Wikipedia talk:PC2012/Archive 1

Archive 1

Archive 2

Can someone come up with a handy checklist?

Okay, I've seen a few proposals so far ... but both the proposals and the discussion tend to be kind of long. If there are no objections, I'm going to create my own page (talk page at WT:PC2012/Dank) where people can suggest summaries that will help us compare and contrast proposals, to help people quickly find the pages they're interested in. I'll start with a quick list of objections and questions from the previous RfC, and people can turn these into whatever kinds of checklists seem useful to you. - Dank (push to talk) 21:23, 5 July 2012 (UTC)

Both User:WaitingForConnection and I have the {{nutshell}} template at the top of the page. I think it would be helpful for everybody to have one of those, summarizing the core components of their policy. For instance, if one were to look only at those, they could assume that my policy deals with the 24 hour "expiration time" while WFC's outlines specific guidelines on when PC should be applied. If I'm understanding you right, though, you want a table or checklist of some sort where we can compare the objections and questions from the RfC with the new "features" in each policy. Is that correct? ~Adjwilley (talk) 22:00, 5 July 2012 (UTC)

Yes ... anything mentioned in Options 1, 2 or 3 that serves to distinguish proposals. Let me see if I can slap something together. - Dank (push to talk) 22:27, 5 July 2012 (UTC) Done, on my subpage. - Dank (push to talk) 12:12, 7 July 2012 (UTC)

One shaky step at a time

I'm going to be kind of busy with Wikimania until Monday. For people working on proposals: it's looking like it's going to be hard to get even 70% (and ideally, we should shoot for 90%) to agree on everything at once; is there any way to break this down into one shaky step at a time? - Dank (push to talk) 11:19, 9 July 2012 (UTC)

So you're saying, instead of writing up a new policy and having people vote on it, write up a list of X number of possible changes to the policy and have X number of mini-votes to see which changes are likely to have consensus? ~Adjwilley (talk) 19:22, 9 July 2012 (UTC)

Well, I'm not saying to stop what you're doing ... writing up a full proposal may attract people to your page who share similar views, and I think it's a good exercise to think through everything. But the 2011 and 2012 RfCs both suggest that it's going to be hard (impossible?) to get almost everyone to agree to any one package, no matter what the arguments, so perhaps we can find small areas of agreement (even agreement on where not to use PC), or limited uses of PC, that no one is terribly offended by. If we could find something to vote on that passes by a wide margin, and if what we approve actually seems to work, then that may build confidence so that we can take another step. - Dank (push to talk) 19:37, 9 July 2012 (UTC)

I agree, what I see developimg here is exactly what I have tried to avoid in the previous discussion. If everyone is just proposing their own finished policy, everyone will support their own draft. If we instead find a way to work on things that need to be adressed point-by-point, we can wind up where we need to be, with one finished policy instead of fifteen of them duking it out. I would again suggest we start with defining the rights and responsibilities of reviewers and clarifying the process for granting and removing that right as these were issues many users expressed grave concerns about in the last two discussions. However, I will also be at Wikimania the rest of this week ( see you there Dan) so I won't be real active in that time. Beeblebrox (talk) 20:05, 9 July 2012 (UTC)

I don't think it's necessarily a bad thing to have a bunch of individual policies at this point. I started mine out with what I thought was a bright idea, but that idea changed significantly as I tried to integrate it into a complete policy. I figure if a bunch of people go through this process, then we'll have a bunch of bright ideas to draw from when we need to pick out individual points to vote on. I imagine that the final policy will have bits and pieces from a bunch of these individual policies that are being written up. (I never saw it as being Adjwilley vs. Rivertorch vs. Waiting for Connection vs. everybody else who writes a policy.) ~Adjwilley (talk) 20:30, 9 July 2012 (UTC)

That's my position too. If someone proposes a finished policy, I'm not going to assume that they're trying to be difficult, I'm going to assume that they're being open about what they want. That's a good thing ... it's just not enough to get us all the way home. Beeble is on to something, however ... discussion kept taking a wrong turn at the 2011 RfC, and what we've done so far doesn't look a lot different yet. But it will. - Dank (push to talk) 20:38, 9 July 2012 (UTC)

In my draft, each level three header should be capable of standing alone. That was intentional on my part – I want to make it as easy as possible for small parts of what I'm suggesting to be discussed in isolation. —WFC— 21:01, 9 July 2012 (UTC)

As a side note, I've made a point of documenting the various changes I've made to the provisional policy (see WT:PC2012/Adjwilley#Documentation of changes to policy). I believe this would be a healthy exercise for those writing the policies, and will make it easier later on to pick out changes that we want to vote on or discuss. ~Adjwilley (talk) 20:30, 9 July 2012 (UTC)

a third closer

Can we have a third closer - I think that would be a benefit to the close - three heads and discussion and consensus agreement seems better to me than two? - I would as always suggest User:Sandstein as a quality closer who I think is uninvolved on this topic and very experienced in complicated closes ... and who I think would be willing to help/contribute to the close. - I left the user a note about my comment here - Youreallycan 05:55, 10 July 2012 (UTC)

Just as an update - the User:Sandstein has commented diff that they have a previously expressed opinion regarding WP:Pending changes and that as such they are unable to accept. I am still of the same opinion that we should seek another closer and welcome suggestions - thanks - Youreallycan 07:02, 10 July 2012 (UTC)

Okay, did you want to discuss this here or at WT:PC2012/Dank#Closer?. Your other thread has attracted comments, so I'll reply over there. - Dank (push to talk) 13:36, 10 July 2012 (UTC)

Mini votes?

I notice Wikipedia talk:PC2012/Adjwilley#First mini-vote (automatically accept changes after some time).

This is put there as a draft so we can work out how to phrase the question. I think this is good because a long discussion about a badly phrased question can be frustrating. However, I'd like to take a step back. Is that how we'd like to proceed? I can see a certain logic in seeing how popular different ideas are but the value of some of them will depend on which other ideas are included in the policy. If we do proceed with taking ideas one at a time I think it is very important to emphasis the NOT in WP:NOTVOTE. It would be useful to know if people have a better way to solve the problem that a given idea is supposed to solve. It would be useful to know of any interrelations between ideas such as "This would only be helpful if we also do X."

Yaris678 (talk) 22:23, 10 July 2012 (UTC)

Agreed. - Dank (push to talk) 23:21, 10 July 2012 (UTC)

There's a related thread at WT:PC2012/Dank#Closer?. The upshot is from the conversation (started above) is that I'm demoting myself to mini-closer :) I changed the main page to say: "Through the end of September, The Blade of the Northern Lights and Dank will be closing a series of mini-votes designed to create a proposal that will be approved, rejected or modified in an RfC in October." Blade will continue to be involved in October, but I won't be.

So, my position is that my COI is mostly out of the way now ... at least enough that I can keep making suggestions about the process (but not take sides) ... but as always, please speak up if I'm crossing any lines. I'm not closing an RfC, and the voters and closers in October can overturn anything we do in the next 3 months (but if we do a great job, they won't want to :) Now I can say what's behind the mini-votes, and feel free to disagree, guys. I think a lot of what went wrong with the previous RfCs was that the whole process from start to finish took too long, so that the people who were voting at the end didn't want to vote in favor of something that they hadn't had a hand in developing at the start. I don't have any specific suggestions for a vote, but I'd like to suggest that it be something simple but important, something that people seemed to think in the last RfC was a significant problem. Also, offer something attractive to the opposers and something attractive to the supporters. We should get feedback first to see if everyone likes the language, then do a story in the Signpost and advertise on WP:CENT to attract voters ... and get the vote over in a week. The comments from the voters will hopefully give us the clues we need to come up with a second successful mini-vote. Thoughts? Tweaks? Rotten tomatoes? - Dank (push to talk) 22:58, 10 July 2012 (UTC)

I see. And I understand Yaris's concern as well. You don't want to do a bunch of little Mini-votes on ideas as they come up. You want to do a big round of Mini-votes all at the same time, close them all after a week, and then we all go back to the drawing board to prepare for a second round. Is this correct? ~Adjwilley (talk) 23:05, 10 July 2012 (UTC)

Great idea. - Dank (push to talk) 23:20, 10 July 2012 (UTC)

Just a thought, if we're going to have a big advertised round of "mini-votes" on this page, perhaps we could have preparatory rounds of informal "micro-votes" on the individual talk pages. For instance, the people watching this discussion could vote and comment on the "first mini-vote" draft on my page that was mentioned above, and then I could tweak, polish, or abandon it based on your concerns, better preparing it for the first mini-vote. (Yaris has already helped out quite a bit on that one...thanks.) ~Adjwilley (talk) 23:19, 10 July 2012 (UTC)

Another great idea. - Dank (push to talk) 23:20, 10 July 2012 (UTC)

What kind of timeframe would we be looking at? End of July for first round, end of August for second round, end of September to prepare the final RfC, and then end of October to close it? Or maybe that's not giving them enough time to close. I dunno. You have more experience with this than I do. ~Adjwilley (talk) 23:26, 10 July 2012 (UTC)

We should do it whenever people are agreed that what we have will probably work. - Dank (push to talk) 11:05, 11 July 2012 (UTC)

Hmmm... I don't think you get my point about emphasising the NOT in WP:NOTVOTE. So to take the auto-approve idea, I think you have done a fairly good job of explaining the problem it is trying to solve (backlog). I think people should be encouraged to not just say support or oppose. They should be able to say "how about solving that problem like this instead" or "that could work but we need to be careful about X". Do you see what I mean? We also need to encourage others to come up with lots of ideas along the lines of:

• Problem we are fixing/opportunity we are taking
• General idea for solving the problem/taking the opportunity
• Specifics for implementing the general idea.

Yaris678 (talk) 23:36, 10 July 2012 (UTC)

Re: emphasizing the NOT... like this?

The three bullets make sense, but I'm not quite clear on what the "taking the opportunity" part means. ~Adjwilley (talk) 23:55, 10 July 2012 (UTC)

I think its the word "vote" in "mini-vote" that makes me think we aren't emphasising the NOT. Maybe it should be called "focused discussion".

In terms of "taking the oportunity" I guess it is just a more positive way at looking at things. Rather than saying "what is wrong?" we say "what does this enable us to achieve?" Maybe its a bit confusing. Maybe it should be be both phrases should be replaced by "what is the challenge we are trying to meet" or maybe we should just stick with solving problems.

Yaris678 (talk) 11:28, 11 July 2012 (UTC)

There are times when brainstorming is required, and so lots of ideas should be encouraged, and times when decisions must be made to narrow down available options to the most suitable and effective ones. When in this latter phase, while contributors should still be encouraged to provide reasoning for their statements, the proliferation of discussion threads and new suggestions needs to be kept under control, or else this process will not be able to be build towards a conclusion. (And as I discussed previously, sprawling conversations self-perpetuate into more and more threads, making them very difficult to follow.) isaacl (talk) 02:50, 11 July 2012 (UTC)

I agree that focus is good, but I don't think the best way to achieve this is to have a very narrow yes/no conversation. As I suggest above, let's have "focused discussions" rather than "mini-votes". Yaris678 (talk) 11:28, 11 July 2012 (UTC)

Yes, as I said, let's be sure to focus the discussion appropriately for the current phase of discussion. isaacl (talk) 12:03, 11 July 2012 (UTC)

An example of a focused discussion

There is something of a focused discussion happening now. It started at Wikipedia:PC2012/Rivertorch#Proposed name change and is continuing at Wikipedia talk:PC2012/Rivertorch#Name change. I have a feeling that we should advertise this on WP:CENT but I would like to check that makes sense with everyone before we do that. Mostly I want to check with Dank, as our "guiding admin". Is this how he sees focused discussions (formerly known as mini-votes) proceeding? Yaris678 (talk) 11:59, 12 July 2012 (UTC)

I was just thinking about that. The issue with advertising anything is that there's quite a lot of vote-fatigue about PC, and we will probably only get a few shots at getting people's attention. (Not that they won't show up the 3rd time we ask, but they may arrive really grouchy.) What everyone has done so far is great ... staking out your position and thinking through all the ramifications is a great way to foster communication. But it's just the first step; we don't have any particular consensus among ourselves, we haven't delineated a path that will be reasonably easy for the masses to follow, and we don't know where we're headed yet. We've got at least several weeks more of work. More people would be great, but something really public, like WP:CENT, at this stage, will probably reproduce more of what we've seen over the last 5 years in PC debates. - Dank (push to talk) 12:10, 12 July 2012 (UTC)

That makes sense... so is your thinking something along the lines of the following?

1. More development amongst ourselves.
2. Think up some specifics we want input on.
3. Get input on those specifics through CENT.
4. Come up with a coherent proposal.
5. Put the coherent proposal to the whole comunity, through CENT and a watchlist notice

Yaris678 (talk) 12:35, 12 July 2012 (UTC)

How would you feel about something in the Signpost instead of a notification at CENT? Reaching a lot of people would be good, but I'd like them to have some context before they get here. - Dank (push to talk) 12:42, 12 July 2012 (UTC)

Do you mean at stage 3? That could work. I guess we'd set up a load of discussion areas for the specifics we want to discuss and possibly work on the Signpost article at the same time. We'd just need to make sure the discussion areas are ready for when the Signpost article hits the press. Yaris678 (talk) 13:00, 12 July 2012 (UTC)

Quick update

Here's where we are, as far as I know: editors are working on and discussing their own subpages. Everyone I asked at Wikimania had strong feelings one way or the other, so we've got a chicken-and-egg problem here ... if we advertise widely before we have any kind of structure or consensus, then we'll probably reproduce the problems at the last RfC. But people will show up to oppose votes they don't like, so we have to find a way to get people to tell us what they want, without the shouting matches. What I'm hoping is that you guys will continue to do what you're doing, defending your own positions ... but also expand it a little bit, try to debate other positions that came up during the last RfC as well, so that we'll have some snappy responses to likely questions. I've asked for help at WP:AN#WP:PC2012. - Dank (push to talk) 13:56, 17 July 2012 (UTC)

Okay, a practical suggestion: would anyone object if we solicit participation here from WP:AN and WT:RFPP? (click if you don't know what those are :) That will notify all the people active in page protection, exactly the people who have data that could be really useful for us. Most of them don't spend a lot of time participating in RfCs because they know they'll get drowned out, but I don't see a lot of shouting going on here. :) - Dank (push to talk) 19:57, 17 July 2012 (UTC)

Go for it. ~Adjwilley (talk) 20:29, 17 July 2012 (UTC)

Agree to advertising it in those places. As I recall, one of the earlier RfCs started because page protectors were annoyed that they had no clear guidance as to whether PC should be used. At the very least it would be nice to know if they think we are addressing the issue with some of the drafts here.

I also agree that this issue is horribly polarising. I think at least part of the answer is introduce it slowly and see how it goes. One day we may get consensus that PC is fine but only for a relatively small number of pages... but right now that just sounds like a weird compromise.

Yaris678 (talk) 12:50, 18 July 2012 (UTC)

The nature of the focused discussions

So here is a proposal of sorts. We should have a number of focused discussions timed to coincide with an article about PC in Signpost. We should prepare in advance some text to kick each discussion off. The text should include a bit of background info as well as at least one proposals. In line with the recommendations of Richard Rumelt, The proposals should be presented in the following format.

1. A diagnosis of the challenge faced
2. A general proposal to meeting the challenge
3. Some specifics in how we can get that done.

The proposals need not be mutually exclusive. The text should invite participants to discuss the ideas with a view to improving them or thinking of completely different but better ones. Some proposals will no-doubt be rejected and others endorsed, which is part of it what we want, but we also want a discussion about why certain ideas will work and others won’t. Here is a list of the focused discussions I think we should have. I am not dead-set on this list so if people can think of a better one then go ahead and suggest it. I may even think of changes to it myself... but it gives you an idea of the sort of discussion I think we should be having. Obviously, the specifics of each item can be expanded on when we get round to writing the text for discussion.

1. Backlog management. To include 3 proposals
   1. Roll it out slowly
   2. Admins to keep the backlog manageable (and associated rule of thumb)
   3. Auto approving of changes as a last resort
2. What situation leads to what type of PC protection. To include 7 proposals.
   1. Use PC level 2 for cases of repeated vandalism by autoconfirmed accounts. Level 1 and SP L2 are for exceptional cases to be discussed at the time.
   2. Use PC for cases that are on the borderline. L1 for almost SP. L2 SP for almost FP.
   3. Use PC for egregious or consistent BLP violations, where other measures have proven ineffective or undesirable
   4. Use PC for low traffic BLP articles with a history of "stealth" vandalism
   5. PC should not be used on pages with a high edit rate
   6. PC should not be used to deal with edit warring
   7. PC should not be used for dealing with BLP violations
3. Centralised discussion of PC. To include 2 proposals
   1. A reviewer’s noticeboard
   2. A page for requesting PC protection that is distinct from WP:RfP
4. Reviewer selection. Maybe this should include two proposals
   1. An editor who meets some relatively tough criteria as per Rivertorch’s ideas in this area
   2. Roughly similar to rollback as per WFC’s ideas in this area
5. How to review. To include 2 proposals
   1. Reviewers can reject edits that are not vandalism but should explain in the edit summary or on the article or user talk page.
   2. Some words that attempt to explain how to review a stack of edits by different users
6. Terminology. To include 3 proposals
   1. Rename PC
   2. Use “go live” to describe when edits are made visible to most users
   3. Use “make visible” to describe when edits are made visible to most users

Yaris678 (talk) 17:50, 18 July 2012 (UTC)

I read the except from the book of the guy you're linking to. I think what he says makes sense, that we shouldn't try to "accommodate a multitude of conflicting demands and interests", but instead figure out what's most important and concentrate our forces on that. The only thing I would add is that he seems to ignore the human factor, that different people come to a big discussion like this with different goals ... and it's not necessary to ignore them all in order to achieve something. The trick is to listen to people and to figure out things you can give people that address some of their concerns without diverting from or undercutting the main focuses and the main goals.

I want to add something here in response to "no high edit rates" ... if that's because that's what you guys want, that's fine. If it's because of an external restriction ... namely, the fact that the draft version seemed to be slow and clunky ... that may not be necessary, we may be able to get a faster version of PC. - Dank (push to talk) 18:32, 18 July 2012 (UTC)

The general structure looks promising. I would probably add the circumstances in and method by which the reviewer right should be removed. Some very powerful points on this topic were made during the RfC. If for instance someone was blocked or warned for something other than the use of the reviewer right, should the reviewer right be removed? Should one admin be able to determine whether the right should be removed? My personal opinion is no to both questions, but I believe there are those that would disagree. —WFC— 19:49, 18 July 2012 (UTC)

About removing the reviewer right, I got a suggestion that WP:AIV might be the right place to do that. (And I'm going to respect anyone's wish to "leave me out of it" ... I will never mention names if people don't want me to.) Are we thinking of Pending Changes being used primarily against vandalism? Don't the folks at WP:AIV already have a long history of making judgment calls on when "restoring vandalism" (approving vandalistic edits, in this case) does and doesn't constitute a problem? Their experience with this, plus the fact that in some of these cases you'd actually need AIV to make the call on whether the reviewer did such a bad job that they need to be blocked themselves for approving vandalism, suggests that we should at least consider the AIV guys to make the call on removing the reviewer right. - Dank (push to talk) 20:32, 18 July 2012 (UTC) Btw, I will immediately stop saying "someone said this" if anyone complains that this is a sneaky way of voting and closing at the same time; I'm not trying to vote, I'm trying to keep the eventual vote from being derailed by people who don't want to talk now but are prepared to smack us when the vote rolls around if we don't deal with their concerns. Let me know, guys.

I also really like the structure of this, as it gives a clear snapshot of what's going on. I would definitely support a format like this for the upcoming rounds of discussion. ~Adjwilley (talk) 20:11, 18 July 2012 (UTC)

I like the structure too. Btw, on "backlog management", 3 different people mentioned at Wikimania that putting Pending Changes on some articles carries a high risk that you'll put off good-faith editors, while other pages are so obviously prone to vandalism that no one is likely to be surprised or offended by Pending Changes. Think: high school articles and porn actor biographies. High school students usually aren't offended by page protection on their schools' page, because they know very well what their classmates are capable of. - Dank (push to talk) 20:50, 18 July 2012 (UTC)

Interesting idea... althought I think I'd stick that under number 2 "what situation leads to what type of PC protection", rather than "backlog management". Yaris678 (talk) 07:31, 19 July 2012 (UTC)

• Maybe this is just me, but I have never really gotten the point of PC level two. I don't see how it would ever be a better option than semi. If registered accounts are vandalising we already have a tool for that, it is called indefinite blocking. I'm tempted to suggest we consider it deprecated before we even begin. It will simplify what needs to be worked out and how and when PC is to be used if, like semi and full, it means only one thing. Beeblebrox (talk) 20:35, 18 July 2012 (UTC)

I could support removal of level two - I never got its benefits to what we have already to deal with such situations either. - Youreallycan 20:55, 18 July 2012 (UTC)

I agree that PC2 doesn't seem very helpful, and I expect it won't be used very much. In fact, one of my proposals is to move PC2 down a level on the table describing the different types of protection, so that it comes after semi-protection instead of before it.

PC2 Semi, however, could be very useful for articles that frequently have to be fully protected. For instance, Barack Obama and Mitt Romney would both be excellent candidates for PC2 Semi: both are very high profile articles and both attract massive amounts of non-constructive editing from auto-confirmed users. ~Adjwilley (talk) 21:00, 18 July 2012 (UTC)

Question: does knowing whether PC-2 is in the picture help us to make decisions about PC-1? (I have no idea.) If not, then we can always just put discussion of PC-2 off; remember that some of the voters are easily confused when you talk about two things at once :) - Dank (push to talk) 21:04, 18 July 2012 (UTC)

Adjwilley seems to understand where the PC 2 would be of benefit - so I think we should keep it in unless its clear that its hampering discussion progress. - Youreallycan 21:08, 18 July 2012 (UTC)

(edit conflict)I think so. My impression was that many of the people voting against PC were actually voting against PC-2. That's the scary one, after all. That's the one that would affect registered editors, putting them on a lower level than reviewers. If I were a protecting admin, I would use PC1 most often, PC2 Semi on occasion, but rarely if never use PC2 by itself. ~Adjwilley (talk) 21:10, 18 July 2012 (UTC)

I never got the point of it myself and it was not used much as I remember (I have a list somewhere of the pages on pending protection when the trial ended. - I must look it out) - if it helps work on Level 1 - I wouldn't object - I did a lot of reviewing but none as I remember on level 2 articles - Youreallycan 21:15, 18 July 2012 (UTC)

If there's a "scary" and a "non-scary" version of PC, you definitely want the "non-scary" version. This has all been mentioned already, but to make sure we're all on the same page: our need for lots of reviewers limits our options. The last thing any of us wants at the end of a day of gnomish edits is to come home to a user-talk-page full of drama because 30% of WPians aren't on board with what we're doing. That's why we have to identify and fix the problems before we start, and try to find pages suitable for PC where there's really not going to be any significant pushback. - Dank (push to talk) 21:37, 18 July 2012 (UTC)

Not sure of this comment - "a user-talk-page full of drama because 30% of WPians aren't on board with what we're doing." I just don't get it, whatever you do at wiki en you will never have everyone on board. I agree that scary is not good though, - Many of the opposers of pending changes will never review a single edit - and that is ok - after time they may decide to have a go - the key to helping and improving that is that we need to focus on creating simple and usable guidelines for the tool - Your comment, "our need for lots of reviewers limits our options" - there was not a shortage of reviewers in the trial with the few thousand pages under pending protection then - Youreallycan 21:44, 18 July 2012 (UTC)

Perhaps we need to spell out that the rules surrounding PC2 are analogous to those surrounding full protection? I.e. that while reviewers and admins have the technical ability to edit freely, it would be highly inappropriate to make contentious edits without prior discussion. —WFC— 21:48, 18 July 2012 (UTC) Monty has put it far better. —WFC— 22:52, 18 July 2012 (UTC)

Yes definitely a good idea - thinking this through - I realize I am not a big supporter of pending 2 - I always dislike and object to admins editing through full protection and I see similar problems and less benefits than the clear benefits imo of pending 1 - Youreallycan 22:08, 18 July 2012 (UTC)

I think the guidelines for reviewing PC 2 edits will be easier to sort out if we nail down the scope when PC 2 should be applied. PC 2 could be used for two main types of issues, edit warring by auto-confirmed editors, or cases where there is persistent Vandalism/BLP Violations/Copyright Violations using auto confirmed accounts. Personally, I think PC 2 is a poor fit for Edit Warring. Our current full protection practices already deal with it, and trying to lessen the bite of edit warring protection will also reduce the incentive to get talking. If then PC 2 is reserved for cases where there is a history of autoconfirmed accounts adding BAD content to the article, the standard for review is much easier: Is it the sort of bad edit the page was protected to stop, if not, approve. (or review more carefully if not autoconfirmed) Using that approach, making a controversial edit through PC 2 isn't an issue, because reviewers should approve ALL controversial edits that don't violate a major policy. Monty₈₄₅ 22:41, 18 July 2012 (UTC)

Far better than I could have put it. I've stricken my previous comment and completely endorse yours. —WFC— 22:52, 18 July 2012 (UTC)

I think I agree with Monty too... although the extent to which this differs from the PC1 reviewing guideline obviously depends on what we agree that PC1 is for. I should also say that I agree with the comment above that PC2 semi would be good for certain articles that might otherwise be considered for full protection... and I should point out that Rivertorch's proposal is to assume that we use PC2!

PS. I am glad everyone seems to like the general idea of the proposal.

Yaris678 (talk) 07:42, 19 July 2012 (UTC)

AIV removal of the reviewer right

AIV removal of the reviewer right - why not - those guys are well experienced and it would be a useful place to report an extreme violator of the tool .. if any of the reviewers acceptances/rejections could be reported as clear WP:Vandalism - especially in regards to any content regardinf living people - but, - more complicated issues regarding a reviewers edit acceptance/rejections would be better reported for discussion/action to an uninvolved admin or to ANI - imo. - Youreallycan 21:08, 18 July 2012 (UTC)

If an account is vandalising with the reviewer function, removal is a no-brainer and shouldn't require discussion. That can be easily covered in policy. The areas that are more likely to cause problems are if a reviewer is accused of disallowing valid edits, or an admin is accused of improper removal. Both of those strike me as ANI territory. —WFC— 21:42, 18 July 2012 (UTC)

How is the rollbacker right currently removed? Is there a noticeboard for that, or is it just whatever forum people decide to bring it up in? If de-reviewer-ing is going to be something that happens frequently it might be useful to have a noticeboard just for that where the "specialists" can hang out. AN/I gets so much activity that admins interested in granting/taking away reviewer rights would probably miss many of the requests. What about having people report reviewer abuse at Wikipedia:Requests for permissions/Reviewer? That's where the admins who give reviewer rights hang out, and they would be excellent judges of when to take away those rights. ~Adjwilley (talk) 22:13, 18 July 2012 (UTC)

I'm of the opinion that reviewer permission removal should generally occur only after misconduct serious enough that the place to discuss it would be AN or AN/I. Last time around there were a few cases were reviewers were wrongly criticized for letting through edits with non-obvious problems. Reviewers should be expected to pass edits if there are no obvious problems, and after checking the edit against the specific reason for page protection. The standard should not require an exhaustive check against everything that could possibly be wrong. To encourage reviewers to pass edits that aren't great, but don't contain a clear reason for protection, we need to make sure that admins are not yanking the reviewer right over some minor issue with an approved edit. To provide sufficient visibility and community oversight, the discussion of removal should occur at a major existing noticeboard. (Blatant vandalism excepted) Monty₈₄₅ 22:29, 18 July 2012 (UTC)

((ec)) -@:Adjwilley - I like that idea. It wasn't a big issue in the trial , as I remember not many users were having the tool removed - occasionally a small handful of users had the tool removed in a "removal package" as they were clearly disrupting and had all their advanced permissions removed - - there was one high profile removal case but it was exceptional imo - run of the mill cases could be well dealt with on such an equivalent page - there were a couple of reviewers that I thought were not reviewing correctly and I would have liked to have been able to raise my concerns on such a page. - Youreallycan 22:35, 18 July 2012 (UTC)

Before deciding how to take away a permission, it is more important to decide what the rules are for the permission. I do not find it acceptable under any circumstances for a single administrator to remove this permission from an editor, as it is a situation ripe for abuse and misunderstanding. Risker (talk) 23:31, 18 July 2012 (UTC)

Risker! Yay. Guys, Risker knows an enormous amount about how the trial went and what opposers are thinking. I'm delighted she's here. - Dank (push to talk) 00:27, 19 July 2012 (UTC)

Oh ... on your point, Risker, are you on board with the idea that since the tool concerns vandalism, AIV is a likely place for relevant discussions? Instead of a single admin, would you just like to see a voting thread opened at WT:AIV to de-reviewerfy someone, to be closed whenever consensus is clear? - Dank (push to talk) 01:28, 19 July 2012 (UTC)

It would make more sense to have the discussion occur someplace that is commonly used for discussion. It is rare that an AIV report elicits more then a single comment or two, and practically never an extended discussion. (Gets refered to WP:AN/I or another board long before that) Monty₈₄₅ 01:34, 19 July 2012 (UTC)

Right, I'm talking about WT:AIV not WP:AIV. There's some discussion there. I don't have a strong feeling about this, and maybe the best way to find out is to try the discussions in different places and see what happens. It's just my experience that at WT:AIV, you get short, focused conversations with people experienced in vandalism-fighting; at WP:ANI, you get some of everything, and the big objection to "de-anything" is that it occasionally turns into a circus. Even when it's not a circus, public humiliation is bad, and WT:AIV is less public. - Dank (push to talk) 01:47, 19 July 2012 (UTC)

I look at the big objection at AN/I to "de-anything" as a feature not a bug, though I can understand your desire to avoid dragging reviewers to AN/I who don't want to be there. Monty₈₄₅ 01:57, 19 July 2012 (UTC)

I wasn't clear ... I think when you say "objection to de-anything" you're saying that voters will show up and oppose removing the userright, and I take your point that that's an advantage. I meant that, whenever the question comes up whether a process should even exist to remove certain userrights, the objection raised is that if you do it in a public place, then everyone wants to get their opinion in, and the whole process tends to drive editors away out of the perception that they're going to get publicly humiliated if they ever slip up. I think that's probably right, so I think quieter is better, but YMMV. - Dank (push to talk) 02:09, 19 July 2012 (UTC)

I'm inclining towards saying that de-reviewer-ing should be discussed at WP:AN... Whatever we decide here shouldn't be the final say. I think that this topic can be added to the list of focused discussions. Yaris678 (talk) 10:03, 19 July 2012 (UTC)

• I think you might be missing my key point here. Deciding how to remove a permission before one decides what the responsibilities of that permission will be is putting the cart before the horse. Decide on how the permission will be used. Once that is fairly settled, then how to remove the permission is easier to work out. Now, if the primary purpose of this permission is to remove vandalism, not to actually review the edit in its fullness for appropriateness, context, referencing and everything else...well, I think it's more realistic, but I also wonder whether it's worth all this time and effort. RC patrollers and RC patrol bots already do most of this work. On the other hand, short of making just about everyone a reviewer after a certain level of activity, we will never have enough editors to review all these edits, if PC is applied as widely as many are suggesting. Risker (talk) 14:00, 20 July 2012 (UTC)
  • I guess I did miss that point in the sense that I didn't go into that area. But since you mention it, yes, it makes more sense to decide how we are to remove it once we have decided what it is for and how you get it. I still think it should be a topic of focused discussion, but yes it makes sense to have that focused discussion after the ones on reviewer selection and how to review. I think some of the focussed discussions can happened simultaneously but this one will have to be after those two... so I guess it is looking like a number of phases of focused discussion. Choosing which topics go in which phase will be a delicate task, as will be choosing when we are ready to start the next phase. I guess you want most of the previous focused discussions to have been closed but some could be left running. This is where we need someone good to keep the show on the road.... like Dank... Yaris678 (talk) 14:28, 22 July 2012 (UTC)

Another mini (or micro) RfC

WT:PC2012/Adjwilley#Second "mini-RFC" (changes to WP:Pending changes/Protection level table).

This is relevant to some of the stuff being discussed above, particularly the idea of moving PC2 down a level to indicate that it's more severe than Semi-protection and should be used less often. Please comment. ~Adjwilley (talk) 22:05, 18 July 2012 (UTC)

No real objection to any of this - "go live" - hm, well why not - the time limit should be "achievable" and I would like to see the slow ramp up ascertain what that is - I would suggest 36 hours to encompass users from all the world having a chance to review as a good place to start - (this imo should be as low as possible to reduce the users waiting time and we can reduce it later when we are aware of the achievable minimum limits) with all automatically accepted edits being placed in a list for interested users to investigate at a later date - I think there may well be tech issues implementing this (such as - if there were three edits in a list then how would the automatic acceptance deal with this, if the third one was a removal of the first two ... etc ... I also support attempting to bring the automatic vandal tool on-board to review desired but not reviewed (not live) additions and reject them as it usually would edits ) and if there is support for the idea - the developers should be nudged asap - Youreallycan 22:17, 18 July 2012 (UTC)

Thanks. I also love the vandal bot idea. Make Cluebot a reviewer and have it revert any vandalism that's pending. This takes some workload off the reviewers as well. I'm not sure if it already worked that way before, but if it didn't, it should. ~Adjwilley (talk) 22:29, 18 July 2012 (UTC)

If I remember from the trial, I think you only need to be a reviewer to approve an edit, not to revert one. I believe that is how cluebot worked then too. Monty₈₄₅ 23:35, 18 July 2012 (UTC)

Anyone can revert an edit... but the new version won't be approved even if it is identical to an approved version. This may seem irrelevent, but:

1. It adds to the reviewer workload
2. Until a reviewer does approve it, the article is in the limbo area where an (auto)confirmed editor can make an edit and it is not shown because there are still some other changes pending.

Therefore it makes sense to give User:ClueBot NG the reviewer right and program it so that if the version before the reverted edits is approved then the version it creates is approved also. Obviously, we don't want CBNG to approve a version if it is not the same as a pre-existing approved version. Yaris678 (talk) 12:03, 19 July 2012 (UTC)

There is no practical use for PC-2; it's intended to keep non-reviewer/auto-confirmed editors from editing an article. If we have an editor in that category who is making inappropriate edits, the correct approach is to sanction the user, not the article. If there are multiple editors inappropriately editing an article, full protection is more appropriate. Risker (talk) 23:30, 18 July 2012 (UTC)

What about persistent auto-confirmed sock puppetry? Sure you can play wack-a-mole or full protect, but wouldn't PC 2 make sense in such a circumstance? Monty₈₄₅ 23:33, 18 July 2012 (UTC)

Persistent auto-confirmed sockpuppetry is in fact quite rare these days. Ironically this might have made more sense five years ago when we first started discussing these issues but in today's climate I don't see the point and completely agree with Risker's statement. Beeblebrox (talk) 20:00, 19 July 2012 (UTC)

That's interesting. Do you know why persistent auto-confirmed sockpuppetry is rare now?

I guess we could just say PCL2 is only for use in cases of persistent vandalism by multiple autoconfirmed accounts. If this does not happen then it won't be used. Yaris678 (talk) 12:09, 20 July 2012 (UTC)

That could work too I suppose. Beeblebrox (talk) 18:39, 20 July 2012 (UTC)

Unmuzzled

Considering the stuff that I said today at WT:PC2012, plus the stuff that I want to say to try to head off some problems, I think I can be more useful if I step down as a closer. I'll continue to support both sides because I think a lot of good people are doing good work on both sides. I apologize for any inconvenience, and obviously, anything that's been said in confidence will remain in confidence. Rivertorch won't be happy, but really, you guys have nothing to worry about from Blade, and I'm not going anywhere. We've asked for more closers at WP:AN.

One of the main things I want to say has to do with YRC's puzzlement that anyone is going to hassle anyone over Pending Changes ... I'm pretty sure that's going to happen unless we know where the potholes are, and avoid them. One thing that supporters are missing: in big, contentious RFCs, voters often lie, in the sense that they don't tell you the thing that's really bugging them; they give some rationale that they think is easier to digest, because they want to be persuasive. (The irony is: when I talk with people privately and they tell me what's really bugging them, I'm almost always more persuaded by what's bugging them than by the "safe" reason they gave in the RfC!) Even if the pilot program goes flawlessly, there will still be significant opposition to Pending Changes, and almost all of the admins I talked with are betting that Pending Changes won't survive in the long run ... they think we'll overreach and get burned. Here's the main objection: Wikipedia is the top content website in the world. (Google and a few other sites get more hits, but they're mainly portals, not content sites.) We went from zero to the very top on a budget of zero in a few years for just one reason: we treated all editors equally (and did a damn good job pretending to even when we didn't :) Now, you and I know that Pending Changes isn't meant to exert editorial control, only to fend off vandalism, but we know that because we're familiar with how things work ... new editors are not, and based on the experiences they've had everywhere else, they have every reason to believe that we're only going to approve their edits if we approve of their edits, which makes us no different in their eyes from the thousands of other content sites out there. That is, for most of the opposers, we're taking the thing that defines us, that we base our "brand" on ("anyone can edit"), and throwing it away so we can try something new and cool, after quick and flimsy testing, and without paying a lot of attention to any of the many objections opposers have raised. So ... yes, the opposers haven't gone anywhere, and they're going to cause trouble unless we pay some serious attention to a variety of potential problems. I'm not pessimistic, I just think we need to get busy. - Dank (push to talk) 23:48, 18 July 2012 (UTC)

I guess I can turn this into a point for focused discussion: I hear there are academics who like to run surveys on questions like this one. Let's ask them to survey new users during the pilot project to find out if semi-protection or PC-1 is more likely to give new users the sense that "not anyone can edit". (This suggests that, for the purposes of the survey, we should have more than just high schools and porn actors in the pilot project, to get better data.) - Dank (push to talk) 10:20, 19 July 2012 (UTC)

Interesting idea. It sounds a little bit like creating a second trial... only making this one a lot more rigerous than the first. Yaris678 (talk) 12:11, 20 July 2012 (UTC)

"Pilot project" is what Beeblebrox, a PC supporter, wants to call the initial rollout. He wants to avoid calling it a "trial", as if it's not real, but he also doesn't want to call it something that implies that we have all the answers already. I like the phrase, but I'm not wedded to it. I just want to be clear, I'm not trying to "demote" PC with the words "pilot project", I don't really have a preference for what we call it. - Dank (push to talk) 13:50, 20 July 2012 (UTC)

I can run over the logic behind the "pilot project" name if that helps. The simplest reason is that it accurately describes what we are endeavoring to do. A limited redeployment aimed at helping us understand the tool in a functional as as opposed to hypothetical environment. It is not a second trial since it is not designed to end with the tool being switched off and reconsidered. "Pilot project" sounds more inviting and is a bit snappier than "limited redeployment" which to me sounds more like we are sending PC off to Afghanistan or something. (I'm sorry, Mrs. Pending Changes, but your husband's unit is under a stop-loss order and he is being redeployed...)

The other reason is of course that the first PC trial is widely (and perhaps accurately) seen as an attempt to stage-manage the community and more or less trick it into accepting PC. It resulted in a lot pf bitterness and a loss of trust between large segments pf our community and the Foundation so the more distance we put between this idea and that one the better. This time around only the commmunity itself is calling the shots, Jimbo and the WMF have kept their hands off recent discussions. In fact after talking to a number of Foundation folks at Wikimania it is clear that this is quite deliberate and they don't want to intervene in any way without being explicitly asked to. So it is both a matter of accurately describing it and disassociating it with the earlier process and the mistakes that were made then. Beeblebrox (talk) 18:38, 20 July 2012 (UTC)

I would prefer "phase one", "version one" or "1.0" to "pilot". The first three imply that we are aware that there is work to do on PC. The latter implies that the introduction is intended to be temporary, and after the way the trial ended we should go out of our way not to give this impression unless we definitely intend to turn it off. —WFC— 20:49, 22 July 2012 (UTC)

No objection here. - Dank (push to talk) 00:57, 23 July 2012 (UTC)

My WP:VPT post

Here's a permalink to my question at the Village Pump/Technical page. (Bottom line: a question about redirecting people when they try to edit a page.) I just want to mention this before it gets archived on that page; it's not urgent. Generally, if VPT folks can think of a reason to shoot something down, they fire away ... if they say nothing, that's a good sign. - Dank (push to talk) 14:10, 20 July 2012 (UTC)

Dank, here are the Engineering Department goals for this year, which lays out what they plan to do, and are budgeted to do. The issue will always be in persuading the department leadership to reallocate resources to PC, which other projects should be deferred, etc. I'm not going to pretend I understand the impacts of many of those decisions (on this or other projects), so it's going to take better-informed minds than mine to make those suggestions. Risker (talk) 14:18, 20 July 2012 (UTC)

Are you saying that tweaking the "edit this page" button so that it can be switched to redirect to a specific page is going to be a major time sink? Holy crap, that must be some convoluted software. - Dank (push to talk) 15:05, 20 July 2012 (UTC)

• - Yes - its a busy developing environment which is part of the excitement of contributing here - I think we have really little work to do with the current situation - we already know where the tool worked beneficially and its limitations - a limited slow roll out of the tool without any immediate development is the way to implement this - switch it on - allow it to be requested at RFPP - the wheels won't drop off - Youreallycan 14:27, 20 July 2012 (UTC)
  • I actually wasn't thinking of that feature as a part of PC. Above, some are making the argument that we have so few autoconfirmed, devoted vandals these days that PC-2 may not be as useful now as it would have been a few years ago. This isn't just a PC-opposer thing; a lot of supporters agree. Nevertheless ... PC-2 is in the draft proposal for a reason: there are a fair number of people who want to have some option other than straight "full protection", and their concerns are valid, and I'm wondering if we can come up with something that might make them happy, without going all the way to PC-2. The idea is: simply making it easier for people to add their edits to a "working version" of any full-protected page might make them happy enough to allow us to scrap PC-2. I don't feel strongly about it, but I'm in favor of scrapping PC-2, for the reason I just mentioned, and because I think it's important in the final vote to keep things simple and not vote on two completely different features (PC-1 and PC-2) at the same time. Also, one common misconception about PC is that people who are pushing it are trying to OWN pages ... with PC-2 off the table, we could say: PC is always the least intrusive form of protection, it's not about owning pages. - Dank (push to talk) 15:17, 20 July 2012 (UTC)
    • When you say "scrapping PC-2" are you including PC2 Semi as well? ~Adjwilley (talk) 16:57, 20 July 2012 (UTC)
      • Yes, as long as we at least experiment with lighter-weight alternatives like the one I just suggested, to make it easier for people to discuss and collaborate on pages that have been full-protected. Beeblebrox and YRC are two of the strongest supporters of PC on this page, and they have problems with PC-2 as well. It might have been fantastic a few years ago, but we just don't have that many determined autoconfirmed straight-out vandals these days ... and the ones we do have, we can just block. - Dank (push to talk) 18:08, 20 July 2012 (UTC)
        • Were Beeble and YRC opposed to PC-2 and PC-2 Semi, or just PC-2? I think it's about time for a section discussing this issue specifically, since the comments seem to be spread between different sections and pages. ~Adjwilley (talk) 19:00, 20 July 2012 (UTC)

Vandalism and reviewership

Two questions that we might be ready to get consensus on: is reviewership a big deal, and is PC targeted only at vandalism? There's strong linkage: if almost anyone can be a reviewer, then we're really setting our reviewers up for failure if we're asking them to tackle much more than identifying vandalism. If there are reasons why PC should be limited to vandalism protection, then there's really no reason not to allow just about anyone to try out being a reviewer; if they're not "getting it" or they're deliberately letting vandalism through, we deal with it. People have mentioned the downside of removing the right, that it implies we don't trust the user. How about the upside of offering the right to fairly new users? Doesn't this say something about how willing we are to trust new editors? In that way, it's similar to "anyone can edit", which has been a huge success. - Dank (push to talk) 16:27, 21 July 2012 (UTC)

Heh, okay, I guess we aren't ready to talk about this :) My guess is the sticking point is that newer users wouldn't know how to handle BLP violations. I don't have an opinion on whether PC should handle BLP violations, but I feel strongly that anyone who wants to see PC survive in any form shouldn't be trying to use it as a BLP-fighting tool from the very start, because BLP questions are the most contentious, least understood, most drama-producing problems anywhere in Wikipedia. Linking PC to that mess, before we have a chance to demonstrate that PC can work as an anti-vandal tool, will kill off PC quickly, I think. Thoughts? - Dank (push to talk) 13:41, 22 July 2012 (UTC)

Hi Dank, I think I agree with you and I think Rivertorch will too - Wikipedia:PC2012/Rivertorch says "[PC] is only for preventing vandalism". He also has some pretty tough standards for becoming a reviewer. At the other end of the spectrum, Wikipedia:PC2012/Wnt says that anyone with 300 edits should automatically be a reviewer. It also requires that reviewers do some things that won't come intuitively to new users so I can see it working in practice.

I agree that there seems to be a trade off between how complicated the reviewer task is and how low the barrier to entry can be. But... maybe in the first instance we want a highish barrier to entry and to say that PC is only for dealing with vandalism (as Rivertorch recommends). That means we are first doing the thing that is least likely to fail. Once we've got that working nicely we then decide where to take it next.

Yaris678 (talk) 14:55, 22 July 2012 (UTC)

Given the controversy that is associated with pending changes in general, the effort its supports have gone through to get it activated again, and how well existing tools fight routine vandalism, I don't think its all worth it if the initial target is vandalism. Routine vandalism is a nuisance, but it really isn't that harmful beyond the minor embarrassment to Wikipedia caused when people notice it before its reverted. Vandalism effects a great many articles, and its always a judgement call whether the vandalism to a particular article warrants semi-protection, just as it would be for applying PC1. But given the nature of routine vandalism, that judgement call will need to be made very often, and we could end up with a large number of articles under protection, taxing the reviewer pool. On the other hand there are really important reasons we need to fight BLP and Copyright violations, they can cause real harm to other people. While it can be more complicated to review BLP or Copyright issues, it is also more important we deal with them. Also, assigning PC1 to articles with persistent BLP/Copyvio violations will represent a much smaller pool of articles being protected, and will allow the community to more carefully consider the suitability of each article for PC1 protection. Monty₈₄₅ 15:07, 22 July 2012 (UTC)

On Wikipedia, BLP is the third rail. If you take any new process that just barely has enough support for a pilot project, and get everyone judging it according to how well it solves our BLP problem, the process will be dead within months. People who are experienced with RfCs, both supporters and opponents of PC, agree. Personally, I've never seen evidence that our BLP problems are fixable, not in any sense that's going to satisfy everyone. But if PC gains support for how it's handling vandalism, then logically, it ought to work against some subset of BLP violations as well ... but what subset, we'll only be able to tell by trial and error, carefully. - Dank (push to talk) 15:55, 22 July 2012 (UTC)

I'm afraid I can't agree with you, Monty. Our copyvio policy is complex to the point of inscrutability for many editors, and I think nuanced enforcement of it is best left to the subset of Wikipedians who (1) understand it well enough to apply it in a fair and evenhanded way and (2) give two hoots about it in the first place. I don't think there's any reason to suppose that that subset particularly coincides with whatever group ends up with the reviewer flag, and I think that's true even if the standards for getting the flag are fairly high (as I've proposed they be). As for BLP issues, they're less nuanced but at least as prone to disagreement among editors as are copyright issues, and officially making them part of the PC rollout will mire it in unneeded controversy from day one (if not before).

In my proposal, I recommend that PC's official scope be limited to vandalism but that considerable leeway be given to reviewers in deciding what edits to accept. In practice, that would mean that reviewers would be free to reject edits that are egregious violations of other policies (not to mention other transgressions, such as writing indecipherable prose). It's also worth noting that many WP:BLP violations also qualify as WP:Vandalism and are already subject to rollback under the latter policy, as they would be under an anti-vandalism-only PC. Rivertorch (talk) 17:37, 22 July 2012 (UTC)

The WMF have set the de facto standard for reviewership, by equating reviewers to rollback in the article feedback tool. Reviewers now have access to the hide tool, which is not unlike revdelete, albeit only for article feedback. I think we should pretty much defer to rollback policy on the granting and removal of the right, with reminders that removal of rights should not be done as punishment, and should only be done in relation to misuse of the right. —WFC— 20:42, 22 July 2012 (UTC)

Okay, I'm wondering why things went quiet, did we take a wrong turn? Is BLP the problem? If so, could we try two different versions of PC at the same time, one targeting BLP and one not? Would that satisfy everyone? - Dank (push to talk) 02:00, 25 July 2012 (UTC)

BLPs require care, yes. However, as has been stated multiple times, it is unknown whether or not reviewers are legally liable for any libelous/defamatory edits they end up accepting. WMF legal has not provided a clear answer on this, as far as I am aware. —Jeremy v^_^v ^Bori! 02:16, 25 July 2012 (UTC)

Jeremy, discussion of hypothetical (and highly unlikely) legal consequences of edits has a chilling effect on discussion, and is generally discouraged on talk pages, particularly when the point has already been made many times. Your views are, of course, welcome, but leave the legal stuff to the lawyers, please. - Dank (push to talk) 02:50, 25 July 2012 (UTC)

Dank, a few opposers of PC did so specifically because of the bolded part. They're afraid reviewers will get sued by someone because they accepted an edit that they weren't aware was defamatory, which is a significant concern requiring clear answers from WMF legal. I'm not trying to chill the discussion; it's a valid concern to some. —Jeremy v^_^v ^Bori! 03:50, 26 July 2012 (UTC)

I heard the WMF's General Counsel speak at Wikimania; he would really prefer that we not try to work out the details of defamation law on-wiki. I promise to email him with whatever the final outcome is of our discussions; he might prefer that we be careful with how we define what it means to "accept" an edit. He prefers not to deal in hypotheticals, so I'd prefer not to email him until we've agreed on something. - Dank (push to talk) 03:52, 25 July 2012 (UTC)

(Warning: the following is not legal nor professional advice.) For whatever protection it brings, Wikipedia's general disclaimer in the page footer remains applicable: "Wikipedia makes no guarantee of validity." The page also contains a section, "No formal peer review," which disclaims responsibility for content. isaacl (talk) 03:19, 25 July 2012 (UTC)

That's not the issue the opposers have been bringing up. They're afraid of reviewers getting sued, not the WMF. —Jeremy v^_^v ^Bori! 03:58, 26 July 2012 (UTC)

The "No formal peer review" section states the following in bold: "None of the contributors, sponsors, administrators, or anyone else connected with Wikipedia in any way whatsoever can be responsible for the appearance of any inaccurate or libelous information or for your use of the information contained in or linked from these web pages." isaacl (talk) 01:56, 27 July 2012 (UTC)

This is not the case with reviewers, who have to approve an edit in order for it to appear in the publicly-viewable version of the article. I would expect that page to be edited come December. —Jeremy v^_^v ^Bori! 04:19, 27 July 2012 (UTC)

Personally, I don't see any reason why the disclaimer would change; I doubt anyone is willing to assume any legal responsibility for accuracy of their edits, either through reviewing pending changes or regular editing. isaacl (talk) 04:54, 27 July 2012 (UTC)

• Pending and the task of reviewing is not simply the removal of WP:Vandalism - we have bots that are pretty good at that these days - and the edit filter. If you restrict reviewers to reverting vandalism you might as well not bother switching it back on - Just a query , is there consensus that vandalism reverting is the sole purpose of the tool somewhere - (discussion is so fragmented with all there different pages as to be falling apart. - What we need to do is lay out how to review - this has been done somewhere previously - and it wasn't/isn't look at the edit and if it is not WP:Vandalism you must accept it - - The first question a reviewer should ask themselves imo is , "is this desired addition beneficial to the article? - if the answer is yes then accept it - if the answer is no but it would be if it was cited the go get the cite and then accept it - if you haven't time to go get the cite either accept it and then add a citation required template or if the content appears a bit contentious/controversial then reject it and move it to the talkpage for consensus discussion/ its a judgment call here for the reviewer - after a quick search you find nothing about it at all the reviewer might feel its uncited and controversial/contentious and they might choose to simply reject it with and explanatory edit summary and not post it to the talkpage at all - these are the type of questions and tasks a reviewer should be asking himself in regards their understanding of WP:Policy and guidelines - Youreallycan 16:04, 25 July 2012 (UTC)

• Responsibility of the reviewer of the edit he accepts - A reviewer has a degree of responsibility for any content he adds to the project just like we all do (as per the foundations previous statements) - There is/would be in the extreme minimal chance that there would be some issue with the addition, mitigation due to the reviewer being an unpaid good will contributor and not the primary ID that is desirous of making the addition. This responsibility is imo/and wasn't in the trial ever going to be a problem in normal review work - Youreallycan 16:09, 25 July 2012 (UTC)
  • Okay, thx much YRC, that's helpful. Can I get another vote for "If no BLP, then don't bother?" - Dank (push to talk) 17:18, 25 July 2012 (UTC)
    • There is consensus for usage of the tool - we need to focus on that and not all the (imo - red herrings such as "am I responsible" and "lets start just allowing reverting of vandalism and see how it goes" and "BLP or not BLP"? ) - we have had the trial and from that experience we know how it works and where it is good - it just needs switching on under similar conditions as the trial - there is really little to talk about imo - the trial is over - we know the benefits/negatives - Its not a danger its just another beneficial tool to add to the box - somewhere just less restrictive and more open and with other subtle benefits than WP:Semi protection . The wheels as we know from the trial, will not drop off. - This position goes diametrically opposite of your statement - "But if PC gains support for how it's handling vandalism, then logically, it ought to work against some subset of BLP violations as well ... but what subset, we'll only be able to tell by trial and error, carefully. - Dank" - You do understand that there is already and on multiple community comments/discussions, clear support for WP:Pending Changes and that we already had the trial and error bit ? - Youreallycan 17:27, 25 July 2012 (UTC)
      • Right, when I said "I'm wondering why things went quiet", I was acknowledging that I might have taken a wrong turn by saying the bit you're quoting. First, I want to see one or two more people say that they think it's important to include BLP from the very start, and we'll go from there. - Dank (push to talk) 18:35, 25 July 2012 (UTC)
        • We do not even need to discuss such a question - there is already majority support for such usage in the RFC - That we now need one or two users to say it again here beggers belief - there are only limited contributors to this discussion, six or seven..? which is clearly an issue after hundreds of users have opined - the majority of the commenter's in this discussion are strong previous opponents of the tool , User:Rivertorch and User:Wnt, to name two ......., have been two of the most vocal opponents of the tool and as such the format created to implement the switch on with low profile multiple user pages, just is not working. The discussion needs to move away from vocal opponents of the tool offering up suggestions to limit the tool as much as is possible, to a post consensus closed RFC in support of switching on the tool, and how can we do that in the most broad beneficial manner. Youreallycan 18:41, 25 July 2012 (UTC)

Okay ... I wanted a little more guidance first, but you're getting upset, so I'll say what's on my mind. You're right that we just had 61% vote in favor of something that seemed to include BLP issues. If you're right that there's still some support for that, and you probably are, then my proposal is that we split the discussion and the proposal in two at this point, and plan to run both programs at the same time starting in December. A lot of people in the discussion so far prefer less-experienced reviewers and simpler issues; your program that involves BLP is obviously going to require knowledgeable reviewers. Wnt's proposal and discussion comes the closest to the simpler approach, so I'm going to start hanging out over there and see what we can come up with; people who want to go the BLP route are welcome to keep using this main talk page. I think everyone gets a win here, because this gives you two shots to succeed with PC instead of one ... either or both plans might work, and the two plans won't weaken each other, since they rely on different reviewers and have different goals. You can put together whatever program you want, and have whatever mini-votes or RfCs you want, without interference from me; if it turns out you need someone to lead the discussion, I'm sure we can get someone who will help, but I doubt you'll need that. Your plan will of course go live on December 1, and hopefully it will work out just like you're expecting. - Dank (push to talk) 19:23, 25 July 2012 (UTC)

I am not upset - please do not suggest that. -WT:PC2012/Wnt - is one of the biggest rejectors of the tool and consensus support and the tool in general. Its not my plan - its community discussion already closed - As per previous discussion, its clear that a user granted WP:Rollback would or should be able to interpret WP:Policy and guidelines and make the simple assessment of a desired addition. There is only one plan - review the edit - properly WP:Reviewer - For WP:Vandalism, we already have good tools and bots for that - Reviewing is not rocket science - is it cited - yes/no - is is contentious, yes/no , do my reviewer investigations support it as verifiable - can I add it now or does it require talkpage discussion and consensus - Youreallycan 19:46, 25 July 2012 (UTC)

Not every edit that goes into an article has to include a citation (think copyediting/rewording), and not every reviewer is going to spend time doing their job if the edit (and any citations included) are above their head. In the latter case, they will punt. —Jeremy v^_^v ^Bori! 03:53, 26 July 2012 (UTC)

Yes, that's the kind of thing we need to get across and lay out for reviewers -we do not expect you to "take a punt" - please avoid reviewing the desired addition if that level of limited investigation is all you have time for - thank you - Youreallycan 18:10, 26 July 2012 (UTC)

You're misunderstanding what I'm meaning. If the statement to be reviewed plus its sources is incomprehensible to a reviewer (Whether because it involves knowledge beyond their comprehension or it's in a language the reviewer can't read) they're not going to bother reviewing it and will more likely than not just let it sit; i.e. they punt it onto the next unlucky CRASH cop in line. —Jeremy v^_^v ^Bori! 04:26, 27 July 2012 (UTC)

(think copyediting/rewording) - yes - these are the easy reviewing work and such clearly beneficial edits should be accepted rapidly - Youreallycan 18:14, 26 July 2012 (UTC)

Okay ... I take the silence as agreement with YRC that Wnt's page isn't the best place to work this out. I'll write something up on my own page, after I catch up on my copyediting and writing duties. - Dank (push to talk) 13:40, 26 July 2012 (UTC)

Maybe I should go into more detail on my subpage, but I do want to say a couple of things at this point. Or three or four things. About punting—yes, this is an issue, and it's one that has little to do with the general competency of the reviewer. Punting is a part of life at Wikipedia. I punt frequently. I do it on edit requests to semi'd articles where the topic is something I know so little about that I can't make an informed decision in a reasonable amount of time. And I punt, in a sense, on vaguely questionable edits to unprotected articles by simply leaving them stand in the same sort of instances. I can only hope that someone else who actually knows something about the topic or who has ready access to reliable sources will come along and either verify the changes or revert them. But with PC, there's a difference: clicking a button to accept a pending edit will make me responsible (maybe or maybe not legally, but certainly ethically) for changing the article. If the policy says that only vandalism may be rejected, then I damn well am going to punt unless I'm pretty sure, because there's no way I'm going to violate policy by rejecting a non-vandalistic edit, no matter how atrocious it may be, and there's no way I'm going to be responsible for making such an edit go live. I see this as being a huge issue even with pending edits that can be easily evaluated. With pending edits that are difficult to evaluate for some reason, there's an extra element of risk. Those who insist that pending changes is precisely analogous to semi-protection seem to miss the fact that most edit requests to semi'd pages are rejected, and rightly so. (That punting may effectively prevent me from making unrelated edits—even really important ones—until a non-punting reviewer happens along just adds an extra element of complexity, if not outright horror. Reviewers' Noticeboard, anyone?)

Notwithstanding all that, I still balk at the idea of making violation of policies other than WP:VAND official grounds for rejecting pending edits. We have a lot of policies, and I do think that slippery slope concerns are legitimate concerns in this case. WP:BLP, in particular, is interpreted very differently by various editors, and the potential for abuse and conflict in applying it to PC is enormous. WP:COPYVIO seems to have a fair amount of leeway as well, for whatever reasons. And even the core content policies have some wiggle room. Edits blatantly violating of any of the above, as well as edits that are so poorly written as to be unsalvageable, ought to be grounds for rejection according to the reviewer's discretion, but only if there's some reason to think the reviewer knows what he or she is doing. And that's why an editor needs to meet some sort of meaningful threshold before being handed the reviewer bit. Rivertorch (talk) 06:02, 27 July 2012 (UTC)

Yes , this is your clear position - You don't support PC but if consensus support has decreed it is to be switched on then you support PC only being able to revert WP:Vandalism - Your position that there should be really high levels of contributory history and then that the tool is only a vandal reverter seem at odds with community consensus implementation of a beneficial switch on of the tool - Your imo extremely restrictive criteria for allowing the protection to be used would result in the cause and effect the removal of the tool in any usefulness at all from the toolbox. Your fears of punting and incompetent users of the protection and opinionated control of content were not apparent in the trial results - as I said above - there is consensus support for the tool and we should be looking for the best usage of it - reverting clear vandalism is not one of them, we already have plenty of good tools and options in current usage for that task - If you are understanding of WP:Policy and guidelines enough to be awarded Rollback then you should also able to review to a decent standard - Desired additions should not be accepted unless the reviewer considers it to be beneficial to the article in regards to wikipedia policy and guidelines - If you are unsure if a desired addition is beneficial or not then please do not take a punt, take your time and do an in depth review or leave it to someone who has more time is a message we need to get over - As far as anyone is free to edit has been a huge success goes in regards to Dank's comment - Yes I agree, I know of a fair few articles that are indefinitely semi protected where pending protection would be a step down the restrictive ladder for allowing editing by unconfirmed users - I made hundreds of reviewer edits during the trial and as I remember almost none of them were to remove WP:Vandalism - I would also suggest that getting users to take responsibility for adding any content that isn't clear WP:Vandalism would be a no starter - I could/would support a reviewers noticeboard as being very likely of a benefit - but the article talkpage is already there for the specific edit - if you reject an edit that is borderline the reviewers should immediately go to the talkpage and open a discussion to explain within WP:Policy and guidelines why they think the desired edit is not beneficial to the article and discussion and consensus will/can/should go from there - Youreallycan 13:51, 27 July 2012 (UTC)

First of all, I think it would be helpful to simply accept that you and I disagree over what exactly what consensus has been reached to date. You see consensus for a broad application of PC; I see narrow consensus. Whatever. I'm basing my proposed policy points both on my sense of consensus to date (if you look at the recent RfC, many Option 2 endorsers mentioned vandalism as their key concern or only concern, so I'm not exactly pulling something out of thin air here) and on what I think will be most beneficial and least harmful to the project. In any event, I don't think it's particularly helpful for you to keep reminding everyone of my opposition to the tool. I oppose all sorts of things, both on- and off-wiki, that I also accept as inevitable. PC in some form is imminent, and a new consensus will have to be reached to determine what exact form it will take.

What you call reviewing to "a decent standard" is frequently not good enough when it comes to determining whether to include or exclude content. If this were just any web site, very likely it would suffice, but we're attempting to build the world's premier encyclopedic reference. Very often it takes specialized knowledge of a topic to avoid introducing not only subtle distortions, such as undue weight, but even rather blatant errors. You're quite right that article talk pages are generally the place to hash out questions over content, but when it comes to little-watched articles it can take months or even years for anyone with specialized knowledge to wander by. With a good-sized pool of reviewers, a dedicated noticeboard would increase the odds of someone with that knowledge noticing the question in time to respond to it promptly. Rivertorch (talk) 20:16, 27 July 2012 (UTC)

I see limited and steady ramp up of the tool n a similar manner/number of articles as during the trial - Yes I disagree with your interpretation of the historic consensus - as for your claim that, "PC in some form is imminent, and a new consensus will have to be reached to determine what exact form it will take.", the tool will be switched on as per the position in the RFC , no new consensus s required for that - There is no consensus at all that that tool is only for vandal reverts - Put it to a new RFC perhaps - Do you agree that WP:Pending changes should only be used to revert WP:Vandalism - Youreallycan

• @User:Rivertorch - What you call reviewing to "a decent standard" is frequently not good enough when it comes to determining whether to include or exclude content. - Youreallycan 20:38, 27 July 2012 (UTC)

As far as I can tell, it's self-evident. Do you feel competent to assess "improvements" to articles as diversely esoteric as string field theory, lymphopoiesis, Siachen conflict, Ainulindalë, Meta-epistemology, and Vedic meter? (Rhetorical question. I doubt that any single editor on this project could correctly identify certain edits to each of those articles as constructive or unconstructive.) You want a citation to something showing that Wikipedia has articles containing utter garbage? Google it for yourself.

As for PC's scope and purpose, those questions are among the ones that will be decided over the coming weeks. Of course that question will constitute at least part of an RfC in the near future. If you think that question has already been definitively determined in the previous RfC, I fear we'll just have to disagree. Rivertorch (talk) 21:53, 27 July 2012 (UTC)

• Let's ignore PC's scope for a moment, and stick to the reviewer's role.

  If a reviewer's scope is limited to rejecting obvious BLP violations and rejecting obvious vandalism (for instance), and we explicitly state that edits which pass these hurdles *MUST* be accepted, then reviewers can confidently do whatever it is that we are asking them to do. The WMF's terms and conditions insist that community policies must be followed as a condition of using the site, and also make clear that responsibility for an edit lies with the initial editor. If on the other hand we follow Youreallycan's bolded suggestion that reviewers take responsibility for the entire content of an edit, it is difficult to argue that they are exercising a form of moderation or editorial judgement, making them just as responsible for the content as the initial editor.

  Perhaps we should make clear that if a reviewer wishes to revert a change which they disagree with, but which reviewer policy does not allow them to reject, the edit must be accepted, and after this they have the option to revert just like any editor? This may seem a bit bureaucratic, but it would reduce the argument that PC equates to censorship, and frankly eliminate the necessity for a reviewer to be impartial. If a reviewer with knowledge of/a POV on the subject disagrees with an edit which is neither vandalism, nor a BLP violation, the obligation to accept and then revert would ensure they are still subject to the same rules (specifically 3RR) as everyone else, rather than exercising some sort of superior editorial right.

  Here are a few examples of how that system might work in practise. If Watford F.C. were under pending changes, and someone who seemingly dislikes the club were to change "Watford's kit has changed considerably over the course of the club's history." to "Watford had little identity in their first 100 years of existence; their early kit colours did not catch on, and were changed frequently". It cannot be regarded as vandalism, as the fact that we changed our kits regularly is reliably sourced, and the "little identity" statement, while probably POV, is not obviously disruptive. I would therefore have to accept the edit. Having accepted it, I would then be entitled to revert as an ordinary editor, on the basis that I believe it violates NPOV. If an edit war were to ensue, then rather than break 3RR I would be obliged to seek external opinions on whether or not the addition does violate NPOV, just as I would be if pending changes wasn't there.

  Another example: if someone were to post, on the club's article, unsourced allegations that a named, living individual is guilty of doping or match fixing, I could reject it: regardless of my POV on the club, that would be an objective breach of BLP policy. Were they later to return with the same allegation, with multiple, reliable sources confirming that the individual has been found guilty of doping or match fixing, I would be obliged to accept it. Having accepted it, I would then be entitled to decide whether that edit constitutes undue weight, and revert if I felt that it was (although having already rejected the unsourced edit, it would probably be wise to instead seek the opinions of uninvolved editors at WT:FOOTY) —WFC— 10:16, 29 July 2012 (UTC)