Wikipedia talk:Pending changes/Straw poll/Archive 2

From Wikipedia, the free encyclopedia
Jump to: navigation, search
Archive 1 Archive 2 Archive 3

No Consensus

I think there is a clear lack of consensus at this poll. There are about as many people agreeing as disagreeing (by my guesstimation), and this doesn't look likely to change. No-one is looking at eachother's comments, so to treat it as a "discussion" would be a fallacy. I'm not suggesting we default to anything, I just suggest this poll be discarded as unhelpful. I think a centralized discussion would be much more effective. We have Wikipedia:Pending Changes/Closure, but we're sending everyone here instead. This isn't helpful at all. ☻☻☻Sithman VIII !!☻☻☻ 01:01, 31 August 2010 (UTC)

There are twice as many people in support of some form of Pending Changes system as there are people who oppose the entire deal. If there isn't consensus for final approval, I'd say there's certainly enough support to extend another trial for the system. BigK HeX (talk) 01:17, 31 August 2010 (UTC)
No, not quite 2:1. —Jeremy (v^_^v PC/SP is a show-trial!) 01:33, 31 August 2010 (UTC)
Oh .. sorry if my rounding is undue. Though I'd guess no one feels misled by that deviation of 01.52% from an exact 2:1 ratio. Meh ... BigK HeX (talk) 01:49, 31 August 2010 (UTC)
Just to be difficult, another way of looking at it is that (following counts are approximate, using search and replace tools, not a hand count), currently, the support breakdown is:
  • Option 1: 178
  • Option 2: 72
  • Option 3: 211
  • Option 4: 74
If we consider each option separately, not as a group, that's a pretty inconclusive set of results; unfortunately, what do with them depends on how you interpret the initial proposal. If the initial proposal was "Test for 2 months, if consensus supports, we keep it," then it seems hard to justify keeping. If, instead, the initial proposal meant, "We want to roll out this feature, so we have a 2 month test, and continue moving unless consensus says stop," then we should keep on with another trial. I have no opinion on which is the correct interpretation of the initial proposal, as both seems plausible to me, and it seems like any interpretation of that proposal is automatically colored by a person's current feelings about the results of the trial. Ugh. Qwyrxian (talk) 01:54, 31 August 2010 (UTC)
Well .. pretty obviously, the poll could've easily been designed as a simple Support/Oppose request. Seems there'd be a tiny bit of absurdity in penalizing supporters because their poll options were more elaborative. BigK HeX (talk) 01:58, 31 August 2010 (UTC)
Currently 66.04% : 33.96%. But this isn't a vote; whether or not the stats for go over 66.66%, that's not the deciding matter here. EricLeb01 (Page | Talk) 04:05, 31 August 2010 (UTC)
The raw vote counts are not a good way to judge consensus. Given the limited trial, only a small number of users may have experience with each drawback. There are not that many votes for options 3 and 4 that acknowledge they have reviewed the cons and still think an expansion is prudent. --UncleDouggie (talk) 06:05, 31 August 2010 (UTC)
Didn't I just say that? "[...] whether or not the stats for go over 66.66%, that's not the deciding matter here."
Maybe because they either have not experienced the cons or have adopted the "I am the World" mentality - if it makes their workload "easier", so be it. —Jeremy (v^_^v PC/SP is a show-trial!) 06:44, 31 August 2010 (UTC)
Judging from the straw poll, everyone seems to feel very strongly about this. Is "almost" 2:1 in favor with much dissent a consensus? I think not. A majority? Certainly. But not a consensus. ☻☻☻Sithman VIII !!☻☻☻ 20:47, 31 August 2010 (UTC)
Badly Designed Poll! Each option should have been its own seperate choice, to reflect actual representation. As is, the poll is unduly weighted towards supporters of the system.--Gniniv (talk) 07:55, 1 September 2010 (UTC)
You could just as well claim the opposite: opposers get one united vote. Choyoołʼįįhí:Seb az86556 > haneʼ 18:00, 1 September 2010 (UTC)
If greater than 50% support it use we should keep it. For those projects who do not like PC they do not need to use it. For those projects who do they will than be able. Obviously though we should keep the trial running to further determine its usefulness.Doc James (talk · contribs · email) 18:31, 1 September 2010 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── That's headcount, though, not argument. Consensus is not defined by headcount, despite what Off2riorob thinks, but by reasoned and rational arguments for or against and how legitimate they are. —Jeremy (v^_^v PC/SP is a show-trial!) 21:00, 1 September 2010 (UTC)

Yes thus even if less than 50% support PC we should implement it as it is just a good idea. Good for the project, good for the encyclopedia.Doc James (talk · contribs · email) 21:18, 1 September 2010 (UTC)
@Doc James: Except it's not a feature you can choose not to use--it fundamentally alters the way edits are done, the way pages are displayed, and thus how both readers and editors interact with the site. For example, I haven't even registered a vote yet, because I'm torn--I absolutely hate what it does to my editing and my watchlist, it hasn't been explained clearly, it gets messy when there's a series of edits some of which are auto-accepted and some of which are not, etc; sometimes I think I'm rejecting or accepting but it doesn't do what I expect, etc. As a reader, however, I can see the value in keeping down potential vandalism on certain types of pages. But as for the poll, and the nature of this feature, my feeling is that we need overwhelming consensus to keep it, not just a 50%+ mark. It shifts not only the method of editing, but further deepens the divide between IP editors and editors with accounts. That's too big a change to let stand at the 50% mark. Just possibly I'm willing to believe that 50% means that a longer trial period is in order, but the new trial period should then have a clear specific target at the end that will define whether the feature is kept or not. Qwyrxian (talk) 01:18, 2 September 2010 (UTC)

On a lighter note

This page has a bot to archive all threads older than 5 days, and nothing has been archived yet! Contentious issue, thou art Pending Changes. ☻☻☻Sithman VIII !!☻☻☻ 20:50, 31 August 2010 (UTC)

That's just because the bot hasn't been around here yet. When it does, I promise you, significantly more than 66.66% of the page will go away. – Smyth\talk 07:13, 1 September 2010 (UTC)

Down with Pending Changes!

Option 1 supporters are encouraged to show their dislike of Pending changes using the following Userbox.--Gniniv (talk) 07:47, 1 September 2010 (UTC)

Is this really necessary? We could also have Support Pending Changes! but reasoned debate seems better.Doc James (talk · contribs · email) 18:29, 1 September 2010 (UTC)
This is reasoned debate?--Bbb23 (talk) 18:32, 1 September 2010 (UTC)

Decision to continue using pending changes should be left to the Wiki Projects

It seems like people are split over pending changes ( with more favoring its us in some form ). Maybe we should leave it up to WikiProjects to decide if this is something they wish to us or not. Twenty of us at WP:MED have a paper in publication which we hope will encourage contribution to Wikipedia from the health care community. During the peer review of the paper a number of concerns were voiced including the issue of vandalism. We have discussed in the paper how pending changes helps address these concerns and all 20 of us editors from WP:MED supported its continued usage.Doc James (talk · contribs · email) 16:45, 1 September 2010 (UTC)

The only problem is that the WikiProjects have no power to persuade the devs to keep this on if much of the community want no part of it, Doc. WikiProjects actually lack political pull. —Jeremy (v^_^v PC/SP is a show-trial!) 20:40, 1 September 2010 (UTC)
Thankfully most of the community do wish to see this used.Doc James (talk · contribs · email) 21:13, 1 September 2010 (UTC)
A headcount isn't consensus, chummer. If this were a consensus discussion it'd be pretty close to no-consensus. —Jeremy (v^_^v PC/SP is a show-trial!) 21:32, 1 September 2010 (UTC)
Support is still well over 60%. PC is currently running and their is no consensus to stop it use. But I agree it probably needs a longer and larger trial now that we have some experience on how it is best used.Doc James (talk · contribs · email) 21:33, 1 September 2010 (UTC)
A headcount isn't consensus. There's no consensus for or against using it at present. —Jeremy (v^_^v PC/SP is a show-trial!) 22:33, 1 September 2010 (UTC)
Yes so you agree we should trial it longer? Which I think all would be fine with.Doc James (talk · contribs · email) 23:10, 1 September 2010 (UTC)
Doc James, read my responses to David Levy in this thread, and next time please don't ask questions I've already answered beyond a shadow of a doubt. —Jeremy (v^_^v PC/SP is a show-trial!) 23:20, 1 September 2010 (UTC)
Just appeared that with discussion you might have changed your position somewhat. I am unsure what you are referring too in relation to David LevyDoc James (talk · contribs · email) 23:23, 1 September 2010 (UTC)
Start reading my responses in the "Include abbreviated list from Closure page?" section. —Jeremy (v^_^v PC/SP is a show-trial!) 23:27, 1 September 2010 (UTC)
Ah yes missed that bit. Doc James (talk · contribs · email) 05:00, 2 September 2010 (UTC)

Potential harm from Wikipedia

This study [1] found that "3% of e-patients say they or someone they know has been harmed by following medical advice or health information found on the internet". Pending changes is one of a few tools which allow us to keep Wikipedia accurate. 70% of junior physicians use Wikipedia in patient care as do nearly 50% of practicing physicians. This is not something we need just to decrease libelous information on BLP pages.Doc James (talk · contribs · email) 18:45, 1 September 2010 (UTC)

This is not something we need at all. Accuracy and quality come from more openness, not less. --Yair rand (talk) 20:07, 1 September 2010 (UTC)
We explicitly note we are not a substitute for a doctor, lawyer, or other similar professional. —Jeremy (v^_^v PC/SP is a show-trial!) 20:38, 1 September 2010 (UTC)
Accuracy and quality comes from more openness from competent editors, or less openness from vandals. And, in any case, pending changes does not close off pages from more editors, unlike it's only current substitute (semi and full page protections, which DO restrict openness). BigK HeX (talk) 20:41, 1 September 2010 (UTC)
Accuracy and quality cannot be achieved using any iteration of FlaggedRevs on topics such as these short of having professionals constantly on tap to approve the edits. This has already been discussed on WP:Pending changes/Closure. —Jeremy (v^_^v PC/SP is a show-trial!) 20:44, 1 September 2010 (UTC)
In addition, ere I forget, using FlaggedRevs to control the article's accuracy would be problematic at best. What's good for one area isn't good for others. —Jeremy (v^_^v PC/SP is a show-trial!) 20:53, 1 September 2010 (UTC)
Flagged revisions is NOT censorship. However, the only substitute we have is to block off pages, which CAN be viewed as censorship, in a prior exclusion sort of way. Pending Revisions is a helluva large step UP towards more openness, given that we resort to outright page protections at present. BigK HeX (talk) 21:03, 1 September 2010 (UTC)
No, BigK, FlaggedRevisions is far more capable of censorship than any form of protection because reviewers are human and have strong feelings about certain subjects. You seem to be in a fantasy land where PC'd articles are protected by intelligent AI that have no bias for or against any subject. While in the real world, humans can - and almost always do - have such a bias. —Jeremy (v^_^v PC/SP is a show-trial!) 21:07, 1 September 2010 (UTC)
I live in a land where those same editors with strong feelings can (and often do) hit the Undo button. Pending Changes has absolutely no bearing on that matter. If we are talking about the issue of reviewers misusing their tools, then the tools should be revoked just as any other abused privilege on Wikipedia.
Are you arguing that Page protection is preferable to Pending changes?? ("Yes" and then why or "No" and then why, please) BigK HeX (talk) 21:15, 1 September 2010 (UTC)
Yes, because then there is no question as to whether an IP can edit a page or not without it seeming like a crapshoot. With Pending Changes - and you cannot deny this - an IP will be completely at the mercy of the POV or anon disapproval of the reviewer and thus must gamble on the chance that his edit, constructive but poorly-worded or well-done and -intentioned but slightly incorrect, will fail any one Reviewer's tests. In short, FlaggedRevisions is anathema to AGF, amongst other things (The revision pollution, ineffectiveness against those who seriously wish us ill, and unsustainable levels of manpower do not help).
And removing a Reviewer's tools? We have a hard enough time as it is revoking users' rights that removing a Reviewer's bit just because of POV will end up turning into a bitchfest between opposite sides that will choke out any discussion and make it impossible to do so. —Jeremy (v^_^v PC/SP is a show-trial!) 21:30, 1 September 2010 (UTC)
ANY edit is at the mercy of the next editor that comes along, and whatever POV they may harbor. Pending Changes does NOT change that. You seem to impute awesome apocalyptic powers to Pending Changes, when it has none. We edit COLLABORATIVELY on Wikipedia, and Pending Changes merely reinforces this fact, and merely does it in areas that are subject to having a far more severe restriction used. BigK HeX (talk) 22:12, 1 September 2010 (UTC)
With all due respect Doc, and to ditto Jeremy's comment, people coming to harm from finding erroneous advice on the interwebz is a problem of human stupidity, not vandalism on wiki Med articles. They just as easily come to harm from hanging out on web MD too much and following treatments without the advice of a doctor. Millahnna (talk) 06:49, 2 September 2010 (UTC)

Comparing difficulty of Pending Changes and the Status Quo for legit editors

Pending Changes is far more severe than the normal editorial process, however, BigK, and I heavily doubt anyone defending it aside from you even realizes this. A Reviewer literally has the power to ensure that an edit, for good or ill, never sees the light of day as the current revision. —Jeremy (v^_^v PC/SP is a show-trial!) 22:17, 1 September 2010 (UTC)
What? Pending Changes makes sure there's a LARGE chance that an edit sees the light of day as the current version. Try as I might, I completely fail to see how you think PAGE PROTECTION is an improvement on the odds, when the cumbersome {{editprotected}} process is really what practically ensures that a proposed edit will not see the light of day. The bottom line is that:
A) Pending Changes requires:
  1. a simple click from any long-term editor.
B) Page Protection requires:
  1. an editor to figure out how to post a talk page section,
  2. to figure out how to get consensus on an edit,
  3. has to actually achieve consensus(!!!!), and
  4. to then realize that {{editprotected}} is the correct Bat Signal, AND then
  5. requires admin intervention of all things.
I am at a complete loss at seeing how a person can reasonably argue that Process A is a tougher barrier than Process B. BigK HeX (talk) 22:27, 1 September 2010 (UTC)
──────────────────────────────────────────────────────────────────────────────────────────────────── Process A also has to go through a human being whom is one of these possibilities: (a) knowledgeable of the subject matter and has no dog in the fight; (b) A Defender of the Holy Wiki whom has become jaded by frequent and unrelenting IP vandalism and whom by nature distrusts IPs and nonautoconfirmeds; (c) unaware of the subject matter and not likely to read up on it before accepting or rejecting the edit; (d) aware of the subject matter and heavily opinionated about it, not even willing to entertain the thought of the "enemy"; (e) a deckhead who cares more about RPM than accuracy. —Jeremy (v^_^v PC/SP is a show-trial!) 22:38, 1 September 2010 (UTC)
You've explained nothing about how these Defenders of the Holy Wiki et al manage to show up in the Pending Changes process, and yet somehow don't get their say in opposing an {{editprotected}} request, which requires consensus.
Pending changes is a far lower barrier to editing than the alternative. BigK HeX (talk) 22:51, 1 September 2010 (UTC)
These "Defenders of the Holy Wiki" are often administrators. Reviewer rights are part and parcel of the admin package (which is why I gave mine up, albeit under a cloud). —Jeremy (v^_^v PC/SP is a show-trial!) 23:00, 1 September 2010 (UTC)
Wait ... so, these Defender of the Holy Wiki Admins somehow thwart the Reviewer process (which is open to any long-term editor), but can't thwart the {{editprotected}} process (which requires an admin)? I'm baffled, to say the least... BigK HeX (talk) 23:06, 1 September 2010 (UTC)
Speaking as a former one myself, BigK, DotHW do not usually serve editprotected requests. —Jeremy (v^_^v PC/SP is a show-trial!) 23:11, 1 September 2010 (UTC)
I agree with BigK PC is not censorship and does increase everyone's ability to edit. Freedom is a balance of "freedom to" and "freedom from". Many editors wish a little more freedom from having vandalism go live thus degrading the view of Wikipedia in the eye of the public and new potential editors.Doc James (talk · contribs · email) 21:15, 1 September 2010 (UTC)

arbitrary break 1

As far as Wikipedia goes, whose who forsake freedom (to edit) for security ("Freedom from vandals") get neither. —Jeremy (v^_^v PC/SP is a show-trial!) 21:30, 1 September 2010 (UTC)

(undent) This thread is about how PC will

  1. Help protect our readers from being exposed to as much vandalism
  2. Protect the people we write about (BLPs) from libelous material
  3. And hopefully encourage more people to join us as editors as we will offer some protection of their efforts

It will help do all this while keeping true to Wikipedia's core principle that anyone can edit. Otherwise as suggested semi-protection may be needed more frequently. This is a gray issue not black and white.Doc James (talk · contribs · email) 21:32, 1 September 2010 (UTC)

  1. What about the circumstances when an oblivious or stupid Reviewer accepts vandalism or blatant disinformation?
  2. What about the edit history chock-full of BLP-violating garbage that anyone could find and sue us over?
  3. What about the obvious lack of instant gratification and Sanger Syndrome?
It's not as effective as you would think. —Jeremy (v^_^v PC/SP is a show-trial!) 21:35, 1 September 2010 (UTC)
Reply to responses
  1. Reviewers who have difficult with the tool will lose privileges to it usage.
  2. Stuff in the edit history exposes Wikipedia to much less liability than stuff that goes live ( seems fairly obvious )
  3. Most pages will not be semi-protected / run pending changes thus still lots of chances for instant gratification, anyone can still edit so this is hardly citizendium. The first page I ever attempted to work on was Obesity. I was unable to edit it as it was semi protected. Thus worked on overweight until I had enough edits to edit obesity. Subsequently brought the page to GA making over 1000 edits.Doc James (talk · contribs · email) 21:48, 1 September 2010 (UTC)
  1. I'm talking about one-off errors, not a campaign of bad reviewing. (I'm not so stupid as to doubt that people abusing the tool will lose it).
  2. The problem is that stuff in the edit history often stays on Google long after it has been reverted, so libelous items in the history can still get us in hot water.
  3. The problem is that the pages that are under PC or prot are the ones IPs are most interested in. During the school year, you'll tend to see more IPs on books, historical events, and the like. During the downtime that is summer vacation, you're more likely to see pop-culture articles edited. So, in effect, there is no instant gratification. —Jeremy (v^_^v PC/SP is a show-trial!) 22:00, 1 September 2010 (UTC)

arbitrary break 2

Where situations arise in which there is too much vandalism for the patrollers to deal with, semi-protection can be necessary in certain places. If an article is vandalized many times over a short period of time, it can be temporarily semi-protected, and if it continues to get large amounts of vandalism over a long period of time, long-term semi-protection can be necessary. Semi-protection is used sparingly, making sure that as few articles as possible have editing restrictions. It is a temporary measure, until the community is large enough to be able to deal with vandalism without resorting to protecting the article. Semi-protection stops vandalism where we can't deal with it regularly. PC does not prevent vandalism. It doesn't do anything to prevent vandalism, it only prevents its visibility without removing some work for patrollers, and it causes the same harm semi-protection does, with none of the benefits, and thus is worse than semi-protection in every way. Thus, PC is useless, and potentially extremely harmful. Which part of this is incorrect? --Yair rand (talk) 22:17, 1 September 2010 (UTC)

I'd say that it's incorrect to state that PC is worse in every way. It IMPROVES accessibility to vandalism prone articles, as page protection obviously blocks access. However, more important is that preventing visibility is not quite the insignificant point that you seem to suggest. What vandal would waste their time vandalizing something that only an anti-vandal will see and certainly remove? Preventing the visibility is the KEY benefit, and a huge deterrent to vandalism where some measure of protection is needed. Pending Changes powerfully deters vandalism, while still allowing a very good chance of many productive edits getting through (and certainly far more productive edits than would happen when subject to the {{editprotected}} gauntlet). This is how it is far superior to page protection against vandalism. BigK HeX (talk) 22:46, 1 September 2010 (UTC)
You seem to be under the impression that Pending Changes magically affects only unproductive edits, and only negatively. Yes, it "IMPROVES accessibility" over semiprotection - to vandals and tendentious editors as well as good faith editors. And no, it isn't true that "only an anti-vandal will see and certainly remove" vandalism - firstly because it is trivial to get bad changes accepted, since some reviewer will approve them if you are smart, and secondly because PC will then get in the way of removing the vandalism - that previously instant revert might be delayed for some time by the need to have it accepted. And "a very good chance" of productive edits getting through is serious breakage compared to the current 100% chance on unprotected articles. Gavia immer (talk) 22:57, 1 September 2010 (UTC)
More like 96%, but still better than PC. —Jeremy (v^_^v PC/SP is a show-trial!) 23:00, 1 September 2010 (UTC)
A "100% chance on unprotected articles"... How is that relevant at all?? We're talking about a tool to thwart vandalism. For vandalized articles that need help, we're certainly not talking about articles that are unprotected.
Also, if a person "has to be smart" to get an edit through, then it probably isn't vandalism .. or at least not simple vandalism. As far as tendentious editors "getting through," we're not talking about content disputes or even POV warriors. That is not the intended function of Pending Changes.
I'm not really sure how your comment is relevant to the intended usage of Pending Changes, which is for vandalism (and sockpuppets) on article that -- with our current toolset -- would end up Page Protected, forcing ALL editors to go through the {{editprotected}} process, which I know that I, personally, never bother with. I applaud editors who do bother with making {{editoprotected}} requests, but it is far too burdensome for me ... which is why I completely fail to see how people believe that Pending changes is worse...? BigK HeX (talk) 23:05, 1 September 2010 (UTC)
You may not, chummer, and therein lies your logical error. I've seen several editprotected requests by IPs for noncontroversial changes that readily get served. ({{editprotected}} would not allow users to put in controversial information anyhow, as in most cases said info is at the center of an edit war.)
You are also severely underestimating vandals' intelligence. I've seen a few IP vandals intelligent enough to post crap that you would need an expert on the subject to wipe out. —Jeremy (v^_^v PC/SP is a show-trial!) 23:11, 1 September 2010 (UTC)
Just because some IPs will do it, doesn't mean all will. Having the "you can edit now" is FAR FAR less a deterrent than "you have to go to the talk page, use the correct template, and how your edit is accepted". I'm consistently baffled how anyone in their right mind could say the later is the better option, even if the first one may require a bit of waiting for it to be 'visible'. Seriously, someone needs to explain how being able to edit is worse than not being able to. ♫ Melodia Chaconne ♫ (talk) 02:05, 2 September 2010 (UTC)

arbitrary break 3

PC while not perfect is much better than were we are now. No reasonable harm comes from using and possibly a lot of benefit. Thus for those who wish to apply this tool to improve the encyclopedia why stand in their way.Doc James (talk · contribs · email) 23:14, 1 September 2010 (UTC)
Because for many of these users "improving" the encyclopedia means getting it to match their weltanschauung, and for Wikipedia, that is far worse in the long run. —Jeremy (v^_^v PC/SP is a show-trial!) 23:27, 1 September 2010 (UTC)
Users may very well be able to find the talk page, the proper template, and flag down an admin to institute an edit, but you've still not explained how this is somehow simpler than just making an edit that ANYONE can then institute. Hell, in fact, an editor can STILL use an editprotected request to flag down an admin if somehow the Pending Changes system is too burdensome for him/her.
I'm aware that some editors can get through the editprotected process, but that's quite besides the point, which is that one is FAR easier to navigate. So could you comment on the relative difficulty of the two? BigK HeX (talk) 23:19, 1 September 2010 (UTC)
I'm not going to comment on the relative difficulty to someone who only knows its name and not how it works. (By your own admission above, you have never used it. I serviced a few requests before I got too extreme.) Try using it first, and come back. —Jeremy (v^_^v PC/SP is a show-trial!) 23:27, 1 September 2010 (UTC)
I know about the process, and that's irrelevant to YOU commenting on the relative difficulty. I've participated in editprotected requests before. I merely refuse to use them myself, precisely because the barrier is far too high for me to waste volunteered time on. I would appreciate a response to my question. What do you think of the relative difficulty?? BigK HeX (talk) 23:32, 1 September 2010 (UTC)
For the most part, the relative difficulty in using editprotected is extremely constant compared to the whims of a reviewer enforcing FlaggedRevs. While an editprotected request is incredibly straightforward and allows the edit to be made with minimal fuss, FlaggedRevisions turns Wikipedia into the Strip. An IP has little chance of getting a sympathetic and knowledgeable reviewer. —Jeremy (v^_^v PC/SP is a show-trial!) 23:39, 1 September 2010 (UTC)
Just to clarify, you are saying that you consider a (potentially new) editor finding the talk page, explaining a proposed edit, and using the correct template" to be a process of "minimal fuss"? BigK HeX (talk) 23:46, 1 September 2010 (UTC)
Compared to having to play good-cop-bad-cop with a reviewer? —Jeremy (v^_^v PC/SP is a show-trial!) 00:12, 2 September 2010 (UTC)
How is that supposed to be 'any different with "playing good cop/bad cop" with the next editor that can simply hit the undo button? Pending changes most certainly does not give editors some amazing new power to review and remove an edit that they dislike. Registered users have had this power since like Day One. BigK HeX (talk) 01:46, 2 September 2010 (UTC)
You are right and you are wrong with that statement. While any user can revert an edit, a reviewer's OK is required for an edit on an FR-enabled article to go live; thus they can take advantage of this to revert edits that run counter their POV without a casual reader (or most editors) being the wiser. —Jeremy (v^_^v PC/SP is a show-trial!) 02:03, 2 September 2010 (UTC)
I have no idea what you're envisioning here. Pending changes show up in the history, just like any other edit. I don't see how editors are any less "the wiser" -- watchlists still work just fine with the system, AFAIK. For Pending Changes (you've started with the presumption of massive rejections by a widespread populace of hypothetical POV Reviewers) and are objecting about problems in making direct edits, but you're raising this objection about the rejection of direct edits in comparison to the status quo where direct editing would be COMPLETELY DISALLOWED by a page protection???
How can you object on one hand about "direct edits probably not getting through the bad cops of the Reviewer system", and yet leave us with the only alternative of COMPLETELY DISALLOWING direct edits, guaranteeing 100% that direct edits will not get through. Moreover, I strongly suspect that editors are still more than welcome to make editprotected requests, even on a Pending Changes page.
If edits "getting through" was a large concern, then there's no way that you could support Page Protection over Pending Changes, especially since it is not mutually exclusive with the editprotected process, so logically I have to conclude that there must be something else you find overwhelmingly objectionable. Difficulty with direct edits "gettng through" seems a self-defeating objection to make against Pending Changes. BigK HeX (talk) 02:56, 2 September 2010 (UTC)
IPs do not have watchlists, rarely look in the history unless they're looking for something specific for a discussion, and would have zero recourse without prior knowledge of how PC specifically works (most will not; IPs don't commonly read the Wikipedia: namespace). Suggesting that they use their non-existent watchlists is insulting.
Meanwhile, with semi-protection, a helpful box shows up explaining to them what's going on and how to correct it. Articles with FR on it have no such explanation, so such a box will actively hurt because the text - and the edit screen below - explicitly contradict each other.
My objection stems mainly from knowledge of a Wikipedian's psychology, BigK HeX. I spent three years as an administrator, and I'm pretty knowledgeable as to how the general populace (or a subsection of it) will act given a certain situation. Thus, any suggestion you make is dependent on knowledge that differs from mine; less or more complete is not the issue, but accurate is. I will make it explicitly clear - once again - I do not support FlaggedRevisions in any fashion. My admin years have taught me well enough that there is no value in FlaggedRevisions to the English Wikipedia. —Jeremy (v^_^v PC/SP is a show-trial!) 04:43, 2 September 2010 (UTC)

If it would help, I'm sure there's no problem with having the "editprotected" instructions appear on a Pending Changes page, to let editors know that the option for that is available, and you seem to suggest that those instructions are easier to follow. If you believe editors would find the editprotected process to be superior, then maybe still including the directions would resolve one of your objections. BigK HeX (talk) 03:08, 2 September 2010 (UTC)

I'm afraid that, given what I know of human behavior in regards to Wikipedia, you will never find me in support of FlaggedRevisions, BigK. Stop trying to convert me and start developing the intelligent computers that you envision as editing. —Jeremy (v^_^v PC/SP is a show-trial!) 04:43, 2 September 2010 (UTC)
I'm no trying to convert you. I'm trying to find the heart of your objection and whether there is anything that we may agree upon.
Also, I don't get your repeated reference to a Turing test. I do vaguely recall one of your edit comments or something making a comment about "automated edit filters", but Reviewers are human, so I'm not sure why you're invoking the Turing Test in this discussion with me. BigK HeX (talk) 04:56, 2 September 2010 (UTC)
The thing is, with every point you attempt to make, you seem to assume that the average editor is a machine, incapable of independent thought or emotion. Thus, you seem to assume that Wikipedia is edited only by computers that have passed Turing tests.
Meanwhile, over here in the land of Reality, all I see are human beings - several whom want to see Wikipedia succeed, some who want to see it fail, and yet others who just want to use it as a playground - editing, discussing, working. These same human beings are ultimately going to be the ones reviewing edits, and to be honest I wouldn't put that much faith in my fellow man unless humanity was free of bias, compulsion, coercion, and just plain hatred. Thus, PendingChanges is fatally flawed by the very people it has to check it. —Jeremy (v^_^v PC/SP is a show-trial!) 05:06, 2 September 2010 (UTC)

arbitrary break 4

@BigKHex: it's very common to claim that PC could replace semiprotection - you say so above. My comment about tendentious editing is just a response to that: PC can't serve in place of semiprotection in such cases. I'm glad we agree on this. It's also common to claim that there's almost no burden to enabling PC - but the inability of non-autoconfirmed editors to make edits without "permission" is a burden. Quite seriously, "a very good chance" of productive edits being accepted is terrible, when any editor who can edit an article can now have their edits applied without burden. As to the "smart" vandalism issue - [WP:BEANS]], but there's a simple method that has an extremely high chance of getting any change that looks like grammatically correct English approved on any PC protected article, because it only takes one reviewer not looking closely enough at what they have approved to accomplish it. Gavia immer (talk) 23:28, 1 September 2010 (UTC)
Gavia immer ... you are still comparing to unprotected articles. We are NOT talking about that need no protection. Currently articles are protected when vandalism becomes a problem, ensuring that non-autoconfirmed editors have a ZERO percent chance to directly edit a page. How do you think Pending Changes stacks up to a PROTECTED page? That's the important question, I think. BigK HeX (talk) 23:34, 1 September 2010 (UTC)
Badly. Protected pages are predictable: they can't be edited, except by a very few people. Even for semiprotected articles this is true; you and I can edit them, but most casual readers cannot. PC protected articles can be edited by anyone who can convince a reviewer to approve their edits, so PC is inferior at preventing vandalism. But they also don't allow good-faith edits unless you can convince a reviewer to approve them, so PC is also inferior at allowing good edits. That's why the comparison to unprotected articles is apt; if only protected and unprotected pages exist, most experienced editors will understand the tradeoffs between the two, but a great number of editors don't appear to understand the tradeoffs that PC protection requires. PC will not (not not not) magically act like unprotection for good faith editors and full protection for vandals; instead, it is inferior at both of those goals. Gavia immer (talk) 23:45, 1 September 2010 (UTC)
Your (unannounced) shift between the comparisons to unprotected pages and protected pages in the middle of your argumentation is difficult for me to follow. Let's deal with them one at a time.
Scenario 1) Let's say .. the Sting_(musician) article has become a heavy vandalism target of anonymous IPs. It is decided that semi-protection is prudent.
  1. For this article, do you agree that the use of Pending Changes instead of semi-protection would improve the chances of productive edits getting through?
  2. Do you think the number of obvious vandalism edits getting through Pending Changes would be significant? BigK HeX (talk) 23:57, 1 September 2010 (UTC)
This can't really be answered without knowing the nature of the vandalism, the frequency of the vandalism, the number of editors watching the article, the general editing climate, etc. Needless to say, semiprotection would most likely put a stop to almost all vandalism - unless it didn't, as sometimes happens. On the other hand, on the most recent day that article has edits, 50% of productive edits and 100% of vandalism reverts are by IP editors (also 100% of vandalism, true enough). Semiprotection would prevent all those edits and inform the editors that the page is protected from editing. Pending Changes would make the vandalism a crapshoot, but if the page was successfully vandalized it would have to wait on a logged-in account for the revert to stick. Based on the same article history, this could be a delay of up to three hours - not good. And the productive IP edits would become a baffling ordeal. You might think that this is superior to preventing the IP edits entirely, but I disagree; an anonymous editor who encounters a semiprotected article gets instructions on how to post to the talk page, but the same editor who encounters Pending Changes gets a confusing mess. Again, though, there's no way for this thought experiment to accurately reflect real article edits. Gavia immer (talk) 00:25, 2 September 2010 (UTC)
So you're not willing simply to concede that Pending Changes vastly improves the chances for productive edits to get through, as compared to a protected page? That doesn't seem to bode well for a good discussion, IMO. BigK HeX (talk) 02:03, 2 September 2010 (UTC)

──────────────────────────────────────────────────────────────────────────────────────────────────── No, Gavia's saying that one cannot determine whether PC would be an improvement over semi-protection because your example has too many variables to make a fair comparison. In addition, while there is a documented procedure to follow for semi-protection that an IP can follow, there is not one for PC, making it far more confusing. —Jeremy (v^_^v PC/SP is a show-trial!) 02:29, 2 September 2010 (UTC)

Thanks, but Gavia did not answer the question, and I'm not sure if he was more focused on the second question. Gavia's references to "variables" didn't really suggest an answer to that question either. I'd like to hear whether Gavia would concede that Pending Changes allows productive edits to get through, as compared to protected pages. BigK HeX (talk) 02:33, 2 September 2010 (UTC)
What do you mean, "concede"? Pending Changes allows edits through, some edits are constructive, so Pending Changes allows some constructive edits through. So likewise would randomly switching protection on and off each minute allow some constructive edits through. And of course you can construct a hypothetical situation in which you simply stipulate that things proceed as you expect, and you will find that things proceed as you expect. My objections have very little to do with these, and everything to do with the whole encyclopedia, containing protected pages, unprotected pages, constructive editors, unconstructive editors, heavily trafficked pages, unwatched pages, smart vandals, dumb vandals, and above all real people acting the way real people really act. That is the case to be considered - not hypotheticals. Gavia immer (talk) 02:53, 2 September 2010 (UTC)

Analysis by voter type

I was interested to see whether more experienced users were more likely to support PC, so I did the following analysis:

Group Anti Pro % Pro PC Straw Poll by user group.png
Anon / unsigned 3 5 63
Sysop 33 82 71
Reviewer 83 228 73
Other 78 63 45
TOTAL 197 378 66

Unique pages edited Anti Pro % Pro PC Straw Poll by pages edited.png
0-9 4 4 50
10-99 26 13 33
100-999 52 85 62
1000-9999 82 187 70
10000+ 30 84 74

Age of account Anti Pro % Pro PC Straw Poll by account age.png
<= 2004 17 32 65
2005 25 64 72

2006 43 87 67

2007 31 68 69

2008 27 58 68

2009 27 41 60

2010 24 23 49

All graphs are based on the version of 07:05, 2 September 2010. Votes have been divided into pro-PC/anti-PC purely according to which section they appeared in. Users have been identified by signature, and user information has been obtained from X!'s edit counter. – Smyth\talk 11:42, 2 September 2010 (UTC)

So, based on those graphs, the newer the user, the more likely they are to dislike PC. But I suspects the newest editors are the least likely to be involved in this poll, which makes interpretation problematic. Revcasy (talk) 12:13, 2 September 2010 (UTC)
On the other hand, the less experienced editors will not have spent much time dealing with vandalism and falsehoods in the encyclopedia, so their opinions would generally be given less weight in a poll such as this. (Unless they're experienced editors in German or another language, of course.) – Smyth\talk 12:32, 2 September 2010 (UTC)
For example, if (following Benford's law) we assume that the average person within each "by unique pages" category has done 3*10^n edits, then the first four bars in that graph represent the following edit totals: 24, 1170, 41100, 807000. – Smyth\talk 12:49, 2 September 2010 (UTC)

Thanks for doing this analysis! The message I see is that the results are dominated by reviewers. However, those in favor of PC to start with were probably the most likely to become reviewers. So we don't have a very good cross section of the editor population represented. Perhaps the notice provided on everyone's watchlist should have said something like "comment on whether PC should be expanded to thousands more articles", instead of just announcing a "straw poll on whether to continue the trial". --UncleDouggie (talk) 12:36, 2 September 2010 (UTC)

"those in favor of PC to start with were probably the most likely to become reviewers" I became a reviewer for a different reason. I didn't want my edits to be marked and needing review and then having to go an accept my own edits. The whole process was flawed from day one with nonsense like that. So by becoming a reviewer my edits are automatically accepted and I don't have to bother with extra unnecessary actions later. HumphreyW (talk) 13:15, 2 September 2010 (UTC)
That is an excellent point. Another reason to disregard this poll. To go even further, it is impossible to make an accurate, scientific poll in this manner, as I am sure everyone knows. Even using the word "poll" is somewhat misleading. This is a vote, and not a valid poll, but as has been said 1000 times, Wikipedia is not a democracy. Revcasy (talk) 12:47, 2 September 2010 (UTC)
I remember reading that certain people were made reviewers automatically (I wasn't), but I don't know what the criteria for this were. Anyone know? – Smyth\talk 12:43, 2 September 2010 (UTC)
I don't know what the criteria was either, but I got a message on my talk page saying I was given reviewer rights. I did not ask for them. ~~ GB fan ~~ 12:51, 2 September 2010 (UTC)
There was no "automatic" (read bot/script) assignment of reviewer rights, but a handful of admins tried to give reviewer access to LOTS of editors. This included most editors who had nominated an article at FAC/FLC/GAC as well as many of the editors listed at Wikipedia:Database_reports/Potential_reviewer_candidates. Karanacs (talk) 14:01, 2 September 2010 (UTC)
What the results show is that in almost every category there are more supporters for PC than opposers (and no obvious examples of Simpsons paradox). The claim that results are biased because a lot of the poll participants are reviewers is somewhat off-base, and particularly ironic in light of the argument several have made that people are unfairly weighing in without having used the feature. Anyway, reviewers, though possibly biased towards PC, are also biased towards being experienced and fairly trusted members of the community. Thanks for doing these! Ocaasi (talk) 16:57, 2 September 2010 (UTC)
Another interpretation would be that even among the most experienced editors there are 25-30% who oppose PC, and that among newer users (who we want to court and retain) opposition is even stronger. Revcasy (talk) 19:24, 2 September 2010 (UTC)

Metrics: age of account and users who provided no input

Age of account is not a reliable metric, because some users may have started out with another username or by editing anonymously for years prior. I, for instance, have been a contributor since 2004, but I did not make this account until 2007. On another note, I'd really like to see what these three graphs look like when the ~30% pro- users that provided no input other than a selection and signature are omitted, as preliminarily done here.   — C M B J   18:19, 2 September 2010 (UTC)

We have the line "Meta-discussion about the poll itself should take place on the talk page". I read this and added my comments here rather than on the polling page thus count as one who of the "Pro-users that provided no input". :-) Doc James (talk · contribs · email) 19:02, 2 September 2010 (UTC)
"Meta discussion about the poll itself should take place on the talk page" refers to discussion about the poll's methodology, not discussion relating to pending changes. Alas, it is no surprise to me that the instructions misguided you -- such is only to be expected when a polling mechanism is this equivocal. Regardless, we can take the names of everyone who commented on this page and factor them into the aforementioned methodology.   — C M B J   07:47, 3 September 2010 (UTC)

Arbitrary break

(undent) A great graph. Obviously those who have dealt with the most vandalism and created the most content wish further mechanisms to defend the accuracy and reputation of Wikipedia this just backs it up with numbers.Doc James (talk · contribs · email) 19:05, 2 September 2010 (UTC)

Nothing of the kind is "obvious". If the reviewers like it more than non-reviewers, that is only to be expected: People like having power. Septentrionalis PMAnderson 19:16, 2 September 2010 (UTC)
I was referring to the analysis of support by edit count. The power to clean up after vandalism seems more like a mop to me.Doc James (talk · contribs · email) 19:25, 2 September 2010 (UTC)
Let's see: that would be the line by number of unique pages edited? That has two results: most people who have edited less than 100 pages don't like it, which suggests that anons won't either; and a trend line among those over 100 which would disappear if half-a-dozen people switched. That's the sort of analysis that would see a trend in plum duff. Septentrionalis PMAnderson 19:42, 2 September 2010 (UTC)
Not to mention that a number of us who are (or were, as the case may be) reviewers and have tried the system have, at the very least, strong reservations about it. In fact at a glance, it looks like the tally of reviewers for and against roughly matches that of total !votes for and against. As I've said before, I see the potential for using it as a tool if it's used properly has certain fixes. But I'm not keen on how I've seen it function during this trial or how the conversation has gone down about it, during the trial or the poll. Millahnna (talk) 19:33, 2 September 2010 (UTC)
Are responses for those that answered under "Other responses" counted? I signed there and included an option because I felt I was being forced to be pigeon-holed into choices (before most of the complaints arose). I think I'd more like to see charts of how strongly people support the poll, such as which options have the most support. I appreciate the time taken to create the charts, but I find them to be reinforcing the inherent problems of polling as apposed to discussion; there are a wide variety of comments for each option and caution should be used to understand the conditions users may have listed with their !vote. —Ost (talk) 20:56, 2 September 2010 (UTC)
Entries under "other responses" were not counted, as requiring too much manual intervention. I agree with a lot of what people have said about the results being open to interpretation, but I maintain that they do show the following three things:
  • The poll is not being significantly skewed by large amounts of very new accounts.
  • Reviewers and admins like the feature a lot more than unprivileged users. I can see how this would worry people who are afraid of creating a two-tier user base which would discourage newcomers. On the other hand, this surely refutes the suggestion that people are voting for it without having experienced its dodgy UI.
  • There is a definite tendency for more experienced users to support the feature more, but this is a genuinely divisive issue with no strong consensus among any group. – Smyth\talk 21:33, 2 September 2010 (UTC)
I also say "thanks" for creating the graphs. I'm a big believer in not over-interpreting data, so the conclusion that I draw from it is: there are significantly more users who support some sort of continuation than oppose it, but there are also a lot of users who oppose, and this overall pattern remains pretty much true across user groups, number of pages edited, and age of account. --Tryptofish (talk) 21:09, 2 September 2010 (UTC)
I agree entirely with Tryptofish's conclusion. Yaris678 (talk) 11:08, 5 September 2010 (UTC)

Everyone is too fast

I have kept trying to use pending changes on the few medical pages in which it is used but everyone is so fast. I see an new pending changes seconds after it appears and try to deal with it but someone keep taking care of it a split second before me grr. Doc James (talk · contribs · email) 20:06, 2 September 2010 (UTC)

It all depends on the computer, internet connection... and unfortunately the user:(. don't feel alone. It happens to all of us.--intelati 20:09, 2 September 2010 (UTC)
You can still revert the reviewers if you disagree. But this is why PC isn't magic; it won't really do much that an active watchlist won't. Septentrionalis PMAnderson 20:30, 2 September 2010 (UTC)
I've had the opposite happen. There have been several occasions where I've checked an edit and scratched my head over whether to accept it or not, and then just left it for the next reviewer to decide. But, aside from wasting a few moments of my times, that's fine too.   Will Beback  talk  20:49, 2 September 2010 (UTC)
Except that it marks the edit as under review. Septentrionalis PMAnderson 20:54, 2 September 2010 (UTC)
Interesting. I hadn't noticed that. I just assumed a page was marked Under Review until it was accepted or rejected; I didn't know that a Reviewer viewing the page changed the status. That seems like an incomplete feature if it can't reset the status if a user "cancels" reviewing. —Ost (talk) 21:06, 2 September 2010 (UTC)
From my tests, if I simply check the diff from my watchlist, and do nothing, it does not affect the review pending status.   Will Beback  talk  03:18, 3 September 2010 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────My tests concluded the same thing, Will. It is only if you view pending changes from the "old reviewed pages" page that the status changes to under review. I can't remember which discussion page had it, but I do remember seeing a conversation about people wanting a "stop review" button for when they found an edit they weren't comfortable accepting. I think I lost track of that thread because I can't recall if there was a reason given for it being tricky to implement or if it was added to a list of possible features. I haven't looks at the summary page for the trial since this poll started so it may be there. The impact of clicking "unaccept" (per my conversation somewhere around here about my example from the atheism page) is also a weird one. If you click "unaccept" on a bad edit that was accepted by another editor (instead of simply reverting), the "accepted by reviewer name" text is removed and the edit goes back to the pending changes queue. Millahnna (talk) 12:55, 3 September 2010 (UTC)

I suppose the workaround to back out of a review is therefore to accept it and then unaccept it yourself. Not transparent, that. Septentrionalis PMAnderson 16:01, 3 September 2010 (UTC)
Huh. Hadn't thought of that. I'll guess that will work for now. If the system gets kept (either in extended trial or permanently down the line) that's definitely going to need a good looking at. </understatement> Millahnna (talk) 16:19, 3 September 2010 (UTC)

So are we going to be disenfranched or not?

I was one of the early ones who left my opinion when the poll claimed there was no need to explain our reasoning and indeed I left none. I supported restarting the poll, but this obviously isn't going to happen. So my next question is are those like me going to be disenfranched? If we are, as I've noted before, some of us (me) did comment in the previous RFC. Are even those people going to be disenfranched? Or only those who didn't leave a comment here, when they were told there was no need and didn't take part in that discussion either? Also am I right that for whoever is going to be disenfranched when they were told what they were doing was fine, no one is going to contact them to see to inform them they are going to be disenfranched for following what we were told and give them the opportunity to offer their opinions in whatever manner is now deemed necessary for them not to be disenfranched? Nil Einne (talk) 08:04, 3 September 2010 (UTC)

Why not edit your !vote, to add your comment? I don't think the poll is closed... BigK HeX (talk) 09:06, 3 September 2010 (UTC)
I agree with BigK HeX here. If you want to be certain that your voice is heard, simply edit your vote and speak your mind.   — C M B J   09:54, 3 September 2010 (UTC)
As I've said before, devaluing people's votes for failing to give an explanation is not reasonable here. This is an entirely subjective question, and "I like it" / "I don't like it" is an entirely reasonable justification. I'm sure the developers, who are the only people able to decide the outcome, will understand that. – Smyth\talk 09:15, 3 September 2010 (UTC)
"I like it" / "I don't like it" may be a reasonable answer to the presented question, but WP:POLL states that "Wikipedia decisions are not made by popular vote, but rather through discussions by reasonable people working towards consensus. Polling is only meant to facilitate discussion, and should be used with care". There's absolutely nothing unequivocal about the message of that guideline. And even if you do want to cite WP:RfA as evidence to the contrary, the actual policy there states that "[consensus] is sometimes difficult to ascertain and is not a numerical measurement". Moreover, WP:DEMOCRACY states that "[straw polls] should be used with caution, and are no more binding than any other consensus decision. Elections and votes are only endorsed for things that take place outside Wikipedia proper, such as when electing the Arbitration Committee." and WP:!VOTE states that "[e]ditors should evaluate the explanations that the participants in a straw poll offer, and should see if those explanations help to develop their own opinions or suggest compromise. In this context, a few well reasoned opinions may affect a debate much more than several unexplained votes for a different course".   — C M B J  
A lot of people are seeing this debate as a fundamental disagreement about what the nature of Wikipedia should be. That is surely a question which lies "outside Wikipedia proper". But regardless of my preferences, if I was in the place of the developers, I would have to conclude that there was no consensus here and turn the feature off. Then I would turn my attention to dealing with the one thing that there was consensus on, namely that the UI sucks. – Smyth\talk 10:56, 3 September 2010 (UTC)
While I seeing the 2/3s support would see consensus to keep it and leave it on during further development. The argument that this is about fundamental nature of Wikipedia I think is false. This is about a tool for increasing the reliability of Wikipedia and decreasing vandalism all while promoting Wikipedia's core principle of allowing everyone to edit.Doc James (talk · contribs · email) 22:32, 3 September 2010 (UTC)
Even places where we count, 66% doesn't usually cross the threshold of being considered consensus. The interesting thing is that the support continues to drop: it's currently below 65% and trending down.—Kww(talk) 22:47, 3 September 2010 (UTC)


Should the poll be closed now? I see a steady decrease in votes (26 yesterday, 12 today), meaning that the September 4 deadline was pretty spot on. And even though there is a considerable amount of protestation over the poll's format, I think it worked fairly well and in an unbiased way. EricLeb01 (Page | Talk) 00:26, 5 September 2010 (UTC)

Go for it. ☻☻☻Sithman VIII !!☻☻☻ 03:00, 5 September 2010 (UTC)
There are about a half dozen more hours before it is September 5th in all time zones on the planet. If WP goes only with UTC; then close it now. But since WP software allows date/time to be shown in local time zones, it might be kinder to wait another few hours before closing the discussion. Cheers. N2e (talk) 04:21, 5 September 2010 (UTC)
I have no problem with that. I think it would be appropriate to close the poll around, say, 09:59 (UTC)? That should cover most of Australia's midnight. EricLeb01 (Page | Talk) 04:26, 5 September 2010 (UTC)
Australia is ahead of UTC so is already a long way into September 5. UTC-10am will cover Hawaii. But perhaps you should hold off until at least UTC-11am (Samoa midnight). HumphreyW (talk) 04:39, 5 September 2010 (UTC)
Oh, shoot. I hate living in the middle of timezones. I think Hawaii would be fine, but Samoa? Well, it doesn't really matter. EricLeb01 (Page | Talk) 05:04, 5 September 2010 (UTC)
Why the rush? It is only one more hour. At the very least just to show a little respect towards Samoa. HumphreyW (talk) 05:12, 5 September 2010 (UTC)
I didn't know Samoa's second official language was English. Shows how much I know. EricLeb01 (Page | Talk) 06:02, 5 September 2010 (UTC)
It's a holiday weekend in the US. Stats may not be representative... BigK HeX (talk) 04:28, 5 September 2010 (UTC)
It's Labor Day weekend in Canada as well. That doesn't mean anything. The poll has been open for 14 days, and we're giving it another ~8 hours. I don't think consensus is going to radically change in the final two days of the poll. EricLeb01 (Page | Talk) 04:33, 5 September 2010 (UTC)

Here is an interesting video I found

Here on YouTube we have some kids making fun of Wikipedia.[2] Just removed the text in question here today [3]. In other words we have a serious problem.Doc James (talk · contribs · email) 06:33, 5 September 2010 (UTC)

So? flagged revisions would have done nothing to help and in the mean time would have killed millions of useful edits.©Geni 09:59, 5 September 2010 (UTC)
If PC was in place when such a change was originally made, it could have conceivably caught it because it was a series of two IP edits. However, if it was caught, the vandals would have just created an auto-confirmed account and added it back. PC as it stands won't fix this problem. There's no substitute for editors asking why there are two red links in the article and figuring out the problem. Also, we need at least one person to adopt every article and watch it or periodically run a diff back to the last version they edited to see what has happened to it. I also just fixed this three year old bad edit in the same article. I said it was vandalism in my comment because the same editor made another clear vandal edit on the article. --UncleDouggie (talk) 13:08, 5 September 2010 (UTC)


I would like to propose a trial to determine the effectiveness of PC. Would appreciate feedback on the design here Doc James (talk · contribs · email) 04:25, 6 September 2010 (UTC)

"Pending changes is more difficult!" Please explain how

One common theme of the !vote comments to oppose Pending Changes is that it will either be more confusing or will vastly complicate editing. I submit that, obviously, compared to a perfect article this is the case. However, many in the community seem to miss the huge point that Pending Changes has no relevance to perfect articles. Pending Changes is meant as a tool to deny vandals recognition, and thus will find employ on pages subject to such vandalism. Editors should be keeping in mind that, currently, the only effective article-level tool for that job is Page Protection. I'd submit that the constant comparisons of Pending Changes to unprotected pages is useless, as Pending Changes is a tool that would be an alternative for Protecting pages. Obviously Protected Pages are a true nightmare to any idea of convenient editing. I've offered this comparison:

A) Pending Changes requires:
  1. a simple click from any long-term editor.
B) Page Protection requires:
  1. an editor to figure out how to post a talk page section,
  2. to figure out how to get consensus on an edit,
  3. many times has to actually achieve consensus(!!!!), and
  4. to then realize that {{editprotected}} is the correct Bat Signal, AND then
  5. requires admin intervention of all things.

I've received some response on this question, but I ask for more input from editors who contend that Pending Changes makes things more complicated. Specifically, I am curious whether many of the editors are commenting with the knowledge that Page Protection is our current method, and how inconvenient that process can often be. BigK HeX (talk) 05:06, 2 September 2010 (UTC)

Semiprotection is far easier to punish for with regards to POV-pushing admins, Defenders of the Holy Wiki whom don't have a good enough reason to prot, and imbecilic Hugglers requesting bogus semi-prots. PC has no such recourse and is far too reliant on the hostile psychology of the community to work. —Jeremy (v^_^v PC/SP is a show-trial!) 05:13, 2 September 2010 (UTC)
Pending Changes and Semi-protection aren't mutually exclusive. So, a page that is subject to POV-pushing admins, Defenders, and Hugglers can still be semi-protected. Other than maybe adding the extra step of trying PC before upgrading to a semi-prot, I'm not seeing this as any insurmountable problem with PC. BigK HeX (talk) 05:20, 2 September 2010 (UTC)
You're wrong. PC and semiprotection are mutually exclusive because Pending Changes can aggravate vandalism issues on a page; semiprotection cannot.Jeremy (v^_^v PC/SP is a show-trial!) 05:23, 2 September 2010 (UTC)
You said basically the exact same thing above. The comparison is meaningless; PC accomplishes none of the things semi-protection does, it's not meant for the same pages, and many pages that are not semi-protected would almost certainly be PC-protected. --Yair rand (talk) 05:14, 2 September 2010 (UTC)
PC accomplishes just about exactly the same delay-until-approval restriction to editing as a page protection with the only difference in function being that Pending Changes's equivalent to an {{editprotected}} request can be serviced by any long-term editor, instead of just admins. BigK HeX (talk) 05:29, 2 September 2010 (UTC)
I agree with Big K many vandals get immense satisfaction from having insults regarding people they do not like potentially appear to million for a brief period of time or have their name listed as the president of wherever briefly. Pending changes will deny them this and thus will probably decrease vandalism over time.Doc James (talk · contribs · email) 05:15, 2 September 2010 (UTC)
How many more vandals will get complete satisfaction from sneaking in a vandal edit past a reviewer or overloading the PC feed? Murphy's Law has always functioned perfectly in the face of Wikipedia's antivandal measures - semiprotection, Checkuser, blocks, edit filters, Pending Changes... —Jeremy (v^_^v PC/SP is a show-trial!) 05:20, 2 September 2010 (UTC)
No one claims that these mechanism are or will be perfect but no reasonable person would say that they are not somewhat useful. Wikipedia still requires human being behind it to verify the actions of the automated processes.Doc James (talk · contribs · email) 05:26, 2 September 2010 (UTC)
No reasonable person would say that the benefits are worth the rather severe logical fallacy it operates on, which is that all Wikipedia editors are inherently both neutral and knowledgeable about the subject matter. —Jeremy (v^_^v PC/SP is a show-trial!) 05:28, 2 September 2010 (UTC)
Yet nearly 2/3 of Wikipedia editors do consider the benefits greater than the risk for implementing some for of pending changes.Doc James (talk · contribs · email) 05:32, 2 September 2010 (UTC)
Nearly 2/3 of Wikipedia editors on this page. There's multitudes of IP editors, users, and administrators whom have not commented. —Jeremy (v^_^v PC/SP is a show-trial!) 05:33, 2 September 2010 (UTC)
I stand corrected :-) Doc James (talk · contribs · email) 05:34, 2 September 2010 (UTC)
RE: "fallacy...Wikipedia editors are inherently both neutral and knowledgeable"
It's hard to understand how you can be so cynical about the capabilities of Wikipedia editors, and yet so vehemently oppose a tool that is meant to curb blatant idiocy on pages that are subject to being targeted for idiocy. I admit that your stance on the matter is hard for me to grasp. BigK HeX (talk) 05:40, 2 September 2010 (UTC)
Because I know enough about the psychology en generale of each user group that I can tgell you right now that you're missing a part of the formula: "A tool that is meant to curb blatant idiocy on pages that are subject to being targeted for idiocy by running the edit through several more idiots." Bluntly, the average reviewer will be no less biased, no less prejudiced, no less destructive than the average administrator, the average Arbitrator, or the average IP. —Jeremy (v^_^v PC/SP is a show-trial!) 05:43, 2 September 2010 (UTC)
Personally, I think that may just be the crucial fallacy in your objection. To contend that a long-term editor [4000+ edits, or whatever criteria for trustworthiness we settle upon] is "just as...destructive" as the average IP on a page subject to vandalism seems to be a pretty dubious proposition. BigK HeX (talk) 05:50, 2 September 2010 (UTC)
Long-term edit count doesn't tell you anything about an editor's overall behavior. A user could have 150K edits and still be unworthy of Reviewer tools because he refuses to disengage from a sociopolitical hotbed (Isr/Pal, Eastern Europe, pseudoscience). Thus, my saying stands true, compared to the logical fallacy supporters like you seem to support (that all editors are neutral and unbiased). —Jeremy (v^_^v PC/SP is a show-trial!) 22:39, 7 September 2010 (UTC)
I see what you mean but I don't think it will decrease vandalism over time that much. For whatever it's worth, I do see potential merit in the tool if used correctly (accepting false information because it's not blatant vandalism has got to go) and if the many interface problems are solved. But it's really not that hard of a tool to get around if you really wanted to. Vandals register "vandal only accounts" all the time, and a series of quick "good edits" and you're above the threshold for PC to stop your edits (that or I'm understanding this all wrong). For that matter, on more than one occasion I saw fairly blatant vandalism flat out accepted by others with reviewer rights. Usually this was of the POV variety (and I can only think of one direct example off the top of my head) but I found that my time of trying out reviewer status was spent cleaning up other reviewers' acceptance of "deliberate factual errors" type edits (I was tracking several IPs that were persistently changing birthdays to be wrong and finding their edits accepted wholesale) and blatant vandalism that was topical in nature ("of course there's a god" on atheism pages). Personally, when I reviewed a page on a topic I wasn't overly familiar with I would either leave the edit (not accept it) for someone else to look at or actually *gasp* research the topic to make sure I could verify accuracy before clicking accept. Usually, by the time I had verified, someone had come along and accepted it anyway meaning it had to now be reverted. This was particularly rampant on the football (soccer to my fellow yanks) pages that were used in the PC trial, some of which were normally low traffic and therefore ideal for the concept of PC in many minds. The concerns of those who feel that reviewers were adding more work to the editors who had pages on their watchlist is an extremely valid one. Unless the guidelines clear these sorts of problems up considerably or we can guarantee that every project/category of pages has a reviewer or two, I foresee a lot more problems. Millahnna (talk) 05:36, 2 September 2010 (UTC)
We have nearly four hundred people who support PC so hopefully as time goes on more editors will get involved with reviewing content. Problems like these will hopefully decrease as people become more familiar with PCs.Doc James (talk · contribs · email) 05:46, 2 September 2010 (UTC)
Trust me: they will only worsen as PC radicalizes them. —Jeremy (v^_^v PC/SP is a show-trial!) 05:47, 2 September 2010 (UTC)
No doubt some will get better. Others will likely use it for POV pushes. Such is the nature of the bell curve of human behavior. Millahnna (talk) 05:51, 2 September 2010 (UTC)
Would a 3-strikes rule be satisfactory, Millahnna? Perhaps a page can be setup similar to the 3RR noticeboard or something, and if an editor can present more than 2 reviewed edits in one month which display a flagrant misuse of the tool, an admin would review the report, and if there is misuse, then the Reviewer is temporarily relieved of the privilege. Maybe 72 hours for the first report, 2 weeks for the 2nd report, etc. in an escalating fashion. BigK HeX (talk) 06:06, 2 September 2010 (UTC)
That would help, I think, or something along those lines. I won't lie; this poll, related conversation, and the various other conversations on PC around the site has soured me to it. Yet I still have not asked to have my reviewer rights revoked. Since I've reverted so many incorrect "accepted edits", I can't seem to bring myself to. Yet I rarely look at the actual "old reviewed pages" page anymore. Millahnna (talk) 06:18, 2 September 2010 (UTC)
The atheism example I noted above. I didn't check to see who accepted it at the time but a quick user history check would have told them this was a vandal only account and not to accept. They accepted anyway. I'd have to dig way back into my own contributions to find the "deliberate factual errors" and soccer examples (and I'm so not interested in doing so). Many cases of players switching teams before a sourced confirmation (on BLPs and not just team pages) were approved without even a cursory google search (I don't know squat all about soccer and it was pretty easy for me to confirm these edits were wishful fancruft at the time). The birthday changing vandal had a slew of warnings on his page for deliberate factual errors and wasn't even editing birthdays correctly and almost every edit of his I cam across was accepted at the time. Millahnna (talk) 05:50, 2 September 2010 (UTC)
The edit was added here] and removed by your here. You appear to be the pending changes reviewer and properly keep this edit from going live. Looks like a success for PC! Doc James (talk · contribs · email) 05:56, 2 September 2010 (UTC)
Further more looking through the last few edits at atheism one good edit was passed [4] and all episodes of vandalism were prevented from appearing.Doc James (talk · contribs · email) 06:04, 2 September 2010 (UTC)
Given that his edit was actually accepted and went live before I came along to unaccept and revert I would disagree. This is not the only instance, as I've said, where bad edits were accepted. I'm still kicking myself for clicking unaccept before I checked the name of the reviewer who accepted. Millahnna (talk) 06:12, 2 September 2010 (UTC)
I would very strongly support that all current reviewers have to sign a project page thoroughly explaining their responsibilities, or that the Reviewers lose the privilege and have to attempt to regain it later. I pretty much assume that the current crop of Reviewers is this way only so that we could have a population for testing the system. BigK HeX (talk) 05:55, 2 September 2010 (UTC)
I am suprised and shocked that Wikipedia is adding restrictions on who can edit. We already have to many regulations on editing. I vote (If PC passes) that we change the motto from "The Free Encyclopedia" to the "The Dead Encyclopedia".....--Gniniv (talk) 06:04, 2 September 2010 (UTC)
If you look at the atheism example above people are editing. The good edit got through. The vandalism did not. If this does not pass maybe we should change it to "The Encyclopedia Anyone Can Vandalise" Sorry I could not help myself :-) Doc James (talk · contribs · email) 06:07, 2 September 2010 (UTC)
Page protection already puts a hard block in place on who can edit. We're actually here trying to determine if this tool is an appropriate substitute. I contend that it is far more accommodating than the hard block of page protection. BigK HeX (talk) 06:08, 2 September 2010 (UTC)
I would strongly agree with that proposal if PC is kept. In my time of using it, and as I've said, I do see the potential merit of it. But the guidelines and such will have to be clear to the point of being explicit point by point rules to keep it from causing more harm than good. Millahnna (talk) 06:12, 2 September 2010 (UTC)
We already have a system that protects against vandalism (WP:Confirmed users); without the need for Page Protection. All PC does is add another slice to the already thick sandwich of bureacracy on Wikipedia. PC is also against the principle of WP:Neutrality because it limits edits to an elite group of reviewers who have a far more limited perspective than all the Wikipedia community...--Gniniv (talk) 06:14, 2 September 2010 (UTC)
God dammit. I swear if I see "PC goes against the anyone can edit philosophy" one more time I bad things. No, it does NOT. ANYONE CAN STILL EDIT WITH PENDING CHANGES ACTIVE, UNLIKE WITH (SEMI-)PROTECTION. IT IS MERELY IF THEY WILL BE IMMEDIATELY VISIBLE. IF YOU CANNOT UNDERSTAND THIS, PLEASE READ THE PROCESS ONCE MORE UNTIL YOU DO. Sigh. ♫ Melodia Chaconne ♫ (talk) 06:18, 2 September 2010 (UTC)
Apparently you do not know what to "edit" means. To my understanding, it is not the same thing as "submitting a possibly used contribution to be reviewed by someone else and then judged on whether it is usable or not". --Yair rand (talk) 06:26, 2 September 2010 (UTC)
Obviously, we do use Page Protection. On pages where the WP:Confirmed users restrictions are enough, Pending Changes isn't going to be a big tool to use. But, if your objection is about "elite groups" then pretty clearly Pending Changes is a huge step up from Protected Pages, where edits can only be approved by Admins. BigK HeX (talk) 06:19, 2 September 2010 (UTC)

Based on the conversation currently happening on my talk page, it is very clear to me that people don't understand what happens if you click "unaccept" on an edit that has been accepted by another reviewer. I have had to do this multiple times in this trial. In every instance, when I clicked "unaccept" the "accepted by 'Reviewer name'" text was removed from the edit and the edit went back into the pending changes queue. The atheism example I gave above WAS accepted by another reviewer before I came and "unaccepted" it. I then reverted so that it was no longer needlessly in the queue. This is documented in the many discussions that have occurred about needing a "do not accept button" that clears edits from the queue of pending changes. This is one area that needs to be addressed in terms of clarity and increased difficulty. Millahnna (talk) 06:27, 2 September 2010 (UTC)

My main point is that this policy restricts Wikipedia to the point where it isn't really The Free Encyclopedia anymore. Besides, approving this policy is like having politicians vote themselves a raise. THE PEOPLE CAN'T VOTE!--Gniniv (talk) 06:29, 2 September 2010 (UTC)
"Anymore"? The policy of restricting edits on a page troubled by vandalism has been in effect for a long time. BigK HeX (talk) 06:32, 2 September 2010 (UTC)

(undent)1% of editors contribute 50% of the content on Wikipedia. Thus yes we seem to already have different classes of editors. We all still have equal potential to contribute.Doc James (talk · contribs · email) 06:32, 2 September 2010 (UTC)

AND THAT SHOULD REMAIN. Just because 1% of wikipedia editors contribute 50% of the edits by no means justifies restricting the other 99%'s right to edit!--Gniniv (talk) 06:38, 2 September 2010 (UTC)
Agreed. Some of our best spell checkers are IPs. Millahnna (talk) 06:45, 2 September 2010 (UTC)
Yes and pending changes will not significantly change this.Doc James (talk · contribs · email) 06:46, 2 September 2010 (UTC)
Any change at all from freedom is to much...--Gniniv (talk) 07:26, 2 September 2010 (UTC)
And do you agree that Pending Changes is MORE freedom than a protected page? BigK HeX (talk) 07:45, 2 September 2010 (UTC)
I know that Pending Changes is less freedom than WP:Right to Edit. I also think that Page Protection is (in many cases) against the original spirit of Wikipedia. The question that will be determined in this poll is how much freedom editors are willing to give up in return for reviewer's privilege.--Gniniv (talk) 07:56, 2 September 2010 (UTC)
Somewhat conspicuously, you did not answer my question. Is it your belief that Pending Changes grants MORE freedom to edit than a protected page? BigK HeX (talk) 09:01, 2 September 2010 (UTC)
I highly suggest that all editors support the below proposal as the best compromise. In answer to your question, Pending Changes is less regulated than Page Protection (Obviously). However, that is not the point, we need to lower all regulation of Wikipedia to the bare minimum balance required to prevent vandals and provide for universal editing. SEE BELOW FOR THE INGENIOUS COMPROMISE TO OUR DILEMMA!--Gniniv (talk) 09:10, 2 September 2010 (UTC)
Thank you for the honest response. BigK HeX (talk) 09:11, 2 September 2010 (UTC)

Change PC to an Edit Filter Option

I propose to not have any articles designated exclusively for Pending Changes. Instead, start with the process as it was before the PC trial began and merely add an option to each filter in the Edit Filter to invoke PC for an edit instead of warn or disallow. In place of blanket PC protection, we would be free to construct elaborate regular expressions in an attempt to flag only high-risk changes for PC. It will never be perfect, but that doesn't mean that we shouldn't try to improve the current system. We need to continually adapt to vandal countermeasures without impacting good edits. Safeguards could be put in place such as firing limits and community review of PC filters. Semi-protection would still remain an option when needed.

It would of course be possible to construct a rule invoking PC on all anonymous BLP edits, but this would not be the intent of the system. Note that all BLP IP edits could be disallowed right now, subject to the edit filter firing limits.

I believe that adding the PC option to the edit filter would reduce the number of cases where semi-protection is needed, reduce the reviewing workload, and most importantly disrupt the vandal workflow on all articles in a way that no other current or proposed method can achieve. The trial feedback indicated that vandals tended to migrate from PC protected articles to those not protected by PC. This shouldn't have been a surprise to anyone, but it bodes potential doom for an expanded PC trial. It simply is not practical to use full PC on all articles as other editions have done. The English Wikipedia is the largest and its visibility will always attract a high degree of vandalism. We need to improve our automated tools and using the edit filter to flag high risk changes for mandatory manual review is a step in the right direction.

The current IP vandalism workflow seen frequently in the Edit Filter log is:

  1. Make obnoxious edit
  2. Reeceive disallow response from the Edit Filter
  3. Slightly tone down the edit
  4. Go to step 2 until just under the limit for triggering any filters that have an action of "disallow"
  5. Receive warning from the Edit Filter
  6. Click Save
  7. Get reverted (hopefully)

It would be much more valuable if step 5 was for the Edit Filter to just accept the edit as a PC. How would this be different from blindly applying PC to every article? Because many useful edits never trigger the Edit Filter at all and will not be subjected to PC overhead/bureaucracy/POV pushing!

Example: The Edit Filter should be analyzed as a complete system, not a list of individual filters. Many edits trigger multiple filters as shown here. Note that the edit at :48 was saved and subsequently rejected by a PC reviewer. This is not an endorsement of blanket PC. Quite the contrary: If filter 61 had an action of PC, the bad edit would have been caught without blindly applying PC to all IP edits. While that might not be the perfect use of filter 61 as it now stands, hopefully you get the idea of what is possible. Looking only at the article edit history in this case gives a distorted view of what vandals are attempting and the impact of PC.

It would be nice to have some stats on how many users ignore the edit filter warnings entirely and just click on save anyway. Also, what percentage of edits that triggered warnings are subsequently reverted? Having these stats will not change my proposal, but they will give us an idea of how much work we need to do on the filter set to make effective use of the new capability. --UncleDouggie (talk) 08:47, 2 September 2010 (UTC)

  1. I Agree: I oppose PC as is, but I support your idea and encourage Option 1 voters to consider this compromise if it is proposed in the future.....--Gniniv (talk) 08:57, 2 September 2010 (UTC)
  2. This makes a pretty good suggestion, if it were maybe a new Level-0 Pending Changes level or something. I would like to see the current Level-1 and Level-2 systems available as options just the same though. BigK HeX (talk) 09:19, 2 September 2010 (UTC)
  3. I like this. I don't know enough about the technical back end of the edit filter to know how tricky this would be but on paper (erm screen?) it looks good. Millahnna (talk) 11:20, 2 September 2010 (UTC)
  4. I Agree I think this is a great idea. And is something we should work towards whether or not PC is accepted. It would probably take some time to set up and customize. Information we gather from a longer trial of pending changes would be useful in it's implementation. Would allow the entire encyclopedia to be covered.Doc James (talk · contribs · email) 18:52, 2 September 2010 (UTC)
  5. Like others above me, I think this is an excellent idea, regardless of whether PC is implemented in a fashion as carried out during the trial. ialsoagree (talk) 20:54, 2 September 2010 (UTC)
  6. An interesting idea. Possibly the next proposal in this field. Septentrionalis PMAnderson 20:56, 2 September 2010 (UTC)

I had suggested something like that in 2009, Wikipedia:Deferred revisions, but there are technical limitations to overcome, and to work well it may require some sort of patrolled revisions, to give reference revisions to the filter. Cenarium (talk) 00:59, 6 September 2010 (UTC)

Ah, thanks for blazing this trail already. I see that it's not so simple, as evidenced by this statement in your proposal: "For reviewers, when the right revision in a diff is deferred, there is in this case the additional 'defer' level (before 'unreviewed'), and the revision can be brought back to unreviewed state if it is a false positive (if the page is not reviewable, there is only defer and unreviewed, if semi flag protected, there is also review)." Please don't try to explain it to me; I've read it four times and I can't take anymore. :-)
Let's step outside of this box in which there are no acceptable solutions for just a moment. There are many automatic and manual ways that edits currently get reverted. The problem is that bad versions can be visible for a time. Often this time is short. However, with the overload of things to look at, the time to revert can be also be very long. Immediate visibility to new users is already an illusion. Their shinny new edit may have been reverted by ClueBot or DASHBot before their browser refresh completes.
It would be nice if all mainspace edits were not publicly visible for two minutes by default. In that time, a bot or trusted user could flag the change as "deferred" pending detailed manual review. This would allow us to run bots of arbitrary complexity to identify suspect changes. Bots could also remove the defer flag on a change in cases when a later revert restores a reviewed version of the page. This will prevent unneeded manual review of changes. It would be best if the software would serve up the most current page to any IP user that has edited the page in the last 2 minutes to retain the illusion of visibility.
Having this capability would enable ClueBot to defer highly suspect changes that would otherwise violate its 1RR policy. When there are fast edits to a page by untrusted users, the visibility of all unapproved revisions should be delayed until two minutes after the latest change. This will prevent bad edits from being displayed just because a revert hasn't timed out yet or been approved. There have been several times that I wished there was a delay on my own edits. It's rather unnerving to refresh a Google search and see the edit I made 30 seconds earlier, which I now realize was horribly botched somehow. —UncleDouggie (talk) 15:12, 6 September 2010 (UTC)
I spent a couple of hours late last night with Huggle when there weren't many other patrollers online so I could easily see the majority of changes in real-time. I'm more convinced than ever that a 2 minute delay would save us an enormous amount of grief. I was reverting repeating vandalism 5 seconds after it happened. Of course, the backlog can be bigger at other times. But if we had a delay in non-autoconfirmed edits being publicly visible, we probably would have had almost no vandalism publicly visible last night. Furthermore, there was a substantial amount of chat style editing between kids. Perhaps regular chat programs are blocked on their computers, or they don't know how to use anything else. Introducing a 2 minute delay would put a damper on all the back and forth, which would mean fewer suspect changes for us to review. —UncleDouggie (talk) 05:25, 8 September 2010 (UTC)

With regard to the Straw Poll - Counting of !votes

Can someone please clarify for me how the votes in the Straw Poll are likely to be counted? From offwiki discussion last night, my understanding is that the votes to keep in some form (Options 2 3 & 4) are likely to be counted together. If this is so, it seems wrong. Surely since there were 4 options, the 4 of them should be counted separately and the totals worked out from there. Also, I don't see how people can get away with supporting more than 1 options, since that also unbalances the total votes. Anyone got any ideas on how it's gonna be done? BarkingFish Talk to me | My contributions 16:39, 5 September 2010 (UTC)

As I've said before .. pretty obviously, the poll could've easily been designed as a simple Support/Oppose request. Seems there'd be a tiny bit of absurdity in penalizing supporters because their poll options were more elaborative. BigK HeX (talk) 16:46, 5 September 2010 (UTC)
Maybe just to even things up a little there should have been three close options also. 1) Close. 2) Close with prejudice. 3) Close with malice. But seriously, I have no problem with summing all the keep !votes as one group. But only as long as each !voter is counted once only regardless of how many keep options they !voted for. However, as was made clear shortly after the voting (it was initially a real vote, not a !vote) started, it is not a vote, it is a poll to gauge consensus, so regardless of how one wants to sum the !votes, the actual figures, it seems, do not matter. HumphreyW (talk) 16:55, 5 September 2010 (UTC)

I'm still waiting for this mythical neutral administrator to decide what happens next. Yaris678 (talk) 18:17, 5 September 2010 (UTC)

I've asked at WP:AN for one to come here (a real one, not a mythical one!). --Tryptofish (talk) 18:30, 5 September 2010 (UTC)
It looks like they are mythical after all! :-D --Tryptofish (talk) 17:36, 6 September 2010 (UTC)
I saw the request, had a look at this matter, & wish I hadn't commented on this straw poll before. Because since I decided no one was listening to my concerns & left, it's clear to me that the situation really hasn't changed. On one hand, a non-binding poll -- which is what a "straw poll" is -- shows a majority of people would like to see the "Pending changes" experiment to continue in some form. However -- & to me this is the more important consideration -- no one has explained in satisfactory detail whether this experiment was a success or failure. We just have a lot of people arguing for or against this experiment from their emotions, not from any sort of measurable results. Now if this experiment fixed some problem we have with articles on Wikipedia, I would at the least be willing to continue the experiment -- more likely, I'd be happy to endorse applying it to more articles. But no one has explained just what problem(s) it fixes. The sense I get from the comments of the supporters for "Pending changes" is that they feel that because the current situation is broken & something must be done, since this is something we should do it. (That same argument could be used to justify banning each & every one of us & protecting every page of Wikipedia from further changes.) At this point, the best resolution to this matter would be to just let the experiment end, spend the time otherwise used making the experiment work figuring out what worked & what didn't -- & why -- then trying it again under new conditions. -- llywrch (talk) 23:15, 5 September 2010 (UTC)
The situation has changed since you last commented (I think on August 25). What was a 3-ring circus has become at least a 6-ring circus. I'm sure most people mean well, but from what I can tell, the comments are mostly going in circles. Of course, honestly, I don't read the comments much anymore because it's not worth it, but they keep coming. My only hope is that someone will put this discussion out of its misery.--Bbb23 (talk) 00:14, 6 September 2010 (UTC)

Straw poll closed

The poll has closed with 65.3% (408) supporting and 34.7% (217) not supporting pending changes. I guess the question is were do we go from here? There have been requests for data on PCs benefit / harm. PC has been used to un semi protect some articles thus making more of Wikipedia free for anyone to edit. PC has also been used to decrease much vandalism going live. Some editors dealing with the system have however found it slow and confusing. Do we have any further data?Doc James (talk · contribs · email) 01:24, 6 September 2010 (UTC)

Doc James, can you break that down into the individual options since they were all separate? I'd like to see how far the Supports for closure outweighed the individual totals. As I said, all 4 options were given separately, thus they should all be counted separately, not just clumped together to make it appear that people want this farce to continue. BarkingFish 03:01, 6 September 2010 (UTC)
There is clearly no consensus for starting another trial of pending changes. As the original two-month trial is over, the current implementation of PC should be removed. More discussion may lead to a further proposal, but the original idea of using PC as is is done with. --Yair rand (talk) 03:07, 6 September 2010 (UTC)
There is not support for getting rid of pending changes thus it should stay IMO.Doc James (talk · contribs · email) 03:09, 6 September 2010 (UTC)
That's not how it works. There was support for a limited trial, and the poll was to determine whether Pending Changes would be extended after the trial was complete. There isn't a broad enough consensus to continue the trial, so it shouldn't continue. Gavia immer (talk) 03:30, 6 September 2010 (UTC)
(edit conflict) Incorrect. There is no consensus that it should stay; the ratio of keep to remove was about 2:1, or about two-thirds (2/3; 67%) in favor of keeping it and one-third (1/3; 33%) in favor of removing it. One-third is a significant objection to keeping the feature and, as there is neither consensus to keep or to remove, pending changes should be removed, as "no consensus" defaults to the status quo—remove. (Note: I voted to remove PC.) fetch·comms 03:32, 6 September 2010 (UTC)

Pretty clearly, with two-thirds of support shown among participants, those voices shouldn't be disregarded, and some form of accommodation should be made for those in support of the Pending Changes system. BigK HeX (talk) 03:50, 6 September 2010 (UTC)

Right now we should deal with the issue that the PC trial has gone far past its end date without being stopped. Perhaps at some later time a compromise proposal may receive consensus, but the existing trial is over, and no continuation of PC has consensus. Admins should be removing PC-protection from articles. --Yair rand (talk) 04:27, 6 September 2010 (UTC)
Against the wishes of 2/3rds of those who commented? Data has not been presented that PC is harmful or does not work. We have developed a little experience with it and it is now time for a more in depth analysis. I have proposed a further trial below.Doc James (talk · contribs · email) 04:44, 6 September 2010 (UTC)
As I said, another proposal may receive consensus at a later time, but the current trial is over. --Yair rand (talk) 04:49, 6 September 2010 (UTC)
Although I personally support keeping the feature turned on, I agree. Back when the trial was initially approved, there actually was a consensus, and it was that the trial would only become permanent if there was consensus to keep it. Such a consensus has not formed, and you can't retrospectively reverse the burden of proof. – Smyth\talk 08:42, 6 September 2010 (UTC)

The original wording of the straw poll stated Consensus will be judged by an Administrator on the basis of a simple head count (as there are no policy decisions to be considered) on the basis that 6 - 4 is a minimum consensus agreement. A no-consensus outcome will default to close.. That 60% threshold was removed after the vote had started by someone who later went on to oppose the poll itself, and never restored. Soap 15:37, 6 September 2010 (UTC)

Excellent so it is clear that we will keep it than. Even the Quebec referendum only required 50% plus one. I still believe it is most fair if we generate better quantitative data and than give everyone a chance to review things.Doc James (talk · contribs · email) 15:42, 6 September 2010 (UTC)
Not that is matters now, but the 6-4 thing was intended to mean a 2/3 (66.67%) super majority. This was clarified early on, and is in the archived page now. Anyhow, as already stated many many times, the figures don't matter, just ignore them. It is only consensus that matters. HumphreyW (talk) 15:48, 6 September 2010 (UTC)
That original wording was removed because people quite rightly objected that the creator of the poll does not have the ability to unilaterally declare what consensus is. – Smyth\talk 19:05, 6 September 2010 (UTC)

It looks to me like there is something approaching consensus about two things:

  1. There were significant methodological and procedural issues about the poll, so it has to be interpreted cautiously.
  2. If anything with PC is to be implemented, there has to be, first, improvement in the software (development) and in the procedures (policies/guidelines) before it can properly be considered. In fact, a significant number of the option 1 !voters said they might support if these things were improved, and a very large number of the keep !voters nonetheless expressed reservations about these things.

Now it seems to me that the argument that implementation requires 66.67% support but since it only got 65.3% support, it follows that we have consensus for option 1 is, with all due respect, one of the most muddle-headed arguments I have ever seen. There is no consensus for option 1. There is also no consensus for options 2, 3, or 4 either. We need to address the development and guidelines issues as editors have been discussing here, and that's the work that needs to come next. As for the argument about whether to leave PC on pages where it is now (a la option 2) or to remove it from those pages, I don't care. It's no big deal. Nothing is irreversible. If it's left, it can still be improved in due course, or it can be removed later. If it's removed now, it can still be improved and re-applied later. Whatever. Let's work on the improvements, and if and when they materialize, all the previous options, including option 1, will still be on the table. --Tryptofish (talk) 17:56, 6 September 2010 (UTC)

It should be removed now and a new trial proposed later after the technical issues raised have been worked out, etc. We cannot let a "trial" go on forever; that's not a trial, it's inconsistency and indecision. We don't leave AfDs open forever if there is no consensus, nor should we leave this trial on until enough people support it. fetch·comms 19:44, 6 September 2010 (UTC)
Sure sounds, here and below, like you are saying there is consensus for option 1. --Tryptofish (talk) 19:48, 6 September 2010 (UTC)

A way forward?

So a majority of people want to keep PC but a significant number of people don’t. There is also some disagreement about what lessons we can learn from the trial and how to interpret the straw poll. Here is what I suggest would make sense in the circumstances.

  • The Pending Changes Protection extension to the MediaWiki software should be left running on Wikipedia (for now).
  • The administrators should be encouraged to remove PC protection from all pages on Wikipedia, except those under WP:Pending changes/Testing. Administrators should use their judgement as to whether semi-protection or no protection is most appropriate.
  • A number of working groups should be set up to look at the issues raised from this trial. Perhaps under the headings
    • policy (e.g. reviewer guidelines, protection policy, avoiding conflict over reviews…)
    • software (e.g. requests for a “stop reviewing” button).
    • terminology and iconography (The pictures and words we use to represent pending changes. This area straddles both software and policy).
  • When a number of the issues have been resolved, a new trial will be considered.
    • The trial will use the new policies, software features etc. (to the extent that they have been developed – some issues may remain and some software may take a long time to develop).
    • Much greater consideration will be given to the methodology of the new trial to enable the collection of meaningful statistics. For example - Does the availability of PC protection reduce the use of semi-protection? How much does PC protection dissuade anonymous users from contributing?
  • After the trial, sufficient time should be given for the analysis of the results. During that period, pages should be converted to semi-protection or no protection.
  • After the analysis of the results, another straw poll should be held. More discussion should be held prior to the straw poll, to gain consensus on options, methodology etc.

Yaris678 (talk) 09:35, 6 September 2010 (UTC)

It should only take a week to analyze the data from a trial such as I have proposed. Thus leaving what ever is running running is not a big issue.Doc James (talk · contribs · email) 15:44, 6 September 2010 (UTC)
We have taken several weeks to look at the current trial and we still haven't worked out what to do. I also think some of the problems could have been avoided if we had rushed less. Let's not make the same mistake again. Yaris678 (talk) 16:33, 6 September 2010 (UTC)

I absolutely, vehemently oppose the concept of "let's just keep the trial running for now." It was a trial of finite length, with a clear end point. The clear end point is in the past. There is no consensus to continue the trial. It has to stop. If people want to restart the trial later after getting a consensus, fine. If people want to do a different trial later after getting a consensus, fine. Letting a finite trial turn into an indefinite trial without consensus to do so is unacceptable.—Kww(talk) 17:54, 6 September 2010 (UTC)

I agree with Kww. This trial was supposed to be two months. There is no consensus to continue it for any length; it needs to end. I am not opposed to starting a new discussion for a new trial later, but this current trial should have ended weeks ago. fetch·comms 19:36, 6 September 2010 (UTC)
So, Kww, Fetchcomms, do you support my idea? It is basically to end the trial now and then have another one when some things have been ironed out. The Pending Changes software will be left running in the meantime but that will just be a technicality since all pages will have PC protection removed. The reason for doing that is three-fold. (1) The developers have expressed a preference for not turning it off then on again. (2) We don't have to go through the hassle of re-assigning reviewer rights for a second trial (3) It means we can do some testing of the feature to aid discussion of improvements. This testing would be done in specific places (such as at WP:Pending changes/Testing) and not on articles or any other "normal" page. Yaris678 (talk) 20:29, 6 September 2010 (UTC)
I see no reason to turn the feature off in the software, so long as pending changes is removed from all articles and is not applied to any new ones until consensus to do so is reached.—Kww(talk) 00:25, 7 September 2010 (UTC)
I agree. It doesn't matter if it is enabled now, just that it's not used until further discussion results in a new consensus on how to proceed. We should have stopped using it after the two months ended, then restart using it after consensus has formed after the vote and these discussions here. fetch·comms 00:34, 7 September 2010 (UTC)
It sounds like you agree with my idea. Or is there some other aspect to it that you don't like? Or are you just waiting for a better idea to come along? Yaris678 (talk) 04:13, 7 September 2010 (UTC)
No real problems with it. I think it's a fairly long-winded way to say "stop, think about what to do next, and then, having thought, do."—Kww(talk) 04:26, 7 September 2010 (UTC)
Cool. Yeah, that is pretty much what I am saying. Although I might say well-thought-out, rather than long-winded.  :-) Yaris678 (talk) 04:59, 7 September 2010 (UTC)

Straw Poll Closure

Given that the poll was closed as 408 Support - 217 Oppose - 45 Confused, it's clear a majority has been formed, I'm not sure why some Support 2 - 4 options are in the Other section... Anyway I don't think this should default to close since that is overwriting the consensus that has been established which is that of keep, while neither party can get what they want I think restarting the poll would be the best option. Ғяіᴆaз'§Đøøм | Champagne? 10:29, 6 September 2010 (UTC)

At least on a numeric basis, 408-217 would rarely be considered consensus. I don't see that the strength of arguments on either side would sway the ratio appreciably.—Kww(talk) 17:55, 6 September 2010 (UTC)
This is a straw poll. 408–217 is not consensus numerically. It's not even 2:1 (although barely). fetch·comms 19:39, 6 September 2010 (UTC)
I for one conditionally support #2 as the best option presented, but basically the options pigeon-holed me to a choice; none of the options were an accurate statement of my feelings. Discussion is much more important than the raw numbers of the poll anyway. —Ost (talk) 19:14, 7 September 2010 (UTC)

Developer Input

At least a fifth of the opposition to Pending changes cited IT concerns such as about the slowness of reverting vandalism using this implementation. It would be good to get feedback from the developers as to whether they could address those concerns. ϢereSpielChequers 13:07, 6 September 2010 (UTC)

Agree. The software is a little slow. If it worked the same speed as usual changes that would be a big plus.Doc James (talk · contribs · email) 15:46, 6 September 2010 (UTC)

RobLa left this message on the talk page of Pending Changes/Closure, in reference to the feature requests listed on the Working Summary. I copied it in full:

"Hi everyone, I thought I'd make a quick assessment of the feature bullet items. As with anything, the devil is in the details, but it's probably helpful to have at least a first pass at this:

  • Ideas that are new to us, but worth considering and don't seem on the surface to be too tough (please file in Bugzilla):
    • - A "stop reviewing" button that marks the pending change as no longer "under review". Alternatively this button could be called "put back in queue" or "defer"
    • - A shorter timeout on the "under review", and/or use of Javascript to determine automatically when the review is no longer actively examining the article
  • Ideas that are already in the queue
    • - Unapprove button should be hidden when innactive, not greyed out.[1]
    • - Clearly labeled "Reject" button tied to the rollback or undo action on the review form using the reason entered as the summary[2]
  • Ideas that have been discussed, but need to be filed in Bugzilla
    • - More common names for PC specialpages, and a quick link to PC specialpages, in addition to notification of PCs on your watchlist
  • Ideas that need further discussion and/or clarification:
    • Different logo for PC/Reviewers
    • Better links to documentation on PC related specialpages
      • Generally a good idea, but specific suggestions would be useful. Also, patches welcome.
    • Incorporate the PC protection rationale (protection log message) into the review screen as instruction to the reviewer.
    • Make it faster
      • There's some performance work that has been done, and more that we know we need to do.
    • - An "accept" button/link on the watchlist or page history without loading the full diffs - using popups for example
      • Some of the early feedback on the feature suggested the exact opposite
    • - Make page protection log more visible (reviewing depends on protection reason, but it's hidden by default)
    • - Add short reviewer instructions to the review page
      • Possibly a good idea, but a mockup of the proposed UI would be extremely helpful here to make sure we're not cluttering things up too badly.
    • - Easier access to preview/edit from the review page, for on-the-fly edits to pending changes
    • - Don't make reviewers who try to directly edit an article they're reviewing have to accept their own edits
      • This only should occur when a reviewer is editing an unreviewed article, which forces the reviewer to acknowledge "yes, I'm accepting the sum of my edits and the person(s) before me"
    • - Ability to leave short notes/questions for other reviewers on the review page itself
      • Seems like a generally good idea, but is more than a minor tweak and could use a good specification. This may not come above the cut line for features we could get to soon.
    • If another user preempts a page under review, make notification to the reviewer more obvious
    • An assisted tool for handling large pending changes queues (something along the lines of Huggle or Igloo)—before we have a large pending changes queue—to prevent a self-defeating backlog
    • Special:OldReviewedPages should separate out those pages that are on a lot of watchlists. It may be better for reviewers to concentrate on pages that they know about and pages on few watchlists.

I put a lot of these in the last category because, as I stated before, the devil is in the details. There are a large number of usability improvements and polish that this feature would benefit from. Furthermore, there are many improvements that can be made to the editing experience on MediaWiki that would benefit both Pending Changes-protected as well as other pages (e.g. diff and history improvements), so I imagine that Pending Changes improvements might be best handled as part of a more holistic look at the editing experience. -- RobLa-WMF (talk)"

User:Ocaasi 18:59, 6 September 2010 (UTC) —Preceding unsigned comment added by (talk)

More or less vandalism?

One clear divide between the supporters of pending changes and the opponents is over whether this does or does not reduce vandalism on Wikipedia. At present we seem to have convinced editors on both sides of this argument - though obviously rather more on one side than the other. Would it help all parties reach consensus if we persuaded the Foundation to commission some research on this? I'm assuming that if the research confirmed that as a result of pending changes less vandalism was visible to our readers it would be easier to get consensus? ϢereSpielChequers 13:25, 6 September 2010 (UTC)

I think most people would agree that less vandalism was visible because of the extra checks introduced by Pending Changes. The issue is quantifying that and also quantifying other things such as reduced participation. I don't think it would be possible to do that based on the trial that we have just had. If I am wrong then someone tell me how. If you look at my proposal above you will see that I suggested a new trial after some of the issues highlighted by this trial have been dealt with. Yaris678 (talk) 15:17, 6 September 2010 (UTC)
There are many people who do not agree that "less vandalism was visible", because PC was almost entirely applied to pages which would otherwise have been semi-protected and thus mostly vandal-free. The question is not how much vandalism was visible, but what is the value of the good anon edits (which would not have happened under semi-protection) balanced against the cost of the reviewers' time filtering out the bad anon edits. – Smyth\talk 19:14, 6 September 2010 (UTC)
Good point. I was contrasting to non-protected, when most of the pages would have been semi-protected were it not for the trial. I still think my proposal is the way forward. Yaris678 (talk) 20:41, 6 September 2010 (UTC)
Thus we need a trial such as I have proposed above to look at a range of article to determine when (nothing should be done, pending changes is useful, and semi protection is best).Doc James (talk · contribs · email) 19:24, 6 September 2010 (UTC)
I don't think there is any way to easily determine this with a trial, because it really depends on the page to which PC is applied. But it doesn't really matter to me; I mean, I'm not sure it reduces vandalism, because people are still going to have to rollback edits and such even with PC. Semiprot means that most vandals cannot even edit the page to begin with. fetch·comms 19:42, 6 September 2010 (UTC)
Don't get me started on the all-too-real possibility some CRASH member with a dopuble-digit IQ on a subject approves a bum edit. —Jeremy (v^_^v PC/SP is a show-trial!) 01:22, 8 September 2010 (UTC)

Consensus is not a numerical concept

There are now two threads above arguing as to whether 65.whatever% is consensus. There have been numerous attempts to quantify responses to the straw poll in other ways, such as word counts and prose size. Consensus is not a numerical quantity, and attempts to make it into one miss the point. Arguments about whether the poll was flawed or not also miss the point. So what is the point? It is this.

  • Hundreds of editors have made comments about the Pending Changes trial in this poll, and their words should not be wasted.

They have in significant numbers supported some continuation of something similar, but those supporting options 2,3,4 have also raised concerns about the current implementation and some of these concerns will likely match objections raised by those supporting option 1. There are all sorts of numerical games that could be played, shifting "!votes" from one place to another according to the nature of the support and/or objection, but that still misunderstands the nature of consensus.

Consensus means that significantly more people are content or accepting than those who are discontent or opposing. Most people don't actually care too much either way. Consensus arises when reasonable concerns on all sides are addressed. For this it is necessary to read what editors say, and find solutions which accommodate differing opinions. Consensus is not reached by listening to the most vocal proponents or objectors, but to the views of those in the middle, so that the balance shifts from discontentment to contentment.

What is needed here is for an uninvolved editor to read through the comments made in the straw poll and draw out the common ground and the issues raised about Pending Changes. Those issues could be interpreted as objections or suggestions for improvements/changes, no matter which option was supported.

An analysis of what editors see as benefits and hazards in Pending Changes will inform what "some" continuation of "something" similar means, so that next time the community is consulted, more editors will be content and we will get closer to consensus in the true sense of the word.

Thats all. Thanks for listening, Geometry guy 20:50, 6 September 2010 (UTC)

This is a straw poll, just a vote, right? I understand where you're going with this, but I thought the point of this poll was to look mainly at the numbers. Looking at many of the keep voters, I see a lot of "keep if the speed is fixed" or "keep option X but not Y or Z". That will have to be factored into consensus, but it's not very easy (are the technical issues fixed yet? will we need to run a new poll for the other options?). I'm also not sure about an uninvolved third party. There aren't many who would be interested in this, I'd think (those who care have voted), and I think that the community would be loathe to let a single editor determine consensus for a huge, 700-user poll. A lot of the issues have already been picked out, and the devs are working on them, but that doesn't say anything about whether this current trial needs to continue. A future one is always an option; the keep supporters just seem afraid that if this fails, there will never be enough consensus to start a new trial. fetch·comms 00:32, 7 September 2010 (UTC)
What is needed here is for an uninvolved editor to read through the comments made in the straw poll and draw out the common ground and the issues raised. This has pretty much already been done: see Wikipedia:Pending_changes/Closure#Working_Summary_.28unofficial.29. – Smyth\talk 08:08, 7 September 2010 (UTC)
That list was originally a list of issues brought up at the closure debate. I have added one new suggestion that I saw next to a !vote in the straw poll and I would encourage other people to do the same. The approach we have taken for the list seems to have been quite successful. It is summarised by the hatnote for the section: This section should briefly list issues. Transfer ideas to this informal summary. Neutral phrasing appreciated. May get edited for clarity. Please keep discussion in other sections. It is not necessary for everyone to agree that every point is valid, just to get a consensus on which issues have been identified. Yaris678 (talk) 18:42, 7 September 2010 (UTC)
Consensus does not mean that "significantly more people are content or accepting than those who are discontent or opposing". The opposite, at least according to policy guideline, is true: "editors who continue to disagree with a majority opinion may not be shut out from discussions simply because they are in the minority" and "a few well reasoned opinions may affect a debate much more than several unexplained votes for a different course". I do agree that we need an unbiased, non-partisan individual to interpret the results -- but it is hard to imagine that someone rational could reach any conclusion other than "no consensus" when looking at this mess. There's simply not enough support for options 2+3+4 to justify continuation by any metric, and there's not enough support for option 1 to preclude continued debate in some form.   — C M B J   18:55, 7 September 2010 (UTC)

I try to keep my replies brief.

  • A straw poll is not just a vote: editors contributing have made substantive comments, and ignoring those comments is wasteful.
  • WP:!VOTE is not a policy, but a section within a behavioral guideline, namely Wikipedia:Polling is not a substitute for discussion. The title of that guideline supports the principle that consensus is not determined by numbers, and that it is more important to read what other people say.
  • The relevant policy is WP:CONSENSUS: see especially WP:PRACTICAL for policy advice on discussions such as this.
  • As I understand it, Wikipedia:Pending_changes/Closure#Working_Summary_.28unofficial.29 is a response to comments made during and after the trial in general rather than this poll in particular.
  • The role of an uninvolved editor reading through this poll would not be to determine consensus, but to articulate the views expressed in a concise form so that other editors would not have to reread the entire discussion.

That's all. Thanks again, Geometry guy 22:20, 7 September 2010 (UTC)

Yes, you're right--the citation I made was just a guideline. Nonetheless, I feel it unnecessary to address the other points of contention because it seems that we agree for the most part.   — C M B J   06:22, 20 September 2010 (UTC)

Data regarding effectiveness

Data is here [5] Doc James (talk · contribs · email) 00:02, 7 September 2010 (UTC)

  • This is only for IP edits; not complete data, so nothing major should be based purely off of this. But, this is what I see for IPs:
  • For a large number of the articles (at the bottom) no difference was seen as no IP edited the page. PC wasn't needed there, even as a preventative measure.
  • A similarly large number had no "good" (nonreverted) IP edits, such as Telephone, with 0 nonreverted vs. 29 reverted over 48 days; semiprot would likely have worked better there to reduce the number of reverts needed.
  • A much smaller number of pages had at least one nonreverted IP edit and no reverted IP edits. No protection would have worked fine here; preventative measures were not warranted.
  • Then we get into the pages that had both reverted and nonreverted IP edits:
  • Dinosaur or Westboro Baptist Church are examples of articles with one nonreverted edit and many (60 and 48, respectively) reverted edits over a span of around 1.5 and 1 month(s), respectively. Semiprot could have worked fine here.
  • Antoni Gaudí is an example of an article with one nonreverted edit and a few (8) reverted edits, over a span of 40 days. Depending on the frequency of these reverted edits, semiprot may have worked better, or it could have just been left alone.
  • Some articles, such as Wrestling, saw more nonreverted IP edits than reverted ones (16 vs. 7, respectively, over a span of just over 1.5 months). PC doesn't seem to be needed here, nor does semiprot.
  • Roller skating is an even better example; no protection would have been needed for 18 nonreverted vs. 1 reverted over 1.5 months.
  • A small number of pages, such as iPhone 4, saw relatively close numbers of nonreverted vs. reverted IP edits (64 vs. 71 over 22 days). PC may have helped here, but from these numbers, it doesn't really say anything over the frequency/spread of "good" vs. "bad" edits timewise.
  • Total Drama World Tour was sort of an outlier; there were 328 nonreverted vs. 85 reverted IP edits, much more nonreverted than any other, and in only 20 days. PC probably helped more here, but very few articles experienced large numbers of good edits and bad edits.
  • I got this by sorting the table under "Nonreverted anonymous under Pending". My interpretation is that PC works well for a "niche" area, articles that are both targets for vandalism and for good edits. However, I also think that it shows that PC should not be used as a preventative measure (such as World Stock Exchange, with no IP edits at all, or Roselyn Sánchez, with no reverted IP edits but 11 nonreverted ones), and that semiprot should be used whenever there is regular vandalism (like Dinosaur), rather than putting on PC and having 99% of the IP edits be reverted anyway.
  • The downside to this information is that it doesn't say anything about how spread apart these edits were (clumped together in one day, or by one IP, or just regular vandalism by different IPs) or show the edits that were later removed if they had been accepted mistakenly and not seen until later. fetch·comms 01:03, 7 September 2010 (UTC)
  • Thanks Doc James for the data. My reading of the table is very different than the above comment. I found 57% of tested articles had both accepted and reverted edits by IPs. Without PC the bad edits would have been visible to readers while with protection the good edits would not have taken place. 7% had only accepted edits while 17% had only reverted edits. For the duration of the test no-PC or semi-protection would have been technicaly simpler solution here, however over a longer test period is likely that some of these articles would have been edited in both ways. 19% of the tested articles were not edited by IPs, many because the test was implemented only for a few days on these. Here PC made no difference for the duration of the test. A large number of articles had a large amount of edits both ways, suggesting PC was effective. -Elekhh (talk) 01:55, 7 September 2010 (UTC)
  • Just having edits both ways does not imply effectiveness. Having 1 reverted edit and 10 nonreverted edits is not effective for PC; no protection would be necessary then, right? Many of the pages (you said 17%) had no nonreverted edits; semiprot would have been more effective there. If 7% had only accepted edits, the no protection would be more effective there, considering the slowness of PC and all that. The unedited ones for which PC was enabled for only a short period can be discarded, but that was not the case for most of the non-IP-edited PC articles. For example, Solanco High School, which had no IP edits at all but was on PC for 48 days. In those cases, neither PC nor any other protection was warranted. I do not agree that PC works for some articles, but that most of the time, it's not really needed or effective. fetch·comms 02:59, 7 September 2010 (UTC)
I don't see how you can look at the numbers under PC without looking at the numbers as they were NOT under PC. Saying "well there was 10 non-helpful edits, obvioously PC didn't work!" isn't effective. If you add "and there was 10 non-helpful ones in a similar time frame right before", then yes. But if it's "but there were 25 non-helpful ones in a similar time frame right before", then obviously PC /was/ working, yes? ♫ Melodia Chaconne ♫ (talk) 04:03, 7 September 2010 (UTC)
If someone can collect that data that would be great. But you will still have objections such as the reduction was down to chance or down to the timing of the trial. This is why a better-planed trial would be useful. I am not saying the first trial was a waste of time. We all learnt something about how PC works and we can use that to make improvements before another trial. See #A way forward? above. Yaris678 (talk) 04:24, 7 September 2010 (UTC)
Melodia, I was simply looking at the data presented. Just because the total number of reverts goes down doesn't necessarily mean PC worked (unless the little notice scared everyone away!), as maybe the total number of helpful edits also went down, indicating PC drove people away? There's no data for that either, though. fetch·comms 04:48, 7 September 2010 (UTC)

(undent) here is the data in another format Wikipedia:Pending_changes/Metrics/Full_table# Doc James (talk · contribs · email) 01:57, 7 September 2010 (UTC)

Oh dear. That's a huge table. fetch·comms 02:59, 7 September 2010 (UTC)

This is looking at the wrong metric. Sure this analyzes how well we freed up articles for editing, which is nice, but doesn't address the original goal of pending changes, to stop the public from seeing major BLP violations. Unfortunately, because the trial was so limited in scope, I am not sure if we can really say anything about that. NW (Talk) 22:02, 7 September 2010 (UTC)

Is there any data on how long it took those edits to be reverted? That would be most helpful in discussing effectiveness as well. fetch·comms 23:57, 7 September 2010 (UTC)

One thing to consider

I admit I'm in the 'pro' camp, but one thing I think should be considered is that a LOT of people who voted "1" in this seemed to do so because of technical issues as opposed to effectiveness ones. Others voted 2/3/4 on the proviso that such issues get fixed. Just something to think about for anyone who might actually try to make sense of the 'votes'... ♫ Melodia Chaconne ♫ (talk) 13:49, 7 September 2010 (UTC)

Absolutely. And a lot of people who voted 1 may have a different opinion after fixes. The most important point for everyone to grasp is that there is no consensus for PC in its current form no matter how you juggle the numbers. People need to stop applying PC to articles, and to remove it where it exists. The trial is over, and it is wrong to let it continue when there is no consensus for it to continue. That doesn't mean a new trial of some modified version can't start later, but this trial has to stop.—Kww(talk) 14:30, 7 September 2010 (UTC)
That was/is my feeling and why I voted 1 or 4 - we either give it a rest or we do it on all BLPs for a further trial and get a feel for it. Casliber (talk · contribs) 14:34, 7 September 2010 (UTC)
As I said above, I agree with what Melodia said here. In the mean time, I have a suggestion: as an alternative to arguing here about whether or not to shut down the existing usage of PC (an argument that appears to me to be going back and forth, without reaching any more consensus than what there was already), people should feel free to go to RfPP and request removal of PC from pages that they watch, changing either to no protection or to semi-protection as the circumstances at the particular page indicate. I'm pretty sure that reasonable requests of that sort will be granted. --Tryptofish (talk) 14:56, 7 September 2010 (UTC)
"no consensus" is far different than consensus to end or consensus to continue. it means the issue is unsettled still. The right answer to "no consensus" is to go back and use the dissenting voices to drive a compromise, then see if you have consensus again. Triona (talk) 18:33, 7 September 2010 (UTC)
In cases of "no consensus", the status quo prevails. The status quo is that pending changes was a two-month long trial: 60 days. The 60 days are over. The thing that there was consensus to do is over. You don't need new consensus to stop a trial that was only supposed to last two months when those two months are up. Once the two months is over, the trial stops. It doesn't coast along in limbo while supporters ignore the fact that they are doing something that there is no consent for. What did "two months" mean if the "trial" is going to continue forever?—Kww(talk) 18:37, 7 September 2010 (UTC)
Hem, a trial is generally conducted in order to get some "Hands on" experience with a certain feature in concept so that man may determine if man wishes to continue its usage, or if man wishes to scrap it. The trial itself has ended, and now we are establishing what we wish to do with the feature now that we have tested it. After all, there is no sense conducting a trial if man just shelves the mechanism and the results without considering them further.
Now, i have to point out that i am in the "keep" camp, but if i read the general sentiment correctly i would point out that most people seem to be in favor of keeping the system itself. A fairly large share of the opposes and conditional supports are related to the performance and user friendliness of the system, and not to the actual purpose of the system. In both cases of these cases there is plenty of room for improvement, but as long as people aren't against the core of the system these issues can be ironed out over time. A secondary note is that little people would support automatic protection of all BLP pages (Option 4), though a fairly large would support expansion of the function to cover low-traffic\blp pages. Even so an equally large number argues that pending changes protection should be kept as-is, so i cannot readily determine if option 2 or 3 would be preferred - though if we consider that the people who voted option 1 would prefer option 2 over 3, we can say that "As is" is preferable of expansion.
If anything i would argue that leaving the system in place - but not expanding it until performance and user interface issues are solved - would be a compromise between both sides of the debate. While straw polls cannot be counted by numbers only (Nor are they actually entirely accurate), i would say that its sentiment seems to be in favor of retention of the system. Of course we could turn the system back of and enter an entire community debate phase to decide if we would wish to turn it back on again, but that really sounds like excessive red tape to me. Excirial (Contact me,Contribs) 20:40, 7 September 2010 (UTC)
If people wanted an indefinite trial with a discussion after two months, that what they should have sought consensus for. If you seek consensus for a two month trial, an important component of that consensus is the time limit on the trial. It's dishonest to tell people you are going to try something for two months and then refuse to stop.—Kww(talk) 20:56, 7 September 2010 (UTC)
What purpose would a trial have if you are going to stop it regardless of the end result? Technically we could finalize the two month trial, disable the feature and discuss if we wish to re enable it - and i would point out that the re-enable discussion was always going to be held regardless, since testing a feature wouldn't make any sense if we didn't at least discuss the results afterwards.
I'm fine with disabling the feature for now and then discussing and improving before we consider re-enabling it - or we could actually disable it pending an immediate review - but the straw poll seems to indicate that the feature itself isn't unwanted, and even acceptable to a large share of the (responding) editors in its current state. A straw poll usually doesn't substitute for discussion which is why i am not insisting on keeping it enabled, but seeing the current state of affairs it would be worth considering to say the least. If anything i would argue that we had a two month trial with a review afterwards in order to establish if we wish to continue using it. If the straw poll would have turned out the other way around i would have argued that it should be disabled altogether until major points of concern are solved (And after that, it would need new consensus). Excirial (Contact me,Contribs) 21:10, 7 September 2010 (UTC)
I'm not asking that the feature be disabled in Wikimedia software, only that all articles currently under pending changes be switched to a standard protection mode and that admins stop putting new articles onto pending changes. People should certainly continue to discuss what to do with this feature, but right now there are about 1500 pages that need to have pending changes undone.—Kww(talk) 21:25, 7 September 2010 (UTC)

Consensus-able ideas.

  • Keep the Mediawiki extension enabled (separate from its use on articles)
  • Don't add new pages
  • Turn off PC on pages where reverts were above 80% (very high vandalism)
  • Turn off PC on pages where reverts were below 20% (very little vandalism)
  • Keep studying stats
  • Keep working on interface
  • Keep developing policy
  • No expansion or permanent establishment without substantial improvements to the feature
  • A community-approved re-evaluation/poll after 3 months
  • Keep the feature running on the remaining ~750 articles, on a trial basis
~bold indicates no major objections

Ocaasi 22:30, 7 September 2010 (UTC)

  • Against simply due to the first bullet. We're already three months into a two-month trial; time to shut it down. —Jeremy (v^_^v PC/SP is a show-trial!) 22:40, 7 September 2010 (UTC)
For purposes of discussion, can you separate the code extension from its actual use as TFOWR suggested we could add another bold point to the list? Ocaasi 23:51, 7 September 2010 (UTC)
I will not do so; you, Levy, Doc, and BigK already bloody know that. —Jeremy (v^_^v PC/SP is a show-trial!) 01:17, 8 September 2010 (UTC)
Keep the feature running on certain articles? This was a trial. It's over. Remember? --Yair rand (talk) 22:42, 7 September 2010 (UTC)
People didn't keep their word about two months, why should a three month promise hold any more weight? Shut the trial off, and then supporters can negotiate with some credibility.—Kww(talk) 22:47, 7 September 2010 (UTC)
Yes, you're against it, but a full 65% of poll respondents are not. This is a workable alternative to either a full close or a full keep (or keep with expansion). Ocaasi 23:20, 7 September 2010 (UTC)
I'm against fraud; I would have been for the trial (at least at level 2) if this poll had been better phrased. As it is, I don't trust you. Septentrionalis PMAnderson 01:08, 8 September 2010 (UTC)
Hear, hear, Septentrionalis. I'm also against fraud, especially where things get messed with to make it look good :) BarkingFish 01:27, 8 September 2010 (UTC)
Precisely the point I'm making. I don't really have a strong view on PC, but I detest fraud.—Kww(talk) 03:38, 8 September 2010 (UTC)

It seems to me that the parts that have consensus are actually: Don't add new pages. Keep studying all aspects that have been criticized. No expansion or permanent establishment without substantial improvements to the feature. --Tryptofish (talk) 23:17, 7 September 2010 (UTC)

I suspect "Keep the extension enabled" may also be consensus-able - as I understand it, the devs aren't happy with turning it off and then back on, but I'd imagine we'd all mostly be happy if we kept the extension on but unused. To be honest, the only one I'm seeing real consensus problems with is the last one: "keep the feature running on the remaining 500-1000 articles", which does seem to be a continuation of a terminated trial. Lose that and I suspect we could get broad happiness. TFOWR 23:33, 7 September 2010 (UTC)
Yes this sounds like a fair compromise. No one has provided data that we should not use this thus we should study it further.Doc James (talk · contribs · email) 23:42, 7 September 2010 (UTC)
I agree with TFOWR; keeping it on is OK as long as it is not used until further research is conducted over the effectiveness and all that. fetch·comms 23:53, 7 September 2010 (UTC)
Seeing that PCs received more than the pre determined 60% I think decreasing it to 500-1000 articles as described and doing further trials is a compromise to those who disagree with PC. A 35% minority does not get to dictate what occurs without any sufficient arguements or data to justify their position.Doc James (talk · contribs · email) 00:08, 8 September 2010 (UTC)
Yes we do, unless you can persuade us to join a consensus to keep this. In my case, this may be possible; but I see no efforts to do so. Claims of this sort are not how Wikipedia works. Nobody ever agreed that 60% was consensus -and it isn't; insisting that it is bad faith. Septentrionalis PMAnderson 01:05, 8 September 2010 (UTC)
Neither is 40% consensus. The right answer is to go back, and find something more of us can agree on, and then see where that stands. Consensus isn't a numerical concept, but a way to make sure that all the arguments are heard. Triona (talk) 01:22, 8 September 2010 (UTC)
As far as I'm concerned, what consensus boils down to is that the arguments that hold the most water with regards to policy and reason are given the most weight, and that straight-up-or-down-!votes are discounted. As there is no way any FraggedRevisions install will pass the latter criterion (because it operates on a glaring logical fallacy) I will never support it. And fixing it to try and win my support is tantamount to eliminating FraggedRevisions altogether. —Jeremy (v^_^v PC/SP is a show-trial!) 01:25, 8 September 2010 (UTC)
If you have an inherent, existential problem with the feature, I'm not sure how you can participate in a discussion about a compromise. Since you seem to want no part in anything PC-related, I'm going to mark keeping the extension active (not its use on articles) in bold, without implying your support in any way. Ocaasi 01:44, 8 September 2010 (UTC)
This discussion is about FlaggedRevs implementations, is it not? If so, then having the extension enabled is going to end up confusing the admins having to apply it, since (IIRC) the PC menu still shows up in the protection lists. —Jeremy (v^_^v PC/SP is a show-trial!) 01:53, 8 September 2010 (UTC)
@Ocaasi: I agree that those who are unwilling to compromise will unfortunately have little voice in a compromise process. BigK HeX (talk) 01:56, 8 September 2010 (UTC)
@Jeremy... This discussion is about what to do, given a trial that was supposed to default to close, the poll for which started off poorly but resulted in 65% support for keep, significant 'minority' opposition, and a feature that almost everyone wants improved before it's expanded. Options are on the table, especially mere technicalities that don't even effect actual articles. If you oppose the use of PC on any pages, that's fine, but the extension issue is just a proxy for that. We're narrowing the disagreement down to a territory of hundreds of articles. That won't be particularly difficult to bridge. Ocaasi 02:23, 8 September 2010 (UTC)

I updated the numbers and bolded the 80/20 changes, as well as a later poll. That leaves articles, unsurprisingly. 750 is less than half of the original trial and about half the amount that are currently in use. Ocaasi 02:53, 8 September 2010 (UTC)

Alternative compromise

I have a great compromise. How about we shut this shit down altogether, and get back to doing what we do best, instead of having even more work to worry about on top of what we already have? Wikipedia is already a bureaucratic nightmare, and having more requests and stuff to do on top of all the AFDs, RFCs, Request boards, task forces and the like only serves to slow down the already sterling work put in by a lot of editors here to keep this site clean. It was a trial to begin with, the trial is finished, let's euthanase this while we have the opportunity to do so. BarkingFish 03:07, 8 September 2010 (UTC)

  1. Oppose Not much of a compromise to the 65% who support PC here. Doc James (talk · contribs · email) 03:10, 8 September 2010 (UTC)
    Comment' - And the compromise above isn't much to the 35% of Wikipedians who think it should have died already, Doc James. What's being suggested is we have another trial (twice the length of the first), when we've already had one, and that we then decide "again" whether we want it or not. As has already been pointed out, concensus is NOT a numerical argument - so why are we still basing this on numbers? BarkingFish 03:18, 8 September 2010 (UTC)
  2. Oppose We could, instead, knock around the idea of accepting Pending Changes permanently, as it has exceeded a very respectable 6-to-4 support ratio that was once considered. BigK HeX (talk) 03:13, 8 September 2010 (UTC)
  3. Support as the support ratio is not a consensus by any stretch of the imagination. They've had three months on a two month trial. Stop pettifogging and turn this feature off already, THEN come back after the data's reviewed. —Jeremy (v^_^v PC/SP is a show-trial!) 03:15, 8 September 2010 (UTC)
    Comment Would you find it a tiny bit odd for someone advocating for 65% of the represented community to be disregarded to invoke WP:CONSENSUS? BigK HeX (talk) 03:24, 8 September 2010 (UTC)
    Comment Wouldn't you find it just a tiny bit odd for Wikipedia to ignore and disregard a standard policy of the site in order to invoke something a third of represented users actually don't want? :D BarkingFish 03:27, 8 September 2010 (UTC)
  4. Strongly support; it is bad faith to insist on Pending Changes, which was only given a trial on the condition that it would be shut down if it did not win consensus. Septentrionalis PMAnderson 19:46, 8 September 2010 (UTC)
  5. Support was a two month trial now past two months. Either shut off the feature or remove it from all content pages until there is some sort of consensus.Sumsum2010 · Talk · Contributions 23:12, 8 September 2010 (UTC)

The real question is will PCs make Wikipedia a better encyclopedia

The next question is how do we determine this? Or have we determined it already? If we havn't than further investigation will be needed. This is what a keep or cancel proposal should be based up.Doc James (talk · contribs · email) 03:29, 8 September 2010 (UTC)

Then start reviewing the data - nobody's even thought of doing that yet. —Jeremy (v^_^v PC/SP is a show-trial!) 03:30, 8 September 2010 (UTC)
Yes we do really need to go through the data we have aswell. This straw poll is sort of meaningless as all those who voted have not had a good presentation of the data from the first trial.Doc James (talk · contribs · email) 03:33, 8 September 2010 (UTC)
Then start going through it. As it is all sides that are already not deadset were responding from a position of ignorance. —Jeremy (v^_^v PC/SP is a show-trial!) 03:34, 8 September 2010 (UTC)

(undent) Here is the data again [6] and [7] Unfortunately the trial was not done in a sufficiently rigorous fashion to answer many questions posed above thus my reason for proposing a smaller more structured trial [8] were the pages are compared to a period were they are not under PC. Doc James (talk · contribs · email) 03:36, 8 September 2010 (UTC)

Given the nature of Wikipedia, any smaller trial is going to end up being as inconclusive, if not more so, as the first, whether due to selection bias, the shift in vandalism levels (vandalism has historically been low from June to August, when the trial took place), or the behavior of editors who have vested interests in the chosen articles. —Jeremy (v^_^v PC/SP is a show-trial!) 03:43, 8 September 2010 (UTC)
So how do you suggest we quantify the effectiveness of PCs?Doc James (talk · contribs · email) 14:39, 8 September 2010 (UTC)
Ask someone who wants to have Community Resources Against Schoolchild Hoodlums around. You should know by now what my answer to that question will be. —Jeremy (v^_^v PC/SP is a show-trial!) 00:30, 9 September 2010 (UTC)

Comments on the situation

Initially, the implementation was supposed to be deactivated at the end of the two month trial, since as the name suggests this was supposed to be a trial limited to two months, we would then analyze the results, the merits of PC, and discuss potential new proposals for implementation. However, the WMF team in charge of the trial having indicated it would be more convenient not to have to turn off then back on the implementation, if a new implementation were desired; the community was given the opportunity, if supported by a compelling consensus, to extend the duration of the implementation beyond the two months, provided a decision would be reached within a month. So, unless a compelling consensus exists to continue the active implementation, it should be halted, though not necessarily entirely turned off.

There has not been any consensus on how the discussions regarding the extension of the implementation beyond the two months trial should be organized, input on this question was requested at different occasions but were not sufficient. Thus, it resulted in incertitude, and this poll being set up by essentially one user. While the poll initially specified how the results should be interpreted (6-4 majority in keep would result in keep, and the specific keep option chosen would be the one with most supports), those specifications were proposed by a single user and not established through consensus, so can't be used to determine the outcome. During the course of the poll, no proposed interpretation of the poll was agreed upon, and considerable objection to the poll structure, from all sides surfaced, with no resolution (other responses, archive). Therefore we should per general practice determine if, based on all discussions, including the poll, there exists a consensus which is compelling enough to continue the active implementation.

Numerically there is almost 2/3 60% (correction counting 'other responses') in support to keep, but it doesn't mean much in terms of consensus, moreover several are conditional, and whether they should be taken in conjunction or separately is disputed, and doesn't appear satisfactorily resolvable. To determine consensus, we should analyze the reasons given for and against extending the implementation and their level of support. That the poll offers different possibilities for keeping without provisions confuses the two different questions of whether we should extend the implementation beyond the two month trial and what kind of implementation we want in one. This creates several problems, it makes interpretation more difficult, and it has the consequence to disenfranchise those opposed to continue an active implementation at this point but who want to have a say on which kind of implementation the community could adopt in the future. Therefore this poll cannot form a basis to valid a specific keep option (2, 3 or 4), and should only be considered as a potential justification for a provisional extension of the implementation under the same terms or in a more limited form.

Moreover the keep options of the poll were offered unconditionally, and while the community can change its decisions, it could be argued that, assuming one of the keep options had been approved, in order to later deactivate the implementation, a consensus for doing so would have to be achieved, thereby reversing the burden of consensus building. The level of consensus here is nowhere high (and informed) enough to justify adopting such unconditional options, particularly as this is a major shift from traditional editing practices, for which an even higher level of consensus than normal seems required (see below) (to handle this, a future proposal could include the provision that if contested, a consensus is required to keep PC - as opposed to a consensus to remove it, at least for some time).

A further justification that the poll cannot support any of the keep options presented is that no comprehensive analysis of the trial occurred, by lack of time. This implies that any non-provisional decision at this point would be uninformed, and hasty. Thus the discussions could only legitimate a provisional extension of the trial, during which we would analyze its results, merits of PC, then decide of a course of action. The question is thus reduced to: is there a compelling consensus to extend the implementation provisionally ? We could theoretically start a new poll explicitly asking this, but there would be no time in the deadline of one month given by developers to organize such a poll.

Considering the various issues raised at the closure page, here and with the statistical data obtained, I think there is enough material to conduct an analysis of the trial, and to form consensus on various points of policy such as the extent of the application of pending changes or the use of 'level 2' protection, which can help to form a consensus on the question of adopting a form of PC or not, so as not to require an extension of the trial. To sum up, the trial served its purpose in giving the community an apercu of what they could expect of PC and will help the community in reaching a more informed consensus on the question of using it or not. And we can always make new trials if there's consensus to do so.

In addition, an extension of an active implementation could lead to considerable controversy, as shown by the above discussions, and it would probably be adverse to a proper and calm discussion on the trial and merits of PC. At worst it could turn into an intractable sitewide dispute, which we don't want. The wide differences of opinion show that there is a need to reconcile them and find a general consensus on a myriad of points. Building consensus on such an issue is complex and should not be precipitated, continuing an active implementation of PC wouldn't bring much in this endeavor and would on the contrary detract from it, due to it being controversial. So in the circumstances, I think that keeping an active implementation would be hardly tenable and counter-productive, but we should still see if there are more substantial reasons not to do so:

Due to the lack of in-depth discussion of the merits of PC, the opinions given have been mostly based on personal experience, or superficial analysis. Both sides gave diverging viewpoints but no consensus developed, for example on the important issue of whether PC is an effective barrier against vandalism, BLP violations, and other, compared to semi-protection. Many arguments of similar strength were given for and against, but for the above reasons, they were hardly ascertainable (such as effect on new users). However those against extending the implementation have given many more arguments, many listed here, and they indicate that various problems need to be addressed, with some clearly identified and recognized, including disturbance of usual editing, increase in the need for vandal-fighting and inefficiency in cases of sockpuppetry or specific article issues; examples which directly affect the quality of articles and work of editors, so must be given particular weight. Since the default was to discontinue the implementation, the burden was on those in favor of extending the implementation to find counter-arguments to those or provide solutions, which in general didn't happen. Accordingly, this is sufficient to show that there is no compelling consensus to continue an active implementation. This should not be seen as a victory or defeat for anyone, because it is what was supposed to happen, it was a two month trial as agreed; and the final outcome of those discussions do not prejudice future discussions of PC by the community.

I would also point out that the WMF explicitly stated that in a poll, a two third majority was necessary in general for deploying 'FlaggedRevs' [9], being understood that the default was deactivation, this applies here and the poll didn't reach 66%, only roughly 65% 60% (correction counting 'other responses'). Though purely numerical, it reinforces the proposition that an implementation of PC is a substantial modification of our editing practices which requires a strong consensus in the community. The very nature of pending changes, by relying on human review, makes it also necessary that it be adopted by a large majority in order to work efficiently.

Therefore, I think that the discussions do not demonstrate a consensus compelling enough to divert from the default option which was to halt the implementation, the active part in any case. Now comes the question of whether we should keep the implementation in a dormant state as opposed to turning it off completely, I think the former would be more convenient for developers and the community, as testing for example could be available in project space, and tackling the numerous usability issues and other technical problems identified in the trial is important. However it should be disabled as an option in mainspace, so as to avoid unnecessary interface load for admins and prevent further uses of PC, and a sunset clause should exist.

With regard to the proposal for gradual shutdown, per my assessment that no consensus existed for extending the implementation, I do not think that the gradual shutdown could satisfy the consequences of this - if uninvolved admins agree that the mandate for the trial is over, we should remove PC from mainspace without delay, which can be done in a matter of days or even hours with a script (guided by those metrics); and I do not find substantial gain in a gradual shutdown, for the same reasons as above. In addition, the one month limit makes that a new poll could not conclude in time. This applies to the other proposed polls, which appear very unlikely to generate a consensus. Now is well past time for determining consensus, not building it. In conclusion, in my opinion:

  1. Pending changes should be removed from any mainspace page and replaced with protection if appropriate, which shouldn't take more than a few days.
  2. If feasible, developers should remove the PC interface from the protection interface; but keep the usergroup, logs, and PC in project space for testing.
  3. In the coming months, the community can calmly analyze the trial, the merits of PC, and discuss of possible new implementation proposals which would have to acquire consensus for adoption, including any trial of any order of magnitude.
  4. If no implementation proposal is accepted after a reasonable period, of say one year, pending changes should be entirely turned off.

Please note that this is not a proposal, but an effective assessment of consensus. In order to move on and as the deadline is getting close, I will seek the opinion of uninvolved admins. Cenarium (talk) 06:41, 8 September 2010 (UTC)

FYI, Already been attempted by others through the admin noticeboard. I'm not sure at this point whether we can find an active admin that is uninvolved - I hope that it is possible, but I'm not holding my breath. Triona (talk) 07:13, 8 September 2010 (UTC)
I know of several active uninvolved admins, we could contact them personally in last resort. Cenarium (talk) 07:42, 8 September 2010 (UTC)

I'd just note that the voices here that have been forcefully demanding a "shutdown" seem to universally be ignoring the fact that there is significant majority support for the system, and that the significant majority support would be completely disregarded if a "shutdown to the trial" was decided. Conceding a shutdown to the trial is tantamount to rejecting PC outright, even despite the significant majority and (perhaps small) possibility of having enough support for permanent approval. There are likely many supporters who are hesitant to make such a large concession, especially when opposers are making few (if any). If the persistent intransigence to compromise continues, and supporters are forced into an all-or-nothing situation, then many likely find the discussion for support strong enough to take "all", though they probably would have happily accepted compromises. IMO, it seems potentially self-defeating to refuse to compromise at this point. Something to consider, perhaps.... BigK HeX (talk) 12:58, 8 September 2010 (UTC)

And I will point out that the people that are refusing to shut it down are going back on their word.—Kww(talk) 13:22, 8 September 2010 (UTC)
I have no idea what the original specifications were, but if (and that's perhaps a big "if", but if) there is sufficient support for permanent acceptance [perhaps exceeding 6-to-4 in support], then delaying a push for final acceptance and even seriously considering a shutdown (in the face of significant majority approval) while compromises are being sought could be viewed as a rather large concession to opposers. Those with pretty strong approval to make their arguments could be just as adamant in pushing for their preferred resolution [Pending Changes goes live permanently.] I'm not really seeing that sort of uncompromising push from PC supporters, though. The unwillingness to come to the table seems a bit one-sided, IMO. BigK HeX (talk) 13:39, 8 September 2010 (UTC)
Wikipedia does not operate on a basis of majority approval. It operates on consensus, and if consensus for something doesn't exist, it doesn't happen. This is not new and it has been standard practice for a long time. The original poll that produced consensus for the trial specifically stated that this is for a two-month trial only. Extending the trial beyond that two months requires a consensus to do so, and the lack of such a consensus means the trial should be stopped. Now this doesn't preclude further discussion on the subject of pending changes, nor does it preclude a different proposal from being considered or accepted (or even the same proposal - the community can change its mind). But it doesn't mean the trial can carry on. Hut 8.5 13:46, 8 September 2010 (UTC)
For the moment, the current use of PC could well not even be about an "extended trial". It could well be an acknowledgment of the significant majority of support for the system, which the discussion may indicate is sufficient for permanent usage. As for consensus, consensus is about finding where agreement lies between supporters and opposers here, and, from what I've seen, oppposers are not participating in that discussion. BigK HeX (talk) 13:53, 8 September 2010 (UTC)
For my part, it's because discussing things with people that are behaving dishonestly has always proven to be a futile effort.—Kww(talk) 14:10, 8 September 2010 (UTC)
There seems to be a failure to assume good faith. I was never involved with the previous discussions to start PCs. Shouting to have PCs shut down is poor form. We should be discussing how we determine whether or not PCs works and how we should go about determining this if it has not been determined already.Doc James (talk · contribs · email) 14:49, 8 September 2010 (UTC)
What's the good faith explanation for a trial that was supposed to last 2 months and require 66.67% acceptance to continue that is still continuing nearly 3 months later with only 60.8% approval? I'm being polite, but I'm furious.—Kww(talk) 14:57, 8 September 2010 (UTC)
It is more than 65% support. And decision is based on reasoned argument.Doc James (talk · contribs · email) 16:05, 8 September 2010 (UTC)
If you measure (support/(support+oppose)), you get 64.96%, which is less than 65%, and certainly less than 66.67%. if you measure (support/total), you get 60.8%. No matter how you measure it, the original conditions for ending the trial have been met. The date has passed. The support level has not been met. I'm not arguing that people can't get a consensus for a new trial, but the first order of business is for people to honor the original agreement. There is no good faith explanation for unilaterally changing an agreement after it has been made.—Kww(talk) 16:45, 8 September 2010 (UTC)
I agree about a possible WP:AGF issue. Moreover, it seems a lot of responses to me personally are of the form "shut it down! RAWR!", which seems kinda futile, since I have no such capability. I kinda wish more of the responses I got were a bit more helpful for the discussion side of matters --- I certainly have no technical involvement in the matter. BigK HeX (talk) 14:59, 8 September 2010 (UTC)
Firstly, I believe it is a bad idea to hinge our entire decision on this poll; the very use of a poll in this situation has drawn controversy from the beginning. Even were it agreed beforehand that a poll was the appropriate means of discussion, I think that we can all agree that this particular poll was less than ideal. Therefore, I think we must take the results with a grain of salt. Obviously, there is majority support for the use of PC in some form, and combined with the limitations imposed by the devs, a straight shutdown is not the appropriate action to take here. I think a soft shutdown per above is the best way forward, because we have not met the conditions (2/3) of either the poll, or even of the original trial, and we have already extended well beyond the 2 months originally allotted. I also think that in debating such a major and contentious issue that we get discussion off of this Talk page and into a more visible and appropriate forum as soon as possible, and ensure that everyone in the community knows that it is ongoing, and is able to contribute accordingly. I am absolutely sure that many people voted in the poll and assumed that was the end of it (yet another way in which the poll itself was tragically, even fatally, flawed). Supporters (vocative case), you are almost certainly going to get your PC in some form or other. Supporters, in the interest of minimizing controversy and backbiting, and ensuring that we have a unified community able to work together toward the best possible implementation, stand by your original agreement and end this trial. Then move forward with good-faith discussions about how to improve PC and implement it in the future. Revcasy (talk) 15:00, 8 September 2010 (UTC)
Note: not directed specifically at you BigK Hex. We need some consensus here to take to the devs. Revcasy (talk) 15:04, 8 September 2010 (UTC)

(edit conflict):We need to end the original trial like originally agreed, then discuss it. I'm open to PC once everything is faster, etc., but this current trial was for two months only. That is an indisputable fact. We have passed two months. Also indisputable. Why is it so hard to say "let's stop using it for now, then focus on analyzing the data, then propose a new trial with a more efficient system"? fetch·comms 15:02, 8 September 2010 (UTC)

RE: "Supporters (vocative case), you are almost certainly going to get your PC in some form or other."

That seems good bit less certain, if supporters concede that the current effort at implementation be killed without first understanding with more clarity what broad areas there may be room for compromise. BigK HeX (talk) 15:07, 8 September 2010 (UTC)

The current implementation was a trial. I am afraid that you are going to have to trust the process in this case. Again, I am not suggesting getting rid of PC (I consider that point settled) we are ending the trial, letting the devs do the work they need to do on the UI, and moving the discussion about implementation to a more appropriate place. Revcasy (talk) 15:16, 8 September 2010 (UTC)
Yes, there was a trial, but it's been pretty strongly suggested that the current situation [the post-trial period] would be discussion to continue or discontinue. Supporters conceding to those who wish to discontinue could well view discontinuance as tantamount to conceding to an outright rejection of PC, with very little idea of how the process would re-initiate (if ever). I, personally, am not confident at all that the discussion for Pending Changes will continue anytime soon if supporters were to agree with discontinuance right now. BigK HeX (talk) 15:23, 8 September 2010 (UTC)
I see your point, and I appreciate your concerns, but I am at a loss for how to address them to be honest. We seem to have a fundamental disagreement about what should happen at the end of the trial. Perhaps we resort to legalism? What was the exact wording of agreement about the trial? I strongly suspect that the wording of the poll has little relation to that earlier statement of consensus. Or perhaps we could say that we are temporarily discontinuing PC for further development and discussion about its method of implementation. Revcasy (talk) 16:45, 8 September 2010 (UTC)
I guess a question in the mind of supporters may well be that if supporters decide to disregard the current majority support that appears to exist for PC, how do they know that some of the more intransigent opposers will suddenly come to the table about what ways they CAN ACCEPT Pending Changes, if the more strident opposers are giving little clue now of what is desired? BigK HeX (talk) 15:12, 8 September 2010 (UTC)
(unindent) BigK HeX, was your first comment directed at me ? I'd like to bring to your attention that I am the author and principal organizer of this trial [10] (although I had little time to edit WP recently), without me there would be no trial. What I have done here is analyze the consensus as I see it, and we actually do have a compromise: instead of entirely shutting down PC as we were supposed to do, there is consensus to place the implementation in dormant state. Yes the situation is less than ideal because the poll was drafted by one user with little discussion and offered unrealistic options, it should have just asked the question of whether we should extend the trial. We would then in either case have proceeded to analyze the trial, evaluate the merits of PC, and finally adopt or not a more definitive implementation proposal. What would have been the point of a trial if we didn't use it so as to achieve an informed consensus on the question of implementing PC ? Now we should do just as planned, we'll analyze the trial (it'll take 1-2 months I suppose), judge the merits of PC (1-2 months) then develop and propose a more definitive proposal (1-2 months), with possibilities of new trials all that time as long as they have consensus (which could add a few more months). I expect that in no more than nine months we'll have developed consensus on a more definitive PC implementation, as long as we can work in a relative serenity and enough people help on the road. Cenarium (talk) 00:00, 9 September 2010 (UTC)
My comments were just a note about the "RAWR" person(s). BigK HeX (talk) 02:45, 9 September 2010 (UTC)

Overall, I think Cenarium's interpretation is reasonable, but with the following caveats. It seems to me that "this applies here and the poll didn't reach 66%, only roughly 65%", even with the subsequent language pointing out WP:NOTVOTE, goes too far in tea leaf-reading the numbers, and disrespects the approximately 65% of participants. I agree with the 1–4 numbered list of actions to be taken, except that number 4 seems gratuitous so long as everything is removed from mainspace. And I also take note of the unfortunate RAWR tone of some of the comments above. --Tryptofish (talk) 15:36, 8 September 2010 (UTC)

If we include also other responses, that's 60% percent, but that's besides the point. I appreciate that a majority of users are in support of keeping some sort of PC enabled, as the author of this trial I'm glad that so many users think it was a good thing. But we should ultimately follow our decision-making process of consensus, and the terms given for this trial. Cenarium (talk) 00:00, 9 September 2010 (UTC)

I can't add much of value beyond Cenarium's thoughtful summary. I have looked at the poll and at some of the meta-debate above and feel generally that there isn't sufficient consensus to either continue the trial or to expand PC to the mainspace. However, there is certainly a broad base of support for PC in general and it is entirely possible that another trial will be better positioned to allay fears of those in opposition. The NOTVOTE vs. tallying debate directly above is largely immaterial. We are talking about a serious change to the core featureset of the encyclopedia. Strong consensus should make itself evident. Protonk (talk) 17:22, 9 September 2010 (UTC)

Arbitrary break (2)

Clearly there is widespread support for PC. However, the current trial has run its course. It should be ended (soft shutdown), and all pages should have PC removed, both so that we can start from scratch, and because this was the original agreement presented to the community. We as a group have clearly not reached consensus on exactly how PC should be implemented and when. Therefore, concurrently with the trial being shut down, a new community-wide discussion should begin, in a more appropriate forum than the closed poll's talk page, about implementation of PC and improvements needed to the trial system. Both supporters and those opposed will have an opportunity (if they choose to exercise it) during discussion to provide input on the nature of implementation, and a consensus of some sort will be given a chance to emerge. Feedback? Revcasy (talk) 15:29, 8 September 2010 (UTC)

How does that differ from just saying let's discuss the proposal in the section just above, but do so on a new page? --Tryptofish (talk) 15:37, 8 September 2010 (UTC)
Indeed it does seem they are both reasonable statements of similar ideas. BigK HeX (talk) 15:39, 8 September 2010 (UTC)
Re:Tryptofish. I am attempting to formulate as succinct a statement as possible so that consensus can be reached on a basic plan of action which includes moving discussion of this very important issue off of this obscure page (as you so astutely pointed out). It also seemed an appropriate place for a break. By all means, discuss it wherever you prefer.Revcasy (talk) 15:49, 8 September 2010 (UTC)
OK, I see, that's fair. --Tryptofish (talk) 16:13, 8 September 2010 (UTC)
Same here, and also a new page please.--intelati(Call) 16:50, 8 September 2010 (UTC)
Question: Would this poll be sufficient to initiate a new trial after improvements are made? If not, can we agree to a 'timeline' for a new poll for such a trial, say within 6 months or a year? Ocaasi 16:52, 8 September 2010 (UTC)
Not for some hypothetical set of unagreed upon improvements. Part of the consensus building process would be identifying the minimum set of changes to be included.—Kww(talk) 18:33, 8 September 2010 (UTC)
You don't think there's enough support here to warrant some continued trial within a year? Ocaasi 18:42, 8 September 2010 (UTC)
Let's see what the improvements are. If there are none - and the speed problem may be intractable - then there is not sufficient support for any trial, any more than there is now. Septentrionalis PMAnderson 19:21, 8 September 2010 (UTC)
For my part, I think there is certainly plenty of time to reach consensus about where and when to have another trial. I suspect that six months is probably closer to the mark than a year. However, as we all know none of us are officially appointed representatives, nor am I sure that this particular debate represents a quorum in any sense that matters (that is, we are almost certainly not going to decide on behalf of the entire WP community what the plan is moving forward with regards to PC). I do however think that we can come to a very limited agreement that: PC should not be closed in a non-reversible or permanent way, and further community-wide discussion about the nature of its future implementation and improvement is certainly warranted. I don't feel comfortable with setting a timetable, or making commitments on behalf of others, since I am speaking as an individual, not a representative. I do however feel comfortable saying that the current trial should be closed in a definitive way, because it would ease opposition and send a signal that all of us are making good-faith efforts to find consensus. I also feel comfortable saying that the discussion, at this point, should be much more inclusive for the very same reasons. Revcasy (talk) 19:36, 8 September 2010 (UTC)
This is how I think it'll go (see comment in above section): we'll analyze the trial (it'll take 1-2 months I suppose), judge the merits of PC (1-2 months) then develop and propose a more definitive proposal (1-2 months), with possibilities of new trials all that time as long as they have consensus (which could add a few more months). Cenarium (talk) 00:03, 9 September 2010 (UTC)
Okay, so soft-close, continued discussion, new poll for a trial within 6 months? User:Ocaasi 20:26, 8 September 2010 (UTC)
Cautious support I can live with that. If we are going to do a poll about another trial though I would strongly suggest it be better thought out than this one, so that we are not continually dogged by controversy. Revcasy (talk) 20:53, 8 September 2010 (UTC)

I don't think there was "widespread support" for pending changes. The instruction was very clear and parties agreed that the trial only lasts 2 months (no if/and/but). We need to end pending changes first, then discuss what to do with it. Not doing so is generally frowned upon because you're pretty much breaking a promise. OhanaUnitedTalk page 22:05, 8 September 2010 (UTC)