Guild Wars Wiki talk:Three-revert rule

Personally, I prefer Guild Wars Wiki:Only revert once. --Barek 11:05, 15 February 2007 (PST)

This looks interesting and perhaps viable, although I'm thinking it might be troublesome if we have to check their number of edits as well as what's being reverted and re-reverted. -- ab.er.rant -- 16:51, 19 February 2007 (PST)

I support the 1RV-rule over this, but it'd be nice to see essential things like 'What's a revert' and 'Exceptions' merged into 1RV. --Erszebet 06:37, 20 February 2007 (PST)

I also support the 1RV-rule over this policy, and the integration of the 'What's a revert', 'Exceptions' and 'Enforcement' with 1RV in some form. The 1RV policy, IMHO needs a bit of a clean-up as well, prior to being implemented. --Indecision 03:46, 4 April 2007 (EDT)

Turnabout is proverbially fair play

Rezyk's entirely to blame for this. He asked.

My largest concern is the 24 hour time period. What changes in 24 hours? 48? 72? Nothing if all you're doing is reverting. You just hope your revert hits just after the other editor has gone to work/school/sleep/whatever so that your version of the article stays up longer than theirs does? That's an edit war just as much as users reverting each other 4, 6, or even 10 times in a few hours.

Actually, the policy recognizes this, stating, "Editors may still be blocked even if they haven't made more than three edits in any given 24 hour period, if their behaviour is clearly disruptive." Granted, I like a little room for admin discretion in policy, but that's pretty darn grey. To use the fence analogy, a policy should have a fence that users shouldn't be allowed to go over (warning, undoing, banning when necessary, etc), and I have no problem with calling a user out for putting up their hands next to the fence going "I'm not touching it! I'm not touching it!" like a 5 year old, but this is more like a dog on a chain that reaches all the way to the yard line in some places but only 3/4 of the way to the line in other places, and drawing a line in the ground at the 3/4 mark as "okay up to this point, mostly". Just put up a fence at the yard line. I don't know how much of that analogy worked, but it's better than the first few I tried.

Moving right along, I think the sooner editors are compelled to move to the discussion page, the better it is for the wiki, for the page, and for the editors themselves. This rule doesn't compel the editors to go to the discussion page at all: if they're stubborn enough, they can literally go on reverting each other forever without ever speaking to each other. Fairly early in my time here, I made an edit to a skill page, which someone else reverted. I thought my position was obvious, but nonetheless I put a note on the talk page of the article, and when I didn't get a response for a day or so, left a note on the other user's talk page. As soon as he saw the explanation, he agreed and self-reverted his revert. Granted, not all disagreements can be resolved that easily, but for the ones that can, I know I would feel pretty darn silly reverting over and over and over when a few minutes on the talk page could've resolved it all. All this policy really seems to do is protect the Recent Changes page from excessive revert spam at any one time, which I should hope is not our only goal.

Honestly, the only cases I can see where it would be more beneficial to revert multiple times are in cases that are already pointed out as exceptions in 3RR, 1RR, and 1RV anyway: vandalism, self-reverting mistakes, someone messing with your user page... If there are further exceptions, they can always be brought up and the exception portion expanded.

That's all I've got for the moment. - Tanetris 00:20, 27 July 2007 (UTC)

I'll try and jump straight to my main point here. I think many users here tend to oppose this for the wrong reasons: that this policy doesn't address some of our potential problems the way other proposals do. As an analogy, imagine if we did not have GWW:NPA in place, and someone suggested a policy which forbade threats of physical violence, but did not address things like racial epithets. Would that be considered a bad policy that should be opposed since it doesn't encompass enough? It may not be as restrictive as we want, but it still fences out some personal attacks and doesn't really create new problems in doing so. It doesn't forbid other policies from addressing racial epithets and such. I'd hope that the main counterargument against it would just be that it's redundant after GWW:NPA has been accepted.

I see the situation here as similar to that. Almost all of the arguments against 3RR have been centered on how it doesn't deal with many potential scenarios as well as the other revert policy proposals. And it's true, it doesn't -- but the reason for that is really not that it handles those aspects badly, but that we're intentionally not trying to "fix" those with this policy. This is just meant as a fence that strictly strikes at one particular aspect of arguments on this wiki (rapid revert warring) and no farther. It doesn't force an end to long-term revert warring, doesn't stop stupid quarrels from happening, and doesn't set up a lasting peace in the Middle East. But...why should we expect that it has to do any of these to be acceptable? It's not like this policy is being pushed as the complete be-all and end-all of revert issue handling. We can have additional policies/practices to deal with the problems that this misses. If someone complains now about how this proposal doesn't solve slow, long-term revert wars, my general stance is: This policy doesn't address that problem but doesn't exacerbate it either; it doesn't make that behavior any more allowable than it currently is, and doesn't stop us from making further policy that does handle them. I think this proposal doesn't deserve many bonus points in that category, but doesn't deserve negative points either."

And that brings us to the other revert policy proposals that are on the table (1RV and 1RR). For those readers who haven't closely followed the discussions, I've been the most vocal detractor of those proposals. I'll re-iterate here though, that my reasons for doing so is not to push us towards 3RR as a solution. It's just that, when judging them by their own merits, I feel that they carry serious dangers which are not readily apparent, and which detract significantly from their many good points. I've generally gone through my arguments on their talk pages so I won't go into further detail here, but please note that mechanically they are not incompatible with 3RR. It's just that they would make 3RR redundant if accepted, much like how NPA makes a separate "no threats of physical violence" redundant. I would like to see if we can distance ourselves further away from the "competing proposals" mentality. (Unfortunately, it's even hard for me to do this myself because I see many of 3RR's twists as motivated by trying to avoid those same problems..)

I'll try to follow this up with some more direct discussion regarding some specific reasons why not to adopt 3RR. --Rezyk 20:48, 30 July 2007 (UTC)

My reason is simple (and I can state it in one sentence as well!). If 3RR gets accepted, 1RV or 1RR (which I feel is superior) will never become policy. --Xeeron 21:29, 30 July 2007 (UTC)

Have another reason. If there is an official policy on what revert wars are fair, users will get the message that as long as they follow the rules, revert wars are OK. Even with no policy, it very rarely gets as bad as 3RR would make the nórm. Backsword 23:17, 30 July 2007 (UTC)

1. Why the 24-hour time period?

   One answer is: it doesn't matter why. We're not trying to accurately model the theoretics of arguments here, but should just get a practical policy in place without causing more problems. If the resulting policy with a time period as a whole is not problematic, and is helpful in some cases, why does there have to be further theoretical justification for its details?

   Another answer is: so that we avoid building around a notion that we're accurately modeling arguments with our policy. Without the time period, we're deterministically deciding the final result of an argument, which isn't good because the nature of arguments naturally defy any such simple model. We'd head into all the same problems of voting...and worse, it'd set up our content to be decided by game-playing and making that culturally okay.

   Another: Because people sometimes develop new ideas or change their minds. Maybe the first 3 attempts at a good neutral version failed because new people joined the discussion and added more issues. After a good night's rest, that same user comes up with a 4th idea that might solve everyone's issues, but doesn't feel like going through the whole get-consensus-approval-first thing. It was mentioned that we can just add exceptions to the other proposals...well, I'd like to see these kinds of things be allowed.

   Another: 24 hours later is better than 24 minutes later on another account.

2. Fence is too far out.

   From a plain mechanism point of view, this is strictly better than no fence at all. Also, I really wouldn't mind moving the fence closer as long as it doesn't do stuff like threaten to usurp consensus. But I'd rather have the electric fence always being on the far side rather than sometimes too far in, where it can really screw things up.

   From a culture point of view about how it makes simple revert warring okay...yeah, there is some problem there. I'll try to rewrite a bit to mitigate that..don't know if you guys will like it or not.

3. Too much admin discretion

   Hmm...agreed, but only because you are (fairly) interpreting it in a way that I wasn't. I didn't interpret that as allowing blocks there, but rather acknowledging that this policy will not protect you from those. Will try rewrite.

4. Doesn't compel the editors to go to the discussion page at all

   Sure it does; it even does so better than policies which strongly favor one side better than the other. If users from one side of a revert war started discussion while the other side refuses to, then the side that doesn't will eventually be operating against consensus, which is a no-no. It even compels outside editors who don't like the revert warring to go in and fire up the discussion themselves. If revert policy automatically favored one side, then that side would not really be compelled to partake in serious discussion because they don't have anything to lose by being simply dismissive/rejectionist.

5. Don't want this because 1RV or 1RR will then never become policy

   Just because of our cultural situation, right? Any ideas/thoughts on how to best proceed? I mean, if we attempt to ratify 1RV and 1RR and they are both rejected, would that eliminate this incentive? And then anyone who wants 3RR might be inclined to reject them for that reason.. =/

--Rezyk 01:11, 31 July 2007 (UTC)

While I don't agree with your general point, one solution to 5. is to make the policy official only for a limited time. Would probalby still give it some advantage, but would force discussion, perhaps more than now, as people wouldn't want to be without policy. Backsword 01:57, 31 July 2007 (UTC)

Fail?

Time to reject this one?

1. It's dead.
2. 1RR seems more popular.
3. Concensus seems to be that rever wars are bad, so a 'fair rever wars' policy would never be accepted.

Backsword 13:09, 14 October 2007 (UTC)