Talk:Global bans/Archives/2011

Please do not post any new comments on this page. This is a discussion archive first created in 2011, although the comments contained were likely posted before and after this date. See current discussion or the archives index.

Current status of global bans

I understand that the discussion above mostly relates to the new proposed Terms of Use, but I'm unsure of the current status of global bans. Given the discussion here in regard to Poetlister, are global bans currently enforceable, and, if so, are they able to be overridden on a local level? My assumption is the the new Terms of Use will clarify this per above, whatever the final wording will be, but my interest is in the current situation. - Bilby 20:50, 4 November 2011 (UTC)

Please note that I explicitly changed to my volunteer account for this post: it should not be read as an official statement of Foundation policy. In my opinion, the actions by this user were so heinous as to justify a global ban. I know there are some who feel they have "investigated" and it's "not that bad" - they are in error. I have seen the full evidence, and I believe strongly that Poetlister is a clear and present danger to the stability of the wikis, and the WMF - should it so decide - would be entirely justified in enforcing any block or ban necessary. Having said that, I should clearly say that I will likely recuse myself from further discussion internally about how to handle this situation on Wikiversity. But it's very important to understand: there is some documentation that simply cannot be made public to protect Poetlister's victims safety and security. I can not stress enough how much I consider Poetlister to be a threat to the wikis. --Philippe 05:03, 5 November 2011 (UTC)

From what I've now read in different fora, I would agree that a ban appears to be warranted. I guess I have two issues. The first concerns whether or not it is warranted, and that seems like yes, (and I see no reason to question the consensus from the ban discussion), although this is more about a particular incidence. The second is about global bans in general, and concerns whether or not it is enforceable in local wikis. If I understand it correctly, the new proposed Terms of Use will allow for WMF-wide bans in conjunction with this policy overriding local consensus, which is a general principal that I would support. What I don't understand is whether or not currently, under the existing system, this is an issue for the local wikis to decide. If not then I presume that the block of a globally banned user should be automatic, but if it is something they can override then the WV community needs to make a decision. I'm being informed on en.WV that a global ban is the equivalent of a global lock, and can be overridden on a local wiki (if they don't support it) by delinking the account, or, as in this case, ignoring the alternate account. - Bilby 08:06, 5 November 2011 (UTC)

The status of the new terms and this document are that they are works in progress. However, that really has zero bearing on how a local community deals with Poetlister, because his ban was the result of a community consensus process that happened before these documents were created. See: Requests for comment/Global ban for Poetlister. That RFC was sent directly both to Poetlister and to the English Wikiversity community, as you can see from Abd's many comments. The discussion was closed and accepted by Stewards as a legitimate community consensus to ban Poetlister, not just lock the account. The reason other cases of account locks were undone locally was precisely because there was no global discussion resulting in a consensus decision. In this case, there very much was. If English Wikiversity doesn't block any local socks of Poetlister, they do so in clear violation of a global community decision in which they were invited to participate. Steven Walling (WMF) • talk 21:14, 5 November 2011 (UTC)

Philippe, I'll take it that you're acting in good faith. I have no reason to doubt you (as I in fact do with other people :-(). But, again, even so, this sort of reasoning raises my civil-libertarian hackles, along the lines of "This is guy is a clear threat to national security. Of course, I can't tell you why, because we must protect our intelligence sources and methods. However, the US, should it so decide, would be entirely justified in assassinating him." Now, again, I'm not questioning your sincerity, and I'm open to the idea that I do not grasp the gravity of the situation. But this all seems over-the-top to me. I mean, seriously, "a clear and present danger to the stability of the wikis, and the WMF"? How does one even get to do that, much less someone who is pretty marginal? When I talk about some of the really nasty stuff WMF and similar people have done against me because I'm often critical of Wikipedia, if I ever described them as "a clear and present danger" in terms of libel and defamation, possibly doing serious harm, I'd get a lot of grief telling me that I should just ignore it, comes with the territory, and they can't hurt me if I just stoically soldier on. He might be a bad character, I'm not disputing that. But, how can he be so bad that WMF needs a whole policy to ban him? Further, if he really is that malignant, a devil made flesh, a fiend in human form, how in the world is eliminating a screen name going to thwart his maleficent scheme? This isn't a rhetorical question. What sort of plot could he hatch that couldn't be done under a different name? (implying sock-hunts galore?). -- Seth Finkelstein 23:20, 5 November 2011 (UTC)

Civil Libertarians do not support an individual who has stolen multiple identifies, committed fraud on a grand level, and used deceit to get into positions of trust so he could get sensitive and personal information on others. Seth, your comments are out of ignorance and you really should have known better than to have made them. Ottava Rima (talk) 23:52, 5 November 2011 (UTC)

Sigh. Please re-read my comment carefully. Did I say anywhere I supported him? No, I did not. In fact, I made it clear what I was addressing. Your response is a common error, the stock "Why do you support those awful people?". Now, perhaps you might address the functional question I bring up - if he has "committed fraud on a grand level", what will stop him from doing so again, under a different name? Pragmatically, sometimes it's better to have someone where the authorities can keep an eye on them, rather than to deliberately lose that ability. -- Seth Finkelstein 00:01, 6 November 2011 (UTC)

I never claimed you did. As I said, civil libertarians do not support such individuals as Poetlister and gave reasons why. There is a lot of information visible on multiple Wikis of dozens of simultaneous accounts operating off of different identifies from multiple Wikis for the goal of gaining personal information on others. This was happening while applying for admin rights and other such things. The sheer amount of socking, identify theft, etc. has already been made public. How can you need more than that? Ottava Rima (talk) 00:28, 6 November 2011 (UTC)

It appears you may be missing my point about civil-libertarian, i.e. many times very bad people are at issue, but the matter is how much power are the authorities to have because of these bad people? Repeating these are bad people, bad bad bad indeed, doesn't address this issue. I heard you. Do you hear me? Moreover, repeating: perhaps you might address the functional question I bring up - if he has "committed fraud on a grand level", what will stop him from doing so again, under a different name? Pragmatically, sometimes it's better to have someone where the authorities can keep an eye on them, rather than to deliberately lose that ability. -- Seth Finkelstein 00:42, 6 November 2011 (UTC)

Seth, I appreciate that you care enough to think deeply about these issues, but to be entirely frank you're neither an active English Wikiversity contributor nor a particularly active contributor to any other project, so it's not really constructive to start a generalized argument here around your opinions of community policy. Especially not based on some notion of civil libertarianism, which isn't a term used in our policies. I think it's totally fine for the projects to make room for someone who has thoughtful things to say but doesn't want to write articles or fight vandalism, but there is a limit to how much such discussion is signal instead of noise. Please keep generalized arguments or hypotheticals to a minimum if not directed at either improving this policy or settling the question of Poetlister. Steven Walling (WMF) • talk 01:45, 6 November 2011 (UTC)

Hmm, as I seem to be pretty active on meta these days, maybe I've found my niche :-). Anyway, you may not agree with my contributions, but on the whole, I'd argue they're worth taking seriously. I also really should make a FAQ about the knee-jerking reply - I said "My civil-libertarian hackles" to orient the perspective I'm using - replying "which isn't a term used in our policies" is at best not grasping the point, at worst a statement you advocate WMF take action by decree from WMF people (which I don't think you meant, well, hope you didn't mean, but the text could be read that way). Maybe I'm misreading you, but my current reading of your reply is that you're denying the worth of a few words of perspective, to an extreme. That seems to me to be wasting far more time on noise. As in, for heaven's sake, given the points I raise, you reply only to deprecate my qualifications and about the introductory "civil-libertarian"? Truly, is that signal or noise? I'm going to start a count on this, since it seems to be ignored (n=3): perhaps you might address the functional question I bring up - if he has "committed fraud on a grand level", what will stop him from doing so again, under a different name? Pragmatically, sometimes it's better to have someone where the authorities can keep an eye on them, rather than to deliberately lose that ability. -- Seth Finkelstein 02:14, 6 November 2011 (UTC)

Civil libertarians are against people lying about their identity, infiltrating a database with personal information, and then using it to harm those people. No matter what you say, no one actually would support such action. So any claims about any ideology or groups falls flat. Merely making up a claim no one believes and trying to attribute it to a group without any basis isn't sound logical reasoning. Ottava Rima (talk) 01:47, 6 November 2011 (UTC)

This point has been asked and answered. I do not support any such actions you list. Repeating myself (n=4) perhaps you might address the functional question I bring up - if he has "committed fraud on a grand level", what will stop him from doing so again, under a different name? Pragmatically, sometimes it's better to have someone where the authorities can keep an eye on them, rather than to deliberately lose that ability. -- Seth Finkelstein 02:18, 6 November 2011 (UTC)

"I do not support any such actions you list" - as I said, no one does. Thus, there is no reason to even assume that it would ever be worth while discussing an unban of Poetlister. And it is silly to say that he can just do it under a new name. He has always operated multiple accounts while pretending one is acting acceptably. You do not unblock because of that. That is like saying if you put a drug king pin in jail and he continues to run his operation from behind jail that we might as well let him out. There is no logic to what you are claiming and it is utterly absurd. Ottava Rima (talk) 02:36, 6 November 2011 (UTC)

The "Thus ..." part does not follow, but I'm not interested in arguing for an unban for Poetlister (as opposed to being unconvinced that the ban was a tactically smart move, please note the distinction). I would say this metaphor is more like jailing one drug dealer of a large operation, and indeed, people do question whether that strategy is effective itself. -- Seth Finkelstein 03:10, 6 November 2011 (UTC)

Seth, it has already been demonstrated that 1. you aren't part of the community, 2. that you don't know Poetlister's background, and 3. that you admit that you don't support his actions. That means that your words are empty and serve no purpose. So why continue? And if you are saying this is like jailing on drug dealer of a large operation, then good. Not being able to get all criminals is not an excuse not to lock away any criminals. Not being able to do everything is not an excuse to do nothing. That is one of the worst logical fallacies. Ottava Rima (talk) 03:34, 6 November 2011 (UTC)

1. For the record, I most strongly dispute this point. I am often critical, and because of that many people, including some at WMF, dislike me both professionally and personally. However, my "membership", such as it is, is a valid as yours, as I act in good faith and make an effort to comply with all relevant policies. You have no right whatsoever to deny it. 2. I'm aware of the various allegations. Finding the truth in these cases is difficult and time-consuming (there's been, elsewhere, Wikipedia incidents which matter to me deeply, and after long investigation I still haven't entirely figured out what was going on). I grant certain negatives for the sake of discussion without taking them as proven. 3. Again, understand the civil-libertarian way of thinking - it's not about the bad people, it's about the powers exercised by authority. I suspect we have an unbridgeable gulf there, which we cannot resolve here. -- Seth Finkelstein 04:07, 6 November 2011 (UTC)

You have no edits at Wikiversity. You have no right to dispute the point that you are not part of the community. Your trying to do so verifies that there is disruption and not proper behavior. You need to stop. Ottava Rima (talk) 14:57, 6 November 2011 (UTC)

I read your original use of "community" as global, since this discussion is taking place on meta, where I have as much right to dispute an issue as you have to advocate it. I would say any disruption lies in trying to render my contributions here somehow illegitimate or even an offense (perhaps it will be clearer why I have concerns about global bans as a process which could be extremely prone to abuse). -- Seth Finkelstein 17:04, 6 November 2011 (UTC)

The question was about Poetlister on Wikiversity and how to handle that. You even acknowledged that a few times. Your feigning difference now when it is obvious that you no longer have an argument is a little unsettling. You overreached, you lost, and now you continue. Ottava Rima (talk) 19:26, 6 November 2011 (UTC)

The issues here are about global bans, and Poetlister's case is cited as an example. Your formulation would restrict people able to comment on meta to only those who are active contributors on Wikiversity, which is not appropriate. Note, your persistent baiting of me reflects badly on you. -- Seth Finkelstein 19:54, 6 November 2011 (UTC)

I understand some people feel there are concerns that need to be addressed because they are in on what transpired. I doubt a healthy relationship between projects/communities can develop or become stronger when all or nothing seem to be the only choices available though. Are there other things that can be done that might bring about compromises? For example, would a way to globally disable sending emails from an account or email address, and maintaining at Meta a blacklist of people that cannot have checkuser tools address some problems? I think having more choices can help with cross-project relations. --dark lama 12:56, 6 November 2011 (UTC)

Restarting the conversation on global bans

Hi everyone! As you are probably aware, WMF General Counsel Geoff Brigham has been working with the community on an update to the terms of use. That’s still in progress, but there are some other policies to which the new terms refer which need creation and polishing. To be specific, I’m referring the policy referred to in Section 10...

“In the interests of our users and the Projects, in the extreme circumstance that any individual has had his or her account terminated under this provision, he or she is prohibited from creating or using another account on the Project for which he or she was terminated, unless we provide explicit permission. Additionally, especially problematic Users who have had accounts terminated on multiple Projects may be subject to a ban from all of the Projects, in accordance with the Global Ban Policy.”

The basic gist here is that the new terms of use entail that you may have your editing privileges revoked if you’re using your account for abusive purposes, but declares that this is decision is normally up to the community unless the community asks the Foundation to intervene. (Which naturally is very, very rare. It's not really our job.)

Current local policies on blocking/banning cover almost all the relevant bases, and the terms support those existing community policies with no change. But in some rare cases of cross-wiki abuse that aren’t just spam/vandalism, the terms refer to a community-owned policy about global bans. This doesn’t really exist yet, although there were bits and pieces around.

I’ve worked up a version of a standalone global bans policy which is ready for further edits and comments or questions. I based it partially on the existing notes on Meta about global blocks/bans and partially on the process that actually worked for the only two existing global bans (Requests for comment/Ban User:PauloHelene globally and Requests for comment/Global ban for Poetlister).

If things change -- such as if the community creates a global requests committee or global arbitration committee to handle this kind of thing -- then the policy can of course be amended accordingly. That flexibility and control by the community is the big advantage of having a separate policy rather than putting details on any bans or blocks into the terms themselves, as the following passage from the terms points out...

“In contrast to Board resolutions or this Agreement, policies established by the community, which may cover a single Project or multiple Projects (like the Global Ban Policy), may be modified by the relevant community according to its own procedures.”

In addition to general wordsmithing etc. from all, I’d especially love to hear feedback from...

Stewards, whose input is absolutely essential.
Folks like Arbitrators, who have experience dealing with this kind of thing
Editors from projects other than Wikipedia
Editors from non-English projects (I am thinking it’d be nice to make translations happen sooner instead of later?)

It’s easier for me to consider how staff and English Wikipedians think about these things, as I am both. :) While myself and the legal team would like to hear from everyone, the groups I listed above should definitely chime in too.

Thanks for your help! Steven Walling (WMF) • talk 22:09, 27 October 2011 (UTC)

A few thoughts:

Make the software more transparent! It must be possible to implement global bans in a form that the people in the projects the user was active get a notice in Special:Contributions and/or the userpage. Global Locks in their current form are "stealth bans", usually nobody will notice that the user account is locked if he was not blocked earlier on that project.
All affected projects in which the user has made a certain number of contributions and is unblocked should be informed about the RFC, not only about a resulting block - only few people check meta regularly.
The critical question is what happens if single projects don't agree with global consensus concerning the ban. Do they have the right to opt out and allow that user to edit locally? If not, that would mean a big step away from the principle of self-sufficiency that most projects adhere to. Currently, an opt-out is possible afaik only if a 'crat renames the user, so that the account is no longer connected to the global account. --Tinz 01:31, 28 October 2011 (UTC)

I concur with Tinz about the need to find a way to "communicate" global blocks/bans/locks when checking user contribs on other projects (just as a local block is visible on their contribs page).
On the point of having projects "opt out" of global bans, one of the issues that has been encountered is users banned on multiple projects is that they used those other projects to continue their disruptive behaviour on the initial projects: emailing users they'd harassed on other projects, etc. In some cases, they persuaded admins or other respected users on other projects to "allow" them to create a new account which they then SUL'd to the projects they'd been banned on and returned to old disruptive behaviours. Risker 01:44, 28 October 2011 (UTC)

Hey Tinz, Risker. Thanks for the thoughtful comments. To address your points...

On transparency: my understanding is that there are some tools (such as Special:CentralAuth) to see the global SUL account of a user on all projects, as well as whether they're locked. Perhaps we should link to that tool from this page, which would be useful for people needing to see where a user is banned or active?
Under "Obtaining consensus for a global ban" steps 3, 4, and 5 include notifying any wikis where the editor is active. If you want, I think it's a good idea to make that language clear that it's mandatory; if you try to sneak a global ban in through the backdoor without notification that's not valid.
On the topic of opting out... the basic point here is that Geoff and I are thinking that because the Terms of Use specifically refer to the global bans policy, this means that people who edit despite being globally banned are technically violating the Terms of Use, which is not negotiable on a per-wiki basis. So, to be concise, I don't think we imagine local communities being able to opt out of a global ban. If someone really is editing in a positive fashion on one wiki but needs to be banned on another, then a global ban is not required. There is one contrary case I can think of: when a community where the editor is active was not properly notified, and comes to object to a global ban after a discussion is closed. Perhaps we should write that case into the appeals section?

Anyway, I'd like to say that the current draft is just my thoughts on what should be covered with a little input from others, and if you think there's a viable way to deal with any unaddressed issues you should go ahead and add some new language. :) Steven Walling (WMF) • talk 02:53, 28 October 2011 (UTC)

Since opt-out is discussed below, just an answer to the first point: I don't think that the toolserver tool is sufficient in the long run. Once you suspect that a user is globally blocked and know that global locks exist, that works, but what if you want to contact him on an unrelated topic? In such a case I would look at his talkpage and contributions, but I wouldn't even suspect him to be globally blocked because global bans are very rare. For my homewiki (dewiki) I have created a template a while ago, but that is not a perfect solution either because I can't monitor all meta locks regularly. I would really appreciate an improvement of the software here. (Is it hard to implement a box with a notice similar to the one for local bans, from a purely technical point of view?). --Tinz 09:21, 28 October 2011 (UTC)

Well sounds like we need to spec exactly what we'd need then, if we're going to make a request of tech resources. I'm wondering if it's easier to make global locks more clear, or actually create a global blocking mechanism? Steven Walling (WMF) • talk 18:50, 28 October 2011 (UTC)

Yes, a new mechanism for global blocking might be optimal, but I'd be already happy with a quick fix that generates local messages for globally locked users. --Tinz 19:26, 30 October 2011 (UTC)

If a user is productively contributing to one wiki, then there shouldn't be a global ban solely based on past editing behaviour. If and when the user starts to be disruptive on that project, then a global ban is warranted because it is clear that they have not changed and some action is needed to prevent further abuse. Ajraddatz (Talk) 03:44, 28 October 2011 (UTC)
Poetlister was "constructively" editing on a Wiki and has a history of doing that - mostly to try and get his other accounts unblocked while operating sock puppets. It isn't always so easy to know if someone is honestly be "constructive" or is just good at hiding their bad behavior. Such a user like Poetlister has been proven too dangerous to allow to edit. Ottava Rima (talk) 19:22, 28 October 2011 (UTC)

Opt-out

The issue of opting out is a much more complicated one than what is being considered. We had 3 cases this year of interesting cross-wiki blocks on Meta-

1.A user from enwp, who was blocked indef. by arbcom. Then started contributing on Meta and was subsequently blocked temporarily twice and unblocked both time after discussion. A very similar block log on commons and another project. But this user currently edits productively on all those wikis except enwp. At one point he might have had blocks on multiple wikis, would that alone have justified a global ban? if so, then we might have lost a productive community member.

2.A user from pt wiki who had intimidating behavior on wiki and was subsequently blocked on 2 wikis, one was a content project. Followed by a temporary block on Meta for intimidating behavior and than on another non-content project. His block on Meta at the time was turned into an indef. per request from a steward. This user also stalked several editors from pt wiki, publicly listing their names, pictures and private information off-wiki, even violating some European privacy laws. This person's block was finally turned into a global lock months later randomly by a steward after looking at the cross-wiki history. No one noticed.

3.A user from nl wiki, who had admin rights and a very high standing within the community was found to be socking across projects. The issue went cross-wiki again, and things got more complicated. The the blocks went cross-wiki on commons, meta, nl wiki and nvwiki. Last appeal was from Langcom to only restore editing privileges on Meta which was rejected by the community.

Those are 3 very distinct cases of cross-wiki blocks. My questions are

Where do we draw the threshold for global bans? 3, 4, 5 wikis?
Will there be a deliberation process?
opting-out might be necessary in cases of local arbcoms? how would that work?

These were complicated issues, the users went cross-wiki, a couple were really productive contributors and still are. Global lock without a proper procedure risks steam-rolling over their good contributions. Theo10011 08:53, 28 October 2011 (UTC)

Just to answer quickly, with mostly just my personal thoughts...

1.A: The policy specifically says in two different places that just because a user is blocked/banned on two projects does not mean they must be banned. Simply that it's the minimum. So no, someone editing productively somewhere probably doesn't merit a global ban, in my opinion. In general, the point of requiring an RFC and cross-wiki notification is so that points like that would be brought up and unnecessary bans could be avoided.
2.A: Sounds to me like no one really objected to that lock, or else it would have been overturned. In any case, the process described currently says that a group discussion, cross-wiki notification, and a clear consensus is required before asking Stewards to make such a lock. So that case would not be really acceptable or ideal if this becomes a policy.
3.A: I don't really know in that case, but my personal opinion would be that it sounds like a global ban would be overkill and unnecessary if it's already being handled. A ban exists to prevent harm, so if it's being handled without a permanent ban, then there's no need for one.

For your other questions:

Where do we draw the threshold for global bans? 3, 4, 5 wikis? The current version of the draft says two is the minimum threshold, but that just because there are two bans doesn't mean a global ban is required.
Will there be a deliberation process? The current version says that you can't have a global ban without an RFC and cross-wiki notifications, so yeah, there should be a deliberation process.
opting-out might be necessary in cases of local arbcoms? how would that work? If a local ArbCom wants to lift it's ban on an editor, then the global ban should probably just be overturned in favor of selective local bans/blocks. It's overcomplicated to retain a global ban but allow opt-outs. If someone is acting in good faith and is trustworthy enough that an ArbCom unbans/unblocks, then there shouldn't be a global ban in the first place.

Anyway, I think there are some answers for your questions already other than the more complicated opt-out issue. Steven Walling (WMF) • talk 18:33, 28 October 2011 (UTC)

It has already been demonstrated regarding Thekohser's account that a project can technically "opt out" by renaming the account and then renaming it back, thus "breaking" the global lock. Whether this is acceptable or not is another matter (Thekohser was not banned by consensus but through Founder action, so that is another matter all together). Ottava Rima (talk) 19:20, 28 October 2011 (UTC)

I completely disagree with setting 2 as a threshold for global bans. If so, Ottava would qualify for it, multiple times in the last year. I believe 3+ should be the ideal threshold, there are a lot of small projects where admins are trigger happy, two blocks are not that hard to accumulate in 5-6 years. I still think you need to clarify the details when it comes to local arbcom over-ruling the ban and how would the two work. Theo10011 22:01, 28 October 2011 (UTC)

Hey hey hey, only -one- of those blocks was a ban. The others were just blocks. I'm not -that- awful. :P :) And by the way, about 50% of the admin here consider me global bannable. It is a shock to everyone that I'm still around. Ottava Rima (talk) 02:51, 29 October 2011 (UTC)

I really don't want to set some arbitrary number, common sense should prevail. I don't want some rule to prevent beneficial action from being taken, as I could easily see this becoming (let's see, they've only been banned on two wikis, let's just wait for the third...) Ajraddatz (Talk) 03:29, 29 October 2011 (UTC)

Heya Ottava, that's my point, if someone looks at one of those temporary blocks and considers a global ones, we might risk losing people for just being opinionated or local issues on much smaller or non-content projects. I would be much more satisfied with this on a case by case basis. Theo10011 03:32, 29 October 2011 (UTC)

To be honest, I expect the Wikis to ban people and global bans shouldn't be a problem. There is very little actual encyclopedia or anything like that. Instead, there is a culture and community that exists on its own. As any culture or community, they would want to get rid of those who interfere with that. Why not let them? It is kind of a farce to pretend otherwise. Ottava Rima (talk) 03:59, 29 October 2011 (UTC)

┌─────────────────────────────────┘
I thought about the example you gave you Theo -- local ArbCom deciding a local ban of a globally banned user wants to be overturned -- and I realized that the way the requirements for a global ban are formed means there is no way a local ArbCom could be unaware of a global ban in a way that wouldn't leave room for them to object if there is the possibility they would want to unban in the future. The current draft states that you must notify all local communities where the editor is active or been banned/indef blocked. If say, someone banned on English and Russian is up for a global ban, then both ArbComs could comment during the RFC and say that they don't think a permanent global ban is merited. If a global ban is actually necessary -- these are people who are so dangerous to Wikimedia communities that we agree they should not ever edit, anywhere -- then creating an opt-out for a community that had plenty of time to comment and object to a ban is overly bureaucractic. There shouldn't be a global ban in the first place for a user who is likely to ever be unbanned locally by an ArbCom. Steven Walling (WMF) • talk 00:50, 30 October 2011 (UTC)

There was a case like this in Russian Wikipedia. A user with a large contribution went wild and was blocked for vandalism, then banned. He went to French Wikipedia and somehow managed to get banned globally for vandalism. After a year, he asked the Arbcom in Russian Wikipedia for a clean start. It was clear that it was easier to either grant a clean start or to rename a local account (I am not sure what was done in the end) than to ask stewards to globally unban him. He was locally unblocked and, last time I checked, he was working fine.--Ymblanter 19:07, 30 October 2011 (UTC)

Sounds to me like a case where a global lock/ban is overkill. FR should just stick with their decision and Russian should be free to make its decision too. Do you feel like it's important to write in something about overturning based on local community consensus Yaroslav? Steven Walling (WMF) • talk 19:37, 30 October 2011 (UTC)

Generally I would be against straight overrule of a global ban by a local community consensus: we had a number of users who behaved disruptively and then found a resort in a smaller project. Having said that we may wish to leave some windows open: for instance, only let the decision to overruled locally by an arbcom decision (which pretty much excludes small project), or to introduce community / arbcom appeal to the stewards which would have more veight than just an individual appeal. Probably more options are possible, I have not thought about this deeply.--Ymblanter 19:42, 30 October 2011 (UTC)

Okay, based on this conversation, I am thinking of adding some language such as this:

While banned users may not appeal a global ban, overturning a global ban (and subsequent global lock) may be considered if a community where the editor was previously active can demonstrate a strong consensus for un-banning the user locally. Requesting an unlock is preferred to local Bureaucrats breaking a lock on the account via renaming. Examples of sufficient consensus to overturn a global ban in favor of local blocks include a local Arbitration Committee decision or a request for comment with greater than 80% support from more than 50 active editors on a project.

What do you think? Steven Walling (WMF) • talk 18:07, 31 October 2011 (UTC)

Does it need to be set in stone? Even if someone is able to be unblocked locally while globally banned, there are many times where they are still dangerous to other projects and need to stay banned. It is better that one Wiki suffer from temporary insanity than to have them all fall together because of a few naive users. I say this because those like Poetlister have proved that they can create a bunch of names and worm their way into positions of power quite easily. Ottava Rima (talk) 20:04, 31 October 2011 (UTC)

I generally agree with your points, but I think we should avoid creating a policy that encourages local 'crats to break global locks whenever they see fit. We should rather be encouraging communities to talk to eachother and come to a consensus. Steven Walling (WMF) • talk 21:24, 31 October 2011 (UTC)

Right, but my point went the opposite way as you assume - I say that we shouldn't allow the local crats, who will try to do whatever they want anyway, to affect us. Look at the drama with the Italian Wiki, with the German Wiki, etc., where they let political people dominate and threaten the Wiki. Combine that with Thekohser and others having their locks broken by local crats and we can only assume that Wiki based insanity will exist in the future. What we should do is ensure that local insanity isn't the cause of us unbanning dangerous people. Otherwise, it will give these groups even another way to annoy and harass the Foundation (and thus harm lots of people who are just regular users). Ottava Rima (talk) 01:20, 1 November 2011 (UTC)

I believe that 65% would work better. enwiki can't even obtain 80% consensus on the Verifiability RfC. It's not possible to gain 80% on something as controversial as locally nullifying a global block. --Michaeldsuarez 03:45, 8 November 2011 (UTC)

I was just trying to provide an example of what is meant by "sufficient consensus". If it seems like that text would be actually suggesting you need to have 80% support in every case, let's just not add it. That's silly and arbitrary. Steven Walling (WMF) • talk 18:32, 8 November 2011 (UTC)

Eighty percent is an extreme example. It would be best if your example simply said "the majority", and leave it to individual projects to decide what "majority" means. --Michaeldsuarez 19:45, 8 November 2011 (UTC)

All projects are connected and are not isolated. However, some projects have little activity and they are easy to take over. Tiny projects would be a source for trolls to take over and continue to disrupt other Wikis, as has been proven elsewhere. There shouldn't be any consideration of any opt-out because it will only breed these problems and cause further disruption. Ottava Rima (talk) 14:30, 9 November 2011 (UTC)

I'm aware that freedom allows room for dangers, but I don't believe that the existence of risk should mean that freedom must be sacrificed. --Michaeldsuarez 21:32, 9 November 2011 (UTC)

According to Hobbes, once you violate your social requirement to not harm others your protection from harm (as a result of governmental punishment) is valid. It is that simple. Freedom exists only until it meets someone else's freedom, and there is no right to use Wikipedia. If you harm people, you are gone. It is that simple. Any attempt to justify keeping those people around is justification of harming others, and no one has that right. Ottava Rima (talk) 23:36, 9 November 2011 (UTC)

The goal of Hobbes' social contact is to protect people from anarchy. Do you really believe that Poetlister's mere presence on Wikiversity will result in anarchy? Since we're talking about global blocks, would Poetlister's presence on Wikiversity result in global anarchy? No, I don't believe so. This is why federalism and states' right exist. Poetlister on Wikiversity is a local affair, and the Wikiversity community is fully capable of handling it on their own. --Michaeldsuarez 13:27, 10 November 2011 (UTC)

Hobbes defended a social system that would keep people from taking the law into their own hands or acting without rules. Yes, that is anarchy, but he was focused on individual anarchy and the rights of individuals, as individuals make up the greater bodies. The only group entity that mattered was the sovereign, i.e. those who control the rights to bring harm to others (i.e. punishment of law). No individual had the right to bring about harm to another unless they were granted that right from the sovereign entity. Poetlister, in taking people's personal information, having multiple sock accounts in which he backed them up with stolen personal information and fraud, and waging war against many, many users, violated multiple people in such an extent that his permanent expulsion is the only way to protect anyone. If the world was fair, he would be in prison to protect society as a whole from such a dangerous individual. Ottava Rima (talk) 14:00, 10 November 2011 (UTC)

┌─────────────────────────────────┘
Sorry, I used a bad example. I also just realized that the Wikiversity discussions concerning Poetlister1 have ended. As you can see, the Wikiversity community was perfectly capable of dealing with the issue without interference from the global community. --Michaeldsuarez 15:11, 10 November 2011 (UTC)

That's not really accurate, considering several Wikiversity community members specifically mentioned the global ban consensus as the reason for why their community should block the sockpuppet. Yes, it was their local decision making process. But no, it was not a decision made in a vacuum. Steven Walling (WMF) • talk 17:22, 10 November 2011 (UTC)

It's also not accurate that the discussions have ended. --SB_Johnny ^talk 22:07, 10 November 2011 (UTC)

If i can just add ,that if someone has a bee in his' or her bonnet, then surely the point in question that makes them angry, then that very point should make it far easier to spot them in future, and so instead of banning ,maybe an alarm of some sort round the very topic of contention of that person, and will make it easier to spot ,because when you start banning people, that must surely be attacking the very principle that Wikipedia is surely setting out to make, by allowing all points of view and all attempts at consensus of opinion around very difficult areas , where opinion might play a very significant part , or maybe all of it.coolpolitealex 17:00, 22 November 2011 (UTC)

Just a note: decisions are taken by majority + one vote; policies are taken by a golden ratio, so the majority and the minority are very clear; two thirds are very difficult to achieve on controversial issues and 80% are impossible to achieve. --Chris.urs-o (talk) 09:05, 26 March 2012 (UTC)

Selective bans

(reposting per comment left on my talk page)

I understand people feel some concerns need to be addressed globally. However I doubt a healthy relationship between projects/communities can develop or become stronger when all or nothing seems to be the only choices available though. Are there other things that can be done that might bring about compromises? For example, would a way to globally disable sending emails from an account or email address, and maintaining at Meta a blacklist of people that cannot have checkuser tools address some problems? I think having more choices can help with cross-project relations. --dark lama 23:19, 6 November 2011 (UTC)

Technically there is currently no way to disallow only the emailuser function, so you'd have to request that. Anyway, I think it's perfectly reasonable to remind people that selective bans are a possibility, so I've added a note about this to the intro. Personally I think if someone is acting in enough good faith that they merit a selective ban, then a global ban proposal should be rejected outright in favor of acting on a case-by-case basis. But others might disagree depending on the circumstances. Steven Walling (WMF) • talk 22:56, 9 November 2011 (UTC)

Responsibilities of "local" sysops

In a now archived but perhaps unfinished discussion, Steven made the following comment:

"If English Wikiversity doesn't block any local socks of Poetlister, they do so in clear violation of a global community decision in which they were invited to participate. Steven Walling (WMF) • talk 21:14, 5 November 2011 (UTC)"

Could you please elaborate on that? I could see describing an unblock (particularly if tied to a SUL detachment) as violating a policy, but I don't think it's fair to demand that local sysops spend time looking for socks and/or engaging with people they would rather not engage with. Blocking is certainly a form of engagement, and requiring volunteers to engage seems a bit much to ask. --SB_Johnny ^talk 17:46, 7 November 2011 (UTC)

Good topic. What are sysops to do should local consensus differ from global consensus? Communities are likely to replace sysops who are in clear violation of local consensus with sysops who will uphold local consensus and do so in violation of global consensus. The global community could in turn feel there is a need to remove or replace community sysops who are in clear violation of global consensus with sysops who will uphold global consensus. If a local community and the global community aren't on the same page, sysops and community relations may be strained as a result. --dark lama 19:33, 7 November 2011 (UTC)

This is an unsolved issue. I still remember when we had a quiet discussion here on Meta regarding the global arbcom proposal, and then all of a sudden hundred or so en.wp users with no or little contribution on meta came and voted the proposal down, so that it was instantly dead. I felt like someone probably canvassing on en.wp just imposed his or her personal views in a way that it looked like consensus. On the other hand, we know that in really small projects weird things sometimes happen, with the full support of local community. In my opinion, both extremes are bad, but I do not how to find a balance.--Ymblanter 11:01, 9 November 2011 (UTC)

I find that a touch disturbing. "A quiet discussion"? There are 143,000 active users on the English Wikipedia; the fact that 100 of them show up on a global issue should not be problematic in the least. Surely there should have been notifications on various projects to get more opinions.--Prosfilaes (talk) 05:43, 25 March 2012 (UTC)

I would not be so sure. If these users would really read what is proposed and leave thoughtful comments, I would not have any problem with this. Especially if they were regular Meta contributors and would understand what it is all about. However, they obviously been solicited somewhere (WR again?), and the majority mostly just voted per xxx and never came back again.--Ymblanter (talk) 01:14, 26 March 2012 (UTC)

Apologies in advance for the wordy reply. I think there's sort of a few questions here:

First, what kind of responsibility do local sysops have? I pretty much agree with you here SBJ. There's a reason the policy doesn't explicitly say local sysops/CUs are obligated to hunt down local socks of a banned user and block them. The open nature of the projects makes perfect enforcement of a global ban very hard, if not impossible. However, the reason I responded the way I did is because here a local sock of a globally banned user is confirmed beyond any reasonable doubt. If it's obvious there's a bad actor around, I think local sysops should act to protect their projects based on the consensus for a global ban.

Second, is the question, What are sysops to do should local consensus differ from global consensus? Short answer: Of course no one here on Meta or at the WMF expects any local sysops to go rogue because of the global ban policy. They would probably just get their rights removed by the community if there really was a strong consensus against a local block. But I think all of you know that consensus is a fuzzy concept with room for admins to use their best judgement. That's your call, as always.

Third, the issue of participants from a larger project showing up and overwhelming the discussion. This is important. I'll start a new thread on this since I have a couple ideas.

To make a long story short: I think when a community learns about a sock of a user who was banned via an open, transparent community process to which they were invited, they have a responsibility to their own project to block. Steven Walling (WMF) • talk 19:26, 9 November 2011 (UTC)

may i inquire how exactly you intend to justify this principle and the underlying hierarchical causality? thx in advance for the elaboration, regards --Jan eissfeldt 23:28, 9 November 2011 (UTC)

I do not accept your assumptions about the principles I just laid out, especially that it implies some kind of immutable hierarchy. What we're talking about here is consensus decision making. Steven Walling (WMF) • talk 04:18, 10 November 2011 (UTC)

i see, thus your notion of having a "responsibility" is not grounded in some sort of "duty" to obey some sort of global decision locally and can be overthrown by everyone at every time for whatever reason at every level (a.k.a. wiki). the trouble with consensus decision-modi without some sort of hierarchical backup is that you can't claim to have reached it in the first place and in an open a.k.a. fluctuating system like ours you can't establish a principle of "responsibility" to be accepted without explaining reasonably _why_ a global decision somewhere down the road should be binding for locals. they may haven't been around as the "consensus" was reached.

simply saying that they could have participated in principle is not a good reason (especially since we all know that substantial parts of the community often feel that they cannot take part for all sorts of practical reasons from a lack of information distribution to language barriers), neither practical nor in principle, regards --Jan eissfeldt 09:54, 10 November 2011 (UTC)

with all due respect to the gibberish (sic) of "consensus talk": may i ask, if you have ever studied the rhetorics of early 20th century later-to-be "communist" apparatchiks? if so: do you see that there are some commonalities with the way you, a paid staff member, treat dissenting unpaid volunteers? Fossa 10:12, 10 November 2011 (UTC)

Fossa, I have been a volunteer for more than five years and a staff member for one. Let's not play that game of "volunteers vs. staff". Steven Walling (WMF) • talk 16:49, 10 November 2011 (UTC)

Four years as a "youth member"/"volunteer" isn't enough to get high positions in most parties. This is an interlingual project, do you speak any languages other staff don't speak? To be blunt, do you speak any language but English? Albert Noniem (talk) 23:20, 11 May 2012 (UTC)

Welcome on Meta, Albert Noniem--Ymblanter (talk) 18:36, 12 May 2012 (UTC)

As long as global ban discussions aren´t shown in every wiki in an highly visible place, a global ban does not necessarily mean global consensus. A lot of users don´t partipiciate because they don´t know there is a site for this.--Müdigkeit 17:59, 10 November 2011 (UTC)

This is absolutely true. Risker and Tinz bring it up in the first thread here too. For now I'll say that the draft requires notice to all communities where an editor is or was active. If you think there should be more or something, please edit or suggest something. Thanks Müdigkeit. Steven Walling (WMF) • talk 19:45, 10 November 2011 (UTC)

Balancing the discussion

So Yaroslav and others have brought up the issue that if a global ban is the product of a cross-wiki discussion here on the Meta, there is the potential for people from big projects like English Wikipedia to just overwhelm the request for comment. I have a couple points here:

Let's not assume that all global bans will be about someone whose primary wiki is/was en.wp.
A request for comment is a discussion, not a vote. Personally I trust Stewards to close this kind of discussion fairly with more weight for thoughtful, constructive points rather than just sheer numbers. We put editors in these positions of trust with the understanding that they are supposed to make judgement calls, not just tally votes. If we want to point out explicitly that a global ban discussion is not a vote, then let's do that.
Currently the policy draft requires notification to all projects where it is known that an editor was or has been active. If we want to be strict about this being truly global decision-making, we could also write in a requirement for comment from members of those projects before a decision is made. That would partially prevent a group of Wikipedians alone could decide something for the other projects.
This is kind of a crazy idea, but we could do a "progressive stack." Sue just talked about this on her blog. The basic gist is that we know people from big projects are liable to dominate a discussion, so why not try to tip the scales a little bit? In the context of IRL discussions this just means letting minorities speak first, but it could also mean something like we (visually) note every time a person who's not from English Wikipedia comments and ask the closing admin to pay special attention to them.

Anyway, just some suggestions to get the conversation going. Steven Walling (WMF) • talk 19:47, 9 November 2011 (UTC)

Extended off-topic replies between Ottava and Seth 04:10, 10 November 2011 (UTC)

"there is the potential for people from big projects like English Wikipedia to just overwhelm the request for comment" Anyone who makes such a claim doesn't understand Wikipedia, Wikimedia, or any WMF project. They are not isolated. We all have SUL. You are a member of every project and every community. There is no "overwhelming" from a "big project". There are people. If you are not part of their project, that is you not doing enough to be active there. Letting people think that small projects should be isolated and do their own thing is a violation of the spirit of WP:OWN, which is something no project should be allowed to do. Such a philosophy is cliquish, cabalish, and utterly destructive. We are one global community, and everyone who is in good standing has the right to defend this global community. Ottava Rima (talk) 23:39, 9 November 2011 (UTC)

"You are a member of every project and every community." Glad to hear it! -- Seth Finkelstein 23:52, 9 November 2011 (UTC)

Seth, that edit was you saying that you knew how the Wikiversity community would have felt, i.e. the topic of the conversation. That has nothing to do with having the ability to say a banned user should be allowed to edit. One is merely saying you are assuming things without having a background, which is not what the above is discussing. You know that. Ottava Rima (talk) 23:57, 9 November 2011 (UTC)

FYI, I never said such a thing. By the way, to make a significant point, I submit the political reversal above as a piece of evidence to observers (if there are any ...) as to why I believe the Global Ban policy will be gamed into simply heaping further punishment on people who run afoul of a certain group. There's really nothing I can do about this. But as I said elsewhere I believe in formalism to some extent, so I'm noting it. -- Seth Finkelstein 00:09, 10 November 2011 (UTC)

You were in a discussion about a very specific situation where everyone else was talking about Wikiversity, and then you started making statements applying to Wikiversity which you lacked the background to make because you never bothered even looking there. If you think the Global Ban policy will be gamed, where is the evidence? You cannot just toss up conspiracy theory. You made a living out of exaggerating things about Wikipedia while ignoring real concerns. If you honestly cared about protecting people, you are focusing on the wrong things. But lets be honest - I've run afoul of every powerful group. Yet I support the ban because the ban is the right thing. Ottava Rima (talk) 00:25, 10 November 2011 (UTC)

I dispute your summarization. However, the idea is that "membership" will argued in an exclusive way to marginalize people opposed to bans, but inclusively to increase those in favor of bans. It is easy to see how this can be applied as a clique mechanism to exclude opponents and include supporters. Oh, as a point of personal privilege to respond to an accusation - "made a living out of exaggerating things about Wikipedia"? HA HA HA. Please, let me know of the lavish munificent sums this has brought me, the big-money lecture fees, the consulting gigs, the stock options, the fame and hero-worship - because I seem to have missed it. My freelance writing earnings wouldn't put me near the poverty line! In fact, there's a good case to be made it's an overall negative to my life. You are sadly misinformed on that matter, perhaps as a projection from people who have indeed made a living "exaggerating things about Wikipedia" but in terms of evangelism. -- Seth Finkelstein 01:11, 10 November 2011 (UTC)

"the idea is that "membership" will argued in an exclusive way to marginalize people opposed to bans, but inclusively to increase those in favor of bans" And I would say that the opposite will be true. Thus, they negate each other. Before, I was saying that you were talking about stuff that had the context of a specific background and history. You could talk about an individual in general, but you are limited to your own experience or philosophy. Acknowledging your experience does not prohibit you from having a say, but it does narrow down what you can use to justify your say. "let me know of the lavish munificent sums this has brought me" - "a living" is a colloquialism that means that you have a habit of doing something. However, you were a journalist/reporter who had attacks on Wikipedia governance published which means that you have a history of focusing on negatives in order to stir up readers. Ottava Rima (talk) 01:27, 10 November 2011 (UTC)

Umm, I think it's going to be unproductive to continue if you equate critical articles with "in order to stir up readers". In addition to the absurdity, it's quite annoying to see those sorts of personal accusations, since they are so untrue. You apparently have no idea how advantageous it would be, e.g. for me to beat the wiki-porn-porn-porn drums. -- Seth Finkelstein 01:38, 10 November 2011 (UTC)

So you are the sole reporter in the world who doesn't wish people to have read his articles? Really? I find that hard to believe. If you can't admit that you sensationalized things for a reader, then that is a little weird. And advantageous to talk about porn? If that was so, Mr. Kohs's already highly detailed articles on the topic would have a lot more hits with out having to bother promoting them at all. Porn on the internet is not something that people care about. Ottava Rima (talk) 01:44, 10 November 2011 (UTC)

Do stewards actually require to close the RfC? I did not count, but my impression is that here on Meta most RfCs do not get closed but just become out-of-date. (Which may also say smth about whether the points addressed in RFCs are valid - but this is a different issue). On the other hand, a global ban RfC has to be closed reasonably soon.--Ymblanter 20:27, 13 November 2011 (UTC)

No, there is no deadline set in the policy. It just says that consensus is determined after a sysop or Steward closes the discussion, but doesn't say exactly when. Steven Walling (WMF) • talk 20:14, 15 November 2011 (UTC)

I think not having a deadline to close is good. I think a minimal requirement for a valid global ban discussion to remain open should be added though. Maybe a global ban discussion is to be open for 31 days or more. I think that would coincide with the typical time requests to create an entirely new project or close a language edition of a project remain open for at a minimal. --dark lama 13:33, 16 November 2011 (UTC)

That kind of feels like instruction creep to me. If we want a discussion closed (including as no consensus etc. because of time) we can simply bug the Stewards or Meta admins about it. Steven Walling (WMF) • talk 19:07, 16 November 2011 (UTC)

RfCs in general on Meta tend to last for more than a few weeks, so it wont really be necessary to worry about them closing "too early". Ottava Rima (talk) 19:09, 16 November 2011 (UTC)

I think people might rush to comment out of some sense of urgency when a project is informed of a global ban discussion if there isn't something in place to reassure people that they have plenty of time to thoughtfully discuss concerns. --dark lama 21:06, 16 November 2011 (UTC)

Without going to every person in the discussion and checking global contribs, it might be hard to tell if the conversation is even being flooded or not. One rather silly suggestion would be to create a bunch of that simply include a small image of the project logo, and allow people to self identify which projects they consider themselves active on. Thenub314 21:23, 16 November 2011 (UTC) (or maybe that should be File:Wikiversity-logo.pngThenub314 21:23, 16 November 2011 (UTC))

Not a bad idea. I wonder if we might just also create some kind of tally of participants from each project on the talk page of the RFC. Steven Walling (WMF) • talk 21:47, 16 November 2011 (UTC)

Not a bad idea, and I can see the appeal. I imagine people might start to feel it comes uncomfortably close to !votes being votes. Of course the tally wouldn't say the views expressed so in that way it should not be a problem, but I will be curious to see what other people think. Thenub314 22:08, 16 November 2011 (UTC)

Some questions

Why is the WMF board creating the possibility of engaging different communities against each other by introdcuting global bans which could lead to the situation in which a "mighty" (in terms of numbers) community is trying to oust members of a less numerous community. (Indeed the malheur with the "sibirean language" version is partly a precedent for such a scenaria.) Other potential war fields for that are the balkans and language versions involving states in which speakers of those languages are involved in conflicts between the ethnicities speaking those languagees. (Or, more directly asked: why the WMF does not keep the Wikipedia and sister projects not out of politics?)
Why the WMF board is creating just another controverse? Why they are so blind-eyed on the real problems regarding several projects? I am utterly irritated by the concentration on creating of more or less unneeded software features (e.g. delivery of cat images, assessment of articles in the English Wikipedia) or on the ongoing incapacitation of local communities (e.g. the boards image filter resolution, now this topic) while on the other side f. ex. the key Wikinews language version, the english language Wikinews, does just crash now or in some days will crash if it hasn't already happened (see n:en:Wikinews:Newsroom).

Wikipedia has existed for ten years without "global banning". I don't see the necessaritiy to invent this in its eleventh year. Honestly, I think that people in the WMF should firstly and most fastly reconnect to the Wikipedia communities. --Matthiasb 14:13, 10 November 2011 (UTC)

If we all have SUL, how are the communities different? We are all capable of moving between the projects, have Meta, Commons, etc., and the WMF provides for our basics (servers, etc). We are one community. Everyone can move between the communities. It is silly to think that any single community can exist in isolation or has the right to endanger other communities by hosting really dangerous users and giving them a platform. As for your claim that there was no "global banning" before, that is absurd. 1. SUL is the only way to truly cross ban someone and 2. from the start of SUL, Jimbo and the Stewards marked out certain users who were globally prohibited from access. It is not the same as a global ban, but does set a precedence that there was a global action keeping users from editing. Ottava Rima (talk) 15:22, 10 November 2011 (UTC)

We (most of us) do have an SUL but are active on one, two, or at most three projects, and have no idea about what is going on other projects. The global contribution button usually gives an idea.--Ymblanter 20:29, 13 November 2011 (UTC)

Hi Matthias, this isn't an initiative of the Board. There is no resolution about it. It's something the Foundation is doing in conjunction with the terms of use update. I can understand your frustration about all the projects being undertaken, but please understand that we are not a single person: just like the community, there is a lot going on at the WMF. The same people working on this are not the same working on say, WikiLove or the article feedback tool. Thanks, Steven Walling (WMF) • talk 00:34, 11 November 2011 (UTC)

I'll add that as community liaison I have been contacted by a number of users from several projects concerned precisely that there *is* no such mechanism in place. While I am speaking now only as to my own perspective, I don't believe a global ban is pitting communities against each other. I believe it is putting them in position to collaborate more efficiently by sharing information and working together to avoid systematic and insidious abuse of multiple communities. If Projects A & B have dealt with the same individual and know he has a history of stalking and threatening other users, the members of other projects should not have to be needlessly put at risk.

You raise good points about the possibility of cross-wiki politics, but this may be a matter to work out through the particulars of the policy rather than through having no policy at all. The current language restricts consideration of global bans to users who are blocked on two separate projects, which may help minimize such abuse of global ban procedures; are there other ways to help ensure that this is not misused as a political power play? --Mdennis (WMF) 11:55, 11 November 2011 (UTC)

Overturning vs. opt-out

I rewritten the last section to say more than just not to appeal a ban directly to the Stewards. It now points to the opportunity for overturning a ban entirely if a local community objects after the fact. Let me know what you think, Steven Walling (WMF) • talk 01:17, 11 November 2011 (UTC)

that's a very reasonable move, thx Steven. i proposed a second (but not final) step for this point according to the concerns raised above regarding a global-local balance, please take a look, regards --Jan eissfeldt 03:37, 14 November 2011 (UTC)

Thanks Jan. I made some copyedits to shorten it a tad, but other than that it looks good to me. Steven Walling (WMF) • talk 03:52, 14 November 2011 (UTC)

The problem is unsolved. It still means that local sysops are forbidden from opting a globally banned out of the block locally. Disallowing opt-out will serve nothing but lead to further tensions between communities. Think of Capsot's global block in the immediate run-up to the 2011 WMF board elections - he had full support of the Occitan and Catalan projects' communities, but was widely hated on Meta and elsewhere such that he got a global block, and that led to a vastly inflamed war between the communities of different projects. It was bad enough without such policy; I can't imagine how bad such cases would turn out should a global ban policy that disallows local opt-out be in place. Der yck C. 10:06, 21 April 2012 (UTC)

Self representation

Looking at the recent discussion of Poetlister and his case a thought occurs to me. The current page says allows for users to be notified if there is a global ban discussion about them. But in some cases the users may already be blocked or banned on meta and hence unable to take part in the discussion. Perhaps, in such cases, we should suggest provisional unblock at meta under the condition that the user edits only the page related to the discussion. Specifically excluding user talk pages and email. Of course any violation would bring about an immediate block and the user would lose the ability to speak on their own behalf. Thenub314 03:58, 11 November 2011 (UTC)

I like this idea. Self representation is good in that it can reveal another side to the story. Ajraddatz (Talk) 04:01, 11 November 2011 (UTC)

Agreed. There is no point in requiring notification to the editor if they cannot participate. ~~Feel free to add something about this in the appropriate section.~~ (Done) Steven Walling (WMF) • talk 07:43, 11 November 2011 (UTC)

I am opposed. Most of the people who would be banned used multiple accounts, mass disrupted, etc. To give them -any- unblock on a project after they were blocked for any reason just to give them a voice opens up that project to harassment. A user does not have the right to defend themselves. This is not a court of law. This is a community with standards. If you are egregious enough to have been blocked AND put up for a global ban, then you don't deserve any ability to respond. Ottava Rima (talk) 14:33, 11 November 2011 (UTC)

I agree with you on most points. This is not a court and could never be one. What constitutes "evidence" will always be circumstantial on a wiki, what determines the outcome is consensus, there could not be a "shadow of a doubt" type of standard. It will (sadly) never be innocence or guilt that is being determined. Thankfully the outcome of any decision is usually pretty minor in any person's life. If we banned Jimbo, he could happily edit wikia, give public talks, go on with his life. He would understandably be upset at having to leave behind a project he put alot of work into, but it wouldn't it doesn't carry the same weight of an actual court case either.

But, if a global ban process exists, in the decades to come someone will try to abuse it. Consider the (admittedly very unlikely) case of a user who hasn't done worthy of a global ban but who is being considered for a global ban. Allowing someone the chance to speak for themselves here at meta doesn't seem like very serious risk to me, global ban cases will be rare and closely watched. The moment any harassment or abusive behavior appeared a block would immediately follow. But allowing someone to speak in their defense bears a great payoff in how fair the process will seem to everyone involved. If they wanted to sock and continue abusive behavior they can and will do so without much difficulty anyways. This is why I suggested temporarily unblocking just for the discussion, and reblocking at the first sign of any disruptive behavior. Thenub314 21:38, 11 November 2011 (UTC)

Let's just throw it all out there - we can all agree that if there will be a global ban then chances are I will be up for a ban. It would be in my personal, selfish interest, to water such a process down so that it wont affect me. However, that would endanger the project and mean that it would water down for people like Poetlister. I would gladly be banned rightfully or abusively if it meant that someone like Poetlister never had access here again. Ottava Rima (talk) 22:45, 12 November 2011 (UTC)

"A user does not have the right to defend themselves."??? As a legal matter, of course. As a matter of what we should strive to do in terms of social fairness and ethical conduct, I disagree. I believe your argument "If you are egregious enough ..." assumes the conclusion. That is, rumors and emotions can run high about people. And I'd say Wikipedia is especially prone to pile-ons and status-based group condemnations. Giving the accused the opportunity to rebut potentially fallacious charges is an important procedural safeguard. -- Seth Finkelstein 23:58, 12 November 2011 (UTC)

"As a matter of what we should strive to do in terms of social fairness and ethical conduct" Nope. We are not a society. Even a society, as I demonstrated, would have everyone follow rules and termination of them for such egregious abuse would be immediate. You don't have rights here, nor are we here to allow for continue harm to the society. If a hand is rotten and gangrenous, one needs to cut it off immediately and not keep it in hopes it will make a persuasive argument on why it should stay. Even in the US, those who are deemed dangerous are remanded or even prevented from re-entering society even after released (the Supreme Court ruled such is acceptable regarding pedophiles). You do not have rights here, so you should not expect even that amount of leeway. Ottava Rima (talk) 02:23, 13 November 2011 (UTC)

Not to be pushy, but personally I consider this issue resolved. If a Steward or Meta sysop feels comfortable enough to fulfill a temporary unblock request, then that's their call. I don't think the policy should explicitly prevent this, and it might happen anyway even if we don't explicitly support it. Steven Walling (WMF) • talk 00:04, 13 November 2011 (UTC)

Given the predominance of the idea that banned-means-banned, I think exceptions do have to be duly written into the policy. Otherwise, anyone who acts to allow an accused to defend themselves risks getting grief over it. -- Seth Finkelstein 00:09, 13 November 2011 (UTC)

As someone banned means banned actually applies to, I think it is utterly necessary to actually enforce such a measure. Otherwise, it would just allow access to areas that those who are banned have been shown do not deserve to have any access. It is merely a flimsy excuse to give dangerous people a way to harm others. Ottava Rima (talk) 02:23, 13 November 2011 (UTC)

What sort of access? The edit button? The move button? Aren't these actions reversible? We're not talking about murder here. If such access has been abused (ie. vandalism), then a sysop can justifiably reinstate the block and repair any damage. Can't you give them a chance? --Michaeldsuarez 16:57, 13 November 2011 (UTC)

Access means ability to edit. That should have been obvious. Ottava Rima (talk) 17:06, 13 November 2011 (UTC)

I know. You're missing the point. Revisions are revertible and hide-able. An user who has been unblocked in order to defend himself or herself will be heavily watched and scrutinized. I don't believe that temporary access to the edit button will lead to a catastrophic outcome. --Michaeldsuarez 03:19, 14 November 2011 (UTC)

Merely being revertable does not mean that they are not harassing. Otherwise, "revertable" would excuse every bad behavior out there. There are people who don't deserve any access no matter what, and giving any is so dangerous and egregious that I would go so far to say those arguing for such people to even be temporarily blocked would deserve to be blocked. Ottava Rima (talk) 04:38, 14 November 2011 (UTC)

You've mentioned that this proposal might led to your own global ban. Do you believe that you're a disruptive user? I doubt that you're the only non-disruptive user being threatened by this proposed policies. Although you're willing to sacrifice your access and freedom, do you speak on the behalf of other non-disruptive users being threatened with a global block? Your account isn't the only one at sake. Sacrificing your account is a selfish thing to do if you don't take others into account. --Michaeldsuarez 15:06, 14 November 2011 (UTC)

I believe I am less disruptive than you are being, but that doesn't mean I am politically favored or supported in any way. Unlike most people around here (in support or opposition to Wikimedia) I am standing up for logic, reason, and sound governance. I'm not here to take jabs at a system that is working well because it may have personally hindered me in some way. That is petty. But being willing to step on everyone's toes means that I am, by definition, "disruptive". That doesn't mean that people who cause major harm should be let go because I personally may think I didn't cause any harm. We shouldn't let proven murders out of jail because we are afraid of falsely convicting a few thieves. Ottava Rima (talk) 15:19, 14 November 2011 (UTC)

Perhaps it's important to bear in mind that the 2 already existing cases where bans have been adopted involve people who engaged in the deception, manipulation, and disruption of entire communities, rather than people who simply tend to rub people the wrong way. --SB_Johnny ^talk 15:47, 14 November 2011 (UTC)

Clarification please

Globally banned editor A is an acknowledged expert in field B and, noticing a factual inaccuracy in the article on this subject on some wiki removes the incorrect information and replaces it with sourced fact.

Editor C reverts editor A, restoring the factual inaccuracy, on the ground that editor A is banned from the project. Is this permissible, since the overriding objective is to provide readers with correct, reliably sourced information, and it is apparent to editor C, from the source provided by A (which of course C has just removed) that in so doing C will be adding false information. 62.140.210.130 16:47, 21 November 2011 (UTC)

Removing reliable information is foolish, but does happen. But the situation your describing is really part of the policy of the local wiki's and deciding how they react to the edits from a globally banned user, and it is not really something we are trying to decide here. That being said, I suspect none of the local communities would support this type of reaction. Thenub314 23:06, 21 November 2011 (UTC)

Agreed. I think it's totally fine to leave the decision about how to deal with the content from a globally banned user up to each community. There's no reason to make a centralized decision on that one. Steven Walling (WMF) • talk 00:28, 22 November 2011 (UTC)

Thenub314 says "I suspect none of the local communities would support this type of reaction", which is obviously true, but when the reaction is that of administrators who use bans, blocks and page protection to override the community what can the community do about it? Individual editors are frightened to speak up because they fear that if they do it will be their turn next. A single administrator can block an editor indefinitely and a single administrator can cut off her talk page access, which is her last community link.

I will give one example. In this sequence an editor corrects the mis-spelling "describtion" to "description" and is reverted five minutes later by an administrator [1]. Two minutes later, to ensure that nobody else can correct the spelling error, the administrator protects the page until 2012. An editor noticing the mis-spelling is then reduced to posting an edit request on the article talk page which is actioned here: [2]. This is not an isolated lapse either - the same administrator has removed references and restored spelling errors in dozens of articles which he has then protected far into 2012 so as to ensure, as far as he is able, that the articles will remain inacccurate for the forseeable future. 62.140.210.130 16:45, 22 November 2011 (UTC)

I am sure you could inundate this conversation with examples of examples of true administrative insanity from any one of the projects. I could embarrassingly point to my own mistakes as an administrator. Being a volunteer organization there will always be people who obtain any given set of tools who are perhaps not qualified to use them, or people get overheated about some subject and make mistakes. These users should have their tools removed. This is most frequently done within the local community, but occasionally it comes up as RfC's here at meta.

The matter under discussion here is something of a different flavor. There are also cases of users who abuse, manipulate, harass and disrupt every community they have been part of as an editor at the WMF. These users might have been administrators, CU's, Overnighters, Stewards or had no special tools at all. But given that massive disruption has in the past occurred, there seems to be some wheels in motion to change the terms of use to allow for the possibility of a global ban. If this is the case we would certainly like to have a sensible policy crafted.

The important thing to keep in mind is that this policy is not meant to override local matters, or even touch upon them. It is about when/how we should try to seek a global consensus to ban users that have been banned from multiple places and who we expect will cause more problems in the future. Thenub314 19:04, 22 November 2011 (UTC)

But isn't this policy designed precisely to override local matters? Without any forewarning a local wiki is presented with a fait accompli. A lot of the trouble is caused by administrators who are impetuous, inexperienced, incapable of interacting with people in the real world and about twenty years old. We've had thirteen year - old bureaucrats before now. Shouldn't there be a lower age limit for these positions? (I believe the President of the United States has to be at least 35). At least the local wiki knows who its crazy administrators are, even if it can't curb their activities. These global bans are going to be implemented remotely from a place where they are not known, giving them an easy method of wreaking their havoc across hundreds of wikis.

Let me give another example of the problem on English Wikipedia. In this sequence editor A quotes extensively from a source and expands the discussion of another source. Impartial editor D reverts editor A and then restores A's edit [3]. Editor C then reverts again, quoting the exact opposite of policy as justification for his intervention [4]. At this point D, rather than get involved in an edit war with C, simply gives up. If A had then offered moral support to D by restoring D's preferred version C would have continued the edit war by reverting A. The way the crazy administrators handle this situation is to revert so that the page shows C's version (rather than D's) and then lock it indefinitely. Editors shouldn't have to work under this kind of intimidation. 62.140.210.130 19:02, 23 November 2011 (UTC)

I'm going to come out and oppose this proposal because it's assymetric. It does nothing to address the problems caused by poorly - performing administrators. People talk the talk about solving this but they don't walk the walk. Result - current situation where people like Ottava Rima and Poetlister are banned on some projects and turn up as administrators or bureaucrats on others. 62.140.210.130 16:02, 25 November 2011 (UTC)

There is normally discretion in reverting the edits of banned users, and that makes sense. If someone is banned for off wiki harassment their actual edits may be quite legit. However if we discover someone whose edits include copyvios or the occasional outright hoax then it is probably safest to revert everything they've done even if that means losing some information that would check out as accurate (some copyvios are very accurate copies of reliable sources). Editors who check the sources should of course be free to revert back to edits that they've verified as good edits. WereSpielChequers 07:31, 29 November 2011 (UTC)

The objection is not to removing content inserted by banned users unless it's a "shoot first" reversion - i.e. made before an SPI has been filed to determine whether the editor is a sock. It's a commonsense objection to removing correct, sourced information added by a sock and at the same time reinserting the false, unsourced information which the sock removed. In my view this is a contravention of the Terms of use (see Talk:Terms of use#Falsification of content). 62.140.210.130 16:17, 29 November 2011 (UTC)

If someone has been banned then we can no longer assume good faith. If they have been banned because some of their edits replace good edits with bad then sometimes it is safest to revert everything they've done, even if in some cases their's was a good edit and reverting it means going back to a worse version. The alternative would be to check each and every one of their edits and only revert the bad ones, and that isn't always a sensible use of time. WereSpielChequers 01:34, 30 November 2011 (UTC)

There's a section in policy (you can read it at en:Wikipedia:Banning policy#Evasion and enforcement) which seems to cover this point. It says

When reverting edits, care should be taken not to reinstate material that may be in violation of such core policies as neutrality, verifiability, and biographies of living persons."

If you revert edits without checking them you cannot comply with this part of the policy. 62.140.210.130 16:17, 1 December 2011 (UTC)