Wikipedia talk:Community sanction

From Wikipedia, the free encyclopedia

Per discussion elsewhere, I think this power already exists; this is simply documenting what the administrators' corps could do--and should do in many instances.

This, and WP:TE, could make the arbcom's life much easier. --EngineerScotty 23:40, 20 September 2006 (UTC)

Procedural inquiries: will all such probations needed to be announced and confirmed at WP:AN/WP:ANI? If so those already contentious boards will possibly get worse. From the current wording I don't get that impression. On the other hand, a user may not be aware when told that they are probation that it can be appealed to WP:ANI (so we would need to make it clear to admins that they should point people to this page when they invoke such probation). A related concern is that this will simply add one more level before things get appealed to ArbCom. People will get probation, they will go to ANI, it will get confirmed and then they'll go to ArbCom. Thus this should only be implemented if we can get a decent guarantee from the ArbCom that they will generally respect community approved probations- the ArbCom currently (even by the observations of one arbitrator) has a history of looking at community bans and then confirmed them after an annoying and time consuming Arbitration. JoshuaZ 03:21, 21 September 2006 (UTC)


Contents

[edit] Current wording matches recent community probation

The current wording of the policy matches the way that I implemented the recent community probation, an one month topic article ban. I suggest we keep this wording until we do a few more and then change it to include any new ideas. FloNight 04:30, 21 September 2006 (UTC)

[edit] RfC

I've opened an RfC for this in order to get more input. The proposal looks good to me right now. Durova 05:55, 27 September 2006 (UTC)

[edit] Community sanctions

One idea discussed on Wikipedia talk:Community probation is to expand this to include the existent, but not-well-documented practice, of community bans. --EngineerScotty 21:07, 28 September 2006 (UTC)

[edit] Changed the name and made policy

The community is now doing community sanctions so a written policy would be helpful. I changed the name per suggestions on the log page. Comments and suggestions welcome.

One area to be addressed is using user conduct RFCs as a starting point for community sanctions. I do not think it should be mandatory but could be a way to make RFC actionable. Short of starting a RFArb little actionable happened in the past even if there was a need. There is a natural link between the two, I think. One RFC is being used that way now and I anticipate more in the future. [1]. FloNight 16:31, 5 November 2006 (UTC)

[edit] Listing of sanction types

I added a list of types of sanctions, thinking that not all admins will be familiar with typical Arbcom remedies. Should be pretty noncontroversial. Personal attack parole is a bit redundant with the kind of thing that happens on the noticeboards already but there might be some times when we would want to specify it. Thatcher131 19:25, 9 November 2006 (UTC)

[edit] No thanks on the merge

I'm going to remove the merge tag for the time being. Community sanction is something we have already agreed to do and have started doing. RFSL is a proposed new procedure in between RFC and RFAR. If RFSL is adopted, it will certainly interact with Community sanction in some way, but it is much too early to consider merging a description of what we already do with an entirely new proposed process. Thatcher131 04:46, 10 November 2006 (UTC)

[edit] Not policy yet

This cannot reasonably be taken as policy - there are far too many major kinks to be worked out, and there exist arbitrators who are actively condemning the policy on the mailing list. It is at best proposed right now, and at worst a fundamental upheaval of the entire way in which we treat sanction and probation on Wikipedia. Phil Sandifer 17:41, 17 November 2006 (UTC)

It would be nice if they would say so in person. Thatcher131 18:28, 17 November 2006 (UTC)
Just starting to read about it. My first impression is positive. Fred Bauder 18:40, 17 November 2006 (UTC)
I support this in line with Charles Matthews recent en-wiki mailing list comment that proactive admins should be the first, second, and third line of defense of the project. Admittedly most of the discussion has taken place at the admin noticeboards, and the regulars there generally support it. Perhaps a more focused discussion is needed here to get it on record. Thatcher131 18:49, 17 November 2006 (UTC)
I think a rough consensus backed by possibility of appeal to Arbcom would be appropriate. Might be hard to overrule a bunch of administrators, but shouldn't unless they got something wrong. Fred Bauder 19:15, 17 November 2006 (UTC)
Yes, but your colleague on the arbcom, Sam Korn, quite strongly disagrees. And remember, community bans are not rough consensus - they're "1000 admins are unwilling to overturn this." Community sanction is quite different- there's not the same weight of "1000 admins decline to oppose this." They're based much more on threats than on a clear consensus. That is, to my mind, quite a problem. Phil Sandifer 19:38, 17 November 2006 (UTC)
Just a point about the 1000 admins. That was added to the blocking policy at the beginning of October. I don't think it was ever there before that, not as 1,000 or any other number. The policy required before, and now, "strong consensus," because we can't give a single admin a veto. SlimVirgin (talk) 22:26, 17 November 2006 (UTC)
There's more to policy than what's written down in myriad WP: pages. The "1 in however-many-admins-we-have" is the de facto policy and has been the whole time I've been around. If one admin -- sane, reasonable individuals -- is convinced enough that you don't deserve to be summarily blocked, you deserve a full hearing. If the admin has appallingly bad judgment, they shouldn't be an admin. Full stop. Sam Korn (smoddy) 17:57, 19 November 2006 (UTC)
Sam, for as long as I've been here, it hasn't been policy, de facto or otherwise. It would be madness to allow one admin a veto, particularly when we have at least one who's prepared to copy deleted material to Wikitruth and Wikipedia Review. I agree that such a person shouldn't be an admin, but at least one of them is. So the one-in-however-many-hundred can't work anymore, if it ever did. SlimVirgin (talk) 18:26, 19 November 2006 (UTC)
Arguing about de facto policy is, I'm pretty sure, going to get us precisely nowhere. You think it isn't; I think it is. I recall it being phrased "if no-one's willing to unblock you, that's a community ban" (approximately) when I was a new editor, which is two years ago and more now. It's never (until recently) been a case of finding consensus or unanimity: it's always been implementing and seeing what happens. I think this is a far better way of implementing the process. It is very unlikely that one admin will disagree with a ban. If it's several users expressing doubt, there should be an ArbCom case. If there is just one, we'd have to see what the most appropriate actions to take would be. It's not a clear-cut thing; we don't have to legislate for it. We need to use common sense. And we are capable of using common sense. Cynicism is not permitted. Sam Korn (smoddy) 18:41, 19 November 2006 (UTC)
Most admins are capable of using common sense, are honest, decent, and care about Wikipedia. But a very small number aren't, and it only takes a very small number to screw up any informal process that relies on common sense and integrity. SlimVirgin (talk) 19:41, 19 November 2006 (UTC)
So the bad admins are flushed out. Excellent. I suggest writing policy based mainly on the good admins, rather than indulging those who are poor. Then desysop those. Sam Korn (smoddy) 19:59, 19 November 2006 (UTC)
You can't desysop someone for disagreeing. You have to show the corruption, but evidence is difficult to find, and even when we have clear evidence of dodginess, the ArbCom isn't necessarily interested. You're presenting too simplistic a picture, Sam. We have bad admins, and I don't mean bad as in not great adminstrators, I mean bad as in bad, and it's very hard to do anything about it. SlimVirgin (talk) 20:35, 19 November 2006 (UTC)
I don't think I'm being simplistic. Perhaps I'm being idealistic, but I don't think that's a wholly bad thing. I am quite happy to concede the existence of some bad-as-in-evil admins (as opposed to bad-as-in-poor). I am, however, reluctant to concede that these poor admins are going to have a significant impact in this problem. I can't prove this, I'll admit, but I'd like to see some evidence first. I would very much like to see a one-against-all incident. I doubt you'll find one. Sam Korn (smoddy) 20:42, 19 November 2006 (UTC)
A recent example was the community ban of Homeontherange, supported by, when I last looked, at least 25 admins and a number of non-admins, but when he was blocked, his friend in real life unblocked him. The one that stands out was the ban on Marsden, one of the most toxic editors we had, and a small number kept unblocking him. In the end Jimbo had to do it. You say common sense is important; common sense rules out giving any random admin — who may know the user personally, may enjoy causing trouble, or may be involved with Wikipedia Review — a veto. SlimVirgin (talk) 21:20, 19 November 2006 (UTC)
The Homeontherange unblock was clearly unjustified and no admin should take such action in such circumstances. That is not changed by this. Interestingly, with Marsden, Phil Sandifer both blocked and unblocked. I don't know exactly what that says, but it does suggest to me that the case wasn't as clear-cut as you suggest. Are you going to tell me that Talrias and Phil Sandifer are bad-as-in-evil admins?
You keep talking about a veto. This is entirely spurious. I do not mention vetoes. You're putting words into my mouth. What I say is that almost always there will not be just one user opposing an indefinite ban, and if there is that user's reasons need to be looked at. I don't see what Wikipedia Review has to do with this. If an admin is a bad egg, these circumstances will make it blindingly obvious. I'm not arguing for a veto, just for a very conservative attitude towards placing indefinite bans. Sam Korn (smoddy) 21:34, 19 November 2006 (UTC)
When Phil unblocked M, my memory is he was only trying to be responsive to the concerns. SlimVirgin (talk) 21:41, 19 November 2006 (UTC)
Hopefully that there is only one example of this problem indicates that it is not a serious one. I don't think we need to legislate for every instance. Sam Korn (smoddy) 21:51, 19 November 2006 (UTC)
I didn't say there was only one example. I said I could only think of that one offhand. I remember that a previous proposed community ban of Homeontherange was opposed by KimvdLinde, an admin and member of Wikipedia Review, who said it wasn't valid because she opposed. Then there was the Saladin1970 thing. I blocked him indefinitely for disruption, anti-Semitism, reverting, terrible editing, the works, and someone unblocked him, so the case had to go to the ArbCom. I had to spend hours searching for diffs and links to anti-Semitic comments of his on other websites to present the case. In the end, the admin who unblocked him withdrew his support and Saladin got fed up and disappeared anyway, because what he wanted was to be a nuisance, and the slowness of the ArbCom case meant he wasn't seeing enough action. So the whole thing was a waste of time. Sam, you talk about common sense and trusting admins. Fine, but then trust them to use their common sense and maturity. A community ban is not a question of numbers, whether you favor the 1-in-1000 thing or strong consensus. Actually, both are wrong. What happens in practise is that it very much depends who is supporting and objecting, not how many. One highly respected editor turning up with an objection and a good argument might be enough to thwart any ban, whereas three or four friends of the user may make no difference. These things rely more on intuitions than hard numbers. SlimVirgin (talk) 22:08, 19 November 2006 (UTC)
Actually, I don't really disagree with you. I absolutely think there should be some discretion for admins, and, as you rightly point out, this is centred on admin discretion. "Who is supporting and objecting" is indeed the principle that should matter. The 1-in-1000 point relies on admins being reasonable, which I think is true on the whole. Maybe there are one or two isolated incidents where they have not been, but they are definitely the minority. If the admins are reasonable, they will not block a user when there is opposition of any reasonable sort. This should be the criterion, I suppose, "opposition of any reasonable sort", and I expect that, in practice, you and I would agree pretty much absolutely. I rather fancy what we are debating is actually semantics. The main difference between us, I think, is my premise that admins will generally do what they should do. Sam Korn (smoddy) 22:50, 19 November 2006 (UTC)
I agree with the "opposition of any reasonable sort" criterion, and that the overwhelming majority of admins are decent and sensible. So it does mostly boil down to semantics. I think it's just that the very small number of times I've seen one admin try to impose a veto, and not for good reason, has made me nervous about formalizing the 1-in-1000 thing. SlimVirgin (talk) 00:20, 20 November 2006 (UTC)
I think SlimVirgin's position may be unduly influenced by her ongoing dispute with another editor. By way of clarification, I should point out that (i) there was no recognized community ban against HotR when I unblocked his account, (ii) the blocking admin did not claim there was a community ban, (ii) the blocking admin identified the account as a sockpuppet, when it had been recognized by all parties as an alternate account, (iv) I am not HotR's close personal friend in real life, (v) the manner by which a community ban was "determined" was anything but free and fair, and (vi) SlimVirgin's account of this situation is extremely selective and highly self-serving.
I'm not going to get dragged into this again, except to say the above account is false: CJCurrie is a real-life friend of Homey's; Homey had over 20 sockpuppets, not alternate accounts; my "extremely selective and highly self-serving account" was supported by over 25 admins and several other editors; CJCurrie should not have unblocked in the face of that strong consensus. If you want the 1-in-1000 admins principle to stand, the best course of action is not to discredit it with that kind of intervention. SlimVirgin (talk) 21:53, 21 November 2006 (UTC)
If you don't want to "get dragged into this again", then you shouldn't misrepresent the situation to advance your argument. There was no "strong consensus" for a community ban when I unblocked HotR, and I doubt such consensus ever really existed.
Despite what you seem to believe, I take COI seriously. If the admin who blocked HotR had argued that he was imposing a community ban, I would not have reversed his decision -- regardless of how much I disagreed with it. But this isn't what happened.
The blocking admin wrote that an alternate account then in use by HotR was actually a sockpuppet. This rationale was demonstrably wrong. The alternate account was openly acknowledged by HotR, and had been recognized as such by all concerned parties, including you. The block was therefore absurd. COI isn't a factor in such situations.
As to the rest ... the "over 20 sockpuppets" assertion is flat-out wrong, and virtually every admin who endorsed the community ban either (i) was a veteran of past content disputes with HotR or (ii) represented an opposing POV in the same matters. The resulting spectacle was not "strong consensus" so much as "mob rule". I'm not going to discuss private matters, except to say that your assessment is incorrect.
To return to the present, I've proposed a compromise option on the "1 in 1000" rule: if a specific number of admins oppose a community ban or sanction, then it should be overturned. This prevents a single rogue administrator from standing in the way of consensus, and it also ensures a fair appeal's process. What do you think? CJCurrie 02:29, 22 November 2006 (UTC)
But that's in the past. Concerning the issue at hand, I should note that Wikipedia:Banning policy upheld the principle of unanimity for community bans until fairly recently [2]. CJCurrie 03:04, 20 November 2006 (UTC)
Now, here's a constructive suggestion: our policy should allow a community ban to be overturned if a specific number of administrators are willing to allow the editor to return. This removes the danger of a "single rogue admin" standing in the way of unanimity, while also permiting a credible appeals process for dubious or questionable bans. CJCurrie 03:12, 20 November 2006 (UTC) And the same policy could be put in place for community sanctions. CJCurrie 03:21, 20 November 2006 (UTC)
As I say, I am against formalising anything. So I don't think we disagree here. Sam Korn (smoddy) 18:25, 20 November 2006 (UTC)
If they could do an article ban or impose probation it would not be "threats" but a remedy. Fred Bauder 19:40, 17 November 2006 (UTC)
(EC) Enforcement still requires strong admin consensus. One admin proposes banning a difficult editor from a particular article, with other admins agreeing, and the threat of a block if the editor ignores the ban. Supposing the editor violates the article ban and the proposing admin blocks him. It still only takes one other admin to unblock. at which point the community article ban collapses. When I added detailed descriptions of the types of sanctions, I was careful to suggest limiting the duration. Perhaps I should have been more explicit. While arbcom can hand down one-year or lifetime topical bans and indefinite blocks, I don't believe any community sanction (article ban, civility parole) should last more than a month, nor any blocks for violations be longer than a week. If it needs to be longer, it should probably be reviewed by arbcom. And there is a provision for arbcom appeal. I also think it is useful to give the community a lesser option than a full ban. As community banning gains wider acceptance, we ought to be able to consider more conservative, less final remedies like 1RR parole and topical bans. Thatcher131 19:51, 17 November 2006 (UTC)
How many cases do you think won't be appealed? I expect the vast, vast majority to be appealed if that's explicitly an option, which is why I prefer the consent-based (and already possible) system I outline below. Sam Korn (smoddy) 18:46, 19 November 2006 (UTC)

(<---)I see no reason 10 or 20 administrators familiar with the problem shouldn't be a "consensus" for this purpose. Fred Bauder 19:43, 17 November 2006 (UTC)

Arbcom doesn't scale well to our current size. This is a useful addition. Further a consensus of noninvolved editors can already per existing guideline call for a block and I've seen such a block's removal be contingent based on staying away from a specific article. This is needed. Its just a matter of doing it right. I recommend two things. One : uninvolved editors make judgements. Two : use repected editors or some such terminology rather than "admins". The two are not the same thing. 4.250.198.106 19:49, 17 November 2006 (UTC) (WAS 4.250, a non admin)
I would expect editors to offer input. Fred Bauder 22:21, 17 November 2006 (UTC)
I want to explain my opposition to this process. I have no problems with the principle, which is that people below ArbCom have permission to form reasonable and focussed sanctions. My problem is with some kind of documented process and requirements for consensus. That is unneeded. It should be a private agreement (i.e. one to one, not in camera) between an admin and a problem user. Perhaps another admin can come and agree with the terms of the agreement. I see this working like this:
Hi. You're being disruptive in articles about Scientology. I think you're being disruptive enough to warrant an ArbCom case, and I'm pretty sure you'd get heavy sanctions. However, I don't want to waste my time with a case if I can work with you to sort out an agreement together. Would you be amenable to this? Admin A 17:57, 19 November 2006 (UTC)
I want to back up what Admin A says. I really think you should think about submitting voluntarily to some restrictions. Admin B 17:57, 19 November 2006 (UTC)
OK, I hear what you're saying. Could you suggest some conditions? I think 1RR on Scientology subjects, NPA parole and an undertaking to discuss controversial changes on the talk page would be reasonable. Problem User 17:57, 19 November 2006 (UTC)
I agree mostly, but I think there should be some kind of penalty attached as well. Say a one-week block for each contravention, and on the fifth block I'll take it to ArbCom. Have you any complaints? Admin A 17:57, 19 November 2006 (UTC)
OK, I can live with these terms. Problem User 17:57, 19 November 2006 (UTC)
If the problem user can't work like this, I suggest that a kangaroo court of admins will not help. If a problem user can't work out a problem like this reasonably, the situation is not going to be helped by an ad-hoc committee. There needs to be some kind of established jurisdiction to make the restrictions credible and, more to the point, workable. Sam Korn (smoddy) 17:57, 19 November 2006 (UTC)
Important point that you are missing, Sam. This is working :-) This policy was written after we began logging cases. So far none of the cases have been appealed. The discussions were all above average for civility. No sanctions have been enforced on any individuals the last time I checked. I think you concerns are overblown based on our experience using community sanctions. --FloNight 19:01, 19 November 2006 (UTC)
Links? (Beyond the "log", that gives no context at all because the links are expired.) Sam Korn (smoddy) 19:04, 19 November 2006 (UTC)
I've also seen this work well. I've only seen it used for editors who were being a horrible nuisance and who in reality should have been blocked. Instead, they were offered voluntary page bans, they accepted, and it worked. Surely it's better to have admins do this than have every case go to the ArbCom with the enormous time and energy drain that involves, not to mention the broken relationships. SlimVirgin (talk) 19:41, 19 November 2006 (UTC)
Um, that is precisely my point -- it should be done with the consent of the user involved and in no way needs formalising. Sam Korn (smoddy) 19:59, 19 November 2006 (UTC)
The main difference between a consent and coercive system is that in a consent system, if the editor refuses to accept a voluntary article ban, it goes to arbitration, and the editor can continue to be disruptive for up to 2 months before a final decision. It requires an affirmative act of the committee to ban him during the case. In the coercive system currently described, an editor who refuses to accept a voluntary article ban may be blocked. The case may still go to arbitration, but it would take an affirmative act by the committee to lift the coercively enforced article ban. Depending on how disruptive the editor is, a lot of good people could get driven off in those two months. I'm not married to the idea of coercive community-imposed probation, but if we're going to drop it in favor of a voluntary model I'd like to see Arbcom much more willing to hand out temporary injunctions. After all, it is the editors who are least likely to accept voluntary curbs on their behavior who are most likely to end up in arbitration. Thatcher131 00:31, 20 November 2006 (UTC)
I don't think we want any "models" at all. We want to see good admins using their discretion. I also fully agree that there needs to be more temporary injunctions by the ArbCom. This is a matter that needs to be addressed and I shall look into how this can be achieved. I also think there needs to be a comprehensive study of the DR process so we can work out what we can best do to improve it. I may look at doing that some time in December. Sam Korn (smoddy) 18:25, 20 November 2006 (UTC)
In fact, I would go so far as to suggest that it can't be formalized and still work. I think the idea works precisely because it's a novel, creative, out-of-process solution. Add the crust of process to it and the idea will find itself open to gaming and rules-lawyering, and will collapse under its own weight. Not every good idea needs a page describing it. Phil Sandifer 20:02, 19 November 2006 (UTC)

[edit] Random section break

If the objection is against formalizing, I should point out that the original intent of this page was only logging. E.g. if it is agreed for whatever reason that user:X should not edit regarding to subject:Y for a while, this is a central spot to write it down, since the usual place for such discussions (WP:ANI) gets archived several times per week. (Radiant) 13:08, 21 November 2006 (UTC)
This page is the product of the formalised process. Why should there be any logging? There only needs to be private agreements between users. If you are going to log it, it makes it a public matter. The choice is between saying nothing and letting admins get on with things themselves to the best of their abilities and making a formal process. What's more, if you read this page, you'll see that it in detail outlines exactly how the process would work. It isn't just about logging. Sam Korn (smoddy) 20:12, 21 November 2006 (UTC)
Perhaps that could be a compromise then. Create a page that is for logging only, with no policy or guideline-type language. SlimVirgin (talk) 21:55, 21 November 2006 (UTC)
Sam - there needn't be logging in all cases, but in some cases it is useful for other people to know what is going on (e.g. someone new to a conversation might be unaware of the history behind it and behind certain behavior). If this page is too complex (as it seems to be) it should be simplified. The original intent was to simply write down here if/when ANI (which is already public) decides it would be best if user:X stayed away from page:Y for awhile. (Radiant) 10:10, 22 November 2006 (UTC)
Why should ANI decide this? My whole contention is that it should not be imposed on a user, whether on ANI or elsewhere, and this page exists to log imposed sanctions. Agreed-upon sanctions obviously don't need logging. This page is only useful if there are to be imposed sanctions. If there will not be imposed sanctions, this page is superfluous. If it can be demonstrated that imposed sanctions will not further antagonise problem users, I am quite happy to see the process.
The process you describe -- decision made on ANI, logging here -- is exactly the same as before. And, Sarah, while I absolutely thank you for your attempt to compromise, do you really think that a page like this can stay clear of policy-like language? People will come, and people will make it overly bureaucratic. Sam Korn (smoddy) 12:26, 22 November 2006 (UTC)
No, I don't think people will make everything overly bureaucratic. Some people will try, and others will counter the effort. Happens everywhere on the wiki. Basically what you're saying is that community sanctions shouldn't happen. However, for better or worse, community bans do happen about once per week now. A sanction is merely a weaker form of a ban, specifically a ban limited to a small set of articles, and this would serve to reduce the number of actual bans. (Radiant) 12:56, 22 November 2006 (UTC)
No, the sanctions are not just weaker forms of bans. They are set conditions that will be tried by an ad-hoc court of admins, few of whom will actually look at the issues. There is an elected committee to deal with these things: if you want quicker results, see if you can reform the existing structures before bringing in completely new ways of doing things. I have no particular problem with "community bans", other than in that people seem to think that a couple of well-respected, uninterested admins cannot stop one of these bans happening. However, these "community sanctions" are extremely unlikely to help matters. Individual agreements with editors will work with half-decent users; otherwise, a body with clear authority is needed. All I am saying is that "community sanctions" should not happen in this way, and they should in no way be settled by a consensus of admins or anyone else on ANI. Sam Korn (smoddy) 13:08, 22 November 2006 (UTC)
Okay, to recap, I think we're in agreement that a formal process here is not such a good idea, and I think we can agree that iff community sanctions happen, it would be reasonable to log them somewhere. The question really is whether they should happen. I see your point about that, but since I'm not a strong proponent of community sanction in the first place, you may want to discuss it with someone who is. Arguably the best place to discuss that is actually ANI, where such things are occasionally proposed. (Radiant) 13:24, 22 November 2006 (UTC)
I absolutely agree with this. As Thatcher131 quite rightly told me off for earlier, this is the page to discuss these matters, although it can be useful to have notices elsewhere, e.g. mailing list and AN/I. I'll have a look at this later on today. Sam Korn (smoddy) 15:54, 22 November 2006 (UTC)

[edit] community or administrators?

There is a problem with the name "Community sanction". If the sanction comes from the Wikipedia community then there should be no special role for administrators in the process above the role played by other members of the community. If this proposed policy is meant to create a new power just for administrators, then call it "Sanction by administrators". —The preceding unsigned comment was added by JWSchmidt (talk • contribs).

Retrieved from "http://en.wikipedia.org../../../c/o/m/Wikipedia_talk%7ECommunity_sanction_7332.html"