Wikipedia talk:Pending changes level two protection for problematic BLPs

This is the talk page for discussing improvements to the Pending changes level two protection for problematic BLPs page.

Put new text under old text. Click here to start a new topic.
New to Wikipedia? Welcome! Learn to edit; get help.

Mark historical

I'm not sure where the idea to start working this up as a new policy came from, but I wanted to jump in right away to try and stop it. We just managed to get rid of pending changes, after considerable time and a rather substantial brouhaha. Pending changes adds nothing that isn't available already, other then adding complexity and additional bureaucracy. Leave it unused, just as it currently is.
— V = IR (Talk • Contribs) 19:01, 2 June 2011 (UTC)[reply]

The point of removing PC protection from all articles was to terminate the trial, so that a long-term policy for its use could be discussed without distractions. It was expressly not to scuttle pending changes forever. The benefits of PC level 2 in regards to BLPs are that it can stop defamation without recourse to full protection, since edits wouldn't be displayed to most readers until reviewers had approved them. Should we instead choose between letting BLP violations continue, or locking everyone except administrators out of articles? Chester Markel (talk) 19:07, 2 June 2011 (UTC)[reply]

There's no problem with our current vandal patrols and page protection facilities, which makes PC unnecessary. There's simply no need to proceed here. PC is dead, and it should stay that way.
— V = IR (Talk • Contribs) 19:16, 2 June 2011 (UTC)[reply]

I've presented an argument as to why conventional page protection is unsatisfactory for certain BLPs. Simply denying my conclusion, without actually refuting the argument, has little logical merit. Chester Markel (talk) 19:20, 2 June 2011 (UTC)[reply]

Ohms law is quite wrong here. The recent RfC was to stop the trial, not to bar any future use of pending changes. I broadly support this proposal. Fences&Windows 00:23, 3 June 2011 (UTC)[reply]

Yea, the recent RfC was about stopping the trial. I never denied that (saying that I'm "quite wrong here" completely mischaracterizes my remarks above). Regardless, PC is currently gone, and it should stay that way. Yourself and Chester may support this in some form or another, but I certainly don't. We don't need PC, which makes this proposal a non-starter in any form.
— V = IR (Talk • Contribs) 00:38, 3 June 2011 (UTC)[reply]

I understand many editors are concerned about the delay in ending the trial, and the recent wheel war/involved block/arbcom fiasco. I'm not pleased with those events either. So I'm going to make a Nixon goes to China argument: examining my comments regarding the injunction and preceding events, it's obvious that I have strongly condemned the recent PC-related disruption. Therefore, in proposing a policy for the limited use of PC protection on BLPs, I am not endorsing power struggling or disregard for consensus. This policy, if adopted, would be implemented subject to consensus, and expressly prohibits the granting or revocation of reviewer rights for wikipolitical reasons, by limiting the criteria considered to mainspace editing. I will personally pursue the desysopping of any administrator who uses a new, limited implementation of pending changes to create another disruption. Also, it should be observed that even without PC protection, opportunities to wheel war, etc. are hardly absent.

How would pending changes protection benefit Wikipedia? Some articles have intractable BLP problems, such that semi-protection cannot resolve. Our choices under the present protection regime are to limit editing to administrators, or let the defamation occur. Neither option is harmonious with Wikipedia's values as an open-editing but BLP-sensitive project. It could hardly be said that adminship is "no big deal" as the number of articles to which sysops have exclusive access grows. And if Wikipedia becomes "the free scandal sheet that anyone can vandalize" in relation to certain articles, our reputation and credibility will suffer. Chester Markel (talk) 01:33, 3 June 2011 (UTC)[reply]

You seem like a reasonable guy, and everything that you're saying makes sense. I can even agree with the underlying idea of enabling something that is less than either semi- or full- protection but more than nothing, in principal. Regardless, pending changes brings so much more baggage to the table, both technically and politically, that any use of it here on en.wikipedia is going to be problematic. Unless and until those issues are addressed to the satisfaction of most editors proposals such as this are going to remain non-starters.
— V = IR (Talk • Contribs) 11:30, 3 June 2011 (UTC)[reply]

A few questions

I think this proposal makes sense for the most part, but I have a few questions:

Can someone give some examples (situations or actual pages) as to when semi-protection is not enough to stop constant BLP violations, but full protection is too much? I was thinking when there are autoconfirmed users edit-warring over some material, but the proposal says, "Genuine, tenable disputes between editors in good standing as to what BLP requires for particular articles do not form a sufficient basis for protection". So would this be mainly to combat autoconfirmed sockpuppets or something? I must admit that I try to stay out of BLP-related disputes as much as possible lest my blood pressure rise too high.
Would "pattern of edit-warring" be a reasonable addition to "Standards for revoking the reviewer permission"? It seems like edit-warring/disputes is the main issue here, as common BLP vandalism can probably be dealt with by semi-protection.
I'm wary of letting the reviewer bit get too much like a free handout as it was during the PC trial. "Standards for assigning the reviewer permission" seems a bit tougher, but does anyone else think that we should set even better-defined criteria? This proposal isn't for PC on just any old page, but for pretty bad situations—so we shouldn't give just any old user the reviewer flag.
Can this be extended from "only to be used on articles with BLP problems which semi-protection cannot effectively resolve" to also including OTRS-involved articles? I can think of several times when maybe semi-protection would handle an OTRS complaint, but PC would be better at tracking the edits to the page.

/ƒETCH COMMS / 03:50, 3 June 2011 (UTC)[reply]

Abi Titmuss (edit | talk | history | protect | delete | links | watch | logs | views) is an example of a BLP that's had problems semi-protection cannot effectively resolve, without an associated legitimate dispute about the application of the policy. It was fully protected[1] on 22:46, 23 March 2011 because of "persistent vandalism", as a result of edits such as [2]. While the account used for the defamation was blocked, there are probably more where that came from. At the time of the offending edits, the article was already semi-protected[3]. Placing articles such as these under semi-protection plus level 2 PC protection would allow more editing, while still stopping the BLP violations.

We could add recent, substantial edit warring warring as a disqualifying feature for receiving the reviewer right. However, "common BLP vandalism" and non-vandalism but obvious BLP violations over which there's no legitimate dispute is what the PC protection is designed to handle. Since most blatant BLP violations can be resolved through semi-protection, most would not result in PC protection. But there are a few situations which would benefit from it.

If an OTRS admin believes that an article has severe BLP problems which merit PC level 2 protection, they could add it without trying only semi-protection first. The criterion is that semi isn't believed to be sufficient for some reason, not that it was actually shown to be ineffective. For instance, if BLP-violating material is being inserted by autoconfirmed accounts, there's no point in using just semi-protection, since they would have edited right through it.

The criteria for assigning the reviewer right could certainly be narrowed. As a preliminary measure, rights assigned during the trial could be removed by a bot, to avoid having to manually review each user holding the privilege and determining whether they still qualify, and any associated personal conflicts. I am wary, however, of turning reviewer rights requests into mini-RFAs, with associated wikipolitical issues. Any administrator should be free to issue or revoke the permission at their discretion, limited by application of the established criteria. We might further specify that any granting of reviewer rights, denial of requests at the permission page, or removal of rights be accompanied by a statement articulating the reasons for the decision, in terms of the affected editor's mainspace contributions, and the guidelines for granting and removal ultimately established. This would normally avoid admins granting the reviewer bit to editors simply because they like them, or removing it because they don't. Chester Markel (talk) 04:21, 3 June 2011 (UTC)[reply]

I just want to comment that the discussion approaches of Chester Markel, Fetchcomms and (section below) TotientDragooned seem very reasonable and their understanding of the issues involved appears "on the money". I agree that a rethink of reviewer rights will likely be needed for any reintroduction of PC. I think it's worth pointing out that the premise to one of Fetchcomms' questions (and Chester's answer), concerning articles for which "semi is not enough, full is too much", is somewhat missing one of they key points of PC. Even if semi were enough, it still locks the article from editing for users who aren't autoconfirmed (which corresponds to the vast bulk of readers) whereas PC means the article remains accessible. There's a swings and roundabouts situation here: arguably the loss of instantly responsive editing that PC imposes, is worse than the process of taking proposed changes to the talk page that semi involves, and there is a case that we should be encouraging more collarborative engagement with talk pages while discouraging the kind of "drive-by" editing that PC allows. That's a bigger argument of course but it's worth being aware "is semi insufficient/full too harsh" rather simplifies the issue, since PC doesn't fit exactly in between the two extremes. It brings its own quirks, dynamics, pros and cons into play. Specifically on the issue of BLPs, what do we miss out on by excluding contributions from non-autoconfirmed readers? On the one hand it's difficult to imagine such a user managing to include a well-cited new fact in an edit - wikicitation is technically difficult for novice editors. But wikignomery such as spelling and grammar fixes, or removal of spurious or misleading information, is quite possible (though the latter would require a convincing edit summary for a reviewer to pass it!). TheGrappler (talk) 15:39, 3 June 2011 (UTC)[reply]

During the PC trial, some articles that were semi-protected due to heavy vandalism were moved to pending changes instead. It was found that PC protection was troublesome in this context, because the number of vandal edits presented a high workload for the reviewers reversing them, and made a mess of the page history. So, in any heavy vandalism situation, conventional protection should continue to be used, to ensure that vandal edits cannot be made at all. We could certainly use level two PC protection instead of semi-protection for BLPs which are subject to infrequent vandalism/defamation. BLPs are protected at a level of bad editing that would not normally justify it, in other contexts, due to the sensitivity of the subject matter. Level two PC has proven to be more effective than semi in this situation for stopping the public distribution of defamatory material, and also allows more editorial access in some ways, though not in others. Level one PC might even be enough, though the higher level should probably be used to ensure that BLP problems are more completely resolved.

So the use of semi plus level two PC, where semi alone isn't enough, is not the be all and end all of pending changes. It's merely the most obvious, most clearly justified situation in which PC protection should be used. Given the controversies that PC has aroused, I suggest proceeding judiciously by starting with PC in this clear cut case. Later, we can have a new community discussion to determine whether it might be beneficial to use PC protection more broadly. Chester Markel (talk) 18:59, 3 June 2011 (UTC)[reply]

Justin Bieber is another example of an article that repeatedly gets vandalized by autoconfirmed accounts. I even have a link on my userpage where a vandal posts "I just need ten edits before I can annoy Bieber"! (He was indefblocked after that edit, which was his nineth.) Reaper Eternal (talk) 13:18, 21 June 2011 (UTC)[reply]

Moving forward

I believe we should leave this under discussion for several weeks while the proposal is refined, then put it to a vote. If there's a poll too early, the policy change will simply be voted down because it isn't quite correct. But we will need a poll at some point, in order to have an ascertainable consensus for implementation. Chester Markel (talk) 04:53, 3 June 2011 (UTC)[reply]

Oppose

Oppose. I'm totally and unalterably opposed to any further backdoor or front door introduction of pending changes.—S Marshall T/C 10:23, 3 June 2011 (UTC)[reply]
Second S Marshall - We only just got rid of it 10 days ago(ish), give us a break. Fish ^Barking? 16:13, 3 June 2011 (UTC)[reply]

It would be helpful to have a more detailed explanation of your reasoning. The proposal isn't up for a vote yet. Chester Markel (talk) 16:15, 3 June 2011 (UTC)[reply]

It would be helpful if we didn't start this again - This happened during the trial vote, people would oppose, others would chime in "please expand on your reasons", etc. I am, like S Marshall, totally opposed to the introduction of Pending changes in any way, shape or form. Period. Fish ^Barking? 19:27, 3 June 2011 (UTC)[reply]

I don't want to flat-out say oppose, but I guess I'd say a sort-of "not now". I realize that this proposal is a good faith effort to, in effect, brainstorm on how PC might be used going forward, and I think that's fine. I also tend to think that the kind of usage described here is actually the best way to go with PC—eventually. But I also think that some things need to be worked out first, before this proposal can be given full, and fair, consideration. For one thing, we really do need to deal with the issues of reviewer qualifications and responsibilities, as has already been raised below. For another, there are several technical issues that were brought up in the recent discussions, things that need to be worked out with the developers, that should also be resolved so that the community can better assess how well (or not well) PC will end up working, before we decide whether we want it working at all. --Tryptofish (talk) 19:01, 3 June 2011 (UTC)[reply]
Couldn't have said it better myself. Well spoken.
— V = IR (Talk • Contribs) 20:21, 3 June 2011 (UTC)[reply]
The standard of review and criteria for assigning the privilege should certainly be firmed up prior to implementation, which I believe we can do. The technical implementation of pending changes is adequate, if not ideal, for use on a few articles, as would initially be the case. Having some active use of PC would encourage the developers to customize the tool as needed for enwikipedia, without worrying that their time is being wasted on something that won't actually be deployed. Chester Markel (talk) 21:32, 3 June 2011 (UTC)[reply]
Oppose having an oppose section - it's way too early. Seriously guys, this is the sort of thing that deserves some brainstorming and discussion. Even if it doesn't necessarily graduate into policy, it'd be good groundwork for future discussions of PC implementation. It seems wrong to kill off one of the more credible and considered contributions to debate I've seen on wiki for a long time. TheGrappler (talk) 22:04, 4 June 2011 (UTC)[reply]
That's all well and good, but this page seems to have been started with the erroneous assumption that using Pending changes at all would be something that the community would be receptive to. Obviously, that's not the case.
— V = IR (Talk • Contribs) 22:42, 4 June 2011 (UTC)[reply]
Ohms law puts it well. Now that we've finally got rid of ~~flagged revisions~~ pending changes, it's only to be expected that there will be several RFCs proposing new ways to bring them back. I'll be opposing them all.—S Marshall T/C 09:43, 5 June 2011 (UTC)[reply]
But it's still a legitimate discussion to have, and I don't see what closing it down early achieves. This is something where there is a divide in the community. I disagree that you can treat "the community would not be receptive" to a return of PC in some form as a given. There was consensus to run a trial, but with considerable opposition too. There was consensus to cease the trial, but with considerable opposition too. Some people liked the idea but disliked the footing the trial was run on (legitimately IMHO). Some people felt the software needed work first. Some people were opposed to PC full-stop. Some people thought it worked well and were disappointed the trial stopped. Discussion of bringing it back is inevitable in those circumstances. A productive and intelligent discussion about what's the best way that it might be brought back, seems the most sensible and appropriate prelude to testing community consensus anew. The proposal that gets presented to the community should be the most carefully thought-through one available. What is gained by shouting it down so soon, even if you have already determined your stance on the issue? Even if you think PC is a bad idea, is there nothing productive you can add to the debate, for instance by identifying what you perceive as key weaknesses and ways they might be mitigated? TheGrappler (talk) 18:26, 5 June 2011 (UTC)[reply]
One key weakness here is focusing on BLP, which has been pointed out below (especially with the "save the children!" attitudes that are exhibited by the most ardent supporters). Another is relying on PC itself, which is problematic for a combination of political and technological reasons. Put them together and this proposal falls flat on it's face. Currently, I'm yet to be convinced that any use of PC is worth the costs. I'm open to being convinced otherwise, but it's going to be a hard sell. PC makes actually contributing to Wikipedia a convoluted and complicated task, which will cause fewer contributions to be made. What's worse is the political ramnifications, because it's clear that there is a small minority of support for PC because it does make it more difficult to contribute. The technical issues are a real problem, but the political issues that this raises are what really sinks the proposal. You make enough noise and you might be able to get some things to change in the short term, but over the long term you're going to see things like... well, this.
— V = IR (Talk • Contribs) 00:13, 8 June 2011 (UTC)[reply]
Oppose lack of professionalism, hyperventilating anti-PC rhetoric, and crimes against humanity. Hans Adler 08:05, 6 June 2011 (UTC)[reply]

Reviewer bit concerns

My greatest concern about use of PC in the future are the criteria for granting and revoking the reviewer bit, which were very nebulous during the trial and caused no small amount of controversy.

My preference would be for there to be an objective set of criteria needed to gain reviewer status, and for this status to be granted automatically, a la autoconfirmed. Something like X mainspace edits and Y amount of time since the last block. These criteria would put up enough of a hurdle that sockpuppets could not easily create new accounts with reviewer permissions, while avoiding the problem of admins with widely differing standards for granting the bit.

I like that the proposal specifically excludes talk page discussions from valid reasons to remove the reviewer bit. It needs to be crystal clear that there is a difference between understanding the BLP guidelines and agreeing with them, that revoking the bit for "ungoodthink" without an actual mainspace violation of policy is completely unacceptable, and that fear of losing reviewer privileges should not chill good-faith attempts at BLP reform. TotientDragooned (talk) 14:57, 3 June 2011 (UTC)[reply]

Any proposal for PC in the future also needs to address how to handle users from the trial who already have the reviewer bit. Are they grandfathered in? Stripped of the bit by bot and asked to reapply? TotientDragooned (talk) 14:59, 3 June 2011 (UTC)[reply]

The problem with completely automatic granting of reviewer rights is that it will be given to people who don't understand, and can't correctly apply BLP. However, granting of the privilege at administrators' discretion shouldn't result in decisions for wikipolitical reasons, with a proper set of guidelines in place. This should be no more troublesome than any existing administrator-granted user right. For instance, if Scott were removing rollback because of "incorrect" opinions expressed at WT:VAND, or revoked the filemover permission because he disagreed with an editors comments at WT:NFCC, his actions would be quickly reversed, and he would be desysopped. The reason User:Scott MacDonald/Removal of reviewer rights from User:Wnt could happen is that there were no written standards for granting or removing reviewer rights, and no willingness on the part of other administrators to reverse the decision when the PC trial was already winding down. If we limited administrators to considering mainspace edits, and require a written explanation when reviewer bit is granted, declined, or removed, then there will be little scope for wikipolitics to enter the decision making process, since even "unilateral" decisions are always subject to being overturned by other admins. This is how we have achieved a reasonable, though not perfect, consistency in applying the CSD, even though they contain some subjective criteria. Scott wouldn't be allowed to start speedily deleting articles written by editors whose comments he disapproved of. Properly implemented, any recurrence of the Wnt situation would be just as untenable.

It would be prudent to have an automated process remove the existing reviewer rights assigned during the trial before starting any new uses of PC protection, since the rights would not have been granted in accordance with the new criteria, much as PC protection was removed from all articles, many if not most of which wouldn't qualify for it under the standards described here. Chester Markel (talk) 16:14, 3 June 2011 (UTC)[reply]

Actually, I think the automatically assigning of the right (assuming it can be revoked) and the existing reviewers are not a problem with the scope of this particular change. The proposal is to apply pending changes to a very narrow set of articles with long term BLP issues where semi-protection wasn't enough. As described above, the pages in question have people who are intentionally violating the BLP policy, and are even going so far as to establish autoconfirmed accounts to do so. If people who already have the reviewer right, or who are automatically assigned it, engage in that sort of conduct that can be quickly dealt with. But as this is not being advertised as a general measure to avoid inadvertent BLP violations, the understanding of those with the reviewer right of BLP policy is not really relevant. Now if PC was brought back with a broader scope I agree the question of reviewer permission would need to be dealt with, but I don't see it being an issue here beyond perhaps setting a threshold for automatic assignment. Monty ₈₄₅ 05:09, 5 June 2011 (UTC)[reply]

Assigning the reviewer right is not simply about believing that a user won't engage in malicious defamation. Reviewers are expected to be able to examine edits contributed by other users to PC protected pages for compliance with policy, and accept or reject them as appropriate. Some editors who have been contributing for years still don't understand what a reliable source is[4]. Obviously, anyone wishing to publish libel in an article could easily supply a reference meeting the standards applied by "Tokerdesigner". While it's possible that he might be more careful on BLPs than an article about methods of smoking pot, there's little reason to be confident of this. The supply of users lacking competence in the application of policy who would, or could receive the reviewer right under automatic assignment greatly exceeds the number of times we are willing to accept the public display of BLP violations on articles which have already had significant problems. Chester Markel (talk) 07:03, 5 June 2011 (UTC)[reply]

Dull

So a tool for the tiny handful of pages that fall between needing semi and full protection. Well utility would be non zero but it's darn close. Thats not really my concern.

It's a newish tool. Surely we can think of some weird and wonderful things to do with it. I don't know a more editable section of the main page or something.©Geni 18:15, 3 June 2011 (UTC)[reply]

The main page is currently comprised of an elaborate assembly of nested template transclusions. This is not a conventional usage, as the content itself is located on various templates. Since PC doesn't protect transcluded versions of a page, it would only be useful for the main page if the template assemblies were dismantled, and all content placed directly on the page. But given the level of vandal attacks there, the main page is probably one situation in which we should retain full protection.

As per my comment above, level two pending changes plus semi-protection for BLPs in which semi by itself isn't enough is not all that PC protection is useful for. It's merely the most straightforward, obvious, highly justified usage, and the best way to start a community policy for PC. Once this is implemented, a new discussion can be had on whether it would be appropriate to use level two PC instead of semi-protection on BLPs that receive relatively infrequent, but still highly objectionable vandalism or defamation, as a more thorough remedy for BLP problems which also, in some sense, allows a greater freedom in editing. Starting the rollout of PC on a small number of articles having the greatest need for it allows the community to determine later whether this is something that we want to be used more widely, or not. Chester Markel (talk) 19:16, 3 June 2011 (UTC)[reply]

I'm well aware of how the main page works.

I'm also well aware of what is being proposed. However it's dull. Worse still even assuming the best case senario it unlikely to be worth the effort needed to get a policy in place and them manage the reviewer permissions and the politics around those etc etc. And please give up the limited use allowing the community to determine anything. If that was the case it would have happened months ago.

But seriously it's the dullness that gets me. New tool and all we can think of is another form of article protection.©Geni 20:47, 3 June 2011 (UTC)[reply]

For more wikidrama... Not all that is interesting benefits the project, and not all that is productive is exciting. Chester Markel (talk) 21:45, 3 June 2011 (UTC)[reply]

It's a new software feature. We've done some hightly non standard things with them in the past The abuse of the uselang function for example.©Geni 22:43, 3 June 2011 (UTC)[reply]

The reason for not doing it, is that it would add an additional layer of complication. A rather elaborate layer of complication, with features many of us found confusing. This is Procedure Creep. We would do better to devote ourselves to improving the articles. I used to add material to one of the WPs that do use Pending changes, when it was straightforward enough for a beginner in the language. I no longer do, because I do not want to deal with it, especially in a language where I have only a little competence. I suggest the same will happen here. However, it is possible that I may be unusually stupid or timid as compared to people at enWP with primary languages other than English. DGG ( talk ) 19:27, 4 June 2011 (UTC)[reply]

We can best improve articles when most editors have access to them. As the popularity of Wikipedia for defamation grows, we can only expect that more BLPs will, or should be under full protection, if pending changes is not available. As far as usability by editors whose primary language is not English, only a subset of BLPs, not the entire project will be affected. By contrast, some Wikipedias in other languages, such as German, use pending changes everywhere. The bottom line is that while enwikipedia doesn't care whether or not editors are native speakers of English, we do require sufficient competence in the language to effectively participate in the project. If an editor can read the text of articles and the standard user interface, I doubt pending changes should impose many additional difficulties. Chester Markel (talk) 20:17, 4 June 2011 (UTC)[reply]

Ah, there's an interesting statement. Do you have any source or statistics to back up the idea that "the popularity of Wikipedia for defamation grows"?
— V = IR (Talk • Contribs) 20:24, 4 June 2011 (UTC)[reply]

When the Wikipedia biography controversy occurred, use of the project for defamation of living people was a rare and notable event. Now, we need an entire policy, OTRS, and a great deal of editors' time just to handle it. The rising interest in Wikipedia for online libel is concomitant with the increasing readership and visibility of the project[5]. Chester Markel (talk) 20:42, 4 June 2011 (UTC)[reply]

5 years ago I may have agreed with that (note that the linked article is from 2007). It doesn't seem to be the case so much these days, that there is increasing popularity of misusing Wikipedia, however. Of course, Wikipedia seems (to me) to have lost it's allure to the serious vandals in large part to the policies and processes that you mentioned above. I just don't see any reason to continue to put more resources into an issue that seems to be well in hand at this point, especially with something that will have such wide ranging side effects. Besides, I just couldn't see a recreation of the "Seigenthaler incident" happening these days. There's quite a bit of familiarity and comfort with Wikipedia these days, unlike in 2005 when Wikipedia was seen by many media outlets as... problematic, at best.
— V = IR (Talk • Contribs) 22:31, 4 June 2011 (UTC)[reply]

re:

The article must have a history of edits obviously not in compliance with WP:BLP. Genuine, tenable disputes between editors in good standing as to what BLP requires for particular articles do not form a sufficient basis for protection.

No. What should be required is that a reviewer or admin with the article watchlisted believe that multiple improper edits have been made on a sufficiently regular basis that PC would help. Articles under massive attack are not good candidates for PC.

It is reasonably believed that semi-protection is insufficient to control the disruptive editing.

No. Though this comes closer. In many cases, PC may be a better solution than semi for articles which get predictable, sometimes seasonal, or sometimes more frequent, attacks.

PC protection may be overlaid with semi-protection.

Why?

Pending changes protection will not be assigned outside of the main namespace.

While mainspace articles are most likely the ones to be affected, I could see template vandalism etc. arising - there is no need for this sort of artificial restriction. Collect (talk) 12:13, 6 June 2011 (UTC)[reply]

Too unclear

The page is a confusing mixture of half-thought-out policy and some steps needed before it could be a policy.

I agree with "the reviewer privilege would be removed from users who received it during the trial" - that should happen anyway; indeed, any proposal for any form of PC is going to run into that barrier. We need to get rid of the permission, for now - so that we can discuss a policy for granting it; but as to the conditions for granting, it depends on how PC will be used. Some folks think PC is a quick 'anti-vandal' tool, and thus 'reviewer' can be given to anyone who has a few hundred edits; others think the reviewer must check the added info in great detail - including fact-checking of the veracity of the source (Jimbo Wales seemed to be saying that) - in which case, the requirement for 'reviewer' would be very high.

"sufficient experience with Wikipedia" - what's that?
"Only mainspace edits [..] will be considered." - why? Users can (and often do) violate BLP policy on talks. Same re. "Revocation may occur when an editor has [..] substantially created mainspace BLP violations" - what is 'substantial'? Would one not be enough then? And would non-mainspace vio's be ignored?

Standard of review - what about edits that *remove* content?
"avoid harm to living people, and ensure the highest level of editorial access for articles" - why is it the highest? Surely full protection is higher?
"a limited number of BLPs" - I've heard such 'limits' before. This would need to be much clearer, about scope, before having a chance.

It's far too imprecise, so I can't support this proposal. Chzz ► 14:25, 6 June 2011 (UTC)[reply]

Tend to agree. I think PC level 2 is a generally far more useful thing than PC level 1, so I tend to favor figuring out a way to use it that we could actually gain consensus for, but this doesn't seem to be the right way to start. Muddling the standards for article together with the standards for reviewer is a bad start, and presupposing that only BLPs would benefit is another bad point. There are a lot of articles that could benefit from having more than semi-protection without having full protection.—Kww(talk) 23:33, 7 June 2011 (UTC)[reply]

Since PC has a controversial history, a limited usage of level 2 on those BLPs for which semi-protection is inadequate seems like the easiest, most obvious, least controversial initial policy for long-term usage. If this application can be implemented successfully, we can have another discussion about whether more broad applications should or shouldn't be allowed. Setting explicit standards for the assignment and removal of the reviewer permission is apparently necessary to avoid situations like User:Scott MacDonald/Removal of reviewer rights from User:Wnt. Chester Markel (talk) 17:44, 9 June 2011 (UTC)[reply]

Provided administrators apply the standards for PC protection in good faith, I see no reason to set a hard numerical limit on the number of articles. This would require some forward-looking statements about how many articles might have future BLP problems that would merit pending changes, for which there's little basis. While full protection obviously remedies any BLP violations that PC could handle, the former substantially limits editorial access. We aren't the "encyclopedia that any sysop can edit." Chester Markel (talk) 17:49, 9 June 2011 (UTC)[reply]

So, basically, PC trial redux.
— V = IR (Talk • Contribs) 18:00, 9 June 2011 (UTC)[reply]

If it's presented and formatted as a trial, I would only support if we had two bots:

One to scan through protection actions and send a nastygram to any admin that attempted to apply it for longer than the trial duration.
Another to automatically return all articles to semi-protection on the agreed-upon end date, and continue to scan for articles under PC2 after the trial and return them to semi-protection.

—Kww(talk) 18:26, 9 June 2011 (UTC)[reply]

I think that recent actions have effectively eliminated the credibility of any future "trial", since it won't really end as scheduled. If we adopt a policy for the application of PC protection to certain articles, the only plausible time frame is remaining in effect until expressly repudiated by the community. Chester Markel (talk) 21:07, 9 June 2011 (UTC)[reply]

This is why I've drawn the initial long-term PC policy as narrowly as possible. If there's no possibility of a further trial, then future application of PC protection should proceed slowly and carefully. Chester Markel (talk) 21:10, 9 June 2011 (UTC)[reply]

Was this proposal not already rejected?

As I understood it, this proposal was floated in the recent RFC on pending changes, but rejected by the community. Am I mistaken? To be clear, I do support this proposal in principle (although I did also oppose the removal of pending changes in the first place). The talk of "reviewers" and other bureaucracy seems a little pointless, at first reading. AGK [•] 17:02, 6 June 2011 (UTC)[reply]

The RFC was regarding discontinuation of the PC trial. It was expressly acknowledged that a long-term policy for the use of pending changes might legitimately be proposed and discussed. Chester Markel (talk) 17:34, 9 June 2011 (UTC)[reply]

Historicize with prejudice

IMO, this proposal reeks of bad-faith forum-shopping. This proposal, along with many others, was floated and utterly failed to gain consensus in the RFC(s) that just recently ended. This page should be immediately marked as historical, since the PC discussion is currently on hiatius. In conclusion: Kill it with fire.☻☻☻Sithman VIII !!☻☻☻ 06:45, 9 June 2011 (UTC)[reply]

The recent RFC was to end the PC trial, not to scuttle pending changes indefinitely. Please see the discussion of this issue above. Chester Markel (talk) 17:32, 9 June 2011 (UTC)[reply]

Comments

Some idea of the sample size for this would be nice, I'd expect it to be less than 500 and probably less than 100 or so. Also for transparency it could well be worth requiring it to be discussed on WP:RfPP or have an ORTS ticket. Other than that this seems reasonable enough. I definitely oppose the idea of any further trials. -- Eraserhead1 <talk> 22:56, 14 June 2011 (UTC)[reply]

Discussion

Moved from page content to talk page 21:05, 20 June 2011 (UTC)

Broad support; I think this is a great idea, which would provide another tool to help protect vulnerable articles. However, I have a couple of minor caveats:
1. When assessing an editor prior to granting the reviewer right, it might not be appropriate to limit scrutiny to article-space and BLP policy. Edits in other namespaces might also be cause for concern, as might violations of other policies. For instance: Good sourcing is vital to controversial BLPs, so I'd be wary of recruiting reviewers who have a history of ignoring WP:V; there's an analogous argument for personal attacks too.
2. Personally, I think that anybody with a recent block asking for the permission should be treated with great skepticism (thought I doubt any block is necessarily justification for removal). bobrayner (talk) 10:54, 8 June 2011 (UTC)[reply]
See talk page for extended comments With regard to reviewers, the layer of bureaucracy is unneeded if reviewers were simply told:
- Other than for clear vandalism or insertion of unsourced contentious claims, reviewers should not act on any edits for articles in which they have made substantive edits. No new "rules" needed, only a small dose of common sense. Collect (talk) 12:55, 8 June 2011 (UTC)[reply]
  - Personally, I agree with you, but others have mentioned a lack of clear rules in their previous opposition to PC; so it might be helpful to put something in writing even if it "seems obvious". bobrayner (talk) 17:31, 8 June 2011 (UTC)[reply]
Comment Not many comments on this, which is unusual, as PC tends to attract a lot of debate. I realise it's listed at WP:CENT, but should this proposal be advertised more widely? bobrayner (talk) 09:48, 13 June 2011 (UTC)[reply]
A link to this proposal could be added to MediaWiki:Sitenotice. That will certainly get plenty of attention, though perhaps not the type we want... Ultimately, interest may be limited simply because this proposal affects a small number of articles. Chester Markel (talk) 17:08, 13 June 2011 (UTC)[reply]
Support - I was recently involved in rewriting a WP biography that had persistent vandalism issues and was the subject of a lawsuit threat by the subject. Pending changes was one part of the solution there. I absolutely HATE the idea of using pending changes across the board and was an opponent of the system when it first popped up, but this is a very appropriate and proper use of the tool. Big problems are out there that need solutions; Pending Changes is part of the answer. Carrite (talk) 15:19, 14 June 2011 (UTC) Last edit: Carrite (talk) 15:20, 14 June 2011 (UTC)[reply]
Strong oppose There is firm consensus against using pending changes for anything at all. The consensus is based upon the dual arguments that the software is a/ inadequate and b/the procedure confusing. To have this procedure for a very small number of articles would be especially confusing: 99.99% of Wikipedia article editing would work one way, and 1 in 10,000 would work another. The smaller the number of article,s the less we need this. It will in a sense serve the purpose of keeping new editors firmly away from an incomprehensible process, but security by difficulty is the worst way of implementing any necessary precautions. For an article under this much pressure, full protection works well, not to freeze the article, but to have the edits suggested on the talk p, and a neutral admin to decide whether to accept them. We already have the remedy under existing mechanisms. It gives the same benefits without introducing additional procedures. (I will admit that my increasing opposition to any form of proposed changes is influenced by my continuing an ability to work at deWP; I used to contribute a bit within by linguistic limits, but I cannot understand their procedure well enough to follow it. I suspect the same problem will be true of the non-native speakers here. I'm not sure they care much about non-native speakers, but we do: we wish to encourage them, because of the wider geographical and cultural representation they give us. , DGG ( talk ) 08:20, 16 June 2011 (UTC)[reply]
- No there isn't. There was a reasonably firm consensus from removing Pending changes from existing articles, but that didn't mean those people necessarily thought it wasn't worth using at all. -- Eraserhead1 <talk> 20:26, 20 June 2011 (UTC)[reply]
A vote? already? Oh well. Support. I recently came across a BLP were semi-protection wasn't enough to stop vandalism: Justin Beiber. The article had to be fully protected just to curb vandalism. But full-protection isn't a long-term solution; PC2 is. Rami R 12:08, 16 June 2011 (UTC)[reply]
Support. This is a process that can limit the visibility of vandalism and contain the damages it causes — as the vandalised version of a biography will not be picked up by Google —, while, at the same time, imposing a minimal inconvenience, which is by far outweighed by the advantages, and upholding the founding principle that everyone can edit much better than semi-protection or full protection do. Salvio ^{Let's talk about it!} 13:38, 16 June 2011 (UTC)[reply]
Very limited Support for another trial in the future. A subset of BLPs are one of the few sets of articles where I think pending changes may work, mostly for rarely watched articles. But I question when semi-protection would fail and blocking the individual users wouldn't be a better solution. I also fear a slippery slope of adopting PC and I still think that the entire pending changes process, if used, needs more clarification on what are acceptable edits. Appropriate statistics should also be agreed upon before a trial for use in comparing the effectiveness of PC. —Ost (talk) 19:48, 16 June 2011 (UTC)[reply]
Changing the venue from the talk page, eh? Oppose, for the reasons that DGG outlined above, and the reasons that myself and others have discussed at length on the talk page. This is a dead horse of an issue for at least the next 6 months, or so.
— V = IR (Talk • Contribs) 19:56, 16 June 2011 (UTC)[reply]
Strong support - This is a perfect use of PC-2, especially considering that the alternative is locking everybody out of the article indefinitely. (And Rami R, Justin Beiber has always been my example of an article that would otherwise need indefinite full protection. Funny that you would use it too!) Reaper Eternal (talk) 20:33, 17 June 2011 (UTC)[reply]
Support seems a reasonable use of pending changes - its clearly better than full protection. -- Eraserhead1 <talk> 20:26, 20 June 2011 (UTC)[reply]
Support I'm particularly concerned about very low-edit-frequency BLPs, an area I spend most of my Wikipedia time trying to assist in. This is a reasonable proposal--and one which is less onerous than an alternate proposal I'm putting together to address my concerns using indefinite full protection on such articles. What both proposals share is the need for eyeballs in the darkest corners. My support is conditional on starting the reviewer list over from scratch, as well as a proviso that this is not an appropriate technology for frequently-edited articles. --joe decker ^{talk to me} 22:20, 20 June 2011 (UTC)[reply]
Humm, given the discussion above I'd figured this was DoA. That might explain the relative silence on the issue. Might be wise to send notifications to participants in the last PC discussion. In any case, Oppose for now. I've lost a lot of faith in PC due to BLP-warriors abusing consensus. As such, I'd like to see more of a bright-line about when to add and remove PC2 rather than the very low and vague bar in this proposal. I'd want to see it on only a handful of articles that would otherwise qualify for full protection but where PC2 would be sufficient. That is, I'd only support this if on the whole it meant more people could edit our articles, not fewer. Hobit (talk) 08:05, 22 June 2011 (UTC)[reply]

I put a notice about this vote at Village pump (policy) to get a wider group of folks who are interested in such things. Hobit (talk) 08:13, 22 June 2011 (UTC)[reply]

Oppose because of the out of the way place this is being discussed and because so recently pending changes was turned off by consensus. We need some stability here. Graeme Bartlett (talk) 11:58, 22 June 2011 (UTC)[reply]

Proposer blocked

Does the fact that the person who proposed this has been blocked as a sock of an intentionally disruptive user have any effect on this proposal continuing? (I'd like to think so, but I'm a partisan here.)
— V = IR (Talk • Contribs) 18:37, 22 June 2011 (UTC)[reply]

IMO, this looks way too premature and should have been trashed at first sight. The only real "official" discussion we've ever had about PC ended just a month ago on the Wikipedia:Pending_changes/Request_for_Comment_February_2011 page. :| TelCo NaSp Ve :| 21:16, 23 June 2011 (UTC)[reply]

Since this appears to have completely fallen off of everyone's radar, I've gone ahead and marked it historical.
— V = IR (Talk • Contribs) 05:13, 27 June 2011 (UTC)[reply]