Wikipedia talk:Non-administrator rollback/Creating a new proposal

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.

---

Untitled discussion

While I think this page may be helpful in improving the existing consensus, I don't think it will have a better chance of consensus since the current proposal is doing very well in the consensus department. 1 != 2 16:04, 6 January 2008 (UTC)

Addressing some of the ideas:

Giving rollback to everyone

There is already a strong consensus against this idea

allow/force custom edit summaries

There is a consensus and a long standing history of rollback being used only when edit summaries are not needed.

Allow rollback only after viewing

Defeats the ability to do mass rollback on automatic bots and vandal only accounts

Preventing abuse

The tools give no new ability that cannot be accomplished through another means and can be quickly revoked.

Strict standards for approval

Why? If they want to meet strict standards they can become an admin. We could set up WP;RfR(request for rollback), but why bother when it is not a powerful tool and can be easily taken away.

Just my 2 cents. 1 != 2 16:07, 6 January 2008 (UTC)

So I take this to mean you don't think the current proposal needs to be changed at all. Equazcion •✗/C • 16:11, 6 Jan 2008 (UTC)

I don't think that is what consensus desires, nor do I desire it. As for me, reasons outlined above explain why, the others have said similar and more on the primary discussion page. 1 != 2 16:17, 6 January 2008 (UTC)

Just clarifying your stance. Thanks. Equazcion •✗/C • 16:29, 6 Jan 2008 (UTC)

Don't get me wrong, if an idea for this proposal would improve things I am sure both the community and I could be swayed. I just don't think any of these ideas would improve things. I do like the idea of specific standards being applied, but strict? No. I also think such standards would be better forged on the new policies talk page after it is created instead of trying to introduce them to a polarized debate such as this one. 1 != 2 16:34, 6 January 2008 (UTC)

This is really for those of us who don't believe a consensus exists, which from what I've seen is most people. So I do understand your opposition to it. The specific standards are not really what we're discussing -- rather we're discussing whether or not there should be standards at all, because one of the points specifically outlined in the current proposal is that there would be no standards. Equazcion •✗/C • 16:40, 6 Jan 2008 (UTC)

No standards? I don't think that is true. The proposed standard is admin discretion which is a standard that has server Wikipedia well for some time. The fact is the over two thirds of the people want this, I just don't see how you can say there is not a consensus. This page is for those who wish to discuss the possibility of a new proposal, not for those who don't think consensus exists. I also find it a bit hard to believe that "most people" think there is no consensus. 1 != 2 16:47, 6 January 2008 (UTC)

Standards usually mean tangible prerequisites. A lack of actual standards was an issue with some voters. As for consensus, all I can say is read through the main talk page (and archives). The straw poll itself came under heavy fire, and a two-thirds majority is not consensus, as far as I know. Equazcion •✗/C • 16:49, 6 Jan 2008 (UTC)

While I disagree with your interpretation of standards and prerequisites, I certainly understand why someone might want more specific standards. I am going to shy away from the "it is a consensus", "it is not a consensus" debate, I don't think we can agree on this and time will tell. I will say that I am pretty sure the existing proposal will be accepted.

That being said it will not be written in stone, consensus can change. Really I don't think we should be laying down specific standards till the need presents itself, per WP:CREEP. We will be in a better position then to know which standards will be helpful. And if a few mistakes are make along the way then it is not hard to fix as rollback is not a dangerous tool. 1 != 2 16:55, 6 January 2008 (UTC)

This is from one of my previous comments above: "The specific standards are not really what we're discussing -- rather we're discussing whether or not there should be standards at all..." Equazcion •✗/C • 16:57, 6 Jan 2008 (UTC)

Move all discussion here

Would anyone object if I moved all the discussion currently on the project page to this discussion page? I think having a discussion section on the proposal page is just going to cause needless conflict and confusion. Discussion is easier if it happens in one place. And this way there won't be anyone complained that their discussion got archived. It will also leave the proposal page more stable and easier to keep track of for actual proposals. Comments? Objections? —DragonHawk (talk|hist) 22:33, 6 January 2008 (UTC)

It does make sense to me that the discussion of the page happen on its talk page. 1 != 2 22:40, 6 January 2008 (UTC)

I think it's better to keep it all on one page, as the rationale stated on top gives the discussion context. The main proposal page (Wikipedia:Non-administrator rollback) has both the proposal and the discussion on a single page too. Similarly, this isn't like crafting an article or a guideline, where discussion on its improvement should be kept separate. This entire thing is basically a brainstorm and there's no separation between proposal and discussion Equazcion •✗/C • 22:44, 6 Jan 2008 (UTC)

As far as "rationale stated on top" goes, I would hope that anyone coming to this discussion page has read the corresponding project page.  :) And I fully understand (and support) the idea that this is just a brain storm at this point. So let the project page remain a short list of ideas and possibilities, and let the storm happen here. As discussion takes place, update the project page to reflect new ideas and any emerging consensus or proposals. This lets the project page converge slowly and neatly towards consensus (easily monitored in the page history), while the discussion can happen in whatever form it needs. Discussion can be archived (by bot, and indexed by bot) as it grows large, and that's expected, so nobody gets upset. And with luck, the project page will eventually turn into the actual proposal, and then the actual guideline. In short (too late), let the discussion page determine consensus, and let the project page document that consensus. Sound good?  :) —DragonHawk (talk|hist) 23:00, 6 January 2008 (UTC)

I'm only agreeing to this because I think you're smarter than me. And I don't think that often about people :) So okay, let's try it your way. Equazcion •✗/C • 23:30, 6 Jan 2008 (UTC)

Thank for you for the complement.  :) I've moved the comments posted on the project page to this page, and put a pointer there to here. —DragonHawk (talk|hist) 02:59, 7 January 2008 (UTC)

Methodolgy for discussion and proposals

I'd like to address some meta- issues here, about how to avoid the... excess of people stating their opinions... which occured at the straw poll. What I basically advocate is that polling should be avoided. A familiar refrain, I know, but hear me out. A poll encourages everyone to respond. That's how polls work. Polls are a vote, by definition; if you don't vote, your opinion isn't counted. This is obviously an important issue to many people, so lots of people responded to the straw poll. At the same time, wiki pages are a poor way to count votes; they get unmanageable at larger sizes. Large discussions erupt in the middle of a lists of votes. If you archive the discussion, people feel their response to the poll is being discounted; if you don't, the page becomes truly unwieldy.

The solution, as I see it, is to 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 400 other opinions.

Finally, polling should be the last option. RFC's, VP messages, watchlist, etc., should all be done in the discussion stage. Turning it into a vote before consensus is reached just confuses things.

—DragonHawk (talk|hist) 22:54, 6 January 2008 (UTC)

Support =D But seriously, I agree, on pretty much all points. Equazcion •✗/C • 22:57, 6 Jan 2008 (UTC)

"Support", heh heh. Assuming the proposal does not get tried "as is", I think the first thing to do would be to refactor the comments in the poll into something as concise as possible. That will at least give us a starting point as to what people supported and what the objections were. I don't exactly relish the though of wading through that morass, but it should be done.

That said: It looks like the edit storm over at the straw poll might be starting to calm down a bit. From the start to the current 380+ comments, it's held pretty steady with the ratio of comments placed under "Support" to those placed under "Oppose" at a little over 2 to 1, or roughly 70% under "Support". But as I've remarked elsewhere, the text in the individual comments seems to indicate things are more complicated than a simple yes/no. I suspect there's going to be some discussion just about how to interpret the poll. (Maybe we should have a poll on that.  ;-) Perhaps the best thing to do at this point is wait-and-see. What do you think? —DragonHawk (talk|hist) 03:18, 7 January 2008 (UTC)

Well, I've attempted to sum up a few of the suggested changes already, but granted it is probably far from complete. I certainly didn't comb through all the votes, but I agree that is probably the best thing to do -- and I also share your dread of doing so. As for whether or not to wait, that's an excellent question. My motivation for starting this page was the fact that I didn't see any end in sight, so I figured I'd quit making suggestions on the talk page about "where to go from here" and instead actually start going somewhere. So in that vein I'd like to get started now. But if you think there's a chance something significant will happen soon, I'm open to waiting. Equazcion •✗/C • 06:14, 7 Jan 2008 (UTC)

My take

• I would support a rollback feature that involved review of the proposed edit (a la the current diff page but without the "Current Version" part), automatically summarized "Undid edits by -User-" and forced an addition to that summary. This would still minimize server load and time to render the user display, but would add the responsibility element. In this form, the button could could be given to everyone, or at least logged-in everyone. Franamax (talk) 15:45, 6 January 2008 (UTC)

Also, any limits on the number of rollbacks in a given time frame would seem to defeat the real value of this button, which would be to help vandal-fighters rolling back a large attack (and would also greatly increase the coding complexity). Franamax (talk) 16:04, 6 January 2008 (UTC)

Not likely

Any solution that sends the developers back to writing new source code for Mediawiki is very unlikely to occur. 1 != 2 16:10, 6 January 2008 (UTC)

Most users' issues with this proposal stem from that limitation. But if we can at least pinpoint what ideal proposal would gain consensus, maybe that would at least get things moving. As it stands, there is no consensus. Waiting for development of new functionality may be the only choice. Equazcion •✗/C • 16:15, 6 Jan 2008 (UTC)

The rollback button has been working great with existing admins for years now. These changes would limit its usefulness and are already available in javascript for anyone to have. It is like re-inventing the wheel. You can keep saying there is no consensus, but that is just not so. Regardless it will not be you are I who judges consensus so time will tell who is right about that. 1 != 2 16:19, 6 January 2008 (UTC)

My point of view aside, something very similar has already been presented at the primary discussion and there has been clear objection to it for the same reasons I have outlined. 1 != 2 16:22, 6 January 2008 (UTC)

What was presented was an alternate proposal. This is not an alternate static proposal. This is a discussion on how to create a new proposal. Equazcion •✗/C • 16:25, 6 Jan 2008 (UTC)

I was referring to the idea of the forced edit summary, not the idea of improving the proposal. When the discussion is closed there will be a first revision of the policy. People will still be able to change that policy through consensus. I would support any idea that improves the existing proposal, I just have not seen one yet. 1 != 2 16:28, 6 January 2008 (UTC)

If you have any of your own, please feel free to add to the list above. Equazcion •✗/C • 16:43, 6 Jan 2008 (UTC)

Ok. I have made a suggested idea, though it does not involve changing the existing proposal. 1 != 2 16:59, 6 January 2008 (UTC)

Read the lead paragraph. This page assumes the proposal won't pass. if you feel it will, and that this page is not needed, I hope you won't feel the need to disturb the effort here. Equazcion •✗/C • 17:04, 6 Jan 2008 (UTC)

Addressing specific concerns

If you want this proposal to have a snowball's chance of reaching a consensus, the first step should be to address the specific concerns raised in the polling and discussion, regardless of how baseless you feel they are. From my reading of the discussion there seems to be the following complaints:

1. Rollback will enable more vandalism
2. Rollback will enable worst edit-warring
3. Rollback has poor edit summaries
4. The process for granting rollback is inadequate

Did I miss any?

Burzmali (talk) 14:11, 7 January 2008 (UTC)

Rollback will enable more vandalism

The vast majority of vandalsim is done by brand new accounts, this is because accounts that vandalize get blocked. Vandals will not have the rollback tool, the vandal fighters will.

Rollback will enable worst(sic) edit-warring

We give the tool out one person at a time, we can just take it away. The tool gives not special ability that could not be achieved without it.

Rollback has poor edit summaries

Rollback is only meant to be used in situations where and edit summary is not needed, such as reverting blatant vandalsim. Any use that would need an edit summary would be a mis-use of the tool and may result in it being taken away

The process for granting rollback is inadequate

If there is a problem, then we can change the requirements. We generally start with simple general rules, then get more strict as the need presents itself.

Really my main point is that the tool is harmless. It can be undone by anyone, even someone without the tool. It achieves nothing that cannot be accomplished without the tool. It can be taken away if shown to be disruptive. It will be a major benefit to those who fight vandalism. Of the four concerns listed the only one I see as having merit is the concerns about how it is granted. 1 != 2 16:02, 7 January 2008 (UTC)

I agree with your points, at least on the first three, however, valid or not, those are the concerns that have to be addressed before any future proposal makes it to polling. Burzmali (talk) 17:13, 7 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.