First off, big thanks to FT2 for getting the ball rolling here! This is a lot of great content to work with! -- RobLa ( talk) 03:59, 5 June 2010 (UTC) reply

FT2, I came away with a better first impression of this older version of the Help page than I did the latest version as of this writing. I think the earlier version covered the topic at about the right level of detail, whereas the latest version may be getting into the weeds. All of the information here needs to go somewhere, it's just probably time to create detail articles. I'll take a stab at that now. -- RobLa ( talk) 03:59, 5 June 2010 (UTC) reply

Generally good stuff. A couple of points that didn't work and I'll have a re-stab at:

• Introduction of minor jargon not needed for a simple help page. "Accepted version" (we seem to have consensus elsewhere that "checked/checking" is a good word), "available to determined readers" (we're speaking to a reader, "determined" is a word that I'd associate with a vandal!). They may be accurate but they're not helpful. No need for extra terms. Minor copyediting. ( copyedits)
• Agree with summarizing process, unsure if that's the best summary. Especially don't like diving into tech explanation (inappropriate focus) and the term "trusted group". Having a go. ( copyedits)

  The tech details are really needed because the "help" and "?" links in a lot of parts of the user interface link to this page. -- RobLa ( talk) 01:03, 7 June 2010 (UTC) reply

• "Some readers might notice a 'Pending changes' tab" - which ones?
• Screenshot of tab bar rather than textual description is needed.
• The extra section header "What does it mean to me?" doesn't add anything. The subheadings are self-explanatory and fit neatly into the flow, the extra header just disrupts it. (After keeping returning to it, removed this header. It just doesn't add anything, and detracts instead)

FT2 ^{( Talk |  email)} 08:02, 5 June 2010 (UTC) reply

Done a bit more copyediting. Back to you. ( All edits including above) FT2 ^{( Talk |  email)} 09:17, 5 June 2010 (UTC) reply

Hrm..."checked" isn't the word we've been using lately. See the terminology guide we've been working from for the past month. The user interface has already been changed to use this terminology, which you can see at http://flaggedrevs.labs.wikimedia.org . So, can we change these back to "accepted"? -- RobLa ( talk) 16:52, 5 June 2010 (UTC) reply

By the way, I should probably explain why we went from "checked" to "accepted". We found we needed to describe three distinct sets of activities: "reviewing", "accepting" and "rejecting". The exact names for each of those wasn't as important as making sure we had terms that were reasonably unambiguous. Under the old usage, "checking" was used interchangeably to mean reviewing in some contexts, and accepting in others. For example, to the layperson, a "checked article" may conceivably be "checked and rejected" just as easily as it could be "checked and approved", since "checking" is much more synonymous with "reviewing" than it is with "accepting". However, an "accepted article" is much less ambiguous.

So, there's some contexts (e.g. "...after an article has been checked") where not using the phrase "accepted" might be ok for a simple explanation (though "...after an article has been reviewed" would be preferable in my mind...not something I think is so important to have a huge argument over, but a stylistic difference worth noting). However, it seems like we should try to be more consistent with respect to using the phrase "latest accepted" over "latest checked". Make sense? -- RobLa ( talk) 20:50, 5 June 2010 (UTC) reply

About the section "...but changes will be visible only to other editors until checked for obvious vandalism and clear errors. The public (99.5% of whom are not editors)..." [emphasis mine]

Firstly, is this technically correct? I mean, there must be users who would consider themselves "editors" but never register - will they see the changes? Unfortunately I'm pretty sure the answer is "no"; secondly, it's not very inclusive-feeling. We want everyone to edit. Maybe "registered users"? Something like that?

Of course, the only reason I'm able to pick up on such trivial points is because it all makes sense to me; at last, a good help page about what we are now calling "pending changes"; so sorry to nitpick :) - Jarry1250  ^{Humorous? Discuss.} 13:12, 5 June 2010 (UTC) reply

I've heard the 99.5% statistic quoted a lot, in WMF sources.Worth checking of course. Maybe clarification of logged out IP readers who click to edit - what will happen. Do they suddenly switch from viewing a checked version to editing an unchecked version (that they hadnt seen). Should we give a heads up on that. FT2 ^{( Talk |  email)} 13:24, 5 June 2010 (UTC) reply

Based on the report card from February 2010 (most recent that included editor stats), there were 375 million unique visitors and 101,730 active editors. I wasn't immediately able to dig up the stat for all editors, but I can poke around if necessary. By those numbers, active editors make up a very tiny portion (1 out of 3686 visitors, or less than 0.03%), and what I've seen elsewhere is that the next ring of participation outward tends to be about 10x greater than the inner ring, so 0.5% seems like a plausible number. -- RobLa ( talk) 22:30, 5 June 2010 (UTC) reply

The article length is creeping upward again (in part due to the change I just made, but that's not the sole culprit). It seems like the 7k versions have about the right level of detail, with 9k being a lot more information than what is probably needed. The subpages, FAQ, etc, can be much longer, but this page should really be short and sweet.

I'm going to take a crack at trimming this back again. -- RobLa ( talk) 01:17, 7 June 2010 (UTC) reply

The page contradicts itself

1. "The only difference is that the version shown publicly to non-editors (including logged-out editors) will sometimes be a slightly older one." Past discussions told me this is the case, and registered editors (non-reviewers) can see and edit the current version, but I was wrong:
2. "All other users will still see the latest accepted revision.".

So, when an editor (not reviewer) opens an article, does he see the reviewed or the current article? The "reviewed" option leads to many more questions... East of Borschov ( talk) 14:20, 7 June 2010 (UTC) reply

All users get two tabs ("Read" and "Pending changes"). The "Read" tab contains the latest accepted rev, and the "Pending changes" tab has the very latest rev. Anonymous users have the "read" tab chosen by default. Logged-in users have the "pending changes" tab chosen by default. You are correct that the "All other users..." sentence wasn't helping. Rather than trying to rephrase it, I just nuked it because I think we're still at risk of this page being tl;dr. -- RobLa ( talk) 07:07, 8 June 2010 (UTC) reply

I came here because of the draft announcement in The Signpost. I started a major copy-edit of this help page, but stopped (and instead saved it in a Word document) when I became thoroughly confused. I fear this will generate negative reactions unless fixed.

My head span concerning the following distinctions, which seem unclear and to cross over each other:

• "Readers" versus "editors" (they are not mutually exclusive).
• Logged-in versus non-logged-in editors.
• Editors who have set their prefs to "Always show latest version" versus those who haven't (are WPians advised not to set this pref, for better maintenance? Something needs to be said either way).

I have a suspicion that there are just two categories: those who are logged in and have not set their prefs to "Always show the latest version", and everyone else. Is this the case?

I saw at least two major repetitions of statements: this clagged up any understanding I was moving towards. The headings are confusing. Some information could be relocated to be more logical. I think the place for justifying the new system is in the lead; it doesn't need to be harped on in the sections

If someone can explain in as few words as possible what the differences in experience will be for the categories I've bulleted above, I could assist with the editing. Possibly, some of those categories could be conflated—if so, that would be soooo much easier for us to understand. Tony (talk) 15:17, 8 June 2010 (UTC) reply

Hi Tony1, thanks for pitching in! Here's my attempt at answers:

• Readers vs. editors: there is no distinction from a software perspective. I'm not sure if Wikipedia users will find that to be a helpful distinction here.
• Logged-in vs. non-logged-in: this is important:
  • Non-logged-in users see the "Read" tab by default (the "latest accepted version"). Logged-in users see the "Pending changes" tab by default.
  • Non-logged-in users by definition are not autoconfirmed, and therefore must have all of their edits reviewed by a reviewer. The story for logged-in users is more nuanced. See the table at Wikipedia:Pending changes/Trial to get a better idea there.
• "Always show latest accepted version": It's generally healthier for the project if logged-in users not futz with the preference for "Read" vs "Pending Changes" tab. We want as many eyeballs as we can get on pending changes.

I hope this helps! -- RobLa ( talk) 16:13, 8 June 2010 (UTC) reply

Thanks; I'll have a go in a few hours' time, but will need scrutiny (I'm quite unfamiliar with the system, which is good in one way, but I could make bad mistakes). When The Signpost edition comes out (it's very late already), there'll be a cascade of visits here. If people care about the prospects of success for the trial, a lean, clear explanation is the starting point. Tony (talk) 09:58, 9 June 2010 (UTC) reply

OK, this is what I think it means, but I'm too nervous to post it yet. I've slashed it down to the bare essentials, removing repetition and anything that seemed unnecessary for dummies like me to understand it. I've copy-edited all but this middle section, pending assistance here:

________________

===What it means for all users===

When you select "edit this page" on a page with "pending changes", you will be able to edit the latest pending revision, which may be slightly different from the page you were viewing. When you save your changes, your new version will become the latest pending revision.

=== If you are not logged in ===

Above the displayed article, you will see by default a "Read" tab (the "latest accepted version"). By definition, you are not autoconfirmed, and therefore must have all of your edits reviewed by a reviewer before the public can see them

If you always want to see the very latest (possibly unchecked) version of all articles, please log in and create an account (it's quick and free, and you don't even have to give an email address). You can then set your preference to always show the very latest version.

=== If you are logged in ===

If you leave your preferences at the default "Read vs pending changes" tab. The article will be displayed as for non-logged-in readers; i.e., you will see only the most recent accepted change, but can click on the "Pending revision" tab to see the changes that are awaiting review. This is generally the most helpful preference for the trial, since the more Wikipedians who can see the pending changes and provide feedback to the trial, the better.

If you change your preference to "Always show latest accepted version". You will see the latest revision by default, whether or not it has been reviewed and accepted.

Could someone please point me in the direction of the "Read" tab? -- THE FOUNDERS INTENT ^PRAISE 18:18, 23 July 2010 (UTC) reply

_____________

Is it mangled? Can people add/change what is above if necessary, please? Also, in the "How does the "pending changes" review process work?" section, I don't quite understand this sentence at the bottom: "All edits will be visible immediately to users who have logged in, and the latest pending revision is what is loaded when any user selects the "Edit" tab." I also don't understand whether this statement will apply to all readers in all categories, or logged-on non-reviewers, or only reviewers: "iewing such an article, you'll see an extra tab: [SCREENSHOT OF RELEVANT TABS]. This means that some of the very latest edits are not shown because they are not yet checked for vandalism. The latest changes can be viewed by clicking the "pending changes" tab. Usually the displayed article will be very similar - the latest changes may be quite minor matters such as punctuation." Tony (talk) 14:55, 9 June 2010 (UTC) reply

Hi Tony1, I say go ahead and make the change...someone can revert and start a discussion if we don't like it, and I'm definitely not in love with what we've got right now. I'd like to see it in context before rendering a final verdict, but generally, cutting this down is something I'm very supportive of, and I hope others are too. -- RobLa ( talk) 22:09, 9 June 2010 (UTC) reply

The main issue here is that this page will not just be used by the cognoscenti (experienced editors). It will be used by an unusually wide range of people in the early days - namely anyone (editor, logged out, non-editor, newcomer, media writer/blogger/opinion shaper) who sees a pending changes tag, button, or reference and looks it up. Most of those are people to whom a technically exact summary with a sprinkling of terms like "logged in" or "autoconfirmed" will be just so much jibberish and incomprehensible. To the wider world (ie non-dedicated editors) the key comprehensible division is "editor" or "reader", not "logged in/logged out". 1/ If you're one of the 99.5% who just read Wikipedia, you're a reader. 2/ If you write articles, you're an editor. 3/ If you mostly read (or are logged out) but occasionally edit this is what you might notice. That's how the wider world will surely best comprehend it. Write for the audience. Right now the main audience (and the most important one) is not dedicated wiki editors. It's everyone else who might need to know. FT2 ^{( Talk |  email)} 22:45, 20 June 2010 (UTC) reply

FT2, I think you make a good point about avoiding the term "autoconfirmed". I don't think "logged in" is necessarily too jargony, though, and there appears to be some very real confusion that's been created by trying to separate these by "editor" and "reader". Also, the article length has crept back up into tl;dr territory. I'm going to continue to implore that everyone working on this article be really be mindful of just how intimidating the article length, the top template, and the general feel of this article would be to a newcomer. -- RobLa ( talk) 23:03, 20 June 2010 (UTC) reply

Other than my relatively minor misgivings about content length (see above), this page is in fair shape. However, Wikipedia:Pending changes is in a lot rougher shape, and there's not any linkage yet between Help:Pending changes and Wikipedia:Pending changes. I'm assuming that the right split would be that the Help page should talk about how the feature works at a more mechanical level, and the Wikipedia: page should talk about the policy, etc. Or, another way to deal with it is to just have one redirect to the other and call it good. Thoughts? -- RobLa ( talk) 22:24, 9 June 2010 (UTC) reply

I think the former option is a better idea. Something similar to what we currently have with Help:Redirect and Wikipedia:Redirect. Also, Wikipedia:Protection policy may need to be updated to inform about this new option as being an alternative to page protection. -- œ ^™ 04:43, 10 June 2010 (UTC) reply

Something that is not clear to me is how the revision process will work if subsequent pending changes conflict with eachother. An example:

• Imagine three 3 pending changes dated (Monday, Tuesday, Wednesday)
• Wednesday is currently accepted. It replaced Monday which was previously accepted. Tuesday has not been accepted.

Questions/observations:

• Can Tuesday still be accepted? If so, since its edits were based on Monday's version, how will conflicts between Tuesday and Wednesday be resolved?
• If there is an edit conflict, then one of two things must happen: either the reviewer is alerted to the conflict and asked to correct it manually; or, the Tuesday version replaces Wednesday wholesale, discarding any intermediate edits between Monday and Wednesday.
• Terminologically, if an intermediate pending change (i.e. Tuesday) has not been accepted, does that mean it has been "rejected"? Or will it be forever pending? Does a pending change ever disappear, and if so, after how long/how many changes posted afterwards are accepted?
• What would happen if a pending change from March (not accepted) gets accepted in August? What happens to all of the intermediate edits?

69.142.154.10 ( talk) 09:04, 10 June 2010 (UTC) reply

The key to understanding this is to understand how the page revision history works. The page revision history is just a linear series of snapshots. Without pending changes installed, the very latest revision is the one that people will see. With pending changes installed, the latest accepted revision will be shown. However, if someone goes to edit the page the very latest version is what they are editing.

So, to answer your question, let's start on Wednesday morning, prior to the Wednesday revision being accepted. "Monday" is the latest accepted revision at this point, but "Tuesday" is pending. When the editor of the Wednesday revision clicks "edit", they are editing the Tuesday revision of the page. The Wednesday editor can choose to revert the Tuesday changes, or can keep them and add to them, but regardless, the Tuesday revision is implicitly dealt with as part of the Wednesday revision. If the Tuesday edits were good, then presumably, the Wednesday editor didn't delete them, and the Wednesday revision is the sum of the Tuesday and Wednesday changes. If the Tuesday edits were bad, then hopefully the Wednesday editor reverted them. Either way, when someone reviews the difference between the Monday and Wednesday revision, they'll be looking in part at how the Wednesday editor dealt with the Tuesday edits. Make sense? -- RobLa ( talk) 15:34, 10 June 2010 (UTC) reply

Thanks RobLa, that goes a long way. I think the reason for my confusion is that the workflow diagrams present pending changes as discrete events--which I assumed were discrete versions--rather than cumulative edits. It also helps that prior pending changes are opt-out, since they are incorporated into the latest pending version. (I recognize that the accepted versions are opt-in, since they must be, accepted).

So,pending changes as a whole is really nothing different. It's just Wikipedia's regular editing process--the only exception is that whereas today the most current version is the displayed version, with pending changes, the last accepted version is the displayed version, which may or may not be the most current. In a sense, Wikipedia develops a facade which "scrapes" accepted content from the underlying edit progression. Also, there may be pending changes in the plural, but there is just one pending version that an editor is initially presented with. If they want to use another pending version as a base for their edits, they have to manually undo them, or copy/paste from the pending version history. Did I get that right?

One question, to see if this fits together. Are edit wars on protected pages going to be more complicated, since they'll be about both the latest pending version as well as the latest accepted version? Or will warring among pending changes be tempered, because gotcha-last changes won't matter until one is accepted. I mean, will people care whether or not their version is the latest pending or just the latest accepted? In short, can there be a war in pending world (obviously there can be on in accepted world). Thanks for the answers. 69.142.154.10 ( talk) 15:27, 11 June 2010 (UTC) reply

I think this may be a good question for the German Wikipedia community or one of the other wikis using FlaggedRevs]. I'm not sure what the exact social dynamic is going to be. I hope there won't be edit wars in pending space, but there is nothing in the software to prevent an edit war breaking out over the latest pending revision. There's only a safeguard to keep that article from being the default for most readers. -- RobLa ( talk) 22:52, 12 June 2010 (UTC) reply

1. The consensus was that this be applied only to BLPs. I suppose we could-interpret it to included articles that were similar to BLPs in their effect upon individuals, but I do not think there was ever agreement for even that--there certainly was not for the application to any other types of articles. This was not ruled out for a future trial, but it was agreed this would require a separate proposal.
2. The consensus was that the initial trial was that autoconfirmed=reviewer. To require anything higher was not accepted, at least not to start with. An additional level was not ruled out for a future proposal if this was successful & an additional or higher level seemed needed. Starting with a higher level, and especially starting with two levels, was rejected as being possibly unnecessary, but certainly more confusing.
3. The consensus was that there be only one level of pending change, one that would require editing or confirmation by a reviewer=autoconfirmed, not another level that would require an administrator. An additional layer was not ruled out for a future proposal if this was successful. snd an additional layer seemed to be needed. Starting with more than one level was rejected as being possibly unnecessary, but certainly more confusing.

I know at least that I !voted to approve the trial limited to BLP, as did many others. I do not think it would have been approved otherwise, especially since the key rationale for using this at all was the effect of vandalism on BLPs. I see the details presented here as an attempt to change the final wording to something that was rejected by the community. DGG ( talk ) 16:45, 11 June 2010 (UTC) reply

Where was the reviewer=autoconfirmed part brought to consensus? My memory was that there was much debate about the specific requirements that would be required for the reviewer flag (including a whole other poll about it) and the majority of consensus was for something much stronger then just autoconfirmed. I do agree that the BLP was what was originally discussed (though I personally think it should be more). James ( T C) 16:55, 11 June 2010 (UTC) reply

This is the version of the proposal that was considered by the poll, and it had two distinct elements to it, flagged protection and patrolled revisions. This "pending changes" version only implements the flagged protection part. Whilst the patrolled revisions part was intended to apply largely to BLPs this wasn't the case for the flagged protection part, and the flagged protection part did include a level of protection where edits needed to be checked by admins. Hut 8.5 18:50, 11 June 2010 (UTC) reply

We dropped the third level for adding too much complication, but they were all present in the proposal which also made clear it could only be used under the same conditions as classic protection. There has been prior proposals to apply flagged revisions on all BLPs but they didn't reach consensus. Cenarium ( talk) 22:07, 11 June 2010 (UTC) reply

" If the user editing the article is a member of the "reviewer" group ( Become a reviewer!), then the new revision is immediately visible to everyone who visits that page."

Erm, I thought that if Level 1 pending changes were in place, edits by autoconfirmed users would appear if there were no pending edits. Has this changed? Risker ( talk) 04:57, 14 June 2010 (UTC) reply

You may wish to refer to the new category Category:Lists of popular pages by WikiProject.— Wavelength ( talk) 21:54, 15 June 2010 (UTC) reply

I am trying to "unaccept" the only current pending change, to Raphael, & it has shown as "unaccepted", but is still showing in the article. Johnbod ( talk) 11:47, 17 June 2010 (UTC) reply

In fact now it has come off the pending changes list, and is showing as accepted, but does not appear in the pending changes log. Johnbod ( talk) 11:52, 17 June 2010 (UTC) reply

Apparently if you unaccept a change, it returns to the pending changes list where the next editor may accept it again. It looks to me like that's what happened here. It seems counter-intuitive to me. Shouldn't unaccepting a change essentially be the same as undoing it? Tim Pierce ( talk) 14:14, 17 June 2010 (UTC) reply

I see now that it's an intentional feature: Wikipedia_talk:Reviewing#Proposal_to_change_.22unaccept.22_to_.22decline.22_in_the_Reviewer. It's definitely counter-intuitive from my point of view. I'll have to think about what to suggest instead. If it is possible to use three-state logic for a pending change (accepted, rejected, and unreviewed), that would seem like the right solution to me. Tim Pierce ( talk) 14:25, 17 June 2010 (UTC) reply

I moved the following from the page; it's probably more useful as a discussion point for the feature than information about the feature.

This is in no way any disrespect to the idea, just that it might be more useful here instead, as the author's edit summary itself suggests. Also keeps the main page shorter. FT2 ^{( Talk |  email)} 16:56, 17 June 2010 (UTC) reply

Pending changes is an evolving concept which will improve through the feedback of users and data collected throughout the trial. The system is designed to maximize the openness of Wikipedia while minimizing the effect of unconstructive edits.

Pros

• All users can edit any article
• New users not discouraged by off-putting locked pages
• Reliable method of keeping vandalism off public pages
• Likely scalable to BLP and Current Events pages
• Reduced gratification for vandalism; clear presence of oversight

Cons

• Increased patrolling load on heavily trafficked articles
• Lowered signal/noise ratio for edits on heavily trafficked articles
• Constructive edits by non-autoconfirmed users possibly lost among vandalism
• Possibly unsuited to dealing with NPOV/content disputes
• Loss of instant gratification by seeing edits immediately
• New layer of temporal and procedural complexity
• New hierarchy of levels, reviewers, and terminology
• Duplication of schemes (Protection, Pending Changes) while trial is ongoing

• add to this list

69.142.154.10 ( talk) 19:03, 17 June 2010 (UTC) reply

Eh, so far, to me this thing seems like a major waste of time. There's one semi-protected article I'm watching that was put into trial. The only effect this had on the article is to produce about ten anon edits that vandalized the page and then had to be reverted, 10 times (of course), by good editors whose time is better spent elsewhere. That's what this whole thing is looking like to me, overall - it just allows vandals to have their edits recorded in the history log, while taking away time from good editors who could be out there expanding articles or reverting vandalism from pages where it is actually visible. I don't think any upsides outweigh that simple fact. All Hallow's Wraith ( talk) 04:30, 19 June 2010 (UTC) reply

I can give a different perspective, from a long-time, i.p. only article reader and copy-editor. It is very refreshing to go to a commonly viewed page like Bible or Ronald Reagan and find that I am not blocked out of editing them. I might be a weird minority, since most people who have been around Wikipedia as long as me are probably already qualified as reviewers, but nonetheless--from the perspective of a reader who takes Wikipedia's stance about "anyone can edit" seriously--Pending Changes removes a glaring exception to that idea. There really is a big difference between not being able to even propose an edit, on the one hand, and anyone being able to while having to wait for approval, on the other. Do I think it will change the quality of those controversial articles? Probably not much, but IMO it's a small big deal. 69.142.154.10 ( talk) 04:28, 20 June 2010 (UTC) reply

It may be fixing a small big deal, but it creates a surplus of time-consuming activity that will detract from quality in other articles (and neither Bible nor Ronald Reagan have thus far received any unreverted edits from non-autoconfirmed users). Pending changes should definitely never, ever be used anywhere other than on semi-protected articles. Putting it in everywhere would be brutal. As for yourself, if I may say, you ought to register an account. Make yourself known! All Hallow's Wraith ( talk) 06:34, 20 June 2010 (UTC) reply

Sorry, but I do find it disturbing to see one of the tens of thousands of anonymous editors who have made positive contributions to this project told that their opinion is without value because they don't have an account. Almost *all* "named accounts" started off editing as IPs, and I know of many, many editors who have named accounts but still regularly edit as IPs. (I've done it myself when doing straight editing on insecure connections.) Every study done has shown that IP editors on the whole make a large percentage of valuable and encyclopedic edits, and they tend not to spend very much time bothering with the "behind the scenes" things, they just invest time in improving our content. Please, 69.142, know that you are welcome here whether with a made-up name, or your IP address, and I for one thank you for taking the time to give us your perspective. Risker ( talk) 06:49, 20 June 2010 (UTC) reply

I never said 69.142's opinion is without value just because they don't have an account - never even suggested it. But I did disagree with their opinion on "pending changes", based on what I have seen so far. And I fully believe that if "Pending changes" (which is really just "Flagged revisions" spelled backwards) was ever applied to all articles, or even all BLPs, it would have a cataclysmic effect - in fact, it would prevent anon I.P.s from making immediate changes to articles all across the board. All Hallow's Wraith ( talk) 07:20, 20 June 2010 (UTC) reply

I am also an "IP editor", i.e. I often make minor improvements, and while I did register a few years ago, I never bothered to log in, and thus abandoned the account. I just "use" Wikipedia and whenever I can improve, I do. But I never visit WP with the purpose of contributing.

Anyway, I wanted to mention that in my view the flagged revisions addon has helped the German WP a lot. Maybe that's because its traffic is a lot lower and vandalism can thus stay undetected a lot longer than here on EN, but I'm pretty hopeful that it'll be a positive change over here too. 87.78.2.191 ( talk) 13:58, 15 July 2010 (UTC) reply

I'm not against this project, and I know this would cut down on a lot of vandalism, albeit with more of a reviewing workload, but doesn't this sort of go against Wikipedia's notion of being an encyclopedia that anyone can edit? If people without accounts need to have their stuff reviewed... -- The Editor1 ( talk) 00:43, 4 November 2010 (UTC) reply

PC Protection. What an unfortunate (but appropriate) name. Protecting Political Correctness. -- Britcom 18:10, 23 July 2010 (UTC) reply

This defacement of a featured article needs to be sorted. The editor who originally added TS to the pending articles already added a template at the top; now the article is being defaced with this template. SandyGeorgia ( Talk) 02:22, 18 June 2010 (UTC) reply

• What makes this article different from the others? East of Borschov ( talk) 02:30, 18 June 2010 (UTC) reply
  • It's a stable, featured article that required semi-protection because of coprolalia-related vandalism-- the nature of the content. Further, if those gross glaring big templates are being added to any featured articles, I generally object-- there has to be a better way to implement this. SandyGeorgia ( Talk) 02:35, 18 June 2010 (UTC) reply
    • This sets a precedent: some objections are heeded. East of Borschov ( talk) 02:55, 18 June 2010 (UTC) reply
      • I suspect there are other featured articles in a similar position. Tourette syndrome is not only vandalized per coprolalia-- it was also a target of the Tourette's guy editors (there were three AFDs on him, he had a huge following-- I'd have to look them up). I doubt that it's the only featured article that is such a vandal target-- there must be others ( Michael Jackson?) This is an opportunity to solve a problem, as Risker points out-- the template is just plain ugly, on any article, and not an optimal way to announce this trial. (I have a dual interest here, as FAC delegate and the main editor of the TS article.) SandyGeorgia ( Talk) 02:59, 18 June 2010 (UTC) reply

• Well, featured articles that have templates at the top of them have always been featured articles that need to go to featured article review. Templates can have an adverse effect on the visual effect of an article. Featured articles that are semi-protected have, by long tradition, not included the semi-protection template; and only rarely have fully-protected FAs had the full protection template.

  I actually see this as an opportunity to consider other ways of informing potential editors of the trial, and the fact that their edits will be subject to review. For example, the edit notice currently in use is minuscule and easy to ignore. Why isn't it a bright, eye-catching colour like the edit notices on a semi-protected page? Risker ( talk) 02:39, 18 June 2010 (UTC) reply

For Sandy (The editor who originally added TS to the pending articles already added a template at the top) I didn't nominate this for Pending changes, I just added a notice. Tbhotch ^{Talk C.} 02:37, 18 June 2010 (UTC) reply

I'm aware of that (one editor added it to the pending list, another implemented that without the template-- declining to add the template-- and you later added the template). This is not a problem or a personal issue, rather an opportunity to solve a problem, as Risker points out. Adding a gross, glaring template to the top of featured articles is not an optimal way to implement this trial. Nothing personal, and I'm sure this will get sorted quickly. SandyGeorgia ( Talk) 02:49, 18 June 2010 (UTC) reply

Don't worry, but like I said at the talk page: "Just because it has a small star, does not deserve other privileges", if it would be the TFA, I could understand it. Moreover, if it would be small I have no problems, though the notice obstruct it. Tbhotch ^{Talk C.} 02:57, 18 June 2010 (UTC) reply

Comment. I agree that the edit notice is a better place for this message (in general). I have no idea if the following is possible to implement technically, but if template code in a "global" edit notice could produce output (a warning) only if the article is in whatever protection mode is associated with the "pending changes" function, then you'd have a maintenance-free implementation of the message across all articles. Riggr Mortis ( talk) 03:23, 18 June 2010 (UTC) reply

• Maybe after the trial is over just having the little "accepted" box at the top will be enough, with the template on the talk page? -- phoebe / ( talk to me) 07:50, 18 June 2010 (UTC) reply

Some sort of talkpage notification would be sensible I think. Ged UK  08:04, 18 June 2010 (UTC) reply

• Here's what I wrote at WT:PROT#Pending Changes article tag:
  
  I say we get rid of it. Every article under PCP already shows an informational message during editing, and shows the status message in the top right corner. If any content from {{ pp-pending}} is missing from that edit notice, it should be added there.
  But showing that intrusive banner at the top of soon 2000 largely perfectly decent articles is not helpful (and showing it on any good or featured articles is right out).
  Amalthea 08:08, 18 June 2010 (UTC) reply

We seem to have the same discussion going in several places, we should probably centralise it. I have just removed the message box from {{ pp-pending}} following a discussion on Template talk:pp-pending which suggested there was not consensus for the message. (I have retained the categorisation which the template was performing and these may be useful.) — Martin ( MSGJ ·  talk) 09:23, 18 June 2010 (UTC) reply

I haven't been able to keep up, but it looks like this is sorted now. Could someone who understands check whether this is the correct solution? Why is move-indef taken off? SandyGeorgia ( Talk) 12:15, 20 June 2010 (UTC) reply

Also, could someone fix the grammar in the template now on the page? "This is the latest accepted revision, accepted on <date>" is just redundant and awkward (accepted, accepted). SandyGeorgia ( Talk) 12:25, 20 June 2010 (UTC) reply

Is there any way to get statistics on how quickly pending changes are accepted/reverted? My impression from the last few days is that pending changes never remain unreviewed for more than a minute or two. If this is the case, it implies that it will be possible to substantially increase the number of articles with pending changes. -- Zvika ( talk) 08:14, 21 June 2010 (UTC) reply

Wikipedia_talk:Pending_changes#How_is_success_or_failure_to_be_judged.3F Tony (talk) 08:51, 21 June 2010 (UTC) reply

Difficult to tell at the moment. Firstly, there's the "novelty factor" - editors wanting to try something new, secondly, there's the relatively small number of articles that have pending changes applied. Once the novelty factor has worn off a bit and more articles are added, a better picture will emerge. Mjroots ( talk) 10:32, 21 June 2010 (UTC) reply

My question was whether there is some automatic statistics generation based on the review logs. It seems there isn't, unfortunately. -- Zvika ( talk) 12:05, 21 June 2010 (UTC) reply

Please see the Pending Changes Stats Page. There's quite a bit of work we need to do on this page, but right now it includes some (poorly labeled) basic stats. The percentile table gives a sense of how long changes wait before they are approved. There's also a mean and media calculation. Howief ( talk) 19:03, 21 June 2010 (UTC) reply

Thanks! This is exactly what I was talking about. I'm not sure I understand what is the difference between "wait for edits by users that have not logged in to be reviewed" and "pages with unreviewed edits pending," though. -- Zvika ( talk) 04:00, 22 June 2010 (UTC) reply

{{helme-nq}} This page is linked prominently from flagged articles (I came form Sarah Palin, and WP users (of whom editors make up a very small percentage) need to be clearly and quickly informed as to what it all means. Instead there's a big orange box aimed at editors, full of backroom technicalities and jargon, which makes this help page totally inaccessible for the casual reader. Please optimise the page for readers instead of editors; one way of doing this would be to {{ hide}} the advice for editors. Thanks 86.41.79.133 ( talk) 17:23, 23 June 2010 (UTC) reply

helpme is not really for requesting edits. monosock 17:27, 23 June 2010 (UTC) reply

The writer has a point. If there's a summary for editors, why not a summary for readers? There are certainly many more of them.

Also, the good-faith attempt at improving a page should't be met with a technical quibble about template choice. It's discouraging for contributors, especially new ones. 69.142.154.10 ( talk) 19:02, 23 June 2010 (UTC) reply

The article states that, "The public (99.5% of whom are not editors)..." Does this mean that only 0.5% of people accessing Wikipedia actually contribute with edits while the rest are non-intrusive readers? Or does it mean that 99.5% of wikipedia hits are for reading purposes only? Politis ( talk) 10:04, 24 June 2010 (UTC) reply

I've hidden the information for editors by default. Cenarium ( talk) 23:44, 25 June 2010 (UTC) reply

I have not applied to be a reviewer but intend to continue to monitor my particular watchlist for vandalism and other issues. One of those articles is Abraham Lincoln and this [1] represents the article history.

When I revert vandalism by an IP, does that still mean that a reviewer will still need to address that IP's edit? In the history linked above, there have been 10 recent reverts of IP or new user vandalism -- do these reverts eliminate the need for a reviewer to become involved? If a reviewer DOES eliminate a case of IP vandalism, how does it look on the article history -- does it disappear completely or does it just continue to show in the way that any reverted edit continue to show? Is there anyway to determine whether a revert is done by a reviewer or simply a regular editor?

The bottom line -- does it serve any purpose for me to eliminate vandalism on articles subject to the new review policy? Tom (North Shoreman) ( talk) 22:10, 24 June 2010 (UTC) reply

seperate question, according to the appointment of reviewer status, it states that edits should be accepted "if they are not obvious vandalism or BLP violations" is this still the case? Should we review only for those violations or should we watch for say unsourced statements, and other policy issues? Sephiroth storm ( talk) 01:35, 25 June 2010 (UTC) reply

If I feel confident in my knowledge of a Subject, I review the statement for all crietria of Wikipedia. Yes I should get administrator pay for such an act, but reviewer pay is ok for now. Otherwise I simply review for blatant retardation. Beam 23:49, 6 July 2010 (UTC) reply

"The bottom line -- does it serve any purpose for me to eliminate vandalism on articles subject to the new review policy?" - YES. Beam 14:35, 10 July 2010 (UTC) reply

Hi all. This new procedure came in while I was away from Wikipedia, and I can't seem to find the details of how to implement WP:PEND on a specific page. I've tried to give the page Stokes Valley PEND-protection, but I'm sure I've done something wrong. Can someone who knows more about the procedure please check whether I've done it OK ? I followed the info at Help:Pending_changes#Advice_for_admins, but I'm not convinced I succeeded! (BTW, FWIW, the page has had ongoing low-level vandalism from anons for about a year. Not enough to need full protection, but enough to need something done with it). Cheers, Grutness... wha? 09:38, 18 July 2010 (UTC) reply

For this edit, should I have accepted the edit and then fix the mistake, or just fix the mistake? I thought I should accept the edit because it wasn't vandalism then fix the mistake, but then I thought if I would do that, then all the readers would see is just a broken reference for a few seconds before I would fix it. I ended up doing nothing and someone else accepted it, but would it have been ok to neglect accepting it and making the fix? エムエックスさん 話 20:13, 24 July 2010 (UTC) reply

Sorry that no one has answered you as yet.

Personally, I would accept the edit, then fix the mistake. Hitting "Edit" on a pending change sometimes causes your own edit to become the one pending (which you can routinely accept, mind, but it looks funny, IMO, approving your own edit). Undo simply reverts an edit, and thus it becomes automatically accepted because you restored it to the prior version. So again, accept and then edit, since yours gets accepted automatically. CycloneGU ( talk) 22:41, 27 July 2010 (UTC) reply

• "Unaccept" is apparently unacceptable, since the button is grayed out. I tried both Firefox and Internet Exploder. •  Ling.Nut 13:34, 14 August 2010 (UTC) reply

I though the same thing for a while. For reading the documentation it seems that you just need to revert the edit by normal means (rollback etc). That could be made clearer, possibly change the functionality so the the "unaccept" button performs a roleback? -- Salix ( talk): 18:38, 14 August 2010 (UTC) reply

A grayed-out button DOES surprise the user, since it is counter-intuitive. Violates principles of software creation. •  Ling.Nut 22:10, 14 August 2010 (UTC) reply

The article refers to a "two month trial" in the lead paragraph, but doesn't give any dates, so I don't know how far through the trial we are. If someone who knows could add that information, that would be great - thanks! ciphergoth ( talk) 06:18, 20 August 2010 (UTC) reply

Is this question not going to be answered? μηδείς ( talk) 06:34, 1 October 2010 (UTC) reply

It's going unanswered because nobody knows the full answer. The two-month trial began on June 15, 2010, (as I recall, although it may have been the 16th), so it would have ended around August 15. However, the software involved was not shut down at that time, nor has it been turned off yet; pending changes is still in force. What has been happening is that a month-long discussion about what to do next achieved no consensus, during which the software remained active. The next activity was an announcement by Jimbo Wales about the status, followed by much discussion, then a straw poll (August 22 – September 4), followed by another straw poll (September 20 – September 27). The trial is nominally over, but it's still going on; the software is still active, and pages are still protected. A new version of software is due around November 9, and it appears that the feature will remain active until (and after) that date. More than that I can't tell you. —  JohnFromPinckney ( talk) 09:52, 1 October 2010 (UTC) reply

Looking at one article currently reviewed - Gilad Atzmon - One thing I see is that in cases of clear vandalism, the entry itself is deleted. Is this true when edits are not allowed because they are, say, not referenced or WP:OR using an existing reference, or whatever? If so, that would remove evidence of patterns of vandalism or disruptiveness by editors who may need sanctions. Is that explained in the sentence "Unaccepted edits are never "lost" for lack of being "accepted". They are simply no longer candidates for wider public display as the current accepted version." If so, I don't understand the explanation. CarolMooreDC ( talk) 03:48, 31 August 2010 (UTC) reply

The way the system works, as far as I can see, is that you "unaccept" by making a new edit, so unaccepted edits are always in the history (unless an admin removes them for copyright violation, libel or something of that order, something which has always been possible, which is extremely rare, and which has nothing to do with pending changes).-- greenrd ( talk) 08:03, 9 September 2010 (UTC) reply

I can't find the "accept button" anywhere. Is it because I am not an admin and only admins get to see it? If so, this Help page should be made clearer.-- greenrd ( talk) 08:03, 9 September 2010 (UTC) reply

Oh, I see - it's because I'm not a reviewer. It does mention this at the top of the page - I read the page too hastily.-- greenrd ( talk) 08:06, 9 September 2010 (UTC) reply

What is the current status of the trial? Should PC be removed from pages currently on trial and protection implemented as needed? Fvasconcellos ( t· c) 19:01, 9 September 2010 (UTC) reply

• See comment by Jimbo here.-- Doug.^{( talk • contribs)} 22:39, 19 September 2010 (UTC) reply

I edited the Condoleezza Rice page, and my edit was put up as "pending review" although recent edits of mine on this page had been accepted automatically. What I did was re-instate some uncontroversial information that had been removed because it wasn't sourced, and added some sourcing. I did this by editing the second last version of the page (i.e. editing an "old" version), rather than undoing and then adding. Is that what triggered the review? VsevolodKrolikov ( talk) 08:04, 15 September 2010 (UTC) reply

• No, that page has pending changes protection implemented. This one doesn't. That's all.-- Doug.^{( talk • contribs)} 22:42, 19 September 2010 (UTC) reply

I just ran into the 'pending changes' thing for the first time. I'm a long-time editor, generally only making minor and cosmetic changes. I don't usually bother to log in, because I think anonymous editing is one of the things that makes Wikipedia as great as it is. I hope this trial ends and doesn't come back; if it ends up being widely adopted I, and who knows how many more people, will probably no longer contribute, and I don't know what that would do to the project. 201.236.144.28 ( talk) 00:40, 5 November 2010 (UTC) reply

Im agree, please delete this stupid new rule of wikipedia. —Preceding unsigned comment added by 190.27.237.33 ( talk) 22:50, 12 November 2010 (UTC) reply

Hopefully it would reduce the vandalism. -- THE FOUNDERS INTENT ^PRAISE 23:14, 12 November 2010 (UTC) reply

The greatest evils in this world and most human misery are the result of good intentions. User 201.236.144.28 is correct.— Quicksilver ^{T @} 09:42, 28 December 2010 (UTC) reply

• Wikipedia: The Free Encyclopedia Anyone Can Edit (if we decide to let you)-- 61.224.46.12 ( talk)

I agree. It is a huge disincentive to edit. In my experience, the admins cannot be trusted to act appropriately, and I don't want them sitting in judgement over me every time I make a minor edit. I'm quitting. Alex Middleton ( talk) 16:17, 8 December 2010 (UTC) reply

Show me a micromanager who doesn't trust others and I'll show you someone who is incompetent both as a manager and as one who might attempt to perform the work of those he purports to manage. Show me an enterprise that is being micromanaged and I'll show you an enterprise that is about to disappear. Proponents of the Pending Changes mechanism would make ideal bureaucrats in the world of George Orwell's novel, 1984. I concur, Alex. If this is implemented in the English Wikipedia, I'll walk away after contributing for nearly six years. This idea has already been implemented for a while in the German Wikipedia, and the result is many stagnating, inferior articles. Why bother?— Quicksilver ^{T @} 09:42, 28 December 2010 (UTC) reply

Uh guys, you seem to have missed the point that you do not have to be an admin to review pending changes. Thousands of non-admin users are reviewers. We have to micromanage because our articles are vandalized all day, every day. There are different levels of PC, the lowest level merely requires you to be an autoconfirmed user, see this handy chart for details. Beeblebrox ( talk) 23:56, 25 January 2011 (UTC) reply

I don't understand how to accept pending review changes? Just made a valid change (with source) to Oldham East and Saddleworth by-election, but was not logged in. How do I approve the change? The instructions really aren't clear, and there's no 'accept pending changes' button anywhere. doktorb _words ^deeds 16:25, 20 December 2010 (UTC) reply

I was just looking at a piece of obvious vandalism that lasted for 30 minutes on a page I watch before being reverted, and found myself wishing that the page was under Pending Changes protection. What happened after the Pending Changes trial? From the review log it looks like reviewing is still happening, even though the trial has "ended". Where can I request for a page to be added? Ryan Paddy ( talk) 00:19, 15 April 2011 (UTC) reply

Hmm, I see the 2011 RfC now and I guess that answers my question about the result: none yet. Ryan Paddy ( talk) 00:23, 15 April 2011 (UTC) reply

Result is in. See info box at the top of this page. Guy Macon ( talk) 09:30, 12 May 2011 (UTC) reply