Wikipedia talk:Autoconfirmed Proposal

Mainspace?
Does the edit threshold include edits made to non-mainspace? If it does, it's quite easy for a user to just make ten edits in their userspace and be done with the matter. We have some seriously persistent vandals on the 'pedia; they will do that kind of thing. A block option that would reset the edits accumulated for semi-protection purposes would be also be great... Okay, okay; I can deal with what we have; I'll put the edit threshold at 17-23. --  tariq abjotu  01:46, 18 June 2007 (UTC)


 * If you're determined, it's pretty easy to rack up a bunch of mainspace edits quickly--just revert vandals, for instance. For this reason the threshold should be on the high end. --Akhilleus (talk) 01:49, 18 June 2007 (UTC)


 * I support this proposal. I think the threshold should only include edits made to actual articles or their talk pages(Mainspace) opposed to user pages or talk pages(Userspace) if that's possible. However if that would be too difficult I believe 30 a edit minimum before auto confirm occurs would deter a good percentage of these types of vandals who go after semi-protected articles. Wikidudeman  (talk) 02:17, 18 June 2007 (UTC)


 * Hey, if you revert twelve vandal edits to make one, aren't you actually helping in the end?  Citi Cat  16:26, 20 June 2007 (UTC)
 * Only if the vandal edits are actually from a different vandal. 64.126.24.12 18:52, 26 June 2007 (UTC)

I'm not liking this proposal. Users who go to the work of creating a sleeper account and waiting four days to vandalize a semi-protected page will be able to get around this in no time - just adding and removing Italic text from the sandbox a bunch of times will do it. Legitimate users will be harmed, while determined vandals won't be deterred. When we make editing Wikipedia more restrictive - like, for example, stopping anons from creating articles - there should be firm reasons that it will help more than harm. I'm not convinced that this will outwiegh the negatives, and I also dislike the idea making editing more like requests for adminship. Picaroon (Talk) 02:33, 18 June 2007 (UTC)


 * It's a lot more work to create an account and make ten trivial edits than it is just to create the account. This is especially true if you have to do this several times a day, because your sleeper accounts get blocked very quickly when they're used. I think it would be worth trying this. However, I would oppose a high edit count limit.- gadfium 03:06, 18 June 2007 (UTC)


 * I hear your objections however they aren't very realistic. Firstly, editing even the sandbox 30 times plus waiting 4 days isn't as easy a thing to do as you make it seem. Making 30 edits no matter in what form isn't an easy task and there might even be a way to exclude edits from the sandbox from counting. Even if not, 30 edits to the sandbox plus waiting 4 days just to vandalize an article once and get banned as a sockpuppet? The number of vandals who are that motivated simply aren't that high.
 * Secondly, Anon IP's can't edit semi-protected articles anyway. How is it restrictive if we simply ask that they make 30 edits during the 4 days they ALREADY have to wait to be able to edit semi-protected articles? I don't see it as restrictive at all.
 * Thirdly, With this implemented we can decrease the number of fully protected articles thus making wikipedia as a whole less restrictive. Fully protecting articles basically indefinitely is what's restrictive, not asking editors to make 30 edits during their 4 day waiting period before being able to edit semi-protected articles.
 * Moreover, With this implemented anon editors who think an article should be fixed could just request the change on the articles talk page and an established editor can make the changes within a few seconds. Wikidudeman  (talk) 03:07, 18 June 2007 (UTC)


 * It seems some people oppose a low edit count minimum and some people oppose a high edit count minimum. In my personal opinion 10 names pace edits is too low for a minimum and 50 is too high for a minimum. I believe the ideal number of minimum edits would be somewhere between 20-40, preferably 30 edits as a minimum. gadfium, Would you oppose a 30 edit minimum plus the 4 day waiting period?  Wikidudeman  (talk) 03:09, 18 June 2007 (UTC)
 * I would vote for a 10 edit minimum rather than a 30 edit one, but I would not vote against the proposal if the count was 30.- gadfium 20:13, 19 June 2007 (UTC)

I can see some positives to this. It certainly would've been nice with the Ararat arev army of socks we had a month or so ago, where he'd create sock armies and age them to get around semi. Wouldn't have stopped him, necessarily, but it sure would've slowed him down and made it a lot less convenient. (The captchas seemed to solve that issue, though, it seems he was making accounts by script.) As always, we should use any form of protection sparingly, only where necessary, and for as short of a time frame as possible, but there's nothing wrong with making sure it's effective when we do. It wouldn't be hard for a vandal to create a sock and make ten sandbox edits, but it would be tedious to make a hundred socks that way. Even if they used a bot or script to do it, it would create a pattern we could watch for in the future. And doing it by hand would quickly overwhelm the "haha fun" factor of it and get them to quit. Seraphimblade Talk to me 03:24, 18 June 2007 (UTC)

I'm all in favor. The only effective tools we have to combat our unacceptably high level of vandalism are those that skew the amount of tedium in favor of those improving the encyclopedia. (In particular, blocking by itself is worthless.) Considering that&mdash;at least as of MediaWiki 1.9.3, which is what I have installed locally&mdash;the user_editcount field includes deleted edits, I'd greatly prefer the minimum edit count to be at least in the 30-50 range. &mdash;Cryptic 07:15, 18 June 2007 (UTC)


 * I think that this will help us deal with the problem. No, it's not a perfect solution, and vandals who are motivated enough can get around it easily, but it will help. Considering the amount of confusion caused by (see here for details), I think that extrra steps which a user must take before moving pages is reasonable. I also think that it should be the mainspace - otherwise all it would take is to find an open discussion and say I agree with User X - this won't get the user blocked. I think that 30 edits seems reasonable - making it too high would discourage new good users. Od Mishehu 07:19, 18 June 2007 (UTC)

Mainspace only would be better as it eliminates the sandbox/userspace issue. I'd set the threshold high enough to filter out 99.9% of vandalism only accounts, which would be towards the higher end. As an aside, can we restrict mainspace page creation as well, as that would also filter out a lot of the "omg look at me im so kool bcuz i have a wikipedia page!!!111!!eleven!1" garbage. MER-C 09:47, 18 June 2007 (UTC)
 * I think this is a good idea, yes.  &gt; R a d i a n t &lt;  13:29, 18 June 2007 (UTC)
 * Mainspace creation won't count, MER-C. Users can't create pages until they're off the semi-protected phase (this is correct, yes?). --  tariq abjotu  13:33, 18 June 2007 (UTC)
 * As far as I know, all registered users can immediately create new pages.  &gt; R a d i a n t &lt;  13:38, 18 June 2007 (UTC)
 * This is correct. Tariqabjotu, MER-C's suggestion was precisely that users not be able to create pages until they're autoconfirmed.  For myself, I think this is an excellent idea.  Among other things, it would reduce the need to salt pages targeted by folks persistent enough to keep re-creating after a block, but not persistent enough to come back and try some more in half a week. &mdash;Cryptic 13:52, 18 June 2007 (UTC)
 * One of my first edits was a page creation. I would not be here if I had been required to go through any level of bureacracy (AFC or similar) to be allowed to help improve the encyclopedia. Please do not make it even harder for us to attract good users who want to fix a red link. Kusma (talk) 15:44, 18 June 2007 (UTC)
 * Actually, that brings up a good point. Is there a way that we could restrict new page creation to autoconfirmed users only if that page has no incoming links?  That way, if a page is linking to it, it would be able to be created by brand new users.  If not, (say, the user's name or band name or some new Stephen Colbert phrase), then they would have to be autoconfirmed.  -- Renesis (talk) 16:12, 18 June 2007 (UTC)
 * Seems pretty trivial to make a redlink somewhere (Sandbox or otherwise) and then start the page immediately in another tab/window. Anything that tries to check if the redlink has existed for more than X hours/days/years is likely to be too labor-intensive to pay off. -- nae'blis 16:18, 18 June 2007 (UTC)
 * Sure, but that's not the problem I'm talking about. It won't be obvious why it's restricted, it will just be restricted.  99% of our new page cruft is created by people who would never know about all the rules or worry about trying to game the system. -- Renesis (talk) 16:22, 18 June 2007 (UTC)
 * (de-indent) Hmmm, that's an interesting angle, actually. I wonder if that would be a good roll-back from the current draconian limits... -- nae'blis 17:03, 18 June 2007 (UTC)
 * What limits are "draconian"? Both Special:Newpages and WP:AFC are both, to put it bluntly, a load of crap with something occasionally halfway worthwhile popping up. But at least AFC isn't currently adding to the speedy backlog. That's not data that says to me "We need less restrictions on newpage creation," it says to me "We need more." Once Special:Newpages is mostly reasonable stuff, and the crapflood is mainly going to WP:AFC or something similar, then we might say "Maybe we could try easing up a bit and see what happens." It's actually far more bitey to let a new user create a page only to speedy it, then it is to say "Look, you can't just create whatever the hell you want, so we ask you stick around a bit and learn a few basics before you create pages on your own. If you're sure you've got it down already, articles for creation is this way." Remember, the people who are having to tag and speedy this crap are people who are known to be good editors with the good of the project at heart. What could they be doing with that time, if suddenly the speedy backlog were cut to half? To a tenth? To be very honest, losing the very occasional good editor who immediately creates a new page would be worth it to free up such a vast amount of our most precious resource&mdash;the time of good editors. Seraphimblade Talk to me 00:41, 19 June 2007 (UTC)
 * No need to be condescending, I'm one of those users that does speedy deletion from time to time. But obviously the current limits aren't very effective if it's still getting a crapflood. Did you read the latest iteration of the proposal above? New pages without incoming redlinks would be limited to autoconfirmed accounts, while we possibly opened up new page creation for existing links to all/more users. There's a decided chilling effect on contribution as an IP lately, which I didn't discover until I started editing while logged out again. -- nae'blis 03:24, 19 June 2007 (UTC)
 * Correct me if I'm wrong, but currently a user needs only to be registered, not autoconfirmed to create a page, correct? If that is the case, I believe my proposal above would be slightly more restrictive but not detrimental, leaving us with a higher percentage of good new pages. -- Renesis (talk) 06:47, 19 June 2007 (UTC)

On abusive socks
This proposal will make abusive socks harder to detect. It will be trivial to get around the requirements and get the requested number of good edits in 10 minutes (or less if using automation). Now the sock has 10 edits to articles unrelated to his topic of interest and starts editing it. When we think about blocking the person, we now have to check all these edits, assume they were bad faith attempts to get around the editing restriction, and then block for sockpuppetry. With the classic autoconfirmed, we can block straight away. Kusma (talk) 09:35, 18 June 2007 (UTC)
 * And what exactly does that immediate block accomplish? &mdash;Cryptic 09:51, 18 June 2007 (UTC)
 * It autoblocks the IP. My point was that we don't ask all puppeteers to cover their tracks better. Kusma (talk) 09:57, 18 June 2007 (UTC)
 * IP autoblocks last only 24 hours, so we have to reassess for sockpuppetry every day for every sufficiently dedicated puppetry anyway. (More for those without fixed ips; for these, blocking them hurts us more than it hurts them, precisely because of the work needed to reassess their next incarnation.)  If we require thirty or fifty edits before being confirmed, then the encyclopedia either gets thirty/fifty good edits, or we have enough time to block and reset their clock before they come back to the article they were disrupting. &mdash;Cryptic 10:09, 18 June 2007 (UTC)
 * How do you know these are good edits? You can make 50 whitespace edits with a bot and nobody will even notice. Kusma (talk) 10:27, 18 June 2007 (UTC)
 * Depends whether our sockpuppeteers have the technical competence required to write a bot, which isn't very likely due to the amount of effort involved - i.e. downloading the framework, compiler (if required), learning the language, figuring out how to use the framework, writing the program and compilation/testing (if required). I'd go as far as saying that at least 99% of vandals can't be stuffed - Pywikipedia's been around for ages and I've only seen a couple of incidents of automated vandalism. MER-C 09:11, 19 June 2007 (UTC)

Retroactivity
Another question: Will this be applicable to accounts registered before the implement date (if there is one) which are older than four days, yet have not made n (mainspace) edits? MER-C 11:33, 18 June 2007 (UTC)
 * Not as currently written. (For reference, the relevant bit of code is getEffectiveGroups in User.php.) &mdash;Cryptic 11:42, 18 June 2007 (UTC)
 * Why would that be important, anyway?  &gt; R a d i a n t &lt;  13:29, 18 June 2007 (UTC)
 * Some random thoughts about preventing previously unknown sleeper socks from circumventing the rule. For example I remember a Willy on Wheels sock that had not edited for two years before commencing move vandalism, though the username escapes me. MER-C 08:50, 19 June 2007 (UTC)
 * Even if inactive currently auto-confirmed accounts remain auto-confirmed, this should help us solve more problems than it creates. There is a small problem which may still be used to try and bypass the edit requirement before we implement it, but I don't think it will be a serious probelm. Od Mishehu 12:10, 19 June 2007 (UTC)

Debate

 * Resounds of the Quasi-protection policy, the problem with that being that it woudl be too much work. If it's autoconfirmed, though, it would be no problem. hbdragon88 04:34, 18 June 2007 (UTC)
 * So they login add some minor punctuation/capitalisation on 10 pages and are good to go, I see plenty of vandalism only accounts which have amassed much more than 10 edits. --pgk 06:31, 18 June 2007 (UTC)
 * It does set the bar a little higher without setting it too high - increases the workload ofr vandals and doesn't seriously affect non-vandals. I don't see what the problem might be. Viridae Talk 06:35, 18 June 2007 (UTC)
 * I don't see a problem, I just don't see a purpose either. If you want to categorise vandals into the most casual who the current system prevents and those who are slightly more dedicated and will happily wait 4 days, then I can't see making 10 edits would make any difference to those people. --pgk 09:22, 18 June 2007 (UTC)
 * On the other hand, Wikipedia would gain 10 more good edits. Uncle G 11:59, 18 June 2007 (UTC)
 * Depends on your definition of good. Add a ".", remove a "." do that 5 times over the 4 days. As I say many vandalism only accounts get blocked after over 10 vandalism edits anyway. --pgk 12:16, 18 June 2007 (UTC)
 * The general problem with auto-confirming is that it is automatic, so by definition it can and will be gamed. It is trivial to get around for a vandal with technical knowledge, and will annoy the crap out of good faith newbies. If you want an access level that actually means something, you have to grant it manually. Kusma (talk) 09:30, 18 June 2007 (UTC)
 * I don't think it would hurt the project, and it might discourage some vandals (if we stop telling them what they need to do to get autoconfirmed). But I expect the effect to be minimal. So why not? -- lucasbfr talk  10:02, 18 June 2007 (UTC)
 * The above moved from WP:AN.  &gt; R a d i a n t &lt;  13:30, 18 June 2007 (UTC)


 * I agree with Kusma; any automated process can be gamed, and any increase in threshold will keep away some number of good editors. Any manual process will be too labor-intensive and require a presumption of bad-faith (remember, the disabling of page creation by anons was originally supposed to be an experiment). Now I can't even make a needed redirect while logged out. Try logging out and editing as an IP for a while, and observe how much harder it has already gotten for good-faith editors to operate here. --nae'blis 15:13, 18 June 2007 (UTC)


 * blis, See my responses to objections. All this proposal is is requiring that new editors be required to make a minimum of 30 edits in addition to the 4 day waiting period before they can edit semi-protected articles. This would NOT affect anon IP editors at all since they can't edit semi-protected articles to begin with. All it would do is make it a little more difficult for vandals to vandalize semi-protected pages and would make wikipedia MORE OPEN to good faith established editors in preventing many articles from being on indefinite full protection. Wikidudeman  (talk) 00:46, 19 June 2007 (UTC)


 * One question which hasn't been addressed yet, is about talk-page bots (such as archiving bots), given that some talk pages are both active and semi-protected. In order to get around the problem, I think taht any approved bot (already requires beaurocrat involvement) should automatically be auto-confirmed. Od Mishehu 12:07, 19 June 2007 (UTC)

Response to objections and remedies
It seems that most people are in favor of a 30 edit minimum before being able to edit semi-protected articles. And please remember that this is all we need it for, to fix the problem of sleeper sock puppets vandalizing semi-protected articles and increasing the workload of potential vandals who want to vandalize pages. I want to address some objections
 * 1.This proposal will make abusive socks harder to detect. I don't believe this is necessarily true and even if it makes it a little more difficult to ban abusive sock puppets, It wouldn't be too difficult. It would be incredibly easy to just click "contributions" and see their edits as building up to a vandalism and then we would give them a 3 or 4 strike warning (as I always do when warning vandals) and if they continue to vandalize we would block them as a vandalism only account, excluding the repetitious edits they made to vandalize a semi-protected page. Though remember, This would be very very rare because the number of vandals who would go through the toil of making 30 edits and then waiting 4 (or more) days just to vandalize a few times is incredibly small.
 * 2.It would be too much work for people to manually confirm all new editors. The confirmation would be automatic as it currently is, But instead of simply having to wait 4 days before one is automatically confirmed as an editor they must wait 4-7 days plus make 30 non-vandalism edits. If their edits are vandalism they would get banned before getting auto-confirmed in the first place.
 * 3.It would be easy for vandals with bots to simply add blank spaces and erase them over and over 30 times just to get the minimum amount of edits. This is true, However what % of vandals targeting semi-protected pages could do this? Actually a very small number of them could and the number of vandals who would be unable to do this sort of thing would be prevented from vandalizing semi-protected articles and we would prevent a lot of vandalism even though some vandals could still get around it in rare occasions in which case we simply block them right off the bat once they are identified as sock puppets. It's a lot easier for us to block a vandal in a few seconds than it is for them to program bots to make 30 edits and wait 4 days just to vandalize a page once or twice which gets instantly reverted.
 * 4. This would prevent new good faith editors from being able to edit. This is simply untrue. With this implemented anon editors who think an article should be fixed could just request the change on the articles talk page and an established editor can make the changes within a few seconds without a problem. This would actually make wikipedia LESS restrictive. With this implemented we can decrease the number of fully protected articles thus making wikipedia as a whole less restrictive. Wikidudeman  (talk) 22:26, 18 June 2007 (UTC)


 * 1. No thoughts on this right now.
 * 2. Moot, proposal says automatic.
 * 3. Why would they need a bot? I could make 30 dummy edits to a page in seven minutes, and I'm sure others could do it faster. Much less tiresome than waiting four days, too.
 * 4. s/seconds/dozen hours. Category:Wikipedia protected edit requests is not that fast. Picaroon (Talk) 22:48, 18 June 2007 (UTC)


 * 1. Ok
 * 2. Right. It would be automatic.
 * 3. Yes, You could make 30 dummy edits in 7 minutes without a but but how many vandals would spend 7 minutes making 30 dummy edits? Not a lot. Moreover, Yes, You could make 30 dummy edits faster than waiting 4 days HOWEVER the proposal is to keep the waiting period as well. That means a minimum edit count of 30 PLUS a 4 day (or more, maybe 6 or 7 ) waiting period.
 * 4.I don't understand what you mean by this. Many pages are kept protected (Brock Lesnar for instance) indefinitely due to a single vandal. Requesting it be unprotected don't work (and didn't work for me). Wikidudeman  (talk) 22:54, 18 June 2007 (UTC)


 * This may be pushing the capabilities of the software and or resources a little, but if we could avoid counting edits under a certain size (e.g. adding and removing a space) it would become much harder to get around it with trivial edits. The vandal would have to actually make substantial edits, which would require more thought and time. To do so without getting blocked for vandalism, they would basically have to try to improve the articles, which would be too excruciating for any of them to bear and none would bother. Of course, this would also mean new editors couldn't get through just for minor corrections, but we could reduce the number of edits required to compensate if need be. The main thing is to create a serious obstacle that vandals just won't bother with, while being of little consequence for people who are actually here to improve things. Richard001 05:23, 20 June 2007 (UTC)


 * If that is possible then It would help however it may not be possible and that shouldn't stop us from implementing this into Wikipedia. You need to remember that it's always a good thing to make new editors make a few edits before editing semi-protected articles to begin with, that's a good thing. It's also a good thing for them to learn about Wikipedia while they are making those preliminary edits. They could still edit all other articles but instead of having to wait a few days to be able to edit semi-protected articles they would have to actually spend time learning about Wikipedia prior to doing so opposed to doing nothing and making no edits. Moreover, As I've mentioned several times, if they really wanted to see a change in a semi-protected article they could easily request the change on the talk page and then another user can make the adequate changes, which is of course a lot easier than going through the whole process of requesting an edit to a fully protected article. Wikidudeman  (talk) 05:34, 20 June 2007 (UTC)


 * This section is not in any way relevant to Navou's response below. Kamryn · Talk 23:25, 6 August 2007 (UTC)

A middle ground?
Obviously this isn't going to solve 100% of problems. Yes some vandals will still be able to get around this, and it's not perfect in any way. All that we're trying to do is improve on what we have now, which is just a 4 day wait and nothing else. I also expected the debate between stopping vandals and discouraging newbies, so the idea is to find a compromise between the two (in the number of edits). Again, this will not solve all problems, but sometimes you need to fix something that's not working (which is semi protection right now).

Anyways, the number I was originally thinking of was around 10-15 edits, because it is still a fair few for vandals to do, and it's not an unreasonable number that will discourage newbies from trying. Comments? -Royalguard11 (Talk·Review Me!) 16:15, 19 June 2007 (UTC)


 * The most important aspect to consider is this will actually increase the openness of Wikipedia by lowering the number of indefinite fully protected articles which no one really bothers to go through the bureaucracy to edit. By adding an edit minimum of around 30 edits this will discourage a good % of vandals and will give new users some experience before editing semi-protected articles in the first place. Not only that, It will allow the rest of us to edit articles and not have them fully protected forever due to one single vandal. Wikidudeman  (talk) 16:34, 19 June 2007 (UTC)

This is controversial...
I feel this may be a bad idea, as it could encourage elitism, and take away the "encyclopedia anyone can edit" theory. It could start off the "I've-got-more-edits-than-you" theory.

Isn't it better to be autoconfirmed via number of days rather than edits??

Articles  shouldn't  be semi-protected indefinitely, as that locks out good faith new editors and anons, but we should be making more use of expiry-timed semi-protection, rather than just protecting pages as:
 * , Monday July 29 2024 (UTC) 18:16 Admin protected Volkswagen Passat (spamming from new users) [edit=autoconfirmed move=autoconfirmed]

This proposal is a controversial one I feel, and you might think my opinion is controversial, but it's true that this proposal will spark a lot of debate - it's like the desysopping inactive admins one. --SunStar Net talk 18:16, 19 June 2007 (UTC)


 * Right now there are tons of articles on indefinite total protection . That's more restrictive than anything this new implementation would result in. With this implemented anon editors who think an article should be fixed could just request the change on the articles talk page and an established editor can make the changes within a few seconds without a problem. This would actually make wikipedia LESS restrictive. With this implemented we can decrease the number of fully protected articles thus making wikipedia as a whole less restrictive. Wikidudeman  (talk) 18:40, 19 June 2007 (UTC)


 * I actually think timed semiprotection hurts us more than it helps - while it prevents administrators from forgetting to remove the semiprotection, it still prevents good faith newbies from editing the article in question, it lets the vandal know exactly when he should come back and piss on our project some more, and if it's longer than four days, it immediately tells the vandal that he's better off creating a sleeper then and there instead of trying to wait it out as an anon. If the semiprotection is untimed, the vandal is more likely to get bored when he comes back and tries to edit the article repeatedly over the next few days. &mdash;Cryptic 18:55, 19 June 2007 (UTC)


 * I disagree that there is anything controversial about this proposal, as far as whether it will encourage elitism or detract from the "anyone can edit" idea. This is just a difference in the autoconfirmed level.  This is no more elitist than the semi-protection we have now, and since this will only affect semi-protected articles (already restricted to users with accounts older than 4 days), it won't be any worse in the "anyone can edit" area. -- Renesis (talk) 19:12, 19 June 2007 (UTC)


 * Cryptic, That is true, however by making it more difficult for vandals to vandalize semi-protected articles and giving new users a chance to get in a few edits before actually being able to edit semi protected articles (which they should be doing to begin with) we improve Wikipedia drastically. Moreover, If a new user wants to see something changed it's as easy as requesting it on the talk page which is A LOT easier than going through the process of requesting an edit be made to a fully protected article. Wikidudeman  (talk) 03:42, 20 June 2007 (UTC)


 * Apologies for starting this thread. On another Wikimedia project, where I'm the only admin, I might ask on IRC for the devs to change it so people can't edit semi-protected pages until 4 days later, as we've had page-move vandalism quite recently. --SunStar Net talk 08:59, 21 June 2007 (UTC)

Any consensus?
I wanted to know if anyone had any more oppositions that I could address or if there was any sort of consensus supporting such a change in 'auto confirmed' policy regarding semi-protected articles. This would be a huge obstacle towards persistent vandals and would actually make wikipedia more open by limiting the number of indefinitely fully protected articles which no one but administrators can edit and rarely see any improvements.  Wikidudeman  (talk) 14:58, 24 June 2007 (UTC)


 * I think this is a good idea but it is worth noting that it will do nothing to stop . This banned vandal regularly makes more than 30 edits before he is caught and blocked.  For example, he did so with the sockpuppet account that was caught and blocked today, .  That does not mean this proposal is not a good idea, mind you, only that it will not help in some very limited cases.  I support the proposal.  --Yamla 20:40, 24 June 2007 (UTC)


 * I agree with Yamla, it dosen't stop Verdict and his socks from vandalizing a group of articles. However, it does slow down the vandalism of sockpuppetry to make certain full-protected articles from vandalism maintainable. Seriously though, who would go to the trouble of creating an account, wait for 5 days AND create 30-40 edits, only to be blocked as a sock of a banned vandal? 80-95% of banned users from Wikipedia would be discouraged by the upgrade of semi-protection. It can also endorse boredom of long-term abusers because it is a tedious process to bypass the upgrade of semi--protection.--PrestonH 21:18, 24 June 2007 (UTC)


 * In case it wasn't clear from my comments, I agree that this will stop 80 - 95% of the banned users. I will refrain from commenting on the sort of person who routinely sets up sockpuppet accounts and makes 30 - 40 edits simply to be able to edit a semi-protected article and have all edits reverted under WP:DENY, but  fits this category.  --Yamla 05:12, 25 June 2007 (UTC)


 * Most of edits were constructive and improved various articles. It's true that it was an example of an evasion of a previous ban and by definition could be reverted to enforce the ban, however there is no arguing that the edits themselves had merit. This means that if this user is going to vandalize for instance Brock Lesnar and must create another account with similar edits as  then he is only benefiting Wikipedia by attempting to vandalize Brock Lesnar and thus we could easily revert his edit once he makes it to Brock lesnar and subsequently ban him. This means that the Brock Lesnar article could be put on semi-protection and it would be vastly easier to stop that specific user from vandalizing the article as well as vastly easier to spot him when he does.  Wikidudeman  (talk) 23:03, 25 June 2007 (UTC)


 * All edits by a banned user avoiding his or her ban are by definition also vandalism (so, not constructive). It's also a pain in the behind to revert the edits from such a vandal, not to mention the joy of receiving threats of physical violence and lawsuits.  I'm also not convinced that it would be any easier to determine that the editor is Verdict, but I'm not saying you are necessarily wrong.  Anyway, we should not let one blocked vandal stop us from implementing this policy.  I mean, the policy also won't clean my fish tank on a weekly basis but that is not a significant strike against it.  I believe the policy has merits.  It doesn't solve all our problems but it is a significant step.  --Yamla 04:16, 26 June 2007 (UTC)


 * All edits by a banned user avoiding his or her ban are by definition also vandalism (so, not constructive). This is a very narrow way to see it. Firstly please stop citing WP:DENY, It's not a policy. Secondly, WP:BAN does NOT say that all edits made by banned users is automatically 'vandalism'. It just says that any edits made in defiance of a ban may be reverted to enforce the ban though it doesn't say they must be reverted.
 * It's also a pain in the behind to revert the edits from such a vandal, not to mention the joy of receiving threats of physical violence and lawsuits. You don't need to revert all of their edits, It's just an "option" to enforce a ban. Unless their edits are clearly nonconstructive you really don't have to revert them. It's only an "option" to enforce a ban. That's it. If this proposal is initiated and the Brock Lesnar article placed on semi-protection then Verdict would have to go through the whole process again which would take days and a lot of effort and it would only improve Wikipedia. He would improve Wikipedia with 30 edits just to make 1 vandalism to the Brock Lesnar article which would then subsequently be deleted and he would be banned as a sockpuppet. As far as "Threats" go, That has nothing to do with the protection of Brock Lesnar. Protecting a page won't prevent threats. If you're receiving threats of lawsuit then you should ignore them.
 * I'm also not convinced that it would be any easier to determine that the editor is Verdict, but I'm not saying you are necessarily wrong. I'm not.
 * Anyway, we should not let one blocked vandal stop us from implementing this policy. I mean, the policy also won't clean my fish tank on a weekly basis but that is not a significant strike against it. I believe the policy has merits. It doesn't solve all our problems but it is a significant step. This policy would be a valid reason to make Brock Lesnar Semi-protected opposed to fully protected which is simply harming the article itself. If the only threat to the article would be 1 vandal who makes sock puppets every now and then and must make 30 constructive edits before making 1 vandalism at which points that sockpuppet is quickly banned then there's basically no problem at all.  Wikidudeman  (talk) 05:23, 26 June 2007 (UTC)

It seems like there is a consensus...
It seems to be like there is a consensus. How do we go about making this into an actual Wikipedia process?  Wikidudeman  (talk) 20:58, 26 June 2007 (UTC)
 * I'm just going to repost it to the mailing list so people can come here and talk more, then maybe a straw poll this weekend (I just need to wait a couple of days to get re-autoconfirmed on the list, because mail from non-subscribers is help for moderation and mine's been help for the last week). -Royalguard11 (T·R!) 01:06, 27 June 2007 (UTC)
 * Updates would be great.  Wikidudeman  (talk) 20:56, 28 June 2007 (UTC)

Advantages and disadvantages
The pros and cons of this proposal: Pros Cons
 * Stops vandalism - may deter run-of-the-mill vandals, but not really determined ones
 * Encourages people to start contributing
 * May stop sleeper accounts.
 * Not useful Wikimedia-wide, e.g on Commons etc.
 * Will not deter spammers/spambots
 * Semi-protection is effective most of the time.
 * Long-term vandals may eventually get bored of semi-protection (which has worked on articles like George W. Bush and Penis)

I would say this proposal is an OK idea, but the current solution of waiting 4 days seems OK for now... --SunStar Net</b> <i style="color:blue;">talk</i> 17:23, 1 July 2007 (UTC)


 * I don't know if it will be useful on Commons or not. Why won't it deter spambots? The bots will have to make the 30 edits and wait 4 days as well as anyone else. Long term vandals will get bored of attempting to vandalize semi-protection? That's a good thing. Then they vandalize normal articles and get banned quickly and thus we have less indefinitely fully protected articles on our hands such as Brock Lesnar which makes wikipedia more open. Some of your "CONS" might be valid but most aren't, moreover, there are a lot more PRO's listed above.  Wikidudeman  (talk) 18:43, 1 July 2007 (UTC)

News
Is there any news on this?  Wikidudeman  (talk) 16:06, 17 July 2007 (UTC)

Update
I would appreciate if someone tried to get this back out there by posting it somewhere for more people to comment on. It seems to have gotten a consensus in support.  Wikidudeman  (talk) 09:44, 27 July 2007 (UTC)
 * WP:RFC perhaps? I only happened on it because of the Brock Lesnar article recently and had never heard of it before that.  So far, the oppose reasoning has been quite unconvincing to me.  —Wknight94 (talk) 17:09, 27 July 2007 (UTC)
 * Right, I think I've addressed all of the oppositions very concisely. Add it to RFC and post a link here for me. Thanks.  Wikidudeman  (talk) 17:11, 27 July 2007 (UTC)
 * Uh. Who was this addressed to?  —Wknight94 (talk) 17:35, 27 July 2007 (UTC)
 * I was asking if you could add it to WP:RFC.  Wikidudeman  (talk) 11:30, 28 July 2007 (UTC)

Some thoughts
To me, there are three kinds of vandals... and this proposal will affect each differently...

So... it will not have any effect on Casual Vandals, will detur Agenda Vandals, but not Long Term Abusers. Still... One out of three ain't bad. Blueboar 18:45, 27 July 2007 (UTC)
 * 1) Casual Vandals - these are vandals who have no real agenda or reason for their vandalism, they just find it amusing to add nonsense to the project. Most of these are school aged kids (or adults acting like them) who think it is the hight of hilarity to add things like "Joe is gay" or "F**K" to an article. These vandals tend to focus on "vandal magnet" articls ... the days Featured Article, articles about subjects taught in school, and some pop-culture articles. For the most part these vandals are IP vandals... so the existing semi-protection is enough to stop them from editing a specific article. Their vandalism tends to be spur of the moment, and if they can not edit the article right now... they will either give up, or move to another article that they can vandalize.Thus, this proposal will not have any effect on casual vandals.
 * 2) Agenda Vandals - these are vandals who target specific articles or groups of articles. They are often users who have previously been blocked (or even banned) for agenda pushing and are editing under a sock account. They are not IP editors, as that identifies them too easily. They want to come across as a completely differnet person, who just happens to have the same agenda. Thus the socks. The whole point of their vandalism is to push their agenda further... which means that they have no interest in going off to vandalize other articles (and probably do not have any interest in editing other articles at all... they tend to focus only on their target area and don't edit elsewhere). With this kind of vandal, having the article semi-protected is a real head-ache... they tend to be editing in the heat of their passion, and don't want to wait until the semi-protect is lifted. However, if they know that all they have to do is wait four days, they are usually willing to do so. This proposal is aimed directly at this type of vandal. The addition of a minimum edit count will discourage them from continuing. They don't really want to edit other articles, and so they may just give up. Either that, or they will slide over into the next category....
 * 3) Long Term Abusers - These are a sub-class of Agenda Vandals... they are so obsessed with their agenda pushing that they will go through any hoop to continue. They are more than willing to pretend to edit other articles legitimately if that means they can continue to push their agenda at one article... so this proposal will not stop them. To be honest, the only way to stop them is to identify them and block them... and even this will be only temporary. They will return. This proposal may make it more difficult for them to edit. It may delay their inivitable return (not a bad thing), but it won't stop them.
 * Especially since nothing that doesn't completely compromise "anyone can edit" can preemptively stop the other two groups. -Amarkov moo! 05:52, 28 July 2007 (UTC)
 * Don't forget that there's no fine line between "Agenda vandals" and "Long term abusers". This proposal might stop a lot of long term vandals as well, or at least slow them down significantly.  Wikidudeman  (talk) 11:30, 28 July 2007 (UTC)
 * From my experience, this won't stop long term vandals... but then nothing will. Their very persistance is really what marks the difference between an Agenda Vandal and a Long Term Abuser.  You can block them when they show up (once you identify them), but ultimately they will come back with a new Sock or a different IP address.  But I do agree that if this proposal can slow them down, it is a good thing.  Blueboar 14:42, 28 July 2007 (UTC)
 * Well logically it would have to slow them down. If they are required to go through an extra process to edit semi-protected articles then that time it takes to do that would be time taken away from their vandalisms.  Wikidudeman  (talk) 16:30, 28 July 2007 (UTC)
 * I thought I'd inquire as to the news on this. It has support, there is a consensus, why isn't anything happening? How exactly does one go about making such changes?  Wikidudeman  (talk) 08:13, 31 July 2007 (UTC)
 * It has nothing like a consensus - your poll shows that, at least. Kamryn · Talk 23:27, 6 August 2007 (UTC)
 * If anything this section shows how pointless this proposal actually is. It'll have no effect on the vandals described in #1 and #3 and a minimal effect at best on #2 depending on how willing each individual is to make several tiny edits (and if they have an agenda, they're probably willing). The only relevant example I've seen is Brock Lesnar and I don't think it's productive to change editing rights on the entire wiki for the benefit of one article. Kamryn · Talk 23:31, 6 August 2007 (UTC)
 * It's not pointless and there was a consensus before hand, some of the people "voting" in the strawpoll had never even discussed it here and weren't part of the prior consensus. The fact is, It would slow down both type 2 and 3 vandals that you mentioned above. It would slow down longterm and agenda driven vandals, and would probably stop many of them. The downsides of this proposal are far outweighed by the potential benefits and in most cases the benefits over-write the downsides, I.E. increasing the openness to wikipedia by allowing more users to edit what are now fully protected articles. Moreover, the Brock Lesnar isn't the only article suffering from this problem. Just look here [] and browse through the numerous articles on basically indefinite protection due to often times a single long term vandal who creates sock-puppet accounts to vandalize it when it's semi-protected.   Wikidudeman  (talk) 23:40, 6 August 2007 (UTC)
 * Consensus can change. You asked for it to get a response from a wider audience; it did. There wasn't really a consensus beforehand, unless you can call support from a very, very small non-representative group of editors 'consensus'. Anyway, despite what consensus you had before, there isn't any now.
 * As for, I couldn't find any articles that were indefinitely fully protected for that reason - I found a lot of protected redirects, articles protected from recreation, and some current ongoing BLP matters amongst other things. Can you please provide specific examples? Unless you can provide specific evidence that this problem is widespread I see no reason for this proposal to pass.
 * Furthermore this proposal seems to me to violate the spirit of Wikipedia, i.e. that anyone can edit. I like the system as it is now - it's simple and it's worked so far (and yes WP:ILIKEIT is acceptable here). I really don't think the problem is big enough to introduce new editing restrictions for what is really a very minimal benefit. It doesn't take a lot of effort to revert, and really we're never going to be able to come up with an automated foolproof system for getting rid of vandals. I think editors who regularly engage in "vandal fighting" (lol) get a skewed idea of how bad the problem really is; out of our almost two million articles only a very, very small fraction are fully protected. However, this proposal will affect the entire wiki. I really don't want to add another exception to the "anyone can edit" concept of Wikipedia unless there is incredibly strong evidence that this is a BIG issue on the wiki. And there isn't. So I guess you could say I am against changing the entire wiki based on the actions of a few :o) I don't really see why a few trolls and vandals should negatively affect the editing experience of newbies. The current 4 days seems to work fine - are sleeper accounts really that big an issue?
 * Of course it works the other way - only a few articles are semi-protected, so you could say that newbies aren't really affected. You could also say that newbies don't need to create or move articles. Sure, this is to some extent true, but really is there any need to remove those rights? Some new users take months to make fifty edits. One of the biggest thrills about Wikipedia for newbies is that anyone can create a new page and help write something new :) Is it really worth detracting from that because of the actions of a few people who are out to spoil everyone's fun?
 * So yes, it works both ways - I won't be righteously indignant if this proposal goes through anyway, but I will be a little dissappointed. Kamryn · Talk 23:59, 6 August 2007 (UTC)

I'm also a little concerned by this. It seems to me we often forget the expectations of people who first show up on Wikipedia. While four days of waiting makes fairly clear sense, I suspect requiring 10 or more edits will make less so. In a way, even something like 500 edits says, ok, you have to be here for a long time before you get to have that feature, fine. Ten edits says "What, I have to add a period to ten sentences to get full privileges?" It took me some time to appreciate the focus that many of us put on edit counts; I suppose I'm partly concerned with presenting this aspect of Wikipedia culture so early. I think there's also some early glow with the idea that anyone can really edit Wikipedia that could be lost. Subjective, perhaps, but things to consider. Adding an edit count requirement would seem like a pretty big step. Mackan79 15:04, 9 August 2007 (UTC)

What this needs...
...first, it needs a concrete proposal rather than armwaving. I've picked an arbitrary number of edits for that. I am aware that it's as arbitrary as any other number, but it should be a reasonable compromise, so please don't replace it with another just-as-arbitrary number for arbitrary reasons, ok? Second, it needs some advertising; perhaps we should drop a line on the village pump et al. No, I don't intent to turn this into yet another pointless vote, but just so people are aware. Give that a week or two, then contact the devs.  &gt; R a d i a n t &lt;  09:28, 2 August 2007 (UTC)


 * 20 edits seems like a reasonable number.  Wikidudeman  (talk) 09:36, 2 August 2007 (UTC)
 * I agree with 20 edits. Sounds like a fair compromise between scaring away newbies and discouraging vandals (which was the reason that I originally brought this up). Although I think that many people wouldn't want to see this implemented without at least a straw poll (like a week long or something). -Royalguard11 (T·R!) 22:18, 3 August 2007 (UTC)
 * My recommendation: Create a page Wikipedia talk:Autoconfirmed Proposal/Straw Poll and take the straw poll directly to the village pump.--<b style="color:red;">Pre</b><b style="color:brown;">ston</b><b style="color:#6495ED;">H</b> 04:39, 4 August 2007 (UTC)
 * Done.  Wikidudeman  (talk) 06:07, 4 August 2007 (UTC)

Volunteer
I do not agree with adding edit based restrictions for the reasons I have outlined in the straw poll. Regards, Navou banter 01:11, 5 August 2007 (UTC)

(paste from) I'm not in favor of adding additional restrictions to editing the wiki. Time based restriction does not compel participation in order to participate fully, however, edit based restrictions will compel an editor to make the required X edits in order to use all the features on the wiki. This is an all volunteer project, lets keep it that way. I'll not support having an editor make the required X edits to create a new article, or use the move feature. While I do understand the frustration with vandalism, these things will happen, and these things are always easily corrected. Let us find another solution that does not involve an edit based autoconfirmed level. Navou banter 01:20, 5 August 2007 (UTC)
 * See []n for my responses to objections and remedies. After reading this you still have objections then just tell me what they are.  Wikidudeman  (talk) 03:30, 6 August 2007 (UTC)
 * None of your responses there are relevant to Navou's comments. And I agree with him. Kamryn · Talk 17:47, 9 August 2007 (UTC)
 * I still object on my reasons above. Navou banter 21:28, 9 August 2007 (UTC)
 * You don't like this proposal, but agree that vandalism is a problem. What is your idea then? We need to use every tool we can (and it seems the devs already have this implemented and were just waiting for us to ask). What would you suggest? -Royalguard11 (T·R!) 05:26, 10 August 2007 (UTC)
 * The current system for 'fighting' vandalism really does work just fine, you know. Kamryn · Talk 08:44, 10 August 2007 (UTC)
 * What about Brock Lesnar? It's been fully protected since March 14 (with a couple on and offs) because it's impossible to unprotect it without getting a very long term vandal hitting the page. The current method is kind-of working for now, but as wikipedia becomes more popular (and I guarantee it will) there will be more and more troublemakers. I know that it's gotten worse already. A year ago I could come to wikipedia and read a few articles and never found any vandalism. Now it's all the time that I come just to read an article, but some idiot has vandalized it and nobody's reverted it yet (and this is several hours sometimes). We used to be better (more like minutes). And we need to deal with it now before it gets out of hand. -Royalguard11 (T·R!) 19:34, 10 August 2007 (UTC)

(outdent)
 * Its very simple. From the foundation principles. "...The process should be virtually invisible for newcomers, so that they do not have to do anything to start contributing to the community."


 * Time based auto confirm is that. Edit based auto confirm goes against that principle.  We should not be working to make the wiki more restrictive.  Regards, Navou banter 20:34, 10 August 2007 (UTC)


 * (To Royalguard): Brock Lesnar is the only example I've seen after repeated requests. I am against changing the editing rights for everyone who signs up to the wiki to deal with one persistent vandal. Kamryn · Talk 21:51, 12 August 2007 (UTC)
 * How about Quebec (I'm sure he'll be back soon enough too). And I'm sure that we're missing some other ones too. We're trying to be preventative here. -Royalguard11 (T·R!) 19:00, 13 August 2007 (UTC)

Deleted edits?
Does anyone know if this counts deleted edits? Mr.Z-man  talk <i style="color:navy; font-family:cursive;">¢</i> 02:26, 5 August 2007 (UTC)
 * I just asked the devs in #mediawiki and Rob Church said the table this would use (in the db) does include all edits, including deleted ones (which is something I didn't know or anticipate before). -Royalguard11 (T·R!) 03:26, 6 August 2007 (UTC)
 * Hmm... that could potentially lead to a way to rack up edits quickly, even in mainspace. As there is a serious possibility that this proposal could be implemented, I won't share the full details per WP:BEANS. Mr.Z-man  talk <i style="color:navy; font-family:cursive;">¢</i> 03:51, 6 August 2007 (UTC)
 * Oops. Perhaps they can change that?  &gt; R a d i a n t &lt;  13:09, 6 August 2007 (UTC)
 * I would hope that this wouldn't be a problem. If a person is making crappy (G1-like) pages, they should get blocked before they rack up 20 edits of any kind. You would hope at least (it of corse depends on whether or not people give out warnings for this type of behaviour because admins don't have time to double check everything, but at least there's a Special:DeletedContributions now).
 * We could file a bug report about this too (although it would be a lot of work and would make deletion really slow since it would have to hit each users edit_count column at the time of deletion). I think that might be unworkable, but we could ask. -Royalguard11 (T·R!) 18:57, 6 August 2007 (UTC)

Most vandals who intend to vandalize semi-protected articles will be blocked before they reach the 20-30 edits (or whatever the final number is) and thus never will get the chance to get to the semi-protected articles. Dedicated and experienced vandals will either make redundant edits, adding / erasing content over and over, or they will make constructive edits to avoid being blocked. In such circumstances vandals are slowed down or wikipedia gets 30 constructive edits and a few vandalism edits in which case the vandlaism edits are instantly reverted and the vandal blocked to start the process over again. The idea that this will deter new users isn't true either. New users can't already edit semi-protected articles and in many cases the articles that would be semi-protected if this proposal were initiated are currently FULLY protected meaning no one but administrators can make edits to them. New users could easily request the edits be made on the talk page, in which case the edits would quickly be made, since the number of editors would have drastically increased due to it not being fully protected.  Wikidudeman  (talk) 21:47, 6 August 2007 (UTC)

Ok, just as a followup to this question, I tested this on a local wiki using the bleeding-edge trunk version of mediawiki. When an article is deleted, the edits are not subtracted from the edit_count table (I did this as a followup to physically look at the db myself after doing an edit and deleting the article). Even though there is this little bug, I think we've had enough discussion to ask the devs to implement this (through ). -Royalguard11 (T·R!) 00:46, 7 August 2007 (UTC)

So...
Has anyone spoken to a dev yet about this?  &gt; R a d i a n t &lt;  09:10, 9 August 2007 (UTC)
 * Do you believe that this has attained a consensus? Kamryn · Talk 09:47, 9 August 2007 (UTC)
 * I've filled a request at buzilla (10864). -Royalguard11 (T·R!) 19:23, 9 August 2007 (UTC)

Bad idea
Remember Delete unused username after 90 days?? Well, this is a repeat of that, it won't happen. And it won't work for Wikipedia anyway. Gotcha there. --Bolusball200 20:02, 10 August 2007 (UTC)
 * # 1, the devs already have this programed in and it may take nothing more than a setting in localsettings.php. DUU was rejected by the Brion (a dev) while I have confirmation from several devs that this is possible. So will you be a little less arrogant and tell us why instead of just a very childish "fail"? -Royalguard11 (T·R!) 01:59, 12 August 2007 (UTC)

I don't like it. Any vandal willing to wait four days will be willing to make 20 sandbox edits. &larr;BenB4 21:29, 12 August 2007 (UTC)


 * I agree, please this proposal seems in line with the spirit of WP:CREEP. As in, we are adding more hurdles.  I'm applying that discussion of policy loosely here. Navou banter 22:19, 12 August 2007 (UTC)
 * There is no possible measure we can implement that will prevent all autoconfirmed vandalism. This would, however, serve the purpose of making it harder for people to create huge farms of new accounts, and then go on a vandalism spree four days later. Some prevention is better than none. -Amarkov moo! 22:33, 12 August 2007 (UTC)
 * Yes, but at what cost. I've quoted a very important principle above. Navou banter 22:38, 12 August 2007 (UTC)
 * The spirit of WP:CREEP is that rules which heavily restrict good things for the sake of preventing a few bad things are bad. Preventing sleeper accounts from all but the most determined vandals is quite a bit of prevention, which is not outweighed by people with less than 20 contributions being able to edit semiprotected pages. -Amarkov moo! 04:30, 13 August 2007 (UTC)
 * I'm speaking of the foundation principle above. Navou banter 19:44, 13 August 2007 (UTC)
 * WP:CREEP is making admins fill out forums to speedy delete pages or creating Protection for Discussion instead of just using WP:RPP and have it done quickly. This has nothing to do with WP:CREEP. -Royalguard11 (T·R!) 18:54, 13 August 2007 (UTC)
 * No, I've applied the spirit of creep here, not the most literal meaning. Navou banter 19:44, 13 August 2007 (UTC)
 * Then you don't understand the spirit of CREEP. Will you even explain instead of hiding behind it? Saying this is instruction creep, say why. -Royalguard11 (T·R!) 23:45, 13 August 2007 (UTC)

(outdent) "...As in, we are adding more hurdles. I'm applying that discussion of policy loosely here." What I mean by this, is we are allowing more hurdles for the new users to jump over in order to use all features.. we are allowing those hurdles to creep in. I don't mean to appear to hide. Regards, Navou banter 23:57, 13 August 2007 (UTC)
 * Your misunderstanding instruction creep. It means hurdles that are demanded for the sake of (we haven't figured out why for some yet, like people who want Protection for Discussion because they don't trust that admins make good enough decisions on their own). It has nothing to do with what pace or weather or not it's transparent. -Royalguard11 (T·R!) 22:48, 14 August 2007 (UTC)
 * I understand it fine. Lets delete instruction creep, and make it security creep.  You are effectively causing a new editor to have to do something to use all features.  A edit based autoconfirm does that.  It will compel participation in order to create new and move.  This is not correct, and I don't think its inline with the foundation principle I quoted above in another header.  It is creepish.  Respectfully, Navou banter 01:20, 15 August 2007 (UTC)
 * So what? I'd like it more if we get people to actually come in and edit and do something more useful than make a page about themselves, it gets speedy deleted, and they leave. Do you want us to stop speedy deletions too now because they infringe on WP:BITE? -Royalguard11 (T·R!) 00:15, 16 August 2007 (UTC)
 * I'm not clear your meaning. However I want to be.  Could you clarify how the speedy deletions and WP:BITE apply here?  Personally I don't see CAT:CSD as a BITE issue, if done properly.  Help me understand what you mean.

I'd love it if editors would provide useful content. I don't mind the test edit that has to get speedy deleted, if that is in fact a gateway to useful contributions. It is a wiki. Navou banter 00:35, 16 August 2007 (UTC)
 * Well, you said It will compel participation in order to create new and move. I don't see anything wrong with that. Sometimes you have to earn the privileges given to you. Before annons could create pages, then it was changes so only users can (due to blp et all). -Royalguard11 (T·R!) 00:44, 16 August 2007 (UTC)


 * This has nothing to do with DUU90, I'm not sure where that came from.  &gt; R a d i a n t &lt;  11:32, 13 August 2007 (UTC)

Consensus
It appears that consensus has not been reached by a wide audience and is unlikely to do so. Mark as rejected? Navou banter 19:24, 12 August 2007 (UTC)
 * It's already been sent to the devs. And I don't think it doesn't have consensus. Several user believe it has enough right now, while only a few (including yourself) disagree. Proposals aren't rejected because 2 people are in opposition. -Royalguard11 (T·R!) 19:39, 12 August 2007 (UTC)
 * I've posted a few places so that this can get wider audience. Thanks, Navou banter 19:52, 12 August 2007 (UTC)


 * It looks like more than two folks are in nonsupport. Also, I'm not counting numbers, just reading the discussion.  I did take a glance a the straw poll, it does however, have less weight in a consensus building discussion.  However, I'm not going to be the best to determine consensus here, I am involved in the discussion, a disclosure I should have made in my above comment.  Regards, Navou banter 20:01, 12 August 2007 (UTC)


 * I did also make a comment about the foundation principles in the above volunteer heading. Navou banter 20:03, 12 August 2007 (UTC)


 * No, it really doesn't have consensus. I hope the devs look here and see that. Proposals aren't passed because three people support it. Kamryn · Talk 21:52, 12 August 2007 (UTC)


 * Nor is "three people support it" necessarily a reason the proposal is bad; it could indicate that people see nothing new to add to the discussion, or don't feel strongly either way, or just haven't seen the proposal, yet. Personally, I'm pleased that it would make it increase the amount of investment and effort needed to build a sockfarm of sleeper accounts, provided that this doesn't become any sort of slippery slope toward more and more restrictions on new editors. Or something along those lines, anyway. – Luna Santin  (talk) 02:47, 13 August 2007 (UTC)


 * I'm tempted to follow up this section with a similar proof by assertion that states "It appears that consensus has been reached by a wide audience. Mark as approved? "  Might I point out that consensus is not unanimity, and that assertions like "this page does / does not have consensus" really don't help discussion in any way?  &gt; R a d i a n t &lt;  11:34, 13 August 2007 (UTC)
 * How about the points I brought up? Navou banter 19:42, 13 August 2007 (UTC)
 * I believe we should bring this issue to the attention of the developers and let them decide whether or not to implement. This is the only viable approach, simply because the community is unable to force the developers to implement, or not implement, any particular feature. I am sure the devs can reasonably estimate whether this is a good idea and whether there is sufficient consensus here.  &gt; R a d i a n t &lt;  10:48, 16 August 2007 (UTC)
 * I believe, the community will discuss this, and if the developers want to give some technical (or otherwise) input, thats fine. The decision to code it is up to the developers.  The decision on implementation, rests with the community.  I would like see any input they have.  Navou banter 12:21, 16 August 2007 (UTC)
 * If you look at the bug report, you'll notice that Rob Church changed the name to Set $wgAutoconfirmCount to 20 for English Wikipedia, which must mean that they've already coded it and just need to turn the option on or not. -Royalguard11 (T·R!) 20:53, 16 August 2007 (UTC)
 * The gain consensus to turn it on. Regards, Navou banter 21:28, 16 August 2007 (UTC)

I like it
It's short and simple. The point is it won't knock off the determined vandals, but it may knock off the lazy ones, who are just hitting Wikipedia because it's quick and easy. This may give a bit of a breather patrolling for vandalism, giving them more time to deal with the determined vandals. The negative responses here appear more reactionary to something new, which is common on Wikipedia, than directed to reasons that such a small increase in the autoconfirm wouldn't work. I ask people to reconsider this as something designed just to give a little breather, and discourage some of the more immediate-gratification-needed vandals, while not doing anything deep to discourage new editors who are here to edit. KP Botany 02:44, 13 August 2007 (UTC)


 * Creating sleeper accounts and waiting for them to mature doesn't seem lazy. &larr;BenB4 04:07, 13 August 2007 (UTC)
 * It's a lot lazier than having to make 20 contributions on every single one of them. -Amarkov moo! 04:26, 13 August 2007 (UTC)
 * Yes, it is, and the proposal is not designed to, nor would it, deter the determined vandal, as I said. KP Botany 04:27, 13 August 2007 (UTC)


 * Really? People are still going to vandalise. Just perhaps not semi-protected pages. Kamryn · Talk 09:40, 13 August 2007 (UTC)
 * Really what? Yes, people will still vandalize, the only thing to prevent that is to make this not Wiki.  KP Botany 16:34, 13 August 2007 (UTC)

This isn't going to stop your random page hitting "PENIS!!!!" vandal. It's meant for pages that have been (and some are still) under full protection because semiprotection just doesn't cut it anymore because they can get around it by registering a bunch of accounts and waiting 4 days. This would make it harder for them. -Royalguard11 (T·R!) 18:50, 13 August 2007 (UTC)
 * Right. At WP:PP I've seen many pages protected from banned users that have an interest in trying to edit it again, having to make 20 other edits for each sock without getting blocked (they must be constructive) is a pain, not to mention the good edits only help. Also, by not saying "get 20 edits" in the UI, it acts as WP:BEANS, and should deter some basic vandals too. The determined, WP-familiar "penis" vandal and Bonaparte of course will just do things to get past the edit limits.  Voice -of- All  02:41, 17 August 2007 (UTC)
 * Sounds a good idea to me. Tim Vickers 01:03, 18 August 2007 (UTC)

Good enough, maybe this is better
The existing proposal is good but far from perfect. I think we can do better.

The problem is repeat sockpuppets who create sleeper accounts.

The best solution would be to neuter the sleeper accounts without impacting other users.

If we can identify sleeper accounts and "reset the clock" on them, that should annoy the puppet-masters enough to discourage their use.

Proposal: For now, adopt the proposed change. It's not very hard.

In the future, find a way to quickly identify such sockpuppet accounts and either block them outright, reset their clock to "0 days old," or mark them as permanently non-confirmed until the owner emails an administrator and asks for confirmation.

A quick way to identify sock puppet accounts might be to search all accounts that share an IP address or range with blocked vandals, discard those from widely-shared IP addresses, discard those with a sufficiently long edit history, say, a total of 3+ edits in mainspace with at least 1 edit on each of 3 different days, then tag what's left as suspected sockpuppets. Of course this method, as with any other method, will generate false positives and false negatives.

davidwr/ (talk)/(contribs)/(e-mail)  00:16, 26 August 2007 (UTC)
 * The only people who have that power are the checkusers. -Royalguard11 (T·R!) 01:07, 26 August 2007 (UTC)
 * What I had in mind can be semi-automated or even completely automated. Complete automation is undesirable due to false positives.  davidwr/  (talk)/(contribs)/(e-mail)  02:45, 26 August 2007 (UTC)

FYI
This has been rejected as WONTFIX. I've already spent quite a bit on this, so if anyone else wants to take this somewhere be my guest, but I'm officially done with this. -Royalguard11 (T·R!) 19:53, 31 August 2007 (UTC)
 * Very well. For those who are interested, this can technically be appealed to Brion Vibber, but he tends to have more important things to do.  &gt; R a d i a n t &lt;  13:13, 4 September 2007 (UTC)
 * I tried appealing to Brion, but no luck. Kaldari (talk) 16:19, 16 June 2008 (UTC)

New poll
The devs have decided to ignore the will of the community and have reclosed bug 14191 despite protests. They have requested that we start the whole process over again in order to establish "unequivocal consensus". To that end, here is the new poll. Please express your opinion "unequivocally" there. Thanks. Kaldari (talk) 16:10, 16 June 2008 (UTC)