Wikipedia talk:Non-administrator rollback

From Wikipedia, the free encyclopedia
Jump to: navigation, search
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.


We keep hearing about Twinkle.

As someone who doesn't have such access, would someone like to list the benefits of Twinkle, so that we might be able to compare Twinkle/rollback/undo?

Also, are there any other similar programs in use atm (For example, can AWB duplicate this as well?)

And if someone really wanted to, perhaps create a comparison chart : ) - jc37 10:09, 5 January 2008 (UTC)

I'll get a list started, anyone should feel free to edit it. I have no idea about AWB though, never used it. PS - Comprehensive info on Twinkle's functions is available at Wikipedia:WikiProject User scripts/Scripts/Twinkle/doc (which appears to be in the middle of an update, probably due to all this attention). Equazcion /C 13:48, 5 Jan 2008 (UTC)
It's a shame that those three options were not given up for discussion. It's essentially what I would agree to. "Rollback" with an edit summary - for non-admins; and "rollback/vandal" which is rollback without the edit summary - for admins. I "think" the AGF one seems to just be a variation which is useful for someone using such a "tool", and so might also be useful for bots? Anyway, thanks for the information and insight : ) - jc37 10:55, 6 January 2008 (UTC)
My pleasure. You're probably right about the AGF rollback. I don't think anyone disagrees that custom edit summaries are a good idea -- the problem is that they would require extra programming. Everyone would rather just use the existing admin rollback function and simply "switch it on" for select non-admin users, rather than having to write a new tool. Of course I see where they're coming from, but I also feel it would be worth the extra effort to write in the ability for custom summaries. I think if we insist that people leave edit summaries for regular edits, we should require them all the more for the much more sensitive act of a rollback. Equazcion /C 11:12, 6 Jan 2008 (UTC)
So it's "I want it now, so let's do something that we probably shouldn't for expediency's sake"? Ouch.
If enough people supported at least this much, I think (or at least I'd like to think) they'd program it. Though I just had a thought... Is TWINKLE open source? and if so, couldn't that just help expedite the process? - jc37 11:37, 6 January 2008 (UTC)
Twinkle is open-source, but it's a javascript tool. It makes rollbacks by running on the user's computer, and performing multiple transactions between the local browser and Wikipedia -- it finds and downloads an old version of the article, then re-uploads it and saves it as the current version. The admin rollback tool performs a single transaction: the user says "rollback this article" and the rest happens on the server. The admin method is faster and more efficient for both the client and the server, with less of a processing/bandwidth burden on both ends (especially for people with slow connections), which is seen as one of its major benefits over Twinkle. To customize it, the server-side code would need to be rewritten -- that is PHP code, not Javascript code, so Twinkle code couldn't be "copied" in order to make the customization easier. Equazcion /C 11:46, 6 Jan 2008 (UTC)
Makes sense. And even though transcribing from language to another might not be a big deal, it sounds like different things are happening in TWINKLE than what would be happening as a result of the internal Wikipedia coding.
I wonder if there is/are any developers who'd be willing to potentially "waste their time" programming this. - jc37 12:11, 6 January 2008 (UTC)
They already have. This function has already been programmed, and is awaiting installation, pending the results of this debate.--Vox Rationis (Talk | contribs) 06:31, 8 January 2008 (UTC)
Let me make sure I understand you. There's already two versions of rollback programmed? One with an edit summary prompt, and one without? (In twinkle terms, "rollback", and "rollback vandal".) - jc37 00:39, 9 January 2008 (UTC)
Three actually, if you include "rollback AGF" (assume good faith). However, when I said "already been programmed, I was not referring to TWINKLE, but that the feature currently being discussed has already been made, and is awaiting installation. --Vox Rationis (Talk | contribs) 02:17, 9 January 2008 (UTC)

Twinkle rollback functions[edit]

The following 3 rollback functions are available on the diff screen of an article, for the current revision only. Following a rollback, a separate browser window is spawned with the rolled-back user's talk page open for editing, so that a notice or warning can be issued:

  • [rollback AGF] - Rollback with the assumption of good faith, adds an edit summary of good faith with optional custom edit summary.
  • [rollback] - Generic rollback with optional custom edit summary.
  • [rollback vandal] - Rollback vandalism, no custom edit summary.

Note that [rollback] and [rollback vandal] are also available from user contribs pages, as links listed next to articles for which this user was the last contributor.

This is the revert function, available in diff screen for all revisions:

  • [restore this version] - Allows restoring a page to any previous revision, regardless of how long ago it occurred. Optional custom edit summary but no user talk page is opened for warning.

Edit wars: prevention[edit]

At Wikipedia talk:Rollback for non-administrators#Some suggestions, I made a suggestion for a way to avoid edit wars. Several others had suggested limiting how often a non-administrator could use rollback; I suggested looking at it from the point of view of the article: "Limit rollbacks on an article. For example, two in a row, or two in an hour. Limiting rollbacks need not limit ordinary editing (including reverting). Also, make rollback subject to the three-revert rule just as a revert is." Would this be compatible with the present proposal? Fg2 (talk) 12:17, 4 January 2008 (UTC)

How about leaving an automated message on an roleback receiving editor's talk page telling them that their edit has been rolledback by editor giving a rollback and if they disagree with the rollback please leave a message on non administration rollback notice board and send an alert to the rollback User notifying him or her that their implemented roolback is beeing disagreed. They do not have to answer it but statistics can be kept and used for evaluation to revoke the rollback tool from an editor if their roolback authority is problematic. So this will be anti exploit review mechanism. Igor Berger (talk) 12:31, 4 January 2008 (UTC)
Well, I am already feeling a little uneasy about the proposal. One RC Patroler has been stalking me with mindless tagging (semi-speedy tags are the patroler's particular favorite), making abusive claims unwarrantedly dropping into discussions, and voting against whatever I vote for on AfDs... but, this clever editor "generally" manages to keep the stalking part under 10% of edits, thus hovering on the borderline and evading detection by outside parties. I have learned to live with this much, but "if" the person can lay a hand on the rollback stuff... I can only shudder at the thought. Aditya(talkcontribs) 14:43, 4 January 2008 (UTC)
This editor could be a sockpuppet or involved in social engineering on behalf of a third party. You may want to bring it to a few admin's attention and see how things fan out. Igor Berger (talk) 14:48, 4 January 2008 (UTC)
Aditya's correct, basing it from an article's perspective doesn't take care of the wikistalking potential. Stalking usually involves many articles, not just one. Equazcion /C 22:10, 5 Jan 2008 (UTC)
Right: this section is not about stalking; it's about preventing edit wars. So, would implementing a limit of this sort help solve concerns about edit wars? Fg2 (talk) 02:39, 6 January 2008 (UTC)
Probably not significantly. Edit wars rely on more than 1 person. For limiting by the hour: unless people are watching the article like a hawk, it is unlikely that someone will make more than 2 reverts in a content dispute in an hour, but it is much more likely that they will revert a vandal more than twice. For limiting it to 2 in a row: that would only prevent edit wars where all parties are using rollback. Mr.Z-man 03:47, 6 January 2008 (UTC)
Any abuse of rollback by a non-admin user would be grounds, I'd expect, for immediate revocation of the privilege. The proposal is explicit, rollback is to be used for vandalism only, not for content disputes. *One* bad rollback, any admin could remove the bit on sight, then it could be argued at leisure if it was a mistake or not. The goal here is to help admins, and no new policies are needed. There is little or no harm in wrongful removal of the bit. That's the beauty of this proposal; admin-granted, not by "vote" or community discussion, simply at admin discretion (and responsibility). It only takes wider attention when someone abuses it, and basic abuse, by a non-admin rollback user, can quickly be addressed by *one* administrator. Only if two administrators disagree does it increase wider community traffic, as with any admin disagreement. Meanwhile, no more harm done than can already be done with heavier load on the servers. Frankly, I'm astonished that we are seeing only 2:1 in favor of this, and I suspect it's another example of participation bias. The vast majority of users, seeing this, are responding, "Huh?" My own analysis of *most* of the No votes who give reasons is that they simply don't understand the proposal, for objections are being made that are addressed by details of the proposal; a few Yes votes also don't seem to get it, but not nearly so many. Again, this is not a vote. This is a request for community input. I presume they will decide, as usual, based on the *arguments*, not the pure number of votes. --Abd (talk) 17:24, 7 January 2008 (UTC)

If someone uses rollback inappropriately then let an admin know and it will be dealt with. There is no point in making hard and fast rules about how many uses per hour or article. If there was a persistent vandal who was using proxies to change their IP, then using the rollback many times would be appropriate. We can handle this is simple sense and consensus instead of numerical rules. 1 != 2 17:54, 7 January 2008 (UTC)

The point of hard and fast rules is that the servers can enforce them. This per-article limit idea was intended to address two points people have raised about rollback: (1) that editors will use it for edit wars, and (2) that it will add to the load on the administrators. Having software prevent the use of rollback multiple consecutive times, or two times in an hour (or in five minutes or in a week, whatever works best), on any given article, makes rollback useless to non-administrators for engaging in edit wars. It removes edit wars as an objection to non-administrator rollback. And having software limit it, rather than requiring running to find an administrator, makes life easier for the members of the administrative team. Or so I envisioned. But I haven't seen enough edit wars to be sure if this is the right approach. Fg2 (talk) 06:13, 8 January 2008 (UTC)


It's quite clear from the poll there is no consensus for this. This doesn't have a snowball's chance in hell of becoming policy. The best that can be done is a new proposal that addresses the concerns the opposers have raised, if that's possible. —Ashley Y 08:36, 6 January 2008 (UTC)

Agreed, but less cynically. I'd say that while there isn't a snowball's chance in hell of this particular proposal becoming policy, there also isn't a snowball's chance in hell that non-admin rollbacks won't be implemented in some form or another. It's just a matter of forming the right proposal. Equazcion /C 08:41, 6 Jan 2008 (UTC)
(to Ashley Y) I think it's too early to say that, but there are concerns about if people who voted earlier would still have the same position, given some changes made to the proposal. So we might want clarify some of that, but either way, the discussion is still very much in progress. -- Ned Scott 08:42, 6 January 2008 (UTC)

I've created a new page to discuss the creation of a new proposal. All are invited to brainstorm: Wikipedia:Non-administrator rollback/Creating a new proposal. Equazcion /C 15:18, 6 Jan 2008 (UTC)

I don't know what discussion you are looking at, but the opposition to this proposal has less than a third of the community behind it and is fragmented in their reasons. What is more many of the given opposition reasons are based on technical misunderstanding about what the tool actually does. Judging by consensus or vote counting I would say there is a fairly good chance of this passing. 1 != 2 17:58, 7 January 2008 (UTC)
The supporters of this proposal are victim to similar misunderstandings and are just as "fragmented in their reasons" (not that each user having a different reason constitutes any lack of validity in the viewpoint). Judging by consensus, which is all that matters, well, there is no consensus. Equazcion /C 18:11, 7 Jan 2008 (UTC)

Who likes polls?[edit]

'Cause I got another one for you. This is very informal, doesn't need to go on very long, and will not decide anything. However I just wanted to get an idea of how many people here think the current poll shows a consensus acceptance of the proposal, and how many feel no consensus is shown. Vote below. Equazcion /C 17:14, 6 Jan 2008 (UTC)

Don't we need a vote on whether the vote below is showing consensus or not? </sarcasm> Carcharoth (talk) 03:12, 7 January 2008 (UTC)
No consensus I think Carch's question should be discussed further to be sure it is supported by the community :) Franamax (talk) 03:19, 7 January 2008 (UTC)
Look, enough with the voting. Seriously. Poling does not help advance consensus. Polls, more-or-less by definition, require one to vote to have one's opinion counted. This actually impedes discussion, since now everyone must respond to have their opinion considered. This turns what could be a discussion into an edit storm of voting. Meanwhile, in an actual discussion, if something is said, you don't have to have 200 more people come in and say "Me too". Indeed, you might actually get some new ideas formed, and improve the state of understanding, rather than being lost in a sea of votes. • Wikipedia is not a democracy. Polling is not a substitute for discussion. —DragonHawk (talk|hist) 03:41, 7 January 2008 (UTC)

No consensus - I don't feel the poll currently shows consensus acceptance of this proposal. Equazcion /C 17:14, 6 Jan 2008 (UTC)

No consensus - Even though I voted for Support, I realize that majority does not equal consensus. Dansiman (talk|Contribs) 20:10, 6 January 2008 (UTC)

No consensus - I don't see consensus at this time because majority does not equal consensus Alexfusco5 20:19, 6 January 2008 (UTC)

No consensus We ought to keep this poll running until we finally come to a conclusion or compromise or someone from the WMF decides. Marlith T/C 00:39, 7 January 2008 (UTC)

No consensus Even if some of the !opposes are using faulty reasoning, a lot of people just don't feel good about this proposal. I'm disappointed at the lack of input at the new proposal page, perhaps the banner invitation has only attracted drive-by's rather than a fruitful discussion of ideas. Franamax (talk) 01:21, 7 January 2008 (UTC)

No consensus There are considerable yea and nay positions on this issue, determining consensus amidst this is problematic at best. Edit Centric (talk) 01:31, 7 January 2008 (UTC)

The worst part about this is that theres so many Supports and Opposes that I can't even remember if I'm one of them! Ferdia O'Brien (T)/(C) 03:46, 7 January 2008 (UTC)

I refuse to sum my comment up in few words of bold print. From the start, through 380+ responses, comments placed under the "Support" section have outnumbered comments placed under the "Oppose" section by a little over 2 to 1 (or roughly 70% in the "Support" section). Of course, given WP:DEMOCRACY and WP:CON, tallying that gets us precisely nowhere.  :) Reading the comments is more informative. People are coming from all over the place. A lot of comments are concerned with abuse. Perhaps that's because when the poll was started, the restriction that rollback was only to be used against obvious vandalism wasn't there. Of course, once that was added, others switched to oppose due to the change. Others complain of the lack of entrance requirements; I note that originally, there were some, but the proposal was changed during the poll. Polling on a moving target; what a mess. But I do see some fairly strong support, overall, for some kind of solution. The poll unfortunately preempted useful discussion of what the solution should be. I note many of the comments use the same rationale, regardless of "Support" or "Oppose": Script tools are available, so (why not give them to everyone|no need to give them to everyone). I'm not sure how to interpret that! —DragonHawk (talk|hist) 03:59, 7 January 2008 (UTC)

This particular point seemed best served by a poll, because you either think there's consensus or you don't. It's not like a proposal where people are likely to say "no, but if you change this, then maybe". But like I said this was entirely informal, just for curiosity's sake -- and just-for-fun, even. But anyway, since everyone has to take themselves so seriously... I don't think there's consensus but I do think there's a chance this will pass into a trial phase, which certain admins may justify by saying there is enough of a majority to enact this, as a kind of "compromise" (no consensus, but, it seems clear most people want it, so let's try it out). I don't necessarily see that as a problem per se, but I do think it would be better to get an actual discussion going in order to fine-tune the proposal according to people's poll comments, rather than this straight-up "either you like it as is or you don't" method. I think it's important that the admission of new tools and rules not fall victim to a trial-and-error trend. There's no reason we shouldn't be able to form a more accepted proposal, before we start a trial run, now that we have everyone paying attention and saying exactly what they want to see. And yes, 70% support is a lot, but it still doesn't equal consensus. For as many of the Oppose voters being accused of making misguided votes, there are even more Supporters whom I would say are pretty misguided as well. Consensus isn't about pie charts. Dragonhawk said this in better words somewhere else, but we'll get more out of the few people who care enough to actually discuss this than we will from the masses of people who've just shown up out of some need to utilize their right to vote. Equazcion /C 05:59, 7 Jan 2008 (UTC)
Consensus is not a binary state. It's not black-or-white. For example, as many have been saying, the poll does appear to indicate a consensus for some sort of non-admin rollback, but the specifics are unclear. I realize you were asking about the current proposal, but as I pointed out, the "current proposal" has evolved over time, so even that's pretty muddy. The content of comments is generally going to be more telling than simple yes/no. As WP:CON says, consensus is about finding a solution that everyone accepts. Voting doesn't help any of that. Consensus is analyzed, not counted. • Perhaps we are all taking this a bit too seriously, but then, I think I can safely say that we're all a bit "sick" of polling, too.  :) Maybe we should have a poll on that.  ;-) —DragonHawk (talk|hist) 02:53, 8 January 2008 (UTC)
I'm not fond of polls - achieving consensus is obviously better. But in order to achieve consensus when one does not have unanimity, one has to discuss the concerns of the people on either side of the debate and work to flip the opinion of enough of them that a clear consensus appears. This may mean using the force of cold, hard logic, or it may require empassioned oration - or you may have to give ground and compromise. However, when upwards of 450 people are involved and something like 150 of them need to be 'flipped' you simply cannot discuss the matter with them one-on-one in order to do that. Rational debate in an open forum becomes impossible when the number of people involved becomes large. Worse still, how many of those people took the HOURS of time it would take to read all of the debate on both sides? How many will ever return in order to be persuaded? Consensus is a great way to proceed when you are in a group of a dozen people and three or four are holding out against the majority. But as a way of making decisions, it's quite utterly useless when the numbers swell into the hundreds. IMHO, a straight vote with a 2:1 quorum is ample evidence of a need to change when this many people are involved and rational discussion is all-but impossible. SteveBaker (talk) 21:08, 8 January 2008 (UTC)
This is one of the best posts I've read about this matter. Perhaps this was implemented to fast but ultimately, I don't know whether we could have done any better. While perhaps we could have improvided the proposal, what likely would have happened would have been a small number of people would have worked on improving the proposal which would have then eventually been put up for community feedback (i.e. vote) again. This probably would have gotten about the same amount of support or worse, less. This process would have been repeated over and over again until either consesus was achieved or there was clearly no consensus not because the proposal had actually improved but simply because of the 450 people who 'voted' this time, we were down to 50 as most people had grown sick of the thing by that time. Proposals should definitely be worked on and improved until they are more or less the best possible. But ultimately there has to come a time when a proposal is either accepted or rejected. And that time comes when the best possible middle ground has been achieved which IMHO is evidence by the apparent inability to come up with something that had more support (since from what I saw, all the alternatives had less support) Nil Einne (talk) 19:00, 11 January 2008 (UTC)

Trial run[edit]

Even though I currently oppose the proposal, I would support a trial run. How would others feel about a trial run? -- Ned Scott 00:53, 7 January 2008 (UTC)

any foothold by the anti-dissident wing could lead to disaster. if you owant to see a trial run in action, visit Kevin Trudeau, criticisms of sylvia Browne, Uri Geller, or any other paprasychology-related article when you have a whole bunch o f peopel pretending to be admins and reverting back any edits that threaten the "mainstream" powerstructure. Smith Jones (talk) 00:59, 7 January 2008 (UTC)
The fact that edit warring already happens now is not a very strong argument for the idea that it would cause edit warring. —Random832 01:02, 7 January 2008 (UTC)
I would still be willing to see how this plays out on a larger scale. I'm betting it would back up my own view, but hey, maybe not. -- Ned Scott 01:04, 7 January 2008 (UTC)
I commented somewhere that if this turns out to be a miserable failure, it can always be reverted; nothing's permanent about this software change. --MZMcBride (talk) 01:10, 7 January 2008 (UTC)
Can't see any harm in a trial run. Are there any volunteers at this stage? --SmokeyJoe (talk) 01:31, 7 January 2008 (UTC)
Em, object to this. You can't seriously give a lot of people rollback, and they think you can take it back from them. A trial run is just a means to push this through without consensus.--Docg 01:33, 7 January 2008 (UTC)
Of course you can. Whatever makes you think that you can’t?
No, a trial run is more than that. A trial run generates real information, as opposed to hypothetical/imagined stuff that people have been basing their opinions on.
This is just a small, reversible, incremental change, expanding functionality, efficiency and trust. It is too inflexible for such proposals need proof before testing. Systemic inflexibility will hurt the development of the project. --SmokeyJoe (talk) 01:54, 7 January 2008 (UTC)
What would a trial run prove? Wouldn't the trial involve the most carefully selected users acting on their best behaviour? Or would there be deliberate selection of careless users to balance it out? The concern here is the long-term possibility of misuse and especially careless use, a trial won't establish that potential. Franamax (talk) 01:49, 7 January 2008 (UTC)
I support non-admin rollback but I agree there seems little point in a trial. The benefits or otherwise should already be apparent to the large subset of regular Wikipedians (admins) who have access to the tool right now. A trial of non-admins would inevitably involve users of good standing, so the risks of misuse would be nil and the trial may not reflect actual practice if non-admin rollback was approved. The best possible outcome of such a trial would be to determine whether rollback was more or less useful than similar tools like Twinkle, which is a mattter of opinion for each individual user. Euryalus (talk) 02:10, 7 January 2008 (UTC)
You will simply not get consensus for a trial, I suspect.--Docg 02:12, 7 January 2008 (UTC)
Euryalus touches on an excellent point here. We already have a case study of the most trusted users. What is the admin experience with the rollback tool? Do they agree that they have all used it responsibly? Have they noted cases where other admins have misused or carelessly used the tool? How have they responded when witnessing misuse - or have they ever looked? Perhaps (another!) poll of admin opinions is in order here. Franamax (talk) 02:39, 7 January 2008 (UTC)
Then people would complain that its only the admins giving opinions. Rollbacks are included in the edit history. It should be fairly obvious if there is significant misuse. Mr.Z-man 03:36, 7 January 2008 (UTC)
Z-man, I won't ask you to forgive my ignorance, since I often display it, and I'm sure I've seen it before, but could you point me to a specific page history where I could see "Rolled back" or whatever in the edit summary? I'm sure I've seen it, I'd just like to see a specific example (a multi-rollback hopefully). That might help with some of my concerns. Thanx! Franamax (talk) 04:28, 7 January 2008 (UTC)
This page has it a few times. Rollback uses the "Reverted edits by..." summary. The summary can be changed at MediaWiki:Revertpage. Mr.Z-man 04:36, 7 January 2008 (UTC)
(copied down here from Dragonhawk's fix) What would a trial run prove? Wouldn't the trial involve the most carefully selected users acting on their best behaviour? Or would there be deliberate selection of careless users to balance it out? The concern here is the long-term possibility of misuse and especially careless use, a trial won't establish that potential. Franamax (talk) 01:49, 7 January 2008 (UTC)
If we "select" the users for a trial, that defeats the purpose. If we were to do it, I'd post a notice in a few places, indicate that it may only be temporary, and see who's interested, maybe allow the first few dozen people who respond to apply. If choose users from that sample based on the same reasons we would choose users on a full scale use (no history of edit wars, previous reverting experience, etc.), it should return helpful results. Mr.Z-man 03:21, 8 January 2008 (UTC)
Yah. If there is a trial run (not saying there should or shouldn't be one), it should attempt to reproduce actual usage. Even better than a simple first-come cut-off might be random selection. Create "Requests for rollback" or whatever. Let the requests pour in (or dribble in, as the case may be). From all the respondents, pick a random sample, and copy them to some other page ("Requests for rollback/Samples" or whatever). (Prolly want to protect that page to keep people from tampering.) Have all the interested admins watch that page (and ignore the main page), and respond to requests as they would if this proposal was in full force. Sit back and watch. If it's a disaster, cancel the whole thing. If it seems to be working, we can keep repeating the sampling process to broaden the trial. If it keeps working, throw the doors open for everyone. Adjust "rules" and procedures as needed. • Just make sure the trial protocol is clearly spelled out on the main requests page, so that petitioners understand that it may be suspended/canceled at any time, they may not be picked right away, etc. • Of course, all that might be overkill. WP:BURO. • Again, not supporting or opposing a trial myself (in this message), just commenting on how protocol might work if there is a trial. —DragonHawk (talk|hist) 04:54, 8 January 2008 (UTC)


I am new to Wikipedia and it is fantastic but a rollback? I am wondering if an ability for Non-Admins. to make easy changes is a good idea? I have read the arguements about editing wars and vandals but what about concerns of losing good editors. If it is really easy to "undo" someones work, and this happens on a large enough scale, will this cause editors or someone qualified that can edit, or that does edit, to stop for fear that a lot of work may be wasted? This seems like a logical point to ponder. I understand that editing can be done anyway but not by just pushing a button. It seems to me, on the face, that a rollback would be a fantastic idea when used correctly--BUT-- how many times will it be abused? Will a person editing a page have to keep copies of his(or her) work and a watch on every page he(or she) edits to make sure "rollbacks" have not been performed incorrectly or unjustly? I do not have enough knowledge of the subject but if Editors police pages and change the works of vandals can vandals then, at least until caught, be able to just log on and push a button also? Oh! I am sure that an IP address is recorded but if someone just fooling around has dial up then the IP address changes. If a user name surfaces more than once it can be stopped from being allowed to use the feature but how hard would it be to just create another name? I am just asking what appears to me to be common sense questions so I hope I do not offend anyone by any lack of knowledge I may have on the subject. There are a lot of individuals that have the time to edit pages for correct reasons, and spend a lot of time doing so, but I can imagine there are a lot of individuals with nothing to do with their time but create problems. Will this be an easy tool for them?Otr500 (talk) 03:41, 7 January 2008 (UTC)


It is common practice in debates to discount brand new users due to the risk of sock puppetry and meat puppetry. Due to the notice above our watchlists we are getting real new users as well. I don't have the time to do it myself, but before considering the results of this poll it may be beneficial to examine the contribution history and account creation dates of those who gave opinions.

If these people are predominantly sincere new users then there should be similar numbers on both sides of the argument, with a bias towards being less experienced with Wikipedia. However if people are using sock/meat puppetry then there may very well be a strong person bias evident in the side the new accounts are taking.

Regardless it is long standing practice to give greater weight to established contributors in debates about how Wikipedia should be ran. 1 != 2 18:05, 7 January 2008 (UTC)

Is that the practice? Never heard of that. Also I find your choice of section header to be troubling in a number of very obvious ways. This is a discussion, not some weighted vote. Equazcion /C 04:46, 8 Jan 2008 (UTC)
Check out Articles for deletions and requests for adminship, read WP:SPA. The wording in the header is how I have seen the issue being refereed to in discussions in the past. I really don't see why you find it troubling. We do weigh opinions, I never said anything about a vote. 1 != 2 04:50, 8 January 2008 (UTC)
Suffrage = vote. Equazcion /C 14:13, 8 Jan 2008 (UTC)
It's more of a bias against brand-new accounts. Such accounts, one reasonably assumes, would not have had much time to learn about Wikipedia culture and customs. So how is it they come immediately to a discussion with an opinion? The concern is these people were recruited by others. Hence WP:SOCK and WP:MEAT. I suppose this is, to some extent, assuming bad faith. But a proper analysis of the poll will examine not just responses, but what they say, and how they say it, with things like numbers and longevity a secondary factor anyway. Someone saying "Support/oppose per <insert user name here>" isn't adding much to the discussion regardless of the age of their account; while a well-reasoned, novel idea from a total newbie might be noticed. • I suspect this question is academic, anyway, as from everything I've seen, the responses were fairly consistent in their patterns from the start. Certainly, the ratio of comments in the "Support" section to those in the "Oppose" section held steady at a little over 2-to-1 the entire time. —DragonHawk (talk|hist) 05:08, 8 January 2008 (UTC)

Why no prerequisites? I propose one.[edit]

If it will be a major benefit to those who fight vandalism, then why there are no prerequisites? A user should be a long term vandalism fighter, i.e. he should have at least X reverts of vandalism every month/day/week during Y months/days/weeks to apply for the tool. X and Y can be small. Any thoughts?   Jhony  |  Talk   21:54, 7 January 2008 (UTC)

it ALSO doesnt look into the small but signifiacnt amount of users who toe the rules but use any ambiguirty or vagueness in order to justify their ongoing totalitarian campaign at stifling alternative theories and promoting drug company and anti-spiritual doctrine. Smith Jones (talk) 02:03, 7 January 2008 (UTC)
Smith Jones, your suggestion that anti-spiritual totalitarians would be in position to exploit this proposed tool seems rather a stretch. Wouldn't they be better off entrenching themselves as admins with privileged access to the existing tool? Arguments about specific malicious groups don't necessarily contribute to the wider discussion and may actually detract from the wider concerns being raised by the opposers. Franamax (talk) 02:46, 7 January 2008 (UTC)
the thing is that many of theise pople hae no interest in becing admins becauses then they they will be more visible and their insidious influence would be easierto catch beside sthey probably lack the qualifications since not everyone is made an admin without checks since they look for your editting record which as I mentioned before is often suspectious. Granting thim this power with no oversight would invariabl ylead to an onslaught against parapsychology and articles relating to faith and spirituality that is absolutely unconsciousnable. Smith Jones (talk) 02:56, 7 January 2008 (UTC)
Do you have any evidence to support your claim that granting rollback to non-admins would lead to an onslaught against said articles? If so, please present it. —DragonHawk (talk|hist) 03:22, 7 January 2008 (UTC)
Seriously, please stop assuming such bad faith on the part of so many users. The oversight is the same as the oversight for any other edit or public action. If you see something wrong, report it. Mr.Z-man 03:36, 7 January 2008 (UTC)
my evidence can essentilly consist of a look at this page. all of my recent contributions have been subjected to a cosntant barath of attacks on my character and my fitness to edit wikipedia. i have been subjected to my fellow editosr trooping up and own all over my user page, threateneng me with adminitsrative action and trying to control every last detail of my contributions here, up and including my right to achive my discussions on my very own talkpage. i know exactly why i am being targetted for this harasment, because I REFUSED to backdown from my defence of alternative scientific leaders such as Kevin Trudeau whose muckraking book can and should be purchased here. i have great respect for wikipedia and all the people who contribute to its, but i am not so naive that i dont recognize a threat to sicnetific and factual integrity wherever i see it. Smith Jones (talk) 04:05, 7 January 2008 (UTC)
At most, I might see that as evidence that your own edits will be reverted more often. That is not evidence of a widespread onslaught against all edits to such articles, merely your own edits. Now, that might be a problem, too, but it's not the same thing. —DragonHawk (talk|hist) 04:24, 7 January 2008 (UTC)
i think you midunderstand me DragonHawk. my edits are revrted and if tehre is a real reason then there is no problem. the problem exists when editors decide to stalk all my edits, reverting with them without even asking why and even going onto my own talk page to spam it with constant flaming. the problme is that when i attempt to discus podsible edits on the talk page of certain dissident science related articles my good faith is attacked andi am treated as if i am a common vandla for daring to not be in complete agreemnt with ever ysingle person who edits in wikipedia. THAT is a problem. Smith Jones (talk) 00:09, 8 January 2008 (UTC)
My point was: That is not a problem with this proposed feature, that is a problem between you and the other editors. This proposal is not the place to raise your complaints about other editors. • To address your complaints about other editors, I did try engaging you in dialog at your talk page, but you deleted my message, so... ~shrug~ —DragonHawk (talk|hist) 02:44, 8 January 2008 (UTC)

Something to think about[edit]

What about the new Wikipedian who wants to do great things for WP but doesn't take the time to learn about policies and templates and usage? I had someone try to speedy delete a Wiktionary redirect because it was too short and they claimed, not notable...its a WP TEMPLATE fer cryin' out loud, its a long as it needs to whats to stop this person from just deleting anything they don't understand/haven't learned about? Errors made with good intentions are still errors that undermine the credibilty we are all working so hard to build.

Will there be a 6 month or 12 month "waiting period" before they can get this tool? Maybe its not such a good idea to have this in the hands of people who aren't quite as accountable as admins so that admins don't have to go behind these people and clean up messes. Let them take the time to learn to use Twinkle or one of the other Vandal Fighters, but giving them rollback would be double plus ungood. Legotech (talk) 00:28, 8 January 2008 (UTC)

Why? We aren't going to give it to everyone, but it sure doesn't take 6 months to get used to vandalism reversion. Users are as accountable as you want to make them. If you want to call out a user for every mistake they make or let a few things slide, that is your prerogative. Mr.Z-man 08:18, 8 January 2008 (UTC)

Again: What's the big deal?[edit]

Rollback is an anti-vandalism tool, undoing edits is an editorial tool. I only rollback if there is blatant vandalism, anything else I undo, because I don't want to use my admin priviliges (on wikinews) for anything else than blatant vandalism.

Is there a real need to provide non-administrators with a new but weak and only slightly faster anti-vandalism tool? No.

Dedicated vandal fighters can install tools and request adminship (which shouldn't be such a witchhunt). This proposal is not going to make a big difference, it is prone to abuse, and it goes against all previous logic. --Steven Fruitsmaak (Reply) 01:00, 8 January 2008 (UTC)

Well, for those of us running out of time, like myself, who are principal contributors and don't want to give up what we've been working hard on, it's a great tool. No big deal here ;D BlackPearl14Pirate Lord-ess 01:10, 8 January 2008 (UTC)

Count the Mistaken[edit]

Quite a number of the Oppose votes seem to think that the proposal is going to make vandalism easier. This is a misunderstanding that has been countered numerous times above, but still keeps occuring. I think that someone should keep a count of the number that have given this as the ONLY reason to oppose. This could be quite significant in deciding whether or not there is consensus. Ideally, someone neutral (ie who has not voted) should count but I will do it myself if others think this is worthwile and fair. SpinningSpark 11:44, 6 January 2008 (UTC)

It's not a misunderstanding, it's an opinion. Thinking rollbacks wouldn't make vandalism easier is also an opinion. It's an ongoing debate, and you can't just declare that the people who disagree with you are mistaken. Equazcion /C 13:56, 6 Jan 2008 (UTC)
I am sure that whoever interprets this debate will discount opinions based on factually incorrect information. I fail to see how having to gain a reputation as a responsible editor then apply for a special button makes vandalism easier when you could just press 3 buttons to get the same effect. Even if you did you would lose that rollback very fast. 1 != 2 16:23, 6 January 2008 (UTC)
how many times has this been hashed out over the course of this page... it's a debate, theres nothing factual about either side's stance. The only way you'll know is with time, if it gets implemented. Probably not even then. Equazcion /C 16:27, 6 Jan 2008 (UTC)
In order that to be factually incorrect, one would need to demonstrate prima facie evidence that the implementation of this proposal could not make vandalism easier. As this proposal has never been implemented, I can't see how such evidence could be found to exist. Therefore, you're making an argument, not stating a fact. It seems a reasonable argument to me, but others obviously disagree, and I can see where they are coming from, too. So please don't declare your own viewpoint to be a fact. —DragonHawk (talk|hist) 22:25, 6 January 2008 (UTC)
If someone thinks a button is going to do something it cannot technically do, then that is a factual error. To give it equal weight would turn this debate into a vote. 1 != 2 16:30, 6 January 2008 (UTC)
If someone says it would make vandalism easier they don't necessarily have any misunderstanding about what the button technically does. The question is whether or not that functionality will make vandalism easier, and that is something that is a matter of opinion and remains to be seen. Equazcion /C 17:36, 6 Jan 2008 (UTC)
I've been interpreting such comments as either an assumption of bad faith of the admins granting it (that we'll give it to vandals) or the users using it (they will get it and just go crazy) or as a sign that they did not fully read the proposal and think this is still like the original proposal to give it to everyone. Either way, the amount of bad faith assumptions I have seen here is shocking. Mr.Z-man 03:31, 7 January 2008 (UTC)
Being worried about users abusing the tool ("going crazy") is a legitimate concern and not necessarily a bad-faith assumption. Firstly, users can be abusive without intending to be. They fall into immature battles with each other over how to best handle an article, a group of articles, or even an entire project. They're just reverting to the way they think is best, which is still good-faith. Also, "assume good faith" applies to judging users' actions after the fact -- it does not apply when we're trying to figure out what tools to enable users with. If we had to assume good faith there, then we would be immediately granting everyone full developer privileges. But we don't. We assume that if a user takes a shot that he had good reason, but we don't hand loaded guns out to everyone. Equazcion /C 05:32, 7 Jan 2008 (UTC)
Being worried that it might be abused/misused by some is fine and quite realistic. Some of the comments however suggest that if this were to be implemented that significant numbers of users would go on rampages rolling back legitimate edits, trying to exert control, and chasing users off the project through careless and/or abusive reverting. Mr.Z-man 08:45, 7 January 2008 (UTC)
Z-MAN , the reaosn whe beleive this to be so is because this is exactly what happens now iwth the tools already available. editing parapsychology an dissident-science related articles is a headache and a knightmare since eery constructive edit is automatically altered beyond recognition, deleted entirely, or reverted so fa rback that is destroys not only the reverted edit but almost every edit immediatle y preceding it. MOST people are responsible with the tools, but the few that therte are -- the few that happened to cluster around certain areas of wikipedia like maggots on rotten meat - that they are ztye ones who will use this ability to wage editrs wars that may or may not be valid in intent but can have the ia unintentionalt sidefefct of driving away legitimate editors. Smith Jones (talk) 03:31, 8 January 2008 (UTC) Smith Jones (talk) 03:31, 8 January 2008 (UTC)
I'm not saying there aren't idiots and POV pushers on Wikipedia, but the presence of a few idiots is no reason to deny a helpful tool from numerous helpful users. If users are making POV reverts, they aren't going to get rollback. Mr.Z-man 08:14, 8 January 2008 (UTC)
Vandalism is just editing in bad faith. Since it is unlikely that any software tool can distinguish between good and bad faith, it is inevitable that any tool that helps editing will also help vandalism. However, I believe that most vandals are casual users who don't take the time to learn all of these features - so I strongly suspect that the existance of this tool will give more power to the good guys than it does to the bad guys. But making it such that an admin has to see a bunch of good editing history before turning it on should defeat most vandals. Almost all attacks that I revert are from IP accounts and almost all of those are from people with either a very short editing history - or one that is 100% vandalism. Since (I presume) this privilage won't be granted to IP accounts - I don't see that abuse FOR THE PURPOSES OF VANDALISM is at all likely. Abuse in other areas (such as revert wars) is possible however. SteveBaker (talk) 15:24, 7 January 2008 (UTC)

The argument that this tool will increase vandalism seems to ignore the basic way that this is to be implemented. The default for users is no rollback right. It takes an administrator to grant the use of this tool. If an administrator starts, for example, granting rollback on request from new users, with no review, that admin could be called on it. It's unlikely to happen, though possible. Rollback doesn't really help much with *creating* vandalism. The argument that rollback will increase vandalism would also apply to granting rollback to administrators. They are merely a group of editors who have attained, at some point, the trust of the community, and they are trusted to use their tools properly. Sometimes they don't, but we do not therefore conclude that nobody but, say, developers should have the tools. What this proposal does, essentially, is to grant a right to administrators, but one much less hazardous than the rights they already have, the right to *delegate* use of the rather minor rollback tool. Do we trust administrators to use this right wisely? I know I do, and I know that the rare exceptions can be addressed specifically. Administrators are unlikely to grant rollback rights to users who are likely to offend. Mistakes will be made, but the real question is how often they will be made. Don't trust administrators? Conclude that they will commonly err, and vote no for this proposal. But even then -- exactly how does this make vandalism easier without at the same time making it easier to fix? Which would you rather revert, one rollback edit -- any editor can do this -- or two or three individual reverts? --Abd (talk) 16:42, 7 January 2008 (UTC)

It ain't broke, don't fix it - but what about a little improvement!![edit]

Let me also add some weight against the argument "if it aint broke dont fix it". Well just because something is not broken, does not mean it does not need an upgrade. Hell whats wrong with swords and arrows? Nothing, they were excellent for warfare and for at least a hundred years between 1350 until 1450, arrows were better than guns - upgrades like this are still essential. We don't and should not wait for something to break before changing. Tourskin (talk) 03:59, 7 January 2008 (UTC)

Poll closure revisit[edit]

As Equazcion remarked above, I think at this point, we've gotten as much out of this poll as we're going to. Response rate seems to be falling off, and I haven't seen anything in the way of new and unique ideas for some time. Are there any objections to closing the poll and then moving on with whatever happens next? • I would ask that people please, for the love of Jimbo, save discussion about how to interpret the poll, what should be done next, etc., etc., for another thread/section.  :-) —DragonHawk (talk|hist) 03:08, 8 January 2008 (UTC)

NO we must continue to market the poll until a specifiac date in the future, preferable within 2-3 days from today. this weill enable any last streagglesrs to get htier opinions in before a drastic decision is made that could easily destroy wikipedia. Smith Jones (talk) 03:21, 8 January 2008 (UTC)
Yes, for the love of crumb cake, close it already. I don't see any difference in outcome between doing it now and doing it 3 days from now or even a month from now. We're not gonna get any more out of this than we already have. Equazcion /C 04:42, 8 Jan 2008 (UTC)
I agree, the poll has achieved its goal. The ratio has been consistent for days and no new arguments have come about in some time. Lets close this thing and determine consensus. 1 != 2 04:44, 8 January 2008 (UTC)
(...not that the ratio is of any consequence...) Equazcion /C 04:53, 8 Jan 2008 (UTC)
I ask that people please argue discuss how to interpret the poll in a different thread. ;-) —DragonHawk (talk|hist) 04:56, 8 January 2008 (UTC)
He started it :) Equazcion /C 05:01, 8 Jan 2008 (UTC)
Sweaking of which, has anyone summoned Jimbo to comment on this? Marlith T/C 05:05, 8 January 2008 (UTC)
I would've done it but I keep forgetting the incantation. He probably wouldn't comment anyway except through a sock, for fear of influencing the outcome too much, and if so then he probably already did. In any event, he does have a watchlist, so he is probably aware of this issue. Equazcion /C 05:14, 8 Jan 2008 (UTC)

Strong Support per above. (sorry, I couldn't resist). Lets close the <expletive> thing, merge the sub-proposals into this talk, or into the new proposal, and begin to edit the current proposal based on the consensus taken so far. I think that too many editors said "oppose" while citing one little issue, which many agreed should be fixed, when they could have discussed it, changed it, and said "support." I agree with the above, that while there is no consensus for how it should be implemented, there is a rough consensus that it should be implemented, and we should there. It might (or might not) help to focus on each individual aspect of the proposal at a time, reach consensus on that, and move on (or have section specific talk on this page). What I do know is that we are getting nowhere right now, discussion is all over the place, and we need to consolidate and reorganize (I sound like a mergist, not the deletionist I usually am). Also, I think most (if not all) discussion up to this point should be archived.<semi-serious>Oh, and as I see it, this small group of people, who have contributed the most to the discussion, have reached a consensus to get rid of the poll, so can we just get it done with, or shall we take another <expletive> poll on whether or not to do it? I swear, its feels like we are in the <insert inefficient democratic legislative body>.</semi-serious>--Vox Rationis (Talk | contribs) 05:59, 8 January 2008 (UTC)

Haha. Classic. The 'opposers' should have discussed it and 'supported.' Strong! But no. the_undertow talk 06:05, 8 January 2008 (UTC)
Hey! Can't you read? "save discussion about how to interpret the poll [and] what should be done next". Honestly, no one pays attention to the OP these days. Back in my day... Equazcion /C 06:06, 8 Jan 2008 (UTC)
Maybe someone should remove the watchlist advertisement. Equazcion /C 06:47, 8 Jan 2008 (UTC)

Based on discussion here and elsewhere, I decided to go ahead and mark the main straw poll, along with the first two alternative proposals and their pseudo-polls, along with the archived votes, all as "archived/do not modify". Please note that I, and I certainly don't think anyone else, considers this discussion to be closed. But enough with the support/oppose me-too-ism. —DragonHawk (talk|hist) 06:56, 8 January 2008 (UTC)

Another idea[edit]

I had this idea today, when I was thinking about how rollback would be more efficient. Would it be possible to enable rollback for everyone, but not include it in the user interface? Basically, one would still need to use a script, app, whatever to preform the actions (as they would serve as the user interface), just as it is now, but at the very least those actions would be less taxing on the servers. I think this might be a good option if other proposals fail, since it wouldn't change anything, except making what already happens more efficient. Thoughts? Would this even be possible? -- Ned Scott 07:03, 8 January 2008 (UTC)

It would be very easy though just a tad bit hackish. Dropping something like the following in LocalSettings.php should do the trick:
if ((((isset($_REQUEST['action'])) and ($_REQUEST['action'] == 'rollback')) or (fnmatch('*api.php*',$_SERVER['REQUEST_URI']))) and ($wgDBname == 'enwiki')) $wgGroupPermissions['user']['rollback'] = true;
-- Cobi(t|c|b) 07:38, 8 January 2008 (UTC)
I like it. I had suggested something similar, that the rollback function be given to everyone but that they'd need to "turn it on" from preferences. This is even better though. Equazcion /C 12:54, 8 Jan 2008 (UTC)

I suggested this at the psuedo-RFC, but only one person has agreed so far (or, as far as I know, noticed). If there is consensus to implement, I think this would be a good idea. It is a compromise, however, and compromises make no one happy :P GracenotesT § 16:21, 8 January 2008 (UTC)

I'm thinking of adding this to the main proposal page for people to vot.. poll on.. or whatever. -- Ned Scott 08:48, 9 January 2008 (UTC)
Rather, I think most people will be happy. The main implementation will inevitably be TWINKLE, and as TWINKLE is already in existence, (and I don't think that there would ever be consensus to remove it) this proposal simply becomes a minor modification (see comments under support on main proposal page). The antibureaucracy people will be happy, the anti-admin people will be happy, the user-scripts-are-good people will be happy, the RC-patrollers-in-support people will be happy, the experience-needed people will be happy, the just-another-RfA people will be happy, and probably others will be happy. Methinks the only people who won't be happy, is those worried about rollback-abuse, and since TWINKLE et al already exists, they are out of luck. --Vox Rationis (Talk | contribs) 14:51, 9 January 2008 (UTC)

Where to go from here[edit]

Okay. In same space of about eight days, we've had the main proposal (including its original form, which evolved into the current form, and the accompanying straw poll, which itself includes around 450 total top-level responses (not counting replies) and over 200 kilobytes of text), at least three counter-proposals, and various assorted discussions (on the project page and at this talk page (some now archived)), those also consisting of over 200 kilobytes of text. And before that, we had Wikipedia:Rollback for non-administrators proposal, Wikipedia:Rollback for non-administrators, and Wikipedia:Requests for rollback privileges and their accompanying talk pages, polls, archives, etc.

Now what do we do?

Some suggestions:

I think the ideal thing to do would be to refactor all the proposals, discussion, polls, responses, etc., into something as concise as possible. There's been a massive amount of text put forward on this, and until we know where we've been, I'm not sure we can really know where we should be going. That said, it will probably take a fair bit of effort just to review, summarize, and refine all that, and I don't expect the world to hold still in the mean time.

So, concurrently: I think we should strongly encourage discussion and consensus over simple voting. Avoid polarizing things into "Support" or "Oppose". Make sure any proposal fairly and completely presents all the pros and cons. Make sure objections (and counter-arguments, and counter-counter-...) are noted. Avoid bias in favor of any particular stance. Refactor it mercilessly. The idea is to get the proposal to a concise state that everybody agrees is an accurate presentation of all sides of the issue. In short, write the proposal like a Wikipedia article. The idea is to avoid legions of people posting "Support" or "Oppose" comments solely because they want their opinion to count. If the proposal addresses their opinions directly, they are much less likely to post a simple WP:ILIKEIT or WP:IDONTLIKEIT comment; they'll be content with the presentation "as is".

If consensus on what the "right" proposal is simply cannot be archived, despite long discussion, I'd say the next best thing would be to get consensus on a presentation of the issues (pros and cons), and multiple, stable proposals addressing those issues. If there must be a poll, people can then vote to indicate their support for a given proposal, or rejection of all of them, rather than trying to get their opinion expressed in a sea of almost 500 other opinions.

Comments? Suggestions? Other ideas? Condemnation?

DragonHawk (talk|hist) 07:35, 8 January 2008 (UTC)

FYI, bugzilla:12534. Speaking of which, the final contents of the proposal should be in a developer-friendly format because the devs have better things to do (hint hint) than to read paragraphs of text. We need to demonstrate that there's consensus for the proposal. MER-C 10:59, 8 January 2008 (UTC)

No consensus[edit]

It seems pretty clear that there is not consensus on this proposal. I think what we need to do is get whoever has the authority to tag this page {{failed}} so that people can go "back to the drawing board" and propose an entirely different solution to what they see as the problem they want to address. Sdedeo (tips) 15:48, 8 January 2008 (UTC)

Are you joking? 100% more than the number of opposers supported this. We've asked the devs to take a look and let them judge consensu, as they are the ones that would be implementing it. Ryan Postlethwaite 15:51, 8 January 2008 (UTC)
Consensus is not a 2:1 in favor. If you want to wait for the devs to judge consensus, I suppose that's OK. Sdedeo (tips) 15:53, 8 January 2008 (UTC)
As I said, they're already looking into it. 2:1 certainly isn't consensus in favour of not giving out rollback. I think it's quite clear consensus in favour. Ryan Postlethwaite 15:55, 8 January 2008 (UTC)
Repeatedly pointing at the 2:1 ratio does not demonstrate consensus. As we all know, "Wikipedia is not a democracy". Majority opinion doesn't necessarily rule, especially if dissenters have legitimate concerns. Now, maybe the dissenters don't have legit concerns, but pointing at the ratio does nothing to demonstrate that. —DragonHawk (talk|hist) 00:25, 9 January 2008 (UTC)
Agreed on all counts. I've seen many, many, many Wiki debates where 2:1 is not believed to demonstrate consensus, and indeed it does not. If 1/3rd of your population is opposed, that's definitely not a consensus. There's a mysterious bias at work here for reasons unknown. Are we changing Wiki into a voting democracy now? If so, that's a fairly sweeping change. --Bishop2 (talk) 15:39, 10 January 2008 (UTC)

I'm surprised you think so especially given the fact that you're an administrator! The AfD rule-of-thumb is that you need excellent reasons to decide consensus on anything weaker than 3:1, for example. I just don't see those reasons here; reasonable people have weighed in on both sides. Sdedeo (tips) 16:00, 8 January 2008 (UTC)

That's an incorrect interpretation of consensus. Ryan Postlethwaite 16:06, 8 January 2008 (UTC)
No, it's not, but thanks for elaborating. Sdedeo (tips) 16:13, 8 January 2008 (UTC)
Am I wrong, or ought there to be a certain bias (just a slight one) towards the more libertarian side, i.e. the one supporting the delegation of tools to more editors? I mean, it is a part of Wikipedia's character as an open encyclopaedia. Waltham, The Duke of 16:14, 8 January 2008 (UTC)
At least with AFD, vote count plays a major factor only in unanimous or near-unanimous (10:1) situations. After that the strength of the arguments are very important. Mr.Z-man 20:22, 8 January 2008 (UTC)

The way I've seen things "go down" in consensus failures is one of four things:

  1. The case was not made. i.e., the proposal should have passed, but the urgency or importance was not made clear.
  2. The proposal is a good idea but does not address complicating factors in the implementation. e.g., not newbie-friendly, not watertight enough against vandalism, &c.
  3. The requires violating some more important principle. e.g., some part of WP:ISNOT, &c.
  4. The proposal was foolish to begin with, e.g., unworkable, inconsistent, &c.

In this situation it seems like a combination of the first three; it's not prima facie bad. The oppose remarks seemed to center on:

  1. We already sort of have the tools anyway. (1)
  2. Too much potential for abuse. (2)
  3. Additional stages of bureaucracy for certification is problematic. (3)

Sdedeo (tips) 16:27, 8 January 2008 (UTC)

I thought there would be no bureaucracy involved. If this proposal passes, certification will be a very simple and rather unofficial process. Unless I am missing something here. Waltham, The Duke of 16:41, 8 January 2008 (UTC)
Well, it could be argued that the proposal is perceived to be bound to create additional bureaucratic load, in which case your point is probably correct. Waltham, The Duke of 16:44, 8 January 2008 (UTC)
Notice how it is those in the minority opinion that claim there is no consensus? There is a very nice consensus. 1 != 2 17:13, 8 January 2008 (UTC)
Notice how it is those in the majority opinion that claim there is consensus. 1!=2, it's fine if you think there's consensus. But please stop declaring it as an obvious fact. Whatever the result of the poll, it was anything but obvious, and your opinion is not fact. Equazcion /C 17:19, 8 Jan 2008 (UTC)
It could be you are looking at a different poll, it could be that I have more experience judging consensus than you. But it is an obvious fact. The opposition is not representative of the communities wishes. The arguments for opposition are fragmented and often logically flawed. The community wants this proposal to pass, and it would be foolish to let a minority withhold this tool. The question you need to ask is "Is there a consensus to cancel the planned implementation of this proposal", I don't see one. I see a consensus to continue forward with it. You might think I am wrong, but I don't think I am. 1 != 2 17:23, 8 January 2008 (UTC)
Regardless, the poll is closed now, so lets just wait and see. Peace. 1 != 2 17:25, 8 January 2008 (UTC)
I'm not sure what meaning you are using for "obvious fact".  :) To me, an obvious fact is something which can be decided quickly, objectively, and unequivocally. For example: "The ratio of comments under 'Support' vs those under 'Oppose' is 2:1" is an obvious fact. But consensus? I think judging any kind of consensus is, almost by definition, not going to be an "obvious fact". It's going to be an analysis. Facts aren't declared by fiat. —DragonHawk (talk|hist) 00:21, 9 January 2008 (UTC)

The main argument of Ryan and 1!=2 here is that they simply know what consensus is because they are very advanced users. 1!=2 adds that the opposition is "logically" flawed, although none of the three reasons I suggest as the main sources of opposition are logical in form. Simply because 30% of the community has a wide spectrum of reasons for opposition is not a reason to ignore them on the basis of "fragmentation." Anyway, my guess given the way things have happened in the past is that this won't go through. Peace. Sdedeo (tips) 17:32, 8 January 2008 (UTC)

  • Fragmented, what does that mean? If opposers each have different reasons for opposing, their side is wrong? Even if so, the supporters are equally "fragmented" and their arguments equally "logically flawed".
  • "Is there a consensus to cancel the planned implementation of this proposal" -- That is not the question. There is no planned implementation until the proposal is accepted. In order for a proposal to be rejected, there simply needs to be a lack of consensus. Proposals don't need "consensus rejection". That's not how it works.
  • "He who knows nothing, and knows he knows nothing, is truly wise." Have the courage to admit that your biased outlook renders you incapable of declaring the facts. I have. Equazcion /C 17:39, 8 Jan 2008 (UTC)
    • Two in one: a quotation, and a paradox. If he knows nothing, then how does he know that he knows nothing?
    • In any case, it would be extremely interesting if we could break up the votes (all right, "votes", if you insist) by groups of users. Think about it; we could examine separately what the administrators think in general about handing out rollback privileges, what vandal-fighters believe about the tool's availability's implications, what vandals think about the tool's potentials... The whole thing would be much more organised then. Waltham, The Duke of 17:27, 9 January 2008 (UTC)

Screening candidates for rollback rights[edit]

I do not know whether it has been noted in the past, but it appears that hardly any vandals take any time or pains to embellish their signatures. Fancy signatures could, therefore, be argued to be a trademark of legitimate editors (perhaps with metapaedic tendencies). Consequently, it makes sense to propose the following: why not give rollback only to those editors with impressive, colourful, or otherwise interesting signatures? We could call it the Signature Criterion. This ought to simplify things a lot.

Not persuaded, are you? Ah, yes, I suppose this suggestion is mostly humourous (seriously, though, the observation I have based it upon is not). Well, I am simply trying to keep the atmosphere here light and friendly. Please don't shoot me. There is a strange gleam in your eyes...

Although the page is over-burdened anyway, I will accept to delete this message if enough editors agree. You can hold a poll for this below if you want to. Waltham, The Duke of 15:55, 8 January 2008 (UTC)

Some of us edit and fight vandalism without seeing any need to embellish our signatures, so should we not get extra tools to help in the fight against vandalism? Jons63 (talk) 21:28, 8 January 2008 (UTC)
No. :-D
Basically, I am mostly saying the opposite: not that all legitimate editors have fancy signatures, but that most vandals don't. Waltham, The Duke of 17:18, 9 January 2008 (UTC)


Hi. Sorry, I oppose this entire idea. No further privileges-- no further editing shortcuts. I feel that all the tools which we have right now are fully sufficient. sorry to the good-faith editors who may support this. Frankly, i wish the idea to distribute the capability widely or collectively would just be discontinued from discussion. thanks. --Steve, Sm8900 (talk) 16:08, 8 January 2008 (UTC)

You are the first to say this. Waltham, The Duke of 16:11, 8 January 2008 (UTC)
I guess so. obviously, I do not mind disagreement with this, or woth any idea of mine. I appreciate your reply. If you wish, please feel free to discuss, of course. thanks. --Steve, Sm8900 (talk) 16:12, 8 January 2008 (UTC)
Please, elaborate. Did you or did you not realise that I was being sarcastic?
Not that I was trying to insult you or anyone else; I just felt mischievous at the moment. I am perfectly willing to continue proper discussion now. Waltham, The Duke of 16:17, 8 January 2008 (UTC)
Oh, that's funny. No I don't mind a bit of occasional irony, droll wit, or anything else. :-) quite alright. actually i was serious in my reply there, as i really was not sure whether i was actually the first to oppose this completely and unequivocally, or whether others felt similarly. so actually i do aprpeciate your reply, which helped me just by providing a bit of context. so thanks. feel free to drop in or to comment any time. thanks. --Steve, Sm8900 (talk) 16:26, 8 January 2008 (UTC)
So you are opposed to selective rollback granting because it's not egalitarian? Wikipedia isn't an experiment in democracy, but it's not one in socialism either :) Certain editors assume certain functions – especially in the case of systematic vandalism reversion – and Wikipedia works very well this way. A shortcut is not necessarily a bad thing. GracenotesT § 16:39, 8 January 2008 (UTC)
Errr, you;'ve actually managed to construe my idea exactly the opposite of what i meant. no hard feelings, as I realize it wasnn't intentional. What i meant is, i feel that no new wholesale edit privileges should be granted to all non-admin editors or to any wide group of ordinary editors. When i said i oppose it, i meant that i oppose the existence of non-adminsitrator rollback capability. period. so actually i agree with your comment. there, now wasn't that the easiest talk page colloquy you've ever been in? :-) thanks. --Steve, Sm8900 (talk) 17:02, 8 January 2008 (UTC)
Rollback is not an integral part of adminship, though. It's just a technical ability that developers decided they should put in a grab bag of other abilities known in MediaWiki software as the "sysop" user group. I think I understand your position better, but I don't understand why you think that. Why is a shortcut a bad thing? GracenotesT § 18:19, 8 January 2008 (UTC)
Okay, I've heard this argument a few times, but I don't quite get it. How are usergroups becoming a hassle? Most users only need to be concerned with one or 2 on a non-regular basis. Mr.Z-man 20:10, 8 January 2008 (UTC)
It's bad if it's given to non-admins (of whom i am one, actually). Anything which allows people to make major changes without doing muich work is bad. Anything which forces them to exert some effort in proportion to the editing which they may want to do can generally be a good thing, usually. thanks. --Steve, Sm8900 (talk) 21:52, 8 January 2008 (UTC)
Anti-vandalism efforts requires people to make major changes without exerting a large amount of efforts – compare the number of individuals who revert vandalism with those that vandalize, and especially the effort required on the parts of both. Exerting effort to make an edit is generally a good thing, and somewhat of a check on frivolous edits, but anti-vandalism is an exception to this rule. GracenotesT § 16:08, 9 January 2008 (UTC)
Do you know what they say, about the race between banknote designers and forgers, and the forgers always being one step behind? I think it's something like that here. The vandal-fighters will always be a few steps ahead of the vandals. And here is a tool that will be available to both, but who will use it better? I should like to think that, again, the "good guys" will. (Except that they are not all as good as they seem—there are some I shouldn't want to meet in a dark alley. But we are all united against the common enemy, so...) Waltham, The Duke of 17:37, 9 January 2008 (UTC)

Settling This[edit]

So let us evaluate this discussion. Did this pass or did it not? Considering that the number of supporters is double that of opposers, I beleive that the Wikipedia community is strongly supportive of this propsal. However, I note that the opposers arguements are much more detaild. However I note that they keep on repeating that "This will create a heiarchy of users" (How is this bad, it'll motivate newbies to move up the ranks) "This will cause edit wars" (Edit wars will always exist no matter what, it is like banning rock music and Wikipedia from schools. Editors will always get into disagreements). The supporters are the vandal fighters, they keep repeating about the faster reversion of vandalisim (however, why care if you beat others, it will get reverted anyway). The prevention of edit wars is more of a evaluation of the use of individualisim on Wikipedia. Let me note that collectivisim, is always a bad thing in all situations. Marlith T/C 19:57, 8 January 2008 (UTC)

Let's not vote on this, I've already told people that the developers are currently considering the consensus here, it's them that decide whether it gets implemented or not. Ryan Postlethwaite 20:00, 8 January 2008 (UTC)
You have almost exactly the same result at the no-consensus poll on this 2 years ago. Despite the discussion, the lack of consensus is static.--Docg 20:06, 8 January 2008 (UTC)
Well if that is the case, then I argue the result of two years ago was wrong. Ryan Postlethwaite 20:12, 8 January 2008 (UTC)
There is no consensus here. The developers don't need to do anything. —Ashley Y 21:17, 8 January 2008 (UTC)
I suggest there clearly is consensus. Ryan Postlethwaite 22:10, 8 January 2008 (UTC)

What happened to the proposal rewrite to reflect the consensus of the poll? Did we pass over that and move straight to dev review? It was my understanding that the devs are having little part in this discussion, that they have already made the feature, and are just waiting on our decision as to whether or not to implement it. I'll b going through shortly and adding or changing minor things to reflect consensus.--Vox Rationis (Talk | contribs) 22:45, 8 January 2008 (UTC)

I wonder what the devs are doing now. Marlith T/C 02:37, 9 January 2008 (UTC)

They are probably watching us on a big-screen plasma television as we fight with each other, munching on pop-corn and sipping red wine.
Wow, the only way you could be more dead on is if you said "wheat thins" instead of "popcorn" :) AmiDaniel (talk) 00:41, 11 January 2008 (UTC)
Ah, I so envy them. Waltham, The Duke of 17:45, 9 January 2008 (UTC)

Watchlist advertisement[edit]

The watchlist notice is causing some confusion now that the poll is closed. Perhaps the notice should be removed now? Equazcion /C 20:24, 8 Jan 2008 (UTC)

Yes check.svg Done. Cheers. --MZMcBride (talk) 21:44, 8 January 2008 (UTC)


The last poll ended with the same percentage and was declared no consensus and the idea was shelved. Why is a similar result being treated differently this time? Wikipedia:Requests for rollback privileges/Poll. Hiding T 22:00, 8 January 2008 (UTC)

How is it that 67% of the community want this proposal, but the less majority (by a considerable ammount) don't want it yet get there way? I question the application of consensus in the previous poll. Ryan Postlethwaite 22:04, 8 January 2008 (UTC)
Because of WP:CONSENSUS. You've closed discussions as no consensus, I take it. On WP:RFA we use 70-80 as a guide for promoting someone to admin, I think it is higher for a bureaucrat. Why do you want it to be lower for a software change? Hiding T 22:15, 8 January 2008 (UTC)
Also, from memory, when we polled on adding speedy deletion criteria we used 70% for the cut off for acceptance. And that was agreed beforehand and I think we didn't allow anons or new users, we had a suffrage on that. Hiding T 22:17, 8 January 2008 (UTC) <edited to insert last sentence> Hiding T 22:18, 8 January 2008 (UTC)
The community has gaged consensus for RfA and RfB - this we have no guage for. Far more community members want rollback than do not and many of the opposes are based on flawed logic that this will cause more vandalism. Look at AfD consensus - that's not a set number like you quote, just like any other process we have here (except RfA and RfB) - the developers can judge consensus for themselves, I hope they can see that an overwhelming majority of people want this. There was no real discussion on acceptance criteria. Ryan Postlethwaite 22:20, 8 January 2008 (UTC)
It's your opinion that it is flawed logic. And we're not talking about AFD, we're talking about a major change in the role of admins and the software. Is it not even beyond the realms of possibility that we should consider whether there needs to be a consensus of admins to do this? If we want to discuss the merits of the poll, I can only point to our policy, Wikipedia decision making is not based on formal vote counting ("Wikipedia is not a majoritarian democracy"). This means that polling alone is not considered a means of decision-making, and it is certainly not a binding vote, and you do not need to abide by polls per se. Polling is generally discouraged, except in specialized processes such as AFD. We're talking about implementing something which many feel violates WP:NOT, specifically not a bureaucracy. I think further discussion is certainly merited, per WP:CONSENSUS. Hiding T 22:25, 8 January 2008 (UTC)
(ec) Many of the support votes are based on flawed logic as well, and whether or not rollbacks would cause more vandalism is a matter of opinion. The vast majority wanting something doesn't mean there's consensus. Whoever makes the decision will (hopefully) go through all the votes to see if a good argument has been made for the implementation. The numbers themselves don't mean much. Equazcion /C 22:27, 8 Jan 2008 (UTC)
Just a question, how do the developers get the right to solely judge consensus on this (or any other) issue? Seems like it's the wrong way 'round. RxS (talk) 22:45, 8 January 2008 (UTC)
they get the right because the ytmade wikipedia. Smith Jones (talk) 22:46, 8 January 2008 (UTC)
Who do you propose should evaluate consensus in this case? I think that the devs are in the best position to evaluate any technical issues and will have to decide in the final analysis whether to implement this or not whatever the consensus might be. Eluchil404 (talk) 23:02, 8 January 2008 (UTC)
The same people that decide consensus on all other policy decisions...the editors (us). Developers don't make policy, they implement policy decisions. This isn't a technical issue, it's a policy debate. They certainly can have plenty of input as to how policy is implemented in a technical sense but they don't decide consensus. As far as whether this has consensus or not, seems to me that it doesn't. I do think that when consensus happens it's a lot clearer than it is here. Sort of you know it when you see it I think. There's been plenty of examples where it's taken several attempts to find a consensus on a policy proposal, off the top of my head I think WP:Prod took a couple tries before it was accepted, several speedy criteria took a couple tries. RxS (talk) 05:21, 9 January 2008 (UTC)
The devs get the right, because they are the ones who make the change - they decide if there is enough support for a particular proposal. Ryan Postlethwaite 23:00, 8 January 2008 (UTC)
So the devs can fundamentally change an admins role without consensus from admins that they want the change? Hiding T 23:03, 8 January 2008 (UTC)
This isn't soley the admins decision - it's the communities. If an admin doesn't like it, they don't have to get involved with it in any shape or form, leave it to the admins who want to. The devs will evaluate this debate, and the "poll" and all other comments then look at the technical aspects and whether or not they feel there is consensus for the change. Ryan Postlethwaite 23:07, 8 January 2008 (UTC)
That seems to contrast with your question to the crats recently. Either it's good for the goose and the gander or it isn't, you ask me. I think if a majority of admins don't want this, then that needs to at least be recognised. Hiding T 23:12, 8 January 2008 (UTC)
Well, technically speaking, the developers and server admins work for the Wikimedia Foundation, and the servers are owned by them, too, so it's really the WMF board that wields the phenomenal cosmic powers. That said, they generally abide by consensus decisions of the community. What's tricky is in situations like this one, where (according to some) consensus is not clear. I suppose in that case, somebody has to make a judgment call. —Preceding unsigned comment added by DragonHawk (talkcontribs) 02:53, 9 January 2008 (UTC)

There's really no need to quarrel here. There is no consensus at the moment (indeed we've had a stable no-consensus for a while). As long as there is no consensus for a change, the status-quo holds. That may frustrate those who are pushing for a change - but, well, sorry. That's how it is. Take it away, work it though if you want, if you make a better case and find a better method, maybe sometime in the future consensus will change. But for now, you need to accept this result.--Docg 00:32, 9 January 2008 (UTC)

No offense to anyone, but, maybe it'd be more helpful for those of us with such strong opinions (for or against) regarding this proposal not to go around proclaiming consensus one way or another. SQLQuery me! 20:29, 9 January 2008 (UTC)

Flawed logic[edit]

How do those suggesting that the oppose logic was "flawed" propose we handle one-liner support arguments? For example, "obviously :-)", "Per common sense", "proposal as written". I'm not saying there is anything wrong with those comments, but I do take issue with one side insinuating flawed logic should result in an ignored vote, when many of the support votes don't even quantify the votes. While I don't agree that some of the logic used in opposes were proposed in a good fashion, if we are going to require a certain standard for oppose votes, we must require the same standard for support votes. So should the final decision maker(s) on this one simply ignore every support vote that didn't quantify or qualify their support?
This is exactly why straw polls should be avoided. Either take each vote for face value, or set the rules on how votes are metered from the very beginning of the debate. Many of the votes in support had very valid reasons for supporting this policy. Many of the votes opposing had very valid reasons for opposing this policy. Wikipedia is not a democracy, and polling is not a substitute for discussion. Specifically, #4 states "Similarly, if a straw poll is inconclusive, or if there is disagreement about whether the question itself was unfair, the poll and its results should simply be ignored." The results of this poll are at the very least inconclusive.
My last comment will be directed at User:Ryan Postlethwaite's comment: "How is it that 67% of the community want this proposal, but the less majority (by a considerable ammount) don't want it yet get there way?" That comment stunned me. WP:CON explains exactly how, and why, it is that 67% of the community wants something but doesn't get it. I've said it once, and I'll say it again, Wikipedia is not a democracy... and as an admin, I'd expect you to know that. Justin chat 23:10, 8 January 2008 (UTC)

Justin dont be a bully, not everyone can rmember bits of the ruels off of the top of thier head. just remind him politely and dont try to make him look foolish for Askign a legitimaite question!!! Smith Jones (talk) 23:34, 8 January 2008 (UTC)
I fail to see how I was a bully at all. Admins are expected to know rules, certainly an admin is expected to know the cardinal rule of policy changes. I don't think Ryan is a fool at all, but I think his frustration with this proposal is overshadowing his common sense. Justin chat 00:17, 9 January 2008 (UTC)
tere was no reason to teaste him and implie that his is foolish because of his ignorance. as you say his frustration is overhsadowing his common sense, band not only that he is still in a humanbeing and if he may be going through things in his life that are draining his neergy he may very well have let something like this slipped his mind. you dont know ifhe is facing any deaths in hte family or the breakup of a relatioship or anything like that, so before you judgehim and imply that he is losing his coolor being unsensical you should put yourself inhis shoes becasue as the wise spiritualist once said "before you judge a manyou should put your feet inhis shoes" and thats what you shoul d do for him even thoughhe is an adminand you may or may not be id ont know. Smith Jones (talk) 02:34, 9 January 2008 (UTC)
The "flawed logic" applies to oppose comments that were totally out of line with the proposal, such as those that suggested that the proposal would give rollback to everyone or ones that say "just reform RFA and they can apply to be admins" (good luck with that). The problem was not that they didn't explain their reasoning, but rather that their reasoning made little sense in relation to the proposal (or in some cases reality). Mr.Z-man 23:55, 8 January 2008 (UTC)
Certainly you can make the case that the policy will give rollback to everyone is flawed logic. But I disagree that the latter of your suggestions is in anyway flawed logic. Alternative proposals aren't flawed... they are the benefits of discussion. While I agree your first suggestion is entirely over the top, the idea that proposing reforming RFA is "flawed logic" is in itself a logical fallacy. Consensus is reached via discussion, not voting. If you want to ignore any opposing vote that makes alternative suggestions, your effectively proposing the opposite of what WP:CONS stands for. And for the record, I opposed this policy not because I think giving editors rollback is a bad idea (I rather think it's a good idea), but because I think the way this policy would implement it is bad. Is that "flawed logic" as well? Justin chat 00:17, 9 January 2008 (UTC)
An alternative suggestion is fine, but people have been trying to reform RFA for a long time and have gotten nowhere. At this point, unless you have a suggestion on how to do that, its not really an alternative idea. Saying "reform RFA" would be like saying "I don't like this proposal, but I would support a better one." Everyone would support something better, but just saying that is not very helpful. Mr.Z-man 05:38, 9 January 2008 (UTC)
I agree. But then, I see a relatively large difference between unhelpful and logically flawed. However, I still don't see how we can require opposers to validate their response and not expect the same from supporters. Is saying "reform RFA" less helpful than "I don't want non-admins having rollback?" It's tough to set a metric for what should be considered an acceptable validation, which is exactly why voting is evil. Justin chat 06:10, 9 January 2008 (UTC)
Those who !oppose saying "you can't give this to IP editors" are certainly wrong - that is not what the proposal is asking for at all. And then there are varying degrees of misinterpretation which result in an oppose. But maybe that's not flawed logic, maybe those people are expressing their discomfort quickly - just because they aren't slapping down the pithiest comment doesn't mean their general concerns aren't valid. Similarly, those who have looked quickly and !support'ed with the comment "looks good to me" may have used some flawed logic. There is simply no way to determine this, maybe another banner notice saying "hey, go back and read it all again"? A quick poll of the general community has to either take each vote at face value or it has to then synthesize and evaluate each and every comment. Either way, it doesn't look like this proposal has achieved a clear consensus - there are obvious merits, but the current proposal hasn't been able to overcome the concerns. Franamax (talk) 07:19, 9 January 2008 (UTC)
In a few words, it is perfectly clear that it is not clear whether there is a clear consensus. Waltham, The Duke of 18:14, 9 January 2008 (UTC)

A New Record?[edit]

Does anyone remember the Wikipedia-space page for "more than 100 Wikipedians supporting any one thing"? I think the new standard is more than 200, but this still qualifies! Franamax (talk) 07:35, 9 January 2008 (UTC)

WP:100 and WP:200. -- Ned Scott 08:40, 9 January 2008 (UTC)
Thanks Ned. Why didn't you tell me that three days ago? Then I would have known I was asking three days too late! :) Franamax (talk) 09:54, 9 January 2008 (UTC)
Speaking of records: Oh, goodie, another proposal polarized into a vote. Is this a record for the number of annoying straw polls? (Nothing personal, Ned. You're hardly the first and unlikely to be the last.) —DragonHawk (talk|hist) 12:44, 9 January 2008 (UTC)
I'm not sure if it's a record, but a poll could probably determine that. Equazcion /C 12:53, 9 Jan 2008 (UTC)
We could write an essay about that: "Polls lead to pollarizing" [sic].
In any case, we have broken neither the record for the most supporters for a proposal, nor the record for the most opposers. This whole thing's a bleeding failure. Waltham, The Duke of 17:58, 9 January 2008 (UTC)


Could someone go ahead and mark this "rejected" plz? And then we can go ahead with the proper consensus thing, which is to come up with a new proposal that works in the issues raised by the opponents, if that's possible. —Ashley Y 18:56, 9 January 2008 (UTC)

Why would we rewrite in on a separate page? If you notice below the main proposal, are two new proposals for implementing this, which are both gaining good consensus. In fact, and this may be a POV, but I think that the top proposal (user scripts) will actually pass. Also, note that as polls are not a replacement of consensus, ending the poll is not a mark of ending the discussion, hence why this page is still here.--Vox Rationis (Talk | contribs) 19:27, 9 January 2008 (UTC)
What's the hurry? You can still edit a new proposal, in the meanwhile. And, as Vox mentioned, there's still activity here. SQLQuery me! 20:40, 9 January 2008 (UTC)
Fair enough. I think the main proposal is dead in the water, but the new proposals further down look like they have consensus. —Ashley Y 21:10, 9 January 2008 (UTC)
Or we can just ignore all consensus and do it anyway. I'm just curious, who made the final decision on this one? Was it the devs? I'm just a little stunned that we got absolutely no explanation as to why the policy seems to have been implemented, when it failed last time with the exact same ratio? Justin chat 00:34, 10 January 2008 (UTC)

Apparently this passed[edit]

WP:RFR is up and running. I was a little confused on what happened, but per the answer to my question it seems the developer on this one saw this as a consensus to pass. So I suppose discussing it any further is pointless. Justin chat 01:01, 10 January 2008 (UTC)

yeah, i guess this the beginning of the end of the wikipedia expierement. i guess it was fun while it lasted. Smith Jones (talk) 01:02, 10 January 2008 (UTC)

Who was the developer? Where is his decision explained? What were his criteria? Where did he announce it? To whom is he accountable? Where do I ask for a reconsideration? --Docg 01:09, 10 January 2008 (UTC)

(ec) Seconded. This change has been bulldozed through. Ryan Postlethwaite's repeated attempts above to rewrite the community standards of consensus are not convincing. — Hex (❝?!❞) 01:16, 10 January 2008 (UTC)
I don't know which developer (perhaps this was a group decision). He/she did not explain. He/she did not list his/her criteria. He/she did not announce it. And frankly, I don't know WHERE one would go to review a decision by a developer. I disagree that this was a devs decision (after all, this isn't simply a matter of adding to the software, but adding RIGHTS to a user group, which should have been up to the community and the bureaucrats). If someone wants to know the next step, I think perhaps starting from the top would be appropriate. Justin chat 01:14, 10 January 2008 (UTC)
Well, I'm considering an RfC as a starting point. Actually rollback worries me less than this disgraceful and plainly manipulative way this has happened.--Docg 01:20, 10 January 2008 (UTC)
Nuff' said. KnowledgeOfSelf | talk 01:24, 10 January 2008 (UTC)
I don't see why everyone is so surprised, there is clear favor towards doing this. 1 != 2 01:25, 10 January 2008 (UTC)
Because favor != consensus (I couldn't help myself). Justin chat 01:28, 10 January 2008 (UTC)
Yes, 1!=2, we know that you and Ryan think that. Lots of people don't think this was at all clear. Carcharoth (talk) 01:40, 10 January 2008 (UTC)
An idea: Instead of filing an RFC, which will likely accomplish little other than create more acrimony and waste a lot of editor time, how about everyone involved with this gets back to work on the encyclopedia and see if this idea actually works out or not? Nothing that's been done here is irreversible. If it works out, great, we have a new tool for helping keep the quality of the encyclopedia up while continuing to be editable by anybody and everybody. If it doesn't work out, that's fine too, it can be retracted with minimal developer effort, and a lesson will be learned.
Put another way: There's no lasting harm in trying this out, but there IS lasting harm in making a big fucking argument out of it. -/- Warren 02:08, 10 January 2008 (UTC)
The lasting harm is letting out-of-process decisions stand. I'm all up for discussing a way to implement rollback. But "consensus be damned" is not a precedent I eagerly welcome. There are a lot of proposed policies I'd love to try... but approve or disapprove I will always complain when something is done without consensus from the community. Justin chat 02:15, 10 January 2008 (UTC)
(e/c) How can you say that? They threw the long established practice of the community accepting or rejecting proposals, right out the window. The lasting harm is that this sets a precedent that is substandard and a general mockery of consensus. KnowledgeOfSelf | talk 02:17, 10 January 2008 (UTC)
At issue is the fact that a lot of editor time has been lost spinning our collective wheels on trying to obtain consensus on this issue, but let's face it: consensus is never going to be reached unless it's actually tried and we see whether it works. Being bold here isn't nearly as bad as you & others make are making it out to be. No, seriously! We've built an entire encyclopedia on the premise of being bold! If we're going to be too chickenshit to experiment with improving the tools because certain people disapprove of how such a policy is introduced, we're fucked.
Anyways, it's probably pointless for me to try and convince anyone involved in these discussions that trying out non-admin rollback isn't that big a deal, but if nothing else, have some faith that the significant majority of people who supported this proposal have weighed the pros and cons, too, and have come to the conclusion that we should try this. -/- Warren 02:37, 10 January 2008 (UTC)
I quite frankly don't give a toss if x number of people conclude it's good and want to try it. You ever hear someone say "E.N.C.Y.C.L.O.P.E.D.I.A good now go build it."? Here now say it with me "The community establishes policy with C.O.N.S.E.N.S.U.S, now go work it out." If you want to ignore the fact that it's been thrown out the door, that's fine. But it means a bit more to some people. KnowledgeOfSelf | talk 02:46, 10 January 2008 (UTC)

Seems no haste was made to put up the request page, have a bot standing by to assist in it, and even putting up a watchlist notice. Anyone else getting the feeling that they're trying to get as many users as they can to request rollback, so that it can be claimed that turning it off is a disruption? -- Ned Scott 01:55, 10 January 2008 (UTC)

Conspiracy theories? Not sure about that, I think this is just a farce the size of the Midgard Serpent. Poor form to be sure. KnowledgeOfSelf | talk 01:59, 10 January 2008 (UTC)
Not a conspiracy, but definitely trying to get their feet in the door before it closes. -- Ned Scott 02:07, 10 January 2008 (UTC)

Drafting an arbcom request for this now. Who wants to be added as a party? -- Ned Scott 02:23, 10 January 2008 (UTC)

Well, the developer who initiated it, everyone who has strongly protested for/against it, and ... well, who really cares. Daniel (talk) 02:25, 10 January 2008 (UTC)
Count me in, this is bloody outrageous.--Docg 02:26, 10 January 2008 (UTC)
I opposed the proposal originally however I doubt that the ArbCom will override developers. I don't see consensus but maybe the developer did and it is not really causing a massive problem so I think a WP:RFAR is overreacting a little Alexfusco5 02:30, 10 January 2008 (UTC)
The ArbCom's authority extends to people's behavior on the English Wikipedia. It has no more right or ability to overrule a sysadmin than it has to punish a user on the French Wikipedia. (What's it going to do, say a user with root access is to be blocked?) If you want to appeal this, I recommend asking a) the sysadmin who did it, or b) Brion Vibber, who's the CTO and actually has authority here. —Simetrical (talk • contribs) 02:35, 10 January 2008 (UTC)
Devs are not empowered to judge consensus, either as an individual or as a group. RxS (talk) 02:36, 10 January 2008 (UTC)
No, they just implement changes regardless of it. Even if we have the tools, we can still enforce admins not to grant rollback requests until consensus is achieved. They give us the tool, they are not forcing us to use it. -- ReyBrujo (talk) 02:38, 10 January 2008 (UTC)
Yes they are, as a matter of convention. The standard for implementing configuration requests is that the sysadmin checks for consensus before proceeding. Of course if they later perceive that they were wrong about consensus, they may undo their changes. —Simetrical (talk • contribs) 02:40, 10 January 2008 (UTC)
Where's the RfD page? Why do some developers have failed RfAs? -- Ned Scott 02:44, 10 January 2008 (UTC)
Some developers have failed RFAs (including me) because development is an entirely separate sphere from the goings-on of the English Wikipedia. I do not have to seek consensus or justify myself to make a software change, even if that very same change (e.g., change to an interface message) would have to be debated endlessly to happen on-wiki. Likewise JeLuF may or may not have any particular authority in terms of editing here, but he's a root administrator and so has full authority when it comes to configuration changes, unless he's overruled by Brion or the Board (or Erik, these days, I suppose). —Simetrical (talk • contribs) 02:48, 10 January 2008 (UTC)
I am not, by the way, commenting on whether JeLuF was justified in performing the change. I'm just remarking on the futility of going to the ArbCom, which has no say in this kind of thing and will just reject the complaint. If you want to complain, you have to go to JeLuF or Brion or someone. If you'd like to file an RFAr, go ahead, just don't expect it to get anywhere. —Simetrical (talk • contribs) 02:51, 10 January 2008 (UTC)
If you are considering arbcom then do so, you will find it most illuminating. 1 != 2 02:45, 10 January 2008 (UTC)
I've been a party in two past cases, so I'm no stranger to arbcom. -- Ned Scott 02:50, 10 January 2008 (UTC)

More bullshit, at least two bots have been granted rollback. WTF. KnowledgeOfSelf | talk 02:51, 10 January 2008 (UTC)

Why not. Since we technically have the ability to grant, and we have absolutely no consensus on to what or whom or how, we can all use our own judgement and grant rollback to anyone (or for that matter recind it). Maybe you should removeit from the bots first time you see them slip up?--Docg 02:56, 10 January 2008 (UTC)
Well, whatever rationales apply to not giving rollback freely to users, there's certainly no reason not to give it to bots. There's literally no difference to a bot between having rollback or not, except that with rollback it can revert edits by sending fewer bytes. The effect is the same, it's just a bit less processing and traffic for the bot and the servers. —Simetrical (talk • contribs) 03:03, 10 January 2008 (UTC)
I was under the impression that much of this proposal arose from recognizing the utility of ClueBot having rollback, since it makes a ton of reverts and its easier on the servers. But vandalbots would never get the full admin bit, so people suggested a rollback-only group, then others said it was silly to have a group for a couple bots, and to give them an editing privilege regular editors didn't have. And here we are... Gimmetrow 07:22, 10 January 2008 (UTC)

I too am disgusted at what has happened. Obviously someone has to make a decision. Someone has to decide if a consensus was agreed or not. But where is the notice? Who, if any users, were formally told? What effort was made to inform users what was going on? And it's one thing to decide there might be in theory a consensus. But entirely another to completely ignore the serious objections of users. Such as, what exactly will be the criteria for granting this faculty? And what efforts will be made to minimise the possible abuse of this faculty? Disappointing.--Gazzster (talk) 03:18, 10 January 2008 (UTC)

Please realize that no sysadmin is currently very active on the English Wikipedia that I know of. Certainly they aren't active most of the places they configure (how many times have you visited ksh-wikt?). They have to look at the information provided to them by the one who files the bug and act accordingly. They aren't going to know anything about what each particular language and site thinks is the correct way to deal with these things, and they aren't going to rummage around through different pages to see if they've been somehow set up (especially if they can't read the language, which is most of the time). If there are subsequently objections, the change can be reversed. —Simetrical (talk • contribs) 03:23, 10 January 2008 (UTC)
Ah, here we get to the root of the problem. There should be developers (or sysadmins whatever that means - it is confusingly similar to sysop/admin) active on the English Wikipedia. Otherwise you get a disconnect between the en-wikipedia community and the developers. The largest WMF projects should have better communnication channels to the developers. Bugzilla just doesn't cut it sometimes. Carcharoth (talk) 03:32, 10 January 2008 (UTC)
Quite simply: there are not sysadmins who are active on the English Wikipedia. There will not be, unless one feels like being active on the English Wikipedia. Access to the servers will not and cannot possibly be given out on the basis of familiarity with any one project. Moreover, by the same logic every single wiki could argue that sysadmins shouldn't touch anything without being part of their project. It's untenable. As a straightforward consequence, any decision that involves sysadmin involvement will not necessarily be done exactly as all of the project's denizens would have preferred.

A more interesting remark is that the software should allow bureaucrats to configure new groups like this. It probably should (within limits), and that would be an improvement. However, for the present time this isn't possible, so you have to involve sysadmins. —Simetrical (talk • contribs) 03:51, 10 January 2008 (UTC)

Oh sure, if there aren't any active there aren't any active. But strategies could be put in place to change that. The Foundation could think "hey, we have this group of volunteers writing software, and these groups of volunteers writing encyclopedias and other free content stuff using that software - we should make sure they keep talking to each other and don't get too out of touch". Is that not reasonable? Sometimes you have to manage things a little bit, rather than keeping hands off all the time. Carcharoth (talk) 03:58, 10 January 2008 (UTC)
What sorts of tactics are you thinking they could use to get volunteer developers to participate in projects they aren't interested in participating in? I don't have the time to participate much in enwiki anymore. Any time I used to devote to that is now spent on development-related matters. —Simetrical (talk • contribs) 04:18, 10 January 2008 (UTC)
The ideal would be that volunteer developers get involved, or remain involved, in the communities. Sure, there will always be some developers who never really get involved with the communities, or who fall out with various communities, but the principle should be there that channels of communication remain open. If not, then the Foundation should appoint a representative to deal with such communication (a poor second choice, but better than no communications, or mis-communications). Carcharoth (talk) 04:25, 10 January 2008 (UTC)
That's a very nice ideal, but it seems not to work in practice. I doubt representatives to deal with communication would be any help unless they were consulted on absolutely everything. If you would like to pursue such an initiative, feel free. —Simetrical (talk • contribs) 04:32, 10 January 2008 (UTC)

I stand by my original point of view: drama :D --TheDJ (talkcontribs) 03:39, 10 January 2008 (UTC)

  • Bottom line: Just because a tool is available doesn't mean that it needs to be used. Whether or not there was consensus to enable the tool, doesn't mean that English Wikipedia has granted its administrators the consensus decision to allow them to hand out the tool. That is the crux issue to take to Arbcom - whether or not there was consensus to grant en-WP administrators the authority to grant rollback requests. Risker (talk) 04:06, 10 January 2008 (UTC)
I completely agree with the above. The first description I received wasn't exactly how things worked out. Initially it seemed the developer took a great leap, but the bugzilla link gives an entirely different light. It appears that none of the editors and/or admins responding to the bugzilla entry made note that the exact same results were previously determined NOT to be a consensus. Now, I'm assuming good faith on this and I don't believe it was intentionally not mentioned, however it certainly should have been. We cannot expect developers to remain in contact with EVERY Wiki, and we have to ensure that if we are requesting they make a change, they have all of the information available to them. I would support taking this to arbcom, to determine if the "consensus" arrived at was flawed or not, but the developer in question certainly should not be listed as a party to it. The most appropriate thing to do at this point would be to protect WP:RFR and wait until this can all be hashed out. It appears an admin did just that, and was overturned, so it looks like it will continue on until someone steps in and shuts things down until this mess can be decided upon by whomever should decide it (the dev, ArbCom, Jimbo etc). In the mean time, instead of fighting back and forth, perhaps just bringing this to ArbCom (the issue of whether or not consensus was reached) instead of threatening to do so is. Justin chat 05:50, 10 January 2008 (UTC)

I would like someone to bring this to RFC or RFA, because this rollback decision against consensus! We are being told to go back to work, but I say let's strike! Say no to non admin rollback! Igor Berger (talk) 22:29, 10 January 2008 (UTC)


I seem to recall that Werdna was going to implement this, but a wiser head suggested the RfC route.

So why now can someone ignore the discussion and implement?

I agree that this was severely wrong.

I don't know if RfAr is appropriate, but as an additional alternative, a petition to the board might be effective.

I dunno. I guess I'm just at a loss for how amazing it is to see this happen. - jc37 03:10, 10 January 2008 (UTC)

As I said above, the correct avenues of recourse are 1) JeLuF, who implemented the change, and/or 2) Brion Vibber. —Simetrical (talk • contribs) 03:25, 10 January 2008 (UTC)
someone should adraft a politelyworded petition elter to try and get them to change their mind before even considering an ARBCOMC. everoyne who thinks that consesnus was ignroed inf avor of this suggestions can affiex their sigs to the Smith Jones (talk) 03:35, 10 January 2008 (UTC) on the bottom of the leter and we can email it to both of the users hta tSimetiraical listes soabove. ~~`~``Smith Jones (talk) 03:35, 10 January 2008 (UTC)
And I would like to find out whether JeLuF implemented this due to language misunderstandings (as I think you implied above) or whether he was "asked" by someone to have a look. The other question is why, in the middle of a debate about whether consensus had been reached, those who supported the process were so eager to jump straight in and get started, and wheel out archive bots and what have you? I would have more respect for those who supported the process if they had said "hang on, something doesn't feel quite right here", let's put this on hold until we found out who implemented this and why, or at least until we can get some more information. Carcharoth (talk) 03:37, 10 January 2008 (UTC)
ASSUME GOOD FAITH USER:CARCHAROTH while is tit possible that they tried to steemroller all over established wiekip ossible its more likely htat they were so excited that they didnt recoutn all the votes and the discussions to see if consensus had actually reached the wrequired amount necessary for the goahed for implementation. i still think it would be better to rais eall of this objections at once in a letter or an email of some sort rather thant to assume sinister motives on the part of outr fellowe ditors and the development staff. Smith Jones (talk) 03:41, 10 January 2008 (UTC)
PLEASE DON'T SHOUT! :-) I'm sorry if I assumed bad faith. I've now read the bugzilla thread, and it does seem to be an honest mistake. I particularly appreciate Gurch's comment at 2008-01-10 03:41:44 UTC. That pretty much sums up my views. Carcharoth (talk) 04:03, 10 January 2008 (UTC)
This was not implemented due to language misunderstandings. JeLuF is perfectly competent in English ― as (necessarily) are all sysadmins and developers ― although it is not his native language. If there were misunderstandings, they were based on his assumption that the poll was conducted under acceptable circumstances, and that two-thirds was enough of a margin to implement it.

As for someone "'asking'" him to have a look, of course someone did. The procedure to get a configuration change is to get consensus on-wiki, then file a request on Bugzilla to have it implemented, linking to the consensus. A sysadmin then reviews the request and implements it if they feel it's reasonable and the presented consensus is sufficient. The request in this case was bug 12534 —Simetrical (talk • contribs) 03:55, 10 January 2008 (UTC)

Thanks for the bugzilla link. I saw that earlier, but no-one seems to have linked to it recently. Carcharoth (talk) 04:04, 10 January 2008 (UTC)

Do you people really find it that hard to believe he sincerely saw this as a consensus? 1 != 2 03:47, 10 January 2008 (UTC)

its kind of farfetched, its mor elikely to assume that he thought that since a majroity of the voters approved of it then that means tiat the resulsts of the poll were binding. he could nto have possibly thought from readin ag any of this discussion that 100% of Wikipedias editors liked this proposal. you'd have to be some kind of moron to actually think that, so its more plausible that he made a simple, forgivable and reasonable mistakes by ased on what he saw. Smith Jones (talk) 03:48, 10 January 2008 (UTC)

First off no need for name calling. Secondly, since when do we need 100% agreement to do anything? We don't agree 100% on a single thing here. A very vocal minority opposing does not mean there is no consensus. 1 != 2 04:09, 10 January 2008 (UTC)

We don't need 100% agreement, but we do need consensus, and with this much participation, a clear consensus. There was none here. And the quality of being very vocal is shared on both sides of this debate. RxS (talk) 04:21, 10 January 2008 (UTC)

There are two parties, those who want rollback and those who do not. Why do you think that one side should need to have such a substantial handicap against the other? The fact is that more people want rollback for good reasons than people don't want rollback for good reasons. Much more, twice as many. Why should the opinion of the few decide that the opinion of the many should not be acted on?

Numbers is not all that needs to be considered in consensus, but you can take into account the substance of the arguments and get the same answer. Disregarding public opinion altogether is a mistake equal to only taking public opinion into account. 1 != 2 04:25, 10 January 2008 (UTC)

Re the actual issue of non-admin rollback: I applied and received non-admin rollback when I saw the offer on the page header. I find it mildly useful but nothing that I didn't otherwise have via VandalProof. I like it but won't particularly care if it is taken away again. I suspect this is the attitude of most people who have now received it.
Re the process debate: Changes shouldn't be implemented without consensus (look at the recent debate over IP page creation). However concern over incorrect processes should be in proportion to the impact of the change. This change makes very little difference - vandals don't need this to disrupt Wikipedia and access to this is unlikely to make any such disruption much worse. By all means make sure any disapproval is passed on but this is surely not the most pressing issue facing the project. Euryalus (talk) 04:36, 10 January 2008 (UTC)
It's true that is may not be such a big issue, but it has led us to discover what is a very big issue--we don't have a good way of communicating with our developers. That is the issue we need to address, so that this doesn't happen again with something even more significant. LeSnail (talk) 05:39, 10 January 2008 (UTC)
You keep asserting that there is consensus for this, there isn't. Not numerically, it's the same result as last time. Consensus may change but what we consider consensus does not. Nor is there any by weight of argument, each side has some (many?) unexplained or irrelevant votes. And each side has arguments that are relevant/rational. The poll gauged consensus on this and it fell short, so instead of accepting it, a few supporters did an end run around consensual opinion here and got a dev to implement it. That's what happened no matter how many times people claim they had consensus on their side. RxS (talk) 13:34, 10 January 2008 (UTC)
As long as you keep doing the tally numerically you will keep getting the wrong answer to the question "what is consensus". If you just compare percents from one poll to another you will see that we don't decide based on percent alone. If you are going to make accusations of conspiracy and fraud provide evidence or don't do it.
The devs are hard working people who do a lot for us and accusing them of corruption is just plain sour grapes. 1 != 2 17:04, 10 January 2008 (UTC)
What I said was a few supporters did an end run around consensual opinion here and got a dev to implement it. That is not an accusation of conspiracy. It is clearly a fraud though...and the evidence is all around these talk pages. I'll say it again because it bears repeating. A few supporters did an end run around consensual opinion here and got a dev to implement it. I have no idea where you got the idea that I was accusing anyone of corruption...rhetorical flourish I suppose.
At the same time you are lecturing me about numerical tallies, you write elsewhere on this page "it would be far less fair to ignore twice as many people because half as many really really wanted it another way." I'd say that's rich...RxS (talk) 21:47, 10 January 2008 (UTC)

permanent departure over administrator/developer consensus fraud[edit]

I have been a strong opponent of proposals on wikipedia that serve to concentrate power, create "classes" of users, and put more decisions in the hands of administrators that contribute very little to the actual work of creating an encyclopedia.

For example, I am a strong opponent of a new proposal, "flagged revisions". You can read about my opinions on this subject at my blog, rhubarb is susan Flagged revisions, which turns "edit this page" into "suggest an edit to this page" for the overwhelming majority of wikipedia editors, was promised for October 2007. The dates have slipped, and now it looks like it will happen in late 2008; I wrote about the impetus behind it thus:

any bureaucratic class will attempt to gain more control, and the administrators are no different. While anonymous and drive-by users contribute the actual substance of value, the administrators have now pushed through a fundamental change in the project that will make them second-class citizens.

A relatively minor proposal, "non administrator rollback" (NAR), I opposed for similar reasons: new powers to admins to "certify" users as trusted or untrusted. A large amount of discussion, with a significant (30%) fraction opposing NAR for reasons such as mine, occurred in early January.

To my great surprise, and against longstanding wikipedia tradition, the 2:1 in favor was declared a "consensus" by a developer without remark, and the proposal was implemented. "Consensus" is a term of art, but only to a certain extent. No consensus-model group I've worked with, including those in the Quaker and academic traditions, would consider 2:1 in favor a consensus sufficiently obvious as to not require a great deal of further discussion.

My feeling in the past was that increasing centralization of power would generally fail because of the strict ways consensus -- necessary for new policy to be enacted -- has been interpreted. What seems clear at this point is that select administrators and other wikipedia "higher ups" are going to implement new policies as they see fit and without regard for consensus.

In the NAR consensus discussions, I've been particularly surprised by the bullying tone of administrators Ryan Postlethwaite and Until(1 == 2); similar behavior I encountered during arguments over flagged revisions from other administrators such as CBDunkerson, and it's re-enforced my belief that as much value as admins contribute in vandal fighting and dispute resolution, many lack the emotional maturity to handle power in a responsible, and responsive, fashion.

I have contributed a great deal to the encyclopedia, both in the field I hold my doctorate in (physics), and in other fields I've had a hand in over the years. It's plenty of fun. But I can no longer contribute value to a work that I feel has in a fundamental sense abandoned the principles that led me to join in. I will no longer contribute to wikipedia; I encourage others to do the same.

Sdedeo (tips) 08:10, 10 January 2008 (UTC)

Wikipedia is fine, l2p Coreycubed (talk) 14:55, 10 January 2008 (UTC)
Do you accept bribes, Sdedeo? Please stay. :-)
Well, seriously, I do not believe this proposal is leading to concentration of power, in the sense that the administrators are attributable to the editors who have elected them. There is no Cabal, and there is nothing preventing administrators from losing their privileges because of their actions. The Signpost publishes stories about desysoppings all the time. On the other hand, this proposal improves the existing anti-vandalism system by giving it a wider base for a useful tool that, until now, only administrators had a right to use. Any editor who believes they have been wronged by an administrator can always ask for further input, and I think we can be certain that it will be given. Everyone is subject to everyone else's scrutiny, and this is especially important in Wikipedia, with vandalism being the number one enemy. You might have noticed how most Western countries have elevated their security measures during the last years due to the threat (real of perceived) of terrorism; Wikipedia must similarly ensure that its policies and institutions are up to the job of safeguarding the encyclopaedia's content and its community's structure against this vague, constant, looming threat.
In any case, even if you disagree with all this (and you obviously do), you could at least not encourage editors to leave... We do need everyone we have, and it's not like they are multiplying these days. I shall stop now, lest I start making apocalyptical prophesies about Wikipedia's slow and painful demise again. Waltham, The Duke of 16:29, 10 January 2008 (UTC)


Hey people. While I personally disagree with the decision to go ahead and implement the proposal despite a number of people with real objections, and a number of people questioning consensus on this, I don't feel a need to get so upset by it. Remember, Wikipedia is not bureaucracy and Wikipedia does not have firm rules. Any claims that Wikipedia is fundamentally different now, that this is an attempt to bypass consensus, etc., are an over-reaction. I'm sure the people moving forward on this were acting in good faith. I see no no angry mastodons charging at us, and I have no reason to think the world will end tomorrow. Nothing here is etched in stone, and even if it was, stone can be broken. There is no need to go to ArbCom, resign from editing, or jump off a cliff. • Personally, I am going to do a wait-and-see. If someone feels they must take immediate action in protest, check out dispute resolution. Most of the steps there can be applied equally well to policy and procedure as they can to articles. —DragonHawk (talk|hist) 13:45, 10 January 2008 (UTC)

Peace can be achieved. Switch the thing off, and let's all calm down and talk it through.--Docg 14:04, 10 January 2008 (UTC)
In the short time I've been here I've often wondered what will ultimately happen to all this wonderful information being accumulated. While it is fun doing and being on WP, it also obviously has great potential for a few to control or "guide" the many, with dissident voices being silenced as "vandals" or "disruptive" or "abusive". Nor am I being an alarmist: a quick look at human history shows this pattern repeating itself within any democratic community. This becomes particularly obvious when social control catch-words are put to use. In that scenario the word "Peace" stands for "Shut up and don't rock the boat!". So whatever the REALITY is, I suggest all parties DISCUSS this issue until a true CONSENSUS IS REACHED. The alternative would be a shame. Shir-El too 15:36, 10 January 2008 (UTC)
My use of the word "peace" was not intended to be "social control", but as a genuine plea for people to maintain a cool temper. There are apparently a lot of people genuinely, seriously upset by this. I really do believe the problem is nothing more than a series of misunderstandings and differences of opinion. Decrying the situation as "the end of Wikipedia", attacking the actions of others, storming off in a huff, shouting (I'M NOT SHOUTING), etc., does not help the situation — indeed, they likely just aggravate it. Finally, remember that Wikipedia is not a democracy. That works both ways. Statements about how "polls are not binding" does nothing to advance consensus any more than treating the poll as a binding vote does. —DragonHawk (talk|hist) 16:35, 10 January 2008 (UTC)
Doc glasgow: The people who oppose this idea are not helping by insisting that It Must Not Be Until We Say So, any more than the people who support it are helping by insisting that It Must Be Because We Say So. Complaining about process is not productive. Making dogmatic proclamations is not the way towards consensus anymore than ignoring objections is. —DragonHawk (talk|hist) 16:49, 10 January 2008 (UTC)
That's not fair. Many of those oppposing (including me) would have been prepared to support something else. I specifically stated what my main sticking point was. I was hoping that another round of discussions would come up with a proposal that would gain greater support. It might still be possible to do that, but not in the current climate. Carcharoth (talk) 17:55, 10 January 2008 (UTC)
Fine, I've got nothing against that. What I was referring to was Doc glasgow's comment, "Peace can be achieved. Switch the thing off, and ...". As I understood it, the implication was that the rollback feature must be shut off before progress towards peace and/or discussion can continue. That's the same kind of stubborn obstinacy that some of the supporters are using. If everybody persists in digging in their feet, we're not going to get anywhere. —DragonHawk (talk|hist) 00:55, 11 January 2008 (UTC)

BTW: found this AFTER writing the above! Wikipedia:Wikipedia Signpost/2007-12-26/In the news, Shir-El too 15:41, 10 January 2008 (UTC)

You found last month's Signpost... okay... did you have point?  :) If you're referring to the PC World opinion piece, did you have some connection you were trying to establish to this situation? If so, it would help if you stated what your point was, and maybe gave some supporting arguments of your own. —DragonHawk (talk|hist) 16:43, 10 January 2008 (UTC)
Several incidents in the past have caused speculation that Wikipedia has an informal inner circle consisting of administrators, outside of which edits might not survive very long. Shir-El too 20:12, 10 January 2008 (UTC)
That's nice. "Citation needed". Got any evidence to support the claim? The author makes absolutely no attempt to substantiate the accusations, and nothing in my experience does so, either. Yes, lots of stuff is deleted, but it's deleted by lots of people and almost always per the policies and guidelines we lay out. I get the feeling Mr. Schneider didn't take the time to understand that. So if you can speak for yourself, please do so. —DragonHawk (talk|hist) 00:55, 11 January 2008 (UTC)
I agree with peace. I do not agree with shutting down a tool that the community wants because of a vocal and stubborn minority. You don't have to agree with this, but you do have to accept it. You can yell "unfair" from the hill tops all day, but it would be far less fair to ignore twice as many people because half as many really really wanted it another way. So lets go for peace instead. 1 != 2 16:57, 10 January 2008 (UTC)
And yet, several people who supported this have made it clear they did NOT agree that a consensus had been reached. People with "twice as many votes" get turned down at WP:RFA all the time. Potential 'crats get turned down with anything short of a 90% approval rate. Are you telling me, implementing policy has a lower barrier of entry than one user requesting adminship? Voting is not a replacement for discussion. The results of the poll were taken into heart, and a discussion on how to implement this AND gain a consensus was JUST beginning. You've used the same argument since this was implemented, but Wikipedia is not a democracy. Justin chat 17:51, 10 January 2008 (UTC)
First off if you use the RfA process then ya it failed consensus, but this is not an RfA. What process are you referring to? Where is it written that the decision is invalid unless it meets certain criteria? What are you basing this idea of it being out of process on? Who said this decision was based on voting? You say the poll was just beginning, but the thread above shows clear favor by both sides in closing it: #Poll_closure_revisit. 1 != 2 17:57, 10 January 2008 (UTC)

Saying, you can have peace if you shut up, is not helpful.--Docg 17:55, 10 January 2008 (UTC)

At what point did anyone say "shut up"? You'll note that I linked to WP:DR, of which the first several steps are entirely about discussion. I'm not asking for silence, I'm asking for productive commentary, rather than two sides drawing lines in the sand and refusing to budge, which is what we have now. —DragonHawk (talk|hist) 00:55, 11 January 2008 (UTC)
I never told anyone to shut up, I said they needed to accept what the community has decided. 1 != 2 17:57, 10 January 2008 (UTC)

But I do not accept that the community has decided it, so no.--Docg 17:58, 10 January 2008 (UTC)

Okay, but please don't be disruptive with your lack of acceptance. 1 != 2 18:00, 10 January 2008 (UTC)
I find it disruptive that someone with a such a clear bias is telling others not to be "disruptive" by disagreeing with him. I really think you should let others judge what's "disruptive" and "fair" in this case, "1 != 2", as your agenda and lack of objectivity has been more than clear. If people want to object to this decision, just let them. If you have such confidence that consensus was as overpoweringly obvious as you claim then you shouldn't have anything to fear from their efforts anyway. Equazcion /C 19:35, 10 Jan 2008 (UTC)
Being a member of a group that does operate by consensus (not WP), every voice must be heard and taken into account! Disagreeing with a vocal minority does not invalidate their argument, any more than the fact that a majority holds a view automatically validates that view. We NEED the minority voice to see an issue from all sides, not just from the side in the mirror of commonly held opinions. Thank you, Shir-El too 20:12, 10 January 2008 (UTC)
1 != 2 and Equazcion: Please let's avoid any accusations of disruption, here, without evidence to the contrary. I think several people are being very stubborn in holding to their own POV (I may be one of them), but that is hardly unique on this talk page, and not against any rules that I'm aware of. :-) —DragonHawk (talk|hist) 01:04, 11 January 2008 (UTC)


By the looks of this - it appear so. Could anyone lead me to where this was decided? Or was it just the devs? Thanks. Tiddly-Tom 19:31, 10 January 2008 (UTC)

The rate limit in question is presumably the default for Wikimedia sites, which, yes, was of course decided by the sysadmins, as all defaults are (unless they're the software defaults, in which case it was developers who decided it). If you would like it changed, you should file a new bug requesting that, linking to consensus to do so. Or else you could try persuading a sysadmin to change the defaults. —Simetrical (talk • contribs) 20:35, 10 January 2008 (UTC)
The only thing I could deduce by that was that it was limited - could someone enlighten me with what those numbers mean? Thanks Tiddly-Tom 23:36, 11 January 2008 (UTC)
rollback' => array(
             'user' => array( 5, 60 ),
             'newbie' => array( 5, 120 ),
Regular users can do 5 rollbacks in a 60 second period, "newbies" (the newest 1% of accounts I believe, different from autoconfirmed) can do 5 reverts in a 120 second period. Mr.Z-man 00:15, 12 January 2008 (UTC)
Actually "newbies" means "not autoconfirmed" in this context, confusingly enough. Also, there are no rate limits on anything for sysops, bureaucrats, or bots. —Simetrical (talk • contribs) 01:04, 13 January 2008 (UTC)
At what point does a user change from autoconfirmed to user? Thanks. Tiddly-Tom 09:22, 13 January 2008 (UTC)
Registered users become autoconfirmed after . . . I forget, maybe it's four days and some number of edits? —Simetrical (talk • contribs) 17:30, 13 January 2008 (UTC)
Sorry, When you originally wrote it - I read it as autoconfirmed - not "not autoconfirmed". I understand now. Thanks. Tiddly-Tom 17:45, 13 January 2008 (UTC)
Here users are autoconfirmed after 4 days. We don't have an edit count restriction. Mr.Z-man 23:01, 13 January 2008 (UTC)
That's because it said "autoconfirmed". It was a typo, I fixed it in my last edit. —Simetrical (talk • contribs) 19:50, 14 January 2008 (UTC)
Ah! Ok, no problem - thanks for the explanation. Tiddly-Tom 19:55, 14 January 2008 (UTC)

The RSS feed of Special:Recentchanges[edit]

I may have completely the wrong end of the stick here - please be gentle if I'm wrong about this.

"Rollback links are displayed on page histories, user contributions pages, and diff pages.". But not in the RSS feed for Special:Recentchanges, even though that does contain the diffs. It seems (please could an admin correct me if I'm wrong?) that admins get such a link, but non-admins with rollback permission don't. This matters, because that is what tools such as Lupin's AVT work from. For non-admins who use RSS-feed-based tools rollback permission seems to give little extra value.

Could non-admin rollback be extended to provide this? I hesitate to ask for this, because while I supported the proposal and believe that many of those who opposed it did so from an incomplete understanding of it, I also agree with those who argue that 2-to-1 is not a consensus. A fortiori, there is not a consensus for widening the power now, even though I believe that such a widening would be in the best interests of the project.

Also, do people feel it would be against the spirit of granting non-admin rollback permission for a developer of an RSS-feed-based tool to use a little Ajax to retrieve the required rollback token and then invoke rollback with it? (No, I'm not sure it would work. But I'm considering experimenting with doing it.) Philip Trueman (talk) 19:47, 10 January 2008 (UTC)

Rollback links are not provided in RSS feeds for recent changes for anyone, as far as I can tell. Notice that there are no edit links, or in fact any links at all beyond a link to the diff (well, and links in edit summaries, but that's a not-really-intentional quirk). It's meant to be a summary, not a substitute for the actual page. If links are being displayed for sysops and no one else, and this isn't due to some custom JS, that's a bug and should be fixed ― there should be no difference between admin and non-admin rollback (except that admins can mark the reverted edits bot, as well as rolling-back on protected pages and so forth).

As for rollback tokens, as far as I can tell, they can be easily constructed using the page name and target username, if you have the edit token. The correct token is md5( $token . namespace-prefixed article title . username of the one being rolled back ) . EDIT_TOKEN_SUFFIX, and all that information should be available to bot editors. Actually I'm not sure why we bother salting it at all, and don't just use the edit token. —Simetrical (talk • contribs) 20:49, 10 January 2008 (UTC)


I have formally requested arbitration on the matter of consensus.--Docg 00:53, 11 January 2008 (UTC)

FYI, Jimbo has apparently directed ArbCom to settle the matter anyway, independent of Doc glasgow requesting it. Jimbo speaks. Personally, I'm pretty disappointed in the whole situation. I think the proposal was implemented overly quick, and I think going to ArbCom at this stage skips several steps in WP:DR. People on both "sides" are not in tune with the wiki way.  :-( Sigh. —DragonHawk (talk|hist) 01:08, 11 January 2008 (UTC)
Can someone point me to where the most current and single good place for this discussion is now? There are so many different pages each purporting in some way to be the right place for this discussion.--Jimbo Wales (talk) 02:36, 14 January 2008 (UTC)
I would suggest WT:RFR for discussion about the process Alexfusco5 02:37, 14 January 2008 (UTC)
Well, let's see:
So I guess Wikipedia talk:Requests for rollback is the way to go right now. --Conti| 16:09, 14 January 2008 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.