User talk:Ambi~enwiki/Thoughts on Dispute Resolution

Ambi, I don't know if you wanted comments, but I read what you had to say, and thought it might be a good idea to chat with you about it. Conversation is key, I think. Plus you're an excellent editor and a wise individual, so I know I have plenty to gain from talking to you. I hope you don't mind.

Arbitration is in a mire: I agree wholeheartedly. I do what I can in my limited time, but it's not enough, and others do far more. It's difficult -- the impression I have is that the only arbitrators who can do the job really effectively are the arbitrators who have given up virtually every other aspect of editing here. I haven't -- I want to keep fiddling with articles and writing new noes and answering the odd question at the reference desk. Maybe I should rethink that. But I think if we're asking good editors to take a position that automatically destroys their ability to edit, there's either something wrong with the system, or else we need to be more up front about how time-consuming it is.

Part of the problem is that we have so many cases, and more each month. Each case is filled with evidence pages that are impossible to wade through. Often complainants wave their hands and say "there's too much to document....just go to these five talk pages (and their archives) and look at the page histories of these seven articles to see what I'm talking about". I know they are frustrated. But as long as people are willing to do that, it's giong to slow arbitration for everyone. Because we're not really arbitrators at that point, being handed specific items of evidence and considering them. We're being investigative reporters as well, and that takes time. Especially because people admit that they include all sorts of really questionable evidence -- I tell someone "that's not really a policy violation" and they reply "oh, I knew that was shaky, just disregard a lot of that stuff. But some of it is real and awful and you need to look at it." How can we move with efficiency through this quagmire? Do you have any suggestions? We ask repeatedly for specific diffs showing the questionable edits (and would prefer them in descending order of importance, so that we see the glaring errors up front and quickly), but few seem inclined to help. I'm looking for answers, and I'm not finding them.

And I would argue that the system we have now works the way democracy works. The enemies of democracy will always find ways of protecting themselves with democracy's fierce commitment to individual rights and respecting the minority. Those in power, however, find that democracy holds them to high standards of conduct. So we have admins who cross the line and are swiftly punished (lest we become a mere oligarchy), while trolls can live more at ease. I don't know how to fix this because fundamentally I think this is a problem in political science. We can have order or individual freedom, but not both. The AC has tended to err on the side of indidivual rights. I think all the cabal accusations tend to make a lot of us react in that way.

Besides, our policies are vague. What is a personal attack? The ones that are clear (like Plautus Satire's) are banned with force. The ones that are unclear (like 90% of the cases we see) are so difficult to handle. It's not true that we've never rehabilitated a "troll" (as some have contended) -- I think many would have described anthony dipierro as a troll this spring, but now he's a productive user (though certainly a maverick who still finds occasional disputes). I don't know if the AC can take any credit for that, but the point is that people can learn. That's one of the guiding ideas of Wikipedia, in my opinion. I don't want to lose that.

I'm writing all this as just a way to sort through some ideas. I hope you'll rpely and help me grapple with some of the issues we've both raised. We both want this to be a better place. We both want to make arbitration better. I hope you'll help me see my way to some things I, at least, can do to help that happen. Thanks for reading this and considering the ideas I'm sharing -- Jwrosenzweig 15:40, 29 Sep 2004 (UTC)


 * Thanks for your response, JW - I do appreciate it. In regard to time, I sympathise with your predicament. None of you should be expected to give up actually editing, in order to fulfil this particular role as well.


 * It seems to me that there's two problems at the root of that - one, that the AC is consistently half-staffed, with numerous users only very occasionally active or on long-term leaves of absences, and secondly, that there may not be enough arbitrators anyway. But also, on top of that, it's my opinion that your job is being made even harder because the mediation system is failing just as badly, which is, in turn, resulting in more cases having to be arbitrated. I'm planning a proposal to try and fix things on the mediation end, which if it works, could help ease this in the long term.


 * With regards to evidence, I think that a bit of give and take is necessary each way. As a complainant, it is my job to provide the evidence. It's not always reasonable, however, to expect that to come in the form of diffs, due to the sheer time it takes to dig these up. For some cases, like that of Plautus, where there are a handful of particularly egregarious edits, then that is quite possible.


 * For some others, the sheer volume of evidence is so large that frankly (and I'm not the only one feeling this way) that it'd be just easier to quit editing in the whole area, rather than collect several hundred individual diffs, and/or go through months upon months of contributions to gather the worst, which would take up enormous amounts of time that would otherwise spent writing articles, or doing things of more use to the project as a whole. If diff edits were to appear in a users contribution page, then this may be possible, but at the moment, I believe that's just not feasible.


 * That said, it's also not fair to just simply point to a talk page and say "stuff happened there! go find it!" In my opinion, the way things were handled on the Rex evidence page is, while not perfect, an reasonable middle ground. All the evidence is there in plain view, and if anyone cares to doubt the context, the timestamp is there, which makes the original post easy enough to find through use of the find option in a browser.


 * As far as even-handed sentencing goes, I think it is possible to have both order and individual freedom. Freedom is important, but, as in any organisation or society, there are rules and laws that need to be abided by. If someone persistently breaks these, whether in a position of power or not, they need to be dealt with. And while it's understandable to pursue such fine ideals, I believe that its vastly superior that one troublesome editor, that spends the majority of their time edit warring and abusing other users (and in the last month, there's been about five of these come up), be barred from editing in a particular area, or even entirely, than seeing one good editor give up in frustration and disgust.


 * It's the same in every area: RK makes people stay away from Judaism-related articles, L6W makes people stay away from the Israeli-Palestinian-related articles, Mr-Natural-Health the alternative medicine articles, and Rex the American politics-related articles. And that, I believe, is inherently bad for Wikipedia. I know that the cabal accusations make some of you guys wary, having had my fair share of discussions with a couple of arbitrators on IRC. However, people who want to cause trouble will do so - and if that means having that we have to ignore yet more cries of "cabal!", instead of dealing with decent Wikipedians being stressed and ultimately leaving, then I'd be inclined to go with the former. It's worth noting that even if we increased the amount of users blocked, that we'd still be far more tolerant than most places online. We give people a fair hearing, and their chance to put forward their own defence. In my opinion, that's enough, as far as fairness is concerned.


 * I, too, believe that people can learn, and I've seen it happen numerous times. I've seen both mediation and arbitration do wonders, and I think the standing order idea in the Anthony case was a masterstroke. Where possible, I believe we should be trying to push similar results.


 * Banning people is a last resort. However, for those that aren't clear cut (which, at a glance of the Requests page, would appear to me to be around half at present), then it may be worth considering increasing the use of standing orders - which are probably the best method the Committee has at this point for trying to get problem users back on the straight and narrow. But for those people that don't show much sign of being turned into productive users - and we do have several of this type - in these cases, we really need some strong action on the part of the Committee.


 * Anyway, thanks a lot for your response, James - and I hope I haven't rambled too much in my own reply (it's 2.30am here). The system is in trouble, but it isn't irreparable, and I'd like to think that through dialog, at least some of the kinks in it can be worked out. Ambi 16:31, 29 Sep 2004 (UTC)


 * Thanks for your thoughts, Ambi -- you make a number of very apt comments, and I find myself with even more to think about. :-) I agree that the kinks are fixable, and I agree that the current system is troubled (I think "in trouble" may imply a more precarious position than I think we're in, but I don't want to get caught up in discussions of definition when it's obvious we have generally similar viewpoints).  I do have a few thoughts in reply to yours, mostly just me exploring a few ideas and refining some ideas I have already expressed.


 * Your comment about too few arbitrators is a wise one. It has occurred to me in the past that we might consider forming tribunals, as a way of limiting the weight on an arbitrator.  For example, if we had 12 active arbitrators, we could have 3 sets of 4.  Each new case would be assigned to the next set of 4 arbitrators in turn.  A majority of 3 votes would be needed for decision-making.  This would cut the burden on each arbitrator by 2/3.  There are serious drawbacks to this system, which I acknowledge.  Deadlocks would be too frequent....perhaps an odd number would be preferable (5? 3?).  And it may be that it would create too many troubles.....a certain tribunal gets known as "the easy ones" and so cases get timed so that you don't take somebody before the "soft judges".  But we have to come up with a method, and I think pure addition of arbitrators is less than satisfactory, since all it's likely to do is increase the length of time before decisions can be made.  It's hard enough now to figure out when we have a majority.


 * I agree that diffs are time-consuming to provide, and that many people find it easier to go elsewhere rather than solve the problem. Here's the one point I want to raise in response: what a person does when they avoid the trouble of finding diffs is place the responsibility on each arbitrator to find them.  It is no quicker as an arbitrator to find the diffs....in fact, it's much more difficult, since I don't know what I'm looking for.  Let's assume for argument's sake I have ten cases that have provided vague evidence, at least in part.  I have to do 10 times as much work as any one complainant would have had to do.  I'm not looking for anyone to cry about how tough my job is.  But I hope people recognize the amount of work they're expecting of arbitrators -- it's too much to ask, especially if you want decisions in days, not weeks and months.  If complainants aren't going to dig up diffs, then someone has to.  And that someone is the AC....each of us individually, of course, since we can't share diffs around with each other because we have no easy way of doing so, and because it would appear that we were trying to advocate for one side or the other, based on the diffs we made public.


 * I agree, some cases have handled it well. But I think generally the community has gotten the wrong idea about evidence.  They think they need to overwhelm us with evidence (Rex's page is 150kb long, as I recall).  I think the easiest cases to decide are the ones that provide a dozen or two diffs of the worst offenses, and then a note that more examples of the behavior will be supplied on request.  It makes it clear that there are real offenses that can be seen at once without forcing me to wade through everything.  You're right, this would be so much easier if diffs were available from the contributions page -- that's a great idea that would help everyone, though I have no idea how tough it would be to implement.  Once again, though, all I can say is that if people don't feel it's worth time to find and present evidence to us, we have to do all that.  We're doing the dirty work for a lot of personal grievances.  I know, we did sign up for the job, but I have to say, the job has gotten worse as it's gone on.  I can't imagine what it's been like for people who've been on since January -- after 2 months I'm already exhausted. We'll burn through far too many arbitrators at this rate -- something about evidence has to be fixed.  I just can't see what.


 * I agree with you that a balance can be found, and that we haven't found it. But I do think balance is difficult, and that ultimately any human system will lean one way or the other.  I do intend to be more proactive in restricting editors who are driving away good contributors from topics.  But I hope you can appreciate it from our perspective -- often the problem is that 5 or 6 good contributors are on one side of an issue, and a lone voice opposes them.  The lone voice is being very rude, and making life difficult.  But as arbitrators, it is difficult to silence the lone voice -- we want Wikipedia to be a good source of information, but more and more, fights seem to be between a pack of contributors and a single person defending an unpopular view.  Yes, that person should behave themselves.  Yes, we should want to keep good contributors, and we do.  But it's hard to draw the line when someone with a legitimate perspective has felt silenced and shouted down, and is lashing back out in frustration.  I know I'm supposed to sympathize with the frustrations of the established contributors more than with the frustrations of the antisocial outsider -- that's what the community cries out for.  But honestly, a lot of very fine, well-meaning established contributors are absolute jerks to people they perceive as being in minority opinion groups.  Many seem to delight in pouncing on new editors who don't understand NPOV, and sniping subtly at them at every turn -- never personal attacks, but consistently marginalizing editors until they feel completely alone.  We manufacture the majority of our trolls, I'm convinced -- many of them could be dealt with kindly, and they aren't.  And it's hard to punish someone for their rudeness when it's based on legitimate feelings that they're being bullied.  They are -- we're full of intellectual bullies here.  Many of them very nice people on most topics, mind you, but with pet peeves that set off fireworks.


 * These aren't popular ideas, I know, so I don't often express them. And I certainly understand the ease with which an editor can say "I know this topic, and my perspective on it is correct....a lot of people agree with me -- why should this confusing and frustrating POV warrior be given the time of day?"  So we don't give them the time of day.  And we don't really have rules against bullying that doesn't involve personal attacks, so we're stuck.  I can think of several cases where a rogue editor is obviously misbehaving, but when I come to examine the evidence, I find that every little change they make is being blown out of proportion and ridiculed.  All of it within policy (except Wikiquette and Wikilove, which sadly are merely principles, not actual rules, and therefore difficult to enforce).  I don't know what to do in those cases.  I hate to ban someone because they were goaded into becoming a troll, especially when (as is often obviously the case) contributors quickly became aware of an editor's buttons, and purposefully pushed them in order to get some bannable actions.  Not all the cases we handle are this way -- maybe it's the minority.  But they're the ones that keep me up at night.


 * I agree that standing orders are preferable to bans. Honestly, though, I think we are moving that way.  We reprimand an editor for misdeeds and announce that a) if they do it again, admins may intervene to warn and then temp-ban, or b) if they do it again, they should be immediately brought before the AC for a long-term ban.  Everyone seems to think this is us wimping out, or acting toothlessly.  They want a ban, not just a censure.  And so we're stuck -- the community doesn't want anything less than a ban, and too often only a punishment less than a ban is appropriate.  We are at loggerheads.  Are there ways of putting together standing orders that you think would be more effective?  And can we adequately combat the community's thirst for blood?  Or do we need a bit more banning?  As you point out, we're far more permissible than most sites (for good reason, I think).  Accepting that we need to be more permissible than most sites, should we be less permissible than we are?  I don't know.


 * I don't mind being called a member of the cabal when I'm not one. Water off a duck's back.  But when I sense that the community pressure is to ban an annoying user, and when I sense that the user has been cultivated into annoying behavior by editors who know they are popular and hard-working, and who don't want to be bothered trying to reason with and compromise with someone who's obviously lacking in several social graces....well, then I do try to avoid banning.  Because I don't want to be called a member of the cabal.  Because I'm more than a little concerned that I'm being told I'm supposed to be one.  And that is my real fear -- not trolls.  But maybe I haven't seen enough trolls in my time here to know how difficult it is to be on the short end of that stick -- I think Irismeister was enough for a lifetime or two, but I'm sure it gets much worse than that.

Jwrosenzweig 20:28, 29 Sep 2004 (UTC)