Wikipedia talk:Flagged protection and patrolled revisions/Poll

Discussion on comments
What about having comment-less votes similar to the recent CU / OS elections? Comments can go up here (or this section can be moved to the bottom). Thoughts? --MZMcBride (talk) 19:52, 17 March 2009 (UTC)
 * That would be good, if this was moved to the bottom. –Drilnoth (T • C) 20:05, 17 March 2009 (UTC)
 * I think we should allow comments with the votes for now. It gives useful feedback, and in case serious problems with the proposal are discovered this poll might be closed and a new one started after modifying the proposal. --Apoc2400 (talk) 20:07, 17 March 2009 (UTC)

Is there a set timeframe for this poll? I think there probably should be, so that it doesn't go on for some ridiculous amount of time. A week, maybe? There's been so much attention already that it shouldn't take long. Maybe a month? –Drilnoth (T • C) 20:05, 17 March 2009 (UTC)
 * A week seems reasonable, esp. for just a trial. --MZMcBride (talk) 20:12, 17 March 2009 (UTC)
 * We need to get a watchlist notice for this poll and I would say giving it one week from the time that the watchlist notice goes up seems reasonable. Davewild (talk) 20:16, 17 March 2009 (UTC)
 * A bit longer would be fair, at least it shouldn't be closed as successful too early. Implementing something that people haven't got a chance to comment on will create a backlash. We could ask the developers to start looking into the technical implementation aspects before we reach a final decision though. --Apoc2400 (talk) 20:17, 17 March 2009 (UTC)
 * (edit conflict) Watchlist notices should be discussed at MediaWiki talk:Watchlist-details, though in my opinion a notice is overkill for just a trial and should not be done. --MZMcBride (talk) 20:18, 17 March 2009 (UTC)
 * (@ Apoc) I would agree were this a discussion on full implementation, but this is just a two month trial. Do we really need to take a month to poll for a two month trial...? :) –Drilnoth (T • C) 20:20, 17 March 2009 (UTC)
 * The last poll was also for a trial and had 720 votes in total. --Apoc2400 (talk) 20:25, 17 March 2009 (UTC)
 * Yes, and that poll had a fairly large majority of supporters. This poll is just on one implementation of the result of that poll. –Drilnoth (T • C) 20:45, 17 March 2009 (UTC)
 * That poll certainly did not receive concensus. Looking at the analysis of the results one sees that admins were in favour and very well represented in the poll. Other editors were just about evenly split. IP editors were not really represented. A one week poll will inevitably skew the representation even further making for an unrepresentative poll. The idea that because something is temporary it needs less discussion is ridiculous: to paraphrase another poster "I didn't think I needed to ask you honey, it was just a two month affair!" Thehalfone (talk) 08:15, 18 March 2009 (UTC)
 * No poll or policy discussion is ever "representative" like that. Claiming that IPs weren't represented is a rather weak argument against a consensus. Mr.Z-man 18:36, 19 March 2009 (UTC)
 * You are right, of course, that IPs and newly signed up editors tend not to take part in policy discussions. It seems more important, however, when it is a proposal that will affect their edits but not those of long term account holding editors. Note that there was no concensus found in the previous discussions or polls. Thehalfone (talk) 09:50, 21 March 2009 (UTC)

The proposer of the trial says "Flagged protection uses similar techniques, but it's very different in spirit from 'classic' flagged revisions." This being the case when was this proposal discussed? Are having a vote on this without discussing it first? If so, then why? I think those who support this proposal should be careful or the will appear to be rushing this through without proper scrutiny: going straight to a vote, making yet another page rather than using the multiple pages already there for discussing implementations of flagged revisions, pushing for the vote to be without comments even, trying to close the poll early. Thehalfone (talk) 08:15, 18 March 2009 (UTC)
 * This has been discussed extensively (or at least the controversial part has). Polls on that page have had a tendency to be lost in the flood of nonsense from zealots on both sides, as well as people asking questions and arguing over technicalities (there have been about 3 polls there on matters of technical implementation, and they were mostly ignored).  A poll on an actual trial will be huge and merits its own page.  This is not "going straight to a vote" by any measure, as the proposal has been around for quite some time, and all it has produced thus far is a large amount of arguing over whether flaggedrevs is acceptable in any form.  Thus a poll on a separate page is warranted.  The combination has also been around for some time, being a descendent of flagged protection.  -- Thin  boy  00  @045, i.e. 00:04, 19 March 2009 (UTC)
 * Thanks for the pointer. One problem is that there are so many pages discussing various implementations of flagged revisions that many people will not come across this even if they are actively looking for it. Note that there is no link to here from Flagged protection. I will add one now. Thehalfone (talk) 09:50, 21 March 2009 (UTC)

Comment on this vs. new page patrolling I've seen some comments here and elsewhere that tries to compare this to new page patrolling. Please correct me if I'm wrong here but the way I've understood this is that all it would take is a minor modification of Huggle. If a reviewer is checking an edit the way she/he would do anyway with the current system that person can either approve it or not which technically isn't much different from rolling it back or not. This is nothing like new page patrolling and those who currently patrol BLPs with Huggle shouldn't be doing so anyway unless they can tell the difference between a BLP violation and a valid edit. Once we have the proper tools in place to review edits this won't be any different from what we're doing now. I'm not seeing the extra workload. I don't think this system is good enough to deal with the problem but at least it's a step in the right direction. EconomicsGuy (talk) 08:46, 18 March 2009 (UTC)

So, when should we close the poll ? It started on March 17, if we say two weeks, we could close this at 00:00, 1 April 2009 (!..). Cenarium (talk) 18:29, 23 March 2009 (UTC)
 * Isn't it a bit late for this decision? It almost sounds like the plan is being made up as you go along. 203.24.135.66 (talk) 21:52, 23 March 2009 (UTC)
 * Well, there was some planning above, but it didn't really get much input. Personally, I think that April 1 is a good time... that's still another week or so, and the number of new !votes already seems to be dying off a little bit. The question then is to determine what the consensus from the poll is... at this time, I'd say "trial" since there's 1/5th the number of opposers as supporters. I know, I know, it's a head count, but I think that it would be impossible to determine consensus based on the quality of a person's argument unless the person determining consensus is completely neutral on the issue... which I think is more or less impossible. –Drilnoth (T • C) 21:58, 23 March 2009 (UTC)
 * Support has been hovering around 85% since the day this opened. What further information do you expect to gain in another week? Better to spend that time sorting out the fine details. 203.213.2.194 (talk) 03:24, 24 March 2009 (UTC)
 * Basically, if we closed this poll now, after just one week, there would inevitably be people saying that they didn't get a chance, because they didn't know about it. If we give it two weeks, that complaint loses a lot of its weight. –Drilnoth (T • C) 13:23, 24 March 2009 (UTC)
 * Still keeping steady at 86.16% supporters. –Drilnoth (T • C) 16:30, 24 March 2009 (UTC)
 * I agree with another week (ending april 1), I just stumbled across the poll via a link through this weeks 'signpost' that I read off another users talk page.....so, there might be an interest/voting bounce. --Rocksanddirt (talk) 19:58, 25 March 2009 (UTC)
 * It's down to 85.8% now, so if this drastic trend continues we'll be below 70% in only 9 more weeks. 203.213.2.194 (talk) 02:37, 26 March 2009 (UTC)
 * OH NO! :) –Drilnoth (T • C) 12:32, 26 March 2009 (UTC)
 * Incidentally, why did the poll only make the watchlist 48 hours before the poll is due to end? Something similar happened with the last poll where there was 3 up at the same time and several people, myself included, went to the wrong one(s). Orderinchaos 15:55, 30 March 2009 (UTC)
 * Because someone added it like that without discussion. Cenarium (talk) 17:11, 30 March 2009 (UTC)


 * I'm curious: What about this proposal is "beuracratic" in the eyes of the opposers? It just allows IPs to edit more pages and provides another means to monitor articles which is no more complicated then what is currently used for new page patrolling (probably less, actually), and it will help cut down vandalism. –Drilnoth (T • C) 16:13, 30 March 2009 (UTC)

The same pattern is evident in this poll as in many RFA's. Positive SUPPORT edits are left alone. Editors who OPPOSE the measure are asked over and over again to justify and defend their opinions. IMO this is a nasty, bullying, biased approach to opinion gathering and should not be allowed to happen. Wanderer57 (talk) 16:46, 30 March 2009 (UTC)
 * I think it's a matter of who pays attention. Supporters are more likely to watch the poll page than the opposers, I'd think. It's kinda the reverse on AFDs -- the deleters will badger the keepers like that. And this is WP:NOTVOTE, there's nothing inherently wrong or bad faith about it. ♫ Melodia Chaconne ♫ (talk) 17:07, 30 March 2009 (UTC)
 * From my understanding most of the opposers didn't seem to be opposing this proposal but something else, be it flagged revisions or a version of this proposal which didn't do what they thought. This will (in my understanding) take the locks off every page, but stop some edits from being viewable to the public (and journalists who delight in Wikipedia's errors). The phrase "anyone can edit" isn't accurate at this moment, if you want to test that log out and hit random article ten times and see how many you can edit. The opposers are getting questions because the first batch kept talking about flagged revs, and editors were seeking to make sure they were opposing this proposal and not something else. And there is no need to question a support as it is obvious what a support vote means. Darrenhusted (talk) 13:44, 31 March 2009 (UTC)

Some of the "opposes" here are very well thought out, but I've started to feel that a number of opposers don't fully understand the proposal (e.g., how will this limit the ability of IPs to edit?). It probably won't matter much (the poll still has 82% support), but is something which I thought was pretty interesting. I just hope that if/when this is enabled for the trial, everyone at least gives it a good try to see how it goes and don't just leave the wiki because "flagged revs was turned on." –Drilnoth (T • C) 21:16, 30 March 2009 (UTC)
 * regarding I've started to feel that a number of opposers don't fully understand the proposal - Perhaps a large number do not fully understand the proposal and that's why they support it? Or that a large number know they don't understand it but are willing to try it out as an experiment. --Marc Kupper&#124;talk 02:25, 1 April 2009 (UTC)

Just a note, I have taken the liberty to transclude the voting section onto a seperate page, in case some people have missed the fact that the poll is already over. Additional comments about this proposal should be made in the comment section instead. 山本一郎 (会話) 08:01, 1 April 2009 (UTC)

Consensus Time
I've flicked the 'wiki tech' mailing list a notification of the above, because it seems to me that this is clear enough, with strong enough support to warrant flicking the switch. If I could work out how to link directly to my message, I would - but basically I've asked the technical folk to have a look at the proposal and Poll again, and if they concur that support is such that the extension can be enabled, to do so. We'll see how that goes :-) cheers, Privatemusings (talk) 02:46, 26 March 2009 (UTC)
 * Current count is 152/27/4, which is 83% supporting the proposal. I think Erik said the bar to pass was 67%.   MBisanz  talk 05:08, 26 March 2009 (UTC)


 * Wait ... we have many things to work out before the technical implementation, see Wikipedia talk:Flagged protection and patrolled revisions. Cenarium (talk) 10:26, 26 March 2009 (UTC)
 * So what is the time frame for that to take place? 219.90.243.174 (talk) 10:31, 26 March 2009 (UTC)
 * Consensus is almost certainly not going to change, but I guess I would support keeping this open for a full two weeks as suggested in the top section. Not a huge deal, but may as well let more votes trickle in and, more importantly, give more folks a chance to become aware this is being discussed. --Bigtimepeace | talk | contribs 06:01, 27 March 2009 (UTC)


 * For starters, whilst I support the proposal and consensus is unlikely to change. If the vote is was meant to last 2 weeks, IT LASTS 2 WEEKS. No if buts or maybes on that one. Privatemusings please direct yourself to WP:TROUT (jk) for being a bull at a gate, you should have waited the entire 2 weeks before the email was sent.  « l | Ψrometheăn ™ | l »   (talk) 11:55, 27 March 2009 (UTC)


 * To be fair to Privatemusings, the poll was opened without a clear time frame for its closure. This fits in with the proposal itself: it is, after all, a poll on a proposal that has not yet been worked out. Thehalfone (talk) 08:55, 29 March 2009 (UTC)

We were also going to get a watchlist notice up about this. That can be suggested at MediaWiki talk:Watchlist-details. Would anyone mind if I open a discussion there have the watchlist notice up for a week or so? There are implementation details to work out anyway, so there is no hurry to close this early. If you want to speed this up, get over to /Trial and /Implementation and help out. --Apoc2400 (talk) 10:07, 29 March 2009 (UTC)
 * ah, well turns out (being an idiot) I'd forgotten to subscribe to the list so I don't think the message got through. I've forwarded it on regardless, and hope the switch can be flicked :-) Privatemusings (talk) 10:37, 29 March 2009 (UTC)
 * Please forward there that we need to clear up and agree on some technical and policy aspects before turning on the implementation, I think it will be done by the second part of April.
 * Too late for a watchlist notice, that wouldn't be fair to add it when it's already at 172/33. The discussion and poll at the end of the trial on the continuation of the implementation will have to go there tough. The trial and implementation discussion will be advertized as soon as the poll ends. I don't see much need in continuing the poll now that the consensus is strong and clear after almost two weeks. Cenarium (talk) 15:20, 29 March 2009 (UTC)
 * a 'bug' is submitted! - I don't expect it'll get flicked on immediately, Cenarium - it's my experience that these things can take a surprising amount of time. However, I hope flagged revisions will be on before very long, and we can dip our toes in...... Privatemusings (talk) 20:24, 29 March 2009 (UTC)
 * I think that we can wait another two days for the poll to be over... –Drilnoth (T • C) 20:30, 29 March 2009 (UTC)
 * Specifically, the devs want to have their Berlin conference first then do big splashy things like flagged revs on en:wp. So the bug is technically two days early, but that's not actually a problem - David Gerard (talk) 21:47, 29 March 2009 (UTC)
 * No point to open a bug if the devs don't know what they have to turn on, so marked as LATER. The implementation details are not completed yet, they will be provided when agreed upon. And as I said some policy and trial conduct aspects need to be discussed before requesting implementation. Privatemusings, avoid screwing up the trial with unconsidered hasty actions, please. Cenarium (talk) 22:27, 29 March 2009 (UTC)

< you've done some fantastic work in helping this forward, Cenarium, and I for one appreciate it - my feeling is that this extension is urgently necessary (for a variety of reasons), and further, that it's been a long time coming. The trial and implementation pages seem in reasonable shape to me, but I'll try and help out where poss. If you'd enjoy being the submitter of the bug which flicks the switch, then by all means delete / remove my request, and when you feel the time is right, submit another (I'd strongly support this co-inciding with the dev.s getting back from their conflab in berlin :-) - just please please please make sure this extension gets enabled asap ;-) Privatemusings (talk) 00:15, 30 March 2009 (UTC)
 * No problem. I'll try to hurry up the discussions on this. I don't know of a way to withdraw a bug request, and it's the oldest one that has priority on duplicates I think. But really, this would have only been for personal gratification, that I can find elsewhere, so nothing important. Cenarium (talk) 00:49, 30 March 2009 (UTC)

NOTE: Consensus does NOT DEPEND ON HASHING OUT TECHNICAL NITTY GRITTY, before anyone tries to play such a gauche card. Support is support, the end. Nitty gritty can be sorted out after, if a binding decision is made per Erik. rootology ( C )( T ) 14:08, 30 March 2009 (UTC)
 * the way I read it there is very little nitty gritty to work out. I say, fire it up and work out the kinks during this trial.  If we decide then that policy changes are needed for smooth implementation of a broader program, we have time to do them.  --Rocksanddirt (talk) 18:50, 30 March 2009 (UTC)

one/two-day watchlist notice?
I'm not sure why the Watchlist notice wasn't up immediately. I voted against FR in January and would have again right away if I knew about this poll, which I didn't. All Hallow&#39;s Wraith (talk) 20:30, 30 March 2009 (UTC)


 * (I added the subtopic over Hallows' Wraith) Yes, I've been wondering what happened. Why's this showing up on watchlist with no time left to study the proposal? Proofreader77 (talk) 21:49, 30 March 2009 (UTC)
 * I think one of the problems with the watchlist notice is that it has a tendency to invite people with strong prejudice who do not bother to read the proposal and think and vote right away based on gut feeling. This has happend even without the advertisement on watchlist details, both on supporter and opposer side. For example, as a oppose reason, it's understandable if the proposal is too complicated since there is so many details that needs to be worked out or the proposal will end up doing nothing productive because the scope is too narrow, etc. But, saying that the proposal is restricting anon editors right's is clearly a sign of not reading the proposal and thinking about the implications of it. This poll has been on WP:CENT, so it should be fairly obvious if you are type of person who keeps track of wikipedia proposals reguarly. 山本一郎 (会話) 23:12, 30 March 2009 (UTC)
 * I'd say this falls under WP:CANVAS. Those of us interested in policy or know someone who is learn about the poll and others don't.  As one could argue this is a poll leading down a path of disenfranchisement of those more "normal" users, the lack of a watchlist notification is a real problem and may put into question the results of the poll.  Do we really think less people care about this than last time?  Or were they just unaware? Hobit (talk) 12:29, 1 April 2009 (UTC)
 * You're right about one thing, this poll produced abosolutely no result whatsoever, I think this should be implemented through discussion/revision of the proposal rather than votes, like how PROD was implemented. I can say with confidence that more than half of the people here are not voting on the same proposal, which explains that most of the arguments don't even relate to this proposal because people are arguing completely different things. 山本一郎 (会話) 21:46, 1 April 2009 (UTC)
 * On the 23rd of January a poll on an implementation of closes. It had been open for a month and placed on the watchlist. It showed that while admins were strongly in favour, other users were fairly evenly split. Some weeks later there is a new poll on a different implementation of the same software. It lasts only two weeks, isn't added to the watchlist (or only by mistake and for a couple of days), isn't in fact linked to from the pages describing this implementation for half of this time. Result: much less participation and a strong majority for going ahead. While the difference in result could be due to the fact it is a different implementation, it could also be because this poll was not so widely known and difficult to find. If we find, as I fear, that the "for" votes in this poll and the last poll have a strong correlation and the same with the "oppose" votes - but with many missing. I think we will have to conclude that many of those "opposers" have been disenfranchised. Thehalfone (talk) 11:07, 3 April 2009 (UTC)
 * Most of those who had opposed the previous poll and !voted here support this poll, check yourself: 1, 2, even some of the strongest previous opposers. Flagged protection is to 'classic' flagged revisions what is semi-protection to banning anonymous editing, the results are not surprising. It has been linked from WP:CENT, village pumps, etc, watchlist notices are less used those days. Cenarium (talk) 18:55, 3 April 2009 (UTC)
 * If I'm counting right, most of those who !voted last time didn't show up here. I think 600+ votes last time and 300+ votes this time.  And watchlist notices are still used (the datelinking had one and is pretty clearly less important this this). Hobit (talk) 00:20, 4 April 2009 (UTC)
 * The date linking issue is blown out of proportion indeed, but the vast majority of discussions are not there, and many are more important than this proposal, that is only a trial. But I said "Most of those who had opposed the previous poll and !voted here support this poll". Cenarium (talk) 02:17, 4 April 2009 (UTC)
 * Ah, so you did. A valid point, but I still think this didn't reach a great percentage of folks who would be interested and those it didn't reach may well have very different opinions than those it did. Hobit (talk) 13:36, 4 April 2009 (UTC)

quick questions by watchlist-notified latecomers

 * Will rollbackers automatically be added to reviewer group (for the trial)? (Excuse my asking rather than interpreting, April 1 is soon upon us) Proofreader77 (talk) 02:23, 31 March 2009 (UTC)


 * (found it/not yet determined): "If this proposal is adopted, requirements for autopromotion to reviewer status would be determined later by the community." Proofreader77 (talk) 04:03, 31 March 2009 (UTC)
 * The discussion on more exact implementation of the proposal for the trial can be found at Flagged protection and patrolled revisions/Trial and Flagged protection and patrolled revisions/Implementation, where discussion is currently open. –Drilnoth (T • C) 12:14, 31 March 2009 (UTC)

Where was the notice?
I voted against this several months ago and did not see any notice this was being discussed again. Why was this not publicized more effectively? I only found out about it via the Signpost, which was published today. I watch many Flagged revisions-related articles and saw no notice of such a discussion on any of them. Was it so that only those in favor of such a change to Wikipedia would know such a discussion was being proposed? The relatively small number of voters, and the tilt in favor of implementing it seems to indicate this. This is highly problematic and most undemocratic. Badagnani (talk) 03:48, 26 May 2009 (UTC)


 * Ditto. I was shocked to read recent news articles about flagged revs and see that 80% were in favor, when I had clearly remembered voting in a poll nowhere near that favorable to flagged revs. Well, I guess this being a new poll explains it...
 * Not to mention, only ~400 votes for something as permanent, entire-wiki-affecting, and antithetical to 8 years of open editing? That seems a little low to me. --Gwern (contribs) 00:00 26 August 2009 (GMT)
 * There were notices all over flaggedrevs-related pages, the village pumps and WP:CENT. You can't always notice everything though. This poll is about a two-month trial detailled at Flagged protection and patrolled revisions, which is not classic flagged revs at all, and nothing else. So this is not pemrnanent, not entire wiki-affecting, and not antithetical to open editing since we replace protection (can't edit at all) with flagged protection (can edit, visible when flagged), and the global patrolled revisions are entirely passive. Cenarium (talk) 00:23, 26 August 2009 (UTC)

Taking into account the four month difference, flagged revisions in not the same as flagged protection and patrolled revisions, for Badagnani (four months too late), and for Gwern, the poll was included as a dismissable notice when users logged on for the period the poll was running, which would have been March 17 to April 1. It was also posted at various other forums. If you look at the main project page you will see that the trial will be narrowly applied and once over will be analysed to determine if it should be rolled out across the whole of WP. And bare in mind that although voting may be evil, wikipedia is not a democracy, even if a thousand editors opposed this it could have still been implemented. Darrenhusted (talk) 00:18, 26 August 2009 (UTC)
 * There were notices all over - you mean, the notices that looked exactly like the ones for Wikipedia talk:Flagged revisions/Trial/Votes (which you'll notice I did see and act upon)? I checked one random screen of the original poll's oppose votes and notice that a good third of them simply don't show up in the second vote. How many of them were like me, I wonder, in not realizing there was a second poll? (I guess flagged supporters didn't like their only 59% support and decided to keep modifying and polling until the numbers finally came out right.)
 * As for them not being exactly the same, yes, that is a tremendously meaningful distinction; I have seen the light and will be voting for it! Give me a break. Flagged revisions is flagged revisions, whatever banner it may fly under or how initially restricted it may be. I remember how semiprotection was approved under the pretense that it would only ever be used to replace full protection, yet I routinely see pages semiprotected that wouldn't've been full-protected. And as for the trial being permanent, well, you'll pardon my skepticism inasmuch as I was around for Seigenthaler; the temporary state of emergency has lasted a good 3 years now. And did the study the Foundation promised us ever materialize? No, of course not. I place as little trust in these similar promises of the results being 'analysed'... --Gwern (contribs) 03:21 26 August 2009 (GMT)
 * I do share some of your concerns. There's been no watchlist notice for this poll. But at the end of the trial period, we'll discuss and decide on whether to adopt the implementation and it'll be massively advertised. If available at that time, we could even use a talknotice or a (dismissible if possible) global editnotice. Cenarium (talk) 03:35, 26 August 2009 (UTC)
 * What makes no sense is that a test has been approved, but what is being tested has yet to be defined. It will be shameful but not surprising if the test starts before there's a defined and operational reviewer right granting process. That is putting the cart before the horse; I hope the community decides not to do this ass-backwards.   --Elvey (talk) 17:30, 27 December 2009 (UTC)


 * Same here. My vote would have been against it if I have know about the poll. Jason Quinn (talk) 19:46, 26 June 2010 (UTC)

what was and was not approved

 * the actual proposal is not the one I thought I was voting to support. I thought I approving a trial of a single version of flagged protection on selected pages; I find instead from Flagged protection and patrolled revisions  that I have voted to approve a trial of three additional levels of   protection  with totally different effects patrolled by different classes of users on variously sleected pages. This is against all principles of a rational trial, to make three separate changes at once. I cannot imagine explain the simultaneous existence of 6 different classes of articles to new users--I am not sure I understand the distinction myself. A trial of a single would make more sense, and I urge the community to reconsider. Perhaps nobody should be allowed to vote to approve the full trial unless they can explain accurately the various possibilities without looking at the text.  I think I know how I am going to handle this: I will ignore the existence of the feature entirely,& edit in my usual & I hope responsible way; what is or is not visible in those articles covered by the system can be the concern of those who think they do understand.    DGG ( talk ) 16:14, 6 January 2010 (UTC)

Scope, deployment, and removal
See: Pending changes -- verdy_p (talk) 05:52, 14 October 2010 (UTC)