Wikipedia talk:Semi-protection policy/Archive 6

Removal of semiprotection

An issue has arisen around removal of semiprotection, and I'm posting this to seek views on how it should be addressed. For the majority of articles, semiprotection requests are made in order to deal with a short-term vandalism problem. Once granted, and the vandalism dies down, it is usually safe to remove the semiprotection. There are some (high-profile) articles which are certain to be vandalised again as soon as the semiprotection is lifted however, the classic example being John F. Kerry — and these articles thus stay in a more or less permanent semiprotected state. The issue is around how we recognise these articles and ensure they stay semiprotected if necessary.

The specific article which has caused a problem is George W. Bush. This was semiprotected about a month ago due to heavy anon vandalism. Semiprotection worked - it gave editors the chance to focus on content instead of reverting vandalism. However, it was then unprotected without a discussion, and the vandalism started again almost immediately. It has subsequently been reprotected, unprotected, vandalised again, reprotected, and today unprotected again; it will no doubt be heavily vandalised again soon if it hasn't already, at which point I'll be listed it for semiprotection again ... and so on. Each unprotection has been carried out unilaterally without discussion.

I believe we need to find a way to avoid cycles like the one I describe above. The root of the problem is that some admins believe, rightly or wrongly, that the policy around unprotection decisions is that they are entirely at an admin's discretion. If this is the case, I'd suggest this is an oversight in the policy, rather than a deliberate intention.

It has been pointed out to me that any solution to this problem needs to avoid being overly bureaucratic: uncontroversial un-semiprotections should be able to be made easily and I agree with that.

Thoughts?

SP-KP 23:35, 14 April 2006 (UTC)

Unprotects are a common sense thing. See an article that can be unprotected? Unprotect it. Don't need to go filling in forms on WP:RFPP for the obvious and then have two admins involved rather than one. Just takes up twice the time for no improvement in outcome. It's not an oversight in policy, since we don't actually have much, if any, policy surrounding unprotection. This is because the current practise is the correct one and there is no need for us to document every click of the mouse. m:Instruction creep rarely achieves much. -Splash^talk 23:41, 14 April 2006 (UTC)

Unfortunately, the problem we have is that some admins are making unprotection decisions in respect of articles that pretty clearly shouldn't be unprotected (if one accepts the premise that semiprotection is the best way to deal with heavy vandalism). What we need is a light-touch mechanism to reduce the incidence of inappropriate decisions, without impacting on admins' day to day work. SP-KP 00:10, 15 April 2006 (UTC)

That's rather disingenuous of you. What you mean is that you don't like my decision to unprotect Tony Blair and think it an inappropriate decision to return a wiki article to free editing. There is no article, imo, that should not be regularly unprotected and so I do not accept the basis of what you say "the problem" is. On the contrary, lengthy protections without any indication that an article might be unprotected ever are inappropriate, not to mention anti-social, anti-wiki, and unnecessary. -Splash^talk 04:12, 15 April 2006 (UTC)

I think its rather disingenuous of you to suggest SP-KP is being disingenuous. Yes he doesn't agree with your action, but that is allowed. Frelke 06:31, 15 April 2006 (UTC)

As to widely-vandalized articles, I think we should write them on paper. Or, agree that this is a wiki and editing it is what people do. It's how it comes to exist. -Splash^talk 23:41, 14 April 2006 (UTC)

Can someone define disingenuous for me? Is this a good example? Frelke 06:34, 15 April 2006 (UTC)

I disagree. If we can see areas where an improvement to our processes could result in a better encyclopaedia, we should at least explore them. Lots of policies and processes on Wikipedia have developed over time; I see no reason why this should be different. SP-KP 00:10, 15 April 2006 (UTC)

The thing you propose to "develop over time" here is a reduction in the editability of the encyclopedia. You might want to try a different ship. Wiki is not paper. Wiki is editable. It is a m:Foundation issue that anyone can edit our articles. You're on the wrong ship if you want a policy to say that freely editable articles are bad. -Splash^talk 04:12, 15 April 2006 (UTC)

Now, I'm not going round in circles with you on this anymore. People can edit articles here, they always have done and always will be able to do so. If you want articles that aren't editable, fork Wikipedia and turn of the relevant variable in the PHP code. -Splash^talk 04:12, 15 April 2006 (UTC)

Honestly Splash, I think you are overreacting a bit. SP-KP's point is a valid one. I mean. George W. Bush is sometimes unprotected but honestly, it's usually symbolic as I don't see us completely unprotecting it until he leaves office. I think that his core point is valid. We should have some special template or something to show that an article is more or less permanently semi-protected. Otherwise, the template that is generally used on GWB and also Jew is inaccurate. It basically says that once the vandalism dies down, the page will be unprotected. Well that's simply not true on those 2 articles. --Woohookitty^{(cat scratches)} 04:40, 15 April 2006 (UTC)

It's just that the editors (few of them, actually) of GWB have some WP:OWNership difficulties. I have never understood why a single editor insists on re-protecting Jew, but it's hardly reasonable to rely on Cecropia's opinion on that article to determine policy. TfD has repeatedly deleted templates of the kind you describe, which goes some significant way to showing that people outside the talk-pages of a couple of articles don't want such a template. I actually do think that those who proceed as if they sky will fall if we allow free editing of articles are the ones overreacting. I mean, Tony Blair has been unprotected for hours now, and the wiki has withstood and will withstand. -Splash^talk 07:49, 15 April 2006 (UTC)

Splash, the deletion of those templates could be because people aren't happy with them as a solution, not that they don't think there is a problem. No-one is arguing for ongoing creeping reduction in Wikipedia's editability, or that we should abandon foundation issues. All this discussion is about is a proposal that we find a way of encouraging admins making unprotection decisions to do so with an appropriate degree of consideration as to the consequences. There are lots of ways in which we could do that. I've freely admitted above that your decision to unprotect Tony Blair was the thing that prompted me to raise this issue; but please accept in WP:AGF that there is nothing personal here, and that my motivation in raising this is to help us produce a better encyclopaedia. SP-KP 09:50, 15 April 2006 (UTC)

heh fight fight fight!!! let disingenuous = n and it is n + 1 of you to suggest User:XYZ is being disingenuous of USER ABC,

let USERS tanversre all the whole set of WIKI users then everyone is disingenuous of everyone recurisvely, problem solved.JUBALCAIN 08:27, 31 January 2007 (UTC)

Potential improvement

Here is the only reasonable improvement I can think of at the moment: Make semi-protecting a page similar to a user block, in the sense that the page is given a set number of days to be sprotected, as decided at the discretion of the admin who's protecting it (based on the rate and severity of the vandalism). It's not a major change but it could help the sprotection/unprotection cycle be more organized, and would allow for somewhat longer sprotected periods in cases where they are believed to be warranted.

The page would remain unlockable at any time by any admin. It should also be unprotected automatically after the allotted number of days has elapsed, because that would probably save time (of admins) in the long run.

(As for completely eliminating the protect/unprotect cycle, I see no possible way to do that while still allowing anons to make useful edits after/if the vandalism dies down. Unless, that is, you can invent something that would predict how many people are about to vandalize an article if it were to be unprotected.)

–Tifego^(t) _{10:32, 15 April 2006 (UTC)}

That's a good suggestion; implementing it would require some developer input, presumably. My only reservation is your comment that the page should "remain unlockable at any time by any admin"; whilst I'd be OK with this being possible, we should have some safeguards in place to prevent individual overzealous admins abusing it e.g. a requirement that an admin who wants to unprotect the page early should take things to Requests for Unprotection. That's a minor increase in effort for early unprotections, but traded against a big decrease in unprotection effort for those that run their natural course.

On the question of anon editors who might be able to make useful edits to semiprotected pages, two thoughts: (i) most of these pages already have large numbers of non-vandal contributors, so the risk of losing important content by preventing anon edits is low and (ii) the talk pages aren't protected, of course, if the wording of the semiprotected message is amended to state this, then anons can place their proposed changes there to be picked up by non-anons. SP-KP 14:36, 15 April 2006 (UTC)

On this topic, one which I do think we might revisit, can I invite interested parties to review User talk:Splash#March admin statistics where some very interesting ideas were mooted in the course of a general brainstormy type thing. For some background, the main reason this was rejected at the time that SEMI was being developed into policy was to avoid asking the devs for anything more than what we were sure we needed: one Christmas present at a time! Anyway, that talk page link is worth reading for this discussion. -Splash^talk 22:33, 15 April 2006 (UTC)

The only way I will go away is if there is a time machine and history changes. Then, I won't have to put in historical FACT in place of revisionist, ideal history. Why are you people trying to pretty it up? If someone wants to know about someone, they should be allowed to know it all. I will keep putting it on here until you accept the truth. The truth hurts, but it is what happened. (Ghandi article) My edits that they keep deleting:

An English-educated lawyer, Gandhi first employed his ideas of peaceful civil disobedience in the Indian community's struggle for civil rights in South Africa. In spite of popular belief of Ghandi as a peace loving man and a champion for human rights, he was in fact what one may consider a racist against native South Africans. The crux of Ghandi's civil rights strategy was to uplift the Indian in South Africa by using white hatred of the South African natives to play in favor of Indians - Indians and upper class Indians only! What Ghandi did is not unlike what many immigrants have done in the United States against African-Americans;playing the race card to benefit their groups from the ruling class. Ghandi constantly put down and used the native plight in South Africa in an attempt to lure the white government into letting upper class Indians enjoy more feedoms in South Africa. Ghandi would support many of the white governments tactics against the natives in an effort to 'ease' the discrimination toward Indians. This "join in with white racism to get ahead" tactic had been repeated throughout history and in the United States from groups such as Italians, Irish, Asians, Indians(from India), Mexicans and now amongst hispanics of clear African origins (mixed of not). This is a common tactic for immigrant groups to get ahead in host countries and has been used against African-Americans every chance a group can get in order to have a group to be better than, or to simply not 'be like them' in an effort to benefit from a new nation. Particularly a nation with known discriminatory practices.[1]

Taking this forward

So, to formalise the proposal, then:

• The sprot template should be amended to include the words "on <date> for <N> days"
• When semiprotecting a page, an admin should decide how many days is appropriate and state this in the tag parameters
• All pages which have reached the end of their semiprotection period should automatically have the sprot template removed
• Any admin wishing to remove an sprot tag before the expiry period is up should take the subject to the talk page and establish whether there is consensus for early removal of the tag or not, and should act according to that consensus (N.B. this procedural change to come into play when the automatic removal is implemented)

Does this have everyone's support? If so, can I suggest we make the changes to the page, and contact the developers to set wheels in motion for their bit of the change?

SP-KP 19:45, 18 April 2006 (UTC)

GWB should be unprotected ASAP, as should Tony Blair. --Sunfazer 20:02, 18 April 2006 (UTC)

If you'll both take a look through the archives, you'll see we discussed both of these propositions. The main issue with telling when an article will be unprotected is that they give vandals that much more chance to just wait for the day. An article will become unprotected once it is naturally ready to become unprotected. We have no way of automatically removing a template (aside from a bot, but this is more of a judgement call, and why we don't do this with ProD). Our current requests for unprotection are handled at WP:PP, although they are usually discussed on the talk page first. In regards to the GWB and Tony Blair pages, we have discussed this as well—there is too much of a temptation to vandalize both pages, and thus they shall stay pretty much constantly protected until both are out of office, or some significant event occurs that new editors are needed or encouraged. Happy editing, Mysekurity ^[m!] 20:18, 18 April 2006 (UTC)

My argument for unprotection as regards GWB and Tony Blair, is, there are no significant events in the news concerning them. If there was an event as such it could be protected easily. --Sunfazer 21:32, 19 April 2006 (UTC)

A BBC News website search gives 4 stories covering Blair today, the same number yesterday, and about a dozen from last week. Perhaps not "significant" stories but enough to keep him in the forefront of vandals' minds. The vandalism figures following each unprotect speak for themselves - go take a look. SP-KP 21:42, 19 April 2006 (UTC)

A 'protection expiry date', so to speak, should NEVER be set. This way it is up to the admin to use their judgement. --Sunfazer 10:07, 20 April 2006 (UTC)

Continuing discussion

This discussion is in danger of fizzling out with no actrion being taken. Clearly there is a consensus for something to be done in respect of high-profile vandal-attracting articles, we just don't seem to be able to agree on what. I want to see if we can rescue the proposal above before proposing a change of tack. To deal with Mysekurity's first concern, this seems to be about visibility of expiry dates - so, can we display the expiry dates away from the articles themselves, so that they are not obvious to the vandals? Any suggestions as to where? Where would they be of most use to editors / admins? The second issue, a template-removing bot ... I think we need to understand how big an issue that is. Can someone expand the reasoning on this one please? SP-KP 10:26, 23 April 2006 (UTC) UNLOCK PS371.225.125.214 21:53, 11 May 2006 (UTC)

Timed article semi protection

I still think that the wikipedia needs to implement: Wikipedia:Timed article change stabilisation mechanism.

It solves quite a lot of the problems that we talk about above; it tackles random vandalisation at root by making it pointless.WolfKeeper 16:20, 27 April 2006 (UTC)

Before your proposal, the first logical step would be a trial implementation of the requirement of user registration, or perhaps universal semi protection. Ben 04:40, 19 December 2006 (UTC)

I'm not a new user, but still can't edit; why?

The 2003 invasion of Iraq page has recently been locked, with a message that new users can't edit. I'm not a new user, or anything like that, and yet I don't see an edit tab there anymore. Any ideas on why this is true? Thanks! --Hermitage 09:03, 30 April 2006 (UTC)

It looks like an admin accidentally fully-protected that article instead of semi-protecting it. –Tifego^(t) _{09:30, 30 April 2006 (UTC)}

Actually it seems to be the other way around; it's supposed to be fully protected, but the wrong notice was put on it. –Tifego^(t) _{09:34, 30 April 2006 (UTC)}

Open up semi-protection to all editors

As is noted above and elsewhere, semi-protection works best when it works fast, and it doesn't need to stay around long. Before I realised that there was a whole procedure to follow to semi-protect a page, I tried just slapping up the {{sprotected}} flag. Guess what-- it worked (sorta). The template alone will put a stop to a lot of persistent vandals simply because the page looks like it's protected. I stopped using it once I realised there was more to it, but I think it would be a Good Thing if the template provided protection all by itself-- dump the rest of the procedure. There are a lot of us non-admins out here fighting vandalism all the time, and remember-- to be effective, it has to be applied quickly (like within minutes).

I'm sure there will be some admins who imagine that it would be misused by non-admins. Think about it though-- who is going to abuse it, and what good is it going to do them? -- Mwanner | Talk 20:20, 3 May 2006 (UTC)

Y'know this sounds crazy... but it might just work. After all, it's not like it could be used in edit wars. --JiFish(^Talk/_Contrib) 20:27, 3 May 2006 (UTC)

Well for a start, I'd say that that template is never supposed to be used as a bluff! Just incase you didn't know. Anyway, all editors? I don't know... If you look at WP:RfPP you'll see that "protection should only be used as a last resort". I feel that if users were given the ability to sprotect on their own, they would not follow protection policy (either due to ignorance or malice) and that'd not be ideal. We don't want a billion sprotected pages. You can see from WP:RfPP that there are plenty of users who request semiprotection for articles that have been vandalised 5 times, or whatever. I don't think this would work. --Darth Deskana (talk page) 20:30, 3 May 2006 (UTC)

See, I knew that wouldn't take long. Consider, though, that it's just as easy to revert an {{sprotected}} tag as it is to revert vandalism (and meanwhile, some vandalism would be prevented). Users who overuse the tag could be warned, and users who place an sprotect and don't remove it quickly could also be warned. My guess is that even with some misuse, it would result in a substantial net reduction in problems. But your mileage may vary. -- Mwanner | Talk 20:42, 3 May 2006 (UTC)

I can see the logic behind your proposal... but my instincts tell me that non-admins being able to semiprotect pages is a bad idea. --Darth Deskana (talk page) 20:44, 3 May 2006 (UTC)

We need to get Wikipedia:Timed_article_change_stabilisation_mechanism implemented. Most of these issues go away; people don't vandalise nearly so much if their changes don't appear.WolfKeeper 16:47, 4 May 2006 (UTC)

people might not write at all if their changes don't appear.Ben 04:45, 19 December 2006 (UTC)

130 articles seems like a lot

I remember when this was first implemented, we had like 4 articles semi-protected, and agonised over each one. Now we have 130 articles semi-protected. Is this not going a little far? I notice that Frédéric Chopin is semi-protected, an unlikely target of continued vandalism I would have thought. Is any kind of review of semi-protection in general planned? Stevage 14:52, 15 May 2006 (UTC)

Protect against New Users?

Hello there. I am actually an admin of an independant wiki. I am wondering how you guys were able to put a protection against new members? It actually sounds like a grand idea on my wiki, as we have a hack installed so only members can contribute. This way those vandals who just make a username to vandalize can't on those important pages that we have. Could you show me or lead me to the hack which lets you guys do this? ~U

Extending sprotection

I've tweaked the page a little to reflect Jimbo's proposal that sprotection be extended on a limited basis to allow pre-emptive continuous protection of some pages, and that such pages need not be tagged but should still be categorized, citing his post as a reference. SlimVirgin ^(talk) 11:24, 21 May 2006 (UTC)

In that proposal, Jimbo says:

...the template at the time is misleading and scary and distracting to readers. I propose that we eliminate the requirement that semi-protected articles have to announce themselves as such to the general public. They can be categorized as necessary, of course, so that editors who take an interest in making sure things are not excessively semi-protected can do so, but there seems to me to be little benefit in announcing it to the entire world in such a confusing fashion.

I agree with this 100% and hope we can now free all those poor sprotected articles from that hideous box. We can do this by either stating that Template:sprotected should be put on talk-pages only (as on the German wikipedia) or to simply edit the template to make it like Template:Featured article; no box, just adding it to the sprotected category and with a lock-icon (or somesuch) in the top-right corner. Shanes 11:42, 21 May 2006 (UTC)

I adjusted the page to say that pre-emptively protected pages shouldn't be tagged, because I was trying to be conservative with the changes, but I agree that it'd be good to get these tags off all the sprotected pages. Adding them to talk instead is a good idea. Shanes, what did you understand Jimbo to mean exactly on this point? Did he mean the sprotected tag should be removed from all pages? SlimVirgin ^(talk) 11:48, 21 May 2006 (UTC)

Maybe we should just ask him, but what he is saying in the above quote is that he finds the box distracting and that we can do without it. I suspect that it's the tags on high profile articles like George W. Bush that made him care enough to make that post, but if it's distracting on those articles, it's distracting everywhere. He does use the word "carefully" in the first sentence, though, so being conservative with the changes to the policy page sounds reasonable to me. And it was just a suggestion of his, so I'm fine with discussing it for a few days first. But I must say I hate that box so intensely that I've never been more tempted to be bolder than ever and just go on a tag-zapping spree. But I must remain calm. Must remain calm. Shanes 12:09, 21 May 2006 (UTC)

LOL!! Well, how about we leave it as it is for a bit to gather opinion? :-) I can't see any objection to having the tag on the talk page and it does sound like a good compromise. SlimVirgin ^(talk) 12:18, 21 May 2006 (UTC)

I like the idea :o) --HappyCamper 11:51, 23 May 2006 (UTC)

I don't like this idea at all. Hiding the fact that this has been done from readers (who are the most important people let's not forget, Wikipedia is not the private club some people treat it as) smacks of a cover-up. I don't remember seeing this mentioned on the village pump, but I have just spotted that it is about to be announced in the Signpost as a fait accompli. I also really don't like to see an all too obedient rush to change policy in response to a suggestion from Jimmy Wales when there is only minimal evidence of backing for it. Wikipedia needs to grow out of being his personal fiefdom. CalJW 22:26, 29 May 2006 (UTC)

It is exactly because of the readers that the sprotected should not be visible on the article pages. The editing status of an article is not relevant to the readers - it is only relevant to the (much less numerous) editors. As I have said elsewhere, there should be a mechanism that explains the sprotected status to non-eligible editors after they click on the "edit" link. - Hayne 16:34, 2 June 2006 (UTC)

I fully agree with the removal of that {{sprotected}} box from semi-protected articles like George W. Bush. This box is distracting. We simply don't need it. And we do not need a replacement for that box either. There is enough explanation for the readers if they click on the "View source" tab. We don't even need a category for semi-protected pages—this is not encyclopaedic. If you like, put {{sprotected}} on the respective talk pages. --Ligulem 12:17, 2 June 2006 (UTC)

Clicking on the "view source" tab is not an obvious thing to do if you want to edit the page but aren't familiar with MediaWiki. I think we do need some form of notice, though I think the current version is unnecessarily cumbersome. Haukur 12:34, 2 June 2006 (UTC)

Agreed. It should be moved to the talk page. Readers don't care about editing; and there are far more people reading Wikipedia than editing. Also, the mere mention of "vandalism" will put into question everything in the article, even though the article is protected from vandalism. Even the term "vandalism" is only for editors; readers are not likely to understand what that term is referring to, as many of them don't even realize Wikipedia is editable. — BRIAN0918 • 2006-06-04 16:24

Yes, the term 'vandalism' is unnecessary and should be removed from the template. But the template itself should not be removed from semi-protected pages. Haukur 10:41, 6 June 2006 (UTC)

Hi, I have comments on how I perceive the psychological approach to wikipedia. To a new user, wikipedia is perceived as free, universal, and honest. For a new user, realizing he can't edit a page only after he maybe made reasonably accurate preparations and thinking for his first edit, and clicked the edit page, whether he has registered or not, may be perceived as dishonest, and have a bad effect on his decision of joining the project. I, for one, have never cared much about registering. I made several edits before registering and, even now, occasionally make edits without logging it. It's not that I have anything to hide, I simply and honestly forget to log in before editing. For the most part, this I think is due to my perception that Wikipedia is simply an universal project, and there is no much point in keeping scores of the number of edits. I know many users disagree, but I also believe this point of view is shared by many. I can understand the rationale behind semi-protection, but I think it should also be remembered that the vast majority of editors are honest and in good faith, not vandals. This is a basic assumption that the very existence of wikipedia as it is now has, perhaps surprisingly, confirmed. I do not think that the logic should pass that wikipedia has to be protected more than it has to be improved. Even good articles should be seen, I think, as potentially being able to spiral towards being better through continual edit. I do think that the current tag is confusing, though. In my opinion the tag should clearly state who can edit the article rather than who can't. For example, clearly and honestly stating "this article has been semi protected because it has been targeted with vandalism (or "because it is a potentially sensible biography of a living person"). Only regstered users whose account is older than 4 days can edit it". Or something like that. Massimamanno 11:47, 16 August 2006 (UTC)

How much vandalism is enough?

What's the general rule of thumb for how many different IPs or accounts need to attack an article in a day before semi-protection becomes necessary? There must be one, but I haven't seen it. Right now I really don't have much idea whether an article should be semi-protected or not. --Sam Blanning^(talk) 18:44, 23 May 2006 (UTC)

The question I always ask myself when evaluating a SP request is whether or not it is at a level that users on that page are able to handle (informed by the way the George W. Bush article used to be). If it's only annoying, but manageable, that doesn't really require administrative intervention -- if we wanted to get rid of the annoyance factor we might as well just ban anon users entirely. But if it's at a level regular editors just can't handle, or that isn't reasonable to ask them to handle, then I'll semiprotect. It has to be significant, concerted, ongoing vandalism at that moment. Averages over time don't make the cut. I'd hate to try to put a number on it, but I generally don't consider anything less than 15 reversions in a day as a minimum. Sometimes I will be less strict with an article on a living person because of the potential for libel, but that depends on the types of edits being made. · Katefan0 ^(scribble) 19:38, 23 May 2006 (UTC)

Permanent semi-protection

I'm all for this on articles that get a high degree of vandalism. However, can we make the indication of semi-protection less prominent on the article? An unobtrusive indication would be great. - Ta bu shi da yu 04:27, 30 May 2006 (UTC)

I like the idea of permanent semi-protection, but I'm wondering why the indication should be less prominent. What's the current standard indication? Perhaps we should use what I put on the top of Jew. — Rickyrab | Talk 19:58, 13 June 2006 (UTC)

It doesn't look good on the page to have a huge warning about vandalism, and with living bios, it could imply that the person is problematic in some way. That's why it was better to have the little keys sign than the words. SlimVirgin ^(talk) 09:10, 7 July 2006 (UTC)

semi-protection and wikipedia stability

I think Wikipedia has a real problem with stability. I know this is going against the party line, but it seems to me semi-protection offers a way to ameliorate this problem.

Right now we are voting to delete a major template because we cannot prevent users from adding to it at random and we are not interested in policing it all the time. Also any time an article loses its primary editors it starts to decay.

I'm going to make a controversial suggestion. I suggest we use semi-protection much more liberally on pages that we want to stabilize. This would include major templates that are used by a lot of pages and perhaps featured articles.

I am new and perhaps that is why I have not bought into the party line. But I thought I would throw this out and see what people think. Ideogram 10:08, 4 June 2006 (UTC)

For some discussion of this issue, see Wikipedia talk:Stable versions#Research / Relevant stats. Ideogram 10:48, 4 June 2006 (UTC)

"new users"

what is a "new user"? At present, less than 10 edits seem to be required before an account can edit a semi-protected article. That is far too little, determined trolls often create throw-away accounts and do a couple of minor edits before having at their actual target. A good faith newbie can be expected to do at least 100 edits to uncontroversial articles to get the feel of Wikipedia before engaging in editing controversial or disputed aritcles. Hell, when I was a newbie, I stayed away from controversies for months. dab (ᛏ) 09:53, 16 June 2006 (UTC)

The only limit is 4 days — the same as that to receive the "move" privilege. There is no edit count requirement. It was discussed during the formulation of the policy, but the devs indicated that it was undesirable to have to count edits on every page save. -Splash - tk 13:34, 16 June 2006 (UTC)

Splash, do you know if there's a way to make it contingent on days edited? I've been fighting off some newly created accounts/anon socks that have been used to stalk me, and would love a more beefed up sprotection policy. IronDuke 05:02, 20 June 2006 (UTC)

I am not a dev, but that's an interesting possibility. I rather suspect that the database does not explicitly store the time of first edit though (e.g. there is no Special:Log/Neweditors, only a Special:Log/newusers. But it has its flaws; one edit a day would clear that kind of hurdle, unless we're looking at X edits per day on average for N days, where e.g. X=10 and N=4 or something. I don't think I like the sound of that: protection should not become a hoop to jump through to 'earn' editing privileges. We should approach it from the perspective of how to limit vandalism whilst making the life of good-faith newbies as easy, transparent and hurdle-free as possible. When getting heat from bad guys, or spending too much time vandalwhacking it's easy to forget that most people aren't bad and just want to help. -Splash - tk 05:28, 20 June 2006 (UTC)

Points well taken. But it seems to me if the system can exclude users under 96 hours, then it could exclude users based on any time period, right? And perhaps I am just being curmudgeonly due to the torture that's being inlicted me just now, but I can't help feeling like we err so often on the side of not hurting a newbie's exquisitely sensitive feelings that we allow smart vandals and trolls to game the system. This would be fine if it didn't lead to decent editors becoming burned out on the project. No new editor "needs" to edit an article on Topic X, do they? Surely, they can be patient, if they're good faith editors? IronDuke 17:08, 20 June 2006 (UTC)

I'm not sure you'd have to count edits on every page save as there is already a record of edits. I'm a dev myself so i'll take a look at the code and see if it's possible. I personally think there should be, say, 50 edits minimum and/or 14 days minimum - it's not that hard to make scripts that generate registrations, so 4 days wouldn't reach the height of impatience and discourage potential vandals nearly as much as 14 would. And why would a new member need to edit controversial articles anyway? Possibly we even have certification where admins take a quick look over a user's edits and give them the thumbs up so that they can edit articles marked with semi-protection, that way we eliminate the bulk of the problem, albeit waste a lot of time on the admins' part. Recruit more admins? :) --Draicone 00:08, 30 June 2006 (UTC)

When SHOULD an article be semiprotected?

The guidelines talk a lot about when a page should not be semiprotected, but they say nothing about when a page should be semiprotected. Say there's an article for which the majority of its edits are vandalism by random IP addresses and reverts to this vandalism. Semiprotection would seem to be in order, except that the guidelines seem to say that semiprotection is only intended to be in place for a limited time; and in this case, the moment semiprotection is removed, the back-and-forth vandalism-and-reversion will start right up again. What's the proper solution in this case? Should WP editors just put up with the fact that vandalism from anons will continue, and they have to remain diligent? Is it ever tolerable to let semiprotection remain in place indefinitely? Does the number of anon vandalisms to the article per day or per week make any difference? It kind of seems like the guidelines discourage semiprotection from being used at ALL, except in cases where a lot of vandalism happens from a wide range of anons over a short time period, and then they all go away and the problem disappears - and I don't see when that's ever the case. - Brian Kendig 13:25, 23 June 2006 (UTC)

The policy says quite clearly: "A page can be temporarily semi-protected by an administrator in response to vandalism, or to stop banned users with dynamic IPs from editing pages.", and that "Jimbo has suggested that semi-protection may be used in the cases of "minor bio[graphie]s of slightly well known but controversial individuals" which are not widely watchlisted, if they "are subject to POV pushing, trolling [or] vandalism".". I don't see much room for misunderstanding in that. Your question is not as hypothetical as you present it, though. In my opinion, no, it is never acceptable to have permanent semi protection, and certainly not on anything other than the very very highes profile articles. The one you do some revering on is certainly not among those. If you've never seen someone repeatedly vandalise an article then get bored when their plaything is taken away, then well....you must not have done much RC patrol in the last year or so - it happens all the time. -Splash - tk 15:34, 23 June 2006 (UTC)

Splash, I don't know if I'm missing something here, but that reply doesn't seem to answer Brian's main point (in upper case in the title, see also his first sentence). I don't see Brian's question as "what does policy currently say" but as a question for discussion about what policy ought to say. He's saying that, at the moment, policy states that semiprotection can be instigated under various circumstances, i.e. it doesn't have to be; what he's asking is whether there are any circumstances - i.e. a subset of the first set, where semiprotection should be instigated. Does that help you to see the distinction? BTW, reading the above, I'm not that far away from you in my views on the use of semiprotection - we disagree only in that I think that the case for near-permanent semiprotection of the very highest profile articles has been convincingly made now; so my response to Brian is yes, let's have a very restricted set of circumstances where near-permanent semiprotection is the outcome, not just a possible outcome if requested. The existing mecanisms seem to work well for everything else. SP-KP 16:34, 23 June 2006 (UTC)

See Brian's first sentence. I answered it, I think, and that sentence is more or less exactly the question you say it isn't. I don't think it makes good sense to allow only a single outcome of permanant protection for a certain class of articles. It's a noose to be hanged with. The existing mechanisms cover this case anyway, George W. Bush has a more or less permanent semiprotect on it by supposedly-common consent among admins. No need to provide bars for ourselves to trip over when we're doing alright without them. -Splash - tk 22:32, 23 June 2006 (UTC)

Here is an idea. I propose that when the History of a given article has, on the average, say more than 90% vandalism edits and reversions compared to other edits, and a minimum of 10 vandal edits per day (excluding anything to do with real content dispute which is never vandalism), it should be a candidate for semi-protection. If an admin sees that the vandalism rate continues unabated, say over 24 more hours after the nomination, then the admin inserts the sprotect for say 1 week, after which the cycle resumes, with an approporiate notice in the Talk page. If the subsequent vandalism rate continues and justifies another sprotect, the minimum period is quadrupled to 4 weeks, and so on. The current vandalism reversion effort takes a toll on people's workload, and this process must be streamlined for future scalability, IMO. Thanks, Crum375 15:00, 23 June 2006 (UTC)

Hard-and-fast rules never work very well because of the fact that few situations actually meet the exact requirements of them. If you want a page protected, list it at WP:RFPP. -Splash - tk 15:34, 23 June 2006 (UTC)

• Definitely there needs to be some discretion, that's why the admin would still be in the loop. But as of now there is no guideline anywhere that I can find as to what is considered 'annoying but normal' level of vandalism vs. 'clearly excessive'. If there were such a guideline, it would be a clear starting point for someone to initiate the process. As of now, we are totally in the dark. Crum375 18:27, 23 June 2006 (UTC)

• That's exactly my point. Codifying in language what constitutes a level for semiprotection won't work well because of the need to apply common sense when faced with a request. Until a certain pro-protection editor removed the words, it used to say "serious" before the word vandalism e.g.[2] and that is really pretty explicit. I will add that back in, I think. The "Adminstrators note" in that link was also obviously useful, but again got edited out since someone wanted to protect something that didn't qualify with that left in. -Splash - tk 22:32, 23 June 2006 (UTC)

To rephrase my question a bit: There are lots of articles out there which are vandalized by random anonymous accounts more often than they're improved. The policy says that they can be semiprotected against the vandalism, but this won't help matters because when the semiprotection is removed the vandalism will start right back up again, and the policy says that semiprotection should only be for a limited time. When should this limited time begin and end? What's the guideline for deciding when an article deserves, and then no longer deserves, semiprotection? I feel that the article should be more clear about about the proper circumstances for semiprotection - not just "Jimbo has suggested that...", but something more along the lines of: "Semiprotection should only be used when an article has suddenly become high-profile, such as by being linked from Slashdot, and is being vandalized a lot right now but the vandalism is expected to die down in a few days. Semiprotection should then be removed three days later." Are there any other circumstances under which you can see semiprotection being called for? (Jimbo's suggestion itself goes against the rest of the guidelines, because it seems to imply a more permanent state of semiprotection.) - Brian Kendig 20:57, 23 June 2006 (UTC)

Yes, well, that particular fudge is because Jimbo did suggest that in a not-very-suggestion-like manner. I don't think there is too much trouble in expecting admins to lift the protection to see if the serious, current vandalism problem has passed after a few hours or days, leave it unprotected if things look ok, and reinstate it if not. Your question effectively stems from the highly editorial changes I linked to above. An article should be protected if there is a serious, current vandalism problem that cannot be dealt with by reverting and blocking the vandals. This seems like a clear enough statement to me. -Splash - tk 22:32, 23 June 2006 (UTC)

I agree, that's clear - except, please define "serious" in this context; how many vandalisms over what period of time make it a "serious" problem? Once that's defined, I'd like to see the policy clearly state what you said: An article should be protected if there is a serious, current vandalism problem that cannot be dealt with by reverting and blocking the vandals. (Right now the policy says nothing about the "cannot be dealt with" part; it seems to paint semiprotection as more of a convenience thing.) And then, the article should state: Lift the protection to see if the serious, current vandalism problem has passed after a few hours or days, leave it unprotected if things look ok, and reinstate it if not. Right now the policy says nothing about when semiprotection should be lifted. - Brian Kendig 04:44, 24 June 2006 (UTC)

Well, like I've said, I personally don't think it useful to codify the meaning of "serious"; I think there are enough other bits there to make the meaning of this in a Wikpedic sense clear enough, but to allow enough room for the key element of judgement. It's that element of judgement and the existence of edge cases and exceptionals that make codification more of a noose to be hung with than a useful move. If an admin is unsure, they should either make a request at RFPP and learn from the feedback, or watch that page and learn what the prevailing feeling(s) is/are about their kind of article/request/etc. -Splash - tk 12:50, 26 June 2006 (UTC)

By the way, Splash - the article you linked is actually not a good example of a need for semiprotection, as that article is only vandalized a few times a week at most, and isn't a high-profile article to begin with. (Though, after looking over Category:Semi-protected, it appears that the policy is really quite subjective - for example, Big Beautiful Woman gets fewer edits and fewer vandalisms than Virtual Magic Kingdom, but the former is semiprotected while the latter is not. And Ronnie Coleman is semiprotected "to prevent sockpuppets of banned users from editing it;" there even exists a template {{sprotect-banneduser}} for this purpose! This only reaffirms my belief that better guidelines are necessary for this.) - Brian Kendig 21:38, 23 June 2006 (UTC)

CAT:SEMI is a good example of admin laziness. No more than about 30-50 of those will actually need protection any longer. Bored schoolkids don't usually have the patience to wait more than about 24 hours. If you see an article that shouldn't be protected, then unprotect it. Banned users may not edit the site period and anyone may revert their any edit without restriction; this is vandalism in all but name, and usually in fact as well. -Splash - tk 22:32, 23 June 2006 (UTC)

Many of those pages are listed because (1) the policy gives no guidelines about how "serious" the vandalism has to be before semi-protection is justified and (2) the policy gives no guidelines about when semi-protection should be lifted. It appears that several of these pages are listed because an article got a rash of vandalism over a couple of days, and the editors don't want that to happen again. I believe this isn't the intent of semiprotection, and I'd like to see the policy more clearly discourage this sort of use. - Brian Kendig 04:49, 24 June 2006 (UTC)

I think your changes to the policy are good and accurately reflect both the intent of the policy and the manner in which it is routinely implemented at present. Well done! -Splash - tk 12:50, 26 June 2006 (UTC)

Thank you very much! I appreciate that. - Brian Kendig 13:39, 27 June 2006 (UTC)