Wikipedia talk:Banning policy/Archive 3

Ban discussions

After seeing ban discussions like Koh's, Kurt's Scetpre's, Steve Crossin's, and Prom3th3n's, I've written up User:MBisanz/RfBan as a proposed way to better handle such discussions in the future. Please comment at the talk page. MBisanz ^talk 01:06, 22 September 2008 (UTC)

This would be very helpful. I am also thinking about how server-level bans should be handled, so I have began developing the Server-level ban policy proposal. Please comment at its Talk page. --Marianian (talk) 19:37, 23 October 2008 (UTC)

Blanking, reverts and talk pages.

What’s the position on the blanking of talk page threads started by banned users? This would seem contrary to all usual talk page practice. Artw (talk) 22:11, 30 December 2008 (UTC)

Instead of blanking, I've seen a hat placed on them, thus collapsing the thread. -- Fyslee (talk) 03:14, 17 February 2009 (UTC)

Related to this is dealing with talk page comments. I have seen and since done it in this manner: Strike through the comment and leave a small note as to why. Here's an example. -- Fyslee (talk) 03:14, 17 February 2009 (UTC)

Ad hominem

If Adolf Hitler says 2 × 3 = 6, the fact that Hitler is evil in no way diminishes the cogency of his observation. To say otherwise is to be guilty of the ad hominem fallacy.

This policy says:

Users are generally expected to refrain from reinstating edits made by banned users in violation of the ban, and such edits may be viewed as meatpuppetry. Users who reinstate such edits take complete responsibility for the content by so doing. It is not possible to revert newly created pages, as there is nothing to revert to. Such pages may be speedily deleted. Any user can put a {{db-g5}}, or its alternative name {{db-banned}}, to mark such a page. If the banned editor is the only contributor to the page or its talk page, speedy deletion is invariably correct.

That means, for example, that if an article says 2 × 3 = 5, and a banned user corrects the error, one must revert to the error, and anyone who corrects it later, even not suspecting that it had been earlier corrected by a banned user, is committing "meatpuppetry". If a banned user contributes a valuable article, readers must be deprived of the information in it. If I create a new article titled William Shakespeare, and it later turns out that a banned user had earlier created an article with that title, which was then deleted because the user was banned, I am guilty of meatpuppetry.

This policy looks like that sort of thing that would be written by a lawyer as opposed to an honest person. Michael Hardy (talk) 04:39, 14 January 2009 (UTC)

Is the use of common sense forbidden? The previous paragraph indicates that it should be used:

• "This does not mean that obviously helpful edits (such as fixing typos or undoing vandalism) must be reverted just because they were made by a banned user, but the presumption in ambiguous cases should be to revert."

-- Fyslee (talk) 01:40, 17 February 2009 (UTC)

Banned editors aren't allowed to edit. If the undoing the edits would introduce errors intothe encyclopedia then they should be retained, but in most other cases the material should be removed. Many editors are banned because of POV pushing, and POVs can be a subtle. So if a banned editor fixes 2x3=5 to 2x3=6, then that's fine. But if they add the correct equation from scratch then it should be deleted. The meatpuppetry issue comes when it's apparent that one user is acting on behalf of a banned user, not simply making similar edits coincidentally. This isn't about making ad hominenm arguments, it's about enforcing legitimate bans on users who violate site policies.   Will Beback  talk  01:59, 17 February 2009 (UTC)

I fully agree. We can use common sense regarding the edit, if it is unarguably an improvement, but no matter the quality of the edit, the block evasion still needs to be dealt with promptly and harshly. -- Fyslee (talk) 02:09, 17 February 2009 (UTC)

Does this policy violate copyright law?

An article is created by a banned user. It is speedily deleted under this policy. Then it's recreated by a non-banned user, verbatim identical to the deleted article. Does that violate GFDL?

If so, we have a problem.

This question was raised at WT:CSD by user:NVO. Michael Hardy (talk) 20:17, 17 January 2009 (UTC)

Yes. That would violate the GFDL. And no, users cannot do that. If a user does do that the content should be marked for deletion as copyvio. JoshuaZ (talk) 00:07, 18 January 2009 (UTC)

Obvious solution - if recreated, simply state in the edit summary or on the talk page that the content belongs to the original user. --Philosopher ^{Let us reason together.} 09:49, 16 February 2009 (UTC)

I can't see any good reason why a user should recreate a properly deleted article verbatim. If they think the topic has merit they should write an article from scratch.   Will Beback  talk  02:03, 17 February 2009 (UTC)

Good point. If the article deserved to be recreated as-is the article should be restored at DRV rather than being re-created. Otherwise it is hard to imagine a situation when this would be a problem. --Philosopher ^{Let us reason together.} 14:39, 22 February 2009 (UTC)

Clarify

The policy says an article by a banned user MAY be deleted speedily, NOT that it MUST be deleted speedily.

This raises two points:

• Someone requests speedy deletion on the grounds that its an article by a banned user with no contributions from others. Should they be told that that's not a good enough reason? After all, it says "MAY", not "MUST", so there should be a reason beyond just that, before it SHOULD be deleted. With the SOPHIA article, no further reason was given. When I declined to delete speedily, the person who requested it asked my why, and repeated that it's by a banned user. It's as if I had the burden to show it should not be deleted, even though he had never given a reason to delete it, except that the rules say I MAY delete it. That the rules say I MAY delete it is not a reason why I SHOULD delete it.
• After the admin declines to delete speedily, is that act of declining enough of a "substantial contribution by others" to the article that other admins may no longer come along and delete without AfD? I'm inclined to say it ought to be so, and I think the rule should clarify that point.

So the rules said that I MAY delete speedily, and the person who pointed that out gave no reason to delete except that—no reason why it SHOULD be deleted. That's enough reason to decline to delete if it's a good article that should exist. Michael Hardy (talk) 01:16, 18 January 2009 (UTC)

The presumption is that we enforce the banning policy. However no admin has to enforce anything. They can simply move on and let some one else do it. But a failure of one admin to enforce it shouldn't prevent another admin from doing so.   Will Beback  talk  02:05, 17 February 2009 (UTC)

The idea is that in order to discourage banned editors from violating their ban, articles solely by them are deleted. Of course as with other speedy deletions, anyone may object to the deletion and remove the speedy tag/contact an administrator to restore the article - at which point it is no longer eligible for speedy deletion. --Philosopher ^{Let us reason together.}

14:33, 22 February 2009 (UTC)

Please!

Would somebody crack down on the people who keep adding small, random, tidbits of information in articles. I have no idea who it is, but its driving me crazy! Bionicle Fan (talk) 22:30, 20 May 2009 (UTC)

Restart and extension of ban duration when evasion is attempted

The following section needs some tweaking:

Restart and extension of ban duration when evasion is attempted

It is customary for the "ban timer" to be reset or extended if a banned user attempts to edit in spite of the ban. No formal consideration is typically necessary. For example, if someone is banned for ten days, but on the sixth day attempts to evade the ban, then the ban timer may be reset from four more days remaining to ten days remaining. So if the user doesn't subsequently evade the ban again, their eventual total duration will have been sixteen days.

We need to step up the consequences more. This is, after all, ban evasion, a pretty serious matter. This amounts to telling someone who stole something that they must return it. They must not only return it, but their existing penalty should be increased exponentially. In this case, it should be restarted and then increased a whole notch, not just restarted.

For example, if the original ban was for two months, then a ban evasion attempt should result in restarting, and then bumping it up to four months (if that was the next notch).

Do we have a list of commonly used penalties with increasing increments? We need to look at it first before doing more with this suggestion. -- Fyslee (talk) 02:06, 17 February 2009 (UTC)

When we block IP users for 3RR violations or petty vandalism the durations typically start at 24 hours, then go to 48 hrs, 72 hrs, a week, a month, 3 months, 6 months, etc. So it isn't a fixed increment or multiple, and mostly comes down to what is offered on the drop down menu. In my experience, after a few cases of block evasion it becomes an indefinite ban. Perhaps that would be a better clarification: "Repeated block evasions may result in an indefinite ban."   Will Beback  talk  02:21, 17 February 2009 (UTC)

That is definitely an improvement. I think the three strikes law is a good precedent to use for these types of cases. I have a subpage which I was playing with that touches on this matter, but deals with vandalism templates:

• User:Fyslee/Vandalism templates

If you're interested in the subject, I'd appreciate comments on its talk page. -- Fyslee (talk) 02:34, 17 February 2009 (UTC)

Strong Code of Ethics is Needed

Community ban is a powerful tool to eliminate bad editors. However given the current process it can be easily abused. First there should be a theme for the community which needs to be endorsed by the members who want to join. Ban should not be initiated to resolve a content dispute. Strong code of ethics are needed. Suggested list for all members of community are:

1. have a good prior knowledge about the topic involved
2. review the past related edits of the editor under target for ban
3. state what motivated them to participate as a member of banning team
4. disclose any dispute they had in the past with the editor targeted for ban
5. not join in ad-hoc just to vote
6. should not participate in the final judgment including Admins. The final decision to ban should be by an independent observing team (say 3) out which at least one is Admin and other is a Sr respected editor related to the topic
7. provide specific reason for voting to ban. Abstract and vague accusations such as annoying, harassment, disruptive Tendentious editing etc. without specific acceptable objective evidences need to be ignored.
8. never join under solicitation
9. should disclose any of their close acquaintance with the editor proposing the ban and immediately excuse if there is a conflict of interest.
10. respond to rebuttal by the editor under target for ban
11. excuse if conflict-of-interest is pointed out particularly for Admins

If there is a violation of code of ethics the ban should be reversed. 75.55.49.63 (talk) 07:55, 22 February 2009 (UTC)

so who were you wgen you were banned? Viridae^Talk 09:10, 22 February 2009 (UTC)

You appear to be a Sr editor. Please assume good faith.76.192.202.111 (talk) 15:07, 11 March 2009 (UTC)

I was ( still I am) a wikipedia reader when editors got banned using unethical methods. Please ask further question fast before someone blocks me. No banned editors are using this system now. 75.55.49.63 (talk) 09:25, 22 February 2009 (UTC)

If you would like to appeal your ban you can write to the Arbitration Committee. If they fail to respond in a timely manner you can contact me. In the meantime, recommend you respect your ban and cease evading it. It makes things simpler to sort out if you respect the process. Even if the process may be flawed, there are legitimate options for resolving things. Durova]]^Charge! 19:59, 3 March 2009 (UTC)

I would like to see examples of "unfair" community bans before we add a bunch of rules. Chillum 00:20, 4 March 2009 (UTC)

Hello User:Durova and User:Chillum. Thanks for your attention about above 'Code of Ethics'. If I am not blocked again I will try to provide you the examples for "unfair" community ban. Meanwhile if you can spare time please visit here. By Not Naadapriya.76.192.202.111 (talk) 07:47, 11 March 2009 (UTC)

Using own talk page vs. sending mails

From the discussion here it is claimed that banned users are not allowed to edit anything (including their own talk page), but yet are allowed to send emails from their banned account. The policy clearly says that banned useres can't edit their own talk page, but it says nothing about sending emails. So, are banned users allowed to send mails? I don't really see much of a difference between sending mails and using your own talk page, especially when it comes to ban apeals, so I would consider it to be pretty odd if one would be allowed while the other would be disallowed. --Conti|✉ 17:11, 29 May 2009 (UTC)

I was analyzing the edits of some banned users, and I found that banned user MyWikiBiz/Thekohser has posted some comments on his talkpages (see the talkpages of MyWikiBiz and Thekohser) this month. Since banned users are not allowed to edit their talkpages, I disabled him to edit his talkpages. The banning policy clearly says that banned useres can't edit their own talkpages, but it says nothing about whether banned users are allowed to send emails or not. Wikipedia:Banning policy#Appeals process states "Generally speaking, the banned user will make the request on his or her talk page, which will be copied to Wikipedia:Requests for arbitration by a clerk." A banned user making request on his or her talk page??? But below, it says "An indefinitely site-banned user loses the right to edit any page of the project, including their talkpage." See Wikipedia:Banning policy#Difference between bans and blocks. I think this page on the banning policy should clearly state the difference between ban and block, and there shouldn't be any self-contradictory statement. It should also make clear whether a banned user should be allowed to send emails or not. AdjustShift (talk) 17:39, 29 May 2009 (UTC)

A banned user can be temporarily granted use of their talk page to file a request (if that hasn't been done by email) and for the time that an appeal or review is running. After that, if the ban continues, their talk page is again off limits to them. Note that this doesn't imply that talk pages should necessarily be protected or blanked. Banned users are normally allowed to send email, but I can imagine that there are exceptions. Guido den Broeder (talk, visit) 18:13, 29 May 2009 (UTC)

It seems reasonable to say that arbcom-banned users should email arbcom for appeal, but I don't see why it's necessary to prevent an ordinary banned user from using the talk page. In particular, it looks bad to change the block conditions on the account mentioned above. The user in question was responding to an AFD notice initiated by someone else; that doesn't seem like talk page abuse. Did the admin who changed the block conditions consult with the prior blocking admin? Gimmetrow 06:21, 30 May 2009 (UTC)

That'll be the day. Guido den Broeder (talk, visit) 08:34, 30 May 2009 (UTC)

Gimmetrow, how do you distinguish between arbcom-banned users and ordinary banned users? We don't make such distinctions. A banned user is a banned user. The banning policy clearly states that banned users can't edit any page of the encyclopedia, including their talk page. This is one of the fundamental differences between a banned user and an indefinitely block user. Even if a banned user doesn't abuse his/her talk page, he/she can't edit it. On 28 May, MyWikiBiz/Thekohser posted this and this comments on the talk pages of his two accounts. If he wants to become a free man on the English-language Wikipedia, how is posting such needless comments going to help his cause? He should apologize to the WP community for his past blunders, and promise to never repeat such activities again. If MyWikiBiz/Thekohser wants to edit the English-language Wikipedia again, he should appeal to the Arbitration Committee. Posting needless comments on the talk pages is a complete waste of time. If he wants to post an unblock request on his talk page, he can send an email to any admin (including me), and ask the admin to let him edit his talk page, so that he can file an unblock request. I didn't consult with the prior blocking admins; I did what the banning policy says. I think the page on the banning policy should state that banned users are allowed to send emails to appeal against ban only. If they send abusive emails, their ability to send emails will also be blocked. I think WP editors should work on the Wikipedia:Banning policy page, make clear what a banned user can do and can't do. Some work is needed on the page. AdjustShift (talk) 09:53, 30 May 2009 (UTC)

An arbcom-banned user is a user banned by an arbcom remedy. That doesn't seem complex. Gimmetrow 18:30, 30 May 2009 (UTC)

Apparently. An admin Someone just claimed that topic bans mean you can't post on user talk pages where by chance the topic happens to be mentioned, as that makes the user talk page a related page. Guido den Broeder (talk, visit) 10:19, 30 May 2009 (UTC)

That admin would be correct. If you are topic banned, you cannot circumvent the ban by discussing the topic elsewhere. But that's not what we are talking about here. This isn't about you, actually. ++Lar: t/c 14:51, 30 May 2009 (UTC)

Please don't try to confuse the issue. If it's not about me, then don't talk about me; I didn't.

The page mentions topic bans, but actually says nothing at all about them, so any interpretation is strictly your own. Guido den Broeder (talk, visit) 15:36, 30 May 2009 (UTC)

I did not see, in the paragraphs above, where " An admin just claimed that topic bans mean you can't post on user talk pages", but I do know that has been pointed out to you recently. Since I didn't see it above, I (perhaps incorrectly, can you give a cite for your assertion) assumed you were talking about your own situation, as you often do. Hope that clears things up for you. ++Lar: t/c 16:01, 30 May 2009 (UTC)

As I am mentioned here (I am the "admin" above), I made no such claim and Guido should stop repeating this slur. Guido's topic ban is clear, he is banned from the topic on all pages; that is all I have said, and is rather obvious. Verbal chat 21:10, 30 May 2009 (UTC)

I'm with Gimmetrow on this one. Actually, there shouldn't be a general rule on whether banned users are allowed or not allowed to use their talk page. The situation of one banned user can be quite different from the situation of another banned user, so let's deal with this on a case-by-case basis. If there's a need to protect the talk page, protect the talk page. If there's a need to prevent a banned user from sending emails, prevent him from sending emails. But don't protect talk pages if your only reason is "because the rules say so". Especially if the rules are kinda self-contradictory on this issue. --Conti|✉ 10:24, 30 May 2009 (UTC)

Agreed. Further, absent a clear and compelling reason to keep the particular restriction in place, I plan to lift it in the case that started this discussion. ++Lar: t/c 14:51, 30 May 2009 (UTC)

The policy is not clear on this. For all banned users, there is a general rule right now. As an admin, I've do what the policy says. I would also like to do this on a case-by-case basis. Some banned users are hopeless; some are not that hopeless. The policy on banned users must not be confusing and self-contradictory. We need to work on the banning policy page. AdjustShift (talk) 15:32, 30 May 2009 (UTC)

Policy should guide us as to what went before, not bind us from doing what common sense suggests is best. If the policy is unclear or doesn't conform to actual practice (remember that policy at WP is descriptive not prescriptive) it should be changed to reflect that practice. This is an ongoing task, as policy at WP evolves. In this particular case, I think it's clear that common sense says there is no harm, and some benefit, from letting Greg talk, at one particular page, as long as it is not disruptive. (critical of WP is not necessarily disruptive). AdjustShift and I have conversed on his and my talk pages and I think Adj. is going to change the block settings him/herself, which I applaud. ++Lar: t/c 16:01, 30 May 2009 (UTC)

I've changed the block setting. Greg can now edit the talk page of Thekohser. I would like to thank Lar for his valuable input. AdjustShift (talk) 16:08, 30 May 2009 (UTC)

Well done, AdjustShift. @Lar: if the policy doesn't conform to actual practice, I believe that it is prudent to discuss both policy and practice, with various possible outcomes. Often enough, they both need changing. Guido den Broeder (talk, visit) 16:13, 30 May 2009 (UTC)

I agree. What makes you think I didn't? ++Lar: t/c 16:22, 30 May 2009 (UTC)

Proposal

Given the above discussion, and existing confusion in the policy page, I propose changing the text at WP:BAN#User_pages from:

Unlike editors who have been temporarily blocked, indefinitely site-banned users are not permitted to edit their user and user talk pages.

to:

Indefinitely site-banned users may be restricted from editing their user talk page if they are disruptive.

This reflects practice and allows options. Isn't any user, when blocked, prevented from editing anything but their talk page, including their user page? Gimmetrow 18:30, 30 May 2009 (UTC)

How about:

Indefinitely site-banned users may be restricted from editing their user talk page or using e-mail if they are disruptive.

That way we cover our bases. Chillum 19:06, 30 May 2009 (UTC)

Sure. Gimmetrow 19:12, 30 May 2009 (UTC)

Works for me as well. ++Lar: t/c 21:03, 30 May 2009 (UTC)

No objections. Durova^Charge! 21:09, 30 May 2009 (UTC)

Agreed. Guido den Broeder (talk, visit) 21:48, 30 May 2009 (UTC)

Yep! Works for me. AdjustShift (talk) 03:51, 31 May 2009 (UTC)

I see the change has already been made, but just for the record, I support it, too. Note, tho, that Wikipedia:Banning policy#Site bans still says "An indefinitely site-banned user loses the right to edit any page of the project, including their talkpage". That sentence should be adjusted (or removed entirely), too. --Conti|✉ 20:03, 31 May 2009 (UTC)

Also changed. Gimmetrow 23:11, 31 May 2009 (UTC)

Thanks for the adjustment. AdjustShift (talk) 17:43, 2 June 2009 (UTC)

Consider evasion to be aggravating factor for a banned user

My two cents:

Ban evasion is ban evasion, plain and simple.

Productive or not, an edit by a banned user is at its core willful defiance of a ban that was presumably issued in good faith, even more so if the arbcom issued the ban. We already have procedures in place to review bans that are inappropriate, so evasion is not proper appeal technique.

Self reversion by a banned editor is little more than spam. At its best, it clutters up the page history with noise, and requires THREE edits for the fixes to be made official. And even though a self-revert preserves the page, it's still log noise. Log noise that only wastes the time of anyone on the lookout for ex-editors trying to cause trouble.

Edits by banned users are rightly treated with suspicion, as the ban itself serves as strong evidence that the user in question has a habit of editing in bad faith. Which means that any such edit will either be reverted as a matter of course, or will waste the time of whoever checks it out. Time that could be better spent elsewhere, especially in cases where the edit in question could just as easily have been made by an editor in good standing, and thus without a time-consuming cloud of uncertainty.

The bottom line is that ban evading is trespassing. Personally, I would consider ban evasion no different from a hacker breaking into the system and editing the database directly.

So, my opinion is thus:

Ban evasion should be treated as a serious offense, at least equally serious to whatever offense the user in question was banned for in the first place. Severe punishment for ban evasion would IMHO be in line with the fact that it is pretty much hacking.

Shentino (talk) 16:42, 10 June 2009 (UTC)

Interesting that this is raised here, just before an actual incident arose yesterday, with me. Proposing self-reversion, I wasn't at all thinking that I might actually apply this. The community has, in a very visible case, as mentioned above, rejected the idea that harmless edits are "violations," other than technically, and this is an example of WP:IAR in action. I argued in that case that there was a problem with such edits, but only because they can complicate ban enforcement. That's why I suggested self-reversion; it's true that it takes three edits to accomplish what might be done with one, at the article, but if one looks at the overall process, it's far more efficient than the commonly suggested approach of suggesting an edit on the talk page (1 edit), or, if the editor is banned from Talk, on a user's talk page (also 1 edit and unreliable), with a response being necessary or else more than one editor will waste time looking at it (2nd edit), and the actual article edit (3rd edit). And these edits are far more complex, hence more impact on the database, not less. What might be a one-character change, as was my edit yesterday, becomes many, with much more editor time invested in it.

A self-reverted edit that acknowledges the ban is a cooperative edit, not a defiant one, unless the content of the edit is defiant. Self-reversion sets up a condition where a banned editor, while banned, can effectively solicit and obtain cooperation wfrom unbanned editors, even those with whom the editor was in conflict, thus helping integrate the user back into the community, and I've seen it work that way in one case.

I have argued, as well, that bots can be set up to revert changes made by a banned editor, this would be simple, and would avoid much fuss and disruption. Any editor, similarly, could revert these changes back in, by taking responsibility for them, and whether or not the banned editor should be blocked for the edit would be a separate question that can be considered at leisure. Abuse of self-reversion would be obvious, actually, and would be disruptive behavior, an offense by itself.

The disruption that happened yesterday took place because this matter hasn't become clear; prior discussion seemed to show that self-reverted edits would not violate a ban, inasmuch as even unreverted harmless edits weren't considered ban violations. --Abd (talk) 14:49, 16 June 2009 (UTC)

Sock puppet

Hi, I am a sock puppet. My original account was closed sometime ago. How can I come clean and start over. Please respond on my talk page. Thank you. Green Squares (talk) 00:05, 20 June 2009 (UTC)

Self-reverted edits by banned editors.

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section.

During a recent flap over a banned editor who was pushing the edges of the ban by making spelling corrections to articles covered, it occurred to me that there was a way that a banned editor could efficiently suggest such changes without violating the ban (more than technically, and it would be an error to sanction mere technical violation that respects the right of the community to ban and which does not defy it). That would be that the editor makes the minor, easy-to-verify or obvious correction, and then self-reverts it. A banned editor did do this just today, and because I saw the edit, I was able to quickly incorporate it into the article; it was slightly more complicated than a spelling correction, but the sources were right there and I checked, and the edit was good. With a spelling correction it would be two clicks and done.

I'd suggested that with the first edit, the editor note in the summary, "will self-revert per ban." Then the editor promptly reverts. In the original case where I got this idea, the editor was free to edit talk pages, but it seems silly to suggest a minor spelling correction on a talk page, there is nothing to discuss, it just fills up the page with talk not needing discussion, and then it takes another editor much more time to fix, compared to a quick revert.

The problem with simply ignoring minor corrections is that it complicates ban enforcement, and arguments can easily arise over what's minor. But if an edit is self-reverted, even if there is a problem with it, the problem is immediately gone, moot.

Meat puppetry is really a red herring: if an edit is a problem, any editor who asserts that edit is responsible, it doesn't matter who suggested it. I've encountered this meat puppetry charge before, it never stood up to community examination. Edits by banned editors may be reverted, but there is no requirement that they be, and thus no objection to restoring a self-reverted edit, provided the editor restoring it takes responsibility for it.

I ran this idea past Carcharoth and he approved it. So I suggested it to the original editor, who quite contemptuously rejected it, as did some of his supporters. ("Insulting!") It actually made for a nice little test of the willingness of the editor to cooperate with the community. Self-reversion respects the ban and the community, shows intention to remove the project, and, the kicker, begins to establish cooperation between the banned editor and the other editors working on articles covered by the ban. Hence I think that some mention of this might belong on the banning policy page. --Abd (talk) 03:43, 18 March 2009 (UTC)

Yes, in my opinion, under IAR, banned editors can make productive changes and immediately self-revert. I would not change the wording of their ban or of the policy, although I'm not opposed to a brief mention of this on the policy page. If it's a topic ban someone might still block the editor in good faith after the first edit and before seeing the self-revert, and should definitely not be criticized for doing so. For this reason the banned editor would be well-advised to very clearly state in the first edit summary that a self-revert is about to take place, and to self-revert as quickly as possible; they should not presume that they have an entitlement to carry out this action. If the edit is unproductive, or even if the edit itself would normally be considered productive but if the overall action is considered counter to the interests of the project for any reason, further sanctions can still occur: for example, if the extra reverts muddy up the page history too much, or if the edit summaries are uncivil, or if it takes too much time to determine which edits are or are not productive, or if the editor has done some unproductive edits in this way and been asked specifically to stop completely, or if the edits lead to a lot of wasted time in discussion over whether they're allowed or not, etc. Therefore even after the self-revert, someone might decide to sanction them and according to the letter at least of policy is entitled to do so; as with any action, according to WP:WIARM they should be willing to explain their reasons. ☺Coppertwig (talk) 16:58, 18 March 2009 (UTC)

I've stated, in practice, that a harmless edit by a banned editor should state the intention to "self-revert per topic ban" in the first edit. It could happen that events transpire that prevent the reversion, but isolated unreverted edits by banned editors that are harmless shouldn't be a problem. It only complicates enforcement when they become frequent. Someone seeing the edit and not wanting to check it, will see the edit summary in Recent Changes patrol and may revert it, no problem. It ends up the same. The value to the project of corrections or other improvements made by banned editors in this way exceeds the "damage" of "cluttering up edit history." If that were really a problem, we'd sanction editors who don't use preview and who make lots of edits for a small change. Self-reversion would not be a protection against incivility unless perhaps there were apology in the revert; otherwise self-reversion could be a way to insult someone and avoid consequences. Self-reversion shouldn't be used by a banned editor to assert controversial changes: the existing procedure of suggestion in Talk should be used for that, because discussion is appropriate. There was substantial debate in the recent ScienceApologist case about "harmless edits," and that's how I happened to come across this idea. Editors should not be blocked for making beneficial edits, if they do not complicate ban enforcement. Yes, if an admin blocks a banned editor based on one of these edits, covered by self-reversion, it would be an error, but presumably made in good faith, and the remedy is simple: unblock with apology. It doesn't have to be a big deal. I do think there should be mention of self-reversion in the policy page, that's why I came here. It should be established that non-controversial edits, edits where discussion is actually a waste of time, self-reverted, do not violate a ban. The exact boundary of "non-controversial" isn't important, what's important is that the edit itself would not be a problem if made by an ordinary editor. This should not create arguments over what is minor, helpful, etc., because there is a big step between a noncontroversial edit and a disruptive one. It's only if a self-reverted edit is still, in itself, disruptive, that debate could ensue, and in that situation, any admin could reasonably block for it. Self-reversion could not be used as an excuse for truly disruptive behavior, only for what should actually require no excuse, helpful edits that don't complicate ban enforcement.

So in determining ban violation, an admin would ordinarily ignore self-reverted edits. They would not have to be reviewed routinely. By the time someone is banned, there are usually editors who will be watching for problems with the editor, and if an editor tries to cover up disruption, say by claiming self-revert, but actually leaving problem text in place, that would quickly be discovered. And that would be it for that editor. Blocked for ban violation. It's hard to screw up a simple undo.

But what is truly interesting to me here is that this can open up channels for cooperation between banned editors and the rest of the community, to rebuild trust in both directions. A banned editor using self-reversion is, every time, demonstrating good faith and willingness to cooperate with what the community needs, including respecting a ban. And every time another editor reverts a helpful change back in, cooperation is being shown in the other direction, and the project benefits.

ScienceApologist rejected this idea because his goal wasn't making helpful corrections, it was to get editors arguing with each other, and to discredit the ban; the idea of cooperating with it? Horrible! Some of the editors who had previously supported him were trying to encourage him to accept consensus, but he blew them off; others clearly sympathized with the disruption and they, likewise, thought the proposal to self-revert an "insult."

It is also possible to extend this to totally banned, blocked editors. A self-reverted, helpful edit by an IP address with an edit summary of, say, sp., will revert per ban of BannedEditor, if not disruptive in itself, should not be considered ban violation and, I'd recommend, while it would easily lead to examination of other edits from that IP, should not result in blocking of that IP if there were no disruptive edits from it. Work with us, we will work with you. I don't know how much difference this could make, but it wouldn't hurt. But I'd suggest starting with topic bans, it's less complicated. (The IP may, of course, simply make those helpful corrections without a banned editor notice in the summary, and they will probably not be noticed, but, then .... the community doesn't see that this editor is being helpful, and perhaps we should notice such things, if we want to heal conflict.) --Abd (talk) 00:07, 19 March 2009 (UTC)

Okay, here's a situation. Banned editor makes spelling correction. Undo takes less than a minute, three or four clicks after the edit, but, in that period, someone else edits the involved section, preventing reversion. Fine. Leave it. The banned editor shouldn't have to rummage through the text and restore the spelling error. That will not happen often and would be covered under "occasional unreverted harmless edits." I'll note that SA made several harmless spelling corrections, and the community roundly ignored them. A complaint was raised only by one of his supporters, and was rejected with derision (the ones rejecting it apparently thought that this editor was actually trying to get SA blocked for spelling corrections). It is not routine to block a banned editor for making a spelling correction, provided it doesn't become so extensive that it complicates ban enforcement. With self-reversion, that wouldn't happen. An editor self-reverting as described should be safe from being blocked for ban violation, and it is very little trouble. On the other hand, an admin reviewing for ban violation could decide to look at any unrestored edits, instead of ignoring them, and, if they were good edits, revert them back in, thus converting a negative activity (scrutinizing for violations) into a positive one (cooperating with banned editor to improve the project, efficiently). --Abd (talk) 00:19, 19 March 2009 (UTC)

Abd, please reconsider these words about an editor who is not at liberty to respond. Durova^Charge! 17:53, 19 March 2009 (UTC)

I think these are essentially the same questions which were previously asked by Abd (and satisfactorily answered) a year or so ago in regard to User:Fredrick_day. — CharlotteWebb 18:01, 19 March 2009 (UTC)

While I appreciate Durova's approach, the comments above aren't, in the end, about ScienceApologist, and he will not be harmed by them any more than what already occurred harmed him, they aren't complaints about him or attempts to increase sanctions. Rather, his case is an example of what is certainly possible behavior, and is adduced as such. If his intentions were, or become, to improve the project cooperatively, the above won't add pressure to him; however, I offered, if he would agree to self-revert should he edit an article covered by the ban, I would fully support him and attempt to protect him from claims of technical ban violation, and, in fact, I reverted one of his spelling corrections back in after it had been removed as an alleged ban violation. Even though this process would be hardly any inconvenience at all for him, and certainly less cumbersome than getting ArbComm permission to edit a specific article -- which he did in one case -- he rejected it, and it was considered, by ArbComm, that his declared motivation for what he was doing was indeed disruptive, that's a settled matter, Durova. I use, where appropriate and possible, real examples from the editorial history of Wikipedia, based on a very sound legal principle: judgment in the abstract is to be avoided. The proposal for text in the guideline here -- which hasn't been made specific yet -- won't be about him, though his case is what led me to develop the concept and to ask Carcharoth about it. --Abd (talk) 05:55, 22 March 2009 (UTC)

However, having said that, Durova, what specifically would you have me consider refactoring? --Abd (talk) 05:57, 22 March 2009 (UTC)

A ban is a blanket statement that editor X's contributions to pages Y are disruptive and unwelcome, that the invitation to edit those pages has been withdrawn, and that the editor may be blocked if they don't respect the ban. Put very simply, a ban is a request to go away and edit somewhere else.

In general, it is a bad idea to create loopholes which allow, because they are "about to be reverted", edits which would otherwise be judged disruptive and might merit a block. Internet and wireless networking connections can go down, batteries can fail, users can be kicked off computers by parents, and the Wikimedia servers can experience difficulties - and even if none of those occur, other editors may still end up viewing the problematic revision of the page in question, simply based on when they access it.

Needless to say, I think it'd be counterproductive to encourage banned editors to edit the pages they're banned from. Creating such a loophole introduces complexity and uncertainty to something that should be very simple, and invites acrimonious dispute in an area which is already prone to drama. It also discourages banned editors from finding something else to do, i.e. edit in an area from which they are not banned. Editing experience away from problem areas can make the difference between an editor becoming productive and being blocked or banned indefinitely.

Finally, this just doesn't look like a good solution to the problem. If you want productive edits to be exempt from a ban, then start a discussion and get consensus as to which edits, if any, are noncontroversial and productive even if made by a banned editor. If you can do that, there'll be no need for banned editors to resort to trickery to circumvent their ban. SHEFFIELDSTEEL^TALK 15:31, 16 June 2009 (UTC)

Bans are preventative, the "go away" is issued, assuming the process was legitimate, as an expression of an expectation that, from history, the edits will engender needless disruption. There then arises the question about harmless edits; that is, edits, that by their nature, are not controversial. If somehow we could allow these edits, we would. And, in practice, the community does. If I wanted to make a spelling correction to Cold fusion, even though I'm banned, I could easily do it without consequence, it's very simple: just log out. Spelling corrections to articles don't establish any kind of pattern that could lead to sock investigation. To use the example of my own ban further, I've not been topic banned, just banned from editing the article and its talk pages themselves. I've been actively encouraged to participate in other cold-fusion related processes, in particular, the mediation. Being involved with the mediation means that I'm regularly reviewing the article, so I may see a spelling error. What should I do?

I can tell you. I won't log out, it's too much trouble. I won't suggest a spelling correction on a user talk page, as has been suggested, too much work. I won't do anything, I'll leave it, now that I've been blocked for making a one-character correction to a reference, then self reverting "per ban." If this had been part of a pattern of testing the edges of a ban, sure. I'm not raising this here to complain, but only as a real example where, as a banned editor, I noticed an apparent error, thought I could fix it quickly, and attempted it. I did nothing more than what I'd seen done many times, and accepted, and I was shocked to be blocked. I won't do it again unless this matter has been clarified to permit it.

What we know is that a popular editor who makes spelling corrections to an article while banned, isn't considered to be violating the ban in any way that deserves sanctions, and that's been very, very clear and actively discussed. Only if the edits rise to a pattern of provocation is a sanction likely. In that case, there was indeed a declared intention to disrupt ban enforcement. Someone filing an Arbitration Enforcement action over a single spelling correction is still likely to get slapped down. It's been tried, too many times. So the alternative here, for popular editors, isn't total ban or self-reversion. The alternative is allowing minor corrections, not controversial, vs. allowing minor changes, not disruptive in themselves, with self-reversion. Self-reversion shows cooperation with the ban, not defiance of it. Routinely, self-reverted edits could be ignored, and only receive attention upon complaint that they were, in themselves, disruptive (such as insulting an editor, then self-reverting, especially if repeated). This makes enforcement easier and less disruptive, not more so.

Blocking someone for a harmless edit is never going to be popular. I did not protest, beyond asking the blocking admin to reconsider, after providing evidence. This was taken to AN by a completely unfamiliar editor who saw it and was puzzled, and thus another long discussion at AN ensued over practically nothing.

Self-reversion is often used by editors to propose an edit that might be controversial. It puts the edit in place so that it can be seen in context, and it is often more efficient than proposing the edit, in detail, in Talk; the edit can then be discussed by reference to history, and easily implemented if accepted. Promptly self-reverted edits that violate policies are normally considered moot. I'm just suggesting that we recognize this more extensively, because there are a number of salutary expected consequences, that have already been seen in actual practice. A ban can be extended, where needed, to prohibit self-reverted edits as well as normal edits, should an editor abuse this in some way. --Abd (talk) 17:19, 16 June 2009 (UTC)

• If it really is a good idea, then you should probably just do it. I don't see why we need to add an authorization for this in the policy, especially since it isn't yet common practice. Do it and we'll see how people react. If it becomes accepted, then we can consider amending the policy. ⟳ausa کui ^× 17:57, 16 June 2009 (UTC)

Look at it this way. If you see an edit that improves Wikipedia beyond any shadow of doubt, go ahead and make it. Of course, if you're banned from editing that page, then you might want to mention WP:IAR in your edit summary, in order to avoid trouble. Good luck!

Having done so, if you then go on to make a second edit (1) to an article that you're banned from (2) that harms Wikipedia beyond any shadow of doubt, don't be surprised if you're blocked. SHEFFIELDSTEEL^TALK 18:21, 16 June 2009 (UTC)

I disagree strongly with the view that it's OK for topic banned editors to make edits to articles in the banned topic. Many times people are banned for disruption and POV pushing. The edits they want to make may not be "harmful" to the project in a narrow sense, but the editor's involvement in the topic is disruptive and that is harmful. Folks don't just get a topic ban for no reason. Very often the other editors have had to put up with the disruption for a long time, and then make a case to get it remedied. To make them continue to fight with the user over his POV edits while he's topic banned is illogical, and just perpetuates the disruption.

If a banned editor makes an edit by mistake, perhaps not realizing that an article is included in the ban, for example, then we should assume good faith. But if it's done intentionally then the ban needs to be enforced with a block.   Will Beback  talk  20:10, 16 June 2009 (UTC)

I can agree in that certain editors should be allowed to make self-reverts in articles where they are banned if the community considers that his edits are good and the information very difficult to convey otherwise. Case in point User:PJHaseldine in Pan Am Flight 103 bombing trial making this edit.

However, I can't agree on making a blanket allowance. Banned editors would start making typo corrections and other minor stuff to show that they are being collaborative, which would to clutter of page history. Every little fix takes 3 entries in the history: the edit by the banned editor, the self-revert, and the undo by someone else to implement the change if it's good. Also, banned editors can simply request the change in the talk page, and, if they are banned also from there, they can use the talk page of any interested user, and this way the change gets reviewed by another pair of eyes before being done. So, to sum it up, why should those banned editors make those self-reverts of minor stuff when they are not allowed to edit in the first place, and when there are multiple way to avoid having to violate their ban, and then those other ways have a better result?

As for invoking IAR, sheffield is correct, but depending on circunstances this invokation can have a very chance of drama and disruption, so banned editors should first make sure with the person that banned them that it's ok for them to edit the page in that way. --Enric Naval (talk) 23:53, 16 June 2009 (UTC)

In the example given above, I don't see a reason why PJHaseldine could not have either used the talk page or contacted an editor directly. While it was an important edit to get made (because it concerned a BLP in a criminal case), there were other ways he could have handled it without breaking the ban.   Will Beback  talk  00:19, 17 June 2009 (UTC)

That's true, and a ban in those conditions looks more like a 0rr restriction than a ban. And, seeing how Abd wikilawyers just below about fixing just one reference, and how he is misrepresenting the support of arbitrators to this practice, I have to retract my support to any exception to a ban. Banned editors were banned in the first place for a reason, and they should keep themselves well away from articles for any reason and limit themselves to suggestions in other places. The sheer amount of wikilawyering and drama that banned editors can generate is just not worth the benefit of letting them make a few edits. If they behave well enough to deserve a 0rr restriction, then they should appeal their ban to get it. --Enric Naval (talk) 03:54, 17 June 2009 (UTC)

The point is that by precedent these edits didn't "break the ban." PJH used this technique for a more complicated edit than I'd suggested, but it does show the effect. By actually making the edit, he could show, in context, exactly what he was proposing. It's much more efficient than describing it on a Talk page, discussing it, waiting for someone else to implement it, etc. As to the efficiency, the problem originally came to my attention with ScienceApologist, who was making spelling corrections to page from which he was banned. The community, quite strongly, opined that this was okay. Some of the same editors complained when I did it and added a self-revert, because of the prior discussions about the SA case. I urge thinking first about a very simple case: an editor sees a spelling error, and is reading the article, and there is the edit link, and it can be fixed in a flash. This is an edit that, except for one problem, IAR would strongly suggest, fix it.

The one problem is that the editor is banned from editing, either that article, the entire range of topics, or, I've argued, is entirely banned from editing Wikipedia, and isn't logged in, because if logged in, the editor won't be able to edit. Should the editor make the edit, and if so, how? What would follow Rule Number One and respect community process?

The practical reality is that we don't go after IPs for making harmless edits.

Now, suppose a banned editor doesn't make the edit, instead, goes to the talk page and suggests it. That takes many more words and time than simply to make a spelling correction. Then an editor who sees the suggestion has to figure out where to find it and then also make the correction, and, to avoid many editors looking at the same thing, has to go back to the Talk page and acknowledge making the edit. Compare this to the editor simply making the edit. In the SA case, he was just making the edits, and it was broadly defended, but my point was that it complicates ban enforcement, and this was also generally recognized. I realized that if the editor self reverted with a note, per ban -- and any editor can revert, on sight, an edit by a banned editor -- this would make enforcement simpler than allowing "harmless edits." So I ran it by an arbitrator -- this is documented on my Talk page in the discussion of my recent block -- and the arbitrator approved it. I ran it by the community, and, until now, nobody opposed it.

This is the reality for me. If I see a minor correction to make and I can fix it on the spot, I'll do it. I'm simply not going to go to the trouble to do something much more cumbersome. But self-reverting is easy. And, in fact, it works. The changes get made, reverted back in, quickly, where an article is watched, and we don't usually get banned from editing unwatched articles.

The idea that the extra edits gunk up article history is spurious. Sure, it's more edits than a single correction, but not much. I don't know how much the database is optimized, but it might not be much disk space at all. In any case, it's much less noise than the common suggestion of discussing a spelling correction! And what is a spelling correction worth? I'd say it's worth the extra edits. Self-reversion shows cooperation with the ban. PJH didn't get into any trouble at all for his edit, it only came up because the editor who took my block to AN mentioned the example. I'd say this should be encouraged, and if any editor abuses it, easy for an enforcing administrator to warn the editor, or even block directly for egregious abuse. I really doubt that would happen much. Most banned editors are not at all interested in cooperating, and self-reversion demonstrates cooperation. --Abd (talk) 02:01, 17 June 2009 (UTC)

"(...) and the arbitrator approved it." and in 15 June another arbitrator came to your talk page after you were blocked and he told you that "the "Committe", as a whole, most certainly does not approve of this method to circumvent a topic ban— and any edit to a page from which one is banned may lead to blocks of increasing duration (...)"[1] and also told you as an editor that you should stop trying to strech the limits of your ban and go edit something else (in the diff I just linked), at which point you should have stopped using as an argument that an arbitrator supported your argument, and much less says that this was documented in the discussion of your recent block without mentioning anywhere that a different arbitrator popped up in that discussion to make clear that the committee does not give its support to this practice at all. --Enric Naval (talk) 03:47, 17 June 2009 (UTC)

I really see no need for this. If you are banned, stay away from the page. That keeps the banning simple, and avoids that edits get into a grey area of "is this a simple typo that is repaired, or does this affect the meaning of the sentence" (there are many words which differ one character which can make a difference in the meaning of the sentence, or how it can be percepted - effect/affect e.g.). If it is is a simple typo, then that is not a pressingly urgent matter that has to be repaired now, especially not by a banned editor. If the case is even the slightest little bit worse, then the banned editor should discuss, or the matter should be discussed on the talkpage (if the editor is also banned from the talkpage, but if the matter is really pressing, then I could see that the banned editor would bring it to the attention by a short message on the talkpage, only mentioning the issue, and nothing else, and then waiting to be asked for further response when needed). The only exception I can think of where a banned editor would need to edit a page, would be in very bad WP:BLP/copyright violations (and even then, the banned editor should take very, very much care, and post to talkpages first, and I would suggest to start a post on a noticeboard to attract others attention to it and not do it at all), all the other errors, however bad, are NOT a pressing matter which needs to be edited or discussed by a banned editor.

Therefore: I strongly oppose this idea: if you are banned, do not edit pages you are banned from, simple, and no exceptions to that. If you do, even in good faith, blocks are in place. Show your cooperation somewhere else (Other subjects? Talkpage?), and when that is met with positive response, then go request removal of the ban (if you are banned from the talkpage as well, other subjects is the way to go, and then request to be unbanned on the talkpage, or both).

Arbitration indeed has not strongly condemned this in the past, though Arbitrators are apparently stronger in opposing this now. That still does not stop the community from having a different opinion and updating policies accordingly.

Abd, you are right, you could log out and do the edit. And I really wonder why you did not do that (well, not really, I think I do know, it is the way to a lengthy, or even a permanent and total ban)! --Dirk Beetstra ^{T C} 08:52, 17 June 2009 (UTC)

Goodness me, but it really is very inefficient trying to change articles you're banned from, isn't it? Perhaps the solution is not to get banned in the first place - or, if you do get banned, accept that and direct your attention elsewhere. Who knows, that might be better for you than spending 30 days (or six months or whatever) staring at the article you're not allowed to edit, looking for typos. I think maybe this discussion has gone on long enough. SHEFFIELDSTEEL^TALK 14:42, 17 June 2009 (UTC)

• Here's what I'm gathering from the above discussion. Topic banned editors who edit outside their ban do so at their own peril, even if the edits are minor and unambiguously beneficial. An admin who spots this kind of editing might choose to "look the other way", but it would be her discretion to do so, and we are not in favor of adding anything to the policy that suggests that banned editors are ever authorized or entitled to edit articles they are banned from editing under any circumstances. Mark this resolved? ⟳ausa کui ^× 20:11, 17 June 2009 (UTC)

  We see here a handful of editors expressing opinions, some in one direction, more in another, but many of the arguments don't seem to recognize all the issues, or make assumptions that aren't actually general. For example, I'm page banned, but not topic banned, and I've been encouraged to continue a mediation on article issues, which involves referring frequently to the article and its Talk page, hence the suggestions of "stay away" aren't appropriate for me. They might be for someone else. A specific change to the policy has not been proposed, but many editors and administrators and one arbitrator have, in the recent past, given opinions on the issue that are quite at variance with what most have said here, hence my position is that there is no consensus at this time for any change, nor for the status quo. I'd say it's important to resolve this, or it will come up again and again; my block is not likely to become a precedent, and neither is the SA case, because in both cases other issues were strongly involved. So, my suggestion at this point: do nothing, see if other opinions are expressed, and I see it as unlikely that this particular discussion will find real consensus, short term, consistent with recent practice. --Abd (talk) 21:03, 17 June 2009 (UTC)

  Luckily, we don't have firm rules, so it's not a big deal if the policy doesn't cover every eventuality. If you are sure that others will find it okay for you to edit an article from which you are banned in this manner, then you should do it. It's possible that the banning admin will look the other way, though I stress that you do it at your own peril, and the banning policy likely won't ever authorize it. ⟳ausa کui ^× 02:19, 18 June 2009 (UTC)

  I disagree, I predict it will, but this is the point: I have not been arguing here that I should be permitted to make "edits" to articles under ban. I will not make such edits, of any kind, and I've stated that over and over, because, obviously, it created much disruption, in spite of any intention to violate on my part. I also advised PJHaseldine, who had successfully relied on self-reversion to efficiently accomplish an edit to an article under ban, to abstain. I'm arguing for the project, following reasoning that was that of a strong majority in the SA case, such as [2]. That's the same admin who blocked me for making a small correction, self-reverted. Leaving this unresolved allows admins to follow, without necessity, personal opinions about editors, rather than the content of the edit, to blocking. Self-reversion removes the ban enforcement concern, it certainly doesn't make an edit more of a problem, and this was cleared with an arbitrator, I wasn't just winging it, and that was when I had no anticipation of having an occasion to use this myself. Again and again in this discussion the idea that I "want" to be able to edit articles under ban has been raised. I don't. I want the community to be clear and to consider this with all the evidence in front of it, and not reacting just to one specific case, but to the whole spectrum of cases. --Abd (talk) 14:37, 18 June 2009 (UTC)

  To make things crystal clear: any edit to a page from which you are banned is a violation of the ban and can (and should) lead to a block. It's that simple. A ban means no editing, not "editing which has a net result of X", or "editing that's only Y". It means your name should not appear in the history of the page for the period in question. — Coren ^(talk) 23:04, 2 July 2009 (UTC)

(reduce indent) Personally, I think this proposal would open a huge can of worms by greatly extending the number and type of edits that - it could be argued - are acceptable under WP:IAR because they're about to be reverted. We alredy have IP editors making "test edits" that they then revert, but which are essentially vandalism. The last hing we want is to introduce uncertainty and the potential for more drama and wikilawyering. That reminds me. I think that as far as the consensus of this discussion is concerned, Causa sui has nailed it, and it's time to close the thread. SHEFFIELDSTEEL^TALK 15:29, 18 June 2009 (UTC)

Back in March when this line discussion first arose I asked that it proceed without singling out a particular individual as a context for discussion. Three months later the discussion has resumed, and the same individual is getting used the same way again. Repeating the request that policy discussion proceed without using this person as a template for discussion. He has chosen not to participate in this discussion, has sat on the sidelines as we ask, and is doing his best to make a quiet return while working on a featured article drive. Please draw upon a wider range of examples, if examples must be brought forward. Durova^Charge! 20:27, 18 June 2009 (UTC)

I'm not sure what you are getting at. If there is a particular example you want us to consider, would you introduce it? I can't go digging up examples, especially if there is one you have in mind. ⟳ausa کui ^× 22:06, 18 June 2009 (UTC)

I think all Durova is saying is that we should not be citing one particular example that has been discussed a good deal above. SHEFFIELDSTEEL^TALK 13:11, 19 June 2009 (UTC)

This issue has also been flogged elsewhere, with the consensus there as clear as it is here. Socrates2008 (Talk) 13:00, 7 July 2009 (UTC)

Flogged by Socrates2008, who filed that report. This issue is headed for ArbComm, I'd say, for there have been three separate discussions of harmless edits to articles under ban, with very different conclusions. One was a set of Arbitration Enforcement requests, with the general conclusion that it was, in the words of one admin, stupid to block for harmless edits, and then in the second situation, underlying the AN filing of Socrates2008, the same admin blocked for a harmless edit. Fun, eh? --Abd (talk) 20:22, 7 July 2009 (UTC)

I'm thinking the only thing we can safely say on the policy page is: if you edit a page that you're banned from, you may be blocked from editing. SHEFFIELDSTEEL^TALK 20:34, 7 July 2009 (UTC)

You may be, indeed. However, the prior consensus on this was that it was "stupid" to block for minor, harmless edits that have nothing to do with the reason for the ban. I proposed a change, then, when it had nothing to do with me, that self-reverted edits, unless specifically disruptive in themselves, do not violate any ban, because they leave behind nothing needing corrective action. It's a little weird that the community had an apparent consensus when it was a different editor involved that harmless edits weren't ban violations in themselves, and that suddenly the reverse consensus was expressed when it was me that made the harmless edit, but that's Wikipedia for you. Sure, you can be blocked for editing in violation of a ban. You can be blocked for practically anything, or for nothing at all, if an admin gets it in his or her mind to block you. So? --Abd (talk) 03:51, 8 July 2009 (UTC)

The only reason there's any ambiguity around making edits while blocked is a direct result of your attempts to encourage blocked editors like user:PJHaseldine to test the limits of their bans in the hope that you can thereby introduce new policy. If you succeeded in this quest, I fear this would give you a license to test the limits of people's patience at the Cold Fusion article even further. Your rallying and lobbying to introduce a complicated exception mechanism to the blocking process does not have general acceptance in the community, so I suggest you drop it now. Socrates2008 (Talk) 23:16, 7 July 2009 (UTC)

No, Socrates, you have the history of this quite backwards. The situation when this topic was begun was that a topic-banned editor was making spelling correction edits to pages covered by the ban. There were regular reports to Arbitration Enforcement demanding that the editor be banned for it, and these were being rejected. However, such edits did, in fact, create a problem, complication of ban enforcement, which is why I dreamed up self-reversion as a solution. I suggested it to an arbitrator. Good idea, the arbitrator said. So I suggested it to the banned editor. He rejected it vehemently, and it was considered "insulting" by some. Why should he self-revert a perfectly good edit? And that was the only objection to the plan, not that it was actually a ban violation. If unreverted harmless edits weren't a ban violation, why should a self-reverted edit be a ban violation? I made the proposal here, then, when it had nothing to do with me. When the ban of PJHaseldine came down, to console him, I suggested self-reversion. PJHaseldine may have a conflict of interest, but he is also an expert on the topics of his interest. Self-reversion would be an excellent way for him to suggest certain changes efficiently. And it apparently worked, and it helped him to accept his ban. And it was only when I was blocked for a self-reverted edit that you took the self-reversion of PJH to AN, apparently seeing an opportunity to tighten restrictions on him. Had any harm been done by those edits? Note that, in the end, PJH wasn't blocked for any self-reverted edit. It was for an article, and he hasn't responded to questions about what he was thinking when he did that. This is a page for discussing changes to the policy. Self-reversion will come back, it's too good an idea to be shuffled off simply because some editors have axes to grind and take the opportunity when they see it. --Abd (talk) 04:02, 8 July 2009 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Appeals of community bans

• Bans imposed by the community may be appealed to the Arbitration Committee (arbcom-l@lists.wikimedia.org). Banned users should not create sockpuppets to file an appeal. Rather, they should contact a member of the committee or an Arbitration clerk by email and ask that a request be filed on their behalf. Generally speaking, the banned user will make the request on his or her talk page, which will be copied to Wikipedia:Requests for arbitration by a clerk. In some cases, a banned user may be unblocked for the purpose of filing an appeal. In such cases, editing of unrelated pages is grounds for immediate re-blocking.

This language seems to imply that the appeal will be handled on-Wiki. However appeals are routinely handled off-wiki, with no notification to concerned parties. I suggest that we add a sentence requiring an announcement of off-wiki appeals, so that interested editors can submit evidence or arguments. Otherwise, the banned party is the only one being heard. It needn't be complicated. I suggest somthing like:

• If an appeal is to be heard off-Wiki, the ArbCom shall post a notification on Wikipedia:Arbitration Committee/Noticeboard requesting input.

Thoughts?   Will Beback  talk  06:48, 26 June 2009 (UTC)

In principle, it's a good idea, though in reality the majority of appeals are swiftly rejected (typically, serial sockpuppeters promising to reform the morning after their latest incarnation has been indefinitely blocked) and in these many instances requests for input would just add delay, unnecessary process, and sometimes high drama. However, particularly in the controversial cases, community input is helpful for assistance in tailoring editing/civility restrictions and so forth. Perhaps something like this might work better,

Where an appeal is to be heard off-Wiki, and there appears to be a reasonable prospect of the appeal succceeding, the ArbCom shall post a notification on Wikipedia:Arbitration Committee/Noticeboard requesting input by email prior to making a final determination.

(This comes with the usual arbitrator caveat that this is my personal opinion rather than an expression of ArbCom consensus.)  Roger Davies ^talk 07:13, 26 June 2009 (UTC)

That's a reasonable caveat.   Will Beback  talk  07:24, 26 June 2009 (UTC)

Also my personal opinion: I'm not overly fond of this as a requirement for two reasons: first, because it doesn't take into account a detail of human nature: many people who have been involved in the original ban may (and often will) oppose a later unban because of all the grief and annoyance that led to the original ban rather than because of genuine concerns looking forward; and secondly because discussion about lifting bans usually require the arbs to dig into the past for faults in a manner that's more likely to cause acrimony and rekindle past anger than anything else if it were done more openly.

Certainly, there are cases where consulting with the editors are large would be wise (and, indeed, we have done so on a couple of occasions already), but mandating that this be aired in public defeats the point of ban reviews by a body different from that which imposed the ban in the first place. — Coren ^(talk) 10:13, 26 June 2009 (UTC)

Agreed. I hadn't noticed this could be interpreted as input in public. I've added "by email" to the draft text. For clarity, I am against any input which might lead to pitchforks and torches.  Roger Davies ^talk 10:25, 26 June 2009 (UTC)

I agree with Coren here. This should not be a requirement, and is just begging for previously offended users to remind everyone of their interpretation of the past grievances, thus poisoning the atmosphere for the banned editor should he return - and they are likely to start this process on WT:AC/N. This is exactly the reason that these requests are usually heard in private. Risker (talk) 11:01, 26 June 2009 (UTC)

Roger Davies' point seems like a good one to me. A ban appeal is essentially an altered RfAr with private instead of public hearings, where the final objective by the filing party is not a desysop, ban, or other restriction, but rather that those restrictions be lifted. If it must be held off-wiki to ensure that everything goes smoothly, fine. But to not allow the other party that will be affected by any unbanning – the community – into the discussion seems to me like a contravention of ArbCom's policy on private hearings. NW (Talk) 13:05, 26 June 2009 (UTC)

This assumes that the majority of the community actually cares enough to comment. Fact is, most of the community don't have an opinion one way or the other, and the only people likely to comment are those with strong opinions one way or the other. We will hear cries of dismay and censorship when we enforce the "email us the response" and remove commentary from the Arbitration Committee noticeboard talk page; and that does not prevent people from (for want of a better way of saying it) pushing their point of view on other pages on and off wiki; in a way, it legitimizes personal attacks against the banned editor in the guise of discussing the proposed unban. Often the Committee is aware of disputes between a banned editor and a current editor that have not been conducted on-wiki, but cannot refer to those disputes in managing on-wiki commentary about a proposed unban. Simply put, it gives opponents (and supporters) a platform on-wiki to promote or refute the proposed unban without giving a similar on-wiki platform to the banned editor.

In many cases, this is a perfectly reasonable manner in which to proceed. I simply do not believe it should be mandatory, and that it must be addressed on a case-by-case basis dependent on the specific facts of the situation. Risker (talk) 13:40, 26 June 2009 (UTC)

This was not the case with User:Thekohser. You folks made a decisions without having all the facts in hand. Jehochman ^Talk 01:13, 27 June 2009 (UTC)

How do you know this, Jehochman? I am not noticing my inbox filling up with people telling the Committee how it missed Evidence X, Y and Z in this case. Risker (talk) 01:24, 27 June 2009 (UTC)

Because I have information I'd have shared with you if I'd known you were considering an unban. At this point, why should I email you? Have you been anything other than dismissive of me? Have you said anything like, "Hmm, you might have a valid point. Can we discuss this further?" Jehochman ^Talk 01:49, 27 June 2009 (UTC)

Something to lighten the mood.
Image by User:David Shankbone

• ArbCom are a bunch of clowns for handling appeals without community input. How do they know that somebody in the community does not have relevant information about the banned user? At minimum they should post appeals and hear input before making decisions. Jehochman ^Talk 13:43, 26 June 2009 (UTC)

• Actually, it strikes me that in many cases we'd be advocating a three-ring circus to proceed with such activities. The image you've added more closely resembles the "community" discussions (usually involving only a handful of users with a vested interest in either upholding or rescinding the ban) that have led to such bans and have historically kept them in place. Risker (talk) 16:46, 26 June 2009 (UTC)

Please list the bans. You're making sweeping generalizations that are not upheld by the facts. Jehochman ^Talk 16:49, 26 June 2009 (UTC)

This is not a power grab on the part of the Arbitration Committee, as you put it in your edit summary, Jehochman. The Arbitration Committee has never been obligated to bring such ban reviews before the community before; it seems our willingness to consider doing so in some cases has led to demands that we do so in every case. That does not encourage me, as a member of the Committee who generally supports transparency where beneficial to the encyclopedia, and in fact has a chilling effect; indeed this is the second time in 24 hours that I have seen a similar comment, both times in areas where we have been moving toward more (but not complete) transparency. Risker (talk) 17:11, 26 June 2009 (UTC)

When people are doing a power grab, they always say, "This isn't a power grab." The community banned somebody. You should not be unbanning them without obtaining input first. Plain and simple. No amount of sophistry will convince me otherwise. Jehochman ^Talk 20:56, 26 June 2009 (UTC)

Ah Jehochman. What input can we reasonably expect? This is the opposite of a consensus discussion in that it starts with an unnatural situation (blocked, banned or otherwise constrained), and there is limited prospect of uninvolved community members participating. Arbcom actions don't attract the attention of truly uninvolved community members the way that AFD or DRV or RFA does. Perhaps the solution is for the community to come up with a "banning" process that requires something more than a 12-hour discussion amongst a half dozen people about an already blocked user. Risker (talk) 21:22, 26 June 2009 (UTC)

Risker, I addressed this in my proposal below: You enforce WP:CIV and WP:NPA on the discussion page, limit the discussion period and that will minimize the kind of input that you're worried about. Put up a nice colorful template across the top of the page warning people that discussion rules will be enforced with a heavy hand (suggested color: black). It will allow for the kind of input that Jehochman is concerned about: good arguments against an unban and relevant facts and links. Your having been elected doesn't eliminate the need for accountability. -- Noroton (talk) 21:44, 26 June 2009 (UTC)

The problem is that community discussions are not a level playing field. The appellant rarely gets a fair chance to present their case and there's a heavy inegality of arms. Usually, they don't even get the opportunity to respond on equal terms. Please don't conflate accountability with this.  Roger Davies ^talk 22:19, 26 June 2009 (UTC)

How does that justify a secret proceeding where the community gets zero input? Your assertion is illogical. There forms of governance in between mob rule and the star chamber. Jehochman ^Talk 12:03, 27 June 2009 (UTC)

Perhaps you are unfamiliar with the fact that, despite similar warnings on Arbcom pages now, any attempts to address uncivil comments or anything short of absolutely blatant personal attacks is met with cries of censorship, regardless of who takes the action; if it is arbitrators or clerks, our recusal is demanded. I'm having a hard time seeing how behaviour would be any better on these pages than (for example) WP:ANI. What do we do with commments like "This account was blocked because of xxx" which, while it may well have been the blocking reason, was subsequently found to be untrue? What about when people allude to off-site activities or personal emails they claim to have, without producing them or linking to them? What about when people use it for a platform to further a personal agenda? Shall we edit all that out? As to accountability, I will personally reblock any account that edits outside of its restrictions without hesitation; it was only by luck that someone else beat me to blocking the last time that happened. If you would like me to add that to every unblock that the Committee makes, I will be happy to do so regardless of what position I took on the discussion, with the exception of recusal.

I am interested to know who you foresee participating in these discussions, if not the same group who were involved in the block in the first place. Nobody has yet identified how we would get neutral, uninvolved editors to participate; without them, there is no reason to believe the community as an entity is involved in the discussion. Risker (talk) 22:16, 26 June 2009 (UTC)

I'm not proposing that the community decide the matter, but that they be allowed to comment, as they are allowed to comment on Arb cases, for the purpose of providing information and arguments. If you avoid blatant CIV and NPA violations, you've done what you need to do. Expect the discussion to be very uncomfortable for the blocked party, but I doubt it would be worse than the original block discussion. Of course there will be complaints. So. What. You'll get them no matter what you do. The purpose is not to get neutral, uninvolved editors to participate but to get what we get now at Arbcom discussion pages: editors who want to make a point and who may or may not have a good point to make. It's always your job to weed out the fair from the unfair accusations and evidence and arguments.

As for people using it as a platform to further a personal agenda, well, you mandate that the comments must be about whether it's a good idea for Wikipedia to remove the block. This would work like a public comment period at a parole hearing. Everyone knows that the arbs will have private information that may be much more important than anything said in public. As a favor to the person unblocked, and for the best interests of Wikipedia, you might delete the discussion page once the unblock is done. Accountability isn't solved by your offering to reblock (although that's good); it's done by making the essential votes public, giving a public explanation (however vague it may need to be) and then having the rest of us consider how well your various unblock decisions held up. That helps maintain faith in the system and helps the system work better. If it turns out that the discussion pages are somehow too horrible to continue, we could do everything else in my proposal and have ArbCom accept emailed evidence, arguments and suggestions, although I doubt that would work as well. -- Noroton (talk) 00:37, 27 June 2009 (UTC)

Noroton, I confess that I had a big smile on my face as I read your comment above. It really is nice to have a discussion with someone who, despite all his time here, is still something of a wiki-idealist. Thanks for giving my cynicism a jolt. At the end of the day, though, I think we both know what will happen. Half a dozen people will comment against the unban (with comments ranging from the sensible to "burn the witch!"), maybe one person will comment in favour of it, the arguments that were presented were considered and the Committee decides to unblock (perhaps with conditions based on the submissions), and then there are screams of failing to (a) listen to the community or (b) properly assess consensus. Incidentally, not once has anyone asked me to explain the reason for my vote on any unblock discussion; I'd be happy to give it, but don't think it is reasonable for all of us to do that every time. We get about 7-8 requests a month, the majority of which are addressed uncontroversially. Risker (talk) 01:01, 27 June 2009 (UTC)

Hey, don't knock my idealism, because much of it involves the ability of a group of elected editors to act responsibly in most cases. Arbcom is perfectly capable of ignoring cries to "burn the witch" (or even ignoring wrongheaded comments that aren't blatant violations of civility or blatant personal attacks). No idealism is involved in seeing value in community input on a discussion page. I think giving concerned editors a chance to comment is a way of recognizing their dignity. Sure, the discussion likely will hurt the feelings of the banned person -- but without blatant WP:CIV and WP:NPA problems, I don't see that as too high a price to pay. In practical terms, the discussion, combined with a final ArbCom decision to unban, might result in a move to overturn ArbCom at AN/I. Well, let that be a check on ArbCom -- one difficult to implement and easily derailed with a reasonable public explanation (and sometimes a community overturn could be justified). Weigh that against the hurt feelings, suspicion of ArbCom mismanagement (or worse) as a result of the present process, as well as ArbCom acting without good information that can come from public input. Seven to eight requests a month? I didn't know there were that many. -- Noroton (talk) 19:42, 28 June 2009 (UTC)

One of the problems with the "let's let the community tell them off, that'll reform them" approach is that many problem editors are very young and emotionally fragile (and that's often the root of attention-seeking or disruptive behaviour). Many editors are frightened of public appeals and refuse to even think about one. Others, who have been through it and been rejected, don't want a repeat performance. They just can't handle it: these people are just as capable of rehabiliation as others.  Roger Davies ^talk 20:07, 28 June 2009 (UTC)

As a description of my proposal or arguments for it, let's let the community tell them off, that'll reform them steps into strawman territory. Otherwise, you make a good point. I'm having a difficult time thinking about community-banned or indef blocked editors as too emotionally fragile to handle a public comment page on their return. But I have to admit, this is not something I have experience and ArbCom members have a lot of experience in considering unbans. At this point, I'd consider your point as something that rightly weighs against a proposal like mine (at least it weighs against having a public comment page vs. emails to ArbCom), while other considerations weigh in favor of it. As a practical matter, I think no policy change is going to get enough community approval if ArbCom members who are familiar with the appeals are mostly against it. I think a lot of editors are going to defer to present and past ArbCom members who are familiar with these behind-closed-doors appeals. -- Noroton (talk) 20:48, 28 June 2009 (UTC)

I'd also be interested in hearing how Noroton would deal with stuff like, for instance, Jehochman's clown picture. Should that be in or out? Does it has a place in reasoned debate? Or is it okay? Just a bit of fun, albeit highly prejudicial, to lighten the mood?  Roger Davies ^talk 22:22, 26 June 2009 (UTC)

The black banner at the top of the discussion page would create such an imposing black cloud of looming doom over all participants that no one would dare put a clown picture on the page for fear that the arb-appointed Executioner would chop off their access to the 'pedia for a day or two or seven, and eventually that mindset would pervade all discussions without participants even thinking about it. I am assuming here that Wikipedia is capable of imposing this kind of very serious order on a discussion page. You would need a very responsible, worthy admin to police the page. Appoint Jehochman for the first case. While that would have a nice sense of irony to it, I'm serious. He'd be fine. -- Noroton (talk) 23:50, 26 June 2009 (UTC)

• There should be some input/feedback whenever the Arbcom overturns the specific and expressed opinion of the community. However, I don't see that we have that much say. . .considering the recent admonishment of Taivo for re-publishing email received [3] even though Wikipedia:Private_correspondence was specifically rejected. Just type up what they do and put it in the policy. (Let me add that I do respect Stephen for abstaining, given his participation in the Priv. Corr. debate). R. Baley (talk) 14:12, 26 June 2009 (UTC)

• ArbCom is there to deal with situations where resolution by the general body of editors has failed. Except in cases where ban appeals have not yielded consensus of any degree, or have resulted in wheel warring, the Committee plainly has no business overruling community discussions, and in general should stick to the job they were appointed to do instead of relentlessly encroaching on the project's non-hierarchical consensus decision-making processes.  Skomorokh  17:05, 26 June 2009 (UTC)

That argument only holds water if editors unbanned by ArbCom go on to cause more trouble. If they reform, and contribute sensibly, then it was unbanning was obviously in the best interests of the encyclopedia.  Roger Davies ^talk 22:09, 26 June 2009 (UTC)

No, "grant Caesar whatever powers he wishes, as he knows what is best for Rome" is not a responsible foundation for governance. By all means, there are cases where off-wiki communicaiton is preferable, due to privacy or accessibility concerns; it might well be desireable to have a designated group of editors to review ban appeals (though I doubt it, and even if there is, I can't imagine it would be the Committee); and limited and self-selected participation in initial may undermine the legitimacy and propriety of some community bans; none of this, however, is sufficient reason to grant the Arbitration Committee power to make the decision to unban an editor. Again, the Committee exists to act where all other avenues have failed, not to overrule those which have worked. This is not a matter of what will yield the best narrow and immediate consequences, but of due process and the distribution of power. It is all too readily forgotten that Wikipedia is not merely an encyclopaedia, but a free and open encyclopaedia ran first and foremost by its editors.  Skomorokh  22:40, 26 June 2009 (UTC)

I'm not quite sure how a ban can possiby be described as a process that has worked. It's simply a stand off. The ultimate object of dispute resolution surely is to create a situation where formerly irreconcilable parties can work together harmoniously.  Roger Davies ^talk 22:59, 26 June 2009 (UTC)

Sometimes a standoff is the best you can hope for. This site has its share of wacky and criminal types (not referring to anybody in particular here). Jehochman ^Talk 12:06, 27 June 2009 (UTC)

Skomorokh, I agree with you about this being a "free and open encyclopedia"; as I have indicated above, the natural state of all editors is to be unblocked and free of personal editing restrictions, and that is all part of being "free and open". Your comment above, however, implies that banning people is evidence of a process that has worked. I'd suggest that it's evidence of a process that didn't work because we now have an editor who wants to edit and cannot. In that sense, I agree quite strongly with Roger.

To be honest, I am not opposed to having a designated community group review at least the majority of community-based blocks and bans,including public discussion where applicable, and have supported such an idea(Wikipedia:Devolution) in the past. Given the appropriate mix of personnel (maybe a combination of community members and those with advanced privileges), this could still work in a lot of cases. On re-reading that page today, it reminds me of how frequently the Arbitration Committee refers to personal, private information when rendering some of these decisions. Making discussion "strongly urged" rather than mandatory would likely address most of my concerns. Risker (talk) 23:09, 26 June 2009 (UTC)

Folks should re-read the current policy. It does not envision off-wiki, private appeals at all. So this is language that would allow what is current practice, and add that the community should have some form of input. Risker seems to think that it would validate drama, but it is standard on Wikipedia to hear all sides. For example, what would DRV be like if only an article's original author could be heard? Less dramatic, certainly, but also less fair. Risker seems to beleive that hearing from only one person with a vested interest is better than hearing from several. We've elected the ArbCom members to resolve disputes, and the best way to do that is to hear all significant points of view, just like when they hear an RFAR.

If this proposal isn't acceptable then an alternative is to request that the ArbCom follow the current policy by holding appeals on-wiki.   Will Beback  talk  18:21, 26 June 2009 (UTC)

Will has a point, and I find it concerning that Risker has such antipathy for input from the "community" - this functions like every other discussion. It's boggling that we have an ArbCom member, elected by the "community" (at least those interested in voting), who is arguing that they should be making decisions with less information instead of more. You can continue with the secret deliberations and wikis, but you should be doing so with a full deck of cards, and that includes what information the "community" may have. -->David Shankbone 18:33, 26 June 2009 (UTC)

In Risker's defense, she also has a good point: if ever there is a time to limit community discussion, it's about sensitive matters that often can create a lot of hurt and where some of the input for the decision is (often) necessarily private. There are legitimate, competing concerns here. -- Noroton (talk) 18:47, 26 June 2009 (UTC)

Noroton, with respect I disagree with that perspective. There is no more hurt involved than in Arbitration cases themselves where people give their input. I don't see ban appeals--where they are seeking to overturn the desire of this community--as any different. I don't buy the argument Risker is making. -->David Shankbone 19:08, 26 June 2009 (UTC)

• Sorry to be so long-winded, but I'm trying to propose something that balances the concerns raised above and elsewhere. Some good objections have been brought up that (a) there's little to be gained from publicizing an unban request if the ban is unlikely to be lifted anyway; (b) opponents may further poison the atmosphere in a public discussion where they hope to influence ArbCom. These concerns compete with some values that the present practice tends to erode: (A) that Wikipedia is run with the input of its editors wherever possible; (B) ArbCom should be accountable to the editors and make decisions, when possible, in public, so that editors can have some assurance that decisions are made with some care and for sincere reasons (this works like a check and a balance on ArbCom's power, which obviously needs to be pretty strong here because we're dealing with sensitive, often private information). Can we all agree that banning policy should sacrifice a, b, A and B as little as possible? So I'd change Will's proposal this way:

If an appeal is heard off-Wiki and if the ArbCom has voted tentatively to unban a user, ArbCom shall follow these steps:

(1) After a vote, ArbCom shall post a notification on Wikipedia:Arbitration Committee/Noticeboard requesting input for at least one week before making a final decision, in which discussion may be moderated by designated Arbcom members, clerks or administrators who will enforce WP:Civility, WP:No personal attacks and other behavioral guidelines and policies in the discussion, along with any other restrictions ArbCom may decide to make in the discussion;

(2) The notice for the tentative decision will identify the ArbCom members who voted for and against or who abstained.

(3) ArbCom, at its discretion, may include in the notification for the tentative decision a brief statement explaining why ending the ban is better for Wikipedia.

(4) ArbCom, at its discretion, may designate one or more arbitrators to answer questions about the preliminary decision.

(5) ArbCom must publicly notify the community of a final unblock decision and that notification must specify which ArbCom members voted for, against or abstained, and it must provide a reason for that unblock (with the level of detail entirely at the committee's discretion).

Government agencies do these requests for public input all the time before issuing final decisions, and what I'm proposing follows that pattern. Yes, it's a bit bureaucratic, but I think only to the extent that it's necessary -- a growing community is going to get a bit more bureaucratic. (The additional, mandated bureaucratic rigamarole: (i) preliminary vote, (ii) notification for input, (iii) one-week discussion, (iv) public notice of an unban that must include the vote, specifying individual ArbCom votes as well as (v) some kind of explanation.) The community doesn't need to hear about every unban request, only the ones that ArbCom itself thinks should be successful. Arbcom should be notifying the community before a final decision to give the community a chance to come up with information and reasons why lifting the ban would be a good or a bad idea. By moderating the discussion on Wikipedia, under Wikipedia WP:CIV and WP:NPA restrictions, less poison is emitted into the atmosphere — but let's face it, Risker, the person who got banned very likely put a lot more poison out into the atmosphere and can't expect to be welcomed back with a lei and an "Aloha!". And if people concerned about the unblock are given a chance to tell ArbCom about their concerns, they're more likely to accept any unblock decision. Forcing the ArbCom to make a statement -- any statement -- on the reasons why the decision was made gives the community assurance that at least some thought went into the decision behind the scenes, and gives some toehold of accountability that the public can later use to assess arbitrator behavior at election time. Having a designated arbitrator to answer questions (it will probably be the arbitarator who dealt with the unblock request the most) means that discretion will be more likely observed than if several arbitrators say several things in an uncoordinated way, leading to confusion and improper revelations. Again, sorry about being long-winded. -- Noroton (talk) 18:29, 26 June 2009 (UTC) several minor tweaks -- Noroton (talk) 18:47, 26 June 2009 (UTC)

I agree with the underlying principles of much of this though I'm not wild about a week long public discussion. This seems to me an awfully good way of raking up stuff that may no longer be relevant (ie assurances made not to repeat etc). My view is that a low-profile low-drama return to editing is probably best for formerly banned users and this doesn't exactly set the scene for it.  Roger Davies ^talk 22:31, 26 June 2009 (UTC)

Great minds think alike, at least to a degree. We both drafted at the same time. Bear in mind that we don't actually need the Committee's permission in order to solve this. The community can review its own bans. We don't have to come to the arbitrators hat in hand. Durova^Charge! 18:54, 26 June 2009 (UTC)

Yes, some of this overlaps with your point #4, below. I don't know enough to have an opinion about setting up a separate committee to do this. I think we both agree that the community is capable of having reasonable discussions about these things. I think these kinds of decisions are so sensitive that much of the deliberating and discussion with the banned person is going to need to be behind closed doors. Electing people to this sensitive position helps with accountability and trust, whether it's ArbCom or a separate board. -- Noroton (talk) 22:05, 26 June 2009 (UTC)

• Responses to many: First off, until very recently, the community did not "ban" a user, it would only support or oppose the indefinite block of an editor. Block reviews are standard Arbcom fare and have been since Day One of the Committee, as a venue of last resort. I don't see this changing; Arbcom will still be the venue of last resort, even if a process for community unblock consideration is developed. We can't get enough admins to staff the unblock-en-L email list now, and the proposals presented here still don't address the fact that blocked/banned editors are not permitted to edit onsite to make requests or participate in their resolution. If, however, an editor wishes to appeal his or her community-imposed ban or indefinite block to the community, there is no Arbcom bar to doing so, except to find a community member willing to put their case forward. During the last six months (the period to which I can speak), every single unblock/unban request that the Committee has dealt with (regardless of outcome) has involved very thorough review of all relevant on-wiki information, including review of contributions and searches through relevant archives, and often consultation with one or more of the individuals directly involved in the block/ban. Administrators who implement longterm blocks/bans are expected to clearly document onwiki their reason for action; lack of such explanation (which could include links to relevant discussions) probably should be a fatal flaw in any longterm block or ban. The natural state of an editor is to be unblocked and unfettered by restrictions, and there needs to be sound justification for changing that status. One of my colleagues is currently working on some statistical information about Arbcom activities that I believe will include the number of unblock reviews that have been carried out. Finally, I do believe that the members of the community support the Arbcom candidates that they believe will show good judgment, even in situations where the judgment of what is best for the encyclopedia is not necessarily all that popular with a segment of the community. Risker (talk) 20:14, 26 June 2009 (UTC)
  • Actually, Risker, it was David Gerard who first added language to the banning policy about community bans in summer 2005.[4] It's been policy nearly four years, and our method of handling it has become archaic. Durova^Charge! 22:45, 26 June 2009 (UTC)

A suggestion

Caveat: this is a sensitive issue for me to comment upon in the current context. People who know the history are aware why. So to be explicit, I've made no formal complaint against the Committee's decision regarding a recent unban that has sparked discussion.

There's a suggestion I've submitted to the Committee list twice in the last seven months regarding community ban reviews. Actually there's no need to obtain ArbCom's consent to move forward with this solution. So summarizing here. If people like it, let's create a proposal.

1. Banned editors who want to return legitimately are curious how to do so. As a website we've been very disorganized about setting up standards and expectations for that. the approach that's been taken has often been haphazard. That's led to a lot of trouble on all sides: extra work for the Committee, bad results of both sorts (meritorious requests ignored, non-meritorious ones granted), and would-be reformers resorting to hardened vandalism in despair.
2. Not every request for unban should be granted (or even considered), yet it ought to be possible to articulate a standard set of expectations that can apply to most cases. A draft already exists at Wikipedia:Standard offer, which may serve as a model.
3. Although community ban reviews ultimately devolve to ArbCom, the community can create a mechanism for reviewing its own bans. This can provide opportunity for feedback and evidence to assist the community and the Committee achieve fair decisions.
4. The basic idea is this: set up a panel to screen unban requests based upon community-agreed standards (this weeds out people who are actively socking to post death threats and other obviously inappropriate stuff). If a request passes basic screening it gets posted in a public location. The community is welcome to submit relevant evidence and discuss parameters for a possible structured return. If nothing obvious arises to knock the matter out of consideration after one week, the proposal moves to AN where the community decides.
5. The screening panel would have a cross-section of established users: at least one CU, respected admins possibly including former arbitrators, and ideally at least one formerly banned user who has regained the community's trust (Rootology would be ideal for this if he's interested).

Banned users could still appeal to the Committee. The Committee would benefit from a lower volume of direct requests, and also from evidence and commentary that comes from the screening phase. The community gets a more mellow and informative decision making process. Banned users get a straightforward set of answers to what they can do and where they stand. And editors who consider themselves to have been hounded (or worse) get a fair chance to participate in these decisions before the outcome is decided.

That's the basics. Shall we move ahead? Durova^Charge! 18:51, 26 June 2009 (UTC)

That is a very good idea in my view. I see no reason why we have always delegated formal unbanning procedures to the ArbCom (or now, BASC) when the community is perfectly able and willing to handle procedures by itself. ArbCom has lots of work on its plate these days; if we can take some off their hands in a reasonable and orderly manner, so much the better. NW (Talk) 18:56, 26 June 2009 (UTC)

There's absolutely no reason something like that cannot be implemented. By definition, ArbCom is the appeal of last resort— it's just that there is currently no reasonable prior venue in which to attempt having a ban lifted. Having a community driven first level of appeal is perfectly compatible with this, and will indeed lighten ArbCom's load to some degree (though, I expect, not as much as one would hope at first glance: most of the appeals that are rejected by the BASC would most likely also be rejected by this community process, and would often end up appealed back to BASC). — Coren ^(talk) 21:40, 26 June 2009 (UTC)

It's far easier to review an appeal when evidence from both sides is already available. A community process would provide a broader picture for situations that hadn't passed through the Committee's hands in the initial stages. That should make it easier to assess the appeals that continue to return to ArbCom. Durova^Charge! 22:41, 26 June 2009 (UTC)

• Durova, this is brilliant insight. I really commend you for breathing some fresh air into this discussion. I agree with you completely and if I can help you to set a system up, I'm at your disposal. -->David Shankbone 22:47, 26 June 2009 (UTC)

• I haven't been following this discussion closely, but I'm a bit confused about why it's being treated like such a complex problem. As I understand the situation, a "community ban" is where an administrator indef blocks a user and no other administrator is willing to lift or shorten the block. Why do we need some bureaucratic oversight body, complete with "panels" and "evidence" and so on, to "review" these "appeals"? If a single administrator is enough to overturn a community ban (as has happened in at least one instance) why are we making it complicated? ⟳ausa کui ^× 02:17, 27 June 2009 (UTC)

  Because, in practice, those situations lead to wheel war. — Coren ^(talk) 02:31, 27 June 2009 (UTC)

  Has anyone tried blocking the wheel-warring administrator? For example, if you indef block a user claiming "community ban", I remove the block, and you reapply it under the same rationale, it seems it should be transparent to everyone who would be in the wrong there. If any administrator removes or shortens the block, the user is not community banned, and you would have to get him banned through conventional means. Therefore if someone re-imposes a contentious community ban after an administrator reverses it, that admin is wheel warring, so block him/her. Problem solved. Is there something I'm missing? ⟳ausa کui ^× 02:34, 27 June 2009 (UTC)

  Coren and I were involved in something like that at Wikipedia:Requests for arbitration/Sadi Carnot. If there is a community ban, there needs to be a discussion before unbanning. If the discussion fails to achieve consensus, then ArbCom is available to break the deadlock. In the alternative, if a ban relies upon confidential evidence, ArbCom could handle the matter in the first instance, but then they should seek input from the community if at all possible Jehochman ^Talk 03:11, 27 June 2009 (UTC)

  I'm sure one of us is mistaken. If a community ban is in place and an administrator disagrees, that administrator can veto it by unblocking the user or shortening the block. The definition of a community ban is that there is no administrator who is willing to reverse it. Therefore, there is no room for discussion; the plain facts resolve the debate. ⟳ausa کui ^× 03:24, 27 June 2009 (UTC)

  No, an administrator cannot single handedly remove a ban. A community ban is established when no administrator objects. Once the discussion closes, a ban needs a consensus to be lifted, or an action of ArbCom, which they are unwise to do without some sort of community input. Jehochman ^Talk 12:08, 27 June 2009 (UTC)

Jehochman: I'm sure you're right, but I'm sure that that's not very fair, especially if the community ban was in error. It seems to fall afoul of WP:CCC :-/ --Kim Bruning (talk) 20:32, 27 June 2009 (UTC)

Reposting from Causa sui's user talk

Consider it this way:

1. A hates B and breaks policy to hound B. After six months of abuse A gets sitebanned.
2. Months later, A appeals the ban.
3. B is unaware of the appeal.
4. Admin X unblocks, not seeing a problem.
5. B learns of the block, and protests "My userspace was vandalized five times in the last week. I think it might have been A."
6. Admin X replies, "Well A is unblocked now. We'll wait and see if there are further problems."
7. A renews vendetta against B, gets rebanned.
8. B retires in disgust.

This is one of the ways that Wikipedia loses good editors. Doesn't it make sense to let B into the discussion? Durova^Charge! 03:39, 27 June 2009 (UTC)

That's certainly one scenario. Consider an alternative view of the same.

1. A and B both have time on their hands and contribute to many topics but both have strong views about one particular topic, where they clash.
2. This develops into deep personal animosity friction between A and B.
3. A is aggressive but naive; B is goading but experienced. Six of one, half a dozen of the other, but B is better at playing the system.
4. A gets blocked a couple of times by an admin sympathetic to B, then site-banned.
5. After a few months of complete absence and missing Wikipedia, A appeals, promising sincerely to avoid B altogether.
6. B makes a highly emotional victim impact statement dismissing A's promise and effectively hijacking the hearing; B's mates pile on, bystanders exprsss shock and horror. A's reputation is comprehensively trashed as the entire focus is on the negative aspects of his behaviour.
7. A's appeal is rejected.
8. A feels he had been treated unfairly, victimised and publicly humiliated, and decides to get his own back. Massive disruption, outings, and drama with off-wiki stuff ensues for years, tying up admin time, checkuser time, and causing many editors to leave.

In fact, the thing missing in both scenarios is an opportunity for A to reform under very tightly controlled/restricted conditions.  Roger Davies ^talk 04:44, 27 June 2009 (UTC)

The scenario the Roger has outlined may be a reason to hold appeals in private, but I don't see how it shows that the appeals need to be held secretly, with no notification of the community or opportunity to email additional evidence.   Will Beback  talk  05:01, 27 June 2009 (UTC)

I'm not advocating "secret" appeals at all (see my proposal above) but as I also pointed out elsewhere ArbCom is aiming to overhaul the current procedures over the next few weeks (it's been on the public agenda for some months now) and I'd personally like to see that include a structured procedure for community input by email. This would be particularly useful in fashioning editing/conduct restrictions. The 2009 committee is, despite the cries of its detractors, much committed to increased transparency (cf. WP:AC/N, AUSC etc) though it will take time to implement this right across the board as it usually means procedures have to be specially developed.  Roger Davies ^talk 05:26, 27 June 2009 (UTC)

The problem with Roger's proposal is its top-down approach. There's no actual need for ArbCom to take the reins with a community review structure; the fact that it's been on the agenda for months without action suggests that the Committee is too overloaded with core functions to be sufficiently effective. The people who already know the background to a case are often the best ones to handle it--and could certainly do so more swiftly than the Committee trying to start from square one each time. Durova^Charge! 05:47, 27 June 2009 (UTC)

Hm; let's see. A and B have been in conflict. A handled the conflict so poorly as to get sitebanned; B was a better Wikipedian and retained editing privileges. So at review time, ArbCom seeks banned editor A's input and presumes so little of B that it shuts out B from any opportunity to offer evidence? What has B done to sink so low as to have all input dismissed in advance as presumptively worthless? The community really does need a self-review system if arbitrators think that way. Durova^Charge! 05:18, 27 June 2009 (UTC)

I don't recognise this interpretation from what I wrote. Who said anything about shutting B out? And why are you ignoring what I wrote about B's behaviour being a contributing factor, and what I said about less than entirely impartial admins? These are very real everyday problems.  Roger Davies ^talk 05:32, 27 June 2009 (UTC)

Shutting B out is exactly what ArbCom often does, including recently. Your post can be read as an intellectual framework for that method of decision making. Durova^Charge! 05:41, 27 June 2009 (UTC)

I still don't see how you arrived at your extraordinary conclusion but never mind.

I suppose the underlying questions are (a) is the purpose of a ban punitive or protective and (b) to what extent should "victims rights" (for want of a better expression") dominate the discussion? For example, many disruptive editors are very young and simply grow out of the disruptive behaviour as they mature. In these instances, the "victim" brings an historic perspective from perhaps one or two years ago that may have not any bearing on current behaviour.  Roger Davies ^talk 06:05, 27 June 2009 (UTC)

I agree with Roger. Community bans were intended to be for the most unambiguous cases only, and this is basically why. An administrator who wants to reverse a community ban should not have to go through all this procedure. ⟳ausa کui ^× 06:34, 27 June 2009 (UTC)

Upon further consideration, Roger's scenario is basically an argument for clerking. Durova^Charge! 05:50, 27 June 2009 (UTC)

Is it? I don't think these sort of discussions can be clerked vigorously enough to exclude, for example, highly inflammatory but perhaps factually inaccurate comments. This would be seen as heavy-handed censorship in some quarters and lead, in and of itself, to even more high drama. In an environment where ArbCom is regularly charactised as a bunch of incompetent controlling clowns, this is probably a realistic prognosis. If, of course, a less strident, more collegiale, atmosphere were to prevail, anything is possible.  Roger Davies ^talk 06:05, 27 June 2009 (UTC)

Roger, just admit that you all handled this incident imperfectly, that you all have heard our point that the community wants to have a chance to provide input, and that you'll do better in the future. The only clownish thing about your behavior is that you stubbornly refuse to hear a legitimate objection to the way ArbCom handled this unban. It's plain as day how things should work and we're laughing because you folks on the Committee seem to be the only ones who don't get it. Jehochman ^Talk 12:12, 27 June 2009 (UTC)

Actually, Jehochman, the response received here has done a great deal to assure me that we handled this correctly. It has confirmed to me that we would have had an extended period of legitimized banned editor bashing, on and off wiki. Causa sui has it right. Risker (talk) 14:02, 27 June 2009 (UTC)

Allow a statement for the record, if this isn't already clear enough, that this is certainly not a screen for "banned editor bashing". There are structural shortcomings in the way this class of matter gets handled. It was in fact humanitarian concern for a banned editor that first prompted this draft proposal more than half a year ago. Durova^Charge! 15:39, 27 June 2009 (UTC)

The 2009 ArbCom understands that the past methods for handling unban requests needed to be changed. So as a part of restructuring the way that the Committee works, an Ban Appeal Subcommittee was started. Promptly addressing the backlog of requests in an effective and thorough way was the top priority. As well, the Committee is beginning to do formal reviews of all our work processes. When the Ban Appeal Subcommittee's work is reviewed, then editors can offer their opinions about how the Community and the Committee can work together to make the process give the best results. I don't see any emergency that require the Committee and the Community to stop there other work and immediately redo the process. The situation is quite complex having evolved over year. Most people need to have time to better educate themselves about the situation so that can offer well thought our opinions during a discussion about a possible redraft of the policy. As well, there never has been agreement in the Community about how to define a Community ban. Since this is a key element of the problem, I think that a discussion about Community bans needs to start with that topic. FloNight♥♥♥ 19:23, 27 June 2009 (UTC)

Since you agree that the current process needs to be improved, in the interim can you reassure us that there will be no more summary unbans without any opportunity for community input? Why can't you filter through all the unban requests, post an announcement saying who you plan to unban, and then allow a reasonable time for community input? Assuming that no unexpected new information comes to light during that comment period, the unbans become effective. You have clerks who can remove any inappropriate commentary, and any private evidence can be submitted by email. Jehochman ^Talk 11:43, 28 June 2009 (UTC)

Banned editor bashing

Can anybody post a diff of banned editor bashing in the current discussion? Risker seems to believe that bashing was such a great risk that it required handling the matter secretly. As far as I can tell, Mr Kohs has received a few warm welcomes back. In the past I supported his unbanning, and still do. Any criticism or "bashing" has been exclusively directed at the Committee for precluding community input on an important decision. Jehochman ^Talk 17:11, 27 June 2009 (UTC)

Should make it clear that on several occasions I've offered to initiate a community unban discussion for the fellow (per WP:SO). Was never ideologically opposed to the idea of him returning. Agreeing with Jehochman regarding the direction of criticism. Durova^Charge! 17:40, 27 June 2009 (UTC)

While you may see your offer to start an unban discussion as being a good approach and generous, the person that is requesting the unban may view the offer differently. You need to respect that idea. Having an editor's return derailed by users that thrive on drama is something that both the editor and the Committee want to avoid. I think that many users need to self reflect about the role that they play in promoting drama on site. Now we regularly blank ArbCom case talk pages because conduct of users commenting is so rude that leaving the comments on site is offensive to the involved editors. The unblock list is very understaffed. Before starting an other venue and process that will require involvement and time commitment from other people, perhaps it would be more beneficial to get more editors involved in this area of Wikipedia that is understaffed, and will reduce the problems at the front end of the situation. FloNight♥♥♥ 18:55, 27 June 2009 (UTC)

Sometimes the commenting of some arbitrators is so rude that the case pages need to be blanked. I've never received an official apology for the abuse I suffered at your (collective) hands. Perhaps that's why I'm still in a pissy mood. I see a lot of lip service from ArbCom, but when I was the subject of a case, I was not well-treated. Jehochman ^Talk 19:54, 27 June 2009 (UTC)

The current ArbCom intends to be professional in the way the it approaches their work. If you see something that does not hold to the standards that we want to present to the Community then feedback is welcome. FloNight♥♥♥ 20:21, 27 June 2009 (UTC)

Yet the ArbCom has made unban decisions which were unequivocally bad: mistakes which could have been avoided if the community had an opportunity for input. Consider the canonical example of User:Poetlister. Bad input can always be discarded; productive Wikipedians appreciate good faith and respect. There's a balance to be struck, and it is unwise to ignore the complaints of the site's most productive volunteers. You have in this discussion David Shankbone--the Wikimedian who interviewed Shimon Peres and who photographed Madonna; also myself, who at 273 featured credits am WMF's most prolific contributor of featured content. We both have stated dissatisfaction with current process, and specifically with the current standards of professionalism to which ArbCom holds itself. There is something slyly parochial about about responses that go over very much like, "Don't you worry your pretty little heads about it; let the wise munificent Arbitrators make their decisions from on high without listening to you. Now toddle along and generate more content." We are intelligent adults and would like to receive treatment as such. Our skills could earn substantial compensation in other venues; consider yourselves fortunate to have (thus far) retained us. Our patience is not infinite. Durova^Charge! 22:31, 27 June 2009 (UTC)

The PoetLister fiasco is due in no small part to the fact that some (myself included) were inclined to disregard views coming from a certain clique that tended to extreme drama-queen-ism when the unbanning of certain users was being considered, with hysterical-toned missives that basically said "They were banned... that proves they're evil, so they must never be unbanned! Anyway, when they got banned, they got angry about it and said rude things about Wikipedians, and became active on those slimy Attack Sites, so it proves the ban was the right thing to do and should be maintained! They're part of the crowd that attacks, outs, and maybe rapes and murders Wikipedians!" Too much of that sort of rhetoric, and I adopt mental filters that may also discount the rare cases where the bad stuff turns out to be true. *Dan T.* (talk) 17:03, 28 June 2009 (UTC)

The Poetlister unblock is a good example of remarkably canny work on the part of the Arbitration Committee. There was considerable disagreement within the community about the appropriateness of the initial block. After extensive review with the wider WMF community, the decision was made to unblock a single account on this project, with strict editing restrictions, which were constantly monitored by the Arbitration Committee. As a result, we now have a solid, widely supported ban on a clearly problematic editor, not just on en-Wikipedia, but on several other projects where that editor was active. Most community members would consider it a net benefit that we were able to clean out this sock nest thoroughly and completely. The community would never have been able to do this on its own because it required careful coordination amongst a range of individuals with access to private information in several projects. Risker (talk) 17:37, 28 June 2009 (UTC)

All is fine

Let me go on the record as one community member - I suspect one of many, but can only speak for myself - who is completely comfortable with the ArbCom handling ban appeals in private, with or without actively soliciting either community or individual input as they feel appropriate in any given case. The role of the committee is to deal with issues of user conduct that the community has proven itself unable to handle. The sheer volume of discussion one recent case has generated, and the stridency into which some of the discussion above on this page has degenerated, is in my opinion prima facie evidence why issues like this should be dealt with as much as possible outside public view and without public participation. I trust that a committee we have elected will have sufficient wisdom to look into the history of a banned user and if necessary solicit opinions from relevant parties before unbanning, or to make the judgement call that they have all necessary information and do not need to do so. The harm done from getting it wrong - someone gets unbanned and proves that they are still impossible to work with, leading to being re-banned soon - is small compared to the needless drama that can be generated from needless re-airing of conflicts and grievances. We've elected a committee of people to take care of the nasty jobs. Let's let them get on with the job, accepting that on any given decision we may not all always as individuals agree with the decision they reach. Martinp (talk) 14:48, 27 June 2009 (UTC)

I wish I could agree with Martinp, but the Committee doesn't always get it right in closed door sessions. cough. Here's what inspired the draft a community review process:

I received an email from Bus stop asking about an unblock on 1 July at the end of April. I said he would have to seek wider authority, and that I would not intervene, but would give him feedback on his attempt. I have lost track of the countless emails he has sent and the gaps between, waiting for a response, being encouraged, and then finally getting no reply whatsoever, so having to move on to someone else. He has been restrained and polite throughout, and displayed the patience of a saint. The process made me feel frustrated and angry, just watching it. This has involved an arb clerk, two three arbs (and I forget who else). I'm sure they're busy people, but if that's the case and, as a result they leave someone in complete limbo, then there's something very wrong with the system. He has been going through this process for five nearly seven months, when it should have been settled in one at the outside. He's played by the rules, sat it out, not socked (I feel pretty sure of that), and, to be quite honest, I'm surprised he can still have any esteem left for the project, but he does and obviously believes in its value. That was his mistake in the first place - excessive and misplaced zeal ...and being on the losing side of the argument. I found some of his points were valid. However, that is not the issue. He has made strong statements in his emails about voluntary boundaries and a desire not to get embroiled in the same problems or the same subject areas. I strongly support his reinstatement of editing privileges. There is no guarantee what the outcome will be, but that is up to him. Ty 22:53, 20 November 2008 (UTC)[5]

The community lifted the ban in three days, and the returning editor has since earned two barnstars with no problems. Yet followup research revealed that over seven months, four arbitrators and an arbitration clerk had all dropped the ball in succession and given him no answer at all. Durova^Charge! 15:34, 27 June 2009 (UTC)

That's a brilliant example, Durova, and perhaps one of the best reasons for having alternative methods for users to return. Developing a more systematic, timely and better tracked process to address requests for unblock/unban was a high priority for the Arbitration Committee this year, and I think it is heading in the right direction so that cases like this won't fall through the cracks. Nonetheless, I too support users having the opportunity to request a review of a community ban from the community itself, and (speaking personally) I appreciate your efforts in this area. Risker (talk) 18:25, 27 June 2009 (UTC)

Thank you very much, Risker. This particular example occurred before Risker joined the Committee. What came to my attention at that time was that two proposals already existed within the Committee itself, which had been on the table for something in the time frame of half a year. So at this point the matter has stretched to approximately a full calendar year without tangible result. With respect extended toward the Committee's hard work, it appears perhaps that this is one of the functions that could most effectively be devolved principally upon the community. The Committee retains its power of review, yet also gets summary reports of recent events--which saves a lot of effort and helps yield good decisions. Seems like everyone benefits. Durova^Charge! 18:44, 27 June 2009 (UTC)

Probably the best way for the user to return is to get the block removed by any administrator, since it's in the power of any administrator to review and overturn a community ban. I am really baffled by this. ⟳ausa کui ^× 19:19, 27 June 2009 (UTC)

It is not in the power of any administrator to unilaterally overturn a community-endorsed ban, only a "defacto" community ban can be reversed without discussion at the whim of any 1 administrator (even then it's probably not advisable to do so without some discussion). Please understand the diff between a community endorsed ban and a defacto community ban. R. Baley (talk) 19:28, 27 June 2009 (UTC)

Many editors have never agreed that a pocket of editors can hold a short ban discussion, and then enforce the ban in the name of the Community in such a way that the ban must remain in place unless another Community unban discussion occurs. It is old school Wikipedia thinking that users that can make good edits should not be easily voted off of Wikipedia.

Originally, the idea was that the admin closing the discussion was doing the block in their own name. So, the "exhaust the patience of the Community" type bans are done today in the name of the Community but that does not stop some people from disagreeing with them and the circumstances surrounding them. We need to respect the various views of the all people. Also, the middle of an unban discussion is a bad time to have policy discussions because of the involved user gets caught up in the minutia of the process. So, it is good to not intermix policy discussion with specific unban appeals. FloNight♥♥♥ 20:13, 27 June 2009 (UTC)

No kidding. I recognize the difference between a "community-endorsed" ban and a "defacto" ban, because the former does not exist, whereas the latter does. You cannot ban people from Wikipedia by hosting an impromptu firing squad on AN/I and use that to browbeat people who log on the next morning and object to what you've done. Maybe I'm in the "old school" as FloNight described it, but I think that this idea of "votes for bans" (in place of votes for deletion) is dangerous at best. I haven't found myself moved to overturn any so-called "community endorsed" bans but I surely won't hesitate to do so if I am. ⟳ausa کui ^× 20:31, 27 June 2009 (UTC)

(EC with above)I came to the conclusion some time ago that all Wikipedia decisions were made by single individuals, and that even ArbComm decisions, which are collectively made through voting rules, are, strictly speaking, only advisory, though the latter comment would take some explanation! The individual making the decision and enforcing it -- same thing -- is responsible for it, in both senses. So in an AfD, for example, a closing admin may decide, based on evidence, or sometimes on snow, to delete. What the admin can decide, the admin can undecide. A deleting admin need not go back to the community to undelete the article, perhaps based on new evidence or arguments that, had they been originally present, the neutral admin would have decided differently. This is highly efficient as a system, it represents how Wikipedia actually manages to be efficient, when it's understood. When we have the illusion that "the community" has made a decision when only a miniscule percentage of the community is even aware of it, and quite possibly a highly biased sample, we then set up a necessity for even more disruption to undo a possibly unwise or unfair decision, and this revisiting is likely to attract the same set of editors who participated originally, at least preferentially. Bans are no different; by this theory, a ban should always have a neutral closing admin who can make detailed decisions as necessary, such as length, whether or not the ban is still needed, what is covered and what is not, whether blocks are needed, etc. There is then a go-to person for attempting to resolve issues related to the ban. There is a manager to accept or reject a possible mentor. (A ban supervising admin should not get involved in assisting the banned editor with content, which could affect neutrality.) There is a manager with the tools to protect the banned editor from harassment. And if disputes arise between the supervising admin and the editor, there is ordinary dispute resolution, which can be highly efficient, as it starts with one more editor as a possible mediator. It all works if editors understand the importance of seeking consensus, which often requires patience. --Abd (talk) 20:46, 27 June 2009 (UTC)

(outdent) The principal issue here is one of providing for informed decision making. Regardless of where the ultimate authority rests, I hope we can agree that having good information in hand is beneficial and worthwhile. Durova^Charge! 22:11, 27 June 2009 (UTC)

I keep bringing this back up because the proposals you've been making seem to have the so-called "community endorsed ban" structure in mind, and so putting them into practice would implicitly legitimize that fraudulent practice. Maybe I've misunderstood you, but I don't think we can discuss that secondary question -- how to get more information and more perspectives into the discussion -- until we've settled this issue of whether an impromptu court martial on AN/I (composed of whoever is logged on at the time, or whoever is paying close attention to the issue) can make binding decisions on "community bans." If the answer to that question is yes, then this is a kangaroo court. ⟳ausa کui ^× 22:14, 27 June 2009 (UTC)

Then we misunderstand each other. To my understanding, the way to prevent kangaroo courts is to open a matter for input and evidence from both sides. How does it work against fairness to welcome evidence? Regardless of who makes the final decision, or how that gets decided, a mechanism for open input should be useful. Your quarrel seems to be with a different phase of process. Durova^Charge! 22:59, 27 June 2009 (UTC)

It is, and that's the point I was trying to make in the comment you replied to, so I'll try again: I don't want to talk about how we're going to get evidence into the process until we have settled what the process is. Your proposal only makes sense if we accept the so-called "community endorsed bans" in the first place. I am trying to head off anyone producing this big bureaucracy in support of that institution, because (A) it would be wasted effort of the institution fails and (B) it would legitimize the institution by grandfathering it into its own charter. ⟳ausa کui ^× 23:42, 27 June 2009 (UTC)

You appear to be out of date regarding what that consensus is. Am not disputing that because either way there is no linkage: evidence from both sides of an issue is equally valuable no matter where the decision rests. Durova^Charge! 23:56, 27 June 2009 (UTC)

I assure you, I have no disagreement with you on the second point. ⟳ausa کui ^× 04:47, 28 June 2009 (UTC)

To MartinP's original point: No disrespect is meant to the fine members of the ArbCom by saying that they are not omnisicent. They have no way of knowing what they don't kow unless they seek input or conduct their own investigation (which would also require asking input from at least selected users). If they don't then they are relying solely on the input from the banned user. I heard that a recent appeal was "crowd-sourced" - the user sought and received considerable help in from users of a forum in preparing his appeal, while at the same time members of this community had no involvement or even knowledge that an appeal was being heard. It's not a radical proposal to simply ask that affected users be notified and given an opportunity to comment.   Will Beback  talk  04:58, 28 June 2009 (UTC)

I agree. It seems that a whole lot of guff is being made about nothing here. I'll make some small but bold edits to the policy and we'll see if that gets us anywhere. ⟳ausa کui ^× 20:34, 28 June 2009 (UTC)

I tend to agree with Martinp … in most cases. There are cases, though, where I do indeed think that wider community involvement – and by this, I mean more broadly than that afforded at AN/I – is a good idea and those cases involve the most disruptive cases. As Roger notes, most appellants are sockpuppet abusers who have matured and are ready to perform as constructive editors. I think most here are not unduly concerned over the rehabilitation of such cases where they have been more annoyances than disruptive. It is in the cases of possibly erroneous decisions and in the readmittance of those who have egregiously "exhausted the community's patience" that community feedback should be sought. The former instance is one of "juridical review" and should be explicitly held to seek out "missing" exculpatory evidence. The latter "parole review" is much more contentious and potentially damaging to the well-being and productivity of the community, so that is what I will concentrate on here.

As R. Bailey noted previously, "There should be some input/feedback whenever the Arbcom overturns the specific and expressed opinion of the community." Let's face it, someone who has 273 blocks for disruption has worked as hard for their banning as someone who has 273 featured credits to their account; however, it's not the former the project needs to keep around. To the extent that they've earned the enmity of the community, the standards for any readmittance should be higher. At a minimum, it should be up to them to convince the community of their "change of heart". And, yes, permitting this will produce wikidrama; it's unavoidable. I'm not overly sympathetic to the idea that those who have worked hard for their disrepute should have their sensitivities weighed higher than the concerns of their past victims among the constructive members of the community. We need to keep in mind that it was the appellant who "poisoned the atmosphere" in the first place. I also believe that readmittance ought to be conditioned, in part, on their acquiring a mentor; if they cannot, then that is itself a strong statement on the community's behalf. If the appellant succeeds in recovering the community's forbearance to return, though, in fairness to them and their new start, they ought to have the option to return under a new name – but still under whatever restrictions set by ArbCom and/or the community – and the mentor should have full access to information on the background to their original banning. Likewise, the community discussion over the reinstatement should be blanked, if they so request, to prevent its being further used against them.

And, yes, the wikidrama of a discussion over the return of a banned editor is an unavoidable and – in terms of fairness all around – necessary, however distasteful. With due respect to those making the suggestion, neutral administrators will not necessarily be able to cull out inappropriate commentary. People handle their anger, frustration and other sources of stress in different ways. Some put it into words pithily, and "censoring" "as unhelpful" it may light their fuse, leading them to explode in a blockable fashion; you're going to have to let each have their say in their own way. There are already enough editors commenting that disruptive editors receive preferential treatment vis-à-vis productive editors

In the end, though, the final decision should still remain with the ArbCom; the discussion should be to inform them, not to serve as a !vote … and the level of objection itself may give ArbCom a valuable forecasting tool for risk management in terms of providing a degree of insight into just how much damage that individual's return may incur. As with other "closures," it's any new evidence of culpability or non-culpability that they should be looking for – and if their decision must rely on privileged information, then that should be part of the final finding (which should itself be public). Askari Mark (Talk) 22:03, 4 July 2009 (UTC)

Wikiproject User Rehab can help here

This is a relatively new project and it needs the help and wisdom of the experienced editors, admins, and ArbCom members here. This project has the potential for dealing with these problems if it gets the proper input, advice, and help. I notice these comments (emphasis added):

• In fact, the thing missing in both scenarios is an opportunity for A to reform under very tightly controlled/restricted conditions.  Roger Davies ^talk 04:44, 27 June 2009 (UTC)

• That's a brilliant example, Durova, and perhaps one of the best reasons for having alternative methods for users to return. Developing a more systematic, timely and better tracked process to address requests for unblock/unban was a high priority for the Arbitration Committee this year, and I think it is heading in the right direction so that cases like this won't fall through the cracks. Nonetheless, I too support users having the opportunity to request a review of a community ban from the community itself, and (speaking personally) I appreciate your efforts in this area. Risker (talk) 18:25, 27 June 2009 (UTC)

• "if the community had an opportunity for input" Durova^Charge! 22:31, 27 June 2009 (UTC)

• NOTE: Comments made here have been moved down; the original formatting made it appear as though this was posted by Durova, when in fact it was posted by Bullrangifer. Risker (talk) 19:51, 28 June 2009 (UTC)

I see some of the people I most highly respect speaking out here, and if the writers in this thread were to actively support and guide this project, it could become an official process that everyone could use, with the community getting the vital input that is missing. Mentors, adopters, ArbCom members, admins, the community, all using an official process to deal with some editors who have really screwed up, but who have also wisened up and are willing to reform. They deserve a process that will help them return properly and efficiently. If this can work and become an official process here, then we might be on our way to solving some of the problems being discussed. That can happen if we start using it, working out the growing pains and initial foibles always associated with new processes, and perfect it. -- Brangifer (talk) 15:39, 28 June 2009 (UTC)

WikiProject User Rehab is a wonderful idea, but isn't ready yet. So far its capacity hasn't developed sufficiently. On June 13 I petitioned the project for assistance seeking new mentors for 3-5 people. Haven't gotten any responses from project participants in two weeks. Durova^Charge! 15:58, 28 June 2009 (UTC)

We need a method for contacting possible mentors. Where is a central place to make invitations for assistance? The mentor system isn't very well-organized. -- Brangifer (talk) 16:10, 28 June 2009 (UTC)

Durova BullRangifer, I am not impressed with your selective quoting and highlighting above; your agenda is rather obvious here. The Arbitration Committee has developed a more systematic, timely and better tracked process to address unblock/unban requests that come to the Committee. If an indefinitely blocked or community-banned editor wishes to request a community review, they can do so if they can persuade another editor to bring their case to the community; that's always been an option. It cannot, however, be a requirement; it requires the editor to have prior social relationships with one or more members of the community, or alternately to create a new account so that they can email someone, which itself violates the block/ban. The processes Durova and Noroton have suggested do nothing to take the pulse of the community as a whole, and are focused to obtain the opinions of people who already have formulated an opinion or are focused on very narrowly watched areas of Wikipedia space. Risker (talk) 16:58, 28 June 2009 (UTC) moved from above, where formatting made it appear as though it had been posted by Durova. Apologies to Durova for not spotting this sooner. Risker (talk) 19:51, 28 June 2009 (UTC)

I think that's right. Both should be available because the community appeal route isn't an option for some users. Cool Hand Luke 19:06, 28 June 2009 (UTC)

Risker, my proposal has nothing to do with taking the pulse of the community. Getting "the opinions of people who already have formulated an opinion" is almost a feature, not a bug of my proposal -- I want people (a) in the know about the banned editor and people (b) concerned, to be able to know a ban may be lifted and to have a public opportunity to tell ArbCom why lifting the ban may be a bad idea. I think there's some value in these editors seeing each others' comments, and having the community see those comments. I think there's some value in the possibility that the community, outraged at some ArbCom decision, may well then take over and reban the editor, and that possibility acts as a mechanism keeping ArbCom on its toes. -- Noroton (talk) 19:58, 28 June 2009 (UTC)

Risker, my apologies for "selective quoting". I haven't read every comment above and just noticed them. I have no "agenda" other than what amounts to a "FYI", since I was unaware of the "more systematic, timely and better tracked process to address unblock/unban requests that come to the Committee" you mention. Please provide a link and I'll do some reading. The project is not intended to replace or interfere with any other existing processes, and if it does, then we need to be made aware of it and coordinate things so as not to create any problems. The project's intention is to deal with banned editors who express a desire to reform and need help to return, and most of them will likely have not been banned by ArbCom. If they have been, then ArbCom's processes should be used, or at least the ArbCom should be involved. I'm sorry if you got a bad impression. Just let us know of any concerns and you'll find us to be receptive. -- Brangifer (talk) 20:28, 28 June 2009 (UTC)

Hi Brangifer. Apologies if I have misread your perspective. The announcement of the Ban Appeal Subcommittee is here, and the procedure is here. Supporting returning editors is a good thing. Risker (talk) 02:44, 29 June 2009 (UTC)

Risker appears to presume that editors in good standing are a net negative to discussion of ban reviews. The opposite is more often true: editors who know a history frequently have noteworthy evidence; former quarrels may be set aside when concerns are respected and addressed. To follow up on two examples above, if any of the arbitrators who received the Bus stop appeals had contacted the blocking administrator (myself), then it wouldn't have taken half a year to discover how gladly I welcomed his return; yet if Poetlister's unban appeal had been publicized before it became a fait accompli then I would have brought forward evidence which might have helped avoid a lot of subsequent trouble. There's a word to describe the formation of subcommittees while failing to adopt the basic principle of hearing evidence from both sides: that word is bureaurcacy. A Committee that desires the community's trust would do well to remember that respect is a two way street. Durova^Charge! 20:56, 28 June 2009 (UTC)

"Risker appears to presume that editors in good standing are a net negative to discussion of ban reviews" is way over the top.

The Poetlister situation is an extremely complicated one, and not a good example of anything I hope you will agree. If you had evidence about the Poetlister situation that wasn't known to the committee, that is in itself a problem. Or maybe you had told the committee, but they had lost the information into the archives. We are trying to prevent that latter situation by keeping better records on the arbwiki so that we can quickly find any emails sent to arbcom-l, related discussions and decisions of prior committees.

We are adding more subcommittees, but not for bureaucracy reasons - we are splitting the workload into different groups, so that we can cope with all the emails and requests that are sent to us. As far as I know, there has been no systematic review of oversight/checkuser, and the WP:AUSC is doing a fantastic job of this, without much involvement from the arbs. The ban subcommittee is having a lot more trouble getting into the swing of things because they are routinely being asked to investigate very old issues. They have been emailing Functionaries-en at times, in order to engage people who probably know what happened. We do need to engage the committee more in these unban situations, and our internal processes for unbans have been progressing slowly as we have time to review them. I havent spent a lot of time working on the unbans, nor have I extensively reviewed the private unban discussions from previous years, however from what I have seen the unban subcommittee is managing to keep on top of the workload far better than would be the case without a subcommittee. We don't want unbans to be dropped because nobody was motivated enough to respond to the unban request, as as far as I know, that isn't happening.

It must be kept in mind that people have many options to be unbanned other than approaching the committee. We are their last option, and we routinely tell them that they should use other methods of unbanning when they havent tried other approaches like {{unblock}}, the unblock list, etc. John Vandenberg ^(chat) 01:26, 29 June 2009 (UTC)

Hold on. You had evidence about Poetlister misbehaving but did not tell anyone? Did you contact a checkuser at Commons, where you are an administrator, and Poetlister had uploaded several images of the women whose identities he was usurping? Please send any evidence of socking or other similar inappropriate behaviour of banned or indefinitely blocked editors to Functionaries-L, so that our full roster of checkusers can review the evidence. It is one of the reasons we set up that email list—so that such information could be reviewed by the most appropriate group without going outside the confines of this project. The Arbitration Committee, as a group or as individual arbitrators, often consults with this team. And please don't put words in my mouth, that is not what I said at all. Risker (talk) 02:44, 29 June 2009 (UTC)

---

This almost deserves a subsection.

It is well known that Poetlister's principal activity of late 2007 and 2008 was occurring at another WMF project where I do not edit. Wikiquote's internal site management is none of my business, and I had been particularly encouraged to avoid sock investigations during the months that preceded Poetlister's February 2008 unban at en:wiki. Of course I contacted an arbitrator promptly after the unban, but by then the decision was a fait accompli.

As one of the people who had assembled the spring 2007 evidence, Poetlister's excuses such as here[6] leaped out as gross misrepresentation. Deceit was the cornerstone of Poetlister's rise at Wikiquote; the question was how much deceit occurred. Poetlister leveraged bureaucrat status at that project into an unban at this one. It was quite well played.

There's a significant difference to whether cross-wiki evidence emerges before an unban or afterward. Beforehand it reflects upon an editor's suitability to return; afterward the standard is to respect the autonomy of separate projects and focus upon current behavior at the site where an editor has been unbanned. So there was little motivation to pursue the leads that were raising my eyebrow. Particularly because, after having been right in dozens of investigations and wrong in one, a part of the community would have ignored any evidence that came from me simply because of who had submitted it.

Although it would be edifying to believe the arbitrators would not ignore evidence out of prejudice, another instance erodes confidence in that hope: during both the Prem Rawat I and the Sarah Palin wheel war arbitrations I submitted offsite evidence regarding a senior administrator. Several times I contacted arbitrators and resubmitted the material. During the second arbitration I also sent it to a former arbitrator, who was alarmed by what it contained and petitioned the Committee to consider it. No pertinent action was forthcoming. So finally, during the Scientology arbitration I published a summarized report in user space. Within a day and a half of publication the subject of that evidence resigned and retired. A sockpuppet of his has since been indefinitely blocked. Yet to the best of my knowledge not a single active arbitrator had read the submission in either of the prior two cases.

Getting back to Poetlister, the sensible thing to do was wait for others to see for themselves that this person was dishonest. It wasn't easy to wait--and if I had been aware of the personality rights issues that existed with that person's uploads I certainly wouldn't have waited. Those uploads and the sole complaint concerning them had occurred long before I became active at Commons. Commons has as many images as Wikipedia has articles: it would be farcical to chide one individual for events from years before their tenure. As soon as the image issues came to light I posted public regrets for not having suspected it myself. Although I may be a Wiki witch, I don't have the crystal ball or time machine to remedy that course of events.

My work on the 2007 Poetlister report was the first time that time stamp analysis was used in an extended and systematic fashion. Mine was one of several contributions that helped resolve suspicions which had existed since 2005. Why did Poetlister single out myself and SlimVirgin for retribution, but not others? Perhaps it's related to the reason he created the Taxwoman impersonation. The pattern is suggestive of gender bias.

The entire Poetlister sockfarm at a critical site was quite diligent for over a year spreading lies and misinformation about me. A substantial part of that damage survived his tenure. One former arbitrator thanked me privately for what I did; none apologized for the additional damage that Poetlister's 2008 return had inflicted upon my reputation. No public statement of support or affirmation has been forthcoming from the people who best know my real work here, and now--regrettably--an arbitrator who was not on the Committee when any of these events occurred uses a policy talk page to post a broadside attack against my competence in this very issue. The hyperbolic suppositions of that arbitrator could have been allayed by running a five minute check on the Commons chronology in advance of posting. A retraction would be most welcome. Usually I hold my tongue about these things, but enough is earnestly enough. Durova^{273 featured contributions} 05:21, 29 June 2009 (UTC)

The Poetlister bit doesn't seem like a great example—FT2 did in fact always suspect socking and unblocked in order to prove it, but your point is taken. Please be careful not to mischaracterize the involvement of Risker and Roger. Cool Hand Luke 15:11, 29 June 2009 (UTC)

Neither Risker nor Roger were arbitrators during that period. Haven't implied anywhere that they were, and if there's been an ambiguous use of syntax that gives such an impression please point it out. Getting back to the main focus of this discussion, the underlying example is very pertinent: when arbitrators suspect wrongdoing and unblock in order to prove it, the Committee makes its own task harder when it excludes bilateral input. That caused collateral damage. SlimVirgin and and I endured months of additional ridicule; possibly it took a greater toll upon Poetlister's well-meaning supporters. One of them who mended fences with me in the aftermath also retired because the outcome left him embarrassed and exhausted: for a while he had really thought Poetlister's unban was vindication.

Yet really, one shouldn't need examples to support the notion that it's good practice to hear both sides before reaching a decision. That's standard in arbitration actions generally, both on Wikipedia and elsewhere. Undisclosed ban reviews are the exception, not the rule. The radical action is the Committee's departure from that norm, and after last year's experiences the undesirability of that course of action ought to be evident (if it were ever in need of doubt). The principal argument against openness is a theoretical potential for disruption: the same potential exists in regular arbitrations and in community ban reviews; we handle that. It is disappointing to see new Committee members who were elected to enact reform argue against one of its cornerstones. Durova^{273 featured contributions} 01:15, 30 June 2009 (UTC)

There are some huge assumptions here that ArbCom decide in a vacuuum based entirely on the contents of the appellant's email. This is just not the case. All the relevant wiki-process stuff and talk pages etc, are reviewed, where of course there is ample evidence of the appellant's misdeeds.

I'm not sure how relevant historical examples of ArbCom shortcomings are to this discussion. Apart from anything else, (i) 11/15 arbitrators are new to the committee this year and (ii) most, if not all, were elected on platforms of reform, which are demonstrably underway.

You keep citing extreme and rare examples, which is considerably skewing the discussion. For the record, most appeals involve sockpuppeteers: many of those don't actually do anything particularly terrible with their socks; haven't engaged in harrassment or personal attacks; and seem to get bored with socking after they've matured a bit.  Roger Davies ^talk 07:49, 30 June 2009 (UTC)

---

Don't we allow banned users to email the unblock list? If so, they can ask to be unblocked, and then the receiving administrator can post the request at WP:AN for discussion if it appears to have a chance of success. It seems like we may need to get more administrators involved with the unblock list. That approach is much more scalable than asking the already overburdened arbitrators to handle yet another task. Jehochman ^Talk 01:20, 29 June 2009 (UTC)

Jehochman, are you offering to man the unblock list?  ;-) John Vandenberg ^(chat) 01:28, 29 June 2009 (UTC)

I would consider it, if help is needed. Is there any chance we could install some open source help desk software? Jehochman ^Talk 01:38, 29 June 2009 (UTC)

IMO, help is needed and your assistance would be greatly appreciated. I would also like to see some analysis of how the list traffic is being handled. There have been discussions about using open source help desk software to handle the traffic; once on the list you could ask them for pointers to the list archives where it has been discussed in the past. John Vandenberg ^(chat) 01:49, 29 June 2009 (UTC)

The current policy seems to say that clerks are responsible for copying ban appeals to an on-wiki page.   Will Beback  talk  01:26, 30 June 2009 (UTC)

We dont often see people requesting a public appeal to Arbcom, however one recent one was an incorrect socking block.[7] IIRC, it was filed by Telaviv1 (talk · contribs) on behalf of Boatduty177177 (talk · contribs), but there was some mixup on how it was filed.

I dont think clerks are responsible for doing this, but users should email someone rather than create a sockpuppet in order to file an appeal - arbcom and clerks are two convenient groups of people to email, as they are all well aware of the process, whereas the average admin may not have email enabled, or may not be very helpful once emailed. John Vandenberg ^(chat) 10:17, 30 June 2009 (UTC)

Yes, banned users can email the unblock list and ask for an admin to post an appeal on their behalf to AN. I'm having a hard time thinking of examples...one is User:Shapiros10 (now User:Shappy) who was community banned for socking and later wrote to unblock-en-l and his appeal was then posted to AN/ANI and he was subsequently unbanned. And yes, the unblock list currently needs help. It used to be a very active list but most of the most active admins helping there either moved on to OTRS or got burned out and several ended up leaving the project. The majority of emails to unblock-en-l are from people caught as collateral damage in range and IP blocks who are confused, alarmed and just need an explanation and reassurance, and a few blocked for reasons they don't understand and also just need an explanation. People appealing community bans are infrequent. Any admins with time and inclination to help such people would be most welcome. Sarah 06:54, 30 June 2009 (UTC)

Banned users may have been unfairly accused

It is possible that a banned user, in the days and hours leading up to a ban, has been improperly or unfairly accused of misdeeds that they either did not commit, or that have been framed in such a way to make them sound more heinous than what actually transpired. This was, in my opinion, a part of the circumstances surrounding my own community ban, leading to an ongoing sense of injustice both in my soul and in the minds of others who generally take a more critical view of Wikipedia's governance model. It would seem that for justice to take place, it would not be fair to take a recently unblocked user, accuse them (without evidence) of something they did not do, use their reaction to that accusation as grounds for re-blocking them, then have a quick (14-hour) community ban discussion about them, in which their right to respond is suspended thanks to the re-block.

Indeed, in a fair system, should such a person ever regain standing in such a community, you'd think it would be their right to similarly "put on trial" those who made false allegations against them leading to the ban; however, I will not elaborate any further, due to the fact that the provisional suspension of my ban is governed by a provision that I not engage –- in either an initiatory or retaliatory capacity -– in any form of feuding, quarreling, or personal attack, which I fear even a reasonable, calm elaboration here would be misconstrued as such. On a good day, we hold no continuing animosity against the sorts of individuals I have mentioned above; we only a wish that justice and fairness might take a more prominent and respected role on Wikipedia and across the Internet. On a bad day, though, our thoughts and retaliatory responses toward those whom we feel have wronged us have been outlandish. That's how insidious is Wikipedia's power to alter human thought and behavior.

In sum, we ought to think about a more formal ban appeal process that allows the accused to present evidence to unaffiliated parties of the various ways in which the subject was wronged by others, and allow for some form of justice to be meted out upon those who unduly mistreated the banned user, if that is the determination of the body of neutral observers. Most of the discussion I see taking place above generally assumes that the banned user was never unfairly provoked by the unjust behavior of others. And we should remember that some people when provoked become downright nasty, and that would be my apology to those who have read this far my insignificant thoughts on this subject. -- Thekohser 20:41, 1 July 2009 (UTC)

Confirming this: During 2008, I saw a clear cut case of exactly this; a user who was accused of actions that a close check on their edits showed to be either unlikely or part of a "group attack" by opposing users with more experience of the "system" framing his words in a dark light. The user concerned from memory had been trying to express concerns over possible non-compliance with core policies as best it seemed. On wikibreak now hence not digging up details but yes I have seen it happen and it is (correct use of the word I think) extremely wrong and shameful if it does. It's doubly so if it goes undetected or unappealed, if experienced editors or administrators become complicit by not checking before opining, or if users and the community in general allow themselves and each other to be a vehicle for legitimizing such allegations and converting them into a ban. It probably only applies to rare or very few cases, but... yes. Not good to come across a ban and realize it might be one of those. FT2 ^{(Talk | email)} 02:12, 2 July 2009 (UTC)

"consensus of editors who are not involved in the underlying dispute"

AdjustShift, what are you trying to achieve by unilaterally closing off the discussion above like that? Abd and Maunus were raising pertinent concerns about this Ban policy, its wording, intent and applicability. FWIW, I roundly agree with the points and arguments raised. There is at the minimum an apparent ambiguity in this page's instruction, if not outright inconsistency. That discussion had not reached a conclusion, I don't see why you should feel entitled to shut it down. The episode with NYScholar is being used as a concrete illustration, an example to compare and better understand the procedure(s) and circumstances under which someone may be declared "community banned".

Before you abruptly cut it off, the discussion had been querying the meaning and intent behind the clause, [t]opic or site bans may be implemented by a consensus of editors who are not involved in the underlying dispute. The clarifications being sought boil down to (a) what is its intended meaning, (b) what is its intended scope, and (c) to what extent should it be taken into account by the closer when they are making their deliberation.

As for (a), I'd agree with Will Beback that the phrase, consensus of editors who are not involved in the underlying dispute seems important, and concur with Maunus and Abd that to interpret bans may be implemented by a consensus [of otherwise uninvolved editors] as meaning "the consensus could be made up of uninvolved editors, or it might not, it doesn't really matter" is an unnatural and counterintuitive reading. If the degree of involvement of editors forming the "consensus" is not relevant, then why mention 'editors who are uninvolved' at all?

Instead, surely the intention of this clause is, at least in part, to provide some safeguards against the outcome being unduly or disproportionately influenced or decided by the !votes of participants from one side of the dispute (or the other)? In the ban discussion, isn't the intention here to see what uninvolved/independent parties make of the issue? --cjllw ʘ TALK 09:56, 6 July 2009 (UTC)

This talk page is for discussing how to ameliorate the Wikipedia:Banning policy page. But, the discussion in the "Banning of User:NYScholar" section is more about the banning of NYScholar rather than how to ameliorate the Wikipedia:Banning policy page. That's why I closed that discussion. Here on this talk page, we should discuss how can we ameliorate the Banning policy page. This talk page is not a place to discuss whether the community banning of NYScholar was correct or not. CJLL Wright, you have raised some nice points. I'm busy in RL now; I squeezed about three minutes to write this comment. :-) I'll give my full explanation after some hours.

I would also like to remind certain editors that this talk page is for discussing how to ameliorate the Banning policy page; this is not a place to talk about the banning of NYScholar. Please discuss how can we ameliorate the Wikipedia:Banning policy page. AdjustShift (talk) 10:43, 6 July 2009 (UTC)

AdjustShift, what are you trying to achieve by unilaterally closing off the discussion above like that? I was trying to stop needless discussion.

Abd and Maunus were raising pertinent concerns about this Ban policy, its wording, intent and applicability. That's not 100% accurate. Some of Abd's comments were more about why the banning of NYScholar was unjust rather than how to ameliorate the Wikipedia:Banning policy page. Maunus doesn't understand the community banning clearly. Read this and this comments of Maunus. AdjustShift (talk) 02:58, 7 July 2009 (UTC)

The question is not that I don't understand the policy - the issue is that the policy is ambiguous and several editors have admitted to the possibility of various different possible interpretations of the wordings. If you understand it so well then why don't you answer any of the questions about what those wordings are supposed to mean.·Maunus·ƛ· 03:29, 7 July 2009 (UTC)

The wordings on the Banning policy page can be improved. AdjustShift (talk) 12:22, 7 July 2009 (UTC)

You are being disingenious uncooperative here - It has been shown beyond a doubt that several editors find the exmaple of the banning of NYScholar to show some problems with the banning polciy. That is why I started this thread - several editors agreed that the points I raised were significant enough to discuss and then you come along and close the discussion. I don't believe you have any right to close discussions among editors who are CLEARLY trying to do exactly what they should - namely improve WIkipedias content and policies. You are now being told by an administrator who is not involved in the thread above that he ALSO sees that the discussion has its merits and instead of arhuing you merely contradict him and say that it hasn't.·Maunus·ƛ· 13:28, 6 July 2009 (UTC)

Please stop using words like "disingenuous". You are breaching WP:CIVIL. AdjustShift (talk) 02:58, 7 July 2009 (UTC)

I'm not uncooperative. I think some wordings on the Banning policy page can be improved. I closed the "Banning of NYScholar" section because it is more about the banning of NYScholar rather than how to ameliorate the Wikipedia:Banning policy page.[8] Some comments of Abd were clearly inappropriate and had nothing to do with improving the Banning policy page. Closing one thread is not the end of the world; anyone can start a thread to discuss things that will help to ameliorate the Banning policy page. AdjustShift (talk) 12:22, 7 July 2009 (UTC)

I am truly puzzled by the differeing interpretations of topic or site bans may be implemented by a consensus of editors who are not involved in the underlying dispute. Perhaps it is differences in British and American English? You see in English the word Should or Must are used when defining tightly what the meaning is. May is a somewhat vaguer word.

Anyway, whilst uninvolved editors should form part of the process, the opinions of involved editors should be taken into account. Not sure how these different groups are to be defined or their "membership" policed though. I don't think anything is served well by legislating for every last nut and bolt. Jezhotwells (talk) 13:20, 6 July 2009 (UTC)

Jezhotwells: look at it like this: If the policy stated that "consensus may be created by editors who have bad teeth" then there are three posible meanings of this sentence. Either consensus may be formed by editors with good teeth and ALSO by editors with bad teeth. Or it means that editors with bad teeth may form a consensus just like any other editor. Or it means that consensus may be formed only by editors with bad teeth. If the first or the second reading is the intended meaning then the stipulation that editrs with bad teeth may participate in forming consensus is completely redundant, and there is no reason to add the phrase about bad teeth at all. If the third reading is intended then it should be made more clear that only editors with bad teeth can form consensus. Now it is in the interest of the community that the policy be as clear as possible so that editors and admins don't have to guess which of the three meanings are intended which is why I propose that we discuss what the passage about uninvolved editors is supposed to mean and clarify the policy based on that discussion.·Maunus·ƛ· 13:28, 6 July 2009 (UTC)

It is pointless discussing what the editor who wrote that sentence meant. Why not ask them? It was User:Jehochman on 13 November 2008. [9]. I can't find any discussion about it. It was just added. Please also note how the word may is used elsewhere in this policy. Jezhotwells (talk) 14:04, 6 July 2009 (UTC)

The question isn't really what it was supposed to mean when written - but rather what we want it to mean. Do we want the policy to say that both involved and uninvolved editors can partake in a banning decision or that that only uninvolved editors can count towards a banning decision or do we want the policy to be somehwre in between? I think like you that it is unreasonable to exclude "involved" editors - above all because there is no good way to define "uninvolved" - but i do think like CJLLWright that it would be a good idea to somehow guard against banning decisions being taken by a group of editors ganging up on an editor with whom they have a dispute - so I think that some how impartial editors' voices should be given more weight in a banning discussion - but i don't know how this balance could be achieved. In terms of the banning policy it could be worded in a way to suggest that uinvolved editors' votes be given special weight in the consensus decision relative to the "involved" editors.·Maunus·ƛ· 14:29, 6 July 2009 (UTC)

(ec) To me, the most natural interpretation of the word may in that sentence is that it is there to signify the contingency of the outcome (the outcome might be that the person is banned, or it might be that they are not banned). It is not there to signify contingency in the makeup of the consensus (possibly they could be uninvolved editors, but they don't have to be).

Also, No-one here has been arguing that the opinions of involved editors should not be taken into account, or that it is improper for them to contribute to a ban discussion. But equally, I don't think we'd rightfully regard any discussion as representing strong 'community consensus', if the only participation came from those heavily invested and committed in the dispute.

Perhaps it would be better to pose the question as, what are the necessary and sufficient conditions in order for a "community ban" to be decided and imposed? --cjllw ʘ TALK 14:50, 6 July 2009 (UTC)

How to close banning discussions: suggestions for making a clearer policy

I haven't found it well-documented, but there are five aspects to a good close: the evidence presented, verification of the evidence and that it isn't cherry-picked or otherwise biased, the overall vote balance, analysis of the !votes according to involvement, and an adjusted vote balance.

• Where there is a snow conclusion, an admin may close based solely on !votes; however, the admin should always, then, be open to making a more complete review upon a questioning of the decision. The admin could ask for another editor to do the more complete analysis.
• There was no snow conclusion in the case cited, but there was an apparent strong preponderance of votes. It was still legitimate to close based on preponderance of votes, however, a more careful close would have uncovered the problem with involvement of editors. Note that even though consideration of involvement reduces the strong majority for site ban to a true no-consensus division or even possibly a minority,[10] the fact that so many editors were offended by the subject editor still establishes that there is a problem, but does raise the question of whether or not total ban would be the best solution. That the closing admin seems to conceive of the closing job as purely assessing a Yes or No answer to the question of site ban is part of the problem here. The community does not make decisions by vote; however, we do assess the level of consensus supporting a decision by such analysis. In claiming consensus for a ban, the admin clearly neglected the policy, and has yet to show understanding of or agreement with the policy.
• "Community ban" is a different decision than "Indef block," or other administrative ban. It's unusual in that it does require a consensus of uninvolved editors, not merely a preponderance of arguments as judged by an ordinary close. That's because it is a site-wide decision, not to be taken without the most thorough of consideration, except provisionally. While the ban discussion may take place at a noticeboard, usually there has been sufficient prior process that the evidence has been clearly gathered and presented, and the positions of editors and their involvement has been made clear. That's what I've seen: a user RfC, which never determines a ban but which can lay the foundation for it, and then an AN discussion to determine an action, if any. If, then, an appeal to ArbComm is necessary, the evidence is already in place, and possible division in the community has been made clear.

We should probably develop a better and clearer "ban procedure." This is not to deprive admins of the right to unilaterally declare a ban, or to declare one based on something short of a consensus as defined -- which can take a few hours of analysis to determine -- but to discriminate these from true community bans, which really do show that the patience of the community has been exhausted, not merely the patience of a relatively small number of editors who pile in to agree with the ban, when some of them, from the record, had little patience at all. Sure, others suffered long. The paradox is that if we had a clearer and less punitive procedure -- but still firm --, the problem of the case we have examined might not have lasted nearly as long as it did. Strong solutions should have been considered a long time ago; instead, the banned editor was led to believe that things were okay and that there were just a few editors who were in opposition. Things were not okay. There was a problem. --Abd (talk) 16:39, 6 July 2009 (UTC)

I agree that the NYScholar case has shown a need for a clearer procedure for bannings. I agree that the current policy is not optimal in its clarity about what kinds of offenses are grounds for bannings, how banning discussions are to be carried out, how a consensus is arrived at or what process should be followed by the closing administrator. However I don't believe this is the right place to question the particulars of the closure of the banning discussion of NYScholar - this should be done at a different venue. Such as a RfCl the villagfe pump or at ANI. It also seems to be problematic to at the same time state that the banning policy is not clear and to fault AdjustShift for his implementation of it: since we agree that the policy is unclear on several important he can hardly be taken to task for having interpreted it differently from how you and I would interpret it. ·Maunus·ƛ· 19:41, 6 July 2009 (UTC)

AdjustShift has presented his view of the situation. It is not our task here to "fault" him for it, but to understand it, and, in particular, to understand if the presentation of policy is unclear. I'm not stating that the policy is unclear, in fact, the banning policy clearly requires, for a community ban based on a noticeboard discussion, a consensus of uninvolved editors. The interpretations that the use of "may" indicates something different have been rather thoroughly deconstructed. What's unclear, though, is overall procedure that would implement the policy. At this point, though, we don't even have agreement on what the policy means and how to apply it. I've shown an analysis of an actual case, and surely it would be relevant to criticize that analysis if anyone disagrees with it. How did I determine which editors were involved and which ones were not? My decisions were to some extent arbitrary. There are other problems with the process that we could and should examine. I find it highly problematic that we didn't have the kind of systematic presentation and organization of evidence and comments that is typical in an RfC. WP:DR would suggest an RfC for the kind of behavioral problem in the subject case, and using a noticeboard to ban prior to RfC would only be appropriate if, indeed, there was an unmistakeable consensus of uninvolved editors, and sufficient evidence was presented in a neutral manner.

This is why I'm concerned with AdjustShift's position: it's not about the ban decision, which is why this is not being taken through an appeal process, it is quite simply about the process. If I decide to appeal the ban, it would be with certain preconditions being satisfied that don't exist yet. Did AdjustShift misunderstand the policy as written? If so, was the defect in the language of the policy, or was it in something missing that should be added, or was it somewhere else, possibly some basic misunderstanding of Wikipedia process by AdjustShift. None of this is about blaming or faulting AdjustShift, of whom I assume nothing but good faith. --Abd (talk) 19:56, 6 July 2009 (UTC)

You raised most of these points here, where I thought the answers you were given were unambiguous. (e.g. that an admin does not have to close a community ban) Socrates2008 (Talk) 13:36, 7 July 2009 (UTC)

Even though anyone can close a banning discussion it seems to me that it would still be a good thing to have a clear guideline for how such discussions should be conducted and closed - somthing we don't have presently.·Maunus·ƛ· 14:13, 7 July 2009 (UTC)

Community Ban

I admit that I have not read through all of the above discussion so if this point is already being addressed above please point me to it. In reading the section on community bans I find the following bullet:

• If a user has exhausted the community's patience to the point where an administrator has blocked the user long term or even indefinitely, and where no uninvolved administrator is willing to unblock him or her, the user is considered to be community-banned.

Now on the surface this statement seems self-evident but something about it bothers me. It suggests that a single administrator can, in effect, ban someone site wide without any explicit community discussion of actually banning them and without any explicit review by a reasonable number of uninvolved editors. This seems fundamentally wrong to me. It seems to be using blocking policy to create bans and these two are typically kept strictly separate because of their distinctly different purposes.

Should there not be a requirement that there actually be an explicit discussion of banning someone and having a community consensus demonstrated and explicitly declared before they are officially declared to have been banned by this policy? --GoRight (talk) 16:57, 7 July 2009 (UTC)

I don't see that problem - the review consists in the phrase "and where no uninvolved administrator is willing to unblock him or her" to me that means that if a single uninvolved editor is willing to unblock then the block does not turn into a defacto community ban. So actually the block must be reviewed by several admins who refuse to unblock before the block becomes an actual banning.·Maunus·ƛ· 17:19, 7 July 2009 (UTC)

Understood, but this all leaves the decision vague and in the shadows and (possibly) excludes community input by non-admins. Although I suppose if non-admins cared enough they would raise the issue visibly on AN or ANI and so a discussion would inevitable ensue but it may not end in a clear declaration either way (i.e. in that circumstance "admin silence = ban" which again seems wrong). Would we not be better off with a clear and explicit declaration that a ban either exists or not based on referenced community discussion with demonstrated involvement by otherwise uninvolved participants? --GoRight (talk) 17:35, 7 July 2009 (UTC)

Actually, the above bullet seems to be at odds with the following bullets from the section discussing the differences between bans and blocks:

• Indefinite blocks may be imposed by any uninvolved admin, after sufficient warnings and/or other shorter blocks
  • A site ban can only be imposed by community consensus, by the Arbitration Committee, or by Jimbo Wales

Personally I question whether having unnamed, uninvolved admins taking absolutely no identifiable action can truly be considered a valid way of determining community consensus. Can it? --GoRight (talk) 18:12, 7 July 2009 (UTC)

Ok, so we have a specific selfcontradiction within the policy, this must of course be remedied. I suggest simply distinguishing between bans and indef block's throughout the policy. However I don't think that if we have this distinction in the policy that we have a problem of "unnamed, uninvolved admins taking absolutely no identifiable action being considered a valid way of determining community consensus". The point is that indef blocks don't need consensus - they just need due process i.e. warnings and a sufficient rationale by the blocking admin. Bans need to be executed through consensus.·Maunus·ƛ· 18:24, 7 July 2009 (UTC)

As long as the bullet above remains as is you do have the case where "unnamed, uninvolved admins taking absolutely no identifiable action" (i.e. unnamed admins NOT unblocking someone) is in fact being used to gauge the required community consensus described in the ban and block differences section. I agree with your proposed manner of dealing with the self-contradiction though. Just keep the two (i.e. indefinite blocks and community site bans) separate. I would suggest that removing the bullet above from the policy resolves the entire problem without any loss of enforcement capability, although I am certainly open to other options. --GoRight (talk) 18:43, 7 July 2009 (UTC)

Ah, you were referring to the way the two bullets were set up - as if one was a subtopic of the other. Yes - they should definitely be two separate topics here. I'll remove the stray bullet.·Maunus·ƛ· 19:00, 7 July 2009 (UTC)

Hmm, not really, although I don't mind the change you made. Let me try to be more explicit and direct. I am saying that if this statement is true (and I think it is):

• A site ban can only be imposed by community consensus, by the Arbitration Committee, or by Jimbo Wales

that this statement in the policy:

• If a user has exhausted the community's patience to the point where an administrator has blocked the user long term or even indefinitely, and where no uninvolved administrator is willing to unblock him or her, the user is considered to be community-banned.

must logically imply that we are taking "no uninvolved administrator is willing to unblock him or her" to be an indication of community consensus (since it is the controlling factor and it has nothing to do with the Arbitration Committee, or Jimbo Wales). This is the equivalent of saying that non-actions (i.e. not unblocking someone) by anonymous administrators (i.e. uninvolved administrators unwilling to unblock) can be cited as community consensus. It just all seems bass ackwards.

I strongly believe that before someone is considered banned that there should be a clearly demonstrated community consensus on that specific point.

You seemed to be saying that we should just let indefinite blocks remain indefinite blocks and not call them bans. If so I agree and the easiest way to make that happen would be to remove this:

• If a user has exhausted the community's patience to the point where an administrator has blocked the user long term or even indefinitely, and where no uninvolved administrator is willing to unblock him or her, the user is considered to be community-banned.

from the Community Ban section of the current policy. Doing so would remove this seeming conflict without affecting the ability to impose blocks or issue bans by demonstrable community consensus. Why is it important that we call a indefinite block a ban? We all know that the effect is similar but the semantics and justifications are operationally distinct as far as I can tell. --GoRight (talk) 21:11, 7 July 2009 (UTC)

One of the problems here is that there are two things being done with the Policy. One is to prescribe, effectively, what clearly has broad consensus, and the other is an attempt to describe more completely the defacto situation by including under "ban" a situation that isn't a ban, but that has some similar effects. That's where the mischief arises. An admin indef blocks an editor. The editor disappears, does not appeal. Is the editor "banned"? No, not really. The editor is indef blocked. Practically speaking, there is a close resemblance to a ban, but the difference is what range of options are open to lift the ban.

My opinion is that we should stop allowing "community ban" to refer to defacto bans through indef block, without a consensus of uninvolved editors. It's much cleaner. A community ban, properly founded on such a consensus, should not be lifted without one of three conditions: consent of the closer, a new discussion at a similar or higher level, or appeal to ArbComm. (Appeal to Jimbo or, I suppose, the WMF Board, are theoretical possibilities, but I wouldn't waste the electrons unless there was a very good reason and a very penetrating argument.) But an indef block can be lifted without such a discussion, under conditions I will not describe now. Because we can't control how editors refer to things, we might still leave in language that explains that some editors use the term "community ban" loosely to refer to an indef blocked editor, but a true "community ban" stands up there with bans determined by ArbComm or Jimbo; it can only be lifted under narrow conditions. [Signed for Abd by --GoRight (talk) 21:55, 7 July 2009 (UTC)]

You, me and GoRight seems to be in agreement about this. I think it would be a good idea to propose this dstinction to the wider community.·Maunus·ƛ· 21:41, 7 July 2009 (UTC)

In second condition, not community banned unless he requests unblock and it is denied

See this, and consider User:JedRothwell, who was indef-blocked, and that no admin would most probably block, but who never put up an unblock template, so no admin ever had the opportunity to refuse an unblock. Under the old text he would be community banned, under the new one he would only be indef-blocked. Under the old text he would be reverted per WP:BAN and under the new one reverted per block evasion. So, which one would be better? --Enric Naval (talk) 03:55, 9 July 2009 (UTC)

That is not an accurate description - under the previous wording we would have no way of knowing whether he was banned or indef-blocked - because there is no way to know whether now or in the future some administrator might be willing to unblock him if he put up an unblock request. The old wording was ambiguous in that it provided no way to see when an indef-block turned into a defacto banning.·Maunus·ƛ· 15:48, 9 July 2009 (UTC)

What if the admin says that the user should be considered banned, but there is no unblock request? In that case it's clear that he's not simply indef-blocked, and the new wording doesn't cover this case. --Enric Naval (talk) 01:21, 10 July 2009 (UTC)

But that case is not founded in policy. The policy as it was before did not give any admin the power to declare a user banned. According to policy banning can only be made by community consensus OR if no admin is willing to unblock a blocked user.·Maunus·ƛ· 01:27, 10 July 2009 (UTC)

That's correct. It's quite enough that an admin can unilaterally block. A "community ban" is a community decision, necessarily, and no set of involved individuals -- unless that set is huge! -- nor even a single uninvolved individual, can establish that. An admin can declare, "You're banned," and can enforce that with blocks. But a community ban is to be enforced by the community, and being able to establish that presumption without discussion is way too much power for an individual to hold, ordinarily. From prior experience, I'd say that true community bans should be logged at WP:RESTRICT for as long as they are active. And one point should be kept in mind: the close of a community ban discussion should be by a neutral administrator, and determination of term should be as decided by that administrator.

Full disclosure: I'm currently considered page banned by an admin who declared the ban unilaterally; it was taken to AN/I by Enric Naval, who has a few axes to grind here, and it started to snow endorse. I saw that there was a major factional dispute involved, most editors endorsing had previously called for me to be banned, and there was no way that it would be resolved short of ArbComm, the most that I'd be able to find at AN/I would be enough support to break consensus, so I asked supporters not to comment there and asked for a rapid close. It was closed by a neutral admin, who determined, when asked (not by me!), that the ban length was one month. That's expired, but the original banning admin claims that he is still in charge of the ban, and he has extended it, giving as his reason my interest in policy. Definitely, that's his reason, it's been his thinking for a long time that I shouldn't be meddling with policy, especially not where he's involved. It's not about the article page in question, and it's blatant, but, folks, I bring this up for the policy issues, not for help with the ban. The ban is not that important, one article page and one Talk page. The point here is that he has declared that I'm banned. Am I? I'd say that at this point, the ban could not be logged at WP:RESTRICT, and if the admin wants to enforce the ban he has declared, he does so at his own risk. I'm not planning on poking him, this will be at ArbComm before I edit the article again.

--Abd (talk) 01:56, 10 July 2009 (UTC)

Hum, so, if an admin thinks that an editor should be banned, and not just indef-blocked, then he would need to open a community discussion on banning that editor. It would then fall back to the third condition. --Enric Naval (talk) 02:23, 10 July 2009 (UTC)

Yes, that seems to be the implications of the policy as it stands - whether this is the way it is enforced or the way the wikipedia community wants it to be is another question.·Maunus·ƛ· 02:27, 10 July 2009 (UTC)

Ok. --Enric Naval (talk) 03:32, 10 July 2009 (UTC)

An admin may unilaterally declare a "ban," but that doesn't make it a "community ban." This is a wiki, and one can't control how people use language, but "community ban" has implications that "indef block" doesn't have. Remember, in theory, anyone could declare a ban, since it doesn't involve tools. But enforcing it would be a different matter, eh? If any editor thinks an editor should be banned, the editor can start the appropriate discussion. What gets tricky, in my view, is when this is done at AN/I, especially. AN/I is a place where we will see lots of ill-considered comments from editors, sometimes in both directions, based on general ideas about who the good guys are and who the bad guys are, or on what policy should be, or on personal experience with the editor, which can be warped. AN is a little better, but not much. RfC is much better as a first step. There are a few editors who used to be very active calling, on AN/I for other editors to be blocked or banned, and often they were successful. They have been banned, and it probably took way too long. Blocks protect the project and the only reason for a ban is to inhibit further discussion, or wildcat admin unblocking, to avoid further disruption. Sometimes trying to seal the matter with a ban causes more disruption that it would prevent.

I have argued that voluntary site bans should be used much more often.

An admin might be ready to block,but, instead of blocking, the admin declares the ban on the user's talk page, and states that it is in lieu of blocking. Or first indef blocks, then offers to unblock on a condition that the editor respect stated rules, which can be completely flexible. It's only necessary to actually block if the editor defies this "voluntary ban." A block and a site ban are blunt instruments; a voluntary ban is less damaging, and, in fact, if the editor cooperates, establishes operating respect. Personally, I'd much rather have access to my user space even if otherwise site-banned. And then the ban, if I wished to dispute it, could be negotiated through ordinary DR, with edits to other pages presumably as permitted by the banning admin for that purpose. This will work, I expect, if both the banning admin and the banned editor operate with AGF and respect for process. Page and topic bans are always "voluntary bans" even if they are community or ArbComm bans, in the sense that I'm using. There is no technical procedure for prohibiting an editor from editing specific pages or classes of pages. Ultimately, this would be a desirable software feature. Once I stated an intention to ignore an admin-declared ban, but only after some days and concluding that it was the least disruptive method of initially "appealing" it; if I was blocked, an unblock template would then solicit a neutral admin to review the matter. I never actually violated the ban, at that point. However, because Enric took the ban to AN/I, that intention was frustrated, and we had the disruptive discussion I was trying to avoid anyway. When that discussion snowed endorse, as I expected from history, I asked that it be closed, and my later block for ban violation was based on a misunderstanding (as discussed here previously); I didn't put up an unblock template, because this was a community page ban now, and it clearly was going to take much more process before I'd be able to make a case.

--Abd (talk) 15:34, 10 July 2009 (UTC)

See "There doesn't have to be repeated declined requests. there are plenty of community banned users who have never requested unblock"[11] --Enric Naval (talk) 02:58, 16 July 2009 (UTC)

Enric Naval seems to think that that change I made to the page, with apparent consent here, and which he reverted, was motivated by a particular case. Not so. It's true that the distinction between indef block and ban came up a few months ago, for me, with that case, but the motivation is general and I was aware of the issue previously. I've seen a number of such cases. For example, Fredrick day was indef blocked. He did not put up an unblock template, he knew it was useless. He created a series of socks which were eventually discovered and indef blocked. Before a series of socks had been blocked, it was claimed that he was "banned." The claim was generally rejected. No ban decision had been made. Later, after a number of socks had been blocked, he was considered banned, and I do think there was a discussion. (If not, I'd argue he is not banned, merely indef blocked.)

In the case Enric raises, there never was a ban discussion closed as described in the policy. Perhaps Enric would debate that, and, if so, if there was a proper discussion and a consensus of uninvolved editors, then the policy would cover it in that section. However, JedRothwell was not blocked for actual blockable offenses, but apparently under WP:IAR and without discussion, "independently," but without explicit review of the evidence, quite apparently and explicitly to support an administrator whose action while involved was being challenged; ultimately ArbComm did determine action while involved, but this doesn't automatically undo any decisions. Nor is there any plan to dispute the block with the blocking admin, and, if such a dispute were pursued, the policy here would not be an issue. Rothwell and Fredrick day are not the point. The point is that any admin can readily undo the effect of an indef block, rather easily, as long as it isn't wheel-warring, which it probably wouldn't be in the Rothwell case. That's not true for a community ban, nor for an ArbComm ban. Otherwise "indef block" and "site ban" are roughly equivalent.

If I wanted Rothwell unblocked, I'd ask for that, specifically by asking the admin who blocked to unblock. And if I didn't get a satisfactory answer there, I'd follow dispute resolution, which I don't do unless there is both importance and expectation of success. --Abd (talk) 15:42, 9 July 2009 (UTC)

For the record, my motivation in seeking this clarification is to reduce pointless disputes raised by situations such as Rothwell's and to make it truly clear that an editor is only community banned after actual and fully resolved discussion on that point. So while Rothwell presents an example of the problem, he is not the motivation for the solution. Regardless of the changes being made here Rothwell's situation remains unchanged. He is indef blocked and he his not permitted to legitimately evade that block unless and until he challenges it. But lacking direct community discussion and a firm decision that he be banned, he should not be considered to be community banned. By requiring repeated requests for the indef block to be lifted and having had no uninvolved administrator step forward to provide the unblock, we insure that the issue of the ban has at least been directly discussed and de facto endorsed before it is considered resolved. --GoRight (talk) 03:10, 10 July 2009 (UTC)

Recording of Bans

As many here will be aware there are frequently disputes over whether a given editor is officially banned or not, well maybe not frequently but frequently enough. Should this policy not have something to say about explicitly recording a ban some place like WP:RESTRICT if it has ever been called into question as a means of reducing the on-going disruption that otherwise ensues? If a particular editor has been legitimately banned there should be no problem requiring that they be logged appropriately to record that fact.

I think such a policy adjustment would benefit the community. It could explicitly state that failure to initially record the ban is not justification for wikilawyering about the non-existence of the ban, but that if the existence of the ban ever comes into dispute that the proper course of action is to either record the ban and thus explicitly declare it to be valid, or not record the ban and thus declare it to be null and void. I believe that such a policy would avoid needless arguing.

Do others agree or disagree with this concept in principle? --GoRight (talk) 16:57, 7 July 2009 (UTC)

It would require a complete redo of the banning policy because there is currently two kinds of bans: ban's by discussion and block's that become ban's because noone is willing to unblock. The second kind of ban is never subject to an actual decision by consensus but only to personal considerations by each admin. As such the second kind of ban could not be "recorded" since the block does not become an actual banning untill every admin has considered whether he/she would unblock or not, and there is no way to know whether all admins have considered a block or or not. Actually I think it would be VERY helpful to make a rewrite of the banning policy to distinguish strictly between indefinite blocks (that may never be unblocked but can) and community bannings made through discussion - but I think it would probably require some kind of large community involvement to make such a rewrite.·Maunus·ƛ· 17:26, 7 July 2009 (UTC)

"As such the second kind of ban could not be "recorded" since the block does not become an actual banning until every admin has considered whether he/she would unblock or not, and there is no way to know whether all admins have considered a block or or not." - I guess that this is sort of the nugget of the point I am raising in both sections. The policy on whether someone is banned, or not, currently relies on something that is inherently unknowable (i.e. whether any uninvolved admin is willing to unblock). One could argue that if the indefinite block is challenged and no admin has unblocked that this is evidence that none are willing to do so, but before that point there is no evidence at all, correct? Yet there are those who would still consider the individual banned under this wording. At a minimum I would think some criteria should be established for how the condition "no uninvolved admin is willing to unblock" is to be assessed. Then if such assessment determines that this is the case then the ban could be explicitly declared and recorded. This does not have to be complicated, merely a single discussion at AN or ANI raised by any editor left open for sufficient time and then closed with the declaration of a ban could be sufficient. The question here is whether this policy should require such action before declaring a block to be a ban. --GoRight (talk) 17:50, 7 July 2009 (UTC)

Actually I don't think such declarations are necessary - admins already have the possibility of indef blocking users who are extremely disruptive. There is no reason to turn an indef block into a ban that I can see. I think the most healthy thing for the encyclopedia would just be to say that bans can only be made through a good solid consensus, and that indef blocks are just a completley different thing.·Maunus·ƛ· 17:54, 7 July 2009 (UTC)

Agreed. I view the two as completely separate animals with they own rules, restrictions, and implications. It would be much cleaner if a ban had to be based on explicitly demonstrated community consensus. --GoRight (talk) 18:15, 7 July 2009 (UTC)

There is no provision at WP:RESTRICT to record unilaterally-declared bans, i.e., indef blocks that, by some diffuse and unexplained process, are determined to be "community bans." And this strikes me as quite wise and appropriate. It's important because restrictions that are logged are then presumed to be enforced by the community as a whole, and administrators with no familiarity at all with the underlying situation may then block simply based on the terms of the ban, on its face, and without warning. For a single admin to be able to determine such a ban gives way too much power to one person, without a good reason. --Abd (talk) 15:51, 9 July 2009 (UTC)

It will re-writing history if we use that approach. Some of the people that are most unwelcome on Wikipedia were some one that was originally blocked by an admin and no one else would unblock them. These were considered COMMUNITY bans because they were not done by Jimbo or ArbCom. By my understanding, the idea of voting people off of Wikipedia through straw polls was not consider a good idea for many years. Only recently (past few years) has the idea of "exhaust the patience of the Community" ban grown to be common. FloNight♥♥♥ 16:49, 9 July 2009 (UTC)

Policy pages are written to reflect existing practice, not to deny past practice or revise it. I'm pretty sure that those old indef blocks that were considered bans were broadly enough known that the effect was the same as a consensus of uninvolved editors. The issue arises now because there have been some indef blocks that were never widely considered, that are, by some those aware of them, considered controversial -- the one of Jed Rothwell mentioned here was based on blocks and a ban declaration by an involved admin, as ArbComm later found. "Ban" is a far deeper statement than "indef block," it is practically perjorative, and shouldn't be used unless, indeed, the "community's patience" has been exhausted, not just that of a small set of involved editors. Yet, unless there is sufficient cause, bringing up some indef-blocked editors for a discussion may cause more disruption than it would prevent; that's the case with Jed Rothwell, who has not requested and does not desire to be unblocked, the blocked account wasn't used for three years anyway. In his case, though, any admin could decide to lift the block (I doubt that the blocking admin would protest), and it wouldn't be automatic that a new account known to be him would be blocked. It would, rather, be decided at the time, through ordinary means, which might include review of the history. The ban of Jed Rothwell was indeed taken to ArbComm for confirmation, and that was quite properly rejected as premature. There is no need for a discussion until the need arises. If Jed Rothwell doesn't care that his IP edits (always Talk page discussion of a topic where he is an internationally known expert) are reverted because, whether they are useful or not -- sometimes they are quite useful -- an editor claims he's banned, instead of merely blocked, who am I to complain?

A true community ban, decided without bias, properly, can't be lifted by an individual admin without discussion. A simple indef block, not so confirmed, can. That's the difference, that is the only practical effect of which I'm aware, other than some influence on the deletion of user pages. Lapse of time since the block shouldn't prevent the ability of an admin to reverse it, by some process of automatically considering an indef block to be a ban because no administrator objected for some period. There is no harm in leaving it as an "indef" block until and unless the discussion takes place, and such discussions should be avoided unless necessary. --Abd (talk) 22:47, 9 July 2009 (UTC)

It is important to note that the original significance in including the word "Community" in the term "Community ban" was because the indef block or ban originated from the Community and therefore could be overturned by the Community "or" ArbCom "or" Jimbo. While bans that originated from ArbCom, could not be overturned by the Community. Jimbo has stated that all of his bans or indef blocks can be reviewed by ArbCom. I don't think we can ignore the significance of including the word "Community" for this reason to every indef block that is a defacto ban.

The main area of conflict now is the best way to discuss and implement the original ban and possible unban that the Community makes. I don't think that having a rigid policy that forces all cases to be handled in an identical way will well serve the Community. My preference is to allow different approaches that make the best sense for the person involved while respecting the idea that the Community has a strong interest in giving input into bans and unbans. As someone that has participated in the discussions as a member of the Community and as an Arbitrator, I don't think ideological thinking about who should be holding the discussion matters as much as some practical considerations. For example, someone indef blocked/Community banned for harassment or making legal threats is not going to be invited to participate in the discussion on site most of the time, and therefor further Community discussion about the person is somewhat unfair. By taking the discussion to an off site venue, then the matter can be reviewed without causing additional problems for either the person or the Community. So, in this situation the Community ban may be best resolved by whoever the person or the Community contacts off wiki that has the time, knowledge, and experience to sort it out. Does that make sense? FloNight♥♥♥ 18:03, 10 July 2009 (UTC)

Well, while i can see that being able to keep some parts of a banning processes outside of the public sphere may be comfortable for the ones wielding the baning tools and in some cases for the banned editor I find it to grate against the ideal of striving for the highest degree of clarity and transparency in policies - in order to avoid abuse of power and speculation about abuse of power, drama and the iability of editors to be able to predict what kinds of behaviour is bannable and what isn't. In short I think as few parts of the process as possible should be discretionary and that those parts that are discretionary should be clearly stated to be so in the policy. You also write "I don't think we can ignore the significance of including the word "Community" for this reason to every indef block that is a defacto ban" - the problem with the current policy is that we have no way of knowing which indef blocks are defacto bans and which aren't. ·Maunus·ƛ· 18:14, 10 July 2009 (UTC)

Thanks, FloNight. I'm suggesting a certain narrowness only in ... a narrow area. There need be, for most purposes, no rigid line between indef block and community ban. But for a few purposes, a difference should be maintained. Indef blocks aren't logged at WP:RESTRICT. Neither are "defacto bans" that are really blocks that haven't been lifted yet, or other admin-declared bans without the consensus required by current policy. It is not normal to delete the user files and user pages of blocked editors, unless they are disruptive or otherwise violate policy. It is much more normal to delete them for banned editors. I'd prefer to see editors who haven't been banned by an uninvolved consensus or by ArbComm or Jimbo not referred to as "banned," but simply as "blocked," if they are blocked. Or, for an administrative article or talk page ban or topic ban, "page banned," or whatever language is accurate, but definitely not "banned by the community," unless there is that consensus of uninvolved editors described in the policy. None of this has any practical effect on those old-time and long-term disruptive editors with whom admins play whack-a-mole. We can call them banned, and the behavior of the community, in that case, shows the consensus, even though there may not have been a formal discussion and formal close.

I'm actually more concerned with how community bans are handled after they are declared. Wikitheory suggests that the closing admin becomes a kind of supervising administrator. Generally, when an admin closes a discussion with a conclusion, the admin can change the conclusion or status quo later, based on new arguments that were not present in the original discussion. If we assume that the closing admin is, in fact, following consensus, the closer becomes, as it were, the representative of the community with regard to the situation, and decides details, such as length, exact scope, etc. And when the ban is no longer needed, the admin may lift it, without a new discussion, presuming that the original discussion was reasonably thorough. ArbComm is not required to notify editors that a banned editor is being considered for unban, and my opinion is that it is generally a bad idea to do so. The comments resulting from notice will generally be biased comment, the very kind that should have been excluded from the consensus with the original ban. Presumably the evidence and arguments for the ban, such as they were, would have been presented in the original ban discussion, and if the editor has not been editing, there isn't likely to be new evidence. --Abd (talk) 21:56, 10 July 2009 (UTC)

Discussion about the banning policy

I think I want to initiate some kind of community discussion on the banning policy. Specifically I would like more input about the following points. I would appreciate if other editors would comment and supply the points of the policy that they would like to have clarified through community discussion.

• Guidelines for proposing a community ban and for determining consensus. Specifically:
  • In which forum such a proposal should be made? Is ANI good enough? Should banning discussions be restricted to being a possible outcome of an RfC where arguments and the any opposing sides present their viewpoints (as suggested by Abd).?
  • Which arguments should count most in favour of banning?
  • Possibly the need for establishing objective criteria for when a users disruption has reached a point where initiating a banning discussion is the best option. (I don't think "has exhausted the community's patience" is a sufficiently objective criteria) ***Establishing criteria for which venues of remedy should have been tried before proposing a banning.
• The degree to which a consensus for a ban should be composed of "uninvolved" editors. Specifically:
  • What it means to be involved vs. uninvolved.
  • The weight that should be given to involved editors opinions (from full to less to none).
• Guidelines for how an admin should weigh arguments vs. numbers of editors "votes" when closing the discussion.
  • The way in which arguments in favour of banning should relate to the banning policy.
• The distinction between community ban and an indefinite block. Specifically:
  • Whether an indef block does in fact become a ban at any point (and if yes how to determine that point) - or whether the two are distinct.
• Should there be a "record" of which users are banned so that a banning is made official only by being adduced to that record? (suggested by GoRight)

·Maunus·ƛ· 21:11, 7 July 2009 (UTC)

Lots of questions, and I have something to say about all of them. I'm going to put my entire response in collapse, and I'll ask that any one who wants to discuss any aspect of it bring what is necessary out of collapse, I give permission to refactor, provided the paragraphs are kept intact. I'll follow with a summary of my basic conclusion.

Full discussion

Absolutely, a change to banning policy should not be proposed at AN/I. Nor should it be proposed, in my opinion, on any noticeboard, though a reference to it on AN would be in order, likewise, the Village Pump. My view is that bans are best determined after there is an RfC, even if actually determined on a noticeboard, because then there is evidence to refer to efficiently, evidence that has been thoroughly vetted and discussed, ideally. This kind of work doesn't happen on noticeboards, using a threaded format; it's only with RfCs that we start to get better and more cautious process. First of all, there are bans and there are indef blocks. Many indef blocks may be easily justified, and they are also easily undone. There is no need for a major process with most indef blocks. An admin makes a decision, with or without a prior discussion. Very simple, and this simplicity is a crucial aspect of Wikipedia process. But a ban is a different animal; a true community ban, decided by a consensus of uninvolved editors -- and I'll add that they should also be informed, which is why, you'll note, that ArbComm doesn't start to vote until long after the evidence has been presented and arguments and proposals debated by the community -- cannot be easily reversed; that's why it's important that it "enjoy" true consensus, and even a mere majority, if there exists a significant number of uninvolved editors who oppose the ban, probably should not be enough, no matter what the "cogency of the arguments" is as judged by the closing admin. As described in the policy, a community ban means that the patience of the community has been exhausted, and the existence of substantial numbers of editors who might be willing to support, for example, monitoring the editor to make sure that disruption does not continue, means that the community's patience has not been exhausted. But that doesn't mean that the editor should not be blocked; that decision is made based on the immediate situation, whereas a ban is long-term. There is no emergency to declaring a ban, when the editor is blocked, and even discussing a ban can be more disruptive than simply letting the matter of an indef block lie where it falls, until there is some necessity and someone wants to bring it up. In our system, the exact standards used are not specified. If a decision is reviewed, the original closing admin may become more or less irrelevant. Good ones often stand aside and don't even participate; unless their decision was blatantly improper, i.e., they were, say, involved, or the ban declaration or block was disruptive and they should have known that, they are not on trial and being reversed in a decision is no shame. (Where they argue tenaciously that their decision was correct, I personally begin to suspect that they had an axe to grind.) It is within the discretion of the closing admin how to weight arguments. In my analysis of the case we've discussed, though, I followed a simple standard, which was easy to apply (and it therefore, only took a few hours of research!): if there had been what looked like revert warring or contentious editing, or other dispute, between the editor whose ban was proposed, and the editor !voting, I set that aside. Further, I looked additionally at the time sequence of !votes. Later !votes are more important than earlier ones. Why? Because, presumably, editors commenting later, on average, have more knowledge of the situation, having read the previous discussion. Further, editors who already have their minds made up, because of previous conflict, are more likely to vote quickly and are more motivated to !vote. They don't need to do no stinkin' research! If it were up to me, we would change RfC procedure, and we would allow bans to be decided by RfC. However, the current early !voting that happens in an RfC would not be allowed. There would be a period for the collection and presentation of evidence and arguments, with no support or oppose votes allowed. Then voting would be opened where endorsements of editor positions is allowed. Still, except for the delayed supports and opposes, this is quite like a present user RfC. The one difference -- and what a concept! -- no !voting until the evidence has been documented. Then, when that is closed, there is no more need for discussion of the underlying situation, a "remedies" section would be opened and remedies proposed and argued. Again, no !voting at first. Finally, !voting would be opened. I read Alice in Wonderland and Through the Looking Glass to my seven-year-old daughter every night that I put her to bed. Remember "Verdict first, trial later"? That's what we do with a ban proposal, too often: propose the ban, the remedy, and vote on it, before having established clearly what actually happened and the boundaries of what's possible. Frankly, all this stuff was worked out over centuries, I'm just applying standard deliberative process, but we imagined that Wikipedia didn't need this kind of "red tape," we thought we could bypass it. It worked when the community was relatively small, and it increasingly has broken down as those who formed that initial core have burned out, overwhelmed by the defects in what they set up. It was very good, but not as good as it needed to be to meet the demands of scale. There need be no specific standard, rather, determining the "consensus of uninvolved editors" is up to the closing admin. It is a separate determination from the remedy itself. In the case we were looking at, the problem was that the closing admin declared a "consensus," with astonishing four-place accuracy, and clearly did not consider involvement at all. Hence my conclusion that the admin didn't establish, by the close, a "community ban," but only an administrative ban. It was not merely a technical error, for, when I did analyze the !votes, I found that a small majority of the clearly uninvolved editors -- at least the ones for whom I didn't find an involvement -- had opposed the ban. When I look at the vote timing, the later votes were, by a good majority, opposed to the ban. That does not mean that the indef block was improper! But it does mean, to me, that a community ban wasn't established, should the difference become important. So how would we tell the difference? Well, first of all, the closing admin should declare the "consensus of uninvolved editors." There need not be one, an admin can decide that the welfare of the community requires that an editor can be blocked, without such a consensus. If the closing admin makes that declaration, we expect that it is either justified or can be justified by some kind of reasonable analysis, whether it is done specifically or not. I'd say, though, if the administrator does the analysis, which should be required for a ban, it should be stated, documented, which takes little more time than actually doing it. The accused editor, as part of the process, could be asked to document possible involvement for review, or this could be done by anyone else. Like all such single-admin decisions, it should be reversible upon review. A close can be reversed; the first reversal may not even be considered wheel-warring, the policy is a little unclear on that.... One of the advantages of making ban decisions with some kind of RfC process, with a separate page for it, is that a review could be handled on the same page, as an addendum. There would therefore be no need to re-establish all the prior evidence and comment; what would be done is simply to review the original close. Really, I'd say, if we were smart, we'd do DRV after AfD this way, instead of with a separate page in a separate place.... FloNight, above, comments on the old reluctance to submit ban decisions to polls. I agree. It's dangerous. We should reaffirm the existing system, and make it clear that a "community ban" is decided by a single administrator based on two findings of fact: first, that the welfare of the project is served by the ban, and, second, that it has been endorsed by a consensus (and I'd make that a strong consensus) of uninvolved editors. Note that for most purposes, the second finding is not actually necessary, and it's quite possible that it could be made later, if the difference matters, or at leisure. For example, a ban discussion could close with an indef block being implemented, or other remedy (I've been favoring "voluntary" site bans, where the editor only edits precisely as permitted, but still has control over their own user space, and may be able to negotiate a mentor), and there could be a scheduled opening of a review in a month, say, where a determination would be made, either based on the prior votes being those of uninvolved editors -- and evidence could be considered at that time -- or on new additional discussion if needed. In fact, with many bans, it might not be needed, for various reasons. Banned editors often simply go away, it's a tad unfriendly to ban somebody, and many people realize that they get more joy out of participating in activities where they are welcome.

There need be no specific standard, rather, determining the "consensus of uninvolved editors," required for a community ban, is up to the closing admin. It is a separate determination from a block decision, which is made based on preponderance of arguments, consensus isn't the issue with the block decision, and by making the declaration of consensus, if that is done, the closing admin is certifying that involvement was considered and that consensus of uninvolved editors was adequate for such an extreme remedy.

--Abd (talk) 01:36, 10 July 2009 (UTC)

I agree that the existing rules for determining when a community ban is placed and how it can potentially be lifted are at least somewhat troublesome. I cannot see, for instance, why a community ban must necessarily go to the Arbitration Committee to be lifted. That would seem to give, somehow, the ArbCom a form of "veto power" over the community, and I'm not sure that is something that many/most of even the frankly very limited number of people who vote in Arbitration elections consider. I'm not sure how to fix it, but I do think that there is a need to change, somehow, the means by which a community ban is determined and the ways of reversing it. John Carter (talk) 14:33, 13 July 2009 (UTC)

A community ban can be lifted in one of three ways: 1/ appeal to the community resulting in a consensus to unban, 2/ an appeal to ArbCom if the first process does not yield a consensus, or 3/ theoretically, an appeal to User:Jimbo Wales, that is almost certain to be referred back to ArbCom. I think it is a mistake for ArbCom to step in and process a bona fide unban request without first seeing if the community can sort it out. Jehochman ^Talk 15:43, 13 July 2009 (UTC)

Not disagreeing with you, but the Banning policy page only indicates the appeal to ArbCom option. If, as you I assume correctly say, there are other options, why aren't they included in the accompanying page? John Carter (talk) 15:54, 13 July 2009 (UTC)

See [12], that type of community ban can be lifted by an uninvolved admin, and the others can't? --Enric Naval (talk) 02:56, 16 July 2009 (UTC)

Its a quality inherent in the wording of the policy it is just made clear. The policy says that an indef blocked user whom no administrator will unblock is de facto community banned - it logically follows that any administrator can unblock and in this way reverse the effect of the de facto ban. ·Maunus·ƛ· 13:39, 16 July 2009 (UTC)

The wording Enric refers to could be improved. An uninvolved administrator "can" reverse a community ban, i.e., one founded upon a consensus of uninvolved editors, but is likely to face an uproar; the same is true, in fact, for a Jimbo reversal. Administrators can do anything the tools allow, under IAR, but the flip side of IAR is accountability. This is the point: there are bans, alleged to be such, where there was no discussion and consensus of uninvolved editors, but these are, properly, administrative bans, not community bans. I've been on the side considering that administrators may unilaterally issue specific bans, but when I examine the implications in detail, it's a can of worms. It's probably better to call them, simply, "warnings" instead of "bans." "Your edits to the project have been disruptive; in lieu of blocking you, I am allowing you to continue to edit for the purpose of finding a mentor. If you continue to edit as you have, see (diff) and (diff), I will block you." This is effectively an administrative site ban, with specified exception, and an unstated exception: the editor may continue to edit, if it is not disruptive. --Abd (talk) 14:19, 16 July 2009 (UTC)

Sorry, but after exhausting the community's patience, and no admin being willing to unblock, an admin shouldn't just go and unblock without giving a good reason (like a mentorship, or good work in other wikis). And he should really make sure that he addresses the problems raised by the community. --Enric Naval (talk) 15:07, 16 July 2009 (UTC)

Enric, your comment is self-contradictory in part and obvious in the rest. If an admin is willing to unblock, the "patience of the community" has not been exhausted, and the condition "no admin is willing to unblock" is demonstrated not to exist. Of course an unblocking admin should consider the welfare of the community and the project, as should any blocking admin, and all of us at all times. Nowhere has anyone suggested that the "problems raised" by involved editors should be disregarded; on the contrary, they should be addressed, with appropriate measures to prevent unnecessary disruption. --Abd (talk) 15:13, 16 July 2009 (UTC)

Past after the community's patience is exhausted, and long time after it's clear that no admin informed of the situation is going to unblock the person, an uninvolved admin could appear there (for whatever the reason, like the user requesting unblock by private email or via the admin's talk page) and unblock without first informing himself of why that person was banned in the first place and without asking him not to repeat his behaviour. Fun ensues as the editor inmediately restarts editing with the same behaviour that had previously exhausted the community patience. --Enric Naval (talk) 16:52, 16 July 2009 (UTC)

I think this shows that there is every good reason to start a wider community discussion about what we actually want the banning policy to be. If thee can be so much confusion and differing opinion about what the policy actually says it simply cannot be a well written policy. However I also see that the policy is controversial and discussing it without passion is difficult for too many editors. I have spoken my opinion of the current policy and will leave the discussion for now. ·Maunus·ƛ· 17:00, 16 July 2009 (UTC)

I have not weighed in much because I believe that you have been doing a perfectly fine job of representing the issues and so my commentary would only have been redundant. I agree with you that the core issue here, exactly WHAT should the banning policy be over-all and in this instance specifically, deserves a much wider audience than it is receiving here. Where would be the appropriate place to raise such a discussion? The village pump? Others? We could simply point people here to drawn in a wider caste of opinions. Thoughts? --GoRight (talk) 18:11, 16 July 2009 (UTC)

@Maunus. The wording is good because it agrees with current practice. It allows the community to ban disruptive users without having to go throught bureaucratic steps of opening a formal discussion, etc, even in cases where the community patience is clearly exhausted. Also, anyone disagreeing with such a ban can simply open a review at a noticeboard. --Enric Naval (talk) 00:48, 17 July 2009 (UTC)

The wording is bad because it is fundamentally misleading. Your point would be more properly worded as "It allows administrators to ban disruptive users without having to demonstrate an actual community consensus upholding those bans." If you want to give administrators this level of power, and I don't, then at least don't try to call it a community action because it is not. It is an administrative action and it should be termed as such. --GoRight (talk) 06:08, 17 July 2009 (UTC)

This is sort of a random thought, I hope it is constructive. It is my sense that the need for this discussion comes out of the fact that the idea of a community ban was developed long before many of our policies and guidelines. I think there was a time (yes, even before ArbCom) when there were situations where an editor was so disruptive at so many articles that after discussion on the listserve and at article spaces it became clear that many people wanted the person banned, and no other editor could reasonably defend him/her any more. At that point it was just a question of an admin (who had the technical means) using his or her own judgment that there was a general community consensus, to block the user indefinitely. This, in my mind, is still how community bans should work: a random (i.e. univolved admin) notices that a vast amount of talk space centers on discussion about how and why x is not doing anything to build the encyclopedia and on the contrary is doing a great deal to disrupt it, and the uninvolved admin blocks the person "per discussion at x, y, and z." Believe it or not this kind of ban was possible in rather uncontroversial ways back in 2002 or even 2003. Two things have changed since then. First, the community now numbers in the millions. Second, we have much clearer policies and guidelines.

I think the "community ban" has morphed into something a litle different: an admin using his or her own judgment to enforce a particular policy or guideline through a block. We call this a "community ban" when there has been a fair amount of discussion, usually at AN/I, that the person in question has violated policy or guideline x and something needs to be done about it. I'd rather call this an "administrative ban" thn a "community ban." I think in the former case, one administrator takes the initiative of blockin someone after there has been some consensus reached among other administrators at AN/I. In the latter case, it really is a bunch of interested editors (meaning, editors, working on the same articles as editor x) who really really really want to see editor x banned, and an administrator who reviews the talk, concludes that all the other editors are being reasonable, and is acting as an instrument of their will. We could still call this a community ban, even if the editors who want to see x banned are just a fraction of the community, if they nevertheless constitute the active editors at the articles x has ben editing. But for me the key thing about a "community ban" is that an admin is simply acting on the behalf of a segment of the community.

I do think that x should always have an opportunity to appeal, the question is "how." For one thing I think it is important that whoever enacts the block puts a link to the discussion leading up to the ban on x's talk page. There is a problem when AN/I discussion gets archived, the link breaks. How do we deal with this? Should the blocking admin cut and paste discussion to x's talk page? Future admins need to be aware of the discussion that led to the block. Slrubenstein | Talk 10:34, 17 July 2009 (UTC)

Exactly.

P.D.: About broken ANI links, I posted here. --Enric Naval (talk) 16:33, 17 July 2009 (UTC)