required to respond fully and expediently to questions put by the Committee and the community about conduct which appears to conflict with their trusted roles."
 * "Expediently" is surely not the right word; was "expeditiously" intended (in which case "promptly" would be better)?


 * Procedures and roles
 * "The Committee may modify its internal procedures, form subcommittees, or designate individual arbitrators for particular tasks or roles. Where appropriate, the Committee will invite community comment on intended changes prior to implementing them."
 * "modify" -->"determine" (otherwise jumps the gun)
 * first "or" --> "and".
 * "Where appropriate" is a hornet's nest: in whose opinion? --> "The Committee may, where it sees fit, invite community comment on any proposal."


 * Inactivity, resignation, and removal
 * "Arbitrators who are, or who are likely to become, inactive for a period of time are expected to ..." --> "Arbitrators who will or are likely to become temporarily inactive are expected to ...".
 * Active voice simpler and places the Committee in a dynamic position: "Arbitrators who are only intermittently active, or who are inactive for an extended period, may be asked by the Committee to remain inactive until the situation is resolved." --> "The Committee may ask arbitrators who are only intermittently active, or who are inactive for an extended period, to remain inactive until the (their?) situation is resolved."
 * "with giving reasons" doesn't work. Perhaps: "with or without explicit reasons"?
 * This needs a rewrite: "Arbitrators resigning voluntarily are entitled to retake their seats at any time during and, for the duration of, the remaining period of their original term of appointment." --> "An arbitrator who resigns may resume their seat at any time during their term of appointment."
 * "Arbitrators who engage in conduct grossly or repeatedly unbecoming to their position may, after being given a reasonable opportunity to provide an explanation for their actions, be removed from office by the Committee."
 * "Arbitrators whose conduct is grossly or repeatedly unbecoming to ...".
 * Bit wobbly about "grossly or repeatedly". So it doesn't have to be gross to qualify if it's repeated. Please check whether this distinction is worth making. "is significantly unbecoming" might be too broad, but the language feels nicer. Tony   (talk)  16:17, 26 June 2009 (UTC)
 * Two random points:
 * In re. BLP, I expected this to fit under the damage to the project heading; I'm not sure trying to give an exhausting list is useful, but that point might indeed need reinforcing.
 * "grossly or repeatedly" makes an important distinction IMO, in that an accumulation of small problems is just as damaging as one big boo-boo; and the committee needs to be able to act in both cases.
 * YMMV. Void where prohibited. My own two cents, may not reflect the views of the committee.  &mdash; Coren (talk) 16:02, 27 June 2009 (UTC)

Comments by FT2
This is a much improved draft. I have added my suggestions to the draft. A handful of comments either need space for discussion or don't easily fit in a footnote.

If the draft will say that "arbcom won't do content disputes" then it needs to state explicitly what a "content issue" is. A previous draft handled it this way, using a footnote. This might fit well in the Arbitration Guide:
 * 1. Content disputes
 * {| style="border:black solid 1px" width="90%"


 * The arbitration process primarily addresses disputes involving problematic user conduct (including the means by which content is determined and untoward activity affecting content decisions). Arbitrators may look at editor conduct in relation to policies governing article content, and may reaffirm and provide guidance on those policies, but will decline requests to help determine the actual contents or presentation of articles ("content disputes").[1]


 * [1] A content issue is one where the committee is being primarily asked to decide as editors, between facts and viewpoints within a mainspace article, or to choose which are 'good' or 'bad' templates, categories, layouts, etc, as opposed to addressing the principles and disputes surrounding these choices, and analysis of the compliance or otherwise of editors' activities with established communal policies and norms.


 * }

The old draft policy included this item, which seems a valuable point:
 * 2. Interim decision pending community consensus
 * {| style="border:black solid 1px" width="90%"


 * In exceptional cases of communal division or exceptional risk of harm to the project, the Committee may determine a interim means to achieve the aims of communal policies or to curtail dispute in the area, until the community has developed a consensus.
 * }

Whatever is decided in regards to Arbitration and Enforcement, they are our final means of dispute resolution and handle our most sensitive, complex cases. They exist because there is communal disagreement what's best and because users (often including admins) are fighting endlessly over the matter. Disruption of such decisions, which are often hard reached, is potentially more damaging since it undermines our ability to finally draw an end to the dispute, and tries to force the dispute wide open again against that decision. There isn't a further route of appeal. To be "binding" it needs to have slightly more force going forward. Proposed policy addition:
 * 3. Disruption of Arbitration process
 * {| style="border:black solid 1px" width="90%"


 * Arbitration processes are by definition difficult and contentious. Disruption of Arbitration processes, including improper disruption of properly imposed measures, may be sanctioned more strictly than comparable disruption of other processes.
 * }

It is foreseeable that after the initial "flurry" of the revised policy discussion and referendum, no amendment will be considered by the community for a long time. As it stands the sole decision will be on Arbcom's version of the proposal. A possible amendment that many users like but which the committee decline to include (which might be for many reasons not reflecting on its merits) is therefore in an invidious position. It cannot gain entry to the draft, nor is it likely to be considered separately on its own merits for a long time after. It's likely that at the point the updated policy is decided, a number of users will have reasonable proposals for amendment that have strong communal interest.
 * 4. User-proposed amendments at time of referendum


 * My suggestion is that any user who feels they have an important amendment that isn't in the final draft, that can show 100+ !votes and Arbcom approval in support of it (as required for a proposed amendment), should have the opportunity to present it to the community at the same time to allow the community to consider it too. Under this proposal there would be a ratification process asking users 1/ whether or not they ratify the main draft policy, and on a separate page 2/ whether or not they ratify some or all of the other user amendments which may have got the strong support (100+ supports and Arbcom approval) required. That way any suggested amendments can be considered for ratification as part of the same process; potentially the policy plus some of these will pass, while others will fail. Proposed addition:
 * {| style="border:black solid 1px" width="90%"


 * At the same time, any amendment proposals that have obtained the support of over one hundred editors and the approval of the Arbitration Committee will be presented on a "proposed amendments" page to be considered for ratification.
 * }

FT2 (Talk 01:35, 2 July 2009 (UTC)


 * Replies, ordered by clause number:
 * 2. While I'm broadly in favor of retaining a clause along these lines, I am concerned that it would be sufficiently controversial to jeopardize the ratification of the policy. If it's proposed, I expect it will be presented as a separate amendment, so that it can pass or fail on its own merits without affecting the core document.
 * 3. I'm not convinced that this is needed; disruption of arbitration is sanctionable more directly (e.g. through clerk- or arbitrator-imposed blocks rather than formal sanctions), but I've seen no evidence that it's sanctioned more severely, or needs to be.
 * 4. This is an interesting proposal, but I have two concerns:
 * a. If we allow simultaneous amendment proposals, I see no reason to require that they meet both criteria (100 endorsements and Committee approval); we can just as easily allow either type to be considered normally, without the need to introduce a special rule for this batch.
 * b. From a logistical standpoint, we need to consider the case where the core policy might fail to be ratified; in that scenario, there would be no provision for amendments via this method, so the amendments could not take effect even if they, individually, passed. I think that for this to actually work, we'd need to (i) explicitly state that the amendments would be contingent on ratification of the core policy, and (ii) prohibit amendments which would change the ratification or amendment clauses in the core policy itself.
 * Kirill [talk] [pf] 02:00, 2 July 2009 (UTC)


 * 2 - fair point, but would it really be controversial? It was proposed last time and reactions didn't seem to pick it up as a deal-breaker or even a controversial point to consider. However it's a valid point of view and YMMV.


 * 3 - Needs saying I think, for reasons given, but on reflection doesn't necessarily need to be in the referendum. Any sanctions will be within usual norms, even if it is towards the "disruption less tolerated and more sternly handled" end of the spectrum. This is perhaps more the kind of warning to users than a matter that needs to be in policy, and might be fine within the Arbitration Guide as a warning to users. Updated: Actually, sanctions by Arbcom do enjoy a privileged position; 1/ the Committee can state they should not be overturned other than by a specified process or by the Committee itself, and 2/ users may be barred from editing pages or restricted from participation in ways that are not open to "appeal to the community" and which may differ from conduct enforcement norms on other pages. This power is currently important to bring a measure of control to disputes that have escalated all the way to arbitration and to ensure remedies can have their intended dispute-quenching effect once the case closes. The Committee's right to create sanctions (both as remedies and in response to conduct on Arbitration-related pages) that are not readily subject to overturn except by a specified route or that are subject to non-standard Committee-determined enforcement measures should probably be reflected in policy, if such a right is likely to be desirable to persist following the referendum.


 * See also proposed edit under "Procedures and roles".


 * 4 - Agree with your basic point and the two conditions (i) and (ii). I didn't spot the "or"; I had figured that the "both conditions" reason might be valuable because potential amendments need to enjoy wide popular support and also the Committee's view they are not in conflict with the aims of policy or likely to harm the Arbitration process. The former ensures only matters with strong evidence of interest may be taken as far as referendum, the latter protects the policy wording against pushing through of ill advised material or agendas. Both are probably both valuable. Requiring both would be an additional safeguard; which may or may not needed. YMMV. (That said, "possible alternates" that are favored but might ideally be better handled separately may need to be presented; the process should allow this).


 * One option for handling such a process (your other point) is that during the main referendum, any user may post a proposed amendment on a separate page (similar to RFC format with amendment and rationale in each section and "supports" below); if the main referendum passes, then those proposals which have gained more than 100 supports by the end of !voting will then be opened to consideration as a follow up. This would involve two stages and some will want it all over at once, but it's probably a conceptually neat way to handle it and an extra week for a 2nd stage to cover this point won't hurt. The alternative is that the final draft for !voting is presented and a period of reflection and discussion and questions is allowed prior to !voting, during which any amendments may be presented for communal support; those proposals gaining strong support will then be presented for consideration at the same time as the main !vote. The problem with the second option would be the hiatus between posting and !voting to allow possible amendments to be mooted and checked for support; conceptually it's better to deal with the main referendum first, then any strongly-supported amendments by way of immediate follow-up.


 * FT2 (Talk 08:07, 2 July 2009 (UTC)

Ratification by majority vote?
Is this "community referendum" through which "majority support" is required to ratify policy intended to be a poll or a discussion? The former would fly in the face of best practices wrt policy changes, while the latter seems inconsistent with the current wording of the draft. [This comment made here with the intention of fostering discussion – if it should be on the page itself, please feel free to move it]. <font face="Goudy Old Style"> Skomorokh 16:42, 3 July 2009 (UTC)

Subject to recall?
Should a provision be added that all members of the arbitration committee sit for their periods of appointment subject to their adherence to the rules governing arbitrators. That some number of editors in good standing may institute a recall procedure to determine whether the arbitrator still has the trust of the community, and such recall election can be held the same way we choose admins and bureaucrats. Too much drama has been made when arbitrators appear to misbehave and leaving it to the rest of their arbitrator peers as the only means of removal leaves insufficient checks and balances. We have had a couple of instances where WP has earned bad press in regards to arbitrators, we should have a way to clean our own house without weakening the close working relationships among the arbcom by forcing them to do their own purges. Carlossuarez46 (talk) 20:52, 9 July 2009 (UTC)
 * No, arbs already have too much to deal with, it would be a big mistake to pile defending against recalls onto their plates. Looie496 (talk) 22:22, 9 July 2009 (UTC)
 * They would have a lot less to deal with if we had clear definitions of what constitutes a case suitable for ArbCom and if cases accepted had clearly defined scope, as keeps being proposed. Obviously they should be expected to defend what they do and be removable if they lose the community's trust, but perhaps since they are Jimbo's appointees, it will be up to Jimbo to dismiss them when they are shown to have lost that trust, and we don't necessarily need a specific procedure for that to happen.--Kotniski (talk) 03:28, 10 July 2009 (UTC)


 * Developing a system for arbitrator recall is actually on our agenda (see WP:AC/A, "Develop an arbitrator recall process"); at the moment, nobody has been able to come up with a system that would reasonably allow a recall to succeed where appropriate while still prevent frivolous and inappropriate requests. Kirill [talk] [pf] 03:46, 10 July 2009 (UTC)


 * Inclined to agree that a self-policing solution is also (unnecessarily) asking for trouble. Given that a separate body is required, and we have such a body already set up for serious concerns over functionaries, I'd be inclined to ask about extending the remit of the audit subcommittee on this. Asking AC to patrol itself (except in blatant cases) may work in some instances but its a stressor and asking for trouble. As for "pile on" - giving the right to a separate group would prevent frivolous complaints (arbs should be obligated to respond formally, whether in private or public, to non-frivolous ones). FT2 (Talk 22:32, 11 July 2009 (UTC)
 * Perhaps the remit of the audit subcommittee should be expanded, but we need to have them be able to act nimbly, we have had too many scandals on what the media calls our "supreme court", which brings WP's self-governance into question and hence, credibility into question. Carlossuarez46 (talk) 00:33, 15 July 2009 (UTC)

Announcing comments
I won't be available for the next few days, but I have a few concerns, comments and maybe proposals on those subjects: I'll comment further and give examples of such actions in a couple of days. Cenarium (talk) 09:30, 11 July 2009 (UTC)
 * 1) overturn of community decisions by ArbCom
 * 2) executive decisions by ArbCom, by this I mean, outside a formal case
 * 3) decisions at the end of (formal) cases with effects beyond the scope of the case
 * Comments on motions are also forthcoming. Cenarium (talk) 02:01, 30 July 2009 (UTC)

Question about functionaries
I have a question about the meaning of:


 * "To appoint[4] those functionaries granted access to privileged information, including the holders of the CheckUser and Oversight privileges."

I'm wondering how and why functionaries would be appointed other than those with CU and oversight privileges. That is, isn't being a functionary simply equivalent to being an ex-Arb, CU, or oversighter? SlimVirgin talk| contribs 15:44, 12 July 2009 (UTC)


 * There are people on the functionaries list who are not one of any of those groups (e.g. developers, WMF staff, etc.); strictly speaking, this item combines two separate ones:
 * Appointment of CU/OS holders
 * Control over membership of functionaries-en
 * I'm not sure that it would be worth splitting them, in practical terms. Kirill [talk] [pf] 17:23, 12 July 2009 (UTC)


 * My concern is that it sounds as though ArbCom appoints functionaries, whereas it actually appoints CUs and oversighters, which makes them ipso facto functionaries. Developers and Foundation staff also become functionaries by virtue of their position, not because of ArbCom, and functionaries are automatically allowed on the functionaries mailing list. So that needs to be reworded. SlimVirgin  talk| contribs 17:42, 12 July 2009 (UTC)


 * I've had a go at a reword, something like this:
 * To appoint, oversee, and if necessary remove local access to privileged information, including [access] to the functionaries-en mailing list and the CheckUser and Oversight functions. (subject to Foundation policies and override), or
 * To appoint, oversee, and if necessary remove local functionaries' access to privileged information, including [access to] the functionaries-en mailing list and the CheckUser and Oversight functions. (subject to Foundation policies and override), or
 * To appoint, oversee, and if necessary remove local functionaries trusted with access to privileged information, including [access to] the functionaries-en mailing list and the CheckUser and Oversight functions. (subject to Foundation policies and override).
 * I think the first is most accurate, as well as being shortest. FT2 (Talk 15:42, 13 July 2009 (UTC)


 * What is wrong with simply saying that the ArbCom appoints CUs and oversighters? If access to the tools are withdrawn, access to the list is ipso facto withdrawn too. What is to be avoided is the implication that ArbCom appoints functionaries in general. It doesn't. It appoints CUs and oversighters. SlimVirgin  talk| contribs 19:02, 13 July 2009 (UTC)


 * Well, the wording needs to cover everyone on the functionaries-en list, since the list traffic contains restricted information, and thus access to it is under ArbCom's direct control. It's possible to not hold (active) CU/OS rights and still be on the list; this is the case, for example, for former arbitrators; ArbCom can remove such a person from the list even though this does not entail removing of CU/OS access in and of itself.  This also applies, incidentally, to arbcom-l itself.
 * To cover this case, the policy would need to say something like:"To appoint holders of the CheckUser and Oversight privileges, and to control access to venues where privileged information derived from these tools is stored or exchanged."This would cover the appointment that you refer to, and, separately, the control over mailing lists, private wikis, and so forth. Kirill [talk] [pf] 01:33, 14 July 2009 (UTC)


 * What I'm trying to avoid the text implying is that ArbCom could say of any random person, "Here, you're now a functionary, and your name's on the list." Saying that ArbCom "controls access to venues" etc, implies that they could do that.  SlimVirgin  talk| contribs 02:02, 14 July 2009 (UTC)


 * In that case, perhaps it would be useful to have the responsibility clause only cover CU/OS appointment, as you suggest, but introduce a clause elsewhere describing the membership of functionaries-en (CU/OS holders, WMF personnel, former arbs in good standing) and then note that determining whether a former arb is "in good standing" is up to ArbCom? Kirill [talk] [pf] 02:08, 14 July 2009 (UTC)
 * Yes, that would work. SlimVirgin  talk| contribs 02:09, 14 July 2009 (UTC)

Question about subcommittees

 * "The Committee may modify its internal procedures, form subcommittees, or designate individual arbitrators for particular tasks or roles. Where appropriate, the Committee will invite community comment on intended changes prior to implementing them."
 * Is this the section that authorizes the appointment of the ACPD? I originally assumed "subcommittees" were made up of members of ArbCom only. Does this mean that subcommittees can be made up of selected members of the community at large chosen by ArbCom? If so, does this mean that the community members on a subcommittee such as ACPD are constrained by the tasks falling under ArbCom "Duties and responsibilities"? Or can ArbCom expand ACPD's duties to cover other aspects of the community not under ArbCom's "Duties and responsibilities"? &mdash; Mattisse  (Talk) 16:54, 14 July 2009 (UTC)
 * ACPD is not a subcommittee. The creation of new projects or processes that are not internal to ArbCom or within the scope of a case is not covered by this policy. Project development is not in the duties and responsibilities of ArbCom. I don't see anything legitimatizing this, in the current policy either (so archaic and strange-looking as it is). Cenarium (talk) 23:48, 24 July 2009 (UTC)

Decisions mandating policy or process
The proposed policy states that ArbCom is not a vehicle for creating policy; it should also states that ArbCom decisions are not a mandate for modifying policy, or processes. This could be amended to the sentence "No decision will mandate the precise content of any article.", so that it is turned into "No decision will mandate the precise content of any article, the meaning of any policy, or the operation of any process." Cenarium (talk) 01:06, 25 July 2009 (UTC)
 * ArbCom could actually be much more useful if it made certain decisions about this sort of thing. As I've said multiple times, the community can make these decisions (by what we call "consensus"), however, in some cases (where the existence and consequences of consensus are disputed) there need to be judges - not involved in the dispute itself - to interpret that consensus. And because these judges get it wrong sometimes, it needs to be possible to have their decisions reviewed by another set of judges of unquestioned standing - that's where ArbCom should come in. So I can foresee that ArbCom decisions might (in effect) mandate some of these things - though preferably only through confirmation or modification of consensus calls that have already been made by admins. (As people may know by now, my view is that ArbCom should restrict itself almost entirely to reviewing admins' specific (in)actions.)--Kotniski (talk) 06:20, 25 July 2009 (UTC)

Case-specificity of decisions
I think it should be stated that in all cases, the decisions are to be case-specific. That is, they should concern only the elements within the scope of the case. For example, if the case is about a content dispute on an article about a fungi species, decisions that would apply to all content disputes in articles on fungi species could not be taken; decisions should only be taken on this content dispute (that is happening on an article about a fungi species). In formal cases, findings of facts should also be case-specific. I don't expect ArbCom to issue a findings of facts or remedy against a user completely unrelated to the case at hand, but for the coherence of the aribtration policy, it is a necessity. Furthermore, when a case is brought before ArbCom, the community expects ArbCom to rule on the case itself, not on elements beyond the scope of the case, and it's what an arbitration committee does in 'real life'. This is fundamental. Cenarium (talk) 01:40, 25 July 2009 (UTC)
 * Yes, I think this has been said before many times in various places, and I quite agree.--Kotniski (talk) 06:22, 25 July 2009 (UTC)

Extension of the scope of a case
The topic of the extension of the scope of a case is not covered. It may happen for long, complex cases, and parties are often added in the course of a case. It should be covered in this policy, as it's a major factor in a case. The scope of a case is generally fixed prior to the opening of the case, but when it happens during a case, the evidence and workshop phases may need to be extended, and the parties (including any new party) notified. Also, for extensions of scope other than adding a party, should they be decided by motion ? Are there examples ? Cenarium (talk) 02:16, 25 July 2009 (UTC)
 * There shouldn't be long, complex cases. There should be specific, focused judgements, which make sense both in isolation and in the wider context. The community has a good process for decision-making - the judgments that need to be made to ensure the smooth running of that process are of the type (1) "the community's current position on this matter is....; users editing contrary to that position may be being disruptive"; (2) "this user's behaviour is disruptive; the following sanction is imposed to prevent continuation of that disruption...". In both cases, the decision needs to be made on specific evidence, and should be made at admin level first, with ArbCom reviewing it if asked. Like in a real appeal court, the evidence will generally have already been presented prior to the original decision.--Kotniski (talk) 06:42, 25 July 2009 (UTC)

Timing of the next version?
I'm seeking confirmation as to when this will happen. The ArbCom schedule says 15 September. Tony  (talk)  07:53, 11 September 2009 (UTC)
 * Any update on the expected timing? Tony   (talk)  11:02, 1 October 2009 (UTC)


 * I was just wondering the same thing, and whether there are issues yet to be clarified, added, or removed. SlimVirgin  talk| contribs 11:04, 1 October 2009 (UTC)


 * I recall asking before what this meant, though now I can't find the question: "To appoint[4] those functionaries granted access to privileged information, including the holders of the CheckUser and Oversight privileges." What kind of functionaries have access to privileged information apart from CU and oversighters? SlimVirgin  talk| contribs 11:07, 1 October 2009 (UTC)

Are any Arbs in fact watching this page? Are any of them working on a draft? (In fact, is there any reason why the drafting has to be done by ArbCom at all?)--Kotniski (talk) 11:42, 1 October 2009 (UTC)


 * Yesterday, we started a discussion on what issues we need to prioritise, and so far this has come up high on the list. Expect action in the near future as soon as we can get someone assigned to it. Risker (talk) 13:35, 1 October 2009 (UTC)
 * Newyorkbrad will be addressing this in the coming weeks. Risker (talk) 17:59, 2 October 2009 (UTC)
 * I would advise for any community consultation to be delayed for after the elections. Cenarium (talk) 22:28, 2 October 2009 (UTC)
 * The community deserves better than to have to elect another set of arbs without, for example, knowing whether ArbCom's policy will close the loopholes that allow it to stray into matters of governance, beyond its traditional role as a magistracy. This needs to be sorted out before the candidates stand, so we know what skill-base to expect of them and whether there will be a political layer to their role. Tony   (talk)  02:27, 3 October 2009 (UTC)
 * Yes, definitely. If this is to be done by an Arb, then NYB is indeed an excellent choice, but any one person's time is going to be limited, and policy is always a matter for the community anyway, so it seems reasonable for others who are taking an interest in these matters to undertake drafting work themselves. I don't know how it was intended, but "community consultation" sounds rather patronizing, as if members of the community don't have a right to be involved in the process of developing the policy, only to be asked what they think as a part of the process by which ArbCom develop their own policy in private (which IMO is not how it should be).--Kotniski (talk) 09:11, 3 October 2009 (UTC)
 * I think many editors would be happy if there were a collaborative effort between community and committee, with a lot of the grunt work undertaken by community members; NYB's time (and that of the other arbs) is precious, and would be better spent saying "aargh, don't like this bit or that bit" as the next draft matures than undertaking the third draft from scratch. Whether this should be simply an evolution of the existing "second" draft or prepared on another page is yet to be clarified. I'm certainly of a mind to fix the obvious things in the existing second draft, given that the pre-existing second draft is perfectly recoverable from the history). Your thoughts? Tony   (talk)  11:21, 3 October 2009 (UTC)


 * To be honest, I'm concerned about the attitude that, even though there are editors here and now wanting to edit this in time for the election, we're told only that Newyorkbrad will be "addressing this" in the coming weeks. Guys, please, this is the attitude we voted against last December. Can we please have a respectful collaboration over this? You have one of Wikipedia's best writers here with Tony. If anyone can turn this into a well-written, meaningful policy, it's him. We need things in place before December so we know what to expect from candidates, and what to ask them to adhere to. SlimVirgin  talk| contribs 12:10, 3 October 2009 (UTC)


 * All I mean is that we won't have time to reach a consensus from the community before the elections. Two months is not enough. Yes, I would like that the community be more involved, and also that we consider proposals for reform. But we won't have time in two months to 'adopt' a new arbitration policy - nor seriously consider proposals for reform, or broaden the scope. A 'ratification vote' should come at the end of a process that should involve much more preparation and discussion. Speaking generally, it's better not to get in a rush for 'constitutional' matters. Cenarium (talk) 13:54, 3 October 2009 (UTC)

Resignation
I strongly disagree with this:

"Arbitrators may resign at any time, with or without giving reasons, by communicating their desire to resign to the Committee in writing. Arbitrators resigning voluntarily are entitled to retake their seats at any time during, and for the duration of, the remaining period of their original term of appointment."

This is unprofessional, and we should be aiming for more professionalism. In the real world, if you resign, you resign. Are there any objections to rewriting this part? SlimVirgin talk| contribs 15:54, 2 October 2009 (UTC)
 * Well yes, I've just put an inline footnote to this effect overleaf, supporting Stifle's comment there. The revolving door scenario is embarrassing; the resign-today, back-again-tomorrow undermines the authority of the voters and the dignity of the Committee. Tony   (talk)  16:10, 2 October 2009 (UTC)
 * Agree with SV - resignation should mean just that. DuncanHill (talk) 16:12, 2 October 2009 (UTC)
 * Newyorkbrad's resignation and return is an excellent example of why that Arbitrators can resign and return. I strenuously object to any wording that would have kept him off of the Committee. FloNight&#9829;&#9829;&#9829; 13:38, 3 October 2009 (UTC)
 * I think it would be possible to craft policy so that you both could be satisfied. For example, NYB could have been required to stand, by himself, for re-election after resigning. I believe he would have passed such an election with ease. IronDuke  23:37, 4 October 2009 (UTC)

If a new draft is imminent, what is the point in continuing to work on this draft? Whilst I applaud the effort being put in, any changes will either be overtaken by the new draft or delay the publication of the new draft for the suggestions to be incorporated. Wouldn't it make more sense to get the new draft published ASAP and then work on that version? EdChem (talk) 16:17, 2 October 2009 (UTC)
 * Please see Risker's comment just above this section. It may be that the Committee is yet to allocate someone to the task; in any case, a reasonable assumption is that this second draft will be the basis for the third. Tony   (talk)  16:20, 2 October 2009 (UTC)
 * Anyone can draft policy - it's normally done collectively anyway. It's not ArbCom's duty, nor their privilege.--Kotniski (talk) 17:08, 2 October 2009 (UTC)
 * To avoid COI, it would make more sense if ArbCom members didn't get involved, except to advise when something is unworkable. SlimVirgin  talk| contribs 17:47, 2 October 2009 (UTC)
 * SlimVirgin, I will simply say that someone who has never worked on the committee writing the policy is like a blind man writing a driving manual: it can be done, but it probably won't be very realistic. Risker (talk) 18:05, 2 October 2009 (UTC)


 * You may be right, but having the policy written by current members of the committee risks it being self-serving. The point about members who've resigned being allowed to return is an example of something that's not in the interests of the project. Ideally (in my view), members and non-members would say on talk what they want to see in the policy, and non-members would write it. SlimVirgin  talk| contribs 18:57, 2 October 2009 (UTC)
 * Sounds very reasonable to me: arbs are complaining they don't have the time to do this kind of thing. Fair enough, the community can and should step in. Tony   (talk)  02:24, 3 October 2009 (UTC)


 * SV, are you saying that the returns of arbitrators Newyorkbrad, Coren, and Rlevse were "not in the in the interests of the project"? That's a legitimate position to take, but you should be clear if you are taking it.  If resignations were irreversible, there would be ten sitting Arbitrators.  Just food for thought.  Happy ‑ melon  13:51, 3 October 2009 (UTC)
 * No, I think SV is saying that the current policy encourages resignations without proper consideration, since the resigning arbs know they can freely walk back through the revolving door tomorrow. It's not fair to voters, who put their trust in arbs to see through their term or to resign if required by circumstances or policy: resignation is a serious matter and should not be treated lightly. Tony  (talk)  05:01, 4 October 2009 (UTC)

(od) Tony, I have a great deal of sympathy with this position and dislike the ever-revolving door. However, arbs resign under all sorts of circumstances (real life pressures/priorities, being the main one). There have also been instances of arbitrators being driven out of office by members of the community badgering them. The way forward is probably that resignations may not be withdrawn months after being made but the ex-arb may request, at the discretion of the committee, to be co-opted for the remainder of their original term. Roger Davies talk 06:33, 4 October 2009 (UTC)
 * Roger, that sounds better than the current situation. However, an environment in which members of the community badger and drive out of office needs to be addressed, too. And I am very concerned at the burn-out factor, and specifically how streamlining and delegation might be judiciously applied to make the workload more reasonable. I've heard horror stories of a hundred emails waiting to be dealt with upon rising each day. That is frightful, and no wonder we lose good content writers and leading admins down a black hole when they become arbs. Does it have to be so full-on? Tony   (talk)  08:01, 4 October 2009 (UTC)
 * Tony, there was a big change in the way that the Committee works this year as compared to last year. There were some rather serious issues with inter-arbitrator communication last year, something that all of us took very seriously this year, so we do tend to keep each other more informed and actively participate in discussions. We have actually started the delegation process, with the AUSC managing concerns about checkuser/oversight use and collection of data related to those two permissions; the current subcommittee includes three community members, and elections will be held in the very near future to replace/confirm the initial appointees. Members of the Arbitration Committee are not the principal checkusers/oversighters anymore: we have established an election process and these elected checkusers and oversighters carry out the bulk of the work. We have initiated a ban/block review subcommittee (currently 3 arbitrators) that shifts that section of the workload to the arbitrators assigned for the period; this is an area we will probably revisit in the next term to see if and how we can integrate community members into the panel. We've done some experiments with different case handling processes to see what works well; these will continue and we can review them more closely next year.
 * None of this changes the fact that the workload remains heavy, and I expect that it will continue to be so. It is not a 5-hour-a-week role if done well; on average, there's at least 10 hours of reading a week, and that doesn't count the responses or drafting processes, and most arbitrators will take on additional short-term or long-term tasks (subcommittees, drafting specific cases, mail management, etc) that will expand the number of hours involved.
 * Having said all that, I agree with Roger that there is a level of badgering that can be highly demotivating, and it seems to be an accepted fact that no matter the decision made by the Committee on a case or an issue, it will be dissected and criticised. Risker (talk) 08:49, 4 October 2009 (UTC)

Perhaps it would be sensible to formally recognise that there will, on occasion, to be need for arbitrators to take a leave of absence? This should be distinct from resignation, as I agree that resigning and returning - especially if it happens regularly - is inappropriate. This approach would also avoid the criticism that would likely follow if the Committee co-opts members as a means of them 'un-resigning'. EdChem (talk) 09:38, 4 October 2009 (UTC)


 * Roger, can you give some examples of Arbs being badgered to leave unfairly? I know that Brad was outed and felt he had to leave for that reason, but then returned, though I don't know what changed his mind. But other than that, I can't think of an example.


 * If this were a volunteer position that no one wanted to do, we'd have to bend over backwards to accommodate those who did step forward. But that's the opposite of the case here. Every year there are lines of people wanting to join. In an atmosphere of stiff competition like that, it isn't really fair of people to take a position, then leave it as they see fit, to be picked up again whenever they see fit. This gentlemen's club atmosphere is something that a lot of us want to see an end to, which is one of the reasons you were elected, Roger, because people believed you'd be part of a process of professionalization.


 * Perhaps, as Ed suggests, we could distinguish between resigning under a cloud (including pre-emptively), and taking a leave of absence for personal reasons, with limits on how many of the latter there can be, and how long they're allowed to last. SlimVirgin  talk| contribs 11:43, 4 October 2009 (UTC)
 * I think if you resign you resign (though the resignation might only be "accepted" after a certain amount of attempted persuasion to change mind has taken place). If you want or are compelled to be inactive for a certain length of time, you should give notice of that fact. That way the remaining Arbs know where they stand, and can make an informed decision on the need for supplementary elections if the number of current and active Arbs is too low. --Kotniski (talk) 12:16, 4 October 2009 (UTC)


 * Do these concerns about burn-out mean we should ideally be electing a larger number of Arbs in future? I've always been concerned that the Committee has seemed too small for the workload, but I don't know who'll be making the decision this year about how many places to fill or create. SlimVirgin  talk| contribs 12:35, 4 October 2009 (UTC)
 * No, increasing the size of the Committee has increased the work load because of the increase in the number of emails and posts that we need to read. Every discussion we have ends with agreement that more arbs is not the answer. The burnout is not primarily from the work load but other factors. FloNight&#9829;&#9829;&#9829; 13:04, 4 October 2009 (UTC)
 * It makes sense that the workload might increase when more people join. Can you give us an idea of what the root cause of the burnout seems to be? SlimVirgin  talk| contribs 13:12, 4 October 2009 (UTC)

Arbitrary section break
←Kotniski, I think it would be unwise to conflate the governance issues with this discussion, which may end up going nowhere if it becomes too wide. I think a lot of people are anxious that the new draft continue to omit the wording that still leaves open the potential to stray into broader governance. However governance is to evolve on WP, it's a different issue from ArbCom's roles as set out at the top of this draft.

Slim Virgin's comment: "I don't know who'll be making the decision this year about how many places to fill or create." In a major statement on 9 September, Jimbo has undertaken to "keep the size of the ArbCom the same, unless there are compelling arguments offered now and with majority support of the sitting ArbCom." Thus, we should all know before the election how many candidates who will take up seats on the Committee after it. I took this to mean the numbers would be topped up to 18, as after last December's election. Jimbo added: "I want my "appoint" role to be purely ceremonial this time around."

It would indeed be useful for arbs to air the reasons for burnout. I'm guessing (1) rather too loose hearings procedure and redundant text on decision pages; (2) insufficient control over the admissibility and length of evidence (although I believe that has been partially addressed); and (3) the need to develop more effective methods of delegation for certain tasks, both to (i) bodies appointed by ArbCom and comprising an arb or two for supervision plus community expertise, and (ii) the clerks. I know there are moves in these directions: I encourage further moves.

It strikes me as unnecessary to have every active and non-recused arb ruling on every case taken on by the Committee; but I don't know how transparency, trust and fairness could be guaranteed in the allocation of arbs to cases. Eighteen, the number at the start of the year, seems far more than necessary for a ruling (even 12 would be a lot, given six recusals). Perhaps the established practice of being intermittently "non-active" could become the norm, planned ahead, to guard against burnout. In addition, one or two arbs might be inactive from the main cases for, say, six months at a time, to chair the newly established AdminReview Subcommittee (here, I'm only half-joking).

Why are so many emails involved? And, as Risker says, 10 hours of reading a week? Phew. Is it not possible to filter more correspondence through the clerks. The clerks should be there to protect the arbs not just from the need to perform the clerical duties, but to act as a buffer. I'm not saying the system shouldn't be open, transparent and accountable, but the more focused the arbs can be on writing and voting on judgements, the better for everyone. Not the least concern is that a little distance is good for maintaining neutrality, and the appearance of it. Tony  (talk)  15:19, 4 October 2009 (UTC)
 * Mail cannot be delegated to the clerks, it is just the nature of the mail. Much of it is discussion of an initiating email, or about a specific case. (Remember that last year, there were issues about arbitrators *not* giving feedback to each other, so this year we have been very communicative and ensuring that we aren't going off on a lark of our own before acting.) There is also the issue of privacy; the clerks have not been elected to their positions but instead are appointed, and they are not Foundation-identified; the amount of email that involves non-public information is significant. Much of the reading relates to cases, and arbitrators need to do it themselves so that they are informed to vote in the cases.
 * Personally, I would oppose the idea of creating arbitrator panels. The community, in its wisdom, has elected a very diverse cross-section of editors to assume responsibilities on the Committee, and I don't see a way to developing panels that would be fair to those coming before the Committee in cases. Recusals are common, and on almost all cases a few arbitrators are inactive, which would unbalance any panels. I think all of us have taken some "inactive" time for personal reasons (travel, work or family related, or sometimes just a break from arbitration work), or because we are deeply involved in another area of Committee responsibility (such as when I was inactive on cases because I was handling the Checkuser/Oversight election tasks, much of which was to lay a foundation for future elections).
 * As to the burnout issue, I can speak only for myself here. We seldom receive positive feedback, but are invariably on the receiving end of negative feedback for every case we manage. I have never been on the receiving end of so much bad faith as I have been in the past nine months; since receiving an 87% support vote, my motivations, abilities, and level of community support and trustworthiness have been questioned almost incessantly by certain individuals and groups. An editor I know well once described me as a "realist Pollyanna", but there have definitely been a few days when I've wondered why I should continue to put up with this. Risker (talk) 16:23, 4 October 2009 (UTC)


 * Question - Do arbitrators share meaningful mail with each other? Or is it common, as happened in the Law situation, for an individual arbitrator to carry out a prolonged email interaction solo? Also, would it not be possible to have a level of security invested in some clerks to filter the mail, much as U.S. Supreme Court Justices have clerks who significantly filter the information an individual justice must address, but this filtering  does not affect a justice's ability to do the job.  I have read that they receive 8,000 requests a year and only consider around 60. Each individual juustice does not go through all this paperwork.   — mattisse  (Talk) 16:43, 4 October 2009 (UTC)


 * Risker, I can certainly see why working in that atmosphere would lead to burn-out. The other danger is that it will lead to "thin blue line" syndrome, where Arbs start to feel, consciously or otherwise, that it's them against the world. That leads to them being overly sympathetic to each other's weaknesses, which in turn decreases the community's trust. Rinse and repeat. The only safeguard is to put checks and balances in place, which both the Arbs and the community can rely on, but there has been resistance to that in the past, ironically from the Committee itself, even though they're one of the groups that would most benefit from it.


 * In your view, Risker, would increasing the number of Arbs increase the workload? It seems to me that it would increase it in the sense of there being more people to communicate with, but with, say, twice the number of Arbs to deal with correspondence and cases, would it not lead to an overall decrease in the work expected from each person? SlimVirgin  talk| contribs 16:56, 4 October 2009 (UTC)
 * There are more checks and balances in place now, largely put in place by this year's Arbitration Committee, than there ever have been before. And no, doubling the size of the Committee would mean having twice as many people to seek agreement/feedback from, and would further delay decision-making, which is not optimal. I waver between thinking that expanding the committee to 18 seats this year was a good thing or a bad thing, and adding another 15-18 voices to the discussion would make it useless.
 * Perhaps a more to-the-point comment is that there are unrealistic and conflicting expectations from individuals and groups within the community, who expect us to be both the equivalent of the Supreme Court and also to have no binding or significant impact on anything other than the narrow confines in which the Committee is tasked. Risker (talk) 17:33, 4 October 2009 (UTC)


 * Well, what should our expectations be? This is what I am asking. Should you not stick to the confines in which the Committee is tasked? If not, then what are the boundaries? I am confused. — mattisse  (Talk) 17:44, 4 October 2009 (UTC)

Risker, can you list what you see as the checks and balances? I can think of three that ought to be, but aren't:
 * 1) We have elections for CU and oversight, with candidates being approved in advance by ArbCom. The policy says, "ArbCom will carefully vet all applications." That's an improvement on ArbCom choosing the candidates themselves. But we currently have a situation where several people were approved for the August 2009 election, even though one Arb knew those people were aware of, or had taken part in, The undertow gaining adminship by evading a ban and lying to voters. That Arb didn't say anything, so the candidates were approved, and some were elected. They would not have been elected if people had known they were involved in this, even passively. And one of these people reportedly used oversight to remove an edit around August 20 (the election ended on August 10) that said The undertow was Law, and that person is now being investigated by the audit committee.
 * 2) The audit committee oversees CUs and oversighters. But there are three Arbs on that committee, including at least one who is a friend of some of the above. Members will need strong personalities to be able to withstand being influenced, and a lot of personal insight, because we're often the last ones to be able to judge when we're being impartial, which is why recusal is so important. We shouldn't put anyone in that position when we have hundreds of uninvolved editors out there able to do the job.
 * 3) A ban appeals committee has been formed, but again one that consists of Arbs, who will only very rarely decide publicly against current colleagues. Why do we continually choose from such a very small pot? This is the opposite of having a system of checks and balances in place. SlimVirgin  talk| contribs 18:19, 4 October 2009 (UTC)


 * This caught my eye: "that person is now being investigated by the audit committee" ... has that been publicly announced somewhere? If not, how do you know this? ++Lar: t/c 20:17, 4 October 2009 (UTC)


 * It was posted on one of the pages we've been discussing this on, I believe by Flo. Even if it hadn't been, why would I or anyone else not know? The details of what the audit committee might examine may be private, but the fact that it is looking can't be. SlimVirgin  talk| contribs 20:35, 4 October 2009 (UTC)
 * If you happen to recall, a link would be appreciated, because I didn't spot it in a quick check of contribs. As per usual at en:wp, discussion sprawls, (just by the nature of it) and it gets worse as it goes on... so thanks for any hints you can provide. As to the second part: If it hadn't been posted publicly yet, then presumably very few people would know. It's not listed here (yet?) Arbitration_Committee/Audit_Subcommittee/Reports but I think that's only for completed things. ++Lar: t/c 20:48, 4 October 2009 (UTC)
 * It's here. SlimVirgin  talk| contribs 20:53, 4 October 2009 (UTC)
 * Thanks! ++Lar: t/c 21:41, 4 October 2009 (UTC)


 * SlimVirgin, I am finding you are not only assuming bad faith but implying that people deliberately went out of their way to try to place people in positions of trust when they should not have been. There is no basis for you to make that implication. Jennavecia passed Arbcom vetting but did not succeed at the community vote level: the check and balance worked. I am aware of the review of the oversights to which you refer; the oversighter involved was elected in February. Simply put, you are saying that the Arbitration Committee should have been making decisions earlier this year based on information that we did not have until last week, and applying standards of behaviour that have not yet been articulated by the community even today to actions carried out in the past. Traveling in the same circles as someone, having worked in similar areas, holding similar wikiphilosophies, or attending the same wiki-meetups does not constitute close personal friendship between two editors, or even imply that they are incapable of making impartial decisions related to those editors. Guilt by association is not an acceptable practice on Wikipedia, and it is time for you to desist. I genuinely doubt your contention that there are "hundreds" of editors out there able to do the job, because I do not believe there are hundreds of editors that either the community or the Arbitration Committee would support in these roles, nor have there been hundreds of editors who have expressed interest in these roles. Fewer than 100 people have put themselves forward for privacy-related roles (Arbitration Committee, Checkuser, Oversight) over the past two years. We shall be holding elections in the near future for the community seats on the AUSC, and it will be interesting to know how many editors put their names forward.
 * I do not understand your making a connection between the Ban Appeals subcommittee and "deciding against former colleagues". No former arbitrators have ever been banned or blocked for a long period of time, to my knowledge; if they have, none have appealed to the BASC. All decisions to lift bans are announced to the community, and several have been discussed with the community in advance of a final decision. There is a recognition that the ban review process does require further examination (it is on our agenda); however, as there are less than two months before the election of new members begins, it is probably not reasonable to make major changes to this process that would not be able to be instituted until after the next election.
 * Other checks and balances include publication of full motions, including recorded voting; centralised and easily accessible announcements for transparency; the community election of checkuser and oversight candidates; the publication of a six-month summary. Hope this is helpful. Risker (talk) 22:19, 4 October 2009 (UTC)

Risker, this isn't a question of assuming good or bad faith. We need to move right away from that mindset. There are legitimate questions, they're being asked politely, and we hope to get answers. There's no point in insulting people are who asking. I remember a time you were in the position of wanting clarity from people in positions of power.

1. You write: "Jennavecia passed Arbcom vetting but did not succeed at the community vote level: the check and balance worked." But not thanks to the ArbCom. Jennavecia got through an ArbCom vetting for oversight when one of the Arbs knew she'd been engaged in recent dishonesty to help a friend gain access to the tools. That shouldn't have happened, so I hope the Committee will publish some information about how you intend to tighten the vetting process.

2. Jennevecia apart, there were at least two oversighters associated with this situation (in the sense of being part of that group and of knowing about The undertow/Law's adminship) who were elected in August. Both were vetted by the ArbCom. So in those cases the checks and balances failed entirely.

3. Editors who don't move in the right circles won't get elected to these positions, so they don't try. You would normally need to be on IRC, or Wikipedia Review, or be a member of the Bathrobe Cabal, or some such. That's the pity of the thing here, and I have no answers.

4. My point about the bans committee is that people banned by ArbCom will be appealing to the same body that banned them, so it's not much of an appeal, because Arbs are usually reluctant to speak out against other current Arbs. It works in the case of a community ban appeal, but not in the case of an ArbCom ban. Why not have an appeals body that is independent of ArbCom?

5. You write: "Traveling in the same circles as someone, having worked in similar areas, holding similar wikiphilosophies, or attending the same wiki-meetups does not constitute close personal friendship between two editors, or even imply that they are incapable of making impartial decisions related to those editors." The question here is: how do you know that? The whole point of having checks and balances in place is that we don't have to wonder. An enlightened institution will say, "Oh look, the Arb who seems to be a wikifriend of the person now being investigated by the ArbCom ought to recuse from that investigation just to be safe, then we don't even have to wonder whether he could have been fair." An institution with problems will say, "Oh for god's sake, witchhunt, bad faith, crucifixion, wikilawyering, for heaven's sake shut up." Please try to see that the people asking questions about this situation are doing so in good faith. Anyone who didn't care about Wikipedia wouldn't waste their time on something like this. SlimVirgin talk| contribs 23:02, 4 October 2009 (UTC)


 * Sorry SlimVirgin, but you are continuing to assume bad faith here, especially in Point #5. The answer to your question is right in our five pillars: Assume good faith. This has been a Wikipedia policy almost since the inception of the encyclopedia, and it is one that I expect every editor to follow. Unless people have a reason to believe that something is amiss, we are all expected to assume good faith. In your points #1 and #2, you are asserting a standard of behaviour that has never been articulated on this project, and even after the revelations of the past week it is still not in any way a fully accepted position: that of a requirement to report certain matters, either to the Arbitration Committee or to the community. There is no policy that comes anywhere close to what you are proposing, and discussions about including such behavioural standards have just begun in the last few days. The Arbitration Committee cannot impose such a standard on the community; we are not the policy makers.
 * The current Arbitration Committee has indeed overturned bans from previous iterations of the Committee, and has even overturned at least one ban of its own.
 * As to your premise that the only way to be accepted as a candidate is to be a member of certain circles, I think that will come as a surprise to many of the Checkuser and Oversight candidates we have brought before the community in the last two elections, not to mention those who did not make it past the vetting stage. Again, you have assumed bad faith that someone outside of those particular spheres would have no chance; several of the successful candidates have none of those links. Risker (talk) 23:39, 4 October 2009 (UTC)


 * You seem to be implying that, because it isn't written down anywhere, it's not clear that Arbs should report to the Committee that people have helped a banned editor gain access to adminship, and it's assuming bad faith to go on about it. Look, imagine this: you know that Jayjg, a wikifriend of mine, is topic-banned from I/P articles. Imagine he were to set up a new account, with my knowledge, so he could start editing I/P articles again. I then successfully nominated his new account for adminship, with the knowledge of an Arb and an oversighter, who oversighted a post saying who Jayjg was. A bunch of people on IRC knew about, and later some of them were given oversight too, but none of them told the Committee, or did anything to stop it, not even its own member. I can only imagine the howling that would ensue. I wonder how many seconds it would be before I was desysopped, and the oversighter de-oversighted, and rightly so. Good faith/bad faith wouldn't come into it. SlimVirgin  talk| contribs 23:57, 4 October 2009 (UTC)
 * This "banned user" meme needs to stop. In fact, on 16 June 2008, the Arbitration Committee announced a block of User:The undertow, and explicitly stated that it was not a ban. There was an implied understanding that when the matters that led to the block were resolved, he could return. An unblock request was made to the Arbitration Committee, and no response to that request was made. There are many issues here, but none of them involves banned users. Risker (talk) 00:42, 5 October 2009 (UTC)
 * Risker, I have no knowledge of the content, and I can assure you that a disinterested party reading this will query why there are multiple accusations of "bad faith". It strikes me that your points could have been put amply without such accusations. Tony   (talk)  03:15, 5 October 2009 (UTC)
 * Tony, when someone says, to paraphrase, that the only way to be a checkuser on this project is to be a member of X group or Y group, that is an absolutely crystal clear example of bad faith. It tars all of the people who hold those permissions, and everyone who has expressed an opinion about them one way or the other. It is patently and obviously not true. I am shocked that it does not disturb you. Risker (talk) 03:35, 5 October 2009 (UTC)


 * Risker, a reasonable person could argue that you've been assuming bad faith yourself throughout this discussion, but it's not a helpful way of looking at things, because it means we go round in circles getting more annoyed with each other. ("You're assuming bad faith!" No, you are!!") People are simply asking questions and expressing concerns. They're legitimate questions. You don't like them. That's legitimate too. Part of the problem here is the massive gap in perception. Any dialogue that bridges it is a good thing.


 * As for becoming a functionary, I think it would be extremely hard in this climate to become one without belonging to one of the usual groupings, but that's something that's easily tested. After a year of having elections, we can look to see who's been appointed, and if I'm wrong, I'll be very happy to retract. SlimVirgin  talk| contribs 04:05, 5 October 2009 (UTC)
 * I agree with Risker that the "banned user meme" is not helpful and should not be used as a rhetorical device. ++Lar: t/c 13:44, 5 October 2009 (UTC)


 * It would be good if we could move on here to deal with the policy, but I want to reply briefly that The undertow was banned for nine months within any definition of ban that I've ever heard, including the one in the banning policy, a ban that was technically enforced via a block, with the option (as always) of approaching the Committee to request that it be lifted when he was willing to comply with whatever the issue was. But I don't see what difference it makes whether we say he violated a block or a ban or a doodah, because the issue is that he violated an ArbCom decision and stood illicitly for adminship with the help of other admins. But I really think it's time to move on to the general lessons learned, and on this page, how to shape a policy that helps. SlimVirgin  talk| contribs 14:32, 5 October 2009 (UTC)
 * I take your point, however I still agree with Risker. Terms can trip things up and get in the way. Unfortunate, but there it is. Perhaps just stop using it and not fret too much, and move on to substance as you say. ++Lar: t/c 18:25, 5 October 2009 (UTC)


 * Both of you: please let's keep it absolutely non-personal. Here we are discussing how to improve a major structural part of the project, which should be exciting and positive, even though though the push and pull will probably result in some disappointments for every individual. I don't want to understand what should be water under the bridge, but can we just be kind to each other (to use a Jimbo phrase)? Risker, I really respect the job you do on ArbCom—I would hate the burden—but I was counting on you to be open to change, even though the daily arb-stress must make that difficult. Can we get back to the topic, which is how to turn this draft into a viable proposal for a new policy? Tony   (talk)  04:22, 5 October 2009 (UTC)


 * Yes, indeed, good idea. What's the next issue we need to look at? SlimVirgin  talk| contribs 04:43, 5 October 2009 (UTC)

Real life is not that strict
This proposal, especially it's first draft, is stricter than most professional COCs, but if you're willing to pay me to do this, remember that public sector arbitrators start at about $250K a year. <span style="font-family:Verdana,sans-serif"> — <b style="color:#060;">Rlevse</b> • Talk  • 23:34, 2 October 2009 (UTC)
 * Yes, this is "just a website". Why do we all take it so seriously and act as if it matters how editors and admins conduct themselves? Seems ridiculous to me. Regards, — mattisse (Talk) 23:48, 2 October 2009 (UTC)
 * Rlevse: so because we're not paying you, a lower standard of conduct applies? Tony   (talk)  02:21, 3 October 2009 (UTC)


 * We can act professionally without money being involved. The question is whether there are lots of people lining up to join the ArbCom, and for reasons I've never quite understood, there are. The fierce competition for places makes them valuable, which means we can request high standards. Rlevse, is there anything in particular in this that you feel is unreasonable? SlimVirgin  talk| contribs 02:27, 3 October 2009 (UTC)


 * Speaking as someone who has a voluntary job which involves managing other volunteers, I am sick of the constant attempts to use "we're just volunteers" as an excuse for low expectations or poor behaviour. Voluntary status is in no way incompatible with professional attitudes and standards, and in my experience the most successful voluntary organizations are those with the highest standards expected of their volunteers. DuncanHill (talk) 12:33, 4 October 2009 (UTC)


 * Strongly agree. Volunteer positions are often necessarily associated with requirements of professional attitudes and standards, the same as if one were being paid, certainly in my profession. I wonder that Wikipedia seeks to compile a reliable and respectable encyclopedia, on the level of the Encyclopedia Britannica, and yet any call for behavioral standards for the contributors is met with the "we're just volunteers" excuse. — mattisse  (Talk) 12:45, 4 October 2009 (UTC)

Dead candidates and interim elections
In my perusal of the start of this draft, the following issues stick out.

"'Candidates must ... not be legally deceased in their country of residence.'"
 * Deceased

Why is this there? Cadavers?

"'The Committee may call interim elections, in a comparable format to the regular annual elections, if it determines that arbitrator resignations or inactivity have created an immediate need for additional arbitrators.'"
 * Interim elections

Three issues:
 * 1) Would these "interim" elections be "advisory" to Jimbo, as are the annual elections? (The "comparable format" of the elections is a different matter (whether secret ballot, the system of voting, preferences used, etc).
 * 2) The text doesn't say how long the term of any such arb would be (until 31 December of that year?).
 * 3) If that were the term, it would not worth holding "interim" top-up elections after the first few months of any year, since the term would be too short for the disruption caused. Who would bother in August?

I corresponded with Kirill about this earlier in the year, suggesting that the terms of candidates elected at interim elections last until 31 December of the following year, not the current year. The response was that the length of such terms had not been thought through, and yes, that it should be thought through. Tony  (talk)  02:54, 3 October 2009 (UTC)


 * There was some concern that a zombie was serving on the Committee so this passage was added. The matter is resolved so the wording can be removed. FloNight&#9829;&#9829;&#9829; 03:01, 3 October 2009 (UTC)


 * Someone, Cirt(?) compiled a list of arbitrator requirements of other language arbcoms on metawiki. All others had a one year term of office. Any reason why English is different? One year seems much more reasonable. — mattisse  (Talk) 13:05, 4 October 2009 (UTC)

Community assistance in the evolution of this page
I address this to both the community and the committee: given that the real-life and on-wiki demands on NYB and the other arbs are, to put it mildly, getting into burn-out territory, it seems practical for the community to work with the committee to turn this page into a proposed policy that can be put to the community for ratification ahead of the ArbCom election in December. Despite Slim Virgin's kind words above, there are many good writers available to assist.

In addition, Kirill (no longer an arbitrator) and his colleagues need to be thanked for getting it to this state.

I wonder whether the following timetable is realistic and acceptable:
 * Right now: advertise this process at the Village Pump and Centralized Discussion
 * By October 31: complete the proposed policy and decide on the method of ratification (voting method, threshold for ratification)
 * November 1–14: watchlist banner call for final community and committee input
 * November 15–30: community ratification by vote
 * December 1: in force if ratified (i.e., before the election process starts). Tony   (talk)  12:25, 3 October 2009 (UTC)
 * It will depend on whether the re-write is found that is generally acceptable to most of the people commenting, I think. We need to have some good so we need to make sure that it is not rushed by deadlines. If the wording is not getting good consensus, then we need to push it back until we get something that gains general consensus before it is put to a vote. FloNight&#9829;&#9829;&#9829; 13:09, 3 October 2009 (UTC)


 * It looks like a reasonable timetable given that much of the work is already done. If it's not ratified, we can try again, but it makes sense to aim for ratification before the election. SlimVirgin  talk| contribs 13:17, 3 October 2009 (UTC)
 * Yes. FloNight, I think a stated schedule is a good springboard, and if there are no big snags it might just work. Of course, we have to be prepared for the worst—that it can't be done in time for the election—if it fails to evolve smoothly. Let's hope? Tony   (talk)  13:27, 3 October 2009 (UTC)
 * I agree with FloNight, we should not be rushed by deadlines for such 'constitutional' matters. Two months will not be enough to review the policy, and broaden the scope, consider reforms, etc. I think we should aim for a ratification vote in early 2010. Disclosure: I intend to propose various reforms related to DR/ArbCom, and I think we should call for proposals in this sense. Cenarium (talk) 14:06, 3 October 2009 (UTC)
 * If we're going to have a long process of consideration of constitutional matters, perhaps we could widen the scope (as Cenarium suggests, in fact) to include the whole area of governance or "how disagreements are sorted out". Experienced Arbs should be in an excellent position to see what's going wrong, what mechanisms need to be added or changed, to make the encyclopedia-building process a smooth and productive one. (Because these vast cases that come to ArbCom are clear evidence that something has gone wrong in each instance - there ought to be regular processes for resolving disagreements and reforming/eliminating disruptive editors without generating whole reams of time-wasting drama on frequent occasions.)--Kotniski (talk) 12:25, 4 October 2009 (UTC)

Issues
Looking through issues to be dealt with:

"The members of the Committee are appointed by Jimbo Wales, in his role as project leader, following annual advisory elections, the format of which is decided by the community."

I'd like to remove the reference to Jimbo, because he has said he sees this from now on as just a ceremonial role. So it would read: "The members of the Committee are appointed following annual elections, the format of which is decided by the community." It leaves open the question of how they are "appointed," but I feel it's neater to leave that open than to raise the issue of Jimbo's role.

Any objections? SlimVirgin talk| contribs 14:40, 5 October 2009 (UTC)
 * Well someone has to appoint them; I don't think we can dodge the issue. We could say "project leader" instead of Jimbo, having defined that term somewhere. (The alternative is for the current ArbCom to appoint the new one, if that's what we want and the Great Man agrees for it to be done that way.)--Kotniski (talk) 15:00, 5 October 2009 (UTC)
 * If the number of arbs to be established after each election is set down (it's 18 at the moment, apparently) and if the voting procedure displays the candidates in order of their popular vote (as the votes have until now shown), an "appointer" is unnecessary.
 * However, the RfC on this matter only a few months ago did not pass, even though 40% of participants wanted a change in the appointment system. I suggest that Slim Virgin's text is a good way forward, since it retains the notion of appointment, but allows the community to be flexible in future years. Tony   (talk)  15:39, 5 October 2009 (UTC)
 * An appointer is necessary in the sense that someone needs to post the final results, notify candidates etc, and this is ideal ceremonial role for Jimmy. It could equally be handled by a presiding officer, but it might be more trouble than it's worth getting consensus on who that is and how they're appointed.  Roger Davies  talk 16:34, 5 October 2009 (UTC)


 * Perhaps we could say Jimbo will announce the appointments (rather than make them), but not say as "project leader," and not describe the elections as "advisory." SlimVirgin  talk| contribs 16:39, 5 October 2009 (UTC)
 * That sounds good, though for the long term it makes more sense to talk of a project leader than of "Jimbo" all the time, since we hope that Wikipedia will continue to function more or less in the same way after Jimbo moves or passes on...--Kotniski (talk) 17:41, 5 October 2009 (UTC)


 * How about this? "Members of the Committee are elected annually by the community, and the appointments announced by Jimbo Wales in accordance with the election results. The format of the elections is decided by the community. SlimVirgin  talk| contribs 19:23, 5 October 2009 (UTC)
 * Sounds good, Slim. Tony   (talk)  01:48, 6 October 2009 (UTC)

Alternative draft
I've been busying myself over the past few days making a rather more radical draft of this policy, to try to address what seem to be the major things wrong with ArbCom as it currently stands. Not sure I've quite succeeded yet, but should anyone be interested and wish to comment, it's at User:Kotniski/ARB.--Kotniski (talk) 16:50, 5 October 2009 (UTC)
 * By the way, I read in the Signpost that German and Spanish Wikipedias have decided to abolish their arbitration committees. Maybe we could consider that option? Does anyone know what those WP's do by way of ultimate authority, or do they get by without? (I was thinking that maybe a convocation of all administrators might be sufficient to issue final rulings in the matters with which ArbCom deals.)--Kotniski (talk) 11:12, 6 October 2009 (UTC)
 * We (ArbCom) supports having more admins intervene in situations earlier. That is the reason that in our rulings we have introduced remedies that focus admins attention to particular articles or topics that are long term problems. But admins not resolving disputes is the main reason that ArbCom is needed. Disagreement among admins and lack of time or interest are key reasons that a situation escalates to ArbCom. Additionally, ArbCom handles privacy related issues. So, I don't think that we can do away with ArbCom anytime soon. FloNight&#9829;&#9829;&#9829; 11:27, 6 October 2009 (UTC)
 * Sorry, I wasn't quite right about German Wikipedia (read for yourselves if you want the facts). But if people have time to be arbitrators, wouldn't they also have time to perform the same tasks, but as administrators? And my impression is that the admin world wastes a lot less time over procedure than the arb world.--Kotniski (talk) 12:44, 6 October 2009 (UTC)


 * I occasionally look at the French ArbCom pages. What strikes me is that they have much fewer cases and the individual cases are much smaller. The pleadings are more succinct than ours and the tone is more laid back. They also don't seem to get the nastiness we often see here.  Roger Davies  talk 13:39, 6 October 2009 (UTC)
 * I have a feeling anglophone culture is more quarrelsome on the surface, and that the other WPs look upon us as a dynamic, unruly lot: they're probably right. I think the same is true of the English language (Clive James once described it as "big and baggy"). The French MoS, for example, is small compared with ours because French is more centrally controlled in the first place (the Academie Francais, even language legislation, as hard as it is for us anglophones to believe); I bit my lip before saying "rigid".
 * We believe we need a strong ArbCom. However, the French evidentiary succinctness is something to emulate; I believe ArbCom has introduced stricter rules on the length (relevance?) of evidence. I hope so. Case deadlines would also be good for the psychological health of the parties and the ability of arbs to plan their lives. Short and strict would help to prevent cases from becoming such big deals. Tony   (talk)  14:22, 6 October 2009 (UTC)
 * I certainly agree with that last part. Arbs (and clerks) must be quite clear about what they're arbitrating on, not tolerate any incivility on the case pages, and make sure everything stays very much on topic. And be prepared to get involved in the dialogue so as to steer it towards finding a resolution.--Kotniski (talk) 14:42, 6 October 2009 (UTC)

Code of conduct
I removed this from the page, because it deals with a much narrower area. If both that and this are passed, then perhaps they can be merged, but for now, it might be quicker to get them passed separately. Any thoughts? SlimVirgin talk| contribs 00:39, 6 October 2009 (UTC)


 * Yes. Strongly opposed to splitting out the two. The area of overlap is enormous and it's entirely arbitrary which is a code of conduct matter (which has considerable instruction creep) and what is a policy matter, especially as you seem to be using the CoC draft to include major matters of policy. The longer these are separated, the longer and more convoluted the discussion, and the higher the likeihood of poor cross-referencing.  Roger Davies  talk 00:44, 6 October 2009 (UTC)


 * Roger, the code of conduct has been proposed as a policy. It may not be ratified, but it is currently a proposal. The intention is that it will be short, and will deal with what's expected of Arbs in very broad terms (honesty, integrity), and what to do when they want to, or have to, leave.


 * This page, on the other hand, covers a great deal more than that, and will therefore take longer to ratify. That's my main concern. SlimVirgin  talk| contribs 02:13, 6 October 2009 (UTC)
 * I see no benefit to having separate Code of Conduct and ArbCom policy. The ArbCom policy already touched on many of the things that you have been adding to the CoC. It makes much more sense to get a well thought out ArbCom policy to be ratified than work on two overlapping policies. FloNight&#9829;&#9829;&#9829; 02:27, 6 October 2009 (UTC)
 * I support the integration of the two, as long as the code of conduct remains substantially intact in the merged version. As a supporter of the CoC, I feel it will have a greater chance of acceptance and authority if included in the Arbitration policy. <font face="New York"> Skomorokh,  barbarian  02:32, 6 October 2009 (UTC)
 * I think integration is a better approach. We have too many policy pages and keeping them in synch is a herculean task (perhaps even sisyphean). Right now just trying to follow the conversation here and on Wikipedia talk:Arbitration Committee code of conduct is quite the job. ++Lar: t/c 02:41, 6 October 2009 (UTC)


 * Skomorokh, I fear the opposite will be the case, which is why I created a separate proposal. This one has been bogged down for a long time, and it deals with multiple issues. The Code of Conduct is much simpler and I think the community will ratify it if we get the wording right. Anyway, the point of a proposal is to let it be worked on for a period before dismissing it, so please let me do that. If I can't produce one within a reasonable time that makes sense, I'll be happy to agree to whatever is left of it being merged. SlimVirgin  talk| contribs 05:01, 6 October 2009 (UTC)


 * I think it does make sense to integrate the code of conduct here, but, at the same time, much of what the draft code now contains should also be retained here, both in terms of expectations for arbitrators, and protections for them. --Tryptofish (talk) 18:07, 7 October 2009 (UTC)


 * Okay, well if there's a clear consensus that it should be incorporated here (and I'd like to hear from Tony1 on that point before acting), I'll go along with it. But we do need to address the issue of resignation head on, and it's important not to dilute the rest of the content. Also, if it looks as though it's going to take a long time to ratify this page, we should consider proposing the Code again as a separate policy. SlimVirgin  talk| contribs 18:24, 7 October 2009 (UTC)

"In writing"
"Notify in writing" should be generally understood to be any means that is stored and which is not likely to be routinely deleted without some effort on the part of the recipient. Email, fax, notifications to the recipient's talk page, and paper-mail are in writing. Personal conversations, phone calls, and non-logged instant messenger messages are not. Voicemail, logged IRC conversations, and, if logged, twitter and other instant messenger conversations might be depending on the context. Common sense applies. davidwr/ (talk)/(contribs)/(e-mail)  21:25, 7 October 2009 (UTC)

Delete this page?
Should this page be deleted? It seems to have become a ghost town. Tony  (talk)  02:58, 28 May 2010 (UTC)
 * Why should that result in deletion? Happy ‑ melon  09:26, 28 May 2010 (UTC)
 * What, then, is its status? The community deserves to know whether a new policy is going to be proposed or not. Tony   (talk)  10:29, 28 May 2010 (UTC)
 * I would suggest either tagging it as historical or reviving it, but not deleting it. --Tryptofish (talk) 18:20, 28 May 2010 (UTC)
 * Sure; and looking at this I appear narky. Sorry, I didn't mean to be. Tony   (talk)  10:16, 29 May 2010 (UTC)
 * In response to your original concern, we are continuing to work on an updated draft—but, as you can imagine, more pressing matters have a tendency of taking priority over that effort. We hope to have something up in the next few weeks, however. Kirill [talk] [prof] 00:53, 31 May 2010 (UTC)
 * Ah, that's excellent news, Kirill. I'm sure it's a difficult job, and I'm glad to see that it's you who are doing it. Tony   (talk)  04:17, 31 May 2010 (UTC)