Wikipedia talk:Dealing with disruptive or antisocial editors/archive1

Initial comments by Mark Richards, Meelar and Isomorphic
Basically I like this - the only problem is that it still has a way to go in terms of specificity. NPOV violations may be obvious in the extreme, but much genuine angst is spilled over this by users in good faith. I worry about situations where admins are empowered to ban people for that. The same with many others on that list. However, basically it is much better than the other suggestion of 'sysops should do whatever they like'! I would support a variation of this. Mark Richards 16:07, 5 Jul 2004 (UTC)


 * Thanks Mark. Feel free to (a) modify it (b) popularise it. The silence is still a bit deafening! Perhaps you would like to expand the definition of what is and is not included. Examples could help.

I am a little worried however that this might be used to silence potentially legitimate criticism. For example, a proposal to limit sysop power could be stopped by two sysops blocking the user making the proposal. Mark Richards 16:58, 8 Jul 2004 (UTC)
 * not sure if i follow this concern Mark. Before two sysops could do anything, other than warn a user, they'd have to identify 10 non productive edits and find a third sysop that agreed with them. (feel free to delete this thread if you see what I mean when you reread the proposal). Erich 17:05, 9 Jul 2004 (UTC)


 * I basically agree with this as well. I think that the sysops have to get a little self-policing done; for instance, I would be willing to scrutinize these blocks, and I suspect you would as well. I think that individually, sysops are more likely to break policy than as a group. [[User:Meelar|Meelar (talk)]] 17:05, 8 Jul 2004 (UTC)

An interesting note: Should this policy be adopted, I believe it would be the first to give admins authority as opposed to merely technical ability. This is a departure from earlier Wikipedia policies. Note that I don't necessarilly think that this is a Bad Thing (see what I wrote here.) I do think we should be aware of it, though. Isomorphic 15:23, 9 Jul 2004 (UTC)

With the caveat above, I think this is a well-written policy. It covers many of the concerns I've had since I got here, and I especially like the warning process. This lets a user know that they have crossed the line, but gives them a chance to stop before any actual sanctions. Isomorphic 15:36, 9 Jul 2004 (UTC)


 * thanks Isomorphic. I have proposed a time line for progressing this. I deliberately want to wait before voting so we can move close to consensus. Does that time line sound reasonalbe?Erich 01:21, 10 Jul 2004 (UTC)

comments from Jallan
Disruptive or antisocial behaviour includes:
 * adding spamming links
 * The deletion queue is very popular because it allows quick judgements about whether an action be taken or not without going through a never-ending battle for or against one person's idea of TRUTH, JUSTICE and hirs OWN WAY. Suggest a "Delete inappropriate link queue" for the same reason. (Anyone could still remove a spamming link or other link they think inappropriate, but if it gets moved back because an editor believes it justified, such a queue would provide quick resolution instead of a long edit war and provide justification for sysops to remove link in future.) Removing links can be as unencyclopedic as adding links. Jallan 15:09, 9 Jul 2004 (UTC)


 * repeated violation of NPOV
 * personal attacks
 * spurious listings on Requests for Comment and Vandalism in Progress
 * blatant attempts to bait other users into a fight
 * I am very unsure about this one. Jallan 15:09, 9 Jul 2004 (UTC)


 * editing others posts on a talk page to change meaning
 * Should be "others' posts. Perhaps reword as "any change to another's statements on a talk page". Why should I be changing someone else's words at all, even if the resultant meaning is the same? Jallan 15:09, 9 Jul 2004 (UTC)


 * vandalism that is not severe enough to warrant immediate blocking under the existing policy
 * misuse of edit summaries
 * There should be rules governing proper and improper use of edit summaries. Maybe there are. If there are, there should be a link to that page. If there arent't, then a page should be created and there should be a link to it. Jallan 15:09, 9 Jul 2004 (UTC)


 * any sort of deliberate attempt to misrepresent the truth
 * This is weak. The word "sort" is not needed and I have doubts about "deliberate attempt". I think what may be meant is simply "misreprensation of truth" or perhaps "repeated blatent misrepresentation of truth", for example insertion of statistics that cannot be verified or without indication that other find them dubious. But I find any example I come up with seem also to be a breaking of "repeated violation of NPOV". For example, insertion of the statement "George Bush is really an alien from Pluto" would be fine if it were an assertion made by a particular group of people and the group was cited. Jallan 15:09, 9 Jul 2004 (UTC)


 * repeatedly adding contested material without discussion on talk pages
 * Perhaps too easily circumvented. Anyone can simply add to the talk page something like "Reinserting material for reasons already given" (and that can actually be quite adequate in many circumstances, especially where someone keeps removing material for reasons that others find weak or wrongheaded). Also repeatedly removing contested material without explaining is surely just as bad as adding contests, that is removing material that the editor contests (or it would not be removed) but others think should be included. But perhaps change to "* repeatedly making changes in areas that are being contested without discussion on talk pages". Jallan 15:09, 9 Jul 2004 (UTC)

Throughout these points I am bothered by the words "repeated" and "repeatedly". How many times does it take to make something "repeated"? Trolls love this kind of thing. Perhaps replace all "repeatedly" by "after being warned"? That is, a different warning for each kind of infraction and reword all so first time for each kind of infraction doesn't count, first time gets a formal warning only. This would apply to each kind of infraction. In effect a warned user would be (for a time?) on parole of having to be more strict about the matter than an unwarned editor. Warned users should be, for a time, flagged.

I've also seen moderated forums which had a demerit point system and moderators could, on being requested or by their own volition, give demerit points. If a user achieved a certain number of demerit points the user was short-term blocked or even banned, depending on a decision by those in charge. The points appeared beside the normal information about the user. Demerit points decreased of their own accord over time. I can't remeber details, but it may have been something like demerit points being decreased by the ceiling of 20% every month. On another forum it was simply: three strikes and you are out of that discussion thread. (Of course there were means to appeal in both cases.)

Perhaps some such flagging of problem users should also be implemented. Jallan 15:09, 9 Jul 2004 (UTC)

Thanks Jallan, I agree completly (I think) with all your comments. Feel free to edit the page acordingly. I think any merit or trust system may have trouble getting through tho, and would need a software modification. Erich 17:05, 9 Jul 2004 (UTC)

First, I haven't understood one important point - does 'user' in this case include an anonymous user? It might be a good idea to make that clear,

Secondly, and importantly, this is way too complicated. If you institute a policy like this then fewer sysops are going to bother tracking down vandals. Sure we might look for dubious edits to reverse, but we're not going to bother interacting with vandals if it takes a lot of work. We'll just leave it to someone else. That will mean more doubtful edits to reverse, and Wikipedia will get closer to the point where keeping it vandal-free is unsustainable.

But, you say, shouldn't we be very careful before we ban people? Shouldn't we be extremely sure that someone really is a vandal before banning them? Frankly no. Long complex legal procedures are needed where the effects of what you are doing are serious or long term. A 24hr ban from Wikipedia is neither. Sure there should be a warning; sure we should wait and see if the user is going to communicate. But if someone is continuing to disrupt Wikipedia then they should get banned.

Thirdly, just how many cases do we think we have where following this procedure would have prevented us giving a 24hr ban to a user who didn't deserve it? If its only a few, and especially if it was a short ban, then this isn't worth it.

Finally, we already have checks and balences in place. A sysop who bans too frequently should and will be taken to task for it, and if they persist will have their syop privileges removed. That's the right way to do it. DJ Clayworth 17:21, 9 Jul 2004 (UTC)


 * This policy is not intended to deal with simple vandalism (e.g. page blanking, or replacing text with gibberish). Rather, it's intended to deal with users who simply have problems editing constructively.  For example, look at the history of Nick Berg conspiracy theories in late May and Talk:Nick Berg conspiracy theories.  That sort of thing could be dealt with, while simple vandalism gets handled in the usual way.


 * As for your comments about checks and balances, well, I'm going to come out and say you're mistaken. As long as I've been here, I don't believe I've ever seen a desysoping. What I have seen is a lot of conflict, directed at sysops and between sysops, about the extent to which banning powers are being used.  Some sysops take it upon themselves to ban problem users; others disagree with their choices.  I'm not taking sides, but the community needs a set of procedures for deciding when bans are viable in cases that aren't simple vandalism, yet aren't acceptable behavior. [[User:Meelar|Meelar (talk)]] 17:32, 9 Jul 2004 (UTC)

I have made changes which I believe fit with Erich's general acceptance of my previous comments, going slightly farther in providing another way to remove warnings that were unjustified. This is based on the fact that the "Votes to Delete" works, in part because action must be taken if the accusation stands and voting puts it outside the necessity of accomodating one or two users who interminably refuse to agree with the consensus of everyone else. If a sysop is issuing too many unjustified warnings, it should become obvious when too many of that sysop's warnings are decided to be invalid by consensus. Any controversy that arises over warnings would be dealt with both openly and in timely fashion. A complaint that a sysop was being unfair can be responded too at once and it can be seen whether the complaint was justified, an empty accusation, or was somewhere in between.

I did not provide a link to any discussion of misuse of editing summaries since I don't know where that is defined. Someone please help, either by adding the link, creating a page, or removing that item.

Sock puppets are not covered here. Should use of concurrent sock puppets be allowed? They are banned on many moderated forums. There are few legitimate uses that I see for having two identities active in Wikipedia at the same time. I suppose if editor Foo has in the past been at loggerheads with editor Bar, and Foo honestly sees a reasonable change that ought to be made in an article which Bar has just been working on extensively, Foo might think it better to assume another identity as, say, Baz, to avoid the possibility of older hostility clouding Bar's vision. On the other hand, if Bar discovers that Foo is Baz, I think Bar has a legitimate gripe that Foo was acting to some degree in bad faith.


 * Using sock puppets is at best shady usage. There is almost always intent to deceive. If sock puppets are not to be condemned altogether, then at least it should be an offense for anyone to edit the same article under more than one alias within a 30-day period (or even to use two or more sock puppets at all in Wikipedia within an overlapping period of 30 days). Accidental anonymous editing might still be permitted. But that could be prevented by a warning dialogue box that would appears for any anymous editing saying something like: "You are performing an anonymous edit which will be identfied only by IP address. If you are a member of Wikipedia, please log in first, but otherwise you are welcome to proceed." If there is some legitimate and non-deceptive reason for sock puppets that I am missing, would that be covered by permitting sock puppets provided that the user page of each sock puppet by a user gave an open list of all such sock puppets in current use. Jallan 17:09, 10 Jul 2004 (UTC)

From "Wikipedia:Sysop Accountability Policy

 * Discussion carried over from "Wikipedia:Sysop Accountability Policy

Gedday all, you may be interested in this proposal: Dealing with disruptive or antisocial editors. Erich
 * Well intentioned, but the definitions of antisocial behavior are too subjective and diffuse and the procedures too cumbersome. The only way for this to have a practical effect is if groups of sysops pre-formed themselves into vigilance committees which we might dub Rolling Torquemada brigades. -- Cecropia | Talk 17:14, 9 Jul 2004 (UTC)
 * Agreed. See Wikipedia talk:Dealing with disruptive or antisocial editors DJ Clayworth 17:39, 9 Jul 2004 (UTC)
 * Note this proposal deals with borderline behavious, not vandalism.


 * Unfortunately justice aint simple. sounds simple... but isn't. Although quite frankly, cutting and pasting a template onto a couple of pages, and adding a list of URLs to the offending behaviour is hardly a big ask. It really is simple. Notice a problem user, warn them, tell others you're watching them, gather URLs of offending difs, warn them again if the problem persists, gather more URLs, everyone agrees, slap a block on. quick, fair, transparent. If one tree-hugging, whale kissing, bleeding heart sysops bleets "no they're not bad, they are just misunderstood", well the three of you gun toting no-nonsense types just need to pop down to the sherrifs office and pick up a couple of supporters, "five to 1" and stick on the block. If the user comes back, they are on a shorter and shorter leash.


 * and Yes the definitions are vague and subjective. Characterisations of human behaviour are, but feel free to clarify them further. Ultimately the subjectivity is resolved by putting the behaviour forward for consideration by a small group of sysops. If it's simple then... well its simple. If it is complicated, (and the people we are talking about will try to make it complicated) then more discussion and more votes are needed. The proposal provides a framework for the sysops to work through these issues in a logical and transparent way. By forcing the collection of the URLs to back up the decision it ultimately builds up a more robust and fairer set of boundaries.Erich (sorry forgot)


 * Please remember to sign your posts, so I know how I'm talk to! Thanks -- Cecropia | Talk 19:22, 9 Jul 2004 (UTC)


 * The whole policy is unworkable. I think the last thing we want is for sysops to have to band together to get things done (one side of it) or encourage them to band together (that's how you actually get cabals--the other side of it). If vandalism is fast moving, you don't have time to raise a posse, podnuh! If it's not, then we're thrown back into the perrenial problem of dealing with persistent bad behavior. The solution is NOT a kangaroo court (even for Australians. -- Cecropia | Talk 19:22, 9 Jul 2004 (UTC)


 * Ummm... at the danger of repeating myself... this policy is not for dealing with vandalism. go on, do me a favour will you? try reading it properly and then comment ? best wishes Erich 21:13, 9 Jul 2004 (UTC)


 * I have read the proposal, which is why I oppose it, however well intentioned it is. My point of vandals is that this is the greatest immediate threat at any given time, and there is no time to raise tribunals of sysops, even if this proposal called for it. In general I oppose blocks except for vandalism, and then they serve as a cooling-off period. The immediate nature of the block, reinstated if needed, tends to either cool vandals or encourage them to go away. This quasi-judicial process will accomplish what for a long-term user? We might get a Wik who is both driven out and allegedly returns to vandalize, or a situation as with 172 where he was nearly driven out (and he is one of the best contributors) but instead reached an accomodation after talk of banning or blocking was talked down.


 * I'm copying other these paragraphs to the Talk page of your proposal. -- Cecropia | Talk 22:18, 9 Jul 2004 (UTC)


 * There are plenty of non-long-term users who cause problems, yet are not strictly vandals. Look at, to name two instances, Phil Gingrey and Talk:Phil Gingrey, or the Nick Berg conspiracy theories conflict. These are not vandals, by any useful definition, and cannot be unilaterally banned under current policy. Yet they need to be.  These are the types of cases where a quick-moving yet non-unilateral policy is necessary. [[User:Meelar|Meelar (talk)]] 22:57, 9 Jul 2004 (UTC)

Power to remove frivouls vexatious complaints
This gives sysops too much power... (snip): Especially this: "If a sysop believes that a non-sysop's request is frivolous or vexatious they may immediately remove it." This seems to make sysops themselves totally immune to this process. Obviously, I'm going to consider any accusation against myself "vexatious." I'm dubious about the whole idea, but if we're adopting it I think it would be more useful to have a way for sysops to weigh in that the accusation was frivolous than to allow one single individual (even a sysop) to delete it. Also, there should be a fast track to banning for any individual who serially brings frivolous accusations of trolling or who vandalizes the Direction Page. -- Jmabel 14:20, Jul 10, 2004 (UTC) (/snip)
 * I've edited that the section. Is that better. Re the "serially brings frivolous accusations", well if they've been warned, then that counts towards their 'ten'... is that fair? Erich 22:58, 10 Jul 2004 (UTC)

Frequently raised objections
This section is for collective editing. Try to use a NPOV style. No need to sign, just chip in add stuff. (This is a bit of an experiment)

No
umm, well yes. It does give them more power. True. But it also makes them justify their actions and requires them to get peer review. In the event of controversy, at least 2/3s of the sysops must agree before action is taken. It seems most Wikipedian seem sysops should be using more power, and are prepared to have 'trolls' (whatever that actually means) banned on site! (see the voting at Dealing with trolls)

Yes
Yes, it gives them more power than they have every traditionally had. Certainly more than was granted to them when the community elected them to administer policy. 2/3 majority is not concensus, and this kind of thing will lead to factionalisation among sysops, with those hungry for more power drumming up ever more panic about a rising tide of trolling. I have yet to see anything that has caused much long term damage that this policy would fix.

This makes it too hard for sysops to do anything
A far as we (well me at least) can tell this adds to sysops powers and doesn't actually take any away (please point us at the policy giving sysop powers that isn't already mentioned in the introduction, if we've missed one)

This is too complicated
Unfortunately justice is not simple. It sounds simple, but isn't. Although quite frankly, cutting and pasting a template onto a couple of pages, and adding a list of URLs to the offending behaviour is hardly a big ask. The procedure really is simple. Notice a problem user, warn them, tell others you're watching them, gather URLs to offending difs, warn them again if the problem persists, gather more URLs, everyone agrees, slap a block on. This is quick, fair and transparent. If one tree-hugging, whale kissing, bleeding heart sysop bleets "no! they're not bad, they are just misunderstood", well the three gun toting no-nonsense type syspops just need to pop down to the sherrifs office and pick up a couple of supporters, "five to 1" and stick on the block.

This makes it too hard to fight vandalism
Response: huh? this has nothing to do with vandalism. Dealing with vandalism already addresses that.

The definition of disruptive and antisocial behaviour is too vague
Response:Yes the definitions are vague and subjective. Characterisations of human behaviour are, but feel free to clarify them further. Ultimately the subjectivity is resolved by putting the behaviour forward for consideration by a small group of sysops. If it's simple then... well its simple. If it is complicated, (and the people we are talking about will try to make it complicated) then more discussion and more votes are needed. The proposal provides a framework for the sysops to work through these issues in a logical and transparent way. By forcing the collection of the URLs to back up the decision it ultimately builds up a more robust and fairer set of boundaries

This is probably the killer. You end up defining 'good behavior' as 'behavior that I like'. I know it when I see it. The problem is that there will be disagreement over this. Engagement with a user, and eventual blocking if they do not engage constructively (that is not the same as come round to my point of view) will solve this.

This idea was developed by an Australian
(gedddayy, 'that's not a knife'.)

This is not addressing a real problem
Response: Have a long slow read through some recent Arbitration Committee requests to find lots n lots of examples of behaviour that it would have been nice to have curtailed earlier. (add your own favourite here) (I will be willing to name and show histories of several, just off the top of my head.) Could we have a few examples? I can't think of any clear ones. Phil Gingrey. Nick Berg conspiracy theories. Plenty of others, I'm sure...

Well, that's part of the problem - I don't think this is clear at all. From what I can see those users had strong opinions, and were pretty rude - they could probably have been temp-blocked for that, but I'd hate to ban them just for having somewhat extreme views. I think current policy can deal perfectly well with them. Mark Richards 20:56, 12 Jul 2004 (UTC)


 * I agree. Rudeness is the key prob. IMO, and not just of "trolls" or nutters. Well known and respected admins are rude sometimes too, and it NEVER helps improve article quality. Respectable debate and well reasoned, polite discourse is the only intellectually honest way to conduct ourselves here, if this is to be a legitimate source of reference materials. Sam [Spade] 21:55, 12 Jul 2004 (UTC)


 * I agree with Sam. Respectable debate and well-reasoned, polite discourse is the only way.  I've got no problem with anyone disagreeing with me; it's when they refuse to discuss or adress the issue that it starts being a problem.  I've had to keep an eye on Phil Gingrey for weeks, after trying again and again, along with other editors, to come to a reasonable solution.  There needs to be some defense against those who stonewall, insisting that their version is the only truth. [[User:Meelar|Meelar (talk)]] 22:00, 12 Jul 2004 (UTC)

oppose; comments
Can I just put a plea in to with simple opposition at this point. There will be plenty of time to vote on the final policy in two weeks! I'm really hoping that we can move closer to consensus first. In two weeks time we have something much closer to what you would like! Erich 05:40, 11 Jul 2004 (UTC)

comments from uninvited company
Erich & everyone,

I reviewed the policy.
 * thanks UC! you say thoughtful things as ever. You'll note my plea above (more for others than you), so I'll try to respond point by point in such a way that can pull us toward a middle ground.Erich 05:47, 11 Jul 2004 (UTC)

I oppose the policy on two grounds:
 * It is legalistic, that is, there are a considerable number of mechanical details and lengthy definitions. This is bad, because it isn't flexible and it doesn't encourage use of judgement and care.  Right now we have the patterns of a few recent troublesome participants clearly in mind.  These particular troublemakers, and the patterns of editing they follow, are unique to the present time.  There have been different troublemakers in the past, that created different sets of problems, and there will be different ones in the future.  No policy can address the wide range of contributor behavior that we will eventually see.  And, the range of things that are troublesome may change as the project expands and matures.
 * well... yes it is legalistic, and fair, and transparent... that has an unavoidable cost. We're kidding ourselves if we think we can have a very simple system that will be fair as well. Just can't be done Wikipedia is too big for peer pressure alone to be adequate (for sysops, yet alone users). Re the flexibility... remember this policy only fills gaps between other policies and it can evolve with time. Its a start...Erich 06:59, 11 Jul 2004 (UTC)


 * I don't like the "three sysops" provision. VotingIsEvil, and besides sysops tend to form peer communities of 5-10, anyway, who often vote as a block.  So, three sysops isn't a particularly effective mechanism, from a standpoint of fairness.  Also, discussion of a proposed ban is in itself damaging, see GetARoom.  Moreover, it is a good deal of work to implement, since at present it is much faster for a troublemaker to create a new account than it is for three sysops to agree on a block (see Wikipedia talk:Sock puppet for more on this).
 * Re the voting... yes voting is bad. i agree that hopefully there will be consensus. The problem is at the momment there is no mechanism to deal with situations when the sysops vehemently disagree. Voting is surely better than mob rule, even if that is mob rule by the sysops... don't you think?
 * I agree about the discussion itself being harmful. Again this is a dilema... all of this is dirty laundry and brings the whole tone down! That is why I propose limiting participants and didn't really emphasise discussion. I would like there to be minimal discussion really. The diffs should speak for themselves. I'm imagining the direction page entries would look like the one I just made below.Erich 06:59, 11 Jul 2004 (UTC)
 * I don't like it either. I think the community should decide if someone is acting like a troll (or whatever) via a vote with the sysops' job only act on whatever the community decides. There is no need for this hierarchy. Wikipeda has thrived because of its anarchistic tradition and we should strive to keep it that way.--GD 18:42, 12 Jul 2004 (UTC)

It is my view that the best responses to troublemakers are:
 * Adoption by the community of a "guidance statement" to the AC to encourage them to relax their evidentiary standards and generally be a little less forgiving. There should, in particular, be no requirement for proof of violation of some written policy for the AC to sanction someone.
 * I'd be happier if we expanded the scope of the policiesErich 06:59, 11 Jul 2004 (UTC)


 * Adoption of the proposal regarding trolls presently being voted upon
 * which one? Erich 06:59, 11 Jul 2004 (UTC)


 * Adopt and give teeth to No personal attacks, which is presently a mere "proposed policy"
 * you mean thats not policy? I didn't realise that. Well I agree completly!! Erich 06:59, 11 Jul 2004 (UTC)

Regards

UninvitedCompany 04:52, 11 Jul 2004 (UTC)

Mock Direction for user:badboytrol
personal attack ++++
 * First warning:
 * Concerning edits:
 * personal attack ++++
 * reversion ++
 * obfuscation +++
 * talk page edit ++
 * reversion ++
 * obfuscation +++
 * talk page edit ++
 * reversion ++
 * obfuscation +++
 * talk page edit ++
 * vandalism +++
 * personal attack ++++
 * personal attack ++++
 * Constructive edits:
 * good new page+++++
 * typos ++
 * copyediting +++
 * typos ++,
 * Final warning criteria met: user:batman 06:36, 11 Jul 2004 (UTC) user:robin 06:36, 11 Jul 2004 (UTC)
 * Final warning:
 * Concerning edits after final warning: personal attack ++++,
 * Block criteria met: user:batman06:36, 11 Jul 2004 (UTC) (24), user:robin 06:36, 11 Jul 2004 (UTC) (24), user:lone ranger 06:36, 11 Jul 2004 (UTC) (24)