Wikipedia talk:Devolution

Okay, none of this is set in stone except the idea that there needs to be a formal transfer/delegation of power, even if limited, from the Committee to members of the community. Could I ask that we not have any polls/votes for at least the first few days? Mackensen (talk) 01:35, 29 June 2008 (UTC)

It's a minor point, but might I suggest renaming this "Delegation of Arbcom powers" or something similar? The title Wikipedia:Devolution actually made me expect an essay on how we are devolving into madness, or some such. But maybe that's just me. Dragons flight (talk) 02:08, 29 June 2008 (UTC)

Who says we aren't? On a more serious note, I'm trying to make a broader point about the need for decentralization, while at the same time emphasizing that power is devolving from a central source. Mackensen (talk) 02:13, 29 June 2008 (UTC)

Which is in itself an interesting point. Though the God-King may be absolute, much of policy and practice on Wikipedia was developed from the bottom-up by the community, and some of what is being discussed here obviously overlaps. For example, administrative actions have always been reviewable by the admin community at large, without need to invoke delegation. In that sense, I'm not sure emphasizing top-down administration even makes sense. However, relying on ANI to solve our problems is obviously dysfunctional. That said, I think it might be more desirable to craft an intermediate structure that extends the traditional review capacity of the admin community at large but adds a definitive decision making process. For example, an administrative review board could be constructed as an adhoc committee of administrators provided that we could agree that reviews had limited time periods, formulated results in a straight foward way, and that decisions explicitly endorsed by a majority of reviewers cannot be overturned by individual admins. Dragons flight (talk) 02:47, 29 June 2008 (UTC)

The system breaks down when consensus isn't obvious--or doesn't exist at all. It's those situations I had in mind here. I'm fairly sure the proposal provides for appeal to Arbcom alone. Mackensen (talk) 02:51, 29 June 2008 (UTC)

I agree with Dragon's flight that the decisions of the ARB should not be overturned by individual admins; which should be achievable if this proposal is adopted with broad consensus and if the ARB's actions are transparent. I'm also thinking that not a lot of cases will come here. ANI frequently sees complaints from this or that editor where two or three people look at it and say, good block or whatever, and that;s the end of it. The ARB would handle cases where the solution is not obvious, the action controversial, and discussion at ANI likely to go nowhere. Thatcher 03:00, 29 June 2008 (UTC)

A more formal way to review indef blocks/bans than AN/I (but less formal than an appeal to ArbCom) seems like it would be a good thing to have, given current scaling issues. AN/I reviews consume incredible amounts of effort per block reviewed, and produce fair amounts of drama (c.f. when is a block a ban, for example), as well as challenges about what exactly got decided... This seems to have merit. It would be good to see it trialled. ++Lar: t/c 04:42, 29 June 2008 (UTC)

I confess I'm a bit concerned by the line "All appeals of indefinite blocks or bans must be made to the Arbitration Committee." That may be a bit heftier than what many have in mind? – Luna Santin (talk) 04:56, 29 June 2008 (UTC)

It is currently policy that users who are indef banned must appeal to Arbcom. Admins may post bans on ANI etc., but the banned user usually has no voice there. Or if a user who was indef banned 6 months ago wants to come back. And they get so many requests that they have admitted that they do a poor job of addressing them. Thatcher 12:29, 29 June 2008 (UTC)

Only if the ban was imposed by arbcom. Viridae^Talk 12:32, 29 June 2008 (UTC)

No, all indef bans. They are sometimes discussed by the community (cf Moulton, Ian Tresman, etc) but ultimately if the banned user's appeal to the community is rejected they have the right to a final appeal to Arbcom. Thatcher 13:02, 29 June 2008 (UTC)

Comments from Risker

I think it would be worthwhile to add "other sanctions imposed by an administrator, absent reference to a specific Arbitration case" to the Administrative review section. I've been seeing more and more of this, and there is not an effective method of review for such unilateral actions at this point.

It seems a little odd to call the group reviewing bans the Block review board, and then allocate responsibility for reviewing blocks to the Administrative review board. I might be more inclined to have the BRB handle blocks only (including indefinite blocks), and the ARB handle bans, page protections, and other sanctions.

I think this has real potential; I like what I see, and will think on it further. Risker (talk) 02:02, 29 June 2008 (UTC)

• Well, the issues in reviewing a controversial admin's action are different than in considering the appeal of a banned/indef blocked user. I think it makes sense to keep them separate, although the naming could be tweaked of course. Thatcher 02:16, 29 June 2008 (UTC)

• Nice tweaking, Thatcher, the update is much more clear in the delineation of responsibilities for the two boards. A noticeboard seems appropriate for the ARB, although it would be difficult for blocked users to post there so an alternate contact method might be required. Perhaps a line needs to be added to clarify that this will not supplant the current process of block review in response to the posting of an {{unblock}} template.

Consider the use of the word "panel" instead of "board", which would turn the acronyms into BRP and ARP - to prevent confusion with 1/ Arbcom and 2/(foundation) board of directors. (The same suggestion applies to the other "boards" being proposed recently by Arbcom.) Risker (talk) 03:19, 29 June 2008 (UTC)

I like boards, but I agree it is potentially confusing. Thatcher 04:21, 29 June 2008 (UTC)

What I suppose we could call "jurisdiction" is at least partially resolved by the manner in which cases are referred to the proposed groups. It looks confusing in theory, but I believe the practice will makes things far more clear. At least as it's currently proposed: with the IRP, cases are referred by arbcom for a recommendation before the committee itself decides; with the ARP, review is requested by the community for a decision. One group is directly accessible by the community, where the other is not. In the event of a conflict, ARP decisions can be appealed to arbcom, which can in turn ask the IRP for advice. Doesn't sound too bad. – Luna Santin (talk) 04:54, 29 June 2008 (UTC)

New watchmen

Who watches the watchmen? That's what I want to know. I don't really have an opinion one way or the other, but doesn't having an ARB or IRB that can keep the Arbitration committee in check just move the problem further down the line? What's to stop a new body from abusing its power? Who keeps the ARB or IRB in check? To be honest, I'm not really bothered either way what happens here, I just thought I'd mention that philosophical point. Deamon138 (talk) 03:58, 29 June 2008 (UTC)

These panels are not "to keep the Arbitration committee in check". They would be created so that the Arbcom could delegate some of its workload. We say, "Arbcom is the last appeal for banned users" which is fine when there are 5 banned users, but when there are 500 banned users all clamoring for attention, the concept does not scale. Arbcom would retain final jurisdiction over both panels can overrule them or dismiss their members. The goal of the IRB is to get timely attention to the appeals of banned users which frequently get lost in the mix. The goal of the ARB is to provide a calm forum to review controversial admin decisions since controversial cases never really get solved at ANI when its a bunch of entrenched interests yelling at each other. And the operation of both panels is to be as transparent as possible. Thatcher 04:08, 29 June 2008 (UTC)

This proposal is designed, I would say, to help arbcom scale; accountability is neither here nor there, with regards to that. – Luna Santin (talk) 04:39, 29 June 2008 (UTC)

Correct. Arbcom is still held accountable via annual elections; this isn't designed to create a new accountability process, although having additional sets of eyes should be helpful. That's the same reason for non-consecutive six-month terms--keep people rotating on and off on a regular basis. Mackensen (talk) 12:47, 29 June 2008 (UTC)

Okay guys, sorry for the misunderstanding! Deamon138 (talk) 14:43, 29 June 2008 (UTC)

Sensitive appeals

Appeals that involve information derived from Checkuser and OTRS will only be handled by panel members who are eligible under the Access to nonpublic data policy.

Is this a good idea? Any such appeal would inevitably be in camera, and if it all possible I think panels would work best when things are carried out in public. We'd wind up having two classes of panelists. Granted, it would be up to Arbcom whether to hand down the appeal for review in the first place. Mackensen (talk) 23:43, 29 June 2008 (UTC)

• I think we have to respect Foundation policy with respect to OTRS and Checkuser. Before you can access non-public info, you have to identify yourself to the Foundation and be over 18. If an appeal involves checkuser data, the data itself could only be shared with other checkusers, unless arbcom makes an exception for the reviewers, but even then they would have to comply with the Foundation policy. If the appeal involves OTRS mail, it could not be shared with reviewers who were not on OTRS, unless they identified themselves under policy or unless the sender waived their right to privacy for purposes of the review. The alternative to having a subcommittee with higher access is to say that reviews involving non-public data are non-delegable unless the appellant waives their rights under the privacy policy. (It may be possible for checkusers or OTRS people to summarize the information for the panel, but if the panelists want to see the actual data for themselves, they either need to be granted access or the editors in question need to waive their expectation of privacy.) Thatcher 03:29, 30 June 2008 (UTC)
  • Note that OTRS members do not have to be eligible under the access to nonpublic data policy; however, I do believe that they should be either eligible under that policy or members of OTRS. Ral315 (talk) 06:38, 2 July 2008 (UTC)

some really good ideas

thanks to all involved with this thus far - I think this page has some really good ideas, and hopefully might represent the silver lining to clouds gathering elsewhere. Whilst the nitty-gritty is important, and will no doubt be tuned and tweaked, for me the central value lies in creating a body of x people (I'd say 50 or so actually - a few more than currently proposed), who form panels of y (I'd say 5 - at the top end of the current proposal) to consider issues. To be honest, I'd have the remit of such panels as quite wide ranging, with perhaps an intentionally fuzzy mission statement ("the role of the panel is to make binding decisions aimed at sorting things out"?).

For me it's also absolutely vital that whatever could possibly be discussed 'on-wiki' is. One of the greatest strengths of such an approach is that it will enable us to analyse competencies, and allow the evolution of 'best practice' (which is in my view a fundamental flaw of the current arbcom - it simply doesn't allow enough inputs to foster evolution). We'll get to see who's good on panels, in what roles, and what approaches actually solve more difficult problems - it's the wiki way!

I'll keep an eye on how this goes in the next few days / weeks, but will hold off editing as furiously as I'm tempted too while the dust all around settles a little... cheers, Privatemusings (talk) 00:42, 30 June 2008 (UTC) it might even be worth mentioning, Kim style, that if this system works well, there might not be much left for arbcom to do! ;-)

Strong support for this proposal

This proposal is inline with my own personal views of how ArbCom should be reformed. I applaud it! Bstone (talk) 05:26, 30 June 2008 (UTC)

Nice theory; no motivation

To comment personally on the admin review panel.

Short version: Please leave me alone, I'm doing fine as an admin and learning from my mistakes as I go.

Long version: This is a meddlesome proposal. It creates a class of uber-admins possessed of an authority to countermand any unter-admins' actions, and those unter-admins may not likewise countermand. It seems to fail to understand the distributed nature of adminning at a pretty basic level. Very rarely do administrative actions that are bad get allowed to stand because there are a large number of admins (and editors) watching. The watchers watch the watchers, if you like. Protections can be 'reviewed' at RfPP or talk pages (and they are; why is there a proposal to overrule admins on an action that is rarely even controversial?). Deletions are handled by WP:DRV which, for all the upset expressed there gets things right most of the time. Blocks are handled by AN(/I) and, where laziness occurs an arbitration case has often followed. If this decentralisation were to proceed, then the block-review board (the BLORB?) would take up these rare cases. Basically, this admin review panel would be invited to execute, and be unable to resist, inquisitions into every-day adminning which is already adequately controlled by sets of admins and the larger corps at the noticeboards. It invites meddlesome reversals of processes which aren't even often in dispute, it demotes the vast majority of admins to those that must effectively seek permission from their superiors and it creates an 'other parent' which will find it impossible to resist dispensing its authority on a regular basis; it will humiliate many admins as it does so and discredit the collective in the process.

Finally, I would observe that, in fact, the proposal is misnamed and misplaced. The items listed are not presently centralised to the committee, save for the trickiest indef-block reviews. They are dealt with distributedly by the admins, editors and peer-pressure. This proposal in fact centralises the process (the lot of them) to a body that doesn't even exist at present! Splash - tk 09:47, 30 June 2008 (UTC)

• Naturally if you think the present system functions fine in all instances there's nothing for you in this proposal. I would however disagree on your second point--the proposal is neither misnamed nor misplaced. The primary motivator here is the delegation of block review by arbcom (a power it possess). There are two proposals and they're not intertwined. The second proposal does not fundamentally alter the power structure; the community retains full control. I doubt editorial actions would fall under any panel's purview; adequate (existing, anyway) processes exist there.
  • Yes, one proposal is designed to interfere with bans; the other is designed to interfere with everything else! Splash - tk 13:03, 30 June 2008 (UTC)
• I would also note that it seems counter-intuitive that a body which, according to you, already handles things just fine would elect a corpus which would proceed to run rampant and screw everything up. I for one don't envision an interventionist panel so much as a safety valve. Mackensen (talk) 10:25, 30 June 2008 (UTC)
  • A safety valve for what, though? The other safety valves? A core/corps of administrators we actually trust, to keep in check those we don't?
  • As an example of interventionism already live is WP:DRV. It is designed to intervene - it has to correct wrong outcomes. DRV largely endorses matters brought to it. It is not, though, at all afraid to intervene in a decision - and (too) often relists the debate (it should just overturn if the decision was wrong, imo). Two things from this: (i) it's not clear to me why there should be a super-DRV able to overrule DRV; and (ii) it intervenes regularly in matters brought to it, even though it mainly endorses. It seems reasonable to suppose that the panel would behave similarly. A body set up specifically to correct admin actions on the entire canvass is a serious devaluation of the strength of all admins' decisions. Splash - tk 13:03, 30 June 2008 (UTC)
• I didn't have DRV in mind when I envisioned the second panel, nor do I think it should usurp it. Broadly speaking, I had in mind blocks and page-specific sanctions; things which are reviewed either at noticeboard or via arbitration. Noticeboards don't scale and arbitration by its very nature either does nothing at all or does too much. I feel obliged to point out at this stage that the proposal enumerates the scope of the panel, and does not mention deletion at all. Mackensen (talk) 15:04, 30 June 2008 (UTC)

Comment in Support

1. In general I like these ideas, and in combination with some of the suggestions at the arbcom rfc, might allow for both less stress, and better focus throughout the dispute resolution, process. --Rocksanddirt (talk) 19:05, 30 June 2008 (UTC)

Jury duty

I wonder if others have read the various suggestions mooted around and about concerning the possibilities of creating a 'jury' system for voting on sanctions. Here's an idea which I think has some merit, and could be worth discussing in this context;

Panels are setup per this proposal, with the roles as described, however they are only authorised to prepare a series of sanctions (editor is topic banned, admin is admonished etc.) - some sort of system (easily technically implementable I'm told...) selects x admins from the entire pool (12 good men and true?) and they are required, by policy, to respond within 7 days with questions, or votes on the sanctions suggested.,

There are many straight forward ways of dealing with absent admins (just select another, basically?!) and I think this concept of separation - tasking trusted, experienced editors to 'make a case' (ask questions, sift evidence, suggest ways forward) and independent admins to make the binding decisions could be a good one. Hopefully it encourages collaboration, sharing of best practice, community investment in smooth relationships etc. etc. thoughts most welcome... if there's some appetite for a bit of a write up (maybe on a sub page?), I'm happy to give it a go.... cheers, Privatemusings (talk) 23:36, 30 June 2008 (UTC)

It is sounds mostly like a pretty good idea to me. One small critique: you shouldn't force people to do the jury duty and make them respond as policy. What would you do if they didn't comply? Ban them? The whole point of Wikipedia is that it is free to use, and as long as everything is NPOV, that should include editing too. Deamon138 (talk) 00:08, 1 July 2008 (UTC)

yeah - I think there'd be a sensible way of facilitating a kind of focused effort that most admins currently active would be prepared to help with - it's a detail which would need sorting out, but isn't central to the concept at this point, I reckon.... cheers, Privatemusings (talk) 00:17, 1 July 2008 (UTC)

A volunteer jury pool might be a good idea. If one isn't available, just go to the next. Bstone (talk) 01:17, 1 July 2008 (UTC)

The jury idea scares me. Reminds me of those mod points on Slashdot. -Jenny 03:31, 1 July 2008 (UTC)

Perhaps they would better be descibed as 'panels,' and have a rolling membership limited to those selected at random and indicating a willingness to conscientiously help out. The number of panelists and length of service needs to be tweaked for 'efficiency' based on experience - between 8 and 15 panelists seems a good strting point. Panels should have access to the archive etc. and help from an experienced knowledgable facilitator. Bug-free coordination and archiving software should be available --luke (talk) 05:59, 4 July 2008 (UTC)

The two proposed review panels

I like the sound of both the Indefinite Ban Review Panel and the Administrative Action Review Panel. I see them more as substitutes for that awful mess on WP:ANI. I'm not sure what it has to do with arbcom, except that one might hope that it will syphon up some work that might otherwise reach them sooner or later.

One reason I think this will be an improvement is because the community is so large. We do still have some really good people around, but the community is so large that they can be difficult to spot when you're surrounded by, well, people who don't really know much about Wikipedia and don't care either. Having panels would help a lot. --Jenny 03:30, 1 July 2008 (UTC)

I think less than siphoning things up to the arbcom it will allow things to be resolved that are not possible through the an/ani/ircadmin system currently. --Rocksanddirt (talk) 22:15, 1 July 2008 (UTC)

Both review panels are excellent ideas and I applaud them. In addition, I think this proposal might want to join with WP:OmbCom and make them all official. Bstone (talk) 06:44, 2 July 2008 (UTC)