Wikipedia talk:Semi-protection policy/Archive 1

Archive 1

Archive 2

Archive 3

→

Archive 5

Voting

Putting this up for a vote. Vote either Support or Reject, and state your reasons why.

• Support - we need some minimum level of protection on our pages evolution and abortion. --Cyde 09:30, 3 December 2005 (UTC)
• Support - Wikipedia slowly evolves to the next stage, with a core of "done" articles and lots of work on the fringes. It is a good idea to have a means of protecting that core. I trust that we Wikipedians will find a way to handle semi-protection in a responsible and working manner. -- Mkill 18:06, 3 December 2005 (UTC)
• Reject. You tried to vote on a proposal that is still being worked on. -Splash^talk 18:25, 3 December 2005 (UTC)
  • There you go, much better. Doesn't Democracy feel nice?  :-P --Cyde Weys [u] [t] [c] 18:29, 3 December 2005 (UTC)
    • No. Wikipedia is not a democracy. Voting here is rubbish. -Splash^talk 18:29, 3 December 2005 (UTC)
      • You're right, it's a mobocracy. --Cyde Weys [u] [t] [c] 18:31, 3 December 2005 (UTC)
        • Whatever. -Splash^talk 18:34, 3 December 2005 (UTC)

Comments

• I strongly recomment a minimum of 25. I have watched vandals and they can easily rack up the ten needed to hit a major page and damage more user's view of Wiki.Voice of All^T|@|ESP 05:42, 29 November 2005 (UTC)
• I believe that we have to assure users that this is just supposed to be used in our most-vandalized pages, like George W. Bush, John Kerry, and Wikipedia. Titoxd^(?!?) 05:56, 29 November 2005 (UTC)
  • Agreed. We protect the front page, but under this, a user with a minimum of ___ edits can edit the templates, or change the front page, virtually eliminating the vandalism. Enough people have the front page on their watchlists (or see it enough) that reverting it would be easy. -Mysekurity 05:57, 29 November 2005 (UTC)
    • I support semi protection for certai articles like George W Bush to keep vandalism to a minimum. Another option, though I don't think it can be done, would be to have some sort of time delay as in have a administrator check (anonymous) edits before they become visible. Is that possible with Wikipedia's software?--Kalsermar 16:59, 29 November 2005 (UTC)
• Minimum of 25 sounds appropriate to me and I agree with Titoxd.--Alhutch 19:27, 29 November 2005 (UTC)
• 25-50 edits before editing a semi-protected page would be a good idea. I support the semi-protection policy![[User:Kevin Hanse|Kevin Hanse ([[User talk:Kevin Hanse|talk]])]] 21:26, 30 November 2005 (UTC)
• After reading most of the discussion on this page, and myself reverting GWB more times in the last several months than I care to count, I support this policy. Personally I prefer about 50 edits for edit capability, but 25 will do. Implementing this will make the GWB article usable again; right now, because of the extraordinary number of vandalisms and reverts, it is impossible to follow the evolution of the article in the history. Antandrus (talk) 02:35, 2 December 2005 (UTC)
• I think this is a very good idea, and that the Main page should be subject to this sort of thing too, possibly with a very high edit count needed. It will stop 90% of vandalism to this page, and coming to Wikipedia and seeing the first words on the page as an insult is not a good idea. However, there is a way for dedicated vandals to get round this problem - vandalbots that can clock up a lot of edits, and then quickly go on to vandalise this page. It's not that hard to create a vandalbot. Nevertheless, Support. --Mark J 19:40, 2 December 2005 (UTC)
  • Just to let you know, if we enforce a minimum amount of days the account has existed, as in 5 days or a month, it should warrant off most vandalbots. --kizzle 20:22, 2 December 2005 (UTC)

Does this help WP:1.0 too?

Wikipedia:Pushing to 1.0 came out with a Wikipedia:Requests for publication feature which is quite similar to this. By freezing content in a special section of the site, they are planning on producing a stable version that can be cited and quoted. While the goals of the two proposals are different, perhaps cross-posting there would benefit this discussion? Titoxd^(?!?) 05:58, 29 November 2005 (UTC)

Questions

• Which is the best level? 10 or 25? 50 or 100? More? Should there be more levels, more incriments of protection? Is it good the way it is? Is this feasable? -Mysekurity 06:14, 29 November 2005 (UTC)

• The following may be viewed more as concerns rather than questions per se.

-Anon IP edits - some of the anon IPs are dynamic, some are shared by many users. e.g. AOL, University IPs. On a given day, some of the more high-profile IPs would contribute 25 edits in a day. How would we deal with it?

-Would the above give a new user an incentive to be an anon IP user rather than a registered user? imo, registeed users are more desirable to the project, so this needs some addressing.

-Will availability of different levels of protection start off wars within editors of each high-profile page, each editor asking for a higher or lower level of protection, thus diverting our efforts from editing, our primary task?? --Gurubrahma 07:33, 29 November 2005 (UTC)

Voice of All(MTG) has proposed we limit the editing of these pages to registered users (see the GWB talk page). I, for one, am having mixed feelings over this; I think it might solve your issue, but is it truly what we're aiming for here? Again, the purpose of this policy is not to eliminate vandalism completely (and keep this as non-anti-wiki-like as possible), merely cut it down greatly. I don't mind reverting the page every once in a while, just not when it happens so often. We could always experiment with different types of protection, which is what semi-protection is all about.

Anon IPs are where you, me, and nearly everyone else got their start, I believe. While it is more desireable to have registered users, sometimes usernames can be mass-created, making it very difficult to stem their vandalism. For more, see Redwolf24's [on the matter].

Again, you ask a good question. I do not believe this is the case, as the only threshold is edit count, an arbitrary number. There will be no requests for permission. We do not have (m)any problems with admins fighting over protected page content, and I doubt this will be much diferent. The point of this is to stop the vandalism, which has consumed our editing resources and, in effect, help editing. Just like the main page, this is a high profile page, and we want to keep it as clean as possible. I don't know about disallowing anons to edit, but I do think if we left the threshold at 25, vandalism would be greatly decreased. Thanks for your questions. -Mysekurity 12:20, 29 November 2005 (UTC)

• If an edit threshold were to be imposed to registered users of a proposed "semi-protected" page, it should be a relateively low number, 100 or below. Personally though, I would prefer we stay in line with how we limit page moves to newly registered users (and prevent page moves from unregistered users). i.e. After an account reaches a certain age, then it can edit a semi-protected page. This would drastically cut down on vandalism, allow non-administrators to freely edit high traffic pages after minimal experience with Wikipedia, and free up administrators time to accomplish other tasks: a win-win-win situation. Hall Monitor 18:27, 29 November 2005 (UTC)

What do you propose the time limit be? Is this only for registered accounts (I agree, by the way). -Mysekurity 22:07, 29 November 2005 (UTC)

Yes, this would apply only to those with newly registered accounts. I propose using the same time limit or aging formula that we use for allowing page moves, which I believe amounts to a few days. Does anyone know the exact formula used for allowing new users to move pages? I have been unable to locate a precise answer on this, but it seems to be approximately 5 days or so. Hall Monitor 23:54, 29 November 2005 (UTC)

Data accurate?

The proposal notes that GWB is vandalised for "hours a day". This probably is a bit of exaggeration, although my data suggests that at worst it could be an hour a day. Pcb21 Pete 08:46, 29 November 2005 (UTC)

I hope to come back with some wider comments on the proposal later. Pcb21 Pete 08:46, 29 November 2005 (UTC)

Okay, I just went through the George W Bush archives of 11/28/2005 and here is what I found:

• 66 instances of vandalism
• a total time of almost exactly 3 hours in which the article was in a vandalized state.
• The subject of the article is called a "Bastard", "co%ksucker", "Dumbass" and "Hitler", to name a few, all within the article text. We also have this fine edit summary by an anon IP: GEORGE BUSH LOVES TO ASSFUCK PORN DEALER WALES AT *(deleted Jimbo Wales address)* CALL HIM AT (delete#) FOR A RIMJOB

I consider the 28th to be a typical day in the life of this article. In fact, there have been much worse as I recollect numerous IP's blocked during a true attack day. I will be willing to concede that much of the reversions are quick, in far less than a minute, but in one case, the page was in a vandalized state for 23 minutes, and in another for 15 minutes.--MONGO 09:31, 29 November 2005 (UTC)

Similarly, among the last 50 edits to George W. Bush: 23 were vandalism. 18 were reversions. The difference is because of vandals in a row. That's, together, 41 edit out of 50. The reverts have been very quick, but it's a problem regardless. The article gets vandalized several times per hour. I took a random hour of today and counted 5 vandalisms in it. Given 1 minute to revert each, that would still be 5 minutes in an hour that the article is vandalized, which amounts, by the way, to 8% of total time.

I appreciate the semi-protection solution if only because of this. Most vandalism is easily dealt with, but Wikipedia is one of the world's most visited sites, and George Bush must be one of the most read articles. Just imagine a first-time visitor whose first impression of Wikipedia is seeing an article on George Bush calling him a "cocksucking bastard" - which wouldn't even be among the worst the article has endured. Solver 13:51, 29 November 2005 (UTC)

I LIKE

I think it is unfair that only admins can edit the Main Page, this is a good alternative. Gerard Foley 14:57, 29 November 2005 (UTC)

YES! This is where a higher threshold would come in. NO vandal, except for someone who really didn't want to be here, would take the time to garner 500 or 1000 edits simply to vandalize the main page. -Mysekurity 22:07, 29 November 2005 (UTC)

THis solves a lot of problem, IF and only if we as users can easily getadmins to mark pages as semiprotected. Many times vandalism comes in the form of people shipped from other forums to edit wikipedia to push a PoV. Dominick ^(TALK) 20:07, 30 November 2005 (UTC)

Yes, I've seen it done a number of times, and I love the idea of opening the main page to those other than admins. The great thing is that this isn't as harsh as full protection, which is unwiki, but isn't as bad as making us look like fools with these vandalism edits, and can be temporary. -Mysekurity 21:00, 30 November 2005 (UTC)

Grilling

Let me give this a grilling, in the traditions of...well...grillings. It's important to beat new concepts up a bit to be sure they are robust:

• Protection creep (1st kind) — Articles will usually only move up the scale. If it's only to be used when the vandalism has become a serious problem, there is unlikely to be a de-escalation short of e.g. a president leaving office. If the vandalism were serious but brief, then full protection for a few hours, as already available, would still be fine.

  Observe this already in practise, where George W. Bush has long been protected against moves only.

• Easing prophylaxis — Wiki is not prophylactic, it does not usually defend itself against imagined threats. This is to keep things as nearly free-as-in-speech as we can.

  An article is greenlighted on Slashdot. At present, we'd just revert the trolls. But why not use, say, level 3 protection, just in case things get serious? At present, we'd maintain the Wiki-way and let people edit it, and slap the vandals.

• Protection creep (2nd kind) — Related to the above, we will, with absolute certainty, end up with more articles in a state of protection. Wikipedia is nowhere near finished, despite the endless push to WP1.0, and low-level protection of more articles will inhibit the process. Articles contributing to this point will also be subject to protection creep of the first kind.
• Stratification of editors — is bad.
• Stratification of articles — Articles are not equal, and they are not created equal. See the various stubs compared to the Featured Articles. At present we have these 3 levels of articles: stubs, normal and featured (and putatively, good). Clearly, there will be a tendency to seek increased protection for better articles. This may not be a bad thing, and I am merely mentioning the point. Asymptotically, however, as Wikipedia approaches completion, everything will be fully protected.
• Vandals aren't the biggest problem — the lack of reliable, verifiable sources and the POV status of many articles renders the Project basically unciteable. If those problems were hypothetically remedied, then the vandalism would be a scratch rather than a scar since people could still cite the article nearly all of the time. Increasing the protection level decreases the number of people who can work on making the article so good it rises above the level of vandalism.

-Splash^talk 15:13, 29 November 2005 (UTC)

If we only allow changes of a page's semi-protection status to run through either arbcom or an admin-only vote, then we don't have to worry too much about every page getting protected left and right. Semi-protecting a page should become a huge deal with only the top 5-10 pages out of the entire article space being protected, and would have gotten there through some kind of due process. Regardless, Vandalism is not just a scratch on pages like GWB, it has crippled the page so that it makes it incredibly hard to view the progress of the page when the history consists of 90% vandalism/reverts. --kizzle 19:45, 29 November 2005 (UTC)

I agree with Kizzle. As one of the people who has had to keep reverting on the GWB article lately, it does get annoying. It would be silly to not do something about this when we have the chance.--Alhutch 20:16, 29 November 2005 (UTC)

Agree with the above - vandalism fighting takes time away from other improvements. Let's at least experiment and see if this proposal makes the positive difference that many of us anticipate. BD2412 T 20:27, 29 November 2005 (UTC)

I think it's a terrific idea myself. Those of us who patrol the Page protection page will also appreciate it because we're kind of tired of rejecting requests for the Bush article. John Kerry also suffers a similar fate. What would be nice with semi-protection is that we could slap it on Community Portal too...and other visible pages that people should be allowed to edit but on the other hand, often fall victim to vandals. I like it...let's give it a whirl, I say. --Woohookitty^{(cat scratches)} 21:42, 29 November 2005 (UTC)

There: protection creep of the second kind, before we even the ability. You must be ultr-careful about applying any level of protection to articles. It is not something that the community at large will support if it is applied in the rather whimsical fashion above. We do not "slap it" on anything. Why? Because of protection creep of the first kind, and the permanence of it. -Splash^talk 22:17, 29 November 2005 (UTC)

I wasn't being whimsical. The problem if we don't have some level of protection, then we'll burn out more people. The Bush page literally gets 30+ vandal hits a day. Talk:Main page is up to 10+ a day. I don't think you quite understood what I was saying. I don't see the harm in limiting access to a small amount of pages. We're not talking 100s here. It would probably be the same amount of pages that are permanently protected...20-30 at most. It wouldn't stop discussion on talk pages. It would just give a threshold for editing certain pages. And btw, I work on the Protection request page every day. I'm very careful about protection. Like I said, I'm not talking every article we have or even .1%. It'd be the more visible ones that get vandalized alot. --Woohookitty^{(cat scratches)} 22:43, 29 November 2005 (UTC)

Agreed with Kizzle, but I also understand Splash's concern, and I'm glad this was brought up at some point by a friendly editor, rather than a hostile one. Not all articles would be protected (I think I like the idea of only having a few pages protected, and have to pass a vote first), and this would aid us in our push to 1.0, by freeing up resources that used to be used on vandal fighting (admittedly very tiring), to be used to edit articles and such. Long-term semi-protection would be decided on at a page like RFPP. Short-term semi-protection can be used at an admin's descresion, much like full protection is now. -Mysekurity 22:07, 29 November 2005 (UTC)

I think changing semi-protection should even be harder than page protections, it should be decided by a certain amount of people, arbcom would be ideal, a certain majority of admins if not arbcom. As long as we stick to only enabling semi-protection changes through a lengthy due process, we avoid Splash's concern of protection creep. --kizzle 22:40, 29 November 2005 (UTC)

I have major objections to this. The arbcom is already bogged down in alot of difficult cases. I don't see where it helps to give them even more to do. I'm not sure why semi-protection couldn't use the same page that protection does. We are very careful with what we give protection to and what we don't give protection to. --Woohookitty^{(cat scratches)} 22:54, 29 November 2005 (UTC)

It doesn't have to be arbcom, maybe just a certain minimum amount of admins voting on it, as semi-protecting a page is too powerful in my mind to be left up to the discretion of a single admin. --kizzle 08:24, 30 November 2005 (UTC)

So I assume you feel that way about protection too? That also just takes one admin. --Woohookitty^{(cat scratches)} 11:17, 30 November 2005 (UTC)

The difference, I see is that with full protection, we have almost no articles or pages that have been protected for a long period. I think that the advantage of semi-protection is it makes longer term (semi-)protection possible, but not neccissary. Short protection can be decided on by a single admin, long-term has to have a consensus.

Is this even possible?

Is this even technically feasible? And if so, shouldn't we not be trying to create a bunch of levels of editors? --LV ^{(Dark Mark)} 22:28, 29 November 2005 (UTC)

I'm no developer, but I think it is. I agree that we should be trying to unify editors as much as possible, but some content should at least be given this (maybe 10 pages at most). -Mysekurity 22:39, 29 November 2005 (UTC)

I'm going to go out on a limb here and say I actually favor protection-creep. I have been watching the Napoleon page and, because of his rather high historical visibility, there are many defacements of the page every day. Despite people's misgivings about keeping wikipedia editors all on the same level, there is a hierarchy here in any case; people who reach a certain number of edits can gain the respect of the community and can become admins and have a greater impact on the day-to-day look of any given page, among other things, which is as it should be, I think. Requiring people to make real edits before they edit sensitive pages is not an undue burden on them. I also think that pages should not be protected because of their "importance," but protected when they are under constant attack by multiple IPs, for whatever reason. I'd be in favor of Level 4 protection for a fair number of pages. IronDuke 22:49, 29 November 2005 (UTC)

I'd agree with IronDuke. I mean the thing is, this would be limited. We're not talking about having 10-15 levels of editors. I don't personally want that. I don't think anyone does. But IronDuke makes a good point. We already have admins, bereaucrats...you could argue that people on the arbcom have more power than other users. We'd be creating one, possibly 2 more levels of users and it would be for just semi-protection and nothing else. And we're not talking 1,000 edits or 100 or even 50. It would just be a way from stopping anons and brand new users from vandalizing some of these pages. IronDuke, the thing with importance is that we already pages protected due to visibility reasons. We'd be talking about similar pages for semi-protection. We'd be talking Community Portal, maybe some of the higher profile templates...basically stuff that usually isn't changed without a major consensus anyway, so there isn't much harm in semi-protecting them. --Woohookitty^{(cat scratches)} 01:25, 30 November 2005 (UTC)

Agreed. Protection creep doesn't have to happen, but it might happen. I personally would hate to see Wikipedia become unwiki, but, like others have said, "Which is worse?" I don't think a new user will care or notice that they can't edit the one page. (Somebody) makes a good point in discussing the Main Page. We will not drive off a new user perminately because they can't edit a few pages (again, we're talking about a small amount here), but we might, and probably will, if they see Image:Human Feces.jpg on pages. Even the "____on wheels!" stuff would probably not bother a new user too much, even though it bothers us. We simply must have a better way of dealing with this. -Mysekurity 21:00, 30 November 2005 (UTC)

Purpose

I'm kind of having difficulty deciding what the purpose of this is. I think this should start small and stay small, but it has some potential for as the wiki grows in scalability. For now, I think we should maybe leave it to high profile pages like GWB and others (only a few), and then maybe we'll increase the number in the future. Any thoughts? -Mysekurity 22:39, 29 November 2005 (UTC)

• Don't know if there's a way to work scalability into the system, but I'm thinking that we can limit protection to pages that are vandalized an average of x times per day over y days (say, 10 times per day for a week). BD2412 T 22:54, 29 November 2005 (UTC)

One problem we're going to run into is that even the protection page doesn't have hardened "rules" like that. Not sure how well that's going to sit with some. Everything we do on WP:RfP is through very informal guidelines. --Woohookitty^{(cat scratches)} 23:01, 29 November 2005 (UTC)

I would urge caution and light-footedness over this. Protection == unwiki == bad. Protection with new algorithms controlling it...I see what BD2412 is driving at, but we do want to retain the element of admin discretion. The admins are a fairly dim bunch at times, but collectively they manage more or less to get it right. -Splash^talk 23:53, 29 November 2005 (UTC)

Yes, but we need more weapons against vandalism and this would be another weapon. And as I said, take a look at the RfP page. We don't give protection liberally. We reject as many (if not more) requests than we accept. And this would be similar. The purpose of semi-protection is to help stop vandalism. It wouldn't be used for edit wars as far as I can see. Just vandalism. --Woohookitty^{(cat scratches)} 00:34, 30 November 2005 (UTC)

Of course we need to urge caution over this. That's why we will have a lengthy process to change semi-protection status that requires multiple admins, if not arbcom. The goal is only to semi-protect about 5-10 pages on all of Wikipedia, if even that. But I'm sorry, protection does not necessarily equal unwiki, especially given the article is completely uneditable, highly vandalized, and a pain to deal with. We should be hashing out what process should justify semi-protection, currently, I think an admin vote of at least 15 people with a 70% pass rate should set semi-protection, in addition to only allowing semi-protection to go up one level at a time, in other words a separate vote for each level. --kizzle 04:13, 30 November 2005 (UTC)

Question for you kizzle. Why would the semi-protection process have to be so much more difficult than protection? I ask this because protection only takes one admin. Semi-protection would actually be much less restrictive than protection. Protection now allows absolutely 0 edits...even by administrators, except in special circumstances (I've seen one edited by an admin all of once). Semi-protection would allow edits by everyone over a certain threshold, which means that basically 90% of the registered users (if not more) would be able to edit the page. So shouldn't it be less restrictive or be the same level (one admin) as protection? I don't quite understand that. --Woohookitty^{(cat scratches)} 16:24, 30 November 2005 (UTC)

Short-term protection would not be so complex. Long-term would require a discussion and admin/experienced user vote. Short-term would work like regular protection, and be undone in about a week, while long term would stay until a big event, or high visibility passes. This is the major difference, I see. The point of this is to be more wiki-like with protected pages than we are currently. -Mysekurity 21:00, 30 November 2005 (UTC)

Crosspost regarding GWB vandalism

(Crossposted from talk page ^[1]) Hi there. You've recently begun something of a determined course of action to have this article protected. Admins have repeatedly unprotected it quite quickly; I was wondering if you were aware of this. It's fundamentally wrong to have our highest profile articles protected, in the same way we never protect the main page article. Most vandalism on GWB is removed in well under a minute, and those in #wikipedia-en-vandalism practically scramble to be the first to revert. Perhaps you should take a look at Wikipedia:Semi protection instead? -Splash^talk 19:47, 29 November 2005 (UTC)

Thank you for contacting me. Yes, I am aware of the discussion at Wikipedia:Semi protection, and have commented there earlier this morning. [2] There have been several recent cases where this page, the most highly vandalised and watched page on Wikipedia, remained vandalised for over one hour and fourty minutes, as originally pointed out by Rhobite in October. [3] As is noted on the Semi-protection policy talk page, the George W. Bush article is now being vandalised to such an extreme that, on average, the article is effectively in a vandalised state 8% of any given time. [4] [5] So, while vandalism may be reverted within 1–2 minutes, it is rather moot when the article is being attacked every other minute. The decision to temporarily protect this page was made [6] while the article was in the process of being simultaneously attacked from multiple IP addresses, meeting my interpretation of Wikipedia:Protection_policy#Uses first line item. The statement that "Admins have repeatedly unprotected it quite quickly" is open for debate after a review of the protection log, but please do note that I specifically requested that someone "please remove and protect from moves only when appropriate" after reluctantly placing this under {{vprotect}}. I believe we both agree that this article needs to be open to well-intentioned editors, but on that same token we are equally beholden to our readers to present them with something more professional than "George Bush is a son of a bitch", or much, much worse. If you have any suggestions on how to better deal with distributed vandalism attacks such as the one made earlier today, please let me know. Best regards, Hall Monitor 22:08, 29 November 2005 (UTC)

---

I've crossposted this here because it raises valid concerns on what to do in the interim while a semi-protection policy is being hashed out and discussed. When is it appropriate to use the current vandalism protection feature with regards to the George W. Bush article? Hall Monitor 00:02, 30 November 2005 (UTC)

This is partially in response to Splash's post. You know what? I'd much rather have a high profile page protected than to make people look at multiple pictures of penises, which is something we got hit with last night. And I don't get this attitude about the vandalism room in IRC. Because people are quick to revert the vandalism, this makes not stopping the vandalism acceptable? That's more fundamentally wrong to me than to protect high profile pages. Anyway Hall Monitor, I think the only reason why we can't really use protection against the Bush page is because in order to stop vandalism using the current process, we'd have to have it protected 24/7 and that's not acceptable to Splash or me or anyone else I've run into. I mean it literally gets hit multiple times an hour. It makes it kind of pointless to permanently protect it, which is why I think semi-protection is a good idea. --Woohookitty^{(cat scratches)} 00:43, 30 November 2005 (UTC)

My particular concern with permanent protection of high-profile pages is embodied in Golbez's comment on Talk:George W. Bush where he basically says that if somone wants a change, they can and I quote, "submit" it on the talk page. The idea that admins should have a content approval role is extremely distasteful and gives them a role they were never considered for in RfA. -Splash^talk 00:38, 30 November 2005 (UTC)

Yes but admins generally cannot touch protected pages except to revert vandalism that occured just before protection and also some other very limited circumstances. And if we agreed to a semi-protection policy, we can pretty easily forbid admins and others from adding content that's proposed by people with less than whatever the threshold is. --Woohookitty^{(cat scratches)} 00:47, 30 November 2005 (UTC)

Previous

I know this has been attempted before (I'm guessing a number of times), and has failed every time, but I'm hoping this could be pushed to the front well enough that it will finally become policy. We could have dealt with this back in july, but we didn't; if we had, maybe we wouldn't have had to have this discussion now. All we can do to prevent this discussion from happening again and again is get it passed now. Looking at the past pages, it doesn't seem like there was a real reason it wasn't implimented; only that there wasn't enough attention brought to the issue.

I'm not feeling well, and I thank those of you who have been editing the proposal for my grammatical and spelling errors (they're showing in my off-wiki work too...). Thanks, and happy editing. Mysekurity 05:31, 30 November 2005 (UTC)

See also:

• {{VotesforSemiprotect}}
• Village Pump proposal
• Wikipedia:WikiProject Semiprotect

P.S. WP:JEFTA!!, I encourage you to make changes, have discussions, start fights, and dress like a superhero in order to get this proposal to pass! -Mysekurity 05:31, 30 November 2005 (UTC)

Should be done without a vote. Votes are evil. Semi-protects should be determined by admins just like regular protected pages. Different levels should exist (as proposed), instead of the un-wiki idea of complete protection, a limit on how much newbies can edit will increase stability on pages such as George W. Bush.

This should not be seen as promoting some sort of elitism, in fact its the opposite. Pages that otherwise would be protected (admin only edits), would with this stay editable even though with limitations.

--Cool Cat^Talk|@ 09:41, 30 November 2005 (UTC)

Instead of full protection? You may have a point...Full could be replaced with 500 or 1000, which is basically admin-level. It would have a tag like full protection, but not just admins could edit. -Mysekurity 21:00, 30 November 2005 (UTC)

Requests for publication

I'm wondering if there is any possible synthesis between this and Wikipedia:Requests for publication? -- Jmabel | Talk 06:32, 30 November 2005 (UTC)

Yeah, it would definately help to push this along. I'm invisioning FAs or {{GA}}s with proposed changes at the bottom, which can be rejected or accepted, (far off in the future, that is, when we're much more complete, but may be useful for GWB). -Mysekurity 21:00, 30 November 2005 (UTC)

requests for advertising

if anyone knows how to do it, can they post the link to this page on the wikipedia mailing lists, along with bugzilla? Thanks. --kizzle 06:59, 30 November 2005 (UTC)

I asked a little while ago about Bugzilla, and was told to wait. I'm not sure if this is the right time, but if someone (possibly me) could ask the devs in the IRC channel, it would be awesome. -Mysekurity 21:00, 30 November 2005 (UTC)

Am I missing something or...

Semi-protection could be temporary right? It doesn't say on the project page, but I would think that it would be an ideal tool against vandalism. Instead of doing vprotect, we could put semi protection on for a day or so until the threat ceases. That way, constructive edits could be done. I don't see that mentioned, but I assume it could be used for that. --Woohookitty^{(cat scratches)} 16:49, 30 November 2005 (UTC)

Yes, I like that idea a lot. This way, the vandalism will subdue, while still allowing good edits. The semi-pro can be removed as is done with regular full protection. I think this would probably be the best idea. -Mysekurity 21:00, 30 November 2005 (UTC)

OK. I'll alter the proposal a tad since right now, it doesn't mention this possibility. --Woohookitty^{(cat scratches)} 06:20, 1 December 2005 (UTC)

Suggestion in the form of a question...

Is there any way Wikipedia can be set up so that only registered users can contribute edits? I'm sure it's obvious to anyone that's ever had to revert an article, that the vandalism almost always comes from unregistered users. Just the usual thought that my penny can buy, odds are good I'll be getting change back from it. -- Cjmarsicano 17:30, 30 November 2005 (UTC)

It is very possible to do this; many wikis using MediaWiki software indeed do make registration a requirement for editing. However, one of the main points of Wikipedia is that all people can edit it, without having to register. There are many anosn who are good editors, perhaps just coming here to look up a topic, but seeing a typo and wanting to correct it without having to go through a registration process. I am positive that making registration mandatory will never happen as long as Jimbo is still an active personality in Wikipedia. - Pureblade | ^☼ 18:19, 30 November 2005 (UTC)

See Wikipedia:Village pump (perennial proposals)#Abolish anonymous users. It is a Foundation issue that will probably never receive support from either the Board, the community or Jimbo.-Splash^talk 18:21, 30 November 2005 (UTC)

Thank you for the link to Wikipedia:Disabling edits by unregistered users and stricter registration requirement. Wikipedia is experiencing exponential growth; I wonder what the outcome of that poll would be two years from today. Hall Monitor 19:19, 30 November 2005 (UTC)

This has been discussed ad nauseaum, and most people decided against it. Redwolf24 wrote a great essay on the subject, that, at some point, we were all newbie anons. Wikipedia is a great way to launch into wikis, so we have to be the best one out there. -Mysekurity 21:00, 30 November 2005 (UTC)

The nut

To me, the real question is a philosophical one. And that is -- which is worse: Someone coming to a vandalized article that says "George W. Bush, president of the United States, is a GIANT PENISHEAD!" and thinking "Man, this Wikipedia sucks, I'll never use it again" -- or someone coming to the article, not being able to edit, and therefore not getting addicted to Wikipedia enough to become a regular contributor? Which is worse? Damage to Wikipedia's reputation as a reference work, or damage to Wikipedia's pool of new users? · Katefan0^(scribble) 18:22, 30 November 2005 (UTC)

It seems obvious that the first situation would be worse. If the person would really "never use it again," then he would not be editing anyway. At least not being able to edit one article would not ruin the credibility of Wikipedia. - Pureblade | ^☼ 18:28, 30 November 2005 (UTC)

Could semi-protected pages come with some kind of disclaimer explaining why the person wishing to edit can not work on that particular page? BD2412 T 18:31, 30 November 2005 (UTC)

Kate, how many times do you think the former will happen as compared to the latter? BD, I like your idea. --kizzle 18:53, 30 November 2005 (UTC)

Considering that I'm reasonably sure more people view Wikipedia pages than actually attempt to edit them, I'm personally more concerned with the former. I'm uncomfortable with the proposals about only letting registered users edit, but this sort of semi-protection policy seems like a pretty reasonable middle ground. · Katefan0^(scribble) 18:57, 30 November 2005 (UTC)

If semi-protection is implemented, we definitely need to provide some sort of communication to our editors what the terms of the semi-protection are. I wish that we communicated something more definitive to newly registered users who wish to move pages as well, but we do not. As to which is worse, the first scenario obviously takes the cake. I am not convinced at all that we would damage our pool of users by protecting an otherwise highly trafficked and highly vandalised page such as this one. Were any users lost after the Main Page was protected [7] or we restricted page moves? Hall Monitor 18:55, 30 November 2005 (UTC)

I agree with Kate, and with BD (I've been invisining some idea along those lines, or maybe just like we have now with view source.) It doesn't matter how wrong the journalists are with their accusations of completely vandalized pages, the fact remains that it hurts our image. We need to make ourselves look respectible before we can look credible. The warnings would be very useful, and even if this fails, I think they should go into effect currently. -Mysekurity 21:00, 30 November 2005 (UTC)

George W. Bush protection

(Crossposted from WP:AN/I ^[8]) To me, the real question is a philosophical one. And that is -- which is worse: Someone coming to a vandalized article that says "George W. Bush, president of the United States, is a GIANT PENISHEAD!" and thinking "Man, this Wikipedia sucks, I'll never use it again" -- or someone coming to the article, not being able to edit, and therefore not getting addicted to Wikipedia enough to become a regular contributor? Which is worse? Damage to Wikipedia's reputation as a reference work, or damage to Wikipedia's pool of new users? · Katefan0^(scribble) 18:18, 30 November 2005 (UTC)

Wikipedia's reputation as a reference is very low anyway. Few articles cite their sources properly, and any high-profile article usually has serious issues of POV. Then, of course, the editors here are, for the most part, non-experts. Vandalism doesn't reduce the reference value of the work because it is so low anyway. Thus, the risk in protection is people not being able to edit the encyclopedia...that anyone can edit. -Splash^talk 18:31, 30 November 2005 (UTC)

Splash, I agree with you philosophically on many things here, but on this we part ways. I can't believe that the answer to point one is "lots of people already think we suck, so we might as well just continue to let them think we suck." ESPECIALLY when Wikipedia starts to get media coverage like this [9] · Katefan0^(scribble) 18:37, 30 November 2005 (UTC)

we can keep protecting in case of waves of vandalism and unprotecting later, lots of people have GWB on their watchlist, so it is extremely unlikely that vandalism goes unnoticed. Hopefully the problem will go away by itself in another three years :) journalists who rant about vandalized article versions simply have not got what we are about here, so never mind. No serious criticism of Wikipedia would even mention vandalism as obvious as "GIANT PENISHEAD" as an issue. Wikipedia's value is degraded much, much more by unreviewed, unreferenced false or inaccurate statements on obscure articles. dab (ᛏ) 18:46, 30 November 2005 (UTC)

You misunderstand something. There are often 50, 100 edits a day on that article - and often, 90-100% of them are vandalisms or reversions. My point? If there were unreviewed, unreferenced, false, or inaccurate statements, they'd be fixed in those 50-100 edits. They haven't been. Very, very few legitimate edits are performed on that article, it's not like there's a flood of them. They can be handled on an individual basis. We have a large enough corps of administrators. Do I suggest protecting it forever? No. Do I suggest protecting it until the devs give us a better option? Absolutely. I don't speak for every admin but I may just remove it from my watch list, not wanting to clean up after the mess of others, especially when my fellow admins encourage the mess. I would not be surprised if other admins are similarly annoyed with always being on vandal watch. --Golbez 19:03, 30 November 2005 (UTC)

We shouldn't worry too much about journalists ranting; it's what they're paid for and, as Dbachmann says, they haven't really 'got' Wikipedia. To Katefan0's specific point, no, I don't think we should give people a "Wiki sucks" first impression and to that extent we should be sure to speedy the worst of the trash and AfD the rest. I think more damaging than journalists ranting are stories in whichever paper it was about Wales having to protect the GWB article for the whole of 2004 (when it was only during the campaign) to which there is no "yeah, but" defence, apart from the fact that it wasn't Wales who protected it. Having an article protected from now until 2009 doesn't seem at all right to me. -Splash^talk 18:56, 30 November 2005 (UTC)

Don't you think this is slightly missing the point? Splash and DBachmann suggest we should ignore media criticism "because they don't get the point" of the wiki-way -- but is Wikipedia's ultimate purpose to be a wiki, or an encyclopedia? Which mandate reigns? The wiki format is a means to an end -- building an encyclopedia. Without the encyclopedia mandate, what's the point of the wiki? High-falutin' ideals are great, and I value the idea of open-source et al, it's clearly important to Wikipedia's success, but it also does no good if Wikipedia drowns in a cesspool of (valid) criticism. I can't imagine any newspaper writing an article criticizing Wikipedia for its articles being protected too much (people outside Wikipedia, frankly, don't realize or care); on the other hand, the USA Today editorial is a perfect example of the kind of things that WILL get noticed. · Katefan0^(scribble) 19:08, 30 November 2005 (UTC)

'the wiki format is a means to an end'? So once an article is complete it should be locked, and one day hardly any articles will be left to do and it will be finished? No, that is the encyclopedia on my bookcase. Locked the day it was published, outdated the day after. I think, if wiki is to work in the long term it can not go the route of locking things down. Otherwise it is exactly the same as any other encyclopedia. It will be necessary to educate readers as to the limitations of an encyclopedia of this nature. Oh, and a decent page rating system would help. Sandpiper 22:42, 3 December 2005 (UTC)

Ah, now that's a more persuasive argument. However, it is our only means to the end. We don't have another one available without becoming Britannica. Closing it off should be done only in emergency. -Splash^talk 19:43, 30 November 2005 (UTC)

One other point to DBachmann -- you suggest that "GIANT PENISHEAD" isn't as serious a vandalism as some other things a vandal could do, like introducing erroneous facts. I agree with that. But I think both are equally damaging -- if I saw something juvenile in an article I was looking at, the entire rest of the article, to me, would immediately be suspect. Same effect. · Katefan0^(scribble) 19:39, 30 November 2005 (UTC)

We're not as far apart as you think. That's more a less the statement I wrote some time ago on my user page. -Splash^talk 19:43, 30 November 2005 (UTC)

Then offer a better solution than simply letting the vandals run wild. --Golbez 19:03, 30 November 2005 (UTC)

Get more good editors, and give them a rollback button. Outweigh the vandals by demonstrating that anyone can fix the problem in a matter of seconds. Setting up admins as ad-hoc committees for editorial approval leaves us wide open to charges of censorship that would be hard to defend against, apart from simple protestation. Possibly, and only possibly, semi-protection is a solution to that. Full blown protection for 4 years isn't. -Splash^talk 19:43, 30 November 2005 (UTC)

I agree with Splash's concerns. I am not a big fan of locking pages and consider it unwiki...but, the page has been locked in accordance with procedure so there is no violation in that respect either. I also agree that if we do develop this semiprotection feature, it must only be used in the most egregious of circumstances...ie the Bush article.--MONGO 20:08, 30 November 2005 (UTC)

Sorry, Splash, we agree on lots of things, but this one isn't one of them. Last time I checked, throwing admins at the problem didn't solve the issue at hand: that we're wasting valuable time in reverting vandalism that we could be using improving the encyclopedia itself. While I sometimes do spend time reverting vandalism "just for the heck of it", I do not think it is the most efficient way to address the problem. I know that if I had seen "GEORGE W BUSH IS A FUCKHEAD" or something like that the first time I visited Wikipedia, I would not have joined. So, the sword goes both ways. Also, if we ever intend on actually fixing the credibility-as-an-encyclopedia problem, we need to take a first step. Perhaps fixing the POV and citation problems of other articles is more needed, but that does not mean that those needs and this proposal are mutually exclusive. Titoxd^{(?!? - did you read this?)} 23:55, 30 November 2005 (UTC)

I agree with Katefan0, that it hurts our image, but I also agre with Splash that we should keep this as wiki-like as possible. The problem is, however, I've seen many good websites bust because of issues within, and I think we should, first and foremost, not make us look like jackasses with images of penises and feces on our pages. We want to encourage new users, but a few pages that are high-profile that can't be edited immediately should not be a concern, they'll simply move on to another page, or maybe come back later to a different page and edit that. -Mysekurity 21:00, 30 November 2005 (UTC)

"Which is worse? Damage to Wikipedia's reputation as a reference work, or damage to Wikipedia's pool of new users?" Three hours of vandalism per day on high-profile pages is a much more potent threat to Wikipedia's "pool of new users" than semi-protection would be. After all, new readers would be likely never return to Wikipedia if they were to find ridiculous statements such as those from GWB quoted above. Perhaps the question should be "Which is worse? Damage to Wikipedia's pool of new users, or damage to Wikipedia's pool of highly experienced users?" --TantalumTelluride 21:39, 30 November 2005 (UTC)

Agreed. We just lost Essjay, what more is going to take? I love this project, but not enough to put countless hours into fighting needless vandalism, and dealing with stuff I really shouldn't be dealing with. There is absolutely no good reason to allow this to happen: we must find a better way. -Mysekurity 21:44, 30 November 2005 (UTC)

Well, I agree with that point as well. Vandalism patrol can often mean lots of wasted wikitime. Some would argue that folks enjoy it and that makes the situation less desperate than it seems. Maybe that's true. Personally, I do vandalism fighting-related activities because they're desperately needed, but it's not really the way I prefer to spend my time here. · Katefan0^(scribble) 21:46, 30 November 2005 (UTC)

I just noticed that the other Kate is on "indefinite wikibreak." Indefinite? There's still a glimmer of hope in that word. Hopefully her break will not be infinite. She has apparently left becuase of the editcountitis epidemic that is sweeping RfA. Essjay left primarily because of the bitterness surrounding the CCW's MfD. But most experienced users who leave, including Essjay and Kate, are motivated not by single incidents but by an overall feeling of wiki-stress and desperation. (I know this sounds dramatic, but it takes a serious problem for such dedicated contributors as Essjay and Kate to abandon Wikipedia.) Undoubtedly the vandalism situation is an important factor. No one can say it better than RickK. There is a fatal flaw in the system, that flaw is my reason for continuing here, and I support semi-protection because it might play a part in correcting that flaw. --TantalumTelluride 23:29, 30 November 2005 (UTC)

From Replies to common objections:

In short, it's not the end of the world when people leave Wikipedia, provided they are replaced with "fresh blood." On the other hand, where there are systematic problems causing many people to leave, that's something we have to address.

Doc glasgow has left now, and part of his departure message is very clear: there is a systematic problem with Wikipedia that is driving off way too many editors (especially points 2 and 6), and I couldn't have said it beeter than him: Fundamentally, wiikipedia needs to decide if it wants to be a useful and relatively reliable encyclopaedia using a wiki format, or a wiki posing as an encyclopaedia – but actually being a graffiti wall and a dump for all the world’s unverified cruft. Isn't it time to address this issue? Titoxd^{(?!? - did you read this?)} 00:54, 1 December 2005 (UTC)

When it comes to "being a wiki" vs. "credability", obviously, I will go with the later. Also, we will loss many possible users if they say penises all over the article they first come up upon. ALso, this issue is causing the loss of current users too.Voice of All^T|@|ESP 00:58, 1 December 2005 (UTC)

I think Katefan hit the nail right on the head by saying "is Wikipedia's ultimate purpose to be a wiki, or an encyclopedia?" Which are we more aiming for? The wiki idea is great, but it is not infalliable. We must be more of an encyclopedia than simply a wiki, as that is our purpose here. I believe in the resiliency of the wiki, but only so much. You cannot honestly tell me that a new user will leave in disgust if they can't edit a page, so please do not even compare it to images of penises, feces, Hitler, or other juvinile vandalism. There is no way anyone can make an argument that these two are even remotely comparable. The issue at stake here is not Wikipedia's actual credibility (or lack thereof) or its ability to revert vandalism quickly. No; to an outside user/journalist, vandalism is vandalism, no matter how quickly it can be reverted. I am saddened to see the spirit of the wiki die, but we must prevent vandalism from even showing up, and thereby keep or restore our credibility. The wiki format is not yet mature enough to defend its practices. What is the disadvantage between semi-protecting an article for a short period of time v.s. full protecting? Keep in mind this is not for for every article, and will be tightly controlled. I just cannot sit back and watch GWB become vandalized 80 times a day. There is no excuse. -Mysekurity 02:37, 1 December 2005 (UTC)

You see the spirit of the Wiki die. I will oppose permanently any policy or technical change that causes that to happen. Wikipedia's purpose is to build an encyclopedia using the Wiki method. The two are non-separable. -Splash^talk 02:42, 1 December 2005 (UTC)

Right now, the only alternative we have is to completely protect an article, which is even more anti-Wiki. Titoxd^{(?!? - did you read this?)} 02:48, 1 December 2005 (UTC)

We've lost the point of this thread. It was: should GWB be protected from now until 2009. Yes/no. -Splash^talk 02:50, 1 December 2005 (UTC)

As our popularity grows, so will vandalism - I fear that the time is not far off when this article (unless protected to some degree) will be vandalised three or four or five hundred times per day. BD2412 T 02:52, 1 December 2005 (UTC)

Until 2009? No. Full protection is evil. But we can try to find something to cut down on the vandalism already occuring, and this proposal fits the bill. Titoxd^{(?!? - did you read this?)} 02:53, 1 December 2005 (UTC)

So, keeping this thread on-topic, when can I unprotect George W. Bush? -Splash^talk 03:20, 1 December 2005 (UTC)

When we have a system in place that can prevent the random vandalism from overwhelming the article - like the semi-protection under discussion, or perhaps bots that are smart enough to automatically revert the obvious things (deletion of large chunks of text, insertion of dirty words, etc.) BD2412 T 03:22, 1 December 2005 (UTC)

I meant seriously. One day, two?-Splash^talk 03:26, 1 December 2005 (UTC)

Well, try it now. Maybe the latest wave has been capped by the protection to this point. BD2412 T 03:37, 1 December 2005 (UTC)

I unprotected it. --Woohookitty^{(cat scratches)} 07:08, 1 December 2005 (UTC)

BD2412, could you please tell me what defines a latest wave? There are waves of vandalism on GWB, there is also the thick background noise of vandalism, and the two have little to differentiate them except volume. I think it's a constant issue, and will not let up, so why pander to them? Why coddle them with the resignation of "It's the wiki way?" --Golbez 15:31, 1 December 2005 (UTC)

This entire thread is proof that we NEED semi-protect.--MONGO 20:59, 1 December 2005 (UTC)

Revoking permission

We could revoke permission to edit protected pages (this gets a little complicated) but would prevent people from abusing the semi-pro part (e.g. a vandal warned with {{bv}}/multiple-blocked vandal/editor who edits a (temporarily) semi'd page. Thoughts? -Mysekurity 21:00, 30 November 2005 (UTC)

P.S. Are we ready to archive yet? We're at 41 kb or something -Mysekurity 21:00, 30 November 2005 (UTC)

Lets not break the thread just yet...give it another day or so.--MONGO 21:36, 30 November 2005 (UTC)

No, hold on. There's no need to archive yet, 32 KB is a subjective limit. Titoxd^{(?!? - did you read this?)} 23:57, 30 November 2005 (UTC)

Percent-vandalism threshold for semi-protection

To address the issue of protection creep (The second kind, as suggested by Splash), I propose that there be some sort of threshold for the percentage of edits that are vandalism-related before a page may be considered for semi-protection. For example, if the threshold for semiprotection were set at 80% in the last 50 edits, then over 80% of the last 50 edits to an article would have to be vandalism-related before an admin could arbitrarily protect it. Pages would be eligible for un-semi-protection after a predetermined interval of time (unless, of course, the vandalism rate were to still exceed the threshold). Even if my proposal is rejected, I still support the introduction of semi-protection. It might help significantly in reducing the wiki-stress that leads to premature wiki-death. And, for the record, I think that the privilege of editing the Main page should be reserved for those editors who have been determined trustworthy by the community. Now, please reveal the fundamental flaw in my plan. --TantalumTelluride 22:27, 30 November 2005 (UTC)

I'm not sure something like semi-protection should be decided on an automatic level. Caution should be exercised, possibly through multiple admins or arbcom, but setting some arbitrary limit doesn't seem to be the answer, as cases may come up that hit right below the mark that still warrant semi-protection. In addition, what you'll get on moderately vandalized pages is legitimate users vandalizing the page in order to get to the semi-protection threshold, I guarantee it. --kizzle 01:43, 1 December 2005 (UTC)

Never mind my idea. I think admins can be trusted to use good judgement when semi-protecting pages. They won't protect too many, and they'll unprotect them when it is appropriate. --TantalumTelluride 21:33, 1 December 2005 (UTC)

Concerned

I'm kind of concerned about this proposal, to be frank. While admittedly some pages (George W. Bush included) do get a large amount of vandalism, that shouldn't be a reason to hinder progress. Progress on making an article better shouldn't be sacrificed or hindered just for the sake of attempting to prevent vandalism. In my opinion, setting arbitrary edit thresholds for editing very public and high-traffic pages is a bad idea. While the principle - attempting to strike a balance between editing and stopping vandalism - is good, I don't like the implementation. What is 25, 50, or 100 edits to a vandal? Vandals, including those who are determined enough to write a 'bot to vandalise, won't let an arbitrary number of edits stop them, especially after they start vandalising other pages first and then turn to the high-profile pages. Potential users, on the other hand, will be deterred by this - Hey, isn't this supposed to be edit-able by me? I know several users that I have talked to would probably not have joined if the first article they tried to edit was restricted. I'm also worried that the increasing proposals to restrict editing will slowly but surely chip away at the spirit and freedom of the Wiki. What's to stop all the articles from being semi-protected like this in one month, two months, or three months? Anyone - regardless of edit counts, ISP, country, or IP address should be able to edit Wikipedia. Finally, I've seen some analogies above to when the Main page was protected. You can't really compare the two; people expect a high-profile website's front page to be locked — besides, it's not an article. People don't expect articles, especially high-profile ones, to be protected, even if it's semi-protection. Thus, in my opinion, we shouldn't sacrifice progress in an attempt to hinder vandalism. Thanks! Flcelloguy (A note?) 23:29, 30 November 2005 (UTC)

What do you think about the proposal to, instead of going by edit count, by age of the account? Similar to what's done with page moves. · Katefan0^(scribble) 23:33, 30 November 2005 (UTC)

I think my suggestion above can prevent too many articles from being semi-protected. It would allow only extremely vandalized pages to be semi-protected, and it would set a time-table for un-semi-protection. The number of admins and highly experienced users leaving recently has convcinced me that the current attitude towards vandalism is not beneficial to Wikipedia in the long run. --TantalumTelluride 23:37, 30 November 2005 (UTC)

Of course, all vandalism won't be stopped. The determined vandals will still be a problem, but many of the thousands of minor vandals that add stupid comments to high-profile articles might be deterred. Vandalism will still occur on "normal" non-protected pages, but semi-protection can help reduce vandalism considerably on such articles as George W. Bush and Saddam Hussein. Legitimate users with the tenure required to edit semi-protected articles will be much more productive with the resulting coherent history pages for the articles, most of which are critical to Wikipedia 1.0. --TantalumTelluride 00:05, 1 December 2005 (UTC)

Age of account certainly is an interesting proposal; however, I would still be concerned about the hindrance of progress in an effort to stop vandalism. Regarding the minimum threshold of vandalism — isn't the line between vandalism and non-vandalism be extremely ambigious with some edits? Who would spend time counting the exact edits? What distinguishes an article that has 79.9% vandalism and one that has 80.1% vandalism? Who likes rhetorical questions? :-) No, seriously though, my point is that I don't think setting an arbitrary percent that can be calculated at any random time would be too effective. If the semi-protection worked out, I would support just a general guideline advocating when to semi-protect, similar to the guidelines we have for protection now. However, I still am concerned about the blocking of progress in an attempt to stem vandalism. Of course all vandalism won't be stopped — that's a fact of life that comes with a wiki. However, the question should be whether the benefits of this proposal outweight the downsides and risks; I believe that there are too many risks, uncertainties and unknowns to overcome the (in my opinion) limited benefits that we can get. Vandals, in my opinion, seem to be more determined than potential users stumbling upon Wikipedia for the first time; vandals won't let account age or editcountitis stop them. We shouldn't risk losing many potential valuable contributors in the hopes of stopping a few vandals. Thanks! Flcelloguy (A note?) 00:35, 1 December 2005 (UTC)

I'm for this proposal in that its spirit is to block blantant vandalism only...Having certain articles editable only by aged accounts would curb a LOT of the major acts of vandalism taking place everyday. Even if the age was low, say 3 days, that would still help. Heck, even if it is nothing more then Requiring any account to edit would help. While of coure any current vandalism IS removable by anyone, it takes diligent editors to remove it as soon as possible, thus keeping the encyclopedia as accurate as possible, as often as possible. xaosflux ^T/_C 04:50, 3 December 2005 (UTC)

Three things to consider

Lets focus for a second on three things which really haven't been resolved yet that are irrespective of whether one believes in semi-protection or not...

Semi-protection Process

How will semi-protection be applied to a page? Should it go through arbcom? A majority vote of at least X amount of admins (10-25)? Should there be a physical limit to the amount of semi-protected pages on Wikipedia, such as 10-15? Should it simply be decided by the vandalism percentage rate? Can something like this be decided automatically or on a case-by-case method? How will your said proposal address protection creep, or the tendency for exponential increases in the amount of semi-protected pages? How should pages become candidates for semi-protection, through the community like request for protection exists today, or through admin discussion? Should semi-protection self-expire after a predetermined time, or should it change status according to a vote? If it self-expires, what about continually vandalized pages such as George W. Bush? How will the enabling of semi-protection as said affect the request protection page? As Coffee talk lady says, Tawk amongst yourselves. --kizzle 01:40, 1 December 2005 (UTC)

Right off the bat, it should not go to ArbCom. Perhaps it should go to WP:AN or some other place that is visited often, but ArbCom is too busy to deal with these sorts of things. Titoxd^{(?!? - did you read this?)} 03:31, 1 December 2005 (UTC)

As I asked you above kizzle, why would the semi-protection process, which is far less restrictive than protection, have to go through a much more restrictive process than protection does? --Woohookitty^{(cat scratches)} 06:27, 1 December 2005 (UTC)

Because semi-protection could end up being a lot more permanent on some pages than mere protection. When a page is protected, there is a clock ticking where either a POV dispute or edit war is resolved, or a flood of vandalism is stopped. Semi-protection, especially on such pages as GWB, may become a permanent fixture of the page, remaining semi-protected for several months, possibly even until 2009. Such permanence requires a hell of a lot more discretion and caution on our parts. Imagine other pages which aren't quite vandalized a lot, and lets say one get a flurry of vandalism. An admin could rightfully up the level to 100+ edits, and then justify keeping it there for a lot longer than the vandalism flurry. It will become highly easy to justify leaving a page with a requirement of 100+ edits, as it would stop vandalism and no obvious signs that any editors had been turned away could be discerned. Thus, without such discretion, semi-protection would be endangered of the very thing Splash keeps talking about: protection creep. A time limit would generally solve this problem, but on some pages like GWB, permanent semi-protection is definetely warranted. Thus, either semi-protection should come in the temporary form up to a maximum of a week, which can be initiated by any admin, or a more permanent form, which should come from a concensus of at least several admins to respect the wiki way. --kizzle 06:40, 1 December 2005 (UTC)

I don't think that Bush would need it permanently. I still think that the vandals would be discouraged by semi-protection so that maybe it would be lowered to the point where it would not be needed. I'm not sure it'll *ever* be needed permanently. --Woohookitty^{(cat scratches)} 06:48, 1 December 2005 (UTC)

Lets be honest: vandals are not going to give up soon. So much of vandalism comes from drive-by vandals that they will never be discouraged by a previous semi-protection notice they never saw. I'll wager you a beer that if semi-protection gets enabled, that after a month of semi-protection, within a week of removing it, vandalism will be at least up to 75% of what it is now. --kizzle 06:53, 1 December 2005 (UTC)

An addendum to my previous post, it would seem that enabling a permanent semi-protection would be hard to justify undoing later. Semi-protection should come in the form of a week (initiated by any admin), a month (initiated and seconded by another admin), or a 3 month period (initiated by at least 5 admins). Thus, if it truly needs semi-protection for longer, it simply can be "renewed" every 3 months. In addition, semi-protecting a page for a week more than two consecutive times requires the consent of 5 admins as well. --kizzle 06:54, 1 December 2005 (UTC)

Actually I can accept that. I'd go for maybe extending it to 2 weeks for a single admin, but you are right. We probably should add more admins to make it permanent. If anything else, it'd gain consensus, which is one of our mantras. More powerful in numbers. Does anyone know how these pages that were permanently protected (like Main Page) got that way? Was it via a vote?? --Woohookitty^{(cat scratches)} 07:03, 1 December 2005 (UTC)

<carriage return> This being protection creep (of the first kind: only ever upwards), there should be a means of de-creeping, too. I have no particular suggestion that doesn't result in the creation of protectionism, unprotectionism and nonprotectionism, but I'm just pointing out that we need a way down as well as a way up. -Splash^talk 07:08, 1 December 2005 (UTC)

Here's something that confuses me. We have...no protection, semi-protection level 1, level 2 and then protection. Why can't we just have 1 semi-protection level? It would decrease the possibility of creep. I don't see why we'd have to split semi-protection into 2 when really, full protection is just one level. Let's not over complicate things, I'd say. --Woohookitty^{(cat scratches)} 07:17, 1 December 2005 (UTC)

I think there should be two, because in some cases, a 10 edit requirement is enough to combat vandalism and help new and anon users as well, whereas in some cases, like GWB, 100 should be used. We can even require that 10 be tried first, and if vandalism continues, upgrade to 100. As for Splash's comment, if we apply a time limit to semi-protection along with increasing amounts of admin concensus for consecutive applications of such protection, it will get progressively harder for semi-protection to expand beyond its essential uses, thus I'm not too worried about protection creep in this implementation. --kizzle 08:34, 1 December 2005 (UTC)

I would prefer that we not do time limits. If you look at the protection cases we have now, there are no rules per se. Time limits are just not practical. In some cases, a day will be enough. In some, 2 weeks will be enough. In some cases, it might be longer. It all depends on the situation. I really dislike having lack of wiggle room. --Woohookitty^{(cat scratches)} 09:01, 1 December 2005 (UTC)

Keep in mind the reasons for protecting and semi-protecting a page when you talk about time limits. Protection generally occurs when there is a POV dispute or an edit war, of which both are observable within the talk page. Some POV disputes may last weeks (as in our good friend Rex at John Kerry), and some may last a day. This variability requires the length of protection to be fluid to adapt to the situation. However, the reasons for semi-protection are different. We only semi-protect a page when there is vandalism occuring, at the unobservable consequence of restricting valid anonymous contributions. There needs to be a time limit set for each semi-protection, otherwise the reality will be that many pages will simply remain endlessly in their state of semi-protection, as it has observable benefits (no vandalism) and unobservable consequences (no valid drive-by users). Forcing time limits along with requiring escalating thresholds of admin concensus strikes a balance between those who wish to combat vandalism and those who wish to cater to the benevolent anonymous drive-by contributor. --kizzle 09:23, 1 December 2005 (UTC)

I just don't see it that way. I don't see why semi-protection would need time limits. I don't think we can assume anything. I don't think we can assume that it will be permanent. Like with the Bush article. I think it's more likely that the page will be semi-protected for a few days...it'll be taken off as a "test" and then possibly put back on. I'm not being critical, but I'd suggest looking at the pages we protect now. Admins are always double checking them to make sure that they should be protected. I think semi-protection will be similar. Every day, we will have admins (probably me, dmcdevit, katefan0 and others) looking at the semi-protected articles to make sure that they should be kept semi protected. I think you are making a faulty assumption. I think we need to start slow with it...if it looks like you are right and it'll be mostly permanent, THEN we can do time limits. I just really don't like putting shackles on this on the get go. I think it's counter productive. And how do you know tht it'll just be used for vandalism? I could see it used to stop edit wars in extreme cases, since there are times that anon IPs and new users (look at Bogdanov Affair) engage in endless edit warring. --Woohookitty^{(cat scratches)} 10:01, 1 December 2005 (UTC)

Seeing as semi-protection has observable benefits (no vandalism) and unobservable consequences (no valid anonymous driveby edits), you really can't see a temptation to leave certain pages semi-protected for a long period of time? Setting semi-protection for pages that engage in endless edit warring is missing the point: in the case of an edit war or POV dispute, we would protect the entire page, even in extreme cases. If we only semi-protected in such disputes, it would allow certain editors in the discussion to edit the page while others couldn't simply because they hadn't been on as long, which would be fundamentally unfair. Just because an edit war involves new or anon users, doesn't mean they are in the wrong. The only reason for semi-protection to be applied to a page is to combat cases of extreme vandalism. POV disputes and edit wars should still warrant page protection until the dispute is over, which could last anywhere from a day to several weeks. However, since semi-protection deals with vandalism, there is no visible evidence to warrant un-semi-protecting a page. This will simply be the case, that many pages become semi-protected long after an initial flood of vandalism. This doesn't mean it can't be semi-protected for longer, it just requires more of an admin concensus than a single person. If we don't set time-limits, then Splash's concerns will become a reality. --kizzle 18:31, 1 December 2005 (UTC)

You make some good points, but. :) I still say...let's start it with no time limits just to see how this works out. If articles are being routinely semi-protected beyond say...4 weeks...then we can change the policy and add time limits. I just don't think we can assume anything with a new policy. It has literally never been tried and we really have no idea how anyone will react. The other idea would be to do a trial for like...a month...analyze the results and then make a permanent policy, which could include time limits. I can see us having time limits. I just don't want to start with them. I want this to be open for awhile just to see how it pans out because we can sit here and argue theoretical and what we think will happen, but we won't really know until we start doing this.

And btw kizzle, I wasn't saying that just new users and anons can get into edit wars. I don't know if you know the Bogdanov Affair case, but essentially, it's an outside controversy...something outside of Wikipedia. It's about 2 men who came out with some new physics theories. It's become controversial because some thing they made the stuff up. To make a long story short, we ended up with countless sockpuppets and new users constantly edit warring on the page. Eventually the arbcom ruled that they can be banned on sight if there only edits have been to the Bogdanov Affair article. Now see on that one, we could've just put it in semi-protection for awhile and that might've let us avoid the arbcom case. That's the kind of case I mean. And as Wikipedia gets bigger and more prominent, cases like that might happen more often. It was just a thought. --Woohookitty^{(cat scratches)} 18:43, 1 December 2005 (UTC)

For the example you give, it is the sockpuppetry which warrants semi-protection, as I consider sockpuppetry a form of vandalism. We can test it out for a month and see what happens, but I'd like to hash out as much as possible to get it right now, so that it has less of a chance of failing and having people cry "See, it didn't work, that's what happens when you mess with the wiki way!" Let me ask you this, Woohoo. Seeing as if semi-protection is doing its job and a previously vandalized page is no longer vandalized, what specific criteria would you use to determine when "making sure that they should be kept semi protected"? If there is no vandalism, how long do you wait before un-semiprotecting the page? Is this amount of time arbitrarily picked out of the blue, or will you wait the same amount of time, keeping in mind the only evidence you have is the lack of vandalism? Do you think other admins will be as conservative as you are in leaving semi-protection on certain pages? --kizzle 18:52, 1 December 2005 (UTC)

That's hard to answer. And no that's not a cop out. :) I say it because the admins have their own standards. Think of it as umpires in a baseball game. For some umps, a pitch can be called a strike by one umpire, but a ball on the same pitch by a different umpire. We have the same general standards, but some allow more than others. So it's really hard to answer that. And again, it's going to depend on the case. And if we keep the talk pages completely unprotected, we might be able to see from there whether the vandalism has gone away. So I can't really say, kizzle. I wish I could. It's just going to vary. On the protection page, we often give pages a "try out" for a few days just to see if the vandalism subsides. We might end up with something similar for semi-protection. There really is no other way to test it. --Woohookitty^{(cat scratches)} 19:00, 1 December 2005 (UTC)

That's ok :) but keep in mind, we're developing policy here. Not everyone is going to treat it as you do, and allowing such variability when it is not needed (unlike protection) is detrimental and increases if not guarantees the possibility of protection creep. We need to assume that both the most dilligent and the most incompetent admins will be using this policy. Though you might be conservative in leaving pages semi-protected, other admins could see it vastly different and keep pages semi-protected for a long long time. If you came along the page and told him that the page should be un-protected and he disagreed, who would be right? How would it be decided? By making this policy as explicit as possible, we reduce the chance of ambiguity in its application. This does not require testing on our part, as it is either going to reduce vandalism or not. Lets say it does. How will the lack of a time limit affect the potential for protection creep? --kizzle 19:12, 1 December 2005 (UTC)

Well I would rather do it like we do the protection page. But...if you are looking for other ideas...Maybe require that 2 admins vote to un-semi-protect. The arbcom uses a voting system I like where they have a requirement of say...4 support votes...but every no vote takes away a yes vote. I can see us doing something like that. There's also the idea of maybe electing a committee of admins to do this. That way we'd always have x number of admins voting. Because like I said, for the protection page, it's voluntary. Right now there is 3 of us on it full time but there has been 5...for awhile they had one guy doing it and that was it. It'd be hard to require a vote of 2 for something if suddenly we only had one admin doing the entire page, which is a possibility. We have 700 admins but many of them do not do "dirty work" like this. --Woohookitty^{(cat scratches)} 19:28, 1 December 2005 (UTC)

I just realized I didn't answer some of your questions. :) As for criteria, there are a few we can use. First of all, we can look at the talk page. What's happened with Talk:Main Page is that because the main page is protected but it's talk page has not, the vandals have vandalized the talk page instead. We might be able to judge vandalism by that. Vandalism on talk pages isn't quite as big of a deal simply because it's not as visible to the average reader of the site. I mean it's not good :) But for our purposes, it's not as bad. We can also look at what kinds of editors are editing the article page. Is it people just over the threshold? If it is, then maybe unprotection isn't the best thing to do. Other than that, time would be an issue. If it's been 2 weeks, then maybe unprotection could be tried.

I do not want a fixed time limit on when to try to unprotect the article, because that would just be too much of a signal to vandals. It's why we don't do it for protection now. I mean if you say, ok...in 2 weeks, we have to try unprotection, well then the vandals will wait until it's unprotected and then hit it harder than before. We've had it happen on the protection page several times. --Woohookitty^{(cat scratches)} 19:15, 1 December 2005 (UTC)

We don't have to and probably shouldn't advertise when the page will be un-semiprotected, the only way they could tell is if they went to the semi-protection page, looked up when it was initiated and for how long, and then waited that long. Keep in mind that a significant majority of vandalism stems from drive-by vandals. If a page has vandals dedicated enough to wait the week, then we can just up it to 3 months with the 5 admin concensus. I'm telling you this right now, and I will bet a large amount of money on it, if we don't set time-limits, you're going to see the amount of pages semi-protected jump a considerable amount and stay that way. The funny thing is, I really am more on your side about semi-protection, I'd rather it be applied very liberally to a lot of pages without restricting the time limit, but I do agree that Splash has a valid point (which I wish he would chime in with) that unlike normal protection, semi-protection has a definite potential to expand quite beyond its original use. By the way, when you get a chance, if you could answer my previous question, how will a time limit or lack thereof affect the potential for protection creep? --kizzle 19:25, 1 December 2005 (UTC)

I understand. But. One thing to consider is something I mentioned up above. We literally only have 3 admins doing the protection page right now. We're going to have to set up a committee or require a "quorum" of admins present to perform functions or something because otherwise there will be times where we won't even have 5 admins working on semi-protection. It's just a fact of Wikipedia. Alot of the admins just do not like doing this sort of work. So I understand the concerns of Splash about creep, but in terms of admins to start semi-protection and extend it, we need to look at stuff like quorum and committees. In other words, that's part of stopping creep as well. I mean if we never have 5 admins working on the s-p page and we have this 5 vote rule, then yes, pages will be semi-protected forever. That's why we need to be careful on time limits and minimum votes and such. --Woohookitty^{(cat scratches)} 19:36, 1 December 2005 (UTC)

I realize that admins won't like doing this stuff, but you didn't go through a popularity contest just to get a rollback button and block people who disagree with you ;). A protection/semi-protection informal committee should be formed, and a push to recruit at least a dozen out of the 700 admins to help with the matter should be made. I personally don't think un-semi-protecting a page should require 5 or any amount of votes, it is the process of semi-protecting that should require these votes. That's like passing another test only if you don't want to renew your driver's license. If this were so, not many people would want to make the effort to get 5 votes just to unprotect a page. The duress of concensus should be placed upon keeping the page semi-protected in order to address concerns of protection creep. Just to clarify, my proposal is that any admin can semi-protect a page up to one week, an admin seconded by another can semiprotect up to a month, and 5 admins semiprotect up to 3 months, where any more than 2 consecutive semiprotections require the consent of 5 admins as well. --kizzle 19:43, 1 December 2005 (UTC)

OK. Well you and I seem to be making progress. Hopefully others will chime in now. :) Yeah an informal committee would be fine. We'd need something. I mentioned all of this because it hadn't been mentioned before. I'd love to say...hey...we're going to have 20 admins working on this around the clock...but we won't. Alot of it isn't just that admins don't like doing it. Alot of it is numbers. Yes we have 700 admins but we have lots and lots of projects for them to work on. RC Patrol, ArbCom, mediation, etc, etc, etc. --Woohookitty^{(cat scratches)} 19:52, 1 December 2005 (UTC)

Criteria to consider within Semi-protection

Should the levels of semi-protection address edit count, or the length of time an account has been used? How do IP addresses factor in the previous question? How will AOL and other proxy accounts be addressed? Should there be more levels than simply 25/100, such as 25/100/500?--kizzle 01:40, 1 December 2005 (UTC)

"Should the levels of semi-protection address edit count, or the length of time an account has been used?" Either one will probably have the same results. "How do IP addresses factor in the previous question?" Anonymous contributors probably shouldn't be allowed to edit semi-protected pages anyway. Requiring them to register before editing semi-protected pages does not significantly reduce their freedom because registering actually makes them more anonymous if they choose not to reveal their real names. "How will AOL and other proxy accounts be addressed?" In my opinion, AOL users should stop whining and just register an account! Unfortunately, the MediaWiki software has no means of identifying anonymous users except by their IP addresses. People with dynamic IP's will just have to register if they want to accumulate edits or account age or even if they want to receive the correct messages on their talk pages. "Should there be more levels than simply 25/100, such as 25/100/500?" I don't think there should be more than one level, at least until the system is tested. We're not planning on using semi-protection on a lot of articles, only a few. --TantalumTelluride 05:06, 1 December 2005 (UTC)

You have to be logged in for any kind of semi-protect anyway. So this is not much of an issue.Voice of All^T|@|ESP 18:51, 1 December 2005 (UTC)

I didn't realize that, I just thought it was minimum amount of edits. I didn't think we decided that already, you're saying if we have a legitimate anon editor with a static IP and over 500 edits, that person would be banned from editing any semi-protected pages? --kizzle 18:58, 1 December 2005 (UTC)

Yeah I have that same question. I don't like that if that's the case. I mean some of our anons are good users. They are far outnumbered by the vandals, but still. --Woohookitty^{(cat scratches)} 19:03, 1 December 2005 (UTC)

One thing to consider, how will AOL proxies factor in the edit count? Since AOL IP addresses are used by multiple people, won't some AOL vandals be able to get over this requirement if many people have edited before on their ip address? --kizzle 19:28, 1 December 2005 (UTC)

Anonymous users shouldn't be allowed to edit SP'ed pages. If they want to edit a page, they should create an account, just as AOL users must create an account if they want to recieve their messages. It's not like it's a big deal to register. All you have to do is choose a name and password. If you're not willing to do that, then you shouldn't need to edit an SP'ed page. (Anons will still be able to edit most pages because only a few pages will be SP'ed.) --TantalumTelluride 21:21, 1 December 2005 (UTC)

If it will be a problem to semi-protect through an edit count against anon AOL ips, then I agree we should enforce only logged-in accounts. --kizzle 21:57, 1 December 2005 (UTC)

If you are a good anon, 25 edits is not hard to get, and it is only on a few of 800,000+ articles until you build up the edits on your new account. Also, IP edit histories naturally build up from random users over time, so we should not try to include IPs here, this on top of all of the AOL issues, lets keep it to logged in users.Voice of All^T|@|ESP 05:52, 2 December 2005 (UTC)

Effect of semi-protection on visible content

Should there be a notice of some sort notifying people that the page is semi-protected? If so, how should the message be crafted without inferring illegitimacy within the text, such as NPOV tags? --kizzle 01:48, 1 December 2005 (UTC)

Good idea.... Well, when a page is nominated for deletion, the template placed on its talk page contains a link to the deletion policy. Perhaps a template could be placed at the top of a semi-protected page's talk page with a link to a project page describing the possible reasons behind the protection, a sort of semi-protection policy. Alternatively, if we decide that a high vandalism rate is absolutely the only legitimate reason to semi-protect a page, then the template can say "This page has been semi-protected because it is a common target of vandalism." It's short and to the point, but readers could find out more information by following the link. --TantalumTelluride 05:26, 1 December 2005 (UTC)

Look at any of the pages on the protected pages page to see what the protection tag looks like. The 2 main ones are Vprotected and protected. I'm sure we could just modify one of those. --Woohookitty^{(cat scratches)} 09:05, 1 December 2005 (UTC)

Agreed - Vprotected is a good model (as the page would only ever be semi-protected, or perhaps Sprotected, for vandalism). BD2412 T 19:48, 1 December 2005 (UTC)

One suggestion, maybe if instead of just showing view source on the top like normal protection, it still says edit page for those who have an edit count under the threshold. Then, if this person clicks on edit this page, it will give them some kind of friendly error message saying this page is a frequent target for vandalism, but that many other pages are available to edit, so that it still encourages those newcomers to remain and simply edit a different page. --kizzle 20:31, 1 December 2005 (UTC)

Sounds good to me. Whether the wiki software can do that is another question. --Woohookitty^{(cat scratches)} 20:43, 1 December 2005 (UTC)

I'm pretty sure the devs can work something out. Titoxd^{(?!? - did you read this?)} 20:51, 1 December 2005 (UTC)

I agree also. --TantalumTelluride 21:27, 1 December 2005 (UTC)

I'd add that said error message page should list all of the semi-protected pages (of which there should only be a handful) so our prospective editor does not jump right to another one closed to his or her contributions. BD2412 T 21:29, 1 December 2005 (UTC)

Good idea. Also, new users would realize that only a few pages are on the list, and that they can edit the other (6,911,996 minus the number of protected pages) pages on Wikipedia. --TantalumTelluride 22:57, 1 December 2005 (UTC)

We could say that too - maybe even point 'em to projects needing assistance. BD2412 T 23:13, 1 December 2005 (UTC)

Just some points

I read the discussion up above (good discusion) and I just have some points. First of all, again, remember that semi-protection would NOT have to be permanent. It could be used on an ad hoc basis when its needed to stop vandalism. It could be on the articles for less than a day in some cases. Honestly, I'm not sure it'd have to be permanent for Bush. I mean, I can see a day where vandals get tired of not being able to edit the page, so they give up.

Secondly, to splash, just because people might enjoy RC patrol doesn't mean that it's a good thing. I mean if vandalism ceased tomorrow, I highly doubt that people would be too sad. I'm not sure how much RC patrol you've done, but it's very draining. We essentially lost Doc glasgow over it and we'll lose others, I guarentee you. I just don't like this idea of "well it sucks, it's always going to suck, so let's not change it". Just because it sucks now doesn't mean that it's right or that it's something that we can't ever change.

Thirdly, this isn't as severe as I think people are thinking it is. #1 we would not be talking lots of articles here. Look at the protected pages list. We have 100 pages protected. And 1/3rd of those are temporary. With semi-protection, we'd be looking at another 50 or so probably? So 150 articles out of 800,000. I think it's a small price to pay to make the encyclopedia more readable and to ease the burden on admins. Bush is literally vandalized every 5-10 minutes. I just don't see where it's a big deal here. We're not talking thousands of articles. We're talking less than 100. I don't see jumping from this to "we're stopping the wiki element of the site". it's alarmist. --Woohookitty^{(cat scratches)} 06:40, 1 December 2005 (UTC)

Agreed with every point except for the conjecture that one day vandals will get tired of vandalizing GWB. After 5 years of being president, vandals have only gotten more determined than before, I highly doubt they will ever give up until he's out of office and a new douchebag is president. --kizzle 06:46, 1 December 2005 (UTC)

We won't know until we try, now will we? :) I dunno. I understand your concerns that we need to have a consensus for long term s-p. I just want this to pass. It's my main goal. --Woohookitty^{(cat scratches)} 07:06, 1 December 2005 (UTC)

I just range-blocked an entire University for GWB vandalism. The attacks were coming way too fast over several proxy servers, so I had no choice. Now, there is something that will irritate potential editors—being blocked—and I wouldn't have had to do it if the page were semi-protected. Titoxd^{(?!? - did you read this?)} 20:08, 1 December 2005 (UTC)

Yep. I totally agree. That's why I'm really going to bat for semi-protection. You're talking to someone who does RC patrol AND the RfP page. :) We really really need an intermediate step. Otherwise, protection is too harsh and having it completely open is too lenient. --Woohookitty^{(cat scratches)} 20:14, 1 December 2005 (UTC)

I added the note at the top...

I added the alert at the top because I'm not sure everyone here (especially the non admins) know the existing policy and tags. I thought it'd be helpful. --Woohookitty^{(cat scratches)} 09:16, 1 December 2005 (UTC)

why not use the publish methodology instead?

• Since this policy is simple to get around (have fun elsewhere to whatever level required, then you can vandalize whatever you want)
• since having a protected "published" page visible will never be vandalized (unless admins go bad)
• since *anyone* can wiki edit the "draft" page
• since moving draft edits to the published page will be less work for admins than the reverting of vandalism today

Why this proposal instead? -Jcbarr 13:19, 1 December 2005 (UTC)

Because it doesn't involve locked forks of articles. -Splash^talk 13:23, 1 December 2005 (UTC)

Yep. And we can build it into the policy...something along the lines of...anyone who is doing that (creating a user account, getting their edit count #s up and then vandalizing the article) can be reverted (and eventually blocked) for doing so. It's not that hard to figure out...we do it when figuring out afd votes all of the time. --Woohookitty^{(cat scratches)} 13:44, 1 December 2005 (UTC)

Semi-protection levels

I just had an idea after looking at Special:Blockip. Instead of having two or three protection levels set at a fixed ratio, how about there is just one level (semi-protected), which then can be set individually? It would be similar to the "Other" selection in Special:Blockip. This has two advantages:

1. Simpler to code, since you would be passing down a variable into the MediaWiki software, which would then be processed as

   IF $User.editcount < $Protect.threshold THEN rejectEdit()

   or something similar (that line would not work in MediaWiki, it's BASIC, not PHP); and
2. Inter-wiki simplification. The levels of semi-protection needed here at the English Wikipedia might not be the same as those needed on Spanish Wikipedia or the English Wiktionary; having a customizable level allows the different Wikis to adopt a semi-protection policy more suited to their individual needs. Meta would like this idea too.

Thoughts? Titoxd^{(?!? - did you read this?)} 21:08, 2 December 2005 (UTC)

I think leaving the threshold up to the individual admins is a very bad idea, we could get anywhere from 5 to 500 edits minimum. What guidelines or policies would we provide admins as to how many edits they should require, keeping in mind that setting the bar too high will result in unnecessary discrimination against possible legitimate editors? Whether its Spanish or English, I don't think it really matters, 25 edits along with a 5 day minimum account life (to counter vandalbots) for level 1 and 100 edits with a 1 month edit time for level 2 (for pages like GWB) will suit all intents and purposes, even as Wikipedia grows in scale. As for simpler to code, I don't see logically how that is simpler than just running a simple check against the user whether their account has existed for X amount of time and they have X edits. --kizzle 21:19, 2 December 2005 (UTC)

It would not be leaving it to admins' individual discretions, there would be limits set in the Semi-protection policy, similar to the ones at the Blocking policy, which would be set in every individual Wiki (we don't know how much vandalism is considered "acceptable" at other Wikipedias, and some of them have a minimum one-month block guideline for vandalism, so there will be other interpretations besides ours) The problem with coding a number in MediaWiki is that it is first, bad programming practice, and second, prevents other Wikipedias from setting their own policies. Titoxd^{(?!? - did you read this?)} 21:27, 2 December 2005 (UTC)

I guess the edit threshold and time limit could be built in as variable within the tool, and let policy (preferably as explicit as possible) be referenced by admins using semi-protection. That way, the time limit and threshold can be decided upon by a concensus from each wiki implementation.--kizzle 01:25, 3 December 2005 (UTC)

Yes, the policy should be implemented in each Wiki as clearly and explicitly possible, on that we do agree. Titoxd^{(?!? - did you read this?)} 01:27, 3 December 2005 (UTC)

Tito, before you claim an idea as your own, how 'bout taking a look at the main page ;). We've been doing so much talk about this on the discussion page (and frankly, it's huge), that we havn't really done anything to the proposal page. I guess great minds think alike -Mysekurity 06:23, 3 December 2005 (UTC)