Wikipedia talk:Blocking policy/Personal attacks

Older proposals archived here and here.

Older discussions archived here and here.

Misc.

Re "implicit" attacks.... can an attack be both personal and impersonal as in the example... seems contradictory? Pcb21| Pete 11:02, 14 August 2005 (UTC)

• What is basically meant is when User:Foo does something, and User:Bar responds with something like "see, once again the evil cabal nazis invoke censorship on poor honest users" (thus implying that Foo is an evil cabal nazi). If you find it a poor example or poor wording, please edit. Radiant_>|< 11:33, August 14, 2005 (UTC)
  • I think that the case you have in mind is quite clear cut. However I am trying to second-guess how some people might try to abuse this extended rule. Pcb21| Pete
    • Thanks. Radiant_>|< 18:16, August 14, 2005 (UTC)

The corrolary says it is not ok to ask an admin to block on these grounds. The majority of attacks take place on Talk: pages or User talk: pages. Supposing the article was something obscure with only 2 or 3 interested editors, how would a user effect a block? Via WP:AN/I? Perhaps there should be another subpage, like there is for WP:AN/3, e.g. WP:AN/PA. For the complex proposal, the styling of it could be imported directly from WP:AN/3. For the simple proposal, something like WP:AIAV would do. Either way, there would need to be an alert mechanism so that Admin X isn't drawn into the dispute with non-neutrality resulting from a pleading message on their talk page. -Splash 15:14, 14 August 2005 (UTC)

• I've removed the "do not ask admins" bit. If people are in favor of setting up a central page (an issue on which I have no real opinion one way or the other), I'd prefer if "telling admins" was done there rather than on their private talk page. Radiant_>|< 18:16, August 14, 2005 (UTC)

About the complex proposal: I've added a clause that prevents the block from 'using up' the previous attacks. Otherwise, when the block expires, the user would have another week to make 3 more attacks. That is, I could attack you on Monday, Wednesday, Friday and Sunday, for a total of 12 attacks per week! The 3RR does not have this problem since the block length is as long as the time period. -Splash 15:37, 14 August 2005 (UTC)

Shorter is better. - brenneman^(t)(c) 14:04, 15 August 2005 (UTC)

A point from WP:NPA

In the Remedies part of WP:NPA, which is official policy, it is provided that you may remove such attacks. A reference is given to WP:RPA. This effectively promotes WP:RPA to policy status. I noticed this because it's being used with that status by a user today to justify removal of mildly attackish comments from other users' talk pages. How should we square the two? Best to remove the remedy from WP:NPA I think, given its disputed nature? -Splash 17:37, 15 August 2005 (UTC)

• Yes, removing that would be best. Policy pages are edited all the time, it is possible that that section was added unilaterally by a supporter of RPA. Radiant_>|< 08:03, August 16, 2005 (UTC)

Interference

If this proposal becomes policy, I am inclined to think it ought to be accompanied by a new policy of "Do not remove personal attacks from other people's comments". WP:RPA is already disputed, and contraversial. If there is another way of dealing with personal attacks, then there is less justification for editing another's comments on a talk page (an inherently bad thing, i feel), just as WP:3RR is suppsoed to reduce the motivation for engaging in revert wars. DES ^(talk) 16:00, 14 August 2005 (UTC)

*controversial Erwin

I tend to agree. Although, I'd go even further, and suggest the concurrent development of a policy prohibiting modifying other users' talk page comments (except on user talk pages). There isn't a prohibition as such currently, but it's strongly frowned upon by the community -- I think that this would be a natural and a well-crafted policy on this score would probably be approved. · Katefan0^(scribble) 16:33, August 14, 2005 (UTC)

I think we should do that the other way around. The comments that result in the block under this proposal should be removed. The diffs will remain available from histories, and the hotheaded editor doesn't get the satisfaction of seeing their comments remain and, usually, get commented on. As to user pages, well, I think the alteration of those is frowned upon strongly enough that we don't need a rule about it yet. And I think to block someone for editing userpages would be a little harsh. Blocking for additions of attacks to userpages is different, of course. -Splash 16:42, 14 August 2005 (UTC)

That's the exact opposite of what I suggested. I said we should think about doing a policy that states that modifying other users' comments on ARTICLE talk pages is prohibited. A person can do whatever they feel like with their user pages. · Katefan0^(scribble) 16:58, August 14, 2005 (UTC)

Oh, sorry, you did indeed say that. Hmmmm. I'll have to think about that, although I'm sure this has been talked about before. -Splash 17:03, 14 August 2005 (UTC)

• If this proposal passes, I would be happy to deprecate WP:RPA - both are countermeasures to the same problem, but this one is better. Katefan makes a good point in that modifying other people's comments is generally cause for an instant block (on grounds of vandalism) but this isn't clearly stated anywhere. Well, it is now :) Radiant_>|< 18:16, August 14, 2005 (UTC)

Asking for help

Why would it be inappropriate to ask an admin to block on these grounds? -- Jmabel | Talk 16:50, August 14, 2005 (UTC)

• The idea was that it sounds like whining, and complaining to admins who may have better things to do ("stop calling names or I'll call my mommy"). I suppose that's not really a good reason, so let's snip it. Further up, some other people suggest another solution. Radiant_>|< 18:16, August 14, 2005 (UTC)

Exception?

I disagree with this addition... "Accusing an editor of being a sockpuppet, meatpuppet, or vandal is not usually a personal attack". I think that accusing any established editor of being any of the above clearly is an attack, unless there is strong evidence to the contrary. I've done several vandal reverts in concert with other editors (e.g. User:Vandal blanks out the RFC page, then within a minute he is warned by one editor, reverted by a second, and blocked by the third) - and afterwards such vandals tend to throw around spurious claims of sockpuppetry. Radiant_>|< 18:16, August 14, 2005 (UTC)

I was trying to avoid the VfD situation of tagging a new user's vote (even without specifying puppetry) and then the wikilawyer new users saying "block the evil editor for he called me a puppet, and I would never do such a thing!". Equally, the various {{testX}} messages certainly imply an accusation of vandalry and I wouldn't want to feel that I was blockable for using them. Unless, of course, I were acting in bad-faith: which is why I included the sentence following the one you mention. Then, the kind of thing you point out above remains blockable as a bad-faith version of the accusation. -Splash 18:22, 14 August 2005 (UTC)

• A statement like "user has 15 edits" is not a personal attack if it's the verifiable truth. Even if the user in fact has 16 or 17 edits, it's a reasonable mistake and not a personal attack. I do believe the various {{test}} messages are neutrally worded, and they are rarely used spuriously. Radiant_>|< 08:06, August 16, 2005 (UTC)

Is "troll" a personal attack?

How about accusing someone of trolling? Do personal attacks on user talk pages count? I certainly think user talk pages are VERY different from article talk pages, or worse yet, article space edit sumaries. ¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸ 19:55, 14 August 2005 (UTC)

I would think that it is not important where the attack occurs. Wherever they happen: article talk, user talk, article, user, edit summary, wikipedia: space, wikipedia talk: etc etc, it is still a personal attack and should not be condoned. As to whether "troll" is a personal attack or not....this is the kind of thing we must take great care over: we should not finish up censoring ourselves for the benefit of....trolls. -Splash 20:03, 14 August 2005 (UTC)

• I think that, like "sock" and "vandal", "troll" is very much a personal attack. None of those terms should be used unless there's evidence of the alleged bad behavior. All three are generally used (in my experience) to speak about newbies that have a strong indication of being a sock/troll/vandal, in which case they're true, and they're sometimes used to speak about known editors that one dislikes. Radiant_>|< 20:15, August 14, 2005 (UTC)
• To Say "that comment is trolling" or "User X has committed vandalism on several occasions" is far less likely to be a personal attack that "You are a troll" or "X is a vandal". I wouldn't say that there is an absolute line here, though. Calling a generally constructive editor eitehr a troll or a vandal is likely to be seen as a personal attack, even if that person is arguably trolling or vandalising in a particular case, and more so if nos such argument holds water. DES ^(talk) 20:23, 14 August 2005 (UTC)

No offense, but isn't this debate a little bit outside the bounds of the policy being proposed? It seems to me that it mostly leaves the decision of whether something is a personal attack or not to the examining admin. · Katefan0^(scribble) 02:23, August 15, 2005 (UTC)

It may be outside the policy proposal, but it is something that needs to be addressed. "Troll" is one of the most common terms used to describe disruptive/unliked editors, and if it is left to the descretion of the blocking admin, then it will result in blocking wars. Admin X thinks troll is definatley a PA and blocks, Admin Y thinks it is obviously not and unblocks, and voila, a blockwar has errupted. If it were any other term (besides vandal and sock, perhaps) it would be no big deal, but blocking over such a commonly used term must have an absolute right or wrong answer in order to avoid causing more problems than it solves. -- Essjay · Talk 03:49, August 15, 2005 (UTC)

I'm with Essjay on this. The four words: "troll", "sock(puppet)", "meat(puppet)" and "vandal" are widely used and are not, when made in 'good'-faith personal attacks. They may of course be used inappropriately and be personal attacks too. I think it is sensible to establish that these particular terminologies (and their derivatives, e.g. vandalising, trolling) are not personal attacks unless made in bad-faith. Then leave bad-faith judgements up to the neutral, blocking admin. We should not censor ourselves for the sake of those who can't behave properly. -Splash 17:29, 15 August 2005 (UTC)

I just don't see how the distinction can be made for one commonly-used phrase but not another. If we leave everything else up to admins' discretion, why not these words? What good-faith remark WOULD be blockable? I can't see how any true good faith comment would be a blockable offense, so doesn't it naturally rely on that admin's discretion to decide? How about "POV-pusher/POV warrior?" That's not included in your list, but arguably could be. I'm searching for a bright line to draw here and finding it mushy. · Katefan0^(scribble) 17:47, August 15, 2005 (UTC)

Additionally, I find it problematic to think of auto-blocking anyone who raises the specter of sockpuppetry. It would seem to put a muzzle on talking about potential bad actors. It seems an unreasonable burden to require someone to know definitively that one user is another's sockpuppet, when there is no real way to know for sure for the average user, and yet good faith editors sometimes do question others on this score. I've seen admins suggest sockpuppetry as well. Is raising the question of sockpuppetry, while avoiding directly calling someone a sockpuppet, a blockable offense? Or are admins the only ones who could raise the question? If this is the road folks want to go down, there needs to be some strong definition on this point, I think. · Katefan0^(scribble) 17:52, August 15, 2005 (UTC)

Ok, I think the trouble is that, by trying to work out if a given word is/not an attack, we effectively trying to write a Wikilibel law. We're not going to manage that; the courts in real-life are full of this discussion. So, how about we just go back to saying something like "any baseless accusation" or any "bad faith accusation". I'm not sure there is a bright line that can be drawn. -Splash 17:59, 15 August 2005 (UTC)

Any comment about an editor's character is a personal attack. Any comment about their behaviour is not. Now, we can argue till we're blue about "troll", but it is essentially commenting on an editors behaivor, even if it is used as a predicate nomitive ("you are a troll") rather in the more behaviour oriented language ("You are behaving like a troll"). I think the thing that needs to be kept in mind is that commenting on someone's behaviour should be allowed, that it allows poeple to vent, and that such comments are really informal "request for comments" on another editor's behaviour. If I cannot state my opinion that some editor's behaviour is troll-like, then I cannot comment as much on an RFC, unless of course, the RFC has somehow been found to be "true". At which point, we've got a catch-22. (how can we know someone is a troll without someone making the first comment? And if someone makes the comment, how do we know its a personal attack unless we really dig into both sides of teh story?) The line is easy to draw. comments on an editor's behaviour should not be the subject of an automatic block by an admin. If someone calls someone else a troll, then the only way to really sort out who's a troll and who isn't is to get an RFC, and get a bunch of poeple to weigh in on the situation, get both sides of teh story, rather than just have an admin take a quick look at a diff and 'pow' block someone for calling a troll a troll. Do not make this rule so that people are suddenly afraid of commenting on another editor's behaviour. FuelWagon 18:42, 15 August 2005 (UTC)

Warnings

I don't understand the purpose of disallowing involved parties from warning the user. Maurreen (talk) 05:15, 15 August 2005 (UTC)

• They can't place an official warning (using the {{npa}} things) because they're party to the dispute and therefore (possibly) not objective. Of course they can still say "don't do that". The intent is to prevent conflict of interest. Radiant_>|< 09:20, August 15, 2005 (UTC)

I know what you're getting at, but I'm not sure it's a great idea. If someone spams or vandalizes my user page, I can put a {{spam}} or {{test}} on their talk page myself. I think the conflict of interest (ie, misinterpretation, or over-interpretation of personal attacks to influence a block) can be solved simply by strongly instructing the blocking (non-involved) admin to check eachof the warnings for truthfulness. If one is frivolous, then it can be discounted. I think it's asking too much for an uninvolved good samaritan to see a personal attack and be sure to warn them while the attacked party can do nothing. Especially since with the warning, they're likely to be involving themselves, and thus it's a self-unfulfilling instruction. It just seems to add another layer of bureaucray is all, andI'm sure we trust admins to check. Dmcdevit·t 09:31, August 15, 2005 (UTC)

• That sounds reasonable. We should probably reword it to make that clear, while maintaining the idea that the blocking admin should not be an involved party. Radiant_>|< 10:19, August 15, 2005 (UTC)

traffic ticket

How do you warn someone and then block him for the same attack? I would suggest some slight modifications to this proposal. First of all, this sould be treated as a traffic violation, running a red light, speeding, whatever. The first incident could be an official warning from an uninvolved admin. All incidents after that can be left to the admin to give an official warning or an official block, depending on the violation. This encourages editors in a dispute to get an outside party to resolve the problem sooner, rather than later. It also gives the editor official warning from an uninvolved party. The point is to snap the editors involved out of their road-rage with an outside voice saying 'knock it off'. having editors invovled in the dispute warn each other will only escalate the problem by making it more confrontational. Imagine if a driver had to be warned by another driver of his behaviour before he could get a ticket for road-rage. You're violating road rage. No, youre violating road rage rules. Treat it like a traffic ticket. Give out official warnings (maybe log it as a 1 minute block) log all the warnings and blocks. Then third adn fourth violations can show a pattern of abuse that may require other measures. FuelWagon 14:35, 15 August 2005 (UTC)

I'm not completely clear where you're driving (excuse the pun) here. The proposal already requests a non-involved user give the warning and a non-involved admin do the block. Though there is nothing much wrong with having an involved user give the warnings, as long as a neutral admin gives the block. This is basically the same process that the 3RR works on, and it seems to do ok. This would be aided by a centralized page for this process so that any outside admin can come along and help and either block or dismiss the request. -Splash 14:59, 15 August 2005 (UTC)

If a warning is required before getting a block, and if users can give other users a warning, then it only encourages escalation. "you're pushing POV". "Yeah, You're a troll." "You are violating NPA and will be blocked if you continue". "Oh yeah? well, consider yourself warned for your personal attack." The design of this system gives the parties involved ammo to exchange fire and escalate the problem. Two editors in a dispute "warning" each other will do nothing to actually diffuse the debate. They can try, and there isn't anything against it. But I strongly recommend that if a warning must be made before a block is enacted, then it must not be a warning from a user, but from an uninvolved admin. There is no point in giving editors projectiles they can throw at each other. "I'm warning you". "No, I'm warning you". If it's a bad enough attack, then get an admin and they'll give out a block, don't have the editors go at each other that much more because a warning is required before getting a block. I'm even more against the "complex proposal" because it turns it into more of a us-v-them situation and mob rule. There is no advantage to have involved parties being able to be part of teh judge/jury/executioner panel. It can only lead to trouble. If someone violates NPA, report the diff and have an admin warn or block, end of story. Log the violations, and if enough accumulate, use that in an RFC or something. But don't create a system that escalates teh confrontation between the involved editors and can be turned into mob rule. Involved editors should report an edit that they think violates NPA, and then an uninvolved admin deals with it. FuelWagon 16:53, 15 August 2005 (UTC)

I see what you mean. However, the 3RR doesn't suffer cripplingly from this problem, and the threat of it is frequently thrown around both in edit summaries and talk pages. It too has templates {{3RR}}, {{3RR4}}, {{3RR5}}, although I'm not sure how often they're used. The difference between these and the {{testX}} templates is that those really serve the purpose of letting anon vandals know they're being watched and that is usually enough. These new templates/warnings would not be intended to have that effect really, and could just become rocks to be thrown. So, perhaps we should do away with the tolerant, complex proposal and simply say "if you attack, you're blocked", no second chances. Set up a page for reporting a personal attack, to keep it off AN/I and to stop admins being dragged in via their talk pages, and escalate the length of the blocks: 1hr first offence, 12 hours second, 24 hours third. -Splash 17:33, 15 August 2005 (UTC)

I strongly support the idea of creating a "noticeboard" similar to 3RR. Beyond efficiency reasons, which are enough, I also think not having one for personal attacks, which are supposedly prohibited in their entirety, sends a message that attacks will get some disapproving looks but won't really be punished. · Katefan0^(scribble) 17:35, August 15, 2005 (UTC)

The complex proposal

This is highly problematic. Asking users who are already involved in a heated debate to be part of the resolution process may only escalate a flame war on an article talk page to spill over to getting enough warnings against the opposing editors to get them blocked. -- FuelWagon

moved to talk page Radiant_>|< 14:55, August 15, 2005 (UTC)

implicit comments

Regarding calling "implicit comments ("The despots here…")" a personal attack:

I see this as highly problematic and subject to interpretation. An editor might complain about the "cabal of editors" or that wikipedia "is a cabal", and allowing implicit comments to be interpreted as a personal attack could almost be used to say "thou shalt not criticize wikipedia". I'd much rather see a simple, direct, objective definition of a personal attack, a simple direct blocking policy for violating NPA, and then say somethign like admins are given discretion or some such thing. Then this doesn't become a blank check to block someone just because they were griping on a talk page somewhere about behaviour in general. Non-specific comments always have some grain of truth anyway. If every editor was perfect, we wouldn't need policy and admins and blocks. So, given all the editors on wikipedia, there is at least one editor that qualifies as a despot. FuelWagon 17:03, 15 August 2005 (UTC)

The other thing is that I sometimes use "implicit comments" as a way to not escalate a situation but still comment on what's going on. "Some editors are pov-pushing" is a way to put people on notice without making a direct accusation. The problem with direct accusations is that it puts puts people on the defensive, and sometimes does nothing but escalate the problem. FuelWagon 18:45, 15 August 2005 (UTC)

• The key is personal attacks. If I say that "some editors are morons", without referring to somebody in particular, that is not a personal attack (similarly, if I complain how Wikipedia sucks in general, that is not a personal attack). If I say that "User:Moron is a moron", that is obviously and explicitly a personal attack. An implicit personal attack would be if User:Moron posted a lengthy treatise somewhere on the Wiki, and I added a single-line comment at the bottom stating that "some editors are morons" - because in that case, it clearly refers to User:Moron above even if it doesn't explicitly mention his name. Of course what I just said may be controversial enough to drop it from the proposal, or at least cause for rewording it. Radiant_>|< 08:18, August 16, 2005 (UTC)

I guess it comes down to a choice: allow for subjective interpretation and catch this one case (and posibly make the whole process more problematic) or prohibit subjective interpretation and catch this one case with an RFC. I'd rather the the system encourage individual integrity for admins and let some editors get picked up with an RFC. FuelWagon 20:36, 16 August 2005 (UTC)

Too vague

This is all a bit vague, and probably unnecessary in the long run. Most of the time, personal attacks get dealt with quickly and effectively by either the parties involved or by other editors. In extreme cases (such as the user who was posting nazi imagery on LGBT editors' talk pages) the offender has been blocked. But this is just going to lead to petty little quibblings over who attacked whom and what's bad enough to be a punishable attack and take away from constructive discussion and editing. In addition, some users are going to glom onto this as a way of muddying the waters whenever possible, so that admins are forever going to be going on wild goose chases....

I'm against any proposal that would identify as blockable any single remark that isn't racist, sexist, homophobic or otherwise discriminatory (all of which are richly deserving of blocking on the first offense). Beyond that, it's up to admins to make judgement calls and individual editors to make use of RFC and RFA, which after all, is why those pages exist. Exploding Boy 23:12, August 15, 2005 (UTC)

"racist, sexist, homophobic or otherwise discriminatory (all of which are richly deserving of blocking on the first offense)"

Thats not current policy, I suggest you propose it so that I can vigorously oppose such discriminatory special rights. You may want to review Wikipedia:Blocking_policy#When_blocking_may_not_be_used and/or Wikipedia:No_personal_attacks#Examples. ¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸ 23:26, 15 August 2005 (UTC) Posted by User:Sam Spade.

What? Sam, please read my posts before you try to respond to them. Exploding Boy 01:14, August 16, 2005 (UTC)

• Sam is correct though, blocking for sexist remarks is not currently policy. Of course such blocks are made from time to time under the 'general disruption' clause. It may be desirable to some to either formalize or forbid that practice. Also, note that people can be very unpleasant and attacking without being discriminatory, e.g. by calling other editors moronic imbecile trolls. Radiant_>|< 08:15, August 16, 2005 (UTC)

If you'll both read my post carefully you'll see that I never said it was current policy. And frankly, I find it rather distasteful that Sam would object to the creation of such a policy in the name of "discriminatory special rights." The very phrase is objectionable. Exploding Boy 22:35, August 17, 2005 (UTC)

• No, the point here is that personal attacks should be blockable whether or not they're discriminatory. If it's limited to that, it opens the door to affirmative action POV pushing. Radiant_>|< 09:06, August 18, 2005 (UTC)

I think what EB is getting at is that discriminatory attacks are often not personal, or directed at any one person. See the talk pages of any high profile Muslim article or the Administrators' noticeboard on any given "Jew-Wiki-admin-cabal" rant and you'll see what I mean. Personally I wouldn't mind including that kind of thing in here, but if I'm reading WP:NPA right, they are not covered under NPA. Dmcdevit·t 09:16, August 18, 2005 (UTC)

A case-in-point

Since the real problem with this proposal has to do with definitions of personal attacks, I think, I would like to point out a discussion I noticed elsewhere. Recently, on the Village Pump, Silverback suggested that we should delete (or at least stop linking to) the wrong version, on the grounds that, in his words, "Whenever it is cited it is invariably an insult and personal attack by the person citing it." In response, TenofAllTrades states: "Please don't make blanket personal attacks in forwarding your arguments—you don't win hearts and minds that way." Both make comments that, if wanted to, could easily be interpreted as personal attacks; and it is not too much of a stretch to assume that in case like this, both of these people could easily find at least three 'uninvolved' people who would back up their interpretation.

Now, that was just an example, and I don't really care who is making personal attacks there and who isn't. The problem is that it seems fairly obvious what would happen under the existing rules--each person would get warned, probably (human nature being what it is) by someone whom they're not friendly with in the first place, or by someone friendly with the person they're in dispute with; they would object, and the discussion would be even uglier than it is already. Once one or the other had accumulated three warnings, an Admin would have to take a look at it and make a decision about blocks--a decision that would be almost certain to be controversial, with more arguments and personal attacks following.

It is also fair to assume that Silverback, for instance, would go around warning everyone whom he saw citing the wrong version in what he saw as an offensive manner, assuming he meant what he said about him seeing it as a personal attack; if any of the people he warned over that were remotely hotheaded (and some of them are certain to be), it seems likely that his warning could provoke sharp words and possibly even personal attacks of its own.

I go over this mainly to give people an idea of how this policy would work out; it sounds good on paper, but looking at a discussion like the one I linked to (or any of the dozens like it that appear on Wikipedia each day), I think it becomes clear how it would play out in reality. Rather than encouraging people to move past personal attacks so they can get back to real discussions and making an encyclopedia, this policy would encourage them to delve into the minutiae of those attacks, wasting even more time and energy discussing, analyzing, arguing over and policing them. It might discourage users from hitting each other over the head with words, sure, but only by giving them a two-ton hammer to hit each other over the head with instead.

I think a far more constructive alternative would be to create a "Wikipedia: Be Thick-Skinned" policy. While personal attacks are certainly bad things, confronting an editor who has already blown their top with a list of warnings from people they've disagreed with in the past and a brief ensuing ban does not seem to me like something likely to make them cool off any time in the next century. Aquillion 00:36, 16 August 2005 (UTC)

"Silverback suggested that we should delete (or at least stop linking to) the wrong version, on the grounds that, in his words, "Whenever it is cited it is invariably an insult and personal attack by the person citing it." In response, TenofAllTrades states: "Please don't make blanket personal attacks in forwarding your arguments—you don't win hearts and minds that way." Both make comments that, if wanted to, could easily be interpreted as personal attacks;" Really? I woudn't have read it either comment as a personal attack. They're not being all lovey-dovey to each other, but they're not breaking policy that I can see either. Maybe I'm already thick-skinned. A lot of people take NPA to mean all sorts of interesting things. being impolite is not a personal attack. neither is having a "tone" that another user doesn't like. Mostly it seems that people see attacks when they're filing RFC's on another editor. I've seen an edit "revert" cited as a personal attack. I think having a clear definition would help alleviate some of this, but I'm sure that some will still cite all sorts of stuff that doesn't qualify. I like teh simple "any comments on the person's character is a personal attack. Any comment on the person's behaviour as an editor is not." Then maybe some of the more common mis-diagnosed "attacks" could be listed on the policy page to clarify that it isn't. i.e. "Reverting your edit is not a personal attack". FuelWagon 01:16, 16 August 2005 (UTC)

That bit about how people behave when filing an RFC was sort of what I was getting at. Since complaints seem, in some ways, like RFCs that can have the immediate satisfying effect of getting your target temporarily blocked (and allowing you to condescendingly to bring up in future conversations that they were once blocked for personal attacks), without the need of convincing anyone more than two other people and a single admin, it seems fair to assume that they would attract even worse feelings. Although I wouldn't characterize any of what either person said in the exchange I pointed to as a personal attack myself, it wouldn't be such a stretch for someone who wants to find personal attacks: In the next line after what I quoted Silverback talks about "a culture of mocking entitlement and hubris among the admins." Any admin who he had been in a disagreement with him over that topic before could legitimately see that description see as being aimed at them. In the line after that he says that "it is an insult to all of us that take wikipedia seriously," implying that there are people who don't take Wikipedia seriously, including everyone who disagrees with him on this and particuarly including, again, the group of admins that he has issues with over the blocking policy.

In his response, TenOfAllTrades make it clear that he, at least, does think that some of these things are personal attacks; and he notes that people whose actions trigger the page protection policy are often "behaving like petulant children", a description that--remembering the history here again--could easily be seen as aimed at Silverback.

If this sort of lawyering over personal attacks makes your teeth hurt, well, that's part of my point, too. You could expect to it every day under the proposed policy. There would even be campaigns (covert if necessary) to catch specific people using personal attacks... heck, there are practically some such campaigns already, and they don't even bring a guarenteed return yet. These things would force all sorts of issues before the admins; admins being human, this would result in at least some bad blocks and many, many, disputed ones (it would inevitably cause far more disputed admin decisions, I think, than any policy to date.) All of these things would create lasting bad feelings, escalating what could have been a one-time exchange of insults if the people involved had obeyed a hypothetical W:BTS into a nasty feud involving several other people. Aquillion 05:40, 16 August 2005 (UTC)

• I think that you're overstating the issue. Many people have broken the 3RR at some point, and this is rarely brought up again ("don't listen to him, he's broken the 3RR last month") and if brought up aren't taken very seriously. If there was a campaign to catch someone on using personal attacks, then that person could easily avoid the issue by not making personal attacks. It's really not that hard to be reasonably civil. It is already the case that some admins occasionally have their logs tracked to counteract their bad decisions; this issue is avoided by said admin learning not to make bad decisions. It tends to work out. Radiant_>|< 08:25, August 16, 2005 (UTC)

so I think the problem with RFC's is that they're so damn subjective that all it really takes is someone to make the accusation of violating "Be Civil", get someone else to sign it, and you've got an RFC that sticks. We're not going to change people. But we can change teh policy so that people can't use it that way anymore. If a "personal attack" is clearly defined, I think the amount of lawyering, the amount of "campaigns" to ding someone for violating NPA, would significantly go down. A directed comment about another editor's personal character is a personal attack. An undirected comment is NOT a personal attack. A comment about an editors editing and posting behaviour is not a personal attack. (comments about trollishness relates to their editing behaviour so is not a personal attack. It may or may not be true that the person is a troll, but editors must be able to expres their opinion about other editor's behaviour on talk pages and article edits. Comments about behaviour outside of wikipedia can be personal attacks. "You suck ****. So, what this results in is three different categories: Comments on an editors behaviour as an editor that that are protected, comments about an editors personal character which are personal attacks, and a gray area that would require an RFC or something to resolve. The tricky ones really need to have some sort of due process. Even a guy ticketed for speeding can challenge his ticket in court. An individual admin who is not involved with the situation should be able to block someoen or specific directed comments against another editor's personal character. FuelWagon 14:43, 16 August 2005 (UTC)

Enforce existing policy

We already have Wikipedia:No personal attacks and Wikipedia:No personal comments. I'm not sure that we need to expand the blocking policy, but I do think we need to make it clear as a community that these things are not to be done. If nothing else, this are disruptive, distracting us from writing an encyclopedia. ¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸ 01:32, 16 August 2005 (UTC) Posted by User:Sam Spade

Excellent point - if editors made a habit of striking out attacks with an edit summary of WP:RPA, we'd hope to soon have less attacks.
brenneman^(t)(c) 05:48, 16 August 2005 (UTC)

• I would be happy to hear other suggestions on how to make it clear that we don't want personal attacks here. WP:RPA is a possible suggestion, only it's pretty controversial because it is sometimes abused to e.g. remove votes, and because some people have a tendency to reinstate their attack and add a snide remark at the editor removing them. Radiant_>|< 08:09, August 16, 2005 (UTC)

Well, one big problem is alot of people have no clue what is and is not a personal attack (which is weird because their pretty well defined at WP:NPA). One way to resolve that would be to adjust the RPA policy to only allow for users not emotionally involved to remove the "personal attack". I have seen an editor remove an entire thread and threaten a nube editor, simply because they didn't like being contradicted, disapproved of the tone, or some such. I think we can all agree that in most cases, there is a positive correlation between perceiving a personal attack, and feeling a comment is contradictory to ones own sentiments. The only thing worse than rampant personal attacks, in my estimation, is popular old timers deleting valid comments from newcomers, simply because they are a bit "rough around the edges". ¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸ 14:55, 16 August 2005 (UTC)Posted by User:Sam Spade

• Okay. Would you say, however, that admins have a clue as to what is (or is not) an attack? Since they'd be the ones doing the blocking. Radiant_>|< 15:11, August 16, 2005 (UTC)
  • Um... no, I wouldn't be able to say that... - brenneman^(t)(c) 15:27, 16 August 2005 (UTC)

Admins are presumed to be trusted, else they wouldn't be admins. Obviously people make mistakes, but anybody with a problem with an admin's actions can dispute their actions. · Katefan0^(scribble) 15:35, August 16, 2005 (UTC)

See Aaron, I think you are a little backwards here. Tony was using RPA, not blocking. That's what's wrong with RPA: the involved "attacked" party can do it, often goes overboard, removes the essence of a comment, even votes. That's one of the good points of this proposal, the involved partyis not allowed to block. We can certainly trust an impartial admin to make the call and see through false warnings. Dmcdevit·t 20:22, August 16, 2005 (UTC)

I trust impartial admins just fine, but in my experience the involved party goes and finds another sympathetic admin, who acts as their proxy to protect the page, do the blocking, or what-have-you. Their are hundreds of admins, and nothing special about them other than that a potent clique or 3 didn;t notice and mobilize against their RFA. Supposedly admins are janitors, how much do you trust your janitor? What we need is counselors, cops, and... lets face it.. net-nannies ;) ¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸ 20:45, 16 August 2005 (UTC)

Hm. I don't much like your generalizations about admins (you don't think any were promoted on good merit?), but of course the possiblility of a "sympathetic" admin's intervention is a real one. But I don't think it is a much of a problem as you make it out to be. I don't think 3RR, working on much the same principle, does. Perhaps having a central noticeboard like 3RR where uninvolved admins can watch would lessen the possibility of sympathetic ones acting first. Dmcdevit·t 20:59, August 16, 2005 (UTC)

I'm strongly supportive of the idea of a separate noticeboard. Beyond helping to remove any "sympathy" issues, it also sends a message that personal attacks are taken seriously here. From the first I became aware of the 3RR noticeboard, it seemed to me to be a message that 3RR's were really, really bad, whereas other stuff (like violating WP:NPA) was sort of smelly but really only got a "tsk tsk." · Katefan0^(scribble) 21:06, August 16, 2005 (UTC)

I like the noticeboard idea too, for alot of reasons. If nothing else it builds on what we already have, and thus would be easy to implement. ¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸ 22:54, 16 August 2005 (UTC) Posted by User:Sam Spade

• To clear up any confusion, the question was do admins recognize an attack. Without prejudice, I provided the only time my comments have been reomved as an attack. Suppose an uninvolved admin had warned me for that edit... is there any recourse? Could I find my sympathetic admin to roll back my DefCon? And as to WP:RPA I support striking out but not removing all but the most extreme attacks. It makes it disapproval clear, along with the shame factor of your comment remaining visible but struckout. It's also harder to justify reverting the strikeout when your comment itseld has not been deleted. - brenneman^(t)(c) 00:21, 17 August 2005 (UTC)

The problem with that on its face is that there's nothing that shows specifically that someone else struck their comments. Because for the most part the commenter is the one who strikes out his own text in a normal situation, it appears as if they have done it. If WP had some kind of red-colored strikeout for an admin action it'd be different. Given the confusion factor, I'll still have to support just leaving them intact. · Katefan0^(scribble) 13:09, August 17, 2005 (UTC)

• • I think that, if there's going to be a central page for this (e.g. WP:AN/NPA) then that would make any admin's actions on this front easily examinable by the community. That would mean that if an involved admin makes a biased judgment call, the action will be seen and undone if others disagree (and the admin taken to task for using his powers when he's not an uninvolved party). This may likely lead to a view discussions in the first month, but that is good as it sets precedent as to what is and isn't acceptable. Note that currently, admins already sometimes block for personal attacks, and go mostly unnoticed. Radiant_>|< 09:14, August 17, 2005 (UTC)

I think that Sam's comments about admins are disingenuous, and smack rather of jealousy. It's clear that Sam would desperately love to be an admin, and tries to behave as much as possible like one, but in the end he just doesn't get it. I find it highly ironic and slightly amusing that Sam, who accuses admins of impartiality, cliquism and inappropriate behaviour whenever he gets the chance, and has been accused of making personal attacks more than once, is even discussing this at all. Exploding Boy 22:40, August 17, 2005 (UTC)

Oh, come on! I find it highly ironic that no one was making personal comments not related to the merit of this policy until you. Now it doesn't matter who is here discussing, as long as there is good constructive discussion. Dmcdevit·t 04:18, August 19, 2005 (UTC)

Stop now. Everyone. Take two WP:CIV and go back to the issue. - brenneman^(t)(c) 04:41, 19 August 2005 (UTC)

Yes, I think you said it better (simpler) than I. Anyway, back to your concern. It is a real one. But I think are some reasons this policy would avoid the problem. For one, I think the reason your comments were removed was not so much because that admin isone we can't trust to interpret personal attacks, but that he was the target, and so interpretted it differently than he might have if he was uninvolved. Now, you have to make three personal attacks in a week's time to be eligible for a block (note: not require). So if there are three warnings from they are all borderline, the admin is unlikely to block. Also, the policy requires warnings from three separate users, making three petty warnings unlikely. I think the point of this blocking policy is that it should be used only when needed to diffuse the situation, and keep it from spiraling away. That's all 3RR is for too (ie, rehabilitation, not revenge). So, I think this wording is probably lenient enough (at least 4 separate sets of eyes and all uninvolved, and even more with a noticeboard) to make misinterpretation unlikely. Dmcdevit·t 05:02, August 19, 2005 (UTC)

Except for wishing you'd put perceived before your first use of attack, I agree with what you're saying. Your arguments about leniancy upwards have swayed me even more towards the "no new policy" feelings I've expressed below. Blocks for vandalism don't require the bureaucracy we're proposing here, just make {{subst:NPA}} templates and start using them. Use two or ten, I trust admins (despite my above observations) enough to use their judgement wisely.
brenneman^(t)(c) 05:31, 19 August 2005 (UTC)

Oh and you don't hold those personal attacks I pasted all over the top of your talk page against me? Dmcdevit·t 06:00, August 19, 2005 (UTC)

after being warned

The proposal still says "after being warned". I don't know if I can edit the proposal or not. (I put something up a while ago, and it was moved soon after) Anyway, I would rather the proposal be that the first offense is an official warning (a 1-minute block) that is logged and tracked. And after that, an admin has the option of warning or blocking any offense that occurs after the first warning. I'm not sure why an admin would block an editor "after being warned", unless the editor violated NPA again, in which case, the first offence could get an immediate warning, and the second would get a block. Otherwise, if an editor violates NPA just once, and is warned, could an admin still decide to block the editor for the same attack that they were warned for? If so, what's the point of the warning exactly other than to hold an arbitrary judgement over the editor's head? "I've warned you, and I'll block you if you don't behave". I would rather see (attack) (warning) (any arbitrary amount of time) (attack) (block) rather than (attack) (warning) (some arbitrary amount of time) (block). Does anyone see what I'm trying to say here? FuelWagon 23:11, 17 August 2005 (UTC)

• That's not really a good idea. First, it makes it so that only admins can warn, which is inappropriate. Second, logging and tracking can be done on a Wikipage anyway. Third, one-minute blocks are more trouble than they're worth. As it currently reads, people can be blocked for making personal attacks after being warned. That means that they must make additional attacks after the warning. Not just that other people may read it later. Radiant_>|< 09:44, August 18, 2005 (UTC)

Well, one of my points of contention is specifically that in the standard case of two editors going head-to-head in a heated debate, each telling the other "you have been warned, and I'll go tell if you keep it up" will make make matters worse. Making it part of the official process turns it into a threat from one engaged user to another. I think you grossly underestimate how this will make fights worse, not better, and make it even more catty than it already is. It would be like writing a "road rage" law saying someone must give a person written warning that they consider their comments to violate road rage laws before the police will get involved. So, two guys are at each other's throats on the highway, and one guy shoves a "warning" in the other guys face. This is escalation, not resolving anything. Don't make the people involved be part of policy enforcement like that. Someone breaks policy, report it. Leave it to an admin to decide and enforce. make it a one-hour block if one-minute is too short. or go straight to a full block. But if a fight has flared up, do NOT make the people involved part of the enforcement process. the point is to DISENGAGE them, not make it worse. FuelWagon 20:01, 18 August 2005 (UTC)

• The problem you perceive would be solved if we were to state that all warning and blocking has to be done by uninvolved people, don't you think? Radiant_>|< 12:04, August 19, 2005 (UTC)

No, that means someone will just ping his buddy to drive in and do the "warning". It still escalates the issue among the editors, rather than diffusing it.

Look, it's very simple. An editor does an attack. If the editor has never been blocked before, ever in his life, the admin gives the editor a warning. After that, all attacks get blocked. Editors involved in the dispute are not required to engage the attacker and escalate the situation by giving them an "official" warning. Nor are they incentivized to ping a friend and have them do it as an "uninvolved" editor.

What this means is worst case, the first attack goes unpunished with just an official warning given by an admin. After that, an admin can block every violation, without pulling involved editors deeper into battle.

This is actually MORE rigid from a punitive standpoint, but it involves LESS editors.

And if you are really concerned about a wave of new editors using up their free "npa warning" without getting punished, then drop the warning completely and have the admin block.

If you get attacked, disengage that editor and report it to an admin.

the system you are proposing is if you get attacked, further engage the editor by giving "official warning", report it to an admin, and then the next attack gets blocked.

Do you see that's no different punitively than if an admin gave the first warning, but from the point of "diffusing the fight", it is much better. There is no incentive for editors to go around engaging their attackers to give them warning so they get blocked next time. FuelWagon 20:26, 19 August 2005 (UTC)

• No, there is a difference. Your scenario hinges on the idea that any semi-disruptive editor can "ping" an admin to do blocks. That is not in fact the case, most admins are more neutral than you would suggest. Radiant_>|< 09:31, August 22, 2005 (UTC)

Sigh

Gawd. This is how it goes: someone calls someone else something near the knuckle. Person attacked goes to admin friend and gets caller blocked. Meanwhile, other admins merrily describe other users as fuckheads. Bitter wrangling ensues.

Here's a good proposal. Try it out. Just ignore it when someone calls you a name. No need to run to mummy. Just get over it and move on. Yes, we'd all rather people didn't, but the bickering over personal attacks is probably more damaging than any name could be. -- Grace Note

"Sticks and stones..." -Splash 19:48, 18 August 2005 (UTC)

Sometimes its hard to ignore. I'll give what I think is the consumate example, Talk:Arthur Wellesley, 8th Duke of Wellington. Comments like:

You are completely disgusting, you know that? The original version of the image which you uploaded didn't have the caption on it. The version I put up was exactly the same as the image which you originally put up and said was fine to use so long as we didn't defame the Duke. You do not get to win an argument by default. Until you put back up the version of the image without the caption, I am going to keep the image out of the article. john k 7 July 2005 15:18 (UTC)

and

I said on the description page that if you didn't like it, I would delete it, you asshole. You are a completely absurd person, fuck you. john k 7 July 2005 15:05 (UTC)

directed towards User:Richard Harvey, Photo Archivist at the Duke of Wellington's Regimental HQ helped cause the wikipedia to lose out on ALOT of images the regiment had donated, and a great deal more they had been intending to donate. So, sometimes personal attacks can be a rather serious matter. ¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸¸,ø¤º°`°º¤ø,¸ 00:42, 19 August 2005 (UTC) Posted by User:Sam Spade

• To me that would be a clear block under disruption (says the backseat armchair admin groupie.) It sound from these arguments that there may be no need for a new policy as opposed to a central notice board combined with use of existing policy. And that could be done straight away. - brenneman^(t)(c) 04:54, 19 August 2005 (UTC)
  • Well, yes. As should be obvious from this proposal's main page, blocks for personal attacks already happen from time to time, generally under the disruption clause. These blocks are sometimes contested on grounds that personal attacks aren't disruptive per se. We basically have three options. 1) keep the current situation. 2) clearly establish that personal attacks are not disruptive. 3) formalize that personal attacks can be grounds for blocking. In case of #3, a procedure may or may not be useful. Radiant_>|< 12:07, August 19, 2005 (UTC)
    • Wouldn't it be much simpler than to revise WP:NPA with "Personal attacks are often considered a disruption of Wikipedia." We go on as before, but with the creation of a notice board, more effectivly? I'd prefer to avoid instruction creep, tons of votes, and the almost inevitable bad blood that a long policy discussion would raise. - brenneman^(t)(c) 04:36, 21 August 2005 (UTC)
• Well, yes. The status quo is that people can be, and sometimes are, blocked for personal attacks. It would be easy to set up WP:AN/NPA and let people report serious abuse. Radiant_>|< 09:31, August 22, 2005 (UTC)

Back to the main point

Okay, time to get back to the main point here. Some people will always agree with a policy on blocking for personal attacks, and others will always disagree. There is sufficient support that I feel warranted in making this an official policy proposal and putting it up for a vote. However, before this happens, I would like to hear comments and amendments to the exact wording of the proposal, so as to make it least likely to be abused and most likely to be acceptable to the community. Radiant_>|< 12:10, August 19, 2005 (UTC)

I added a little more on how it's not an entitlement, adapted from 3RR. Can I suggest that we widely publicize this now so we get as many more people in on the discussion beforehand as possible. That is what will most likely make it successful. Dmcdevit·t 16:37, August 19, 2005 (UTC)

• Yes. Excellent idea. Radiant_>|< 09:31, August 22, 2005 (UTC)

comments/ammendments/options: (1) drop the warning completely. (2) require the first violation get a warning. (3) allow an editor to give the warning.

While I'm thinking of it, if Alice attacks Bob, Charlie warns Alice, no further attacks occur. Then a month later, Alice attacks Dave. Does Eve have to warn Alice before an admin can block her on her next attack? How do you keep track of Charlie's warning from the month prior? FuelWagon 20:35, 19 August 2005 (UTC)

• I would suggest that any warnings "time out" after a week. Bygones. The intent here is not to block as many people as possible; the intent is to discourage people from attacking. Radiant_>|< 09:31, August 22, 2005 (UTC)

The proposals look great and we need something like this adopted ASAP. I would suggest to add to the "super super simple" and "super simple "proposals, a tag that is added to the user page that will displayed text such as "you have been banned for XX hrs due to violatiing the No personal attacks policy", or similar. That tag alone in a user's page during the 24 hour period, is quite a stigma and will be a much better deterrent than just a block which will be only visible to the banned user. ≈ jossi ≈ 04:01, August 21, 2005 (UTC)

Uninvolved

In general, I believe that policies that require an "uninvolved" admin should be considered harmful because it is an area that chronic problem users learn to exploit. The problem is that there are a very limited number of admins who are a) willing to deal with behavior problems, b) active on any given day, and c) following a particular dispute. Any policy requiring "uninvolved" admins could effectively allow admins only one "adverse action" in the lifetime of a user, because after that, the admin is involved. The number of admins who are able and willing to deal with a chronic troublemaker then becomes vanishingly small over time.

I believe that "uninvolved" restrictions should be more clearly worded to make it obvious that past casual contact or disciplinary action taken against a user does not necessarily make an admin "involved." The Uninvited Co., Inc. 02:34, 26 August 2005 (UTC)

Could you tell me (and this is an honest question, even though it sounds leading written down) how you feel about the "uninvolved" admin in 3RR? I tend to view these two as analogous. Dmcdevit·t 05:58, August 26, 2005 (UTC)

I don't feel as strongly about the "uninvolved admin" in 3RR because it is widely interpreted as an admin who is uninvolved in editing the content of the article. Though there are still accusations, the context makes it clear. Since personal attacks tend to occur in discussions whose scope is not limited to particular articles, I find the language more troubling here. The Uninvited Co., Inc. 13:31, 26 August 2005 (UTC)

'uninvolved' would at a minimum mean that the admin is not the target of the alleged violation. A violation would be a single, specific edit on a specific page. if an admin hasn't edited that page for some period of time before the alleged violation, (days, maybe a week?), then the admin can be considered 'uninvolved' in that incident. Once an admin accuses an editor of violating NPA, whether found guilty or innocent, warned or blocked, that admin should probably be considered "involved" for a week after the warning or a week after the block is removed, and that admin shouldn't block or warn the same editor for that time on any page. If you ever get pulled over by the police, when you get back on the highway, the police turn around and go the other way or they take the first exit. They leave the driver alone. It isn't a perfect definition of "involved", but at least it's completely objective. and objective is good when emotions flare. FuelWagon 16:18, 26 August 2005 (UTC)

I like the idea of an objective defintion, but I fear that the more defined it is, the more likely it is to being gamed. For example, what if a user makes a personal attack on WP:AN/I? Would dozens of admins be considered involved because they've recently edited that page? What if a user declares "All Wikipedia admins {fill in personal attack}"? Would all admins be considered involved as the target of the attack?

(chuckling) uh, if someone takes it upon themselves to insult 1,000 admins so as to make them "involved", then it would be worth the day delay to get an uninvolved admin to lay down 1000 consecutive 24-hour blocks. The time in between would be great entertainment value. The only "gaming" i can see is that Alice insults Bob. Charlie comes in as an admin and issues a "warning". Alice then insults Charlie in an attempt to get charlie "involved" so he can't block her. This is what I was talking about for the "warning" problem. How do you warn someone for violating NPA and then go back and block them for it at the same time? That's why I'd prefer the cleaner 1-hour warning block for first offence, and up to 24 hours for any violation after. This means Alice attacks Bob. Charlie gives Alice a 1-hour warning block and should walk away. If Alice then attacks Charlie, admin Dave can give her an immediate 24-hour block and walk away. This means you would need at most 6 admins to keep a relentless attacking editor on revolving block, but most editors will either learn after the first two or three or the problem should get kicked up to arbitration. The point is to keep the low-level implementation as objective as possible and as free from abuse as possible so that admins can implement it easily and fairly and accusations of admin-gaming by the blocked editor are easy to disprove. If you design the system with objective measures to that an admin cannot easily game the system, then the system can be allowed to operate with little oversight from arbcom. That's the point. Unload the work from arbcom by designing the sytem right. If this fails to stop an editor, you can still take it to arbcom for a temporary injunction until they reach a decision. But then if the system is designed well, you shouldn't have to worry about having to investigate charges of the admins gaming the system. This is a standard reply to all admin actions. If you design the system with objective measures the make admin gaming difficult if not impossible, then arbcom won't have to spend its time investigating charges of a group of admins gaming the system. FuelWagon 17:27, 26 August 2005 (UTC)

It might be more workable to have uninvolved simply defined as "Not the specific target of a specific attack". This would be less prone to gaming the system or wiki-lawyering, two things which often follow personal attacks. Carbonite | Talk 16:46, 26 August 2005 (UTC)

Hm. Good points all around. I guess all I want is an admin that 1) is not the target of the attack (obviously) 2) is not in an active dispute with the offender and 3) has not made a disputed or controversial block of the same offender before. If an admin has blocked someone before in an uncontested, clear-cut case, than they are probably still uninvolved in my book (especially if time has passed). If an admin has approached a person about personal attacks, or even issued them a (undisputed) warning, I still wouldn't really call them involved. I do think that this kind of block is much less urgent than a 3RR, just because of the more long-term and and not immediate outcome. So, while a NPA noticeboard will help, I don't think we need to worry if it takes admins a day or more to get around to this. Dmcdevit·t 16:49, August 26, 2005 (UTC)

Actually, the recent examples of wikilawyering I've seen involved all sorts of evidence, diffs, and what not, that were used to support a rather subjective policy of "harrassment". If you take away the subjectiveness, then the wikilawyering evaporates. The beauty of 3RR is that it is specific, objective, and it can be quickly determined in a yes/no fashion whether it was violated. You don't see much wikilawyering around 3RR except perhaps to argue that one revert wasn't the same as another revert and stuff like that, but there again, that's a subjective interpretation, and that's were teh wikilawyering pops up. Making an objective policy for admins to enforce NPA that makes accusations of admin-gaming fall flat. And an objective system for individual admins still allows a subjective interpretation to be taken to arbcom if the need arises. The point is to design a system that works with little oversight, something that makes less work for arbcom, not more. Making it objective, and making it hard for a rogue admin to abuse means that arbcom can focus on problem editors. FuelWagon 17:35, 26 August 2005 (UTC)

Well, I added a definition for "uninvolved" to the proposal. FuelWagon 04:04, 29 August 2005 (UTC)

I removed this: "An admin shall be considered "involved" with that editor for one week after a warning is given by the admin and for one week after a block imposed by admin against the editor has expired. This post warning/block "involvement" shall apply to any page, not just the page that the personal attack took place." I wasn't sure what it meant, but it seems to say that, once an admin has warned an editor for something, or blocked that editor, the admin is regarded as "involved" with the editor and may not block them for a personal attack. If that's what it means, that's not a good idea. Admins who've already warned or blocked an editor are often the best people to continue dealing with them, as they're aware of the pattern of behavior. We want admins to become "involved" in that sense, just not in a conflict-of-interest sense. SlimVirgin ^(talk) 05:40, August 29, 2005 (UTC)

If every admin is perfect, that would be fine. Since every admin is human, then the reality is that once an admin blocks or warns an editor, they have become involved and are no longer unbiased. If an editor clearly attacks another user, then it should be easy enough to get another admin to do the second block. Once an admin performs a block, the user may be less than cordial to the admin but not actually violate NPA. If the admin can take an impolite statement and turn it into a NPA violation and block the editor again, then the admin seeing a "pattern of behaviour" is seeing it with bias. If the admin withdraws and calls in a new, uninvolved admin, and that admin sees a clear NPA violation, they can do the block. After three back-to-back blocks from three different admins, you now have sufficient firepower to get a temporary injunction from arbcom. But I can point to cases where an admin blocked an editor, the editor was impolite but didn't actually violation NPA, and the editor was blocked again by the same editor. Admins are human. Designing the system with the assumption that admins will remain neutral even after they've engaged an editor is setting the system up for failure. FuelWagon 13:58, 29 August 2005 (UTC)

• One week sounds reasonable. It is not automatically true that by blocking someone you become biased against that person, and setting a one-week bar stops people from gaming the system by claiming that the blocking admin was biased. Radiant_>|< 14:01, August 29, 2005 (UTC)

Radiant, I agree that not that all admins becomes biased after blocking an editor. The point is that any editor blocked by the same admin two or three times in a row will almost certainly generate a RFA against the admin for abuse of privledges. That the great majority of those RFA's will be baseless won't change the fact that it will generate a lot of paperwork for arbcom, and will generate a lot of bad feelings all around. The point is to make this clearly about policy, and to make it clear that nothing personal is going on between a single admin and an editor. FuelWagon 14:17, 29 August 2005 (UTC)

• Well, it sounds like a good idea like that (I was actually responding to Slim earlier but got an edit conflict). Radiant_>|< 14:31, August 29, 2005 (UTC)

I strongly disagree with that section, as it would change the whole way admins deal with editors. It's important that a blocking admin have knowledge of a pattern of prior behavior. As I said above, we want admins to become "involved" in this way. SlimVirgin ^(talk) 21:10, August 29, 2005 (UTC)

Ed Poor and I have a situation going with one editor, for example, where we're currently blocking him for every violation of policy, as a way of seeing whether he can continue editing while the arbcom decides what to do with him long term. There's another editor I'm currently approaching in the same way. FuelWagon is trying to turn an admin's valuable experience of an editor into inappropriate "involvement." I doubt very much if other admins would ever support this wording, and I fail to see any benefit in it.

Furthermore, editors blocked by the same admin do not "almost certainly" cause an RfAr to be generated against the admin for abuse of privileges, as FuelWagon claims. Can he provide examples? SlimVirgin ^(talk) 21:15, August 29, 2005 (UTC)

So, you and Ed are working as a tag team. Are you saying that only you and Ed would make the same blocks? Would a randomly chosen third admin disagree with your blocks? If only you and Ed would make the blocks that you're making, then that should be cause for alarm. If not, then there is little disadvantage for pulling in another admin. Does the editor violate policy as soon as the block expires? or does it take a while. a 24-hour block, followed by a 24 hour period before the next violation occurs means that you only need three admins in rotation to keep the guy blocked. If the guy does back-to-back personal attacks, and they're handed down by different admins every time, then arbcom should be able to issue a temporary injunction because 6 blocks from 6 different admins already reflects 6 different opinions on the matter, rather than 1 admin doing 6 blocks and having to investigate themselves. FuelWagon 22:18, 29 August 2005 (UTC)

• I think Slim makes an excellent point - an "uninvolvedness" clause could easily be misconstrued to game the system. Admins tend to have a lot of work, and do not generally call in other admins to review their work. This system usually works. For instance, I've become involved with a certain user who was uploading copyrighted images. I patiently explained our policy a couple of times, but his responses were not very pleasant and he kept uploading the images (and revert warring to insert them into pages). Eventually, I blocked him (for 24 hours) for repeatedly uploading copyvios after being told not to. His response was to call me an 3vil cabal member and threatening to complain about my actions.
• Anyway my point is that that situation could easily be construed as me blocking him for personal attacks while being involved with him. However. If he had, then any admin looking at it would have concluded that that interpretation was incorrect. So the question really is, do we want a policy that states "uninvolved admin" by some definition (which will allow the occasional troll to complain about it) or do we want a policy that does not (which may be feared as abusible by high-handed admins). Radiant_>|< 13:15, August 30, 2005 (UTC)

The system usually works. But when it fails, it fails miserably. If an editor violates NPA, then any admin should be willing to block them. If the only admin who sees cause to block the editor for NPA is one admin, then there is a problem. Handing the block decision off to another admin when you're already involved shouldn't be a problem. If the other admin decides "no, that really wasn't an NPA violation", or decides it wasn't strong enough to warrant a full block, then that means the reporting admin is biased against the editor, and the editor really shouldn't be blocked. having an objective definition of "involved" is like requiring police officers to get a warrant before breaking down someone's door to search their property. Yeah, the system usually works without a warrant because most police are operating on good faith. Yeah, criminals could "game the system" and go free because some requirement of the warrant wasn't followed. But there is a reason for warrants. You want to have a system designed so that hundreds of thousands of police officers can do their job day to day without every damn case going to the supreme court to get a subjective interpretation of whether or not the cop acted in good faith or not. Same with this policy. You want to design a system that will allow admins to operate with little oversight, will give editors the confidence that policy is being enforced neutrally, won't result in escalating problems rather than diffusing them, and won't end up generating a lot of RFAr cases as the number of editors and admins continue to increase. FuelWagon 16:56, 30 August 2005 (UTC)

Instruction creep

This policy is pointless. Admins should be doing it anyway. --causa sui ^talk 15:21, 27 August 2005 (UTC)

• If you say that admins should be doing it anyway, that means you actually support this proposal, doesn't it? The simple version of it, anyway. The easiest way to go about this is stating that since admins already block for attacks and go uncontested, the blocking policy should be reworded a bit. Codifying current practice is somewhat important for future admins. Radiant_>|< 08:17, August 28, 2005 (UTC)

Explanation of my edit

I deleted this: "In particular, to avoid potential abuse, an admin should never block for a personal attack if he is an involved party, or the target of the attack - just like admins should never block for revert warring if they are an involved party in the edit war."

There's no reason an admin shouldn't block if s/he's the target of a personal attack. It's often the case that admins block trolls and vandals and then have their talk pages attacked with personal abuse. We should be able to extend the block for this, or block again as appropriate. These sections about "involvement" seek to change the whole way admins relate to people we block, and the policy risks not being approved because of them. What's important is that we shouldn't be involved editorially with users we block, but so long as that's not the case, involvement can be a positive thing i.e. can mean simply that the admin has experience of dealing with that editor. SlimVirgin ^(talk) 21:23, August 29, 2005 (UTC)

I strongly disagree with that section, as it would change the whole way admins deal with editors. It's important that a blocking admin have knowledge of a pattern of prior behavior. As I said above, we want admins to become "involved" in this way. SlimVirgin ^(talk) 21:10, August 29, 2005 (UTC)

If it would "change the whole way admins deal with editors", then it might indicate just how much this sort of restriction is needed. Guilt should not be determined based on past convictions. If admins are using a "knowledge of a pattern of behaviour" to convict an editor, then that's like convicting someone of robbing a bank because he was in the neighborhood during a recent robbery and he had been convicted of three bank robberies before. Either the editor broke "No personal attacks" or they did not. If they did, get an uninvolved admin to make the block. There are hundreds of admins. It isn't like we'll run out. The very thing I would want to avoid is an admin who takes it upon themselves to accumulate a "knowledge of a pattern of behaviour", become super-sensitive to an editor launching personal attacks to the point where the admin blocks the editor for something that wasn't a clear personal attack. That sort of thing will generate an accusation of admin abuse, and it might be deserved, meaning that arbcom or someone will have to spend time and energy investigating every charge of admin abuse that gets generated out of this policy. If the editor made a second personal attack, the clearest and most objective thing to do would be to get a second admin to do the second block. and so on. That way, rather than charge a single admin with admin abuse, the editor will have to show that two or three editors abused their powers, and that's going to be a hard sell.

Well, I do think that it is important to know an editors history in order to determine whether certain things are good or bad faith in that context (though that doesn't really work for personal attacks). But I'm starting to think we should get rid of all this definition and other stuff like that. Not say that an admin can or can't be the target. If we put it iinto practice, I tend to think that good practices will evolve, and those that try to game the vagueness will be called out o it. Let's avoid instruction creep. Dmcdevit·t 21:40, August 29, 2005 (UTC)

There's no need to get other admins involved for a personal attack on an admin. If someone (as they recently did) writes "burn your ass, you c***" to me, because I protected a page, I want to be able to block him and not have to call for help, which would be a waste of my and the other admin's time. And if he returns with another personal attack, whether against me or someone else, I want to be able to block him again without involving others. If you want this proposal to become policy, there's no point in risking sabotaging it with clauses that increase the hoops admins have to jump through to get simple things done. SlimVirgin ^(talk) 22:26, August 29, 2005 (UTC)

Since several definitions have been proposed, I listed them all on the article page as possibilities. It seems to fit the different proposals, as in super super simple proposal, super simeple proposal, and simple proposal. FuelWagon 22:09, 29 August 2005 (UTC)

• I think that the main difference, per Slim's suggestion, is between an established user and a new account that has mostly abusive/vandalistic/trollic edits. If there is a long-standing grudge between two existing users, and one of them happens to be an admin, then he shouldn't be blocking the other for (perceived) personal attacks against him. How about rewording it to reflect that? Radiant_>|< 13:19, August 30, 2005 (UTC)

Why would this be unfair?

I'm beginning to wonder whether some of this is being written by someone with a personal interest in the issue. Why would this be unfair, for example? "Removing personal attacks is a disputed issue, because it involves editing other people's comments. While this proposal makes no claim either way regarding the validity or consensuality of removing personal attacks, it would be unfair to both remove a perceived attack and block someone for it."

If someone writes an obscenity on a talk page, are we seriously suggesting an admin has to choose between removing the comment and blocking the offender? SlimVirgin ^(talk) 22:44, August 29, 2005 (UTC)

"it would be unfair to both remove a perceived attack and block someone for it." I know editors who removed their own comments and were still blocked for NPA after the fact. maybe this should be clarified in the policy. FuelWagon 22:48, 29 August 2005 (UTC)

Well yes, you do. If a personal attack is serious enough to be blocked for, whether or not the comment being removed is sufficient should be left to the blocking admin to judge. You're trying to pre-empt people's judgment here. For example, if an editor immediately reverted his personal attacks without being asked to, I personally wouldn't block him. But if there were multiple attacks left on the page for days, and he only removed them when someone else asked him to, then I'd say it was blockable. SlimVirgin ^(talk) 22:56, August 29, 2005 (UTC)

• Okay... "If the attacking party has apologized, or his remarks have been stricken out or removed by anyone, then this can be seen as a mitigating circumstance". Or something like that. Better? Radiant_>|< 13:21, August 30, 2005 (UTC)

Please stop undoing my changes

FuelWagon, you're undoing every change I make to this page. Please stop it. I can add what I want to my proposal. SlimVirgin ^(talk) 22:58, August 29, 2005 (UTC)

No other proposal but yours advocates for its adoption. I simply removed the part that was attempting to say "this is the best proposal". I appreciate your neutrality in presenting the various options. whatever. FuelWagon 23:34, 29 August 2005 (UTC)

It doesn't say it's the best proposal, but explains the reasoning behind it. I'd appreciate it if you would leave it as it is. SlimVirgin ^(talk) 23:47, August 29, 2005 (UTC)

No other proposal "explains the reasoning behind it". Only yours. All the others simply say "the proposed definition for 'involved' is blah". You give a whole advertising campaign for your definition. FuelWagon 00:00, 30 August 2005 (UTC)

You wrote the other involvement proposals, so it's not appropriate for you to write mine too. There's no advertising campaign. It seems to me that you're trying to control the content here, and in a way that I believe does not enhance the proposal's chances of becoming policy. SlimVirgin ^(talk) 00:06, August 30, 2005 (UTC)

Just stop with your accusations for once. I didn't write them. I took the different definitions that everyone tried and put them in as multiple choice alternatives. and this "Admins are often the subject of personal attacks as a result of admin actions (e.g. blocking for vandalism, protecting a page on the "wrong version,") and therefore have to be able to block for personal attacks without being required to involve a second admin." does not define the proposal, it justifies it. It says THIS IS WHAT WE NEED. ADMINS HAVE TO BE ABLE TO BLOCK WITHOUT BEING REQUIRED TO INVOLVE A SECOND ADMIN. Yeah, lets present the multiple choice options and then tell everyone which one we HAVE to have. And you accuse ME of trying to control? You've skewed the entire ballot. FuelWagon 00:19, 30 August 2005 (UTC)

I agree with FW here, keep it barebones. But why'd we delete my counterproposal too? Dmcdevit·t 00:23, August 30, 2005 (UTC)

Dmcdevit, I tried to word your counterproposal in neutral/positive language, saying "involvement" would be left to the admin to determine. "instruction creep" cast the other options in the negative and the part about "gaming the system" is sort of a justification for that proposal. I may have failed in my translation. It was not intentional. [1].

'Sokay, that's what I figured. I just wasn't going to add it while you two were still reverting. Now, about your change... you made it into a perfectly good proposal that we should keep, but not what I was getting at (which may have been my fault). I just want to see an additional proposal there that we don't need to include a definition of "uninvolved" at all, and that community will through general practice should be allowed to shape it (like we have for many other policies, where it is assumed involved parties should probably not use admin powers, but not necessarily spelled out). Dmcdevit·t 02:59, August 30, 2005 (UTC)

Well, I'm not sure how you want to word that. I think what you're proposing is basically "at the admin's discretion" with the added piece that the community will determine what is acceptable. But I'm not sure. I would look at our current discussion to be part of the "community" trying to determine what is and is not "involved", so I figure why not iron something out now, and then the community can adjust it later if it doesn't work. Sort of like there was no definition before, and wikipedia has been using that approach, and now some editors in the community are trying to codify a policy that works. Or, it could be I'm not understanding what you're proposing here. FuelWagon 03:23, 30 August 2005 (UTC)

I think having the option to not include a definition at all would make it more fluid, rather than concreting the definition at whatever the community decides right now. I'd like it to be more flexible, and more interpretable in specific cases. Dmcdevit·t 03:40, August 30, 2005 (UTC)

FuelWagon, perhaps you could allow other people to write their own proposals (so long as they're not too long), and you could write yours? SlimVirgin ^(talk) 01:09, August 30, 2005 (UTC)

This is a wiki, isn't it? That's what you usually tell me. Hey, at least I tried putting everyone's proposal in. You just wiped it out and replaced it with your proposal. All I did was to try to present the proposed definitions without any advocacy and with just the facts and a neutral tone. I could modify the definitions to advocate for the one I want, but then everyone will want to advocate their way, and the definitions will get overwhelmed by campaigning. FuelWagon 01:15, 30 August 2005 (UTC)

• Both of you - take a deep breath, count to 10, and realize that you're both in good faith trying to create the best policy proposal. Okay? Radiant_>|< 13:23, August 30, 2005 (UTC)

Warrant

So, I view the idea of having an objectively defined rule for "involved" along with an objectively defined "cool down" period to be similar to the way a warrant works. A cop thinks someone broke the law. Before they can break down that person's door, they need a warrant. The point of the objective "invovled" policy is like a warrant. It makes sure that policy enforcment remains neutral. Yes, a warrant is an extra step and isn't needed when police are acting in good faith. Yes, a criminal could game the system and avoid punishment on a technicality if the police don't follow the warrant. but a warrant is a separation of power that has several intended consequences. First it keeps the honest cops honest. It will impede dishonest cops and possibly uncover them. It builds community trust in the law enforcement system. and it avoids the troubles that a single bad cop can generate in a whole population. Do you think wikipedia is served in the long run if good editors stay away because wikipedia is surrounded by charges of being a cabal? That policy enforcement is arbitrary? that admins go unchecked? I don't. I think that the damage done by a few bad admins creating the sense of arbitrary justice and an 'old boys' network to protect themselves is far worse than what would happen if good admins had to wait a week before they could block an editor again. That's what we're talkign about. Either give admins full discretion to decide if they're neutrally enforcing policy, have a few bad instances taint all of wikipedia, or impose a minor restriction that admins wait a week after they block an editor, have some editors game that system but watch charges of "cabal" evaporate from the community at large. This really isn't that restrictive of a policy. one week before on the page that contained the attack, and one week after. Are you saying that wikipedia would dissolve into chaos if this were implemented? Or is it a matter of admins who already have the power not wanting to give it up? FuelWagon 17:10, 30 August 2005 (UTC)

Policy in action:

If this is what gets an admin warning now, implimenting this policy will serve only to make things worse. I am thus, opposed.

Uh, Ed Poor has an interesting definition of what violates "No Personal Attacks". Remarks about an editor's behaviour on wikipedia should not be considered a violation of NPA. However, if Ed thinks said remarks will hurt another editor's feelings, then he may consider it an attack. So, I would say Ed is administering a very loose interpretation of what constitutes a personal attack. One of my reasons for supporting the requirement that admins distance themselves for a while after they do warn/block is to get multiple admins involved so that one admin's loose interpretation doesn't become a sequence of undeserved blocks against an editor. FuelWagon 13:18, 1 September 2005 (UTC)

The editors in question just want another way to block users who don't agree with them. The whole proposal is a waste of time. Bad admins are just about always endorsed anyway, regardless how fair or otherwise their actions are. Why don't SlimVirgin and EdPoor simply carry on doing whatever the hell they like without bothering with codifying it? What's the big deal? This sort of thing encourages friction. Admins will attack others with impunity. Others will find themselves blocked when admins who don't like them interpret their comments as attacks. There will be another noticeboard that features bickering. Editors such as Slim and Ed, who could be making useful contributions or bickering over content instead, will spend yet more of their time bickering over the rights and wrongs of what they've done. Just get over it. If you are so sensitive that someone calling you a turd really upsets you that much, you probably shouldn't be wandering around the interwebnet anyway. Someone who does nothing but call people turds can safely be blocked as things stand. There is no real need for yet more policies for bad editors to bash each other with. -- Grace Note

Trying to extract something from that broadside:

• You're thus supporting Wikipedia:Blocking_policy/Personal_attacks#The_non-proposal?
• You're engaging in a free kick at a couple of admins with whom you disagree?
• You're suggesting people should simply toughen up?
  brenneman^(t)(c) 06:53, 30 September 2005 (UTC)

Changes that don't require changes (to policy)

• Wikipedia:Personal attack intervention noticeboard

Created all the following above, it required no changes to policy. Make it clear (on the talk page i.e. through peer pressure) that an admin who is involved (however that is interpreted) is expected to place a username here as opposed to dealing with the individual themselves.
brenneman^(t)(c) 06:53, 30 September 2005 (UTC)

Given the somewhat cliqueish nature of the wiki admin community, it's very important that any sanctioning admin not be involved with any disputes with the other parties. Otherwsie a sense of piling on may occur Klonimus 09:42, 11 November 2005 (UTC)

A recent case on WP:ANI shows a possible course of events. User RingFoo is making attacks left and right. Admin BigCheese trys to talk to him about how it's not acceptable, and cops a spray of his own. Now BigCheese is involved, and thus would be wiser to not do the block himself, but to use WP:PAIN. I'd oppose anything that explicitly said they couldn't do the block, by the way, peer pressure should be enough. - brenneman^(t)(c) 11:40, 11 November 2005 (UTC)

Next step?

What's the status of this proposal? When is it going to be voted on?

I'd strongly support it, especially since it codifies existing practice. -- nyenyec ☎ 18:29, 21 December 2005 (UTC)

• Since it codifies existing practice, that means it's already a policy or guideline, and there's no need to vote on it. Policy and guidelines are descriptive, rather than prescriptive. Radiant_>|< 22:33, 21 December 2005 (UTC)