Wikipedia talk:BLP/3RR

''This is a talk page. The discussions listed here began at Wikipedia Talk:BLP. This can be considered a subpage of that talk page, separated off to reduce page size.''

WP:3RR current best practice
The three-revert rule does not apply to such removals, though editors are advised to seek help from an administrator or at the BLP noticeboard if they find themselves violating 3RR, rather than dealing with the situation alone. Content may be re-inserted only if it conforms to this policy.

Appears to be wishful thinking. The 3RR noticeboard does take such situations under consideration, and seems to be in favor of blocking when they occur.

Often in situations where things are marginally unclear, people also think that this wording means taht it's ok to edit-war (which it isn't), such edits have certainly lead to page protection (on the wrong version, no less) in at least one case which I recently checked.

Either way, I do not think it is a current best practice to overlook three revert rule violations on biographies of living persons. The practice is neither current (3RR board considers anyway), nor best (people just edit war and don't get any closer to resolving anything). --Kim Bruning (talk) 15:57, 6 May 2008 (UTC)
 * The editor who reverts other people's work in a 3RR case sometimes claims to be reverting BLP violations. This claim needs to be investigated along with other recent conduct of that editor on the article. In an article about person X, if the claim that is being reverted is that 'X eats children' then chances are there will be no block issued for doing that revert. But usually the issues are more subtle. There was at least one 3RR case on the article of a leading politician alleging BLP in a case where the stated info just made the candidate look bad, though it was well-sourced and did not appear defamatory.  EdJohnston (talk) 16:17, 6 May 2008 (UTC)


 * Agreed - it happens all the time. Look at Talk:Ashley Alexandra Dupré where a battle ensued between an administrator and several other editors over whether someone's height at 5'5" versus 5'6" was a BLP violation.  The administrator wins with his admin tools by blocking the contending party for "violating BLP" in what is merely a content dispute to which the administrator is just as culpable of a party.  Or Talk:Nigel McGuinness where the inclusion of a guy's real name (publicly available on public records) was a BLP violation, and an admin blocked somebody for a week for including it with absolutely no repercussions whatsoever.  It seems BLP can be used as a catch-all to engage in edit warring, escaping the intent of the 3RR rule... I think 3RR should be enforced unless the BLP violation is blatant. Reswobslc (talk) 16:26, 6 May 2008 (UTC)


 * Ah yes, you again. WP:BLP is crystal clear - if you have to dig into public records to find something out about someone, we don't publish it. Do not use, for example, public records... or trial transcripts and other court records or public documents, unless a reliable secondary source has already cited them. End of story. FCYTravis (talk) 16:53, 6 May 2008 (UTC)


 * Speak of the devil! I don't think Ashley Alexandra Dupre's height came from a public record.  Nevertheless, it is no surprise that you, later in this section, are arguing that any revert or action taken under the name of "BLP" should stand if it's done in good faith.  Your edit history is peppered with edit summaries of "revert, BLP", and you are the biggest abuser of this! Reswobslc (talk) 17:06, 6 May 2008 (UTC)


 * It's no surprise that someone who has repeatedly edit-warred to include privacy-violating, poorly sourced and unsourced material in BLPs, of which removal has been requested via OTRS, would appear on this thread either. FCYTravis (talk) 17:08, 6 May 2008 (UTC)


 * This is a situation where we should formalize the distinction between simple BLP violations (unsourced negative content) and BLP-penumbra issues. 3RR exemption makes sense for the first, not for the second. JoshuaZ (talk) 16:31, 6 May 2008 (UTC)
 * 3RR says that "multiple reverts may not constitute a breach of this policy" if they are "reverts to remove clearly libelous material". My reading of this is that BLP is a valid defence to use, but the clarity of the libel is a high bar to clear, and the 3RR board is the right place to determine that.  Regardless of whether the 3RR technically applies to a situation, edit wars are rarely an effective means of dispute resolution. Bovlb (talk) 16:50, 6 May 2008 (UTC)
 * If edit warring does not solve the situation, and things always get brought to the 3rr noticeboard (and this is the typical procedure you mention), then you have just verified both prongs of my rationale. :-) --Kim Bruning (talk) 16:55, 6 May 2008 (UTC)
 * Well, yes. I was trying to interpret the policy as written, not advocating against change.  It's a thin reed to stand on.  Oh, and consensus (or even good faith debate) on the article's talk page or WP:BLPN would go a long way towards demonstrating that the libel is clear. Bovlb (talk) 16:59, 6 May 2008 (UTC)
 * It only said that after Kim Bruning edited it. The policy has long said "or unsourced or poorly sourced controversial material about living persons" - at least back to 2007, and probably longer. The BLP 3RR exemption for controversial material has been in existence for years and is an important tool in the fight to clean up BLPs. I strenuously object to any weakening of it, even in good faith. FCYTravis (talk) 17:04, 6 May 2008 (UTC)
 * And patent trolls strenuously object to any weakening of the current patent system, "for the good of inventors". Yeah, whatever.  If you didn't engage in so many edit wars where you toot the BLP horn to fend off scrutiny, then you wouldn't be here fighting for the rights of these oh-so-poor unnamed people who are being defamed by having their height misrepresented on Wikipedia by up to 1 inch. Reswobslc (talk) 17:11, 6 May 2008 (UTC)
 * (edit conflict)
 * I object to weakening policy too. :-) If you've used the 3RR exemption successfully several times, could you please provide links/diffs to a couple of those situations, like I asked below? Thank you! --Kim Bruning (talk) 17:14, 6 May 2008 (UTC)
 * I used it in the current case under contention, obviously. Usually, I just revert and protect, but I was afraid that would spark a wheel war in this case, and I'd prefer to avoid those. Those are bad, we can all agree. I've probably run past 3RR in other cases before, but I can't recall them offhand. I'll have to dig. FCYTravis (talk) 17:17, 6 May 2008 (UTC)
 * Even in the current case under contention, there was a VERY recent edit war involving FCYTravis as to whether or not Ashley Alexandra Dupré belonged in Category:Political sex scandals versus Category:Sex scandal figures. That edit war was merely over the semantics of a category name, but FCYTravis played the BLP card to avoid accountability for 3RR.  See the history, that 3RR was within the last 50 edits.  There is merit to his category argument, but it has nothing to do with abusing BLP to "force" his position on the community. Reswobslc (talk) 17:23, 6 May 2008 (UTC)
 * "Played the BLP card?" Reswobslc, do you think this is some kind of silly fucking game? This is not your Interwebs playground. These are real people's lives we're talking about, on one of the 10 largest Web sites on the planet Earth. We have a moral, ethical, legal and public opinion obligation to get BLPs right - to be accurate, fair, sensitive, even-handed, non-sensational, objective and respectful of those we write about. That's what I'm here to do. If you're here to make scandals and talk about "omg BLP card," you need to go find some other site to play on, because this one really isn't for you. FCYTravis (talk) 17:31, 6 May 2008 (UTC)
 * Fair enough. :-) I would like 3 situations with diffs, showing that the 3RR exemption is actually achieving those aims (without a need to seek recourse to further DR steps... which is a basic indication of failure). If it's as important as you claim and works as you claim, this should be fairly easy to do. --Kim Bruning (talk) 18:04, 6 May 2008 (UTC)
 * I hear crickets Reswobslc (talk) 19:42, 6 May 2008 (UTC)

BLP vio removal and 3rr
BLP vio removal is ALWAYS 3rr exempt unless the person removing it is told by consensus that their removal is not covered, at which point they MUST stop or be subject to 3rr sanction. A good example is the Mantanmoreland RFAR case, when Crum375 was blocked twice by the Arbitration Committee for trying to remove unflattering information about User:Mantanmoreland (I forget the specifics now) and at least 7-8 people told him to stop--which he ignored, and got blocked in turn. In general, however, barring if people tell you to knock it off, anyone can remove BLP violations with no fear of 3rr. Those re-adding the BLP vio however, should be blocked if they exceed 3rr, or immediately if it's particularly egregious BLP violations. Lawrence Cohen §  t / e  16:00, 6 May 2008 (UTC)
 * Yep, the policy is clear. 3RR does not apply to BLP issues. That has been part of BLP from the very beginning. It's policy. Whether or not the issue is truly BLP-related is a matter for debate and discussion, but a user who, in good faith, asserts that he or she is reverting to keep material out that is covered by BLP should never be blocked unless there is a broad consensus that the matter is not BLP-related and they continue to revert anyway. FCYTravis (talk) 16:39, 6 May 2008 (UTC)
 * Policy is indeed clear that 3RR does not apply to BLP issues. Kim has not convinced me as to why this accurate section should be removed. undefinedUntil  16:41, 6 May 2008 (UTC)

(edit conflict) Fair enough, but we do need to be thorough. Should be pretty easy then: Please show me 2-3 situations in which the policy was carried out successfully in that manner. (where the situation was not escalated up the dispute resolution chain) --Kim Bruning (talk) 16:50, 6 May 2008 (UTC) Probably we should start adding references to diffs like that everywhere

I have thought for a long time that the "no 3RR" language is misguided and should be removed. It aims to protect BLP articles by enabling editors to remove things that are being added against consensus - but we have numerous other ways to remedy that, including involving other editors via ANI or the BLP noticeboard, protecting the page, and blocking the person (or people) who keeps adding inappropriate material to the article. There's no reason one editor should need to revert more than 3 times, if these other methods are employed as soon as it is clear there is a problem. But the ability to revert often encourages people to revert, rather than encouraging them to do the right thing by bringing in other editors. Also, in borderline cases, the frequent reverting leads to long arguments over whether 3RR should be enforced or not, which waste productivity. &mdash; Carl (CBM · talk) 16:48, 6 May 2008 (UTC)
 * Your missing the point of the 3RR exemption for this or copy right violations. No editor that in good faith reverts in these situations should be blocked for a 3RR violation. With BLP reverting, often the editors doing it are the subject of the article or BLP content. Blocking in these instances is not for the best in the big picture. In any case, the standing wording is working fine for a long while with little problems, I think. FloNight&#9829;&#9829;&#9829; 17:39, 6 May 2008 (UTC)
 * Right. But that's all theory. Can you provide solid situations and/or diffs? (as requested above) --Kim Bruning (talk) 18:01, 6 May 2008 (UTC)
 * Let's shift the onus, Kim - can you provide a single instance where a 3RR block on BLP material was sustained? Sarcasticidealist (talk) 22:02, 6 May 2008 (UTC)
 *  I was anticipating that question and had my diffs ready... but it looks like I've been scooped! (see next section below) --Kim Bruning (talk) 17:14, 7 May 2008 (UTC)
 * Re FloNight, I completely agree we shouldn't block IPs or new editors who are making the reverts in good faith. But we should also discourage experienced editors from using this provision to keep reverting rather than using other methods for handling disputes. Perhaps some more descriptive language can be found here. &mdash; Carl (CBM · talk) 18:06, 6 May 2008 (UTC)


 * It seems to me some users are missing the point. The primary reason for the 3RR exemption is not because it's recommended that editors ignore normal dispute resolution policies. The reason is because regardless of dispute resolution policies, when an editor has a genuine concern of a BLP issues, until and unless there is consensus there is not an issue, an editor should feel free to revert in good faith a misguided (or worse) editor who keeps re-adding the material of concern. The usage of additional measures, including page protection, blocking and an other dispute resolution procedures is implicitly NOT a sign of failure of the 3RR exemption but part and parcel of it. BLP (and copyvio) issues are serious enough and have sufficient urgency that we CANNOT afford to wait the few hours or more until other editors (or an admin) is able to take the time to consider the matters. Note that the 3RR exemption also serves as an important reminder to those who think BLP is a joke that it is not and very likely discourages editors from getting into revert wars in the first place because they think they can 'win'. Editors should NOT be discouraged from using the 3RR exemption if an when necessary, but instead should be encouraged to take all means necessary, including other DR procedures, requesting page protection, requestion a block etc and ultimately, if and when reverting beyond 3RR to ensure BLP violating material stays out of wikipedia. I myself don't think I've ever used the 3RR exemption, but I can see the need for it and would definitely not support it's removal. From my experience, other then a few cases where it very quickly becomes apparent there is little or no support for the supposed BLP concern, most BLP issues which result in edit wars (not necessarily 3RR ones) usually come down in favour of BLP in the end, either in the form of removal of the information or better sourcing. If anything, the bigger problem is that far to many editors are not taking BLP seriously enough in which case the problem is to make editors aware that BLP is indeed a serious matter and the way they should deal with it is not to edit war by reinserting material that is of BLP concern in the absence of consensus itt isn't but to keep it out until there is consensus it is. If editors would understand that when it comes to BLP, it is nearly always going to be more harmful to leave it in when it should be out then to leave it out when it should be in then I think there would be far fewer problems. Remember it takes 2 to edit war. The thing which is semi-unique about BLP is that the general consensus is that we should usually err on the side of caution and therefore the default behaviour is to leave material out even if you have to violate 3RR to get there. Now if others want to improve the wording to remind editors to use other methods of DR and solving the problem in as much as possible, I'm fine with that but not with a removal the 3RR exemption. Since people are asking for examples, I would like some examples of my own. Show me some examples (say at least 3?) where the 3RR exemption has been mis-used, in other words used to remove something which did not violate BLP (meaning the material was eventually kept without additional sourcing) and where the was consensus that the 'offender' was acting in good faith in using the 3RR exemption. It seems to me if someone wants to change a long established policy, they should be the ones showing it needs to be changed. I'm not normally a supported of inertia (for example I still believe it should be WP:MP) but in cases like this I do think it's up to other editors to show why it needs to be changed. Nil Einne (talk) 20:57, 6 May 2008 (UTC)


 * I think you are thinking of articles where the reverts are unarguable. On the other hand, I am thinking of the articles where the reverted material probably should be removed, but there could be good-faith disagreements about interpretation of the BLP policy. That may account for different perspectives.
 * In the case of a true emergency, like sustained libelous vandalism, I think it would take under 5 minutes to get the article temporarily protected by posting on ANI or going to IRC to ask. Can anyone give me some examples of articles where the exception to 3RR has been used where there was no good-faith disagreement about the appropriateness of the material being removed? &mdash; Carl (CBM · talk) 21:13, 6 May 2008 (UTC)
 * The whole point of BLP is that even in "good-faith disagreements," negative information about living persons which is the subject of a reasonable dispute should be left out pending resolution of the discussion, based on the "first, do no harm" principle. The onus is on the person wanting the information in, to demonstrate why it should be included. Without the 3RR exemption, that proviso has no teeth and disputed information can easily be edit-warred back in by any semi-organized group.
 * I would support this only if it was made clear that in such cases, any administrator may revert to a version without the disputed information and protect the article until discussion is completed, and that a person edit-warring to keep disputed information out will not be sanctioned unless that person has made no effort to contact admins for intervention via ANI or BLPN. FCYTravis (talk) 21:58, 6 May 2008 (UTC)
 * I would support that proposal as well, as a compromise. I completely agree that the point is to get the bad stuff out of the articles promptly, and I support that goal. I don't think organized groups are a major concern, because if the material is clearly a violation there will be plenty of previously uninvolved admins around to handle it once they are notified. My concern with the 3RR exemption is that it encourages people to take the matter into their own hands rather than expanding scope. &mdash; Carl (CBM · talk) 22:54, 6 May 2008 (UTC)
 * Something appended that says, "However, if the dispute becomes an edit war and you find yourself in danger of breaching the three revert rule, instead of continuing to revert, request administrator assistance at the biographies of living persons noticeboard or the administrator's noticeboard for incidents. Administrators should protect the page on a version that omits the disputed information, until consensus has been reached on the article talk page." FCYTravis (talk) 23:17, 6 May 2008 (UTC)
 * Well Travis the things you choose to edit war are trivial and non-contentious and I think you're out of place to be shaping the rules here when people feel you abuse them. Like when you recently edit warred over whether Ashley Alexandra Dupré is 5-foot-5 versus 5-foot-6, and called it a "BLP" dispute.  If there were any sanctions to hand out, you deserved one.  You, as an administrator, should not be abusing the BLP policy to engage in run-of-the-mill edit warring.  And you, as an administrator, should not be sitting here moving for policy changes just so you can abuse them.  In any REAL scenario involving a REAL BLP violation, no sane administrator would actually block someone for 3RR for repeatedly introducing contentious information.  But, you on the other hand block people for much less (such as this week-long block citing "BLP" for this edit which appeared to be in good faith and was only ever made once). Reswobslc (talk) 22:56, 6 May 2008 (UTC)
 * I'm not here to rehash your ArbCom-rejected insistence that we include Mr. McGuinness' real name sans any reliable source. We have nothing to discuss. FCYTravis (talk) 23:22, 6 May 2008 (UTC)

Data on BLP and 3RR
Here are eleven fifteen recent reports on WP:AN3 where the page in question was a BLP (hence excluding those which merely mention living people). I have undoubtedly omitted important details in each case.

After a very cursory examination I would say that 1, 4, 5 & 10 sound like cases where the editor could have laid claim to the BLP exception. In two of those cases, however, the editor was blocked. Additionally 6, 9, 11 & 12 are "anti-BLP" cases where opposing edit warriors could plausibly have exploited the exception. Bovlb (talk) 00:47, 7 May 2008 (UTC)
 * Updated with three more. (I must say that most of these edit wars seem pretty lame.)  I see no clear consensus here not to block in BLP cases (or to block in anti-BLP cases).  Does someone want to make the argument that immunity from blocking would have made any of these cases end better?  Do I need to go back further to find the juicy BLP cases? Bovlb (talk) 04:48, 7 May 2008 (UTC)
 * Update: Apparently Ayn Rand died. Why does no-one tell me these things? Bovlb (talk) 04:56, 7 May 2008 (UTC)
 * The result of #4 was a block. So of the four cases with arguable BLP defence, 3 resulted in a block, and one in protection. Bovlb (talk) 20:57, 12 May 2008 (UTC)

The BLP 3RR exemption is vital because it changes the balance of power in favor of keeping material that violates BLP policy off Wikipedia. In my experience, editors who are pointed to WP:BLP generally don't get into revert wars, but either accept a deletion or find good sources. The BLP 3RR exemption, like most good policies and guidelines is intended to prevent disputes from escalating to the point they make the notice boards. So a list of cases that have appeared there is not indicative of the value of this provision. --agr (talk) 14:38, 7 May 2008 (UTC)


 * If the 3RR exemption is to be of any value, it should be reserved for blatant BLP violations (such as Joe Schmoe's home phone number is 1-888-5551212 or that he molested little girls), and not for insignificant stuff like somebody's height or the numerous other edit wars carried out in the name of BLP. Reswobslc (talk) 15:44, 7 May 2008 (UTC)


 * WP:BLP says "unsourced or poorly sourced contentious material about living persons — whether the material is negative, positive, or just questionable — should be removed immediately and without waiting for discussion..." The BLP 3RR exclusion puts teeth into that statement. Limiting it to "blatant" violations would seriously dilute the policy. If a living person's height is a point of contention and we don't have a reliable source that settles the question, their height shouldn't be in the article. Note that the exclusion currently reads "The three-revert rule does not apply to such removals, though editors are advised to seek help from an administrator or at the BLP noticeboard if they find themselves violating 3RR, rather than dealing with the situation alone." So there should not be a prolonged edit war over BLP without admin intervention. --agr (talk) 16:39, 7 May 2008 (UTC)


 * "Should", that's what we wish. But the actual numbers seem to suggest that that is not quite what's happening. Possibly the current wording is not effective at obtaining the desired results? --Kim Bruning (talk) 17:16, 7 May 2008 (UTC)


 * 'should' may be seen as a "non mandatory recommendation", if it is supposed to be mandatory then maybe 'must' would be better? - Nabla (talk) 18:05, 7 May 2008 (UTC)


 * Must just makes a recommendation seem stronger to people, in this case we'd have stronger random behavior. ;-) --Kim Bruning (talk) 18:07, 7 May 2008 (UTC)


 * "You should take the left road". "You must go by the left road". Which will make more people go by the left road? Which will make more people know that the idea is going there (unless there's good reason not to)? But, please, if you plan to reply with another 'funny' remark, save your time to use elsewhere. - Nabla (talk) 23:58, 7 May 2008 (UTC)
 * The problem is not the use of "should" or "must". That's a red herring. The numbers above show that people clearly interpret the rest of the instruction randomly. --Kim Bruning (talk) 00:29, 12 May 2008 (UTC)


 * I'm not sure which "should" you guys are talking about, my comment that there should not be prolonged edit wars or the advice to notify and admin or the BLP notice board when you are going to violate 3RR.--agr (talk) 18:47, 7 May 2008 (UTC)


 * The problem is when the exclusion is abused to unilaterally act against consensus. In the "height" case, the edit war involving FCYTravis was not over whether the material was unsourced, but which of two conflicting sources was more likely to be accurate.  One was multiple sources in the mainstream press, FCYTravis vouched for another website whose "about us" page clearly stated it was a website run by two guys.  There were merits to the validity of both sources... but THAT in itself is not a BLP dispute, particularly since a 1-inch margin of error in someone's height is not a contentious item.  However Travis had no problem reverting three plus times, enjoying the license to edit war invoking BLP to continually revert to the source he (absent consensus) felt mattered more, and abuse the very exclusion he is saying is so important.  It is the community's job to decide which source is more reliable, not Travis's to decide and force his way. Reswobslc (talk) 18:19, 7 May 2008 (UTC)


 * Without commenting on the specific case, there are always going to be situations where people abuse policies and guidelines. See Wikilawyering. Constantly trying to tweak policies to cover all possible abuses is usually a losing game. The longer the policy, the more raw material for Wikilawyers to be creative. See Avoid instruction creep. Perhaps there needs to be more clarity about what a 3RR exemption means. Here is how I see it: Normally 3RR is a procedural violation. All that is needed is to count the number of reverts in the time period. More than three and the editor is blocked -- there is no consideration of reasons for the reverts. In a BLP case, the justifications for the reverts are considered. If they are not sufficient, warnings or sanctions can still be issued. I would suggest that in a case where the multiple reverter's actions were in good faith and they made a timely notice, they should be given at most a warning at first if the reverts are deemed unjustified. If they continue the reverting, sanctions are then in order. -agr (talk) 18:47, 7 May 2008 (UTC)
 * Can you show me where people have done it that way? --Kim Bruning (talk) 19:09, 7 May 2008 (UTC)
 * I'm suggesting a way to go forward. I don't think this stuff has been made clear.--agr (talk) 20:31, 7 May 2008 (UTC)
 * Well moving forward, I would suggest that removing the exemption would be the easiest way to simplify things. That would also lessen the "wikilawyering" problem by giving the lawyers less to bite.  From a practical sense, an administrator exercising good judgment wouldn't block someone reverting multiple genuine BLP violations even if the 3RR rule were violated by its letter (WP:IAR would apply).  By the time the 4th revert comes around, it's time for WP:BLPN anyway, and if BLPN doesn't give a response, then it probably wasn't a BLP violation to begin with. Reswobslc (talk) 22:21, 7 May 2008 (UTC)
 * I agree. The exception is essentially instruction keep.  If they are good reverts, then no one's going to get handed out spurious blocks.  All it seems to be really useful for is making the BLP regulars elitist and for fueling drama.  Celarnor Talk to me  23:27, 7 May 2008 (UTC)
 * The idea was expressed above that minor violations of BLP policy do not warrant use of the 3rr exemption. Excusing minor violations because they are "minor" leads to discussions like "yeah, it's a vio but so what, it's not a big deal, it's minor".
 * It's a lousy idea to have editors arguing whether the effect of dubious information on the subject might be serious. In many cases, we do not know and will never know how "serious" the impact of a violation is.
 * Some editors are not concerned about BLP policy and will look for ways to ignore it. It is important not to start weakening the policy. Further information is here.  CBHA (talk) 16:55, 12 May 2008 (UTC)

Instruction creep?
One could argue that the entire BLP policy is instruction creep. WP:V and common sense should suffice. But WP:BLP was put in place precisely because that was not enough. The 3RR exception puts teeth in the policy. Without it I certainly would not advise anyone to ignore 3RR in the expectation that the reviewing admin would IAR. As I understand it, 3RR applies even if your edit are absolutely appropriate. --agr (talk) 03:15, 8 May 2008 (UTC)


 * agree its policy creep. If there's edit warring over a BLP it can be dealt with at the noticeboard or if immediate action is needed at AN/I, under the ordinary rules. DGG (talk) 18:12, 8 May 2008 (UTC)
 * Instruction creep is unnecessary elaboration of rules. In this case while it is true that BLP policy follows from the pillars, and from common sense, and from ethical behaviour, it is not true that implementing it was working. The BLP policy and specific proscriptions on behaviour were introduced to address that failing. Therefore BLP policy is not instruction creep because it is necessary. ++Lar: t/c 01:10, 12 May 2008 (UTC)


 * You could say that, and I'd agree with you. V, RS, N, NPOV, and COI are all we need to craft a BLP article, just the same as any other.  Just like any other, it can be taken to ANI if something is going on.  Celarnor Talk to me  02:33, 10 May 2008 (UTC)


 * Precisely "But WP:BLP was put in place precisely because that was not enough. The 3RR exception puts teeth in the policy." Nil Einne (talk) 23:00, 11 May 2008 (UTC)


 * Why isn't it enough? Subjects must be notable.  Nothing that isn't verifiable in reliable sources can be added.  That takes care of the most simple forms of slander.  The page must be NPOV.  That takes care of defamation and unnecessarily long "Controversy" sections in articles that should be about the subject.  Subjects can't edit their own articles.  That prevents spam and conflicts of interest.  What more do you need, really?  Celarnor Talk to me  23:08, 11 May 2008 (UTC)
 * Ask John Seigenthaler, perhaps. BLP is established policy precisely because before BLP, Wikipedia often acted like its pages had no impact on the real world, and acted cavalierly toward the fact that people were being libeled and slandered and having their privacy invaded, all over the encyclopedia. The Seigenthaler incident woke us all up. We have a moral, ethical, public opinion and legal responsibility to ensure absolute accuracy, neutrality and sensitivity in all coverage of living people. That is what BLP mandates, and the 3RR exemption ensures that mandate is enforced. FCYTravis (talk) 23:42, 11 May 2008 (UTC)
 * I think that this is a tangent. Get your own section, folks :-) --Kim Bruning (talk) 00:29, 12 May 2008 (UTC)
 * Um, no. V, NPOV and UNDUE do that.  BLP is just extra creep.  If the other policies are followed, it's completely unnecessary.  Celarnor Talk to me  01:21, 12 May 2008 (UTC)
 * Absolutely agree that IF other policies are (perfectly) followed that the BLP policy is completely unnecesary. However, they are not. Since they are not, and since in this case the matter transcends normal practice, the BLP policy is necessary. FCYTravis has it exactly right. We break lesser rules and customs in order to do the right thing about more important, more major ones. And BLP is the most major one there is. ++Lar: t/c 01:30, 12 May 2008 (UTC)
 * BLP was established because there are numerous, endless examples of other policies not being followed everywhere - but they are generally harmless and when they are harmless, there's no urgency to fix them. In BLPs, this is not the case. We have a moral, legal, ethical and public opinion-based mandate to ensure that BLPs are accurate, fair, non-invasive and encyclopedic at all times. FCYTravis (talk) 01:33, 12 May 2008 (UTC)


 * The answer to these problems should be enforce existing policies, not create stronger ones that apply only to a subset of articles. BLP, as a "stronger" copy of N, V, NPOV, UNDUE and COI then becomes unnecessary.  Celarnor Talk to me  01:39, 12 May 2008 (UTC)
 * When every article is BLP compliant, at all times, I will be among the first calling for BLP to be un-fiat-ed. With 250K articles covered by it, and any sort of rough analysis you care to name giving us hundreds or thousands of problematic articles, there is a long way to go. ++Lar: t/c 01:42, 12 May 2008 (UTC)

Data?
Alright, so we've gotten several people providing data, and the data shows that the 3RR exception does not have consensus in practice, and is therefore not policy. All people commenting specifically on that issue with the data at hand seem to agree (apologies if I missed some pertinent comment). Side tracks are not relevant, Those can be discussed in a different section, thank you.

I've left this question up several days. No one has provided evidence to the contrary. Therefore, understanding that some people have strong feelings and having put this up for far more due deliberation and consideration than I'd ever normally do,I will now remove that statement.

While people are permitted to edit policy pages, I would strongly recommend against it here, unless you can somehow provide new evidence that this statement does have consensus. --Kim Bruning (talk) 00:29, 12 May 2008 (UTC) "Before editing this page, please make sure that your revision reflects consensus", as such an edit clearly would not reflect consensus, such an edit is not permitted.
 * Kim: BLP transcends consensus, and transcends other policy, by fiat. There have been arbcom cases that pretty clearly establish that. I suggest you not do anything to soften any statements making policy in this matter very clear. I see User:ArnoldReinhold has reverted your removal, which action I applaud. You should let that reversion stand. Your recommendation to others about not editing should probably apply to your changes, not theirs. ++Lar: t/c 01:03, 12 May 2008 (UTC)


 * I have reverted ArnoldReinhold, because his action does not follow the currently documented Consensus best practice. The requirements to make an edit are as documented by policy, they are not as presumed by ArnoldReinhold. (While his position is becoming more common, it is not most common, nor is it best practice.) Consensus across wikipedia rejects that particular procedure (both at Consensus, and in rather stronger wording at Governance reform. ) I am extra strict here today, because this page has specifically been mentioned in the Governance reform context.
 * While rules can be introduced by fiat, they can be tested and maintained by normal consensus (no other mechanism exists, nor has any other mechanism been proposed or declared by fiat or consensus).
 * As seen above, the 3RR exemption is not a current practice on the english wikipedia. It would be inaccurate to state that it is.
 * As argued above, the 3RR exemption is not a best practice at any rate. It would be inaccurate to state that it is.
 * We therfore find that: in the field, the 3RR exemption is not currently a best practice on the english wikipedia (and thus cannot be said to have consensus)
 * On 6 May, people were challenged to provide evidence to the contrary.
 * It is now 12 May. No evidence to the contrary has been provided.
 * Therefore, we can conclude that current consensus is as stated. (unless someone now provides evidence to the contrary.
 * Policy pages must reflect consensus on current best practice on the english wikipedia.
 * --Kim Bruning (talk) 01:15, 12 May 2008 (UTC) To alter my position, you must provide evidence that the 3RR exemption has consensus in the field.
 * BLP transcends consensus, and transcends other policy, by fiat. Take it to ArbCom if you disagree. If you remove the material in question again, you may find yourself blocked for a short time to prevent disruption. Normally, policy is descriptive. BLP is not normal policy. That it has not been perfectly followed is not evidence that what it states is inoperative. ++Lar: t/c 01:22, 12 May 2008 (UTC)
 * The BLP exemption is policy and practice. Whether it is "best practice" is a matter of interpretation. What is not a matter of interpretation is the need to enforce the provisions of BLP. An edit that violates BLP may actively threaten the encyclopedia by including defamatory, libelous and privacy-violating material. Ensuring that such material is not republished on the encyclopedia is a far more important short-term goal than merely preventing the development of an edit war. Wikipedia's reputation is not at stake when someone reverts four times. Wikipedia's reputation is at stake when we screw up biographies of living people. For this reason, breaking the 3RR to prevent a BLP violation from showing on the public encyclopedia is an example of disobeying a lesser law to uphold a higher principle. FCYTravis (talk) 01:23, 12 May 2008 (UTC)
 * While that is commendable and I have no issue whatsoever with those aims, the test as written empirically fails to reach the objective you state. In fact in my personal experience it has even managed to achieve the opposite. Can we think of some other approach? --Kim Bruning (talk) 01:37, 12 May 2008 (UTC)
 * There is no other approach which can ensure that inappropriate material is removed from the encyclopedia as quickly as possible. We should add a sentence requesting that people seek emergency protection on the non-violating version as quickly as possible, but that is not a substitute for immediate reversions of inappropriate material. FCYTravis (talk) 01:43, 12 May 2008 (UTC)
 * Clearly inappropriate material is vandalism and not subject to 3RR in any case. Anything else should be reverted once, and if it returns, it should be discussed, whether on the talk page or on the appropriate noticeboards, not continually removed and reinserted.  That doesn't get anyone anywhere.  <b style="color:#629632;">Celarnor</b> <sup style="color:#7733ff;">Talk to me  01:52, 12 May 2008 (UTC)
 * No, "clearly inappropriate material" is not considered vandalism. Read the vandalism policy. "Any good-faith effort to improve the encyclopedia, even if misguided or ill-considered, is not vandalism. Even harmful edits that are not explicitly made in bad faith are not considered vandalism." There are often good-faith efforts to improve the encyclopedia which violate WP:BLP. Under your system, they could not be reverted. FCYTravis (talk) 02:00, 12 May 2008 (UTC)
 * (ec) Reversion and fast protection are fine. Edit warring is not. Apparently more BLPs are harmed by inappropriate edit warring than are helped by "appropriate" edit warring. (some of the situations in the above sample are discussed in more detail, and the discussion shows that that is the case). I'm surprised that your position does not take the empirical evidence into account. --Kim Bruning (talk) 01:54, 12 May 2008 (UTC)


 * @Kim: I don't interpret the data the same way you do, Kim. From the table given I see some cases where the teeth worked, and some where they did not, but not that the teeth "achieved the opposite". I see imperfect teeth. Feel free to suggest BETTER ones, but do not remove them without such a suggestion. ++Lar: t/c 01:55, 12 May 2008 (UTC)
 * Ok, I'm willing to look at different interpretations, and will modify my opinion as necessary. Can you point out which cases were effective, according to you? --Kim Bruning (talk) 03:09, 12 May 2008 (UTC)
 * I'm at my client and can't get into this in great detail until later (for which some are sure to be grateful!) but let me make a few points all in this one place, with apologies for scattershotting. First, let me advance this theory... By looking at the 3RR noticeboard, we're looking in the wrong place! On review of some of the cases in more detail, they show that editors were using BLP as cover for advancing other agendas. That's bad. Some other cases show that editors were incorrectly blocked, in contravention of this policy that exempts from 3RR. That's bad as well. About the only case there in that chart that supports this working as designed is #5... the person reverting did exceed three and wasn't blocked, but even that case has a tangled history of what went before that doesn't make it clear cut. So if that's the wrong place, what's the right place? I think it's more at the coal face... how many times at actual articles did we have any edit warring at all and how did it come out. Most edit wars don't get reported to the 3RR noticeboard, after all. The one that Kim and others were engaged in last night didn't! As FCYTravis says below, the analysis isn't broad enough. It might be a herculean effort to actually get valid data that is likely to be statistically accurate and significant enough to draw conclusions from. So THAT makes me ask, is that the right way to approach the problem? Maybe my asking for more data last nite was wrong, and instead I should have been challenging the underlying assumptions of using data at all? Maybe a more logic driven approach is needed? I'm not sure. Second, in general I am feeling a bit less sure of a lot of this. I remain certain that BLP is not a normal policy, that despite the initial discussion of it that went the way normal policies do, that it was imposed by fiat and takes its place among the foundational principles, but I don't have the cites for that assertion that I was sure last night would be easy to find. Maybe ask Jimbo? I think it came from him and from OFFICE actions. I know there have been a lot of OFFICE actions over the years that assumed BLP was not a negotiable matter. ++Lar: t/c 15:20, 12 May 2008 (UTC)
 * It's hard to reply to a scattershot, so I'll concentrate on the key point, which is the demand for data. Basically I'm an empiricist. If there is no observational data, then the claimed phenomenon simply does not exist for me.  Indeed, I don't believe in ghosts or fairies, for the same reason. While I certainly find the possibility that they might exist entertaining, I don't base any real-world decisions on that possibility.
 * The same here. If there is no hard data, you are asking me to take your position on faith. I find your ideas about 3rr exemption very interesting and entertaining, but I do not wish to base my decision on those ideas.
 * I am quite flexible. I understand that sometimes strict statistical analysis is not possible. That's too bad, as I can certainly live with anecdotal evidence and etc, just as long as there is at least some evidence.
 * Well... none of that is happening here.
 * If you want to convince me; Show Me! If you show me a ghost, I'll believe in ghosts; if you show me a fairy, I'll believe in fairies; if you show me the overwhelming success of 3rr-exempt, I'll believe in the overwhelming success of 3rr-exempt. But no one has shown me any of these things so far, so I can't believe in them. Is that fair enough? --Kim Bruning (talk) 16:19, 12 May 2008 (UTC)

Kim, this is getting tenditious, and I advise you to lay off. BLP is a gun that Jimbo gave us when we he still had community support to unilaterally do whatever he wanted or needed to, and it's a gun that we fire when we need to. If the community is fine with it's present interpretation then that's what it is. We don't have to provide data to you to justify it, but people can if they want to. One editor cannot and will not be allowed to hold policy theoretically hostage, and if you want this changed in the way you seem to want to, you're alone in that. You're fine to argue this, but you have zero power to enact policy change a solo editor. No one editor, no matter who they are, if they started in 2001, 2002, or 2008, has power like that over the community. You have this backwards: sorry. If you want it changed, you give us the evidence or data to change it. Otherwise, this is, again, sorry: policy wankery, and let's end it. No amount of political, policy, or tenure capital gives anyone power to change things on their own. Once you have consensus support of multiple editors, maybe this can happen. In the meanwhile, you need to demonstrate for us with your requested data why BLP is not right. Get to it, or lets archive all this as a non-starter. Lawrence Cohen §  t / e  16:32, 12 May 2008 (UTC)
 * eh? Kim has data. It's right up there in the table. His interpretation of that data is that the 3RR exemption isn't working. Last nite my spot checks suggested he was interpreting the table wrong. Today, after I looked harder and thought harder, I agree, the table of data presented contains a majority of incidents where the exemption apparently didn't work (editors got blocked anyway) or worked in the wrong direction. ... let me repeat... the table of data presented. I assert not that Kim hasn't provided data, but that he's provided the wrong data. The right data set to sample is all the actions carried out in support of BLP, not just the ones that ended up at the 3RR noticeboard. This is an anecdotal example, but if you look at my own edit history you'll see I've reverted, and warned. I've never myself gone over 3RR even in matters BLP, but I could have, and knowing I could have gives the words of warning I gave more weight. I expect other editors that work at the coal face of reverting stuff day in and day out, much more hard working and experienced than I am in this area, can chime in with more anecdotes. I gotta "get back from lunch" so I'm done for now. Again, sorry about the scattershotting. But I'll leave you with this thought, Kim is not, in my view, "tendentious". He's just very patient and won't let you slide on things if you're assuming too much, or trying to subvert the wiki way. I consider him one of our best and most experienced editors in that regard. He happens to be wrong about this matter, but do not dismiss him. ++Lar: t/c 17:09, 12 May 2008 (UTC)

Recent edits to Biographies of living persons
I have just now become aware of the discussion regarding the removing the 3RR exception from the BLP policy. I don't have a position on this issue at the moment, but I would like to register a strong procedural objection to Kim Brunning's edits that removed the 3RR exception. The rationale given is flawed: it is not enough to say that there is not a current consensus to have the 3RR exception clause in the BLP policy. That clause was a part of long-standing previously established consensus. Therefore the right question to ask is this: Does there exist a current consensus to remove the 3RR exception from the BLP policy? Based on reading through the above discussion, I have to say that the anser is definitely "No". A more comprehensive discussion, with a post at the Village Pump, and probably a straw-vote or two, is required here. Nsk92 (talk) 01:18, 12 May 2008 (UTC)


 * Even though I strongly agree with the removal of the exemption, this isn't the proper way to go around doing it. The community as a whole has to discuss the issue to get a general consensus.  <b style="color:#629632;">Celarnor</b> <sup style="color:#7733ff;">Talk to me  01:20, 12 May 2008 (UTC)

Hmmm, That is not the correct question. The correct question is whether the policy page correctly reflects practice in the field. (policy is descriptive, not prescriptive, as you probably know. :-) )

Here's the procedure I'm following:

In this case it was clearly shown that there is no consensus in the field. If you cannot show consensus for a particular part of a policy page, then it is effectively a dead letter (no one is actually doing it, so why would you say that they do? ).

The discussion above was in two stages. First people made general comments, then certain people actually came forward with evidence that indeed the 3RR exemption does not have consensus. No one came forward with evidence that the 3RR exemption does have consensus.

If you can show evidence now that the 3RR exemption has consensus, I will agree to reinstate the sentences that were removed. However, I'm not sure you'll be able to do that.

--Kim Bruning (talk) 01:23, 12 May 2008 (UTC)
 * Most policy is descriptive. BLP is an exception to that. If you disagree, take it to ArbCom. ++Lar: t/c 01:24, 12 May 2008 (UTC)
 * The community has made it abundantly clear that they no longer tolerate filibustering and edit warring on policy. If you attempt to filibuster or edit war with me, I certainly shall take it through all steps of dispute resolution, yes. Are you dead certain that your position is unassailable? Otherwise it might be wiser to negotiate and reach consensus, as is standard wikipedia procedure. --Kim Bruning (talk) 01:31, 12 May 2008 (UTC)
 * I am dead certain that ArbCom is not going to sanction any weakening of the BLP. If anything, they will move to strengthen it. FCYTravis (talk) 01:35, 12 May 2008 (UTC)
 * Yes. It may well be that taking this back to ArbCom will be the only thing that will convince you, Kim, that BLP transcends normal practice. So be it. Intermediate forms of dispute resolution and negotiation are irrelevant. BLP is much bigger than 3RR, and your using 2RR or 1RR or whatever to try to unmake that fact is not prudent, in my view. I don't normally speak quite this strongly, as most can attest, but you are in the wrong or I am very much mistaken. ++Lar: t/c 01:38, 12 May 2008 (UTC)
 * I will simply grant for the sake of argument that BLP is as important as you say it is. Now we have found a part of the policy that -in practice- actually sabotages the goal of BLP. Should it not be imperative that this part be removed immediately then? --Kim Bruning (talk) 01:57, 12 May 2008 (UTC)
 * Again, I don't interpret the data the same way you do, I see imperfect teeth, but teeth that do work a good part of the time. But more importantly I'm not interested in "for the sake of argument" or "BLP is important" sort of concessions. Unless you accept that BLP is a fiat policy, unchangeable without a major change to the entire project (perhaps forcing a fork), we don't have common ground yet. But for the sake of argument, what other teeth do you propose as "better", given that we had a problem with BLP articles before BLP was introduced and we still do. ++Lar: t/c 02:03, 12 May 2008 (UTC)
 * Well, I can see why we still have problems with BLP, if you don't allow normal feedback mechanisms to improve wikipedia process. My initial position on this has been that BLP should be banned from wikipedia entirely. The current BLP policy has several counterproductive elements, so I'm not surprised that we still have issues with BLP articles. If you wish to make it impossible to resolve remaining issues with BLP articles, then keep in mind that we DO have those obligations, and I think I'll start pushing for my initial position. Alright? --Kim Bruning (talk) 02:22, 12 May 2008 (UTC)
 * Again, I don't interpret the data so far presented the same way you do, and at best I think that more data is needed before empirical claims can be made about the ineffectiveness of the 3RR exemption. I'm certainly willing to see that discussed, but the outcome, should the empirical evidence come out the way you think it has, would be to ask the fiat-lux for a new fiat. Not a change to the policy. ++Lar: t/c 02:53, 12 May 2008 (UTC)
 * I would hope Arbcom would not directly effect policy changes. Funny though that "BLP" is the exception to (literally) everything. — CharlotteWebb 01:51, 12 May 2008 (UTC)


 * The sentences already have been reinstated by another administrator. Your next revert would be your third. Ironic. FCYTravis (talk) 01:28, 12 May 2008 (UTC)
 * I do not intend to edit war. (and you somehow miscount?) I have only made 1 revert total today. (note that both reverts incorrectly believe that this change was undiscussed. It was discussed and left open for 6 days) --Kim Bruning (talk) 01:31, 12 May 2008 (UTC)


 * Right. The BLP policy was not intended to reflect current practice on english Wikipedia, but to cause significant change in our practice, for the important reasons clearly stated in this policy. --agr (talk) 01:29, 12 May 2008 (UTC)


 * Regardless, the way to get this done is by a posting at the pump involving other people to get involved for its removal. Then we can get consensus to get this extra BLPcruft out of policy.  <b style="color:#629632;">Celarnor</b> <sup style="color:#7733ff;">Talk to me  01:31, 12 May 2008 (UTC)
 * Though it is not a part of the current documented procedure for obtaining consensus, feel free to post on the village pump :-) --Kim Bruning (talk) 01:32, 12 May 2008 (UTC)
 * BLP transcends consensus. It's bigger. SRSLY. You need to internalise that. ++Lar: t/c 01:34, 12 May 2008 (UTC)
 * Fair enough, so knowing that it's so big and important, I'm sure you agree that it is of paramount importance that it actually achieves its aims. The 3RR exemption part of the policy is at best neutral, and at worst actually works against the aims of BLP, we have found. So how do you propose we set about fixing that problem? --Kim Bruning (talk) 01:40, 12 May 2008 (UTC)
 * Under the assumption that you are now prepared to accept that BLP transcends consensus (if not, we have nothing to discuss yet, as we lack common ground on this matter) ... You are the one arguing for the change. What do YOU suggest be done instead? If I were asked, I'd say start rapping the knuckles of those handing out incorrect 3RR blocks. But you're the one taking away teeth without providing a replacement. Suggest away. ++Lar: t/c 01:46, 12 May 2008 (UTC)
 * I'm willing to grant your position for today. I reserve the right to take you to task over it at some later date ;-)
 * .Now, having granted that, I think I should point out that it is empirically shown that the 3RR exemption does not provide teeth, rather the opposite! (Else why would I be wasting my time here?) . There are several other mechanisms by which BLP can be (and hopefully is) successfully applied. Removing 3RR exempt makes BLP a stronger policy overall, with no other action required. --Kim Bruning (talk) 02:04, 12 May 2008 (UTC) (unless you can come forward with further data that disproves the previous observations)
 * I'm not sure how many times I need to say this. I don't get the same result from the data you do. I don't see a thorough analysis of the data, just some assertions of what it shows by some people. (so I'm not asserting I absolutely am right, just refuting that you absolutely are, and casting doubt) Usually when I don't get the same result from data as someone else, my suggestion is to collect more data and do a more thorough analysis. You keep asserting that the data supports you, but that's just argument by assertion. ++Lar: t/c 03:06, 12 May 2008 (UTC)
 * BLPcruft? Wow. Just wow. FCYTravis (talk) 01:35, 12 May 2008 (UTC)
 * Can you point to something that makes BLP immune to consensus? The only thing I can come up with would be an office statement, since we can't create something by consensus that is immune to it; that's simply inane.  <b style="color:#629632;">Celarnor</b> <sup style="color:#7733ff;">Talk to me  01:37, 12 May 2008 (UTC)
 * BLP was not created by consensus, any more than RS or NPOV were++Lar: t/c 01:40, 12 May 2008 (UTC)
 * "There is general agreement that reliable sources are needed for the facts presented in Wikipedia articles." . Looks fairly consensussy to me (and there's wikiediting going on past that date too, which is the normal consensus system at work by the looks of it). NPOV was actually defined on a wiki preceding wikipedia (using wiki based consensus), and wikipedia picked it up as the most ideal system for an encyclopedia. --Kim Bruning (talk) 01:45, 12 May 2008 (UTC)
 * "We the people of the United States, in order to form a more perfect union," ... is softening text for what comes next, not a statement that consensus was reached. The basics here, NPA, RS, NPOV, are foundational principles imposed by fiat and are not subject to change by consensus regardless of any preambulatory softening text. ++Lar: t/c 01:51, 12 May 2008 (UTC)
 * Foundational principles? I know of 5 Pillars and Foundation issues. Which do you mean? --Kim Bruning (talk) 02:28, 12 May 2008 (UTC)
 * The latter. (of which RS is not among, it seems, oh well :) )... 5P is more a distillation of a lot of policies, some of which are subject to consensus. But Foundation issues clearly states "some things are beyond debate" and "ArbCom can make binding decisions". ++Lar: t/c 03:06, 12 May 2008 (UTC),
 * Actually, yes we can. Much as the United States Constitution was able to place certain portions of itself above amendment, the BLP is a permanent part of policy. If it didn't exist in its current form, the Wikimedia Foundation would impose something by fiat. This is not an Internet playground. These are real people's lives we're talking about, and it is a moral, ethical, legal and public opinion imperative that we respect that, and ensure that we do not do harm to the people whose lives we chronicle. FCYTravis (talk) 01:41, 12 May 2008 (UTC)
 * For some reason, you seem to think I don't agree with most of this (short of BLP being unamendable... that would be a very serious problem, though I believe that the evidence is strongly against you on that point - and continuing). --Kim Bruning (talk) 01:48, 12 May 2008 (UTC)
 * You're missing the point. Of course they're real people, and of course we can't say whatever we want about them (V), we need to be able to back up what we do say (RS), and it has to be neutral (NPOV, COI, UNDUE).  All those statements already have policies to go along with them.  The view that we need an additional mirror of existing policies doesn't really make sense to me.  <b style="color:#629632;">Celarnor</b> <sup style="color:#7733ff;">Talk to me  01:49, 12 May 2008 (UTC)
 * That's partially because you've only been a regular user for two months. You don't have any sort of perspective on this issue. That's not necessarily your fault. We should do a better job (on this talk page) of explaining the history and genesis of BLP, and why it is now regarded as one of the most important policies on the encyclopedia. FCYTravis (talk) 01:52, 12 May 2008 (UTC)


 * I'm well aware of the Brandt, Siegenthaler, and other associated incidents. I know that policies exist that require material to be neutral and verifiable in reliable sources.  I know it started as a proposed guideline in 2005, later became a guideline, and later became a policy.  How long I've been a registered user has nothing to do with it.  I still haven't seen anything not circumstantial (i.e, not editors just ignoring core policy) anywhere that suggests this mirror is necessary.  <b style="color:#629632;">Celarnor</b> <sup style="color:#7733ff;">Talk to me  01:58, 12 May 2008 (UTC)


 * The last bit might be slightly exaggerative. I've been here two years, give or take, and I haven't yet needed to use "BLP" to justify an edit or action. That doesn't mean I consider this page useless, only that less contentious policies have better served my purpose in every case. Mileage may vary of course. — CharlotteWebb 02:06, 12 May 2008 (UTC)
 * Have you spent any great deal of time dealing with OTRS or BLP Noticeboard issues? Your perspective might change a little. FCYTravis (talk) 02:07, 12 May 2008 (UTC)
 * At 18:04, on 6 May 2008 (UTC) I requested that you provide evidence to support your position. With your stated experience, this should have been an easy thing to do. I still have not seen that evidence. Can you provide it now please? --Kim Bruning (talk) 02:15, 12 May 2008 (UTC)


 * I spend time at the BLPN, yes. What I see there are people requesting sourcing (N, RS, V), discussing an especially long controversy section (UNDUE), the subject editing the article (COI), or some heavy editing in favor of one position (NPOV).  I've never seen anything where BLP has served some kind of special purpose that other policies didn't already provide.  <b style="color:#629632;">Celarnor</b> <sup style="color:#7733ff;">Talk to me  02:35, 12 May 2008 (UTC)

Consensus can change
<<< Consensus can change ... and to remove an important portion of the policy requires an effort to build consensus for that removal. The fact that it has been already reverted twice, is a good indication that such consensus does not exist as we speak. ≈ jossi ≈ (talk) 02:00, 12 May 2008 (UTC)
 * As documented, there is global consensus, and there is local consensus. I have shown that global consensus is that the 3RR exemption does not exist. I have also gained local consensus that the finding on global consensus is correct.
 * I now merely need to translate that into a modification to the page. This latter part is now being done, as per WP:BRD. --Kim Bruning (talk) 02:09, 12 May 2008 (UTC) in short, I'm being very careful to follow documented procedure as much as possible here. 


 * Personally I have greater confidence in ignoring rules in situations where they don't make sense than in injecting numerous kludges of verbiage just to keep policies from eating each other alive. Maybe I shouldn't admit to things like that (shrug). — CharlotteWebb 02:16, 12 May 2008 (UTC)


 * It's the part where a good editor then needs to document what they learned while ignoring the rules that drives me round the bend sometimes %-). Someone's got to do it! --Kim Bruning (talk) 02:44, 12 May 2008 (UTC)

(ec) I came here because I watch Kim Bruning's talk page, and saw him being warned by Lar, and I found that so thoroughly offensive that I almost restored Bruning's edit. However, I did cool down and read the arguments. Bruning is following an important principle, and his middle name is "God," but perhaps he is testing us mere mortals in this case. Seems to me that there is consensus that there is an exception to 3RR when removing material that an editor, in good faith, believes is improper for BLP. That in actual practice some users are being blocked for violating 3RR in what are arguably such situations only shows what most of us already know, there are some improper blocks; but I didn't track down the situations. The exception is not a carte blanche. The presumption of good faith is rebuttable. 3RR is normally a bright line, and we can be blocked for violating 3RR regardless of our absolutely appropriate application of WP:IAR. Being blocked is not the same as being flogged. I've often said that if you've never been blocked, you aren't working hard enough to improve the project. I keep trying, but, so far, no blocks for me. Sooner or later, I'm sure.

But BLP has some to be considered quite unique, and there is clearly consensus on that. I was removing edits from a sock puppet, a particularly nasty one. And he started deliberately editing porn star articles to remove what seemed, on the surface, to be BLP violations, knowing that he could then call down opprobrium on me for reverting him. I did, in fact, restore stuff that on casual examination, looked like violations, but which wasn't, necessarily, and I came as close as I've ever come to being blocked for it. Since I didn't really care that we had full sourced information in those particular articles, and since I preferred *not* to be doing source research in that field, and debating the details, I didn't push the point, though I could have. I'll save point-pushing for truly important situations, where it is *worth* getting blocked. Someone else could do it, putting in what was amply sourced and easily available outside.

There is an interplay between consensus and practice. Actual practice may *not* reflect consensus; to discover actual practice here, though, we'd have to know how *often* 3RR edits are allowed, not merely that sometimes they are not. There is practically no way to collect that information. Normally, if there is enough discussion, consensus and actual practice will converge, in any case. It appears to me that there is consensus that we need to make it easier to remove possibly damaging and unsourced material from BLPs, and so it makes total sense to me, and apparently to most commenting here, that we allow, for a time, relatively unrestricted reversion to remove such material, thus placing the burden on the one trying to insert it or maintain to obtain consensus first. The down side: some proper material might go missing for a short time. It will be there in history. Further, I'd not be particularly tolerant of someone continuing to revert against all comers, for days, and without obtaining community support at AN/I or BLP. That 3RR is no longer a bright line with BLPs doesn't mean that the principle no longer exists. So, God Kim, did we fulfil our duty here in the face of your trial of us? --Abd (talk) 02:14, 12 May 2008 (UTC)


 * Actually, it turns out that people use the 3RR exemption to edit war on BLPs, as opposed to its intended purpose of removing inappropriate content. This actually makes it much *harder* to keep BLPs neutral, reliable, and verifiable. So that particular recommended practice has actually been having the opposite effect to what was intended. Does that make sense? (you don't need to agree, but can you see why I would want to make the edit that I did?) --Kim Bruning (talk) 02:19, 12 May 2008 (UTC)
 * How does it encourage edit wars? It only protects *removal* of material.  If somebody repeatedly reverts to re-add material (good or bad), that's still blockable.  In *all* edit wars admins can block.  All the exception does, is specify which edit is to be blocked (namely the one adding the harmful material).  When has blocking a good faith use of the 3rr BLP exception been beneficial.  BTW, do you oppose all of the several exceptions to 3rr, or just the BLP one?  --Rob (talk) 02:32, 12 May 2008 (UTC)
 * All I know at the moment (per best information) is that the 3RR exemption for BLP is not working as intended in practice. (see: ). I can't comment on the other issues as I don't have data on those (yet). Those other issues might be interesting to look at too, though as you can see, my hands are currently a bit tied ;-) . Would you be willing to take a look? --Kim Bruning (talk) 02:48, 12 May 2008 (UTC)


 * Ah, Abd, you're the fellow on their third marriage lecturing me about my first, aren't you? If you were offended by my warning Kim, who I consider to be a good friend, and who I would rather not see blocked, that he was treading where eagles fear to go, I'm confused. But then much that you say confuses me.
 * In this case though, I find myself in substantial agreement with most of the rest of what you say. There is indeed a tension between processes and the things that exist outside the space where they are operative and nevertheless intrude into the space where they do. And the 3RR exemption is indeed applied imperfectly, and in some cases used by those who would edit war for its own sake or rules lawyer for its own sake. But I'm not convinced that the sample set we have so far, drawn by a few people, empirically demonstrates that it is a net negative. I think we need FAR more data, drawn by more people, and subject to more rigorous analysis, before we can conclude that. ++Lar: t/c 02:27, 12 May 2008 (UTC)


 * And in the mean time I have acted on best information. :-)


 * I'm always willing to change my opinion if you can provide data in support of your position. (though it would have been nice if folks had done so last week already, when we asked nicely.:-) --Kim Bruning (talk) 02:35, 12 May 2008 (UTC) You can't claim that I made any edits without giving a massive amount of warning a week in advance here ;-) 


 * This ties into the meta-meta debate we've also had recently at MoS on just how prescriptive and/or descriptive policy can be/should be/is. It seems to me that BLP is too serious a matter to water it down by removing the 3RR exemption, however worthy the philosophical reasoning for that may be. --John (talk) 02:50, 12 May 2008 (UTC)


 * There is no philosophical reasoning. The empirical evidence is that the 3RR exemption waters down BLP. Removing it will make BLP more effective. --Kim Bruning (talk) 03:01, 12 May 2008 (UTC)
 * So you assert. I assert otherwise. More data, and more thorough analysis of it, is needed because we can't both be right. ++Lar: t/c 03:08, 12 May 2008 (UTC)
 * Sure. Please provide data in your favor. :-) --Kim Bruning (talk) 03:14, 12 May 2008 (UTC)

RfC: What kind of consensus is required to change BLP
What kind of consensus is required to remove the 3RR exception from the BLP policy? -- (unsigned)

Answer: Mu! Wrong question. The question is whether policy pages must actually describe global consensus. In this case the policy page does not accurately describe the consensus of the wiki, and I'd like to correct it. --Kim Bruning (talk) 02:11, 12 May 2008 (UTC)

Answer: Mu! Wrong question to both the original and Kim's questions. The question is whether BLP, as a policy imposed by fiat, is subject to change by the consensus process at all, and further, whether aspects of the process or enforcement mechanisms that implement policy are part of the consensus process, or are imposed (by fiat, or by arbcom findings) and therefore also not subject to change by the consensus process. ++Lar: t/c 02:16, 12 May 2008 (UTC)
 * PS - If you accept that this is the correct question then I suggest that it is not one that can itself be answered by the consensus process of an RfC either... why do I feel like Gödel? ++Lar: t/c 02:21, 12 May 2008 (UTC)
 * During wikimania boston, we played a game called calvinball (a variant of nomic), and actually discovered ways in which systems based on hard rules could fail. I'm guessing that you've just managed to fry your brain in that exact way :-P --Kim Bruning (talk) 02:39, 12 May 2008 (UTC) Good, saves me writing out a whole reductio ad absurdum myself ;-)
 * I've played (and won) Nomic before and am familiar with Calvinball... I even watched it being played at WM 06. My brain isn't fried at all. It's just clear to me that this matter won't be resolved by this RfC, because rules based systems have areas that are outside the rules. Which is why no rules based finite system ever functions perfectly in a universe that is larger than it is. Hence the Goedel reference. That's not at all a reductio ad absurdum, it just is. ++Lar: t/c 02:48, 12 May 2008 (UTC)
 * Again, can you show us where it is said that BLP is above the rules? It was proposed as a guideline in 2005, later made a guideline, then later voted into policy, just like everything else.  I don't see anything in the history that says "This can never be changed by consensus."  <b style="color:#629632;">Celarnor</b> <sup style="color:#7733ff;">Talk to me  02:56, 12 May 2008 (UTC)

For the benefit of bears of little brains like myself a little bit of context from my perspective, per the above. This is 'my take' on the situation, so usual disclaimers apply;

I think it's fair to say that Kim believes that BLP articles are not 'protected' by the existence of a 3RR exemption, but harmed, and that further, there is no global consensus for the 3RR exemption to remain in place. Lar sees the 3RR exemption as a useful tool in giving the 'BLP' policy some teeth, and further believes that because it has been 'ruled' (by fiat) that discussion of consensus is not material.

My tuppence; Wikipedia is jaw droppingly irresponsible in the area of biographies (come by for a chat about this any time), and has shown little ability to respond accordingly. The 3RR exemption should be removed, and subjects of Wikipedia articles should be offered a way of removing themselves from the currently dysfunctional system. Privatemusings (talk) 02:49, 12 May 2008 (UTC)
 * I agree that "Wikipedia is jaw droppingly irresponsible in the area of biographies". Personally, I very much like Kim's other suggestion that we be shut of them entirely since we cannot do a good job with them (the empirical evidence for that is easy to find)... it would have to be implemented by fiat as well, I suspect. I also think that "subjects of Wikipedia articles should be offered a way of removing themselves from the currently dysfunctional system". But removing the 3RR exemption does not logically follow from any of that, even if this were a normal consensus matter. Which I contend it is not. ++Lar: t/c 02:57, 12 May 2008 (UTC)
 * Yeah - sorry about the complete lack of explanation for my opinion! In reading the above, and in taking some time to think about it a bit, 2 criticisms of the 3RR exemption seem to be worth examining at this stage - Firstly that it's not the 'wiki way' to disallow evolutionary consensus processes, and that we may be losing the baby along with the bathwater, and secondly that the 3RR exemption can be seen to be an unhelpful process - as in is in practice counter productive to its intention (more evidence / discussion on that point might quickly convince me that it's in fact a good thing - but it doesn't seem like a slam dunk on a '20 minute look' basis).
 * ps. if you do support the 'optout' - would you (and others!) mind taking a look at OPTOUT/Long_Term_Straw_Poll? cheers, Privatemusings (talk) 03:04, 12 May 2008 (UTC)
 * Tomorrow. Tonight I'm for bed. ++Lar: t/c 03:10, 12 May 2008 (UTC)
 * The only argument I see for the removal of the 3RR exception is that "it is used/abused to edit-war". Well, almost every other policy can and has been used to edit-war. What's new? ≈ jossi ≈ (talk) 03:16, 12 May 2008 (UTC)
 * Nothing. When other policies are misread we correct them. I don't see why this is different. (But let's give Lar some time to provide more data in his favor. :-) ) --Kim Bruning (talk) 03:19, 12 May 2008 (UTC)
 * Lar, do I understand you right--are you suggesting that we be the one encyclopedia in the world that does not cover GW Bush? DGG (talk) 03:21, 12 May 2008 (UTC)
 * well let's not get too distracted (and I think Lar's gone to bed) - but no, I don't think anybody is saying that we shouldn't have an article on GW Bush - that would not be a result of the opt-out proposal. The proposal might however go some way to undoing some of the harm caused by Wikipedia's jaw dropping irresponsibility in the area of biographies. I see you signed as opposing the idea, but I think you might have misunderstood it a little? - maybe give it another look if you're so inclined... Back to the RfC topic however - I think the most valuable area of discussion right now is the strength of the data above - I'd be particularly interested to hear others' thoughts on that (it would certainly help me make my mind up!) cheers, Privatemusings (talk) 03:26, 12 May 2008 (UTC)

Evidence of efficacy, use, or fiat
I see a lot of assertions above that essentially boil down to one of: While I do not agree with Kim in all respects, I have to agree that despite repeated requests over a prolonged debate, no-one has yet provided substantiation for any of these three points. For the first two, if they're empirically true, then it should be easy to provide some examples. I realise that people may fear providing examples because real life is never clear cut and it's always possible to raise some counter-argument, but surely there are some cases where these are at least arguably true? For the third, presumably we wouldn't find ourselves claiming it with such vigour without some documentary basis. Is there a link to some office action, or pronouncement by ArbCom or Jimbo or something to demonstrate that these aren't empty words? I have all the sympathy in the world for attempts to balance our mission with the legal and ethical considerations of BLPs, but I think we need to be clear on what we're doing and why. If I've missed somewhere above where someone has responded to any of these, please accept my apologies and point me at it. Thanks, Bovlb (talk) 05:09, 12 May 2008 (UTC)
 * 1) The BLP 3RR exception improves Wikipedia's handling of BLPs;
 * 2) The BLP 3RR exception is commonly used; and
 * 3) The BLP policy in general, and its 3RR exception in particular, are imposed "by fiat" and therefore trump both consensus and best practice.
 * Nobody's saying it's commonly used. It's the threat of its use (or the unspoken knowledge that it can be used) which is the point. It is the teeth of BLP - it is what says "if there is a legitimate concern about something because of BLP, remove it and keep it out until it's discussed, period." Without it, BLP has no force. "Well, I disagree with you on this issue, so I'm just going to revert-war you, and if you revert me, you'll hit 3RR and be blocked, so you can't remove it, nyah nyah." The 3RR exception stops that scenario from happening. It says "No, you simply reverting the potential BLP violation back in won't do you any good, because 3RR doesn't apply." I have used that threat on Star Wars kid. It is a deterrent to problematic content that has substantive use in BLP matters. FCYTravis (talk) 06:36, 12 May 2008 (UTC)
 * Thanks for the example. I intended "used" to cover cases where the existence of the exception affected events, not just where it was actually exercised.  I suspect you would agree that the exception would not be worthwhile if it rarely had any (positive) effect in this broader sense.
 * To summarise the example: Some kid became the star of a cult internet video without his permission, and at least one lawsuit resulted. The edit war was about the inclusion of his name.  It is in court records, and was mentioned by some news sources, but he has no other notability, and has not sought public attention  There is a strong case that the article loses no context by excluding his name (see WP:BLP).
 * The article has been subject to frequent insertion of his name (by both edit warriors and drive-by editors), and has been under long-term (semi-)protection. You (and others) argued strongly against including the name from (at least) 2007-08-23, and on 2007-10-14 you said "Anyone inserting his name will be reverted without recourse to 3RR".  It's hard to say for sure what effect the exception (and its invocation) had on the edit war, but so far as I can see the name had not been inserted for 20 days before this remark, and was not inserted again until 41 days afterwards.  The effect seems to me to have been largely on the course of the talk page discussion.  I can only find one BLP/N report related to this article, and it does not (directly) concern the inclusion of his name.
 * So ... I think it quite possible that this invocation had the effect of discouraging talk page contributors from restarting an edit war, but I can't quite convince myself that posting the issue at BLP/N might not have worked just as well.
 * On a tangent, I note that of the eight interwiki links on Star Wars kid, at least seven name him (it's not obvious in the Japanese). I realise that I'm stretching your example beyond where you intended it to go, but this makes me wonder about how BLP is applied on other Wikimedia projects.  If this is a fiat, presumably it is one that applies to all.  Does anyone know whether other projects have both 3RR and a BLP exception? Bovlb (talk) 08:22, 12 May 2008 (UTC)
 * A 3RR exception provides a shield for editors enforcing BLP...that shouldn't be that controversial. An example would be | 9/11 conspiracy theories and Shel Silverstein. But there are more I'm sure. If you take a look at the history on April 21, you'll see several editors trying to insert his name in association with conspiracy theories on the basis of one conspiracy theory web site. In this case no one had to go past 3 but you had several editors "tag teaming" the insertion of unsourced implications of involvement/fore knowledge in the 9/11 attacks. But this 3rr shield is needed where tag teaming going on and only a single editor trying to fight them off. Of course that editor should also take to to a noticeboard, but while that's being done the they shouldn't be afraid of pushing the rollback button for the third time on a BLP issue...especially since we should be removing BLP content immediately (no one disputes that). RxS (talk) 12:54, 12 May 2008 (UTC)
 * That's a lot closer than most folks have gotten. Can you also show evidence of situations where the 3RR exemption was actually used? --Kim Bruning (talk) 17:46, 12 May 2008 (UTC)
 * Two cases where the 3RR exemption was actually used. I searched the 3RR archives looking for 'No action' cases where BLP was mentioned as a factor by the closer:
 * The Alansohn-Xcstar case. A clean case, no block issued, BLP was argued. There were four reverts by Alansohn within a 24-hour period: 20:32 Jan 6, 16:39 Jan 7, 16:56 Jan 7, 17:08 Jan 7 (all times UTC). This one could have justified a block if not for the BLP exception. User:Kafziel mentioned the need for sourcing BLP claims in his decision.
 * The Gamaliel-DJ_Creamity case. This is an almost clean case except that the closer gave credit for a self-revert. There were enough reverts for a 3RR block without use of the BLP exception, and the closer did mention the BLP exception. In deciding not to block, User:Heimstern Läufer suggested that the editors take it to the BLP noticeboard for vetting of the BLP issues.
 * I did not include any cases where 'Full protection' was the result even though there were a number of such cases where BLP was mentioned by the closer. When a case ends in protection no blocks are issued, and we have no insight as to how the closer counted the reverts. They may not even have counted up to four; they just felt the situation called for protection. EdJohnston (talk) 19:59, 12 May 2008 (UTC)
 * <Phew> Finally folks are showing some evidence. :-) I actually see what's actually going on there now, and I might be able to tweak things so that the situation doesn't end up on the 3RR noticeboard as often even. Thank you so much, EdJohnston! --Kim Bruning (talk) 20:27, 12 May 2008 (UTC)

Pre-BLP policy
The limited discussion about removing the 3RR excemption from this policy missed several important considerations. I only have time to address a few now.


 * 1) So far no data supports the idea that edit warring on BLP content is made worse not better by the "3RR exemption" in this policy. What has been presented so far does not take into consideration that editing waring happened on BLP related content long before the "BLP 3RR exemption" was added.
 * 2) Is the intent of removing this section of the policy to make it best practice block more users? If so, I want to see strong evidence that there is consensus for that idea. In general, any policy that opens the door to blocking more users needs more than a few days discussion.

Before the BLP 3RR exemption is removed, data needs to be presented that pre-BLP policy or some other alternative approach is better than this policy. FloNight&#9829;&#9829;&#9829; 14:05, 12 May 2008 (UTC)


 * Here is a quote from the intro to the December 19, 2005 version of this policy, when it was still being proposed:


 * "Place yourself in the shoes of a marginally notable person who wakes up one morning to find that someone has written a Wikipedia entry about him. He learns that the page may be edited by anyone in the world, including entirely anonymous editors, and that it may be picked up by mirror sites and copied all over the Web. He tries to correct what he sees as mistakes or an unfair slant, but his edits are reverted, and he realizes he has no control over the article's contents. He continues to revert, but finds himself blocked for 3RR. He discovers there is no editor-in-chief he can easily appeal to, so he leaves a belligerent legal threat on the talk page, and is blocked indefinitely."
 * The genius of 3RR is that it keeps any one editor, IP to admin, from having too much say over an article. Article contents reflect group process. Our bureaucratic structure controls behavior, not content. You don't have to look any further than this discussion for an example. Kim felt strongly that the BLP 3RR exception should go and removed it. I reverted. Kim reverted back. Another editor reverted that and the issue is now being discussed again, at length, on this talk page. Meanwhile the exception remains in the policy. That works fine for most Wikipedia content, but not for articles about living people. For them the legal and ethical calculus changes. Leaving material in place, even with a "disputed" tag, can cause serious harm. The current BLP policy properly recognizes that. If the exception creates problems that cannot be managed by processes already in place, then perhaps some additional clarification may be needed (note that such a clarification was made recently, advising editors who exceeding 3RR to contact an admin or the BLP noticeboard). A few abusive incidents is not a reason to neuter an important policy.--agr (talk) 15:04, 12 May 2008 (UTC)


 * 1) I'm sure the Rosalind Picard article was mentioned in the evidence (and that's where I started) so your "no evidence" statement is false. Are you saying that 3rr-exempt has no discernible effect at all? Currently no evidence has been brought forward to show that 3rr exempt is even applied very often, let alone that it works. (I have reason to suspect that it wouldn't do what people hope it would do anyway, but that's a different discussion)
 * 2) Although making predictions is always slightly iffy, I'd say that based on the available evidence, I'd lay odds that the nett effect of removing the section might in fact prevent a (small) number of blocks.
 * --Kim Bruning (talk) 16:30, 12 May 2008 (UTC)
 * I am baffled by the frequent mention of the Rosalind Picard case. Since it ended in full protection, by my analysis it proves nothing. There may not even have been four reverts counted. We have no idea whether the closer took BLP into account, even though BLP was mentioned by other editors. A change in the BLP rules about 3RR is supposed to influence the closer. We don't know how the closer was counting the reverts in the Picard case, or whether he was excluding any of them for BLP reasons, so I don't believe it constitutes data on whether closers ignore BLP. EdJohnston (talk) 20:50, 12 May 2008 (UTC)

No amount of data will ever authorize removal of the 3RR exemption WITHOUT overwhelming consensus support. I can present 100 links showing this, that, or the other thing. Unless there is community buy-in to remove 3RR exemption, which I do not see, this is not going to happen. Discussing it in ever-widening avenues of discussion or approach are all reaching the same conclusion, that there is no support to remove it. So why are we pursuing this? How many people here actually even support this, beyond Kim? One editor has and is entitled to no power here over the masses. Lawrence Cohen §  t / e  16:35, 12 May 2008 (UTC)
 * Excuse me? I interpret the data as saying that there is already no global consensus for the 3RR exemption. Currently no other data has been supplied. I have been acting on best information. (If new information or interpretation emerges, I shall change my behavior). Read the tag at the top of every policy page: "Before editing this page, please make sure that your revision reflects consensus". Now Consensus states that local consensus is overridden by global consensus. In short, to the best of my current knowledge, anyone may remove the 3rr exemption statement, based on current information.
 * If you interpret policy differently, please explain your interpretation, based on current best practices documentation.
 * --Kim Bruning (talk) 16:47, 12 May 2008 (UTC) I agree that no single editor or small group of editors is entitled to power over everyone else. But can you explain why are you acting against your own belief?
 * I'm not acting against my beliefs, Kim. I believe absolutely in a strict BLP policy, and any attempt to weaken it even slightly, even in good faith, goes against my beliefs. Removing 3RR exemption will harm Wikipedia and living people, and there is no support to remove it. Lawrence Cohen  §  t / e  17:22, 12 May 2008 (UTC)
 * You are arguing in the face of evidence to the contrary. Removing the 3rr exemption is more likely to lead to better behavior on BLP pages. Leaving it in has harmed and will continue to harm people. We don't need to wait for Lar. Show me new evidence that says the opposite, and I will change my mind. --Kim Bruning (talk) 17:40, 12 May 2008 (UTC) I hold that you are acting against your own beliefs, because you refuse to allow the page to reflect global consensus.
 * ...but where is the consensus to remove the 3RR exemption? How many users are on the record supporting that? Lawrence Cohen  §  t / e  17:45, 12 May 2008 (UTC)
 * See . Many BLP incidents, no application of 3rr exempt. Policy must reflect actual global (best) practice/consensus. Show examples of situations where it was applied, and I'll concede, and we will be done. (I've been waiting, and no one has done so so far the past week so far, so please forgive me for becoming a bit skeptical already ^^;; .) --Kim Bruning (talk) 17:52, 12 May 2008 (UTC)
 * But that isn't consensus. I can provide any amount of data on any topic, but without users expressing support for a change, there is no consensus. How many users there in that section agreed with the remove of the 3RR exemption? 1? 5? 10? 100? Lawrence Cohen  §  t / e  18:01, 12 May 2008 (UTC)
 * What was measured is a form of consensus. It is called "global consensus", "wiki-wide consensus", or "consensus-in-practice". Such "larger" consensus has precedence over local consensus. That's what prevents the tyranny of the minority. --Kim Bruning (talk) 18:30, 12 May 2008 (UTC) this is also why calling in more people helps, as the local consensus will tend to move towards the global consensus in that case.
 * Anything like that automatically is deprecated if a change is contested. Myself and others contest any removal of 3RR exemption, so it would need to be decided by a real consensus, and nothing less. I'm sorry. I see no data, rationale, or consensus today that supports removal of 3RR exemption. Anything that would neuter or muzzle attempts to defend BLPs is abhorrent to many, and you're going to be extremely hard pressed to ever draw up a convincing consensus of multiple users to support this. Any attempts to remove it from the policy as "written" will similarly fail. I think the idea will never have consensus, and I again challenge you to demonstrate a consensus and reasons why we don't need this exemption. Sorry... Lawrence Cohen  §  t / e  18:37, 12 May 2008 (UTC)
 * And this is why I accuse you of being in favor of the tyranny of the minority, you see. I have yet to see you accept global consensus. You prefer minorities of size 1? 5? 10? 100? over the best practices used across the entire wiki.
 * Now, in the mean time, global consensus does exist, and as more people have been visiting this page, more people have in fact been speaking up against 3rr exempt (have a proper read, and confirm what I'm saying! :-) ).
 * I'm not sure I'd go as far as say I'd support "defending" BLP's... that sounds like old fashioned wiki-war, which we did away with years ago.
 * I *am* quite in favor of making BLP stronger, of course, because that's an action that improves the encyclopedia. Now will you help with that, in a reasonable, rational way, or (if you do not wish to cooperate in such a fashion) will you hold your peace? --Kim Bruning (talk) 18:46, 12 May 2008 (UTC)

@Kim: The 3RR rule is obviously important to Wikipedia. It is policy and goes a long way in dealing with edit-warring. It helps editors do their job better, it helps admins do their job better, it helps people understand what WP is all about. However, it is an in-universe wikipedia term and concept. When it comes to publicy printing potentially damaging information about a living person, any wikipedia in-universe policy is irrelevant. That is basically what the BLP policy is stating. It's telling anyone that looks at it that Wikipedia's internal policies on how we run the joint will never trump real-world, grown-up stuff. Like possible defamation. It's saying "we have these two policies that are basically contradictory, and for the sake of clarity, don't worry about this in-universe 3RR rule when you believe in good faith you are protecting a BLP". You ask for examples of its effectiveness. I was wondering if in my mere 1,300 edits I have ever had to break 3RR when BLP reverting. I don't believe so. But it doesn't matter... because I would do it, in a heartbeat. I would do it even if for some reason "3RR" says I can't. I abide by wikipedia policy, but never will I let it trump common sense, or defamation. IAR indeed. The data above gives me concerns. It shows that not everyone is on the same page with this, including admins. It shows that many are abusing BLP protection when it doesnt apply, and then hiding behind the exemption. I concede that. You can remove the sentence for all I care, but it doesn't mean knowledgable users who understand both the BLP policy and the 3RR policy will all of a sudden give equal weight to both. I won't, and neither will most. 3RR is a tool for effective editing and enforcment, BLP, whether typed up as a pretty policy in Wikipedia, or common sense from the governing laws of the countries we live in, will ''always trump that. Gwynand | Talk•Contribs 19:05, 12 May 2008 (UTC)


 * It would make no sense to remove the 3RR exemption. What we'd be saying is that our 3RR policy is more important than the reputations of the people we write about. SlimVirgin  talk| edits 19:20, 12 May 2008 (UTC)
 * It would be saying that we actually don't suddenly turn reckless, the moment we deal with people's reputations. ;-) --Kim Bruning (talk) 19:27, 12 May 2008 (UTC)
 * Kim Bruning finds itself with defamatory material one day. I go in and remove it three times, and three times it's readded by random users. If I remove defamatory material a 4th time, I should be blocked? Are you honestly suggesting 3RR, an internal policy that affects only WP users, is more important than BLP, which can affect real-world issues, including hurting real people? It's more important for me to back down after my 3rr than to continue undoing egregious BLP violations that can hurt someone's career if Google picks it up, or that can get us sued? Lawrence Cohen  §  t / e  19:32, 12 May 2008 (UTC)
 * Yup, because you're going about it all wrong. Your fine edit war is making it hard for others to maintain the page against further BLP-vios. How about blocking the adders, or semi-protecting the page? Maybe others have had better experiences though. Have you encountered the situation you describe often? Why not add some examples to the /log ? :-) --Kim Bruning (talk) 19:41, 12 May 2008 (UTC)


 * I've encountered non-admins many times having to hold the fort by violating 3RR while they wait for an admin either to block the user adding the defamation or protect the page. People are advised to contact an admin asap rather than revert endlessly, but in the meantime, 3RR does not apply. SlimVirgin  talk| edits 20:05, 12 May 2008 (UTC)
 * Ah! <gentle_sarcasm>"wikipedia has enough admins"</gentle_sarcasm> :-). See! If you just keep on asking, and continue to assume good faith, at some point people will provide useful information.


 * What you're saying coincides well with what Gwynand is saying below. Can you provide some data for /log too? --Kim Bruning (talk) 20:22, 12 May 2008 (UTC)


 * For the avoidance of doubt, I for one certainly don't want to see more blocking. I'd far rather find ways to turn problematic editors into productive ones.  There seem to be two distinct scenarios that people have in mind for exploitation of this exception: one is the experienced editor (possibly an admin) facing a tag team of inserters; the other is a newbie/anon editor (likely with a strong connection to the subject) who removes content clumsily, and who might otherwise face rapid blocking.
 * In the former case (and no disrespect intended to the examples given), I can't shake the notion that help is only as far away as the appropriate noticeboard, and empowering a single editor to wage edit war does not deal nearly as well with the possibility of error. We all make mistakes after all; when everyone disagrees with you is precisely the right time to bring in fresh eyes.  I would certainly take into account any attempts to bring in help when assessing a 3RR report that might fall under the BLP exception.
 * In the latter case, the newbie is likely unaware of both 3RR and BLP, so the effect would have to be on the other editors that interact with them. Allowing the newbie to blank repeatedly while blocking those who break 3RR in restoring the page (reverting "simple and obvious vandalism") may satisfy some short-term BLP goal, but is not good for the long-term health of either the article or our base of editors.  Perhaps we could avoid getting into these situations if we augmented uw-delete1, uw-coi, uw-3rr and MediaWiki:Blockedtext to point people in this situation at WP:BLPN or WP:OTRS sooner. Bovlb (talk) 17:04, 12 May 2008 (UTC)


 * Gwynand. I love your attitude! We need more editors like you. :-) I too would ignore the 3 revert rule in a heartbeat, if I thought it would improve the encyclopedia. (In fact, I used to hate it as a limitation on what I could do). But can you honestly say you've experienced situations where going past maybe at most 2 reverts on a day was actually a good idea? I think I've maybe deliberately done it only once or twice, and in at least one case, people later invented better ways of doing what I did with reverts before.
 * Rules and advice tend to alter the dynamics of interactions between users. In the case of BLP, I think that making an exception to the 3RR is more likely to lead to an old fashioned adversarial "wikiwar" atmosphere, exactly at the time and place where such an atmosphere will do the most damage.
 * So that's why I asked people to provide evidence that I'm wrong. I'd love to be wrong! But so far, no one really has come forward all week. I'm kind of disappointed. --Kim Bruning (talk) 19:26, 12 May 2008 (UTC) Well, except for Lar, so far. Maybe he'll take a look this coming week, at least.
 * Kim, one thing I can state about you for a fact, you are amazing when it comes to keeping things civil. As for the evidence, I do see you repeatedly asking for it and no one really coughing up diffs. I'm a little confused with this, as I think there is plenty of evidence. I watch the article David Archuleta. Let me give you two examples.
 * 1. Somewhat recently, someone wanted to add an ethnic category to the intro sentence, based on research of his family, but not that term ever being used in any source. I didn't think it was defamatory, rather just incorrect. I reverted I believe once, when I was reverted back, I went to RfC. Eventually neutral editors came and agreed that we really shouldn't have the ethnic category without a source. Falling behind the 3RR rule in this case would have been disingenuine, although I do see the flaw in the current policy in that someone thinks they can. "Honduran American" or "Has Honduran ancestry" is an editorial content dispute, but far from defamation. On the other hand...
 * 2. There are random internet rumors, without even a basis in facts, regarding, let's say, what kind of person Archuleta might want to go on a date with. It's come up on the talk page, and it the controversy section of the article. I believe there have been several good-faith reverts of this information in the past that have broken the 3RR rule, and for good reason. This article is watched by many, and in cases where defamation is not occuring, I'll partake in talk page discussion, maybe warning the editor about false info, or simply letting another editor revert. But I, and seemingly others, wont do this when there is a falsely sourced section, especially about a 17 year old, stating controversy over their sexuality when there is none. It is especially bad when it isn't clear vandalism, because those that come to the page might take it as fact, and seeing those mini source numbers (that of course lead to some defamatory blog), they really might believe it. In this case, revert, revert, revert, revert, do it as many times as neccesary. Also take other steps of warning, engaging on the talk page, etc, but continue to revert that section right out of the article. This is a situation where 3RR is not only the best option, it's the only immediate option of keeping non-vandalism defamtory material out of an article. Gwynand | Talk•Contribs 19:44, 12 May 2008 (UTC)


 * Exactly. Anyone removing problematical BLP content can run right past 3rr without a glance. Surely no one is disputing that? And since that's true (and it is), the ability to protect BLPs should not be limited to those editors who are confident enough in IAR or know the spirit of BLP or who are secure enough to make an edit that looks like a policy violation. It should be included here so that editors feel on solid ground when protecting BLPs. They don't need it listed here, but for new editors or those that don't work with BLPs much, it's nice to have it spelled out so they don't have to worry as they hover over the rollback button deciding about a 3rd revert. RxS (talk) 19:59, 12 May 2008 (UTC)
 * *raises hand* Disputer #1, present and accounted for. I actually thought this would be a minor check, until people failed to provide evidence. Gwynand is actually the first to do so. ^^;; --Kim Bruning (talk) 20:18, 12 May 2008 (UTC)


 * Aha! You're actually applying what's called soft protection. Interesting. Now why couldn't anyone provide such an example last week? :-/ Oh well. At any rate, that leads to an interesting path towards a useful solution/compromise. --Kim Bruning (talk) 20:18, 12 May 2008 (UTC)
 * I don't think it's uncontroversial that blowing past 3RR without a glance is a good thing for an experienced editor or admin to do for several reasons:
 * As Kim says above, it sets a bad example and promotes a wikiwar atmosphere.
 * There are adequate ways (e.g. BLP/N, protection) to deal with the situation that don't involve one person deciding that he cannot possibly be wrong. (We can all be wrong sometimes.)
 * Those other ways may provide less interference to the unrelated development of the article.
 * A good faith inserter of BLP violations is more likely to question his own actions if faced by multiple people reverting than by just one person who is quite likely getting terser and tetchier with every revert. Involving more people may result in better ways to explain what the problem is, or the development of a compromise.  Who knows, someone may even find a reliable source.
 * I came to this discussion believing that we needed to impose a BLP policy with as many teeth as possible. Reading through Kim's arguments and the lack of evidence in response, I find to my surprise that I am changing my mind. Bovlb (talk) 20:33, 12 May 2008 (UTC)


 * It's commonly understood that 3rr's in this context happen along with other actions (AN/I, BLP notice board etc) and not in place of it. I've said that, others have said that. It's to be used along side of, and during an effort to find a more permanent solution. One person won't be deciding that they cannot possibly be wrong, they are just performing a holding action while others are notified of the issue. And just so we're on the same page, BLP policy states thatUnsourced or poorly sourced contentious material about living persons... should be removed immediately and without waiting for discussion. As with anything else, thought should be put into any rollback, including this type, and communication with whoever you're rolling back is expected. No one is claiming that mindlessly reverting someone indefinitely is a good thing. But frankly speaking, whatever this page says about it, 3rr will never apply to removing BLP violations. RxS (talk) 20:54, 12 May 2008 (UTC)
 * I confess that I have not cross-checked every example cited against ANI and BLPN archives, but I just don't feel confident that everyone who might rely on this exception is following your excellent precept. If I thought that no-one would revert for the fourth time without first filing a BLPN report, then I would be far happier about things.  By the way, I don't mean to be contentious, but your last comment is the reverse of my observation.  The exception is documented on this page, but the evidence seems to be that 3RR violations often result in blocks even when the exception might be argued.  Of the four such cases in the table above, three resulted in blocks.  I note that the fourth case is the only one where the editor concerned was an administrator, which I hope you'd agree does not give the best possible appearance. Bovlb (talk) 21:12, 12 May 2008 (UTC)
 * But then that only points to a broken system of warning. Admins noticing an editor properly reverted BLP violations beyond 3RR, but did not place a notice on the proper noticeboards, should (hopefully gently) warn the editor in question that they only completed one step of the process.  It doesn't mean the rule itself is broken. This would be like saying the second step after committing an act of self-defence that resulted in a death is calling the police. But if someone was shaken up and didn't phone the police, then we should throw out the ruling of self-defence, regardless of the fact that it wasn't murder.  The subsequent actions of violation of process should have no bearing on the law/rule itself. Or perhaps better stated, violation of process should not influence retention of law. Faith (talk) 07:26, 13 May 2008 (UTC)

Straw Poll on removing the 3RR exemption from BLP
I think that a straw poll, to see where people stand, would be useful at this point.

The specific proposal is to remove the following passage from the BLP policy: "The three-revert rule does not apply to such removals, though editors are advised to seek help from an administrator or at the BLP noticeboard if they find themselves violating 3RR, rather than dealing with the situation alone. Content may be re-inserted only if it conforms to this policy".

Please state briefly your reasons for supporting or opposing the proposal. Thanks, Nsk92 (talk) 18:46, 12 May 2008 (UTC)
 * On careful re-reading, I see what's confusing people. Such a proposal does not actually exist. There exists an observation that the 3RR exempt does not have consensus. That's something else. I was wondering why I needed to correct so many people. No one is proposing anything. Can we close the poll now? --Kim Bruning (talk) 16:40, 13 May 2008 (UTC)
 * Absolutely and categorically NOT! I disagree with your interpretation of the the actual dispute. If you like, you can consider this to be my proposal and I could remove my vote. But closing it after just a day, when people are still voting? Certainly not! Nsk92 (talk) 16:45, 13 May 2008 (UTC)
 * You're proposing something, but opposing it? That's typically considered disruptive... maybe I don't understand your intent.  You have me confused now. What are you trying to achieve? --Kim Bruning (talk) 16:50, 13 May 2008 (UTC)
 * I am trying to see if there is a current consensus for removing the 3RR exception from the BLP policy. If you think that this question is irrelevant, you can say so and not participate in the poll. Even a simple AfD is supposed run for at least 5 days before closing. We are talking here about a major policy change, which requires no less. This is a policy discussion, not an express train. I take a strong exception to your suggestion that I am being disruptive. Nsk92 (talk) 17:12, 13 May 2008 (UTC)
 * Can you please tell me who is trying to make a major change to policy?  --Kim Bruning (talk) 17:29, 13 May 2008 (UTC) Note that polling is not a suicide pact. It would be rather silly to keep on going asking a question that no one needs answered. No one will be enlightened :-). We can certainly stop at any time (in fact around half the respondents actually already ~state that that's what we should do for now. We can always come back later.)
 * "Can you please tell me who is trying to make a major change to policy?". Well, you are. Here and here. Nsk92 (talk) 22:53, 13 May 2008 (UTC)
 * See my discussion with Faith, in the "oppose" section. --Kim Bruning (talk) 00:03, 14 May 2008 (UTC)

I currently Support removing 3rr exempt at this time.

 * 1) Strong Support. NPOV ought to mean that edits that portray an article subject 'positively' are not privileged over edits that portray an article subject "negatively".  See any definition of "neutral"!  As it stands now, a user can't revert a totally unsourced and spurious "positive" edit without being bound by 3RR, but a sourced "negative" edit can be reverted by citing WP:BLP! (and calling the sourcing "poor"... no matter how "good" a source is, there is always SOMEONE who considers the source "poor")Bdell555 (talk) 02:55, 13 May 2008 (UTC)  I'd also note that WP:FAITH means starting with the assumption that BOTH sides of an edit war are acting in good faith, NOT just the deletionist side.Bdell555 (talk) 20:18, 13 May 2008 (UTC)
 * Something about what two wrongs make springs nimbly to mind. (I agree there should be more exemptions from 3RR though, but then I don't like 3RR to begin with.)--Kotniski (talk) 10:32, 13 May 2008 (UTC)
 * 1) Support. I've seen 3RR exemptions such as this one abused on several occasions by editors with axes to grind who frame their pet issue (whatever it may be) as a "BLP violation" and edit war to that end, even when they're the only one who sees it that way against multiple opposing good-faith editors and admins.  When the BLP violation is clear and obvious, there should be no problem finding multiple editors and admins to ensure it stays out and whoever keeps adding it gets blocked if necessary, and WP:IAR would be validly invoked in the unlikely event it becomes necessary for the same editor to revert more than three times in the course of doing this.  No 3RR exemption is needed, as it's likely in practice to be invoked only in highly controversial cases where it causes drama rather than quelling it. *Dan T.* (talk) 23:23, 13 May 2008 (UTC)

I am currently Opposed to removing 3rr exempt at this time.

 * 1) Oppose. Haven't yet seen any good reason for the proposal. It seems to be an attempt to make some sort of point, but I'm not clear what.--Kotniski (talk) 18:55, 12 May 2008 (UTC)
 * It's called demanding evidence and not getting it. (by the way, can you provide a diff to your contribution to the discussion on this matter? I can't see your contribution there, and would like to respond and discuss.) --Kim Bruning (talk) 18:58, 12 May 2008 (UTC)
 * 1) Oppose, See my comments above. (And I strenuously object to attempts to belittle editors with opposing views. See WP:CIVIL)--agr (talk) 19:35, 12 May 2008 (UTC)
 * That's not good. Can you show me where editors with opposing views have been belittled? Was it me? Should I apologise? --Kim Bruning (talk) 19:50, 12 May 2008 (UTC)
 * I found your response to Kotniski, immediately above, to be belittling and, yes, I think an apology would be in order.--agr (talk) 19:56, 12 May 2008 (UTC)
 * I'd say that's a stretch. Gwynand | Talk•Contribs 20:00, 12 May 2008 (UTC)
 * I'll admit to feeling some level of frustration after a while, but see below where Kotniski seems to have taken things amicably. But alright. @Kotniski: do you require an apology? --Kim Bruning (talk) 20:14, 12 May 2008 (UTC)
 * No, not at all, but if you were suggesting that people shouldn't take part in a "poll" if they haven't contributed to the preceding discussion, I would have to disagree (unless that's the custom around here, in which case I would still disagree).--Kotniski (talk) 20:30, 12 May 2008 (UTC)
 * *shakes head* You're good. :-) --Kim Bruning (talk) 20:33, 12 May 2008 (UTC)
 * 1) Oppose I oppose anything that can be seen to weaken the strength, authority, utility, or power of BLP in any perceived, actual, or literal way, shape, or form at this time.  Lawrence Cohen  §  t / e  03:11, 13 May 2008 (UTC)
 * If you oppose weakening BLP, then on the basis of evidence, shouldn't you be supporting removal of the broken text? You're not being rational or logical here. --Kim Bruning (talk) 11:43, 13 May 2008 (UTC)
 * Correct! Kim, real people that we accuse of who knows what--pedophilia, sedition, criminal acts, etc. are not going to be rational or logical when they sue, nor will they be when people google them and find massive defamation. The community seems to want a very strong handed BLP policy "written". We don't do prescriptive policy--great! But the community clearly wants one in this case, so because of that: fuck the wikiway. The community can at any time decide to change anything, and in this case, they apparently have. We're not going to suck the wikicock if the community has already decided, is the point. Or, we can call it IAR, I suppose. Maybe this will be the one prescriptive policy. Maybe there will be more. If the community decides it is so, it is, even if that slaps around some of our sacred old-timer cows. Our long-term survival and the well-being of BLP subjects is more important than our internal religious practices, which don't matter at the end of the day. To hell with a Wikipedia where our Wikiway is more important than doing the right thing. Lawrence Cohen  §  t / e  15:43, 13 May 2008 (UTC)
 * I'm correct that you're not being rational or logical? :-( --Kim Bruning (talk) 16:30, 13 May 2008 (UTC)
 * Some things so far transcend our wikiway of prescriptive vs descriptive that you can't even see them from so high up above. Like BLP. Lawrence Cohen  §  t / e  17:10, 13 May 2008 (UTC)
 * I can't parse that. Can you please explain. How do you think your current position improves BLP, or the wikipedia? --Kim Bruning (talk) 17:14, 13 May 2008 (UTC)
 * Again: editors need to be assured that if they shoot past 3rr to remove BLP violations, that they will be in NO risk of sanction. If users get sanctioned for removing BLP violations, the person sanctioning them is wrong--not the policy. Any attempt to remove that guarantee from is wrong because it takes teeth out of BLP. I am 1000% opposed to ANY attempt to weaken BLP in any literal or perceived way shape or form. How else can I say that? 3RR is irrelevant in the face of BLP. We need to codify that so no stupid people can misintepret. Period. Lawrence Cohen  §  t / e  17:36, 13 May 2008 (UTC)
 * Right, several people are saying that, but as far as we've been able to ascertain, they've been saying that on faith, rather than based on solid evidence that 3RR exempt actually works that way. Can you help us out there? --Kim Bruning (talk) 19:47, 13 May 2008 (UTC)
 * I can't help you. If consensus is that 3rr and BLP work as I described, thats what the accepted use is. If some people don't do it right, that doesn't mean the accepted policy is broken: it means the few people are doing it wrong. Rather than try to fix what isn't broken, the BLP policy, you should redouble your efforts to fix the users who are clueless. Lawrence Cohen  §  t / e  20:11, 13 May 2008 (UTC)
 * Consensus in the field seems to be that it does not work that way. If you think that that is because everyone in the field is clueless, and only visitors to this page have a clue ... wait a minute... are you advocating tyranny of the minority again? --Kim Bruning (talk) 20:18, 13 May 2008 (UTC)
 * Kim, I think you live on another planet sometimes (don't take that as a personal attack). As far as I'm concerned, as their is no consensus to remove the 3RR exemption, any further discussion is fruitless here. Lar, if you're reading this, I'm not dimissing Kim, but Kim has this squarely backwards, and I will not spend the next week debating policy wonkery. Consensus supports including 3RR exemption. From now on, as far as I'm concerned, Kim has to demonstrate consensus for it's removal. This policy is prescriptive, on purpose, and nothing is going to change that. I urge Kim to spend his policy karma more fruitfully on battles that mean something. I'm done here, and my !vote in the consensus battle is set in stone, and no one is going to change it, and I will vehemently object to anyone attempting later to deprecate the weight of my opinion if I am unwilling to further debate it. Discussion will never, ever, never allow for the weakening of BLP without overwhelming consensus demonstrated in support of it. I'm sorry. Kim, you're misfiring widly here. I will not respond again to this thread, my vote is cast, and any attempt, again, if someone does it, to minimize my !vote due to my unwilling to further debate it is out of line. Sorry! Lawrence Cohen  §  t / e  20:42, 13 May 2008 (UTC)
 * Well, you are acting in exactly the way you claim you are against at Governance reform. You reject global consensus in favor of local consensus, you reject consensus and compromise entirely in fact: "my !vote is set in stone", and if it were up to you, it would be impossible to update the policy, no matter what evidence or information is brought forward.  --Kim Bruning (talk) 20:52, 13 May 2008 (UTC)
 * Not at all. Demonstrate for me an overwhelming number of users saying "I support removal of 3RR exemption", far more than say "Lets keep it", and you'll have consensus. You simply cannot arbitrarily gut a major policy with real world implications. Ever. You have the weight of consensus squarely backwards. And this is totally in-line with my views at governence reform: no binding policy change with overwhelming support for the change. Add in my (some have called too-extreme) views on BLP, and I've been absolutely as I've always been in this discussion. If I had my way and got to set one binding policy change that everyone had to obey, I'd take BLP further and apply 1rr for all users to include unsourced or bad BLP info, to make it even more likely that BLP abusers be blocked. But thats just me. If you look at what I've argued in favor of at governence reform more closely, all my desire was to see that consensus alone--not one user presenting some holy grail of information--decide. If the masses decide that bit of magic information is relevant, it would sway them. If not, not. You on the other hand are doing the same thing you did at governence reform, micro-arguing any points that don't align with your own views for change. You seem to be unwilling to stop arguing, in fact, any viewpoint that does not follow with your view, and you insist on getting in the last word at all times to try to sway my views or perhaps others' perceptions of my views to your own ends. Sorry--that's just not on. Users here are certainly entitled to binding views, to their own mind. I believe what I believe, and I'm 100% entitled to that belief. Unfortunately, it appears that only one other person currently believes in your views on this matter. Again, this appears to be a dead non-starter at this time. Again, if you want this changed, YOU have to demonstrate consensus in support of that change: and that is 100% in alignment with every firm stance I've taken in every major debate I've been involved with here. Consensus needs to put up, or shut up, as we say here in America. Lawrence Cohen  §  t / e  21:10, 13 May 2008 (UTC)
 * Sure. There's a large number of people (over 15) who "voted through action" . They actually took actions that were not in line with what's on this page, so they clearly don't support it. We managed to interview one today. He said for the record that he does not support 3rr exempt . So we're definitely looking at the right group of people. This information is out there on the wiki for anyone who is willing to look. I've currently asked some people to help increase the sample size, but with current information, it doesn't look good for 3rr exempt.
 * --Kim Bruning (talk) 21:39, 13 May 2008 (UTC) Note that consensus is not the majority, nor is it your personal will above everyone else. Consensus is where you cooperate with everyone and convince them with solid well reasoned arguments, and also where you are willing to compromise and let others convince you too.
 * Um... No Kim, the " large number of people (over 15)" (actually it's under 15 according to that table but I won't quibble) didn't necessarily act to "not support" the policy. They may have just acted out of ignorance or not thinking things through. You would need to do root cause analysis to know whether they do support it or not. Remember, when if I run a red light, it doesn't necessarily mean I don't support the concept of red lights. I may have supported the concept but forgot it was there, or felt the circumstances justified the run (as it happens I have some very recent personal experience with deliberately running a very large number of red lights for what I felt was a very justifiable reason, so maybe that's where this analogy is coming from... but I was careful about it, and I generally don't do that because I support red lights in general). The true measure of my support for red lights is not the relative few that I run, but the thousands I stop for. The true measure of my support for emergency vehicles being immune to red lights is not the wrecks I witness when someone gets in their way, it's the mass of cars that typically get out of the way and yield as soom as they see them coming. Hence, the true measure of support for the 3RR BLP exemption among the general populace is how infrequently it's used. ++Lar: t/c 00:51, 14 May 2008 (UTC)
 * "Hence, the true measure of support for the 3RR BLP exemption among the general populace is how infrequently it's used.", would lead us to conclude that if there are 0 uses of the 3RR BLP exemption, that it is wildly successful, and if there are hundreds of uses, it is not successful at all. Correct? --Kim Bruning (talk) 04:28, 14 May 2008 (UTC)
 * No. It's more complex than that. If out of the population of all BLP actions (there are tens if not hundreds of thousands of them a month, I would estimate) the 3RR BLP exemption is used 0 times, it's wildly successful. If it is used, correctly, hundreds of times a month, it is STILL wildly successful. If it is used correctly, hundreds of times a month, and also used incorrectly, hundreds of times a month, there is an issue with correcting the incorrect use of it, but it is STILL wildly successful. (a 1% error rate is not bad, when we consider that somewhere between 1 to 20% of all BLP articles are likely, at any given time, to have BLP related problems) If on the other hand, a large majority of the tens, if not hundreds of thousands of BLP actions a month ended up getting overturned, or the wrong people blocked (a large majority!) then I'd say maybe you have a case. Heck, if 10% of them are wrong, I'll give you that you have a case. But there is no way there are 10% wrong, because most BLP actions don't even result in one revert. Not even one. I have dozens of actions and no failures, remember. How many actions do you have that are BLP related? Oddly I don't see any of yours in the /log. ++Lar: t/c 04:41, 14 May 2008 (UTC)
 * Consensus, if many many people oppose something or support something, is consensus. You don't buy into that--most people do. Your view of consensus, god bless ya, is the minority I think. I don't think 3RR is going away, and I've asked Jimmy to weigh in as well. Did you canvass off-wiki? I hope not. Canvassing is very bad without naked transparency. Lawrence Cohen  §  t / e  22:01, 13 May 2008 (UTC)
 * Eh? No, the requests for more information are all over this page. The results are supposed to go to /log. --Kim Bruning (talk) 22:54, 13 May 2008 (UTC)
 * 1) Oppose Blocking somebody for re-re-re-removing unsourced (or badly sourced) contentious BLP material, has to be one of the most foolish things an admin could do.  While people can argue there are better methods of fixing BLP problems, nobody has argued the benefit of blocking as a *standard* means of response.  3RR is rather unusual for a Wikipedia rule, in that admins regularlly impose blocks based on the letter of the rule, with little interest in the article content, except for what we specify as the allowed exceptions.  It seems those wanting to remove the 3RR BLP exception assume admins will still use good judgment, and not automatically block people for appropriate re-re-re-removals.  But, this ignores the fact, that admins have always blocked for 3RR violations even when they agreed with the edits of the blocked editor.  3RR is operated with a blindfold, and there are several cases, where we need to force the admin, to take off their blindfold.  If we can make an exception to 3RR to protect "Gdansk", surely we can also protect BLPs.   --Rob (talk) 06:24, 13 May 2008 (UTC)
 * Subtly, I don't want to remove the 3rr exemption from policy. I'm simply stating that it doesn't appear to be policy (after giving people a week to show that it is). If you'd like admins to change their behavior, adding or removing text from a policy page has no effect (as I think has been adequately shown). You do make some useful points about 3RR being operated with a blindfold, though. Perhaps we could discuss methods to reduce that? --Kim Bruning (talk) 11:50, 13 May 2008 (UTC)
 * I think it's time to put the "gave people a week" thing out to pasture. Not calling anyone negligent but I imagine I'm not the only person on this wiki who doesn't necessarily watch every page that might possibly be relevant every day, or even every week. A week went by, there was not much traffic at all. We get that. Absence of evidence is not evidence of absence. So drop that refrain, ok? We're now in a different phase, one in which your previous conclusion is called into question, and your previous sample set is called into question as well. I would agree that in the general case you don't change behaviour by changing policy pages, because in the general case policy is descriptive, but in this specific case, we've had arbcom rulings that were prescriptive, so the general case doesn't necessarily apply. And repeating the assertion that the exemption doesn't work won't make it necessarily true, no matter how many times you assert it. Further, as more people are bringing forth examples, it seems that the original sample set is less and less supportable as an accurate, statistically valid sample of the correct population. Perhaps what we really need to discuss is reform at the 3RR noticeboard to cut down the number of bad tickets being handed out? ++Lar: t/c 13:09, 13 May 2008 (UTC)
 * All I've ever asked for is evidence. Now there's the suggestion that some folks do some logging, and we can give it another week before anyone does anything too drastic. We'll see where we end up at. Now it's there, now an adequate number of people is aware, so the same excuses can't apply twice. Alright? --Kim Bruning (talk) 14:11, 13 May 2008 (UTC) btw, I've deliberately only been asking for evidence in favor, so that's what we're getting, though it's still somewhat thin.
 * If I'm not misreading things here, others are being asked to prove that the system works as it stands now. Is that correct?  If so, shouldn't it be the one stating the system is flawed showing practical examples of why it is flawed?  Are people being blocked despite the clause?  Theorising is fine, but I haven't seen an overwhelming amount of logging and evidence showing the exemption is no longer viable to effect any need for change from an established policy. Faith (talk) 17:04, 13 May 2008 (UTC)
 * I'm actually neutral on the topic believe it or not, and I've been working off of current evidence. ( as I've also just posted below). Originally, I just posted a short challenge here on this page, because I've had some trouble with edit warring on a BLP article while the subject of the article was looking on. (ooops). I sort of hoped it'd take all of 7 minutes for someone to come up with counter examples. Well, someone did come up with many examples where the exemption didn't work out... but there's still only very few examples where it did work out. I'm hoping that getting people to make entries in the log might help them defend their position more easily. --Kim Bruning (talk) 17:21, 13 May 2008 (UTC)
 * 1) Oppose per Lawrence, who stated the case clearer than I would have Faith (talk) 07:32, 13 May 2008 (UTC)
 * FYI, Lawrence states oppose, but based on similar arguments, I'm actually under neutral (leaning towards support) --Kim Bruning (talk) 11:50, 13 May 2008 (UTC)
 * As is your right, but IMO, that's an untenable position. BLP removal is supported by ArbCom (to the point of allowing Administrators complete content removal if necessary), by Jimbo ("Real people are involved, and they can be hurt by your words. We are not tabloid journalism, we are an encyclopedia" & "should be removed, aggressively, unless it can be sourced. This is true of all information, but it is particularly true of negative information about living persons"), and by a consensus that appears to have been place for a while now and for good reason (all the previous and current violations are proof enough). Again, as I said above, violation of process should not influence retention of law. If not all the proper steps are being followed, that's a policing issue, not a broken policy issue. You don't throw out the law because the police are ineffective at enforcing all the letters of it. You get better police, a stronger police force, etc. When an admin does amble along, the vandal can be blocked and the article can be protected, but until that time, immediate removal of real BLP issues by editors without Administrator "teeth" is a viable alternative to WP being sued for sullying someone's reputation. Editors should not be penalised for making WP safer for the subjects of the biographies, and for WP itself else it get closed down by too many lawsuits. Jimbo's "aggressively" settles the matter in my mind. And I see no consensus that speaks toward changing it. Faith (talk) 15:06, 13 May 2008 (UTC)
 * But this discussion is about examining one particular mechanism for removal (3rr exempt), which at the moment appears to be flawed. The discussion is not about removal itself, which can be achieved by several means. No one has come forward in opposition to removal of bad materials in BLP situations, in fact. I think we are in agreement that bad material should be removed Agressively. My aim is to ensure that this happens as efficiently as possible. What's your aim?   --Kim Bruning (talk) 15:30, 13 May 2008 (UTC) Note that wikipedia has no laws, and no police, and no parliament, and is not planning to introduce any of those things.  Instead (to make a long story short) there's just us; so to achieve aims, we need to form consensus and cooperate.
 * My aim on this particular issue? Not WP:STICK :). In general? To improve the wiki (which is the obvious answer, but still factual). "Policing" was a metaphor for administrative action against disruptive editors; Admin make those judgements all the time, and if some are falling down on not warning people who vio 3RR for good BLP reasons, but who fail to present the information on BLP/N for whatever reason (has anyone shown this is a pattern?), it's not the fault of the policy unless it doesn't inform the editor in a clear manner that is a step they need to take.  I guess what it boils down to is, can you show practical examples, since you are pressing for the change, of how the system is flawed enough to no longer be viable? Faith (talk) 16:55, 13 May 2008 (UTC)
 * Re-iterating that I am not pursuing any other problems with the BLP system at this time. I therefore support the rest of the BLP system in general at this time. I am not pressing for any change to the system as applied in the field at this point in time. My one and only interest in this discussion to to ensure that the policy page reflects consensus, as that is the requirement.   The 2 criteria for 3rr exempt to be considered policy are: 1. Is the 3RR exempt mechanism employed in the field? 2. Is the 3RR exempt mechanism a best practice? Data on the particular 3rr exempt mechanism is at: . More data is requested at /log (empty at time of writing).  Feel free to draw your own conclusions based on evidence so far, but please do share them! :-)  --Kim Bruning (talk) 17:13, 13 May 2008 (UTC)
 * (1)The sampling is too small, (2)The sampling doesn't take into consideration problems which took place before the exemption was added (i.e., there is no comparison measure), (3)The data is not representative of the whole of the wiki, only of cases found in one area (i.e., only if they were reported on the noticeboard). Shall I continue? --Faith (talk) 21:07, 13 May 2008 (UTC)
 * You're doing fine. :-) I wasn't planning on writing a paper on this. If there are a bunch of clear data points that point in a different direction, then that's good enough for now already. Can you provide some? --Kim Bruning (talk) 03:13, 14 May 2008 (UTC)
 * (haha) I've replied on this point elsewhere, so I won't repeat myself except to say that your attempts to change the policy (by edit) have to be supported by evidence that the exemption is no longer viable. You haven't made that case; if anything, my opposition to this change is even stronger than when I first started responding here based on the lack of strong conviction for change, and the lack of evidence that the change is necessary, while the need for the exemption is obvious to the majority weighing in. --Faith (talk) 03:47, 14 May 2008 (UTC)
 * 1) Lean towards opposed. I don't have a strong opinion on this issue, but from what I have read here so far, I am leaning towards opposed. First, I have not seen substantial evidence that the 3RR exemption in BLP is creating serious problems in terms of edit warring, etc (or at least some substantive evidence that such edit warring would go down if the 3RR exemption is removed). I tend to view the 3RR exemption clause in BLP as primarily a preventive tool, which strongly discourages people from adding poorly sources and contentious BLP material to WP articles in the first place. As such, the exemption might well prevent more edit wars than it creates. Second, it seems that dealing with the BLP issues is one of the greatest weakenesses of Wikipedia. Much of the ArbCom stuff is about that and, from what I understand, most legal threats that WMF receives also arise from BLP issues. So it is probably better to be more agressive, not less, in removing contentious BLP material. But, if there is substantial evidence that the 3RR BLP exemption is creating significant problems, perhaps other tools of achieving the same goal should be explored. My 2c. Nsk92 (talk) 15:02, 13 May 2008 (UTC)
 * The "serious problem" is not edit warring but the removal of sourced NPOV edits. Edits can be 100% consistent with WP:RS and WP:NPOV and get removed anyway (and have been removed) by a user citing WP:BLP (why does WP:BLP exist if NPOV and RS standards are to be uniformly applied to all articles?).  The exemption undoubtedly DOES reduce edit warring, simply because it sends the message to one side that resistence is futile.  You'd get the same reduction in edit warring if you put your thumb on the other side of the scale, such that inclusionists were not bound by 3RR and deletionists were .Bdell555 (talk) 18:38, 13 May 2008 (UTC)
 * This argument is indicating something not part of the policy, though. If it's sourced NPOV, then it shouldn't be asserted under the exemption because then it's a content dispute. Someone falsely claiming it does to the point to edit warring should be sanctioned equally. The policy protects BLP violations being added repeatedly to the point of 3RR to avoid potentially libelous, unsourced (or poorly sourced—i.e., blogs, yellow journalism, "heard it some where", etc. that do not meet WP:RS—from remaining in the bios or other articles. The 3RR BLP exemption, properly used, protects the editor trying to protect the project as a whole by protecting the article and the subject of the article. --Faith (talk) 21:03, 13 May 2008 (UTC)
 * It IS part of the policy. The policy says "poorly" sourced without defining what "poorly" means.  It follows that there are no limits to what can be considered "poor".  Again, if applying WP:RS is enough to solve the problem, why does WP:BLP exist?Bdell555 (talk) 02:14, 15 May 2008 (UTC)
 * 1) Oppose this change, it goes in the wrong direction with our BLP problems. People need to be free to remove BLP violations without the fear of being blocked for it. No real reason has been given for this change other than "nobody has demonstrated we should not change". 1  !=  2  16:37, 13 May 2008 (UTC)
 * Hmm, the poll question seems to have been stated incorrectly, which is why I'm ending up correcting everyone on the same issue each time, it looks like. :-/ There's no proposal to change policy. There never has been. Alright? --Kim Bruning (talk) 16:44, 13 May 2008 (UTC)
 * What was this edit intended to signify, then? Removal of a portion of the policy seems to be a clear indication that someone wanted to change the policy. --Faith (talk) 17:42, 13 May 2008 (UTC)
 * Come on Kim. Clearly you are attempting to encourage a change. You changed the policy and when reverted you came here and attempted to convince others of the merit of this change. This is all well and good, but I don't see any grounds to close the poll. Perhaps it is true that you are not "proposing" it, more it seems you are saying that it is just so. Regardless a discussion leading to consensus is the way to go about it which is what we are doing.
 * While it does seem rather clear at this point that this change is opposed by the community, I say we let the discussion/poll continue so as to be more certain of consensus. 1  !=  2  17:50, 13 May 2008 (UTC)
 * It is clear right now that a small subset of the community is not convinced of anything yet. I did apply 1 step of BRD to 2 different policy pages, but policy wasn't changed by that, obviously. --Kim Bruning (talk) 18:18, 13 May 2008 (UTC)
 * @Faith: You don't change policy by editing policy pages. You edit policy pages to reflect (what you think is) the current policy (see the policy tag, which actually states that requirement). --Kim Bruning (talk) 19:50, 13 May 2008 (UTC)
 * Semantics. Edits indicate attempt to change existing text of the policy through deletion of an exemption. You might have an argument if you were merely copyediting, but those two attempts deleted policy, and ergo, were attempts at changing policy. --Faith (talk) 20:58, 13 May 2008 (UTC)
 * Surprisingly to some, it actually has to do with how the consensus process works. As Bovlb explains beautifully at "", policy pages describe what the current consensus (and thus policy) is on the wiki. Policy pages are intended to reflect consensus, but they are not the policy/consensus themselves ("the map is not the terrain"). So you cannot actually alter policy simply by editing a policy page. You do edit policy pages to keep them in line with (what you perceive to be) current policy. There's more depth to the consensus process yet, and there's more depth to why I made those specific edits, but that's a good start, I think. If you want a positive explanation: I made the edits to initiate the process of updating documentation. --Kim Bruning (talk) 21:27, 13 May 2008 (UTC)
 * I call BS. While the description is accurate for descriptive policy, it only holds for BLP if BLP policy actually is descriptive. Others assert it's prescriptive. In the prescriptive case, editing the policy page is changing the policy itself, not changing the map to match the terrain. That's been pointed out to you already. Multiple times. You can't blithely keep blowing past points that are in contention to make assertions this way. It won't work. Stop, please. ++Lar: t/c 04:26, 14 May 2008 (UTC)
 * @Kim: Are you trying to adjust perception of reality to match reality?? Society will not tolerate such heresy! 1  !=  2  20:14, 13 May 2008 (UTC) My smart ass comment for the day
 * Every day dude! I think you get me now :-) So... when do we organize the burning at the stake? --Kim Bruning (talk) 20:24, 13 May 2008 (UTC) "First you will be baked, and then there will be cake"
 * 1) No Way. Wikipedia needs to eradicate BLP problems not institute process over doing the right thing? Spartaz Humbug! 21:29, 13 May 2008 (UTC)
 * Exactly. So why are you opposed to eradicating the institution of process over doing the the right thing? ;-) --Kim Bruning (talk) 22:55, 13 May 2008 (UTC)
 * I think the source of confusion is that most people don't see this exemption as "institution of process over doing the the right thing", rather the exemption provides allows the avoidance of "institution of process over doing the the right thing". If you interpret other people's comments with your own assumptions they are bound to appear contradictory. 1  !=  2  23:01, 13 May 2008 (UTC)
 * 1) Oppose I'm puzzled by some of the presentation of this. The exemption is not a carte blanche for unlimited reverts. Rather, it is a removal of the "bright line" of practically automatic blocking for 3RR. BLP has, by consensus, a number of variations from normal procedure. There are two conflicting desirables: the prompt removal of defamatory material in BLPs, and the avoidance of edit warring. Edit warring is *never* a long-term solution to content issues, even with BLPs. Someone using reverts to keep material they think is negative out of a BLP cannot use a "3RR exemption" to indefinitely justify reverts, especially of a registered editor. We might word the exemption better to make this more clear. It is not "unlimited reverts," rather, it is "not strictly limited to three, pending review." It can take some time to get admin assistance to protect an article, when necessary, and to seek and find consensus on whether or not to include possibly negative material. Pending involvement of editors with buttons, or broader consensus, 3RR shouldn't be the limit. But admins can still use discretion to block if it seems to them that an editor is attempting to abuse the "exemption." WP:IAR still stands, and what the "exemption" actually does is *remove* a rule -- in a special case --, not create one. This exemption should be stated under 3RR policy, really. --Abd (talk) 23:48, 13 May 2008 (UTC)
 * 2) Oppose. I was blocked for removing a controversial category on a living person (case #1 listed in WT:BLP). This particular category was clearly inappropriate for categorizing living persons: it did not contain any living person in the past; it was not used in any of the non-English Wikipedia pages for the same article; and it could easily be thought to suggest the person has a poor reputation. Prior to the decision to block me and subsequent denial of my appeal, the contentious category had already been removed by another editor, and has not been resurrected ever since, even by its original proponents. The justification for blocking me and denying my appeal was as shaky as the use of this category to categorize living people. But I think the way forward is not to lower the bar and change BLP policy to accommodate administrators' current practices and presumptions, but rather to educate them to better implement Wikipedia's policies.  --Zhenqinli (talk) 02:04, 14 May 2008 (UTC)
 * Thank you so much. That actually covers much of what I'm saying! --Kim Bruning (talk) 02:39, 14 May 2008 (UTC)
 * I can only shake my head at a reply like this, Kim. How is that anything close to what you have been saying?  Zhenquinli opposes removing the clause and "not lower[ing] the bar" (while you removed it without providing the necessary data to show why that part of the policy should be eradicated), and then shows how poor implementation needs to result in better education (while, again, you advocated removal with your actions, if not your words).  This is exactly the point I made above about the need for better "policing", where administrators are educated properly on the existing policy, not the policy is hacked up to what an editor thinks it should say because some people aren't applying it as it is written. It's replies like this that make me think I stepped into an episode of the Twilight Zone, where black is said to be white and vice-versa. It's just "heads you win, tails the opposition loses", and that's really not an effective way to continue discussion. --Faith (talk) 03:09, 14 May 2008 (UTC)
 * (EC) See my response to Kim here, and on which Lar also has corrected him. Kim is now trying to reverse the burden of consensus on those he disagrees with, and is working to redefine the views and perceptions of those he disagrees with as the valid authoritative weight of consensus has summarily (so far) rejected what he is selling here. The more that editors see this kind of bold manipulation of consensus, the more we need to call out those who do it, so that everything is above board. Consensus and discussion are not manipulation, and I'm actually fairly upset with Kim at this point for attempting to completely manipulate and subvert this discussion to his own ends. Lawrence Cohen  §  t / e  03:27, 14 May 2008 (UTC)
 * Ummmm... to which end would that be? --Kim Bruning (talk) 03:33, 14 May 2008 (UTC)
 * You're losing the consensus struggle, and challenging any and all views that run counter to your goal to remove 3rr exemption from the policy. People only vigorously challenge all people that oppose them in these debates (you see it all the time on AFD) when they are adamant about not losing. You're coaching and gaming the debate, by trying to exhaust everyone with demands for evidence and data, rather than letting their opinions stand, and have repeatedly attempted to subvert raw consensus by reversing the burden of "proof" to those that disagree with your aims to remove 3rr exemption. Are you so afraid to let a raw, honest, unmanipulated consensus form? Is a straight up and down "I agree"/"I disagree" such a horrible thing? If you had faith in your convictions and ideals that your idea was righteous and good for Wikipedia you would let it roll, but you clearly don't believe the community will support you, or you wouldn't be strangling this debate (as you did at Governence Reform) so tightly. Is the idea of trusting the masses to make a decision without the blessing of data, flowcharts and "senior wikipedians" so abhorrent to you? Between this and what you pulled to work to sink Kirill's governence reform proposal (which would have explicitly "de-powered" people like yourself to be just one of the peon masses) I've seriously lost a bit of respect for you. You will not win this by attrition, and debating opposers until they get sick of arguing and move on. Lawrence Cohen  §  t / e  03:41, 14 May 2008 (UTC)
 * (ec) The alternative and more likely explanation is that we have a communications problem somewhere. :-) Zhenquili points out that something is REALLY wrong with 3RR Exempt, which tends to confirm certain suspicions I've been having. The main thing I disagree with Zhenquili about is how to go about fixing it. I think his/her suggestions are a bit tricky to implement. (How does one go about educating administrators, or forcing them to implement anything? ) --Kim Bruning (talk) 03:26, 14 May 2008 (UTC) Note where I placed my opinion, it's not in support.
 * "(while, again, you advocated removal with your actions, if not your words)" is the answer to your small text comment "How does one go about educating administrators, or forcing them to implement anything?" By making them accountable if they are blocking against existing policy, warning them to follow policy, making RfC/Arbcom cases for the worst cases of continued flouting of policy, etc. I would think that was common sense. Faith (talk) 03:40, 14 May 2008 (UTC)
 * Absolutely spot on. The answer to a problem with a traffic light is not to rip it out, it's to write more tickets. ++Lar: t/c 03:50, 14 May 2008 (UTC)
 * 1) Well, we have a guideline, it might even be a policy, that administrators don't block users for being uncivil *to that adminstrator*. It's widely ignored, but, at least sometimes, when an admin does this, it is confronted, as with User:Physchim62, who resigned his bit under a cloud over it, or User:Tango, who is currently facing ArbComm over it. It's pretty clear: if a user crosses the 3RR line, it's been said, in some places, that this is a bright line and is to be blocked. Well, there is an exception, and if admins don't pay attention to that exception, and especially if they err and then stubbornly defend the error, they could lose their bit. By making a fuss, that's how we "educate administrators." Note that I'm not advocating desysopping adminstrators who make mistakes. But if the mistakes become a pattern, and the admin defends it and doesn't apologize or otherwise give us some sense that it won't continue, then it's time to protect the community by taking action. The BLP policy documents this. Without that exception being documented, an admin can simply say, 3RR is quite clear; maybe it was a mistake, but I was following policy. Frankly, everything has been turned upside down in this discussion. The exception is a *relaxation* of a strict rule: block for 3RR, not a creation of a new rule requiring something or even permitting edit warring. It provides a limited protection to editors who, in good faith, revert what they consider defamation. The possible harm is little: perhaps some legitimate material is *temporarily* removed from the article. This exception does not establish permission to edit war indefinitely, on the one hand, and, on the other hand, that some admins have blocked for 3RR in violation of the exception doesn't at all establish this as policy "based on actual practice." Is there any admin here claiming that actual practice is to block for 3RR without regard to whether or not the user was removing BLP violation? Is there any administrator arguing that they are going to disregard this exception? If the exception were *not* actual practice, I'm sure we'd see it. Did I miss it?
 * 2) Oppose. I have seen nothing in the discourse here that suggests this change is a) needed or b) helpful. --Fred Chilton (talk) 11:18, 14 May 2008 (UTC)
 * 3) Oppose. All this rule does is say that BLP is more important than 3rr. 3rr says don't fight, this is only a wiki, we can take our time to sort it out and get it right in the end. BLP says real people's lives are involved and real harm can occur so priority one is to put the questionable BLP claim in safe-mode right now and if there are issues to be sorted out then do that while the article itself is in safe-mode. WAS 4.250 (talk) 13:41, 14 May 2008 (UTC)
 * 4) Oppose - purely on the grounds that I'm seeing no evidence that 3RR-exempt is causing any trouble that its removal would correct (and yes, I saw the evidence listed above so don't bother responding to this vote with evidence - I simply don't buy it) and so in the absence of clear evidence of a benefit of the change, I support the status quo.

As a secondary note, I actually agree with the "non-Wikipedian notable person editing their own article and getting blocked" rationale behind the exemption - I think Wikipedians lose sight of what it's like to face this monolith as a neophyte to wikis and to the WP community itself, and so since edit warring is a fact of WP life anyway, any policy which increases or facilitates edit warring (and I do not concede that this exemption does) would *still* be okay by me, if it makes things easier for neophytes, *especially* in the case of BLP, which is an absolute shambles anyway and where one more incentive to edit war can hardly be seen as really impacting the 'jaw dropping' quality of the articles in that project.

I'd only support a removal of the 3RR exemption for this project if BLP were completely reformed, if new policies came into place to give people editing their own entries a recourse beyond the community - a super-editor or dedicated arbitrator who could handle issues raised by the subjects of BLPs. Only in that case would I support the removal of the 3RR exemption. CastorQuinn (talk) 04:28, 18 May 2008 (UTC)


 * 1) Oppose I've yet to see strong evidence that the 3RR exemption is causing harm beyond a few extreme cases. However from my experience, the enforcement of BLP can be difficult, and although I've never used it, the 3RR exemption gives the necessary teeth to a vital policy and helps ensure potentially harmful information will stay out until any problems can be resolved Nil Einne (talk) 06:25, 19 May 2008 (UTC)

Neutral/Other

 * 1) Voting is evil.  Corvus cornix  talk  18:51, 12 May 2008 (UTC)
 * Nobody is voting, I see plenty of discussion beside everyone's post, and I don't think we are going to do a head count at the end to determine consensus. 1  !=  2  18:01, 13 May 2008 (UTC)
 * Voting is not evil, as we voted for 3RR, the main page, our arbiters, and our WMF board. But no, in this case this is not a vote. Lawrence Cohen  §  t / e  20:47, 13 May 2008 (UTC)
 * 1) The ob- standard polls are evil poll question. ;-)  In this case it applies more than usual. We've already been using a stronger level of evidence (to wit: actually looking at what people are doing on the wiki, taking global consensus samples) so a mere poll is rather weaker than that. :-). Also I never made a proposal to remove the 3RR-exempt wording. I observed global consensus and removed it. (that's a bit different) --Kim Bruning (talk) 18:52, 12 May 2008 (UTC)
 * 2) *Is your point simply that this policy is not being adhered to in practice? If so, there's something wrong with practice, not with the policy. This is a good and important rule which enjoys wide consensus; if you know of cases where it is not being applied correctly, then draw the offending admins' attention to the actual policy.--Kotniski (talk) 19:08, 12 May 2008 (UTC)
 * 3) ** Policy is descriptive of consensus, it doesn't prescribe it. If it is not followed in practice, the policy is potentially wrong right there. While I agree that that is potentially, it *is* a red flag of course, and it warrants further investigation. On further investigation, we find that the 3RR exemption actually seems to hurt more than it helps. --Kim Bruning (talk) 19:14, 12 May 2008 (UTC) (see discussion over past week: one half keeps discussing theory, the other half keeps requesting evidence in support of that theory. The divide is quite interesting!)
 * 4) ***Can you give an example then of where this exemption, correctly applied, can be seen to have done harm? If the harm is through mis-application of the rule, then the solution is presumably to reword the rule to make it clearer, or to ensure more people are aware of it.--Kotniski (talk) 19:31, 12 May 2008 (UTC)
 * 5) **** Hmmm... Correctly applied. That's a good question. Define correctly applied? If you define correctly applied as where nothing has gone wrong, I'm not sure I can provide that. There's some evidence of people attempting to actually apply it, but almost every time something seems to have gone wrong. --Kim Bruning (talk) 19:46, 12 May 2008 (UTC) Though in the section below, I'm asking for more data, so maybe we'll discover something that meets your question
 * 6) *****So what is it you actually propose doing to make things work better? Reword the exception? Put it up in big letters on the 3RR pages? It seems clear that it shouldn't be deleted outright, as nearly everyone supports the intention behind it. Personally I find the whole concept of the 3RR seriously flawed (as I've said in a few other places), but given that people seem to consider it a good thing, we have to protect the valid exemptions which prevent it from becoming an even worse thing.--Kotniski (talk) 20:50, 12 May 2008 (UTC)
 * 7) ****** Originally I was going to just throw it out. Some people have now provided at least some examples, where we can see what went right and what went wrong. I hope to see some more examples, and then folks can start the thinking part. :-) --Kim Bruning (talk) 21:12, 12 May 2008 (UTC)
 * 8) We need more examples to dissect, and more discussion.  We're not ready to poll. Bovlb (talk) 21:14, 12 May 2008 (UTC)
 * 9) While I'm leaning towards supporting the removal of the exemption, I think we need to see more examples to examine to determine whether its really useful or not.  It's not ready for a formal, examined consensus yet.  <b style="color:#629632;">Celarnor</b> <sup style="color:#7733ff;">Talk to me  22:33, 12 May 2008 (UTC)
 * 10) Mu. Not a valid question per arguments advanced above, even if polls were absolute goodness. ++Lar: t/c 10:21, 13 May 2008 (UTC)
 * 11) Register not to vote. — CharlotteWebb 13:47, 13 May 2008 (UTC)
 * 12) I would support something alongside to handling of reversion of vandalism in WP:3RR - reversion of simple vandalism does not count toward 3RR, on the other hand removal of a referenced info is not a simple vandalism and it counts. The same way reversion of simple violation of WP:BLP (e.g. adding unreferenced negative info or info referenced only to a blog) should not count towards 3RR, while reversions because FOX News is an unreliable source or All Russian newspapers are biased, etc. should counts Alex Bakharev (talk) 01:22, 14 May 2008 (UTC)
 * 13) Some BLP violations are clearcut and akin to vandalism; others are a matter of judgment and interpreting policy. In general we ought to discourage revert warring, even if motivated by a desire to get rid of undesirable material.  I must admit to relying on the 3RR exception on occasion, but it's not too useful - there's a fear of getting blocked anyway if an admin decides it was a content dispute rather than a BLP matter, plus it's just not cool to edit war.  There are other, better alternatives, e.g. reporting the claimed violation on a noticeboard, asking for a third opinion, etc.  If the problem is serious enough that it's a BLP issue then in most cases cooler heads will prevail sooner or later.  If it's such a bad BLP vio that it needs to get removed quickly, I doubt anyone would block a good faith editor with or without the exception - an admin taking a look at it would probably block the BLP violator or protect the article in a good state.  Perhaps we could clarify this process.  But I don't see the 3RR exception as a problem that's crucial to fix either.  It doesn't get abused very often. Wikidemo (talk) 00:39, 14 May 2008 (UTC)
 * 14) This is unnecessary - if someone has added a BLP violation, you remove it, and they re-add it twice, don't continue edit-warring - report it to WP:ANI or the BLP noticeboard. There's never a need to edit-war over this. If they are breaking BLP, then they can be dealt with by a neutral admin. <u style="text-decoration:none;font:100% cursive;color:#060">Neıl <u style="text-decoration:none;color:#060">☎  11:38, 14 May 2008 (UTC)

Keeping a log
/log

Ok, well, let me help the folks who still want to provide evidence. (that and keeping logs is a smart idea anyway)

I propose we keep a log (like is being done for IP block exemption too ... see Wikipedia_talk:IP_block_exemption/log), where people can take note of their BLP actions, and how they work out. Since you're saying this particular rule (3rr-exempt) is so important and oft-used, you should have an easy time filling it up. We can then come back in a week and see where we stand. :-)

--Kim Bruning (talk) 19:06, 12 May 2008 (UTC)


 * Kim, that sound highly impractical. It would take much longer than one week to reach the majority of editors and administrators to alert them to the need to log. BLP issues are not handled in a central location.


 * Plus, I'm not at all clear what you are trying to show with the data. Also, under close observation, editors and administrators actions would change so the data would not be meaningful. FloNight&#9829;&#9829;&#9829; 23:05, 12 May 2008 (UTC)
 * There is no need to log, it's just a voluntary. You don't need to reach a majority of editors, you just need to ask a small number. It doesn't need to take very long either.
 * The objective is simple, I'm not trying to show anything anymore. This is up to you and/or others. It's like Lar says, right now we only have negative information; so try show a small sample of situations where BLP 3rr-exemption was successfully applied at least some of the time. Folks find the exemption very important, they say. So if that's true, it shouldn't be hard to show it being applied successfully. Alright? --Kim Bruning (talk) 23:24, 12 May 2008 (UTC) it's also ok to throw in old information... anything. Just so we have something to work with, and not just ghosts and stories. It's obviously possible, several folks have already provided examples. Folks could just pre-fill with those even. Whatever floats your boat.
 * If I show you a situation where matters stopped at 1 revert, or at two, who's to say that is or isn't a successful application of this? I say it is. Why? ... When traffic engineers study the effectiveness of a traffic signal, they measure all the traffic, not just the times the signal was disobeyed. Stretching the analogy further, if on the times there are accidents at the intersection, the police are giving the wrong motorists the tickets, that's not a condemnation of the traffic signal, it's a condemnation of police procedure for giving out the wrong tickets. Perhaps all those cases that were reported above, where we found the person doing the reverting under BLP was censured, are an indictment of processes at the 3RR noticeboard, or of the admins handing out the blocks, not of the exemption. Hence my suggestion that the wrong data is being collected here, the net is not cast wide enough. I'll, I guess, start logging all my BLP incidents and count each and every one of them as data in support of the 3RR exemption, unless I get blocked after reverting. Since I never ever exceed 3 reverts, (and rarely even exceed 1, I subscribe to 1RR... it's a personal matter, not a disagreement with the exemption) I'm going to have a 100% success rate. ++Lar: t/c 10:19, 13 May 2008 (UTC)
 * If you subscribe to 1RR, you're following a different procedure. You should log as success for 1RR. :-) Don't discount the procedures you yourself are using in the field, just because they are your own. We're actually interested in procedures that work in the field you see, because that's something we can work with and improve. --Kim Bruning (talk) 11:41, 13 May 2008 (UTC) At the end of the day, this is not a game, nor is it a theoretical exercise.
 * Nope. If I always stop at yellows, that does NOT mean that the concept of traffic lights, or that the concept that people normally HAVE to stop at red, or the concept that emergency vehicles get an exemption from reds if their lights are flashing, are flawed concepts. You haven't grasped the key point I'm making yet. Measuring only traffic accidents is no way to measure effectivness of traffic laws. I'll be logging all my BLP actions as successes for the BLP 3RR exemption policy, just as I would log all my stopping at yellows as successes for the red light exemption for emergency vehicles policy. (that is, if I stopped at yellows and we were carrying out such an analysis via logging... I don't stop at yellows, I subscribe to a somewhat different policy in practice, but I digress :) ) ++Lar: t/c 13:02, 13 May 2008 (UTC) (PS, I highly recommend against stopping early in the yellow cycle, at least in the US, it's a sure way to get rear-ended on a regular basis, since hardly anyone else does and you will catch people off guard. But I digress. Mentioned only to highlight that no analogy is perfect. Not even my almost uniformly awesome ones! :) ) ++Lar: t/c 13:12, 13 May 2008 (UTC)
 * Wikipedia does not use a code of law, it uses consensus. Now, to get a valid result, you log what you do, not what you hope the outcome is. It goes without saying that I expect you to be honest. Others will log their behaviors, and we shall see where we end up. Alright? --Kim Bruning (talk) 14:17, 13 May 2008 (UTC)
 * Perhaps I'm not following you here. I'm willing to log my BLP activities for a while, sure. I'm about to carry one out later today in fact. I'll report honestly what happens. I'll count each activity that results in the removal of BLP violating material without my getting blocked for doing so as a success of the "3RR BLP exemption". Regardless of how many times I had to remove the material, or how many times anyone else reverts. I'll count each activity that results in my getting blocked, or that fails to end up in the BLP violating material ending up being removed when all is said and done, as a failure of the "3RR BLP exemption". Regardless of how many times I had to remove the material, or how many times anyone else reverts. As I have explained, that's the correct population to be sampling from. The correct population to sample from is number of cars through the intersection, not number of crashes, or number of tickets, or number of red lights run without further incident... All that said I'm not sure what you're driving at exactly about needing good faith or ethics or whatever. (as an aside, the red light thing is just an analogy... but BLP is more like law than consensus, would be my assertion (without evidence directly provided, so feel free to dismiss it :) ) ) ++Lar: t/c 14:31, 13 May 2008 (UTC)
 * Ah! I think I get it. You're pointing out that I said "I will log all my actions as successes". That's merely because I assume they all are going to be, just as all of them have been so far. But ya, I may end up marring my perfect block record here, or may end up failing to remove something that needs to be removed. Oops! Those would be fails, yes. Excuse me for my cockiness! :) (but there won't be any fails from ME :) ) ++Lar: t/c 14:44, 13 May 2008 (UTC)
 * Just objectively write down what you did and what happened (diffs always appreciated), we can figure how things work based on that data, alright? --Kim Bruning (talk) 14:46, 13 May 2008 (UTC) While binary (succeed/fail) data is nice, to build/update/maintain/retain consensus, we need more information than that. :-)
 * ps. Thank you! :-) --Kim Bruning (talk) 15:31, 13 May 2008 (UTC)
 * I put one in there from before. It's a successful use of the policy by my metric. I've got more. ++Lar: t/c 18:18, 13 May 2008 (UTC)

Global Consensus?

 * Kim wrote above: "Also I never made a proposal to remove the 3RR-exempt wording. I observed global consensus and removed it. (that's a bit different)".
 * Please what does global consensus mean in this context, and how did you decide that it existed? CBHA (talk) 02:28, 13 May 2008 (UTC)
 * I believe he's referring to common/best practice. The argument is that if there's scant evidence that some policy is being followed in the field, and even less that it actually helps when it is followed, then it doesn't deserve to be documented as policy.  In this view, policy is a description of the things that are actually done by the community, or at least the bits that are seen to work out well.  Thus, policy is not a way to impose novel restrictions, and it does not change practice except by building on things that have already been tried.  As a process engineer, I can attest that a process with buy-in generally works out better than a theoretically superior process without.  When I isolated three assertions above, two built on this (that the exception helps, and that it is commonly used) and one explored an alternative to this view (that BLP with its 3RR exception were imposed "by fiat" in order to change practice).  Cheers, Bovlb (talk) 03:57, 13 May 2008 (UTC)


 * It seems to me that WP:BLP is the poster child for prescriptive policy. And if there was a significant disconnect between what it says about 3RR and actual 3RR enforcement practice, I would expect at least some editors who had been sanctioned for 3RR while enforcing BLP to have appealed to Arbcom by now. As far as I know, that has never happened. The most likely explanation is that, for the most part, the existence of the 3RR exception serves to deter edit wars over BLP violations. The rare situations where an editor accused of a BLP violation persists are the ones where that editor believes BLP does not apply to his edits and the other editor is abusing the 3RR exemption. Those few cases are the ones we see in the table above. --agr (talk) 15:31, 13 May 2008 (UTC)
 * For the three 3RR cases in my table where the BLP exception was arguably applicable, but which resulted in blocks, I'm not aware that anyone invoked the BLP exception explicitly, Two of the blocks were by one admin, who has already commented in this discussion. Maybe we should invite the three editors and the other admin to contribute their insights.  Bovlb (talk) 16:50, 13 May 2008 (UTC)
 * Agreed. --Kim Bruning (talk) 17:00, 13 May 2008 (UTC)
 * Done. I now see that in case 10, something very much akin to a BLP exception argument was made, but rejected.  Bovlb (talk) 18:21, 13 May 2008 (UTC)
 * Which is a good indicator that wise admins made the process work by recognising bad usage of the exemption. Faith (talk) 21:14, 13 May 2008 (UTC)
 * Commenting on my block of Pia L - I specifically chose not to address the content issues of that case. I believed that content could still be worked out on talk and that reasonable people would sort the mess.  I chose to act, instead, strictly on the 3RR violations by that user.  In this case, because the user was "blind reverting", I felt that there was a likely possibility that they'd keep reverting all day unless I got involved.  For the record, I do not support a 3RR exemption for BLPs.  Rather, the processes we have in place should be sufficient - and if they're not, we ought to work on those processes.  - <span style="font-family:Papyrus, sans-serif; color:#775ca8;">Philippe  18:59, 13 May 2008 (UTC)
 * In an impartial (and quick) review of this particular case, it appears one side was adding well-resourced information about court judgements and negative information about the subject, while removing/severely reducing positive well-sourced information about the subject's advancements, and the other side was claiming BLP and doing the reverse. This would not be a case for the exemption clause, but for improper use of BLP as a sword. The exemption is for instances of "Bob is a pedophile", when Bob is a famous whatever who was never accused, convicted, mentioned as a pedophile in any way, shape, or form, not for content disputes. This is where proper "policing", as I mentioned before, works. Faith (talk) 19:28, 13 May 2008 (UTC)
 * Faith, no "well-resourced information" or simply "negative information" was removed from the article by me, unless you count fraud accusations found solely on self-published serial "fraud accuser" web sites to be suitable "facts" in a BLP . Another such site, (informath.org) was finally agreed to be an inapproriate replacement for the reliable published sources which had been removed after April 15 (see talk page here), and the original sources were reinstated here, although the official documents on the Swedish Parliamentary Ombudsmen website still has not been reinstated (see for example the link to informat.org in note 50 under Notes). Concerning the Swedish TV-documentary, it was also discussed on the  talk page. The program contained a dramatization of the shredding, to establish if the shredding of 22 meters of material could really have been done in the time frame indicated at the trial. Apart from that, the majority of the airtime was dedicated to interviews with the adversaries of the WP biography subject (Gillberg), who himself had declined to participate (as did the group of researchers who had shredded the documents). The absense of their input presented a big problem when it comes to balance. Another reason the program with the interviews may not be well suited for the lead  of a biography on Wikipedia is the severe criticism the interviews and the lack input from the "other side"  have elicited in the major newspapers of Sweden (, even resulting in the program being called  Tabloid TV in a DN editorial). The fact that the program is under investigation by the Swedish Broadcasting Commission is also of some concern. (See for example this article about the reports to the Broadcasting commission and the general criticism). However, contrary to what is implied by your "whitewashing" comment, I never intended to remove the documentary (as indicated by my edit here see note 51) but reverting to the last stable version of course meant that all new material was temporarily removed , but only until the BLP notice board had time to intervene regarding the issue with the false statements. In my opinion the docudrama should certainly be part of the article, but the speculations expressed in the interviews need to be properly attributed and the critique of the interviewees' speculations as expressed in published sources (especially by representatives of the media) should not be excluded from this BLP - they are not "white-washing", as implied. In order not to overpower the article once both sides are added, the section is better suited for a section in the running text, or, if very detailed, perhaps in the article about the study. Please note that it is not a question of excluding facts about the dismissed allegations either. The claims by the different parties in the controversy were not added by the anonymous users - they were described already and still remain. What was introduced after April 15 was an emphasis on fraud speculations in the lead, with no mention of the dissmissal of the allegations and no mention of the fact that Gillberg was not involved in the shredding. Also, the post-April 15 edits about an "outside scientic scrutiny" are incorrect and contradicted by the fact that the unanonymized raw data (including taped interviews, info about family relations, sex, criminal records, school records etc) from the longitudal psychiatric study, which had followed a rather small group of children from the 1980s forward, was sought under the principle of public access, but with the confidentiality requirement intact (=the persons who got access could not talk about or publish what they learnt about the study and the children). The investigation regarding the scientific misconduct allegations was officially over before that access was granted. Pia (talk) 02:36, 16 May 2008 (UTC)


 * I would disagree. "...adding well-resourced information about court judgements and negative information about the subject, while removing/severely reducing positive well-sourced information..." sounds like a BLP violation to me. We are not supposed to be neutral in that kind of content dispute. An editor that appears to be trying to damage someone's reputation should be sanctioned, not the editor who is trying to restore balance.--agr (talk) 20:10, 13 May 2008 (UTC)
 * That's the problem. It wasn't restoring balance, but rather whitewashing the problem out of the article or reducing it to such simplicity as to hide it in great amounts of text.  So it was a content dispute with both sides going heavily toward pos-POV or neg-POV.  This is not what the exemption was designed for.  If it were false information being added/removed/added/removed/etc., then the exemption might be validly used. Faith (talk) 21:12, 13 May 2008 (UTC)


 * Part of the problem is the conflation we've had between simple BLP issess (unsourced or poorly sourced contentious claims) and more general BLP-penumbra issues (such as due weight, npov phrasing, privacy levels appropriate for notable public figures etc.). I suspect that many more people are ok with a 3RR exemption for the first type than the second type. JoshuaZ (talk) 03:58, 14 May 2008 (UTC)
 * Which supports my point as valid that education, not eradication, is the answer. --Faith (talk) 04:05, 14 May 2008 (UTC)
 * Perhaps some clarification of procedure should be in order too. I suggested way back in this thread that editors who exceed 3RR and notify an admin or the BLP noticeboard in a timely manner should be initially warned, not blocked, if it is determined that their use of the exception was not appropriate. The specific case we are talking about in this section, Christopher Gillberg, raises a number of questions. Some of the issues are what JoshuaZ call penumbral, but I would argue that the article is not in an acceptable state at the moment. We normally tolerate competing pov-pushers in non-BLP articles and end up with articles that are imperfect, but valuable presentations of both sides of an issue. Is this good enough for BLP? Also the BLP noticeboard was informed about this situation by the editor who was blocked on April 22, but there is no record of action, calling into question the idea that the 3RR is never needed since help is immediately available. Finally, one of the disputed sources is a television documentary in Swedish.   Should we allow contentious BLP material on English wikipedia that English-speaking readers cannot verify? --agr (talk) 12:13, 14 May 2008 (UTC)


 * I don't think the system worked very well in my case, but perhaps that was just an isolated incident. I appreciate the efforts to put the policy issues in focus (although I don't feel it's appropriate for me to participate in the survey, as I am one of the examples above). I was blocked for violating 3RR while removing non-factual and damaging material by reverting to a version that had been stable since August 2006. The compromising edits started April 15 2008 . I reverted to the last stable version in order to have the BLP notice board look over the new additions that I consider to be BLP violations. Even the editor who reported the 3RR, User:TheSeven, admitted that one of the newly added statements was indeed false: "Your point 1, above, was valid; I corrected the article" . However, this was only after my removal of the false information had been reverted 3 times. Apart from the more obviously non-factual statements, the article had other items indicating culpability and participated in fraud, including a scientific misconduct tag. In a BLP of a scientist against whom the fraud allegations have been officially dismissed by the authorities, this would not appear appropriate. User:TheSeven also admitted on the 3RR page that, "Gillberg was acquitted of fraud, but some people believe him to be guilty" - what "some people think" would seem a weak argument for inclusion of fraud allegations in a BLP.  I posted an alert on the BLP notice board as soon as the problems started . I don't think I failed to state my concerns either: I explained in detail why I considered the edits to be violating BLP policy on the talk page of the article, ,  and I explained the problems again on 3rr notice board . In addition, the BLP problem in this article was a repeat occurrence: it had previously been to arbitration (August of 2006) , but since that time, the article had remained stable -  up until 15 April 2008, when anonymous users, using a blog as a source  reinserted the non-factual fraud accusations. I must admit that an added worry in my mind is the possible impact of certain passionate, religiously motivated campaigns, especially since one of the Gillberg adversaries has issued favorable statements regarding one organization's enlightment efforts, after which her cause has become championed extensively by this organization in Sweden. This becomes a concern when anonymous groups collaborate in introducing rather troubling POV-slants in biographical articles about people involved in the fields of neuroscience, genetic research or psychiatry, since, unfortunately, these fields seem particularly vulnerable to that sort of group activity. Pia (talk) 02:36, 16 May 2008 (UTC)


 * Commenting on being blocked by Philippe: No, reverting to the last stable version was not "blind reverts", but used to "hold the fort" until the non-factual and damaging new additions that an anonymous editor, a single use account and User:TheSeven were reverting to (after having introduced them on April 15),  until these additions could be looked over for BLP violations by BLP notice board. I agree with the blocking administrator that the issue at hand is complicated and hard to grasp because of the language barrier, but I also want to state for the record that I very much resent being made an example of, especially since the block was implemented without even the courtesy of a warning that the BLP policy would not applicable in my case. If the intent was to demonstrate a point about how BLP policy is ineffective and needs to be changed, there must surely be better ways than to block someone who acts in good faith -  I have always been very careful about following rules and had a clean record up until this "blind" encounter, but I am now permanently marked as a disruptive editor which is very discouraging to me personally. If the BLP 3RR exemption policy is to be considered negotiable at the 3RR board or if the policy is meant to imply that you should not revert false statements or questionable sources on the spot but instead turn to an administrator to ask for the article to be protected until people at the BLP notice board has a chance to check your violation concerns, it definitely needs to be stated clearly. Pia (talk) 02:36, 16 May 2008 (UTC)


 * Did your posting to the BLP noticeboard result in an action? If so when? --agr (talk) 03:05, 16 May 2008 (UTC)
 * Agr, no result as yet. I think that after my block, my posting with the plea for help and input was simply moved to an archive. When someone finally does have the time to check the article, and if I'm not able to respond and explain my concerns again, I have marked the POV problem spots that are left in the article today, using "clarify" tags. The new headline introduced on 15 April 2008, with the words "Alleged scientific misconduct" is still there and both that section and the lead could use some work. Pia (talk) 04:07, 16 May 2008 (UTC)

Simplicity its own sake: When should the 3RR apply?
The 3RR should apply when none of the [edits or subsequent] reverts violate any policy other than 3RR. Sorted. — CharlotteWebb 15:16, 13 May 2008 (UTC)
 * You mean when none of the edits being reverted violate...?--Kotniski (talk) 15:24, 13 May 2008 (UTC)
 * Do you edit war with yourself? No, actually you're right. There may be a situation where none of the edits being reverted are reverts themselves. See update above, but try not to make it needlessly more complex. — CharlotteWebb 15:43, 13 May 2008 (UTC)


 * --Kim Bruning (talk) 15:38, 13 May 2008 (UTC) though odds that it will stick might not be so good. See for instance this similar statement at CSD, which got reverted out.. Still sounds like part of a solution though. I just hope people provide more examples, so that we get more ideas.
 * Well, the other approach to say "the following types of edits may be reverted till doomsday" followed by a list that is too for anybody to read before deciding whether to block for 3RR. Are those the sort of examples you are looking for? — CharlotteWebb 15:58, 13 May 2008 (UTC)


 * <GRIIIIIN> A kindred spirit? \o/ That list is actually too short. Check out the table at WP:CSD for true horror ;-)


 * Basically what I'm pointing out here is that probably we *shouldn't* have exceptions to the three revert rule (which I detest as much as visiting my dentist, mind you). At least, very few people seem to actually be coming forward with examples of an exception working very well for BLP. ^^;; (/log is still empty at time of writing)


 * --Kim Bruning (talk) 16:26, 13 May 2008 (UTC)
 * OK, now I call BS. I have lots of them available but you haven't conceded they're in the sample population yet. Others not following what I mean can see /log for themselves. ++Lar: t/c 03:48, 14 May 2008 (UTC)
 * So list them, and let's take a look. --Kim Bruning (talk) 05:01, 14 May 2008 (UTC)
 * Oh I see. Note that my comment about the log being empty was made at 16:26 UTC. Your first post to that page was at 18:13. --Kim Bruning (talk) 05:40, 14 May 2008 (UTC)

My replacement for the three-revert rule would be something like this: before doing anything for a second time, you must answer all the arguments which were raised against it when you did it the first time (and if there were no such arguments, state your arguments against its being undone).--Kotniski (talk) 16:51, 13 May 2008 (UTC)
 * That doesn't quite work (it allows infinite filibuster). If you make the requirement that you should try to reach compromise with one person at a time, (and bypassing people who are blocking consensus), well, that's Bold revert discuss ;-) --Kim Bruning (talk) Remember, it's a cycle. By forcing people to discuss everything on the talk page beforehand, you break the cycle, and you might never get around to editing :-P

The issue is already very simple, when removing problematic BLP material (...removed immediately and without waiting for discussion), 3rr does not apply. There's no long list of exceptions...just this one simple concept. Reverting (in this context read: removing) Unsourced or poorly sourced contentious material about living persons is something we have to do. If you've removed/reverted something 2 or 3 times in a row already and are faced with another insertion, BLP policy doesn't go away...we're bound by it before most other editing restrictions like 3rr. For those familiar with Active Directory or similar concepts, think of it as effective permissions.

Of course, you may find trouble if you're wrong about it being a BLP issue, it has to be applied correctly. But that's true while enforcing any policy here whether it's blocking, deletion or whatever policy. BLP is no different in that sense. RxS (talk) 17:23, 13 May 2008 (UTC)
 * Seeking assistance is a very good way to determine whether you're wrong. Bovlb (talk) 18:22, 13 May 2008 (UTC)


 * Yes it is, and since we err on the side of caution when dealing with BLP, it's perfectly acceptable (mandated in fact) that we keep contentious material about living persons out of articles while soliciting help/advice/assistance. Time is a | factor. RxS (talk) 20:27, 13 May 2008 (UTC)


 * Well, yes... there seems to be a very particular way to do it right, doesn't there? But we don't have a large enough sample to be sure. Can you provide some examples at /log? --Kim Bruning (talk) 21:04, 13 May 2008 (UTC)


 * Actually we can be sure. It says it right at the top of WP:BLP, there's only one way to immediately remove something without needing any discussion. If you find a better way to immediately remove something without any discussion, more power to you. But that doesn't depreciate the justification for exceeding 3rr when faced with a BLP problem. RxS (talk) 22:27, 13 May 2008 (UTC)
 * It doesn't. It just says that it should be done. I think there are some very clear issues with the 3rr exemption that need to be resolved. People still are very skimpy on evidence. What little evidence I *have* seen, shows a particular Pattern that's very familiar to me, and which I might enjoy describing. Application of a clear pattern would help prevent a lot of the problems that have been reported. But I don't dare go that far until I have a bit more information. I'm hoping the pattern will show up some more. --Kim Bruning (talk) 23:09, 13 May 2008 (UTC)  I'm letting you see into my kitchen a bit with this statement, but I'm not done cooking yet, so hands off the ingredients! ;-) 
 * And I guess this is my problem with your request for everyone to run around looking for evidence for you to interpret. IMO, any evidence found using recent examples would not illustrate the situation before the implimentation of the exemption, as the exemption is already in place.  There needs to be strong, strong, strong evidence for removing the clause brought to the table, along with a proposal for a viable substitute.  The exemption was not created in a bubble of instruction creep; it was created to address a problem, so it would need to be shown (1) the problem no longer exists, or (2) the exemption is no longer viable as a solution to the problem.  I've seen no evidence of either point presented, aside from a data sample far too small to be statistically viable, and which doesn't take into consideration when it works and isn't being reported for 3RR (as has been pointed out already).  --Faith (talk) 03:36, 14 May 2008 (UTC)
 * I think you have some things a little mixed up. I'm not trying to change policy. I'm figuring out what it is, and then documenting it. What I'm asking is for some more evidence so that we can keep the 3rr exempt in. So far I've only seen some evidence for removing it.  --Kim Bruning (talk) 03:55, 14 May 2008 (UTC)
 * My perception is starting to shift to one where I view your actions in this matter as trying to game the system, or even to being a bit disruptive. You can't analyse data however you see fit, draw whatever conclusions you like from it and then claim that you are the only person able to divine consensus from it. You have been challenged a number of times now to defend your interpretation of the data, to defend your definition of the sample set population, and you are failing to do so. Therefore I suspect it is your conclusions here that may be spurious, even if it did make sense to say that this particular policy was descriptive, which it does not. Kim, I am one of your biggest fans, you know that, but you've lost the plot on this one. Stop dodging the valid objections raised and deal with them. Stop using the question game to prolong discussion and hiding what point you are trying to make. Stop taking people's statements in support of one point and reframing them to claim they support the opposite point. This really all comes down to this: you made some unsupported edits to a very important foundational policy, they were reverted, because the policy isn't what you think it is, and now you're dragging everyone else through hill and dale chasing chimeras. Stop. Please. ++Lar: t/c 04:19, 14 May 2008 (UTC)
 * BLP is not an foundation issue at this point in time. But ok. I never set out to do a full statistical analysis. If you read back, you'll see I just asked for a couple of (anecdotal) examples. I expected that to take 7 minutes. Maybe 70 on a slow day. So far I've gotten quite some anecdotes that there's something wrong, and only a couple that this 3rr exempt actually works as advertised. (I respect you too, but I'm not really someone who figures things out on the basis of warm fuzzy feelings ). So that's not really what I expected to happen in this case. When I made the BRD edit on the 12th, there was actually no evidence to support 3RR exempt at all at that point in time. Since that time, I have made no further edits.
 * I don't think we've ever set up a large scale statistical analysis for updating just 2 or 3 sentences before. (or in fact so much hoopla ever at all). Is that what you'd need me to do now? --Kim Bruning (talk) 04:38, 14 May 2008 (UTC)
 * No. I need you to drop this. You are wasting the community's time. If you have other suggestions for improving the BLP 3RR exemption, so that we don't get as many bad blocks when it was validly used, that would be great. But the fact some ambulances get struck by careless motorists when going through red lights is absolutely no reason to take their exemption from red lights away. It is a reason to educate the careless motorists not to get in the way, because we're not going to get rid of the need for ambulances, or the need for red lights any time soon. Similarly, the need for the BLP policy based edits (the ambulance) and the 3RR policy (the red light) is going to remain with us for some time. You need to internalise all this and move on. Because this is truly a massive waste. For shame. ++Lar: t/c 04:48, 14 May 2008 (UTC)
 * Why do we need this, if there is no popular support for what you want? You're starting to act like a pit bull that refuses to loosen it's grip until death. Lawrence Cohen  §  t / e  04:44, 14 May 2008 (UTC)


 * And, as has been pointed out, it's semantics. Edits to the policy are changes to the policy as it is written, regardless of the equivocation that you aren't trying to "change policy". So, semantics aside, I stand by my points before your reply: "it would need to be shown (1) the problem no longer exists, or (2) the exemption is no longer viable as a solution to the problem". Neither has been accomplished. --Faith (talk) 04:02, 14 May 2008 (UTC)
 * It's not merely semantics. It's a fundamentally different approach to things than you might be used to, it looks like. The way this works *is* documented in policy, but maybe I should just explain it to you someplace. This might take some explaining. :-/ --Kim Bruning (talk) 04:24, 14 May 2008 (UTC)
 * The problem is your "fundamentally different approach" is counter-productive to the discussion. Using "I'm not trying to change policy" as a reply to valid objections is not helping the process move forward. It's just annoying.  Regardless of your intent, in practice, your arguments in talk and your edits to the policy attempted to make a change from how it was currently written.  Your equivocation of this point, coupled with your statements of neutrality on the issue, show a complete lack of conviction for the differences you attempted to implement, and as I said, this made my opinion on the matter even stronger, because if you lack conviction enough to support it from a solid position, how can you hope to address the opposition appropriately or support your point effectively.  The neutrality is harming your case, and as Lar points out above, the process you are following isn't all that great either. Faith (talk) 04:34, 14 May 2008 (UTC)
 * I think we have seen that there are some clear issues with the implementation of 3RR Exempt, that bear further investigation. Will you be doing that? --Kim Bruning (talk) 05:09, 14 May 2008 (UTC)
 * No, Kim, because, as I've said several times, I do not see it as an issue with the policy but with the implementation of the policy, the "policing". Please stop pressing this onto other people.  If you have an issue with the policy, in whatever shape or form of terminology you wish to express that issue, then you are going to need to give strong evidence for change and achieve consensus for that change, because as it stands right now, your attempts were reverted and your "attempts to make the policy read differently than it currently reads" (which is a long-winded and very annoying way of having to say "changing the policy", but hopefully this will avoid the unnecessary stress being placed on a preferred version of wording "change") have been met with strong, supported opposition.  (See my unaddressed points about the poor data sample, as well as Lar's general comments regarding the need for looking at the big picture). Faith (talk) 05:42, 14 May 2008 (UTC)


 * I've got one for you to go fix yourself, Kim... take a breeze through the header material on Administrators' noticeboard/3RR. Nowhere in there that I could spot did the exemption get clearly explained. It gets MENTIONED but it doesn't explain that admins can get in trouble for not heeding it, as it should. Be Bold, Kim. Go fix that. Then add some of your BLP related article edits to the /log so we have more data. I'm quite curious to see some of yours. ++Lar: t/c 05:26, 14 May 2008 (UTC)
 * Again, if consensus supports a PRESCRIPTIVE policy in this case--as the stated consensus of many editors does--then it stays in, and is prescriptive. If you don't like that consensus, Kim, YOU need to develop a new consensus that BLP should not be prescriptive. I don't care how senior or veteran you think you are. Your voice carries not one penny more weight than the rest of us. Lawrence Cohen  §  t / e  04:04, 14 May 2008 (UTC)
 * Both our voices carry as much weight as the strength of our arguments, no more, no less. Once again just so we're clear on this: I'm not challenging this policy in its entirety at this point in time. I'm trying to document the current situation and explore fixes for certain issues that have shown up. Alright? --Kim Bruning (talk) 04:24, 14 May 2008 (UTC)
 * (ec) No. You in fact were trying to change a prescriptive policy without consensus for that change and were properly reverted. That's my assertion, which you have not refuted. BLP is prescriptive, not descriptive. Further, your arguments throughout this matter are weak, and Mr. Cohen's are strong. ++Lar: t/c 04:33, 14 May 2008 (UTC)
 * I've been applying all normal documented procedure. BRD, 1RR, CONSENSUS, etc. and have then backed off appropriately and gone to discussion. --Kim Bruning (talk) 05:05, 14 May 2008 (UTC)
 * Except it appears to me that "discussion" has left the building to be replaced by a version of WP:IDIDNTHEARTHAT. --Faith (talk) 05:09, 14 May 2008 (UTC)
 * Kim, as Lar and I have detailed, you have not engaged in discussion here. You've engaged in trying to game consensus to the end you wanted. That will not stand, nor will it be allowed. Your behavior here is exactly the sort of thing that Kirill Lokshin's Governance reform proposal that you fought tooth and nail to sink was designed to stop: one editor from dominating and shaping policy decisions to his own views. Well, I think I will end up doing what I said on your talk page, and build the entire reform platform in user space, where tenditious meddling is restricted, and then put it to the community. Your actions here have been the utmost clearest demonstration that "old timers" need to be reigned in from gaming discussion and consensus. Thanks! Lawrence Cohen  §  t / e  05:19, 14 May 2008 (UTC)


 * Now I get to call bullshit as Lar did. My voice carries weight, period, as does yours, as does User:Joe Schmoe who posts Oppose in the straw poll and never shows up on this page again. Does his voice carry less weight if he's not willing to sing to your tune and do your song and dance? Nonsense. There is nothing broken with the 3RR exemption--all that is broken is some people do stupid things with their admin tools and incorrectly block when the blockee is exempt under our standing policy. If we have a stupid admin or five, that doesn't mean this policy is broken. You can't fix stupid. Lawrence Cohen  §  t / e  04:30, 14 May 2008 (UTC)
 * I think it's a bit more than 5 admins doing this. --Kim Bruning (talk) 05:07, 14 May 2008 (UTC)
 * Please present data and evidence to support this claim. Lawrence Cohen  §  t / e  05:19, 14 May 2008 (UTC)

I look at Talk:Fritzl incest case, and see one editor editing warring and taking this as justification, only being stopped by the article being protected. I fail to see why this disruptive action helped anything; right, wrong or indifferent, it could have been handled in a less high-handed manner without this rule, probably just as quickly provided that someone besides Lawrence believed that it was a BLP issue. A notice on WP:ANI would have brought a dozen editors running and would have effected the same result without starting an edit war ... or not, which would have brought into question whether this really was a BLP issue.--Prosfilaes (talk) 03:48, 15 May 2008 (UTC)

Ok, that was amazing... let's not do it again
A week or two ago, some people told me about the difficulty of handeling BLP policy. Well... I didn't believe them. Later I found a minor flaw, and decided to try and fix it. How bad could things possibly be? Well... 8 days later, I think I've found out. ;-)

Since there are now 3 people drawing a bead on me and preparing to assume bad faith, I think it might be a good idea to back off, and figure out what happened.

One thing of interest was that in the not the wikipedia weekly chat (unfortunately not recorded) someone else decided to demonstrate what was up. They simply asked the same questions I did, and the NTWW channel ended up acting exactly like here (the questioner was even pushed into the same role of "no no! Slow down, I didn't mean that, I mean..."). It was quite hilarious, and quite interesting to see. Also... it probably means a bit of a rethink of strategy.

So I'm going to go off and ponder on what happened for a couple of days, and figure out what happened. Feel free to either archive or continue the discussion here, as you see fit.

Thank you very much for your time and patience. If you'd like to drop me a note and just chat quietly some more, that'd be nice too.

Have a nice morning!

--Kim Bruning (talk) 05:37, 14 May 2008 (UTC)


 * Ouch. My summary of the situation is that (1) an exemption was removed without supporting evidence that: (a) it was no longer viable, and (b) its inclusion was harmful to the policy; (2) discussion points supporting the removal consisted mainly of: (a) prove to you it should remain, (b) do all the work for you by searching for specific examples where the exemption was invoked, and have them meet your interpretive approval; (3) many of your replies degraded to WP:STICK and WP:IDIDNTHEARTHAT. That doesn't assume bad faith; it points out that we are running around in circles trying to accommodate you, while asking you, in return, (1) to support your points through more than equivocation and claims of neutrality, (2) to prove how the sample data you pointed to several times could in any way be considered an accurate representative of the whole when: (a) it didn't sample from across the wiki, but only from the noticeboard, (b) it had a very limited number of examples comparatively, (c) did not present occasions when the system worked (again, by only drawing from the noticeboard where listings only show when it wasn't working); and (3) to please stop answering points with questions, shifts of responsibility, rewording of opposing points as if they lended support to your point, and pedantic arguing. --Faith (talk) 06:07, 14 May 2008 (UTC)


 * I think we were running around in circles around each other. I don't think either party was deliberately avoiding questions by the other, but it did look that way to both sides. Like I said, the NTWW channel folks actually got themselves into a similar situation with the same questions (applied to a hypothetical BLP-like policy).


 * I think that that session suggests that what I did was not a useful approach to use here. So I'm going to go off and think about that.


 * --Kim Bruning (talk) 06:20, 14 May 2008 (UTC)

Prescriptive versus descriptive
In a section above, I identified three assertions that seemed to me to be being made without any supporting evidence. Two of these related to the 3RR exception being either useful or used. Since my post, several people have posted evidence of these, which I very much appreciate. The third assertion related to the idea that BLP policy (and its 3RR exception in particular) is imposed by fiat, and is therefore prescriptive and hence not required to describe actual practice. So far as I can see, this claim is being made independently of whether the policy does describe actual practice. Since my post two days ago, I cannot see that anyone has provided anything in support of this, yet the assertion has been repeated several times. (Again, please excuse me if I'm missing some explanation in the mess of threads,) I have also seen a variation that appears to claim that the policy is prescriptive, not because of a fiat, but because of a consensus.

I think this distinction lies at the heart of much of the confusion above, so for those of you making a claim that BLP and its 3RR exception are prescriptive, I renew my request for evidence, and add a few more quesions. Is it prescriptive by fiat or by consensus? If by consensus, please show me where it was established. If by fiat, is there any public link to it? If by fiat, then why is it not followed by all foundation projects? If by fiat, then why does the header on the policy page refer to reflecting consensus? Thanks, Bovlb (talk) 23:08, 14 May 2008 (UTC)


 * Policies are by definition prescriptive. Here is the most relevant definition from Wictionary: "A statement of commitment to a broad requirement, often used in an organisation to instruct personnel as to a required outcome." That a policy may be adopted by consensus does not mean it has to reflect actual practice, it may establish an important goal. The BLP policy speaks for itself on this. As to its origins, you can slog through the talk archives. It started as a proposal, went through many revisions and was finally adopted as a policy. Influential people weighed in. There is no confusion here, just an attempt by a strong-willed editor to push a bizarre view of policy formation that has been roundly rejected and which he now admits was unhelpful. Enough is enough.--agr (talk) 12:01, 15 May 2008 (UTC)
 * A small correction to my comments: I don't think anyone has made the "prescriptive by fiat" claim since 2008-05-12 when I asked for evidence. It seems to have been replaced with the "prescriptive by consensus" claim.  Further, the latter claim appears to have two variations, regarding whether BLP/3RR is special in this respect, or if it applies to all policies.  I know this has been a long slog, but we do seem to be making some progress in working out how people see things and why.  Bovlb (talk) 18:17, 16 May 2008 (UTC)

Policies_and_guidelines says:


 * "Policy change comes from three sources:
 * Documenting actual practices and seeking consensus that the documentation truly reflects practices.
 * Proposing a change in practice and seeking consensus for implementation of that change.
 * Declarations from Jimmy Wales, the Board, or the Developers, particularly for copyright, legal issues, or server load. "

I think WP:BLP was adopted by a combination the latter two sources. It was developed through a long history of community discussion (which I have only sampled) but it also cites Mr. Wales' input numerous times. --agr (talk) 21:01, 16 May 2008 (UTC)

I too have seen many comments from Jimbo (some linked from WP:BLP) about the importance of having the right BLP policy because of its real-world consequences, and I wholeheartedly agree. Not unreasonably, his remarks tend to be high-level, and do not set out our current policy in all its detail. In particular, while he certainly speaks to the important of removing certain material promptly, I have see no comment relating specifically to a 3RR exception.

So far as I can see the 3RR exception was added (in some form) to both WP:BLP and WP:3RR on 2006-05-17 by. Looking at the contemporary talk pages, WT:3RR had some discussion of exceptions, including the importance of not over-applying the rule, but there is no direct discussion of BLP and the editor did not participate. I liked Geni's comments in this discussion, "The 3RR must be applied equally or not at all." and "Almost no one breaks the 3RR without thinking they are in the right." Bovlb (talk) 23:44, 16 May 2008 (UTC)


 * UninvitedCompany was Arbcom at the time (and still is, IIRC), most likely acting on the committee's decision. You will need to do more than just slog through the talk pages and history here, because this decision was not made in a bubble, nor can it be changed in one. --Faith (talk) 14:11, 17 May 2008 (UTC)
 * According to Template:ArbitrationCommitteeChart, UninvitedCompany was appointed to ArbCom from 2004-01-16 to 2004-02-16 and elected from 2007-01-01 to 2009-12-30. Clearly the links above are not sufficient to prove that there was no discussion anywhere, just that there was no discussion in the obvious places.  Looking just at the edits, I can't tell whether he was being bold, or acting on the basis of some unreferenced discussion.    By the way, I'm not sure how to characterise this discussion page, but I would not describe it as a bubble. Bovlb (talk) 17:04, 17 May 2008 (UTC)
 * What I meant was that most article changes (and probably policy changes as well) have their discussion mostly on the related talk page. However, BLP is and was a very big issue, and there were a lot of circumstances surrounding the changes made that are probably not evident in the archives of the BLP policy page. The history of the decisions may encompass discussion and decisions beyond what is found in the archives of this talk page; that's all I was saying. --Faith (talk) 17:47, 17 May 2008 (UTC)
 * I absolutely agree (except for the many changes that have no prior discussion, of course). I'm making an effort here to find support for the claims that the BLP/3RR has a strong background in consensus, and that BLP is prescriptive in some special way that ordinary policy is not (whether founded in consensus or in fiat). I came to this discussion as a supporter of the BLP/3RR exception, but I am now undecided, largely because I see proponents of the exception making very strong claims without the evidence to support it.  While we did eventually get into discussing some specific examples of it working and not working,  there seems to be some strange barrier that makes it hard to discuss aspects of BLP policy on their own merits, even when the discussion is whether they contribute to the goals of the policy.  Bovlb (talk) 19:19, 17 May 2008 (UTC)
 * Yes, of course. Another place to look would be within the archives of 3RR (such as Wikipedia_talk:Three-revert_rule/Archive_4). There seems to have been heaps of discussion, which makes sense. --Faith (talk) 19:58, 17 May 2008 (UTC)

There is no strange barrier here. Regardless of the history, WP:BLP is the current policy and the 3RR exemption is part of it. Some think it's not helpful. I and others disagree. Without a strong consensus for change, it stays. (Maybe even with a strong consensus change will be vetoed, but we are far from confronting that.) That some admins are not respecting the 3RR exemption is not a reason to change the policy. Admins are not the community, they are servants of the community, and when exercising their special powers they are bound to follow policy. Sure, there is room for discretion, but that does not extend to punishing editors who act in good faith according to written policy. I'm not trying to second guess admins who are trying their best, but if there is confusion here, better communication is what is needed, and perhaps some clarification of policy as I have suggested above. --agr (talk) 02:07, 18 May 2008 (UTC)

What you say about needing consensus to change policy, and the need to encourage everyone (admins included) to follow best practice are both good points. I feel I need to expand on what I meant by "strange barrier". An editor started this thread with two questions, essentially (my paraphrase): While we did eventually get some discussion going on one or both of those points, there were numerous responses that fell into the following categories (again, my paraphrase): Now, you could probably argue that I'm exaggerating one or more of these points, but I'm afraid that that's the way it comes across to me. I don't want to be contentious, but frankly it seems to me that repeatedly making these claims adds a lot of unnecessary friction to what ought to be a relatively simple issue. If the exception is highly beneficial, then that ought to be easy to show. If the exception is followed in practice, then again that should be clear. If the exception is sound, but it's not being followed in practice then that's something we should know about and act on (for the sake of BLP). If the exception is unsound or has mixed results, then we should definitely consider tweaking it. If the exception is unsound but is being forced on us from above, then again this is something we should know, and there are some actions we could take. None of the enumerated claims seem like good reasons not to discuss these questions. Trying to quash discussion with strong claims like 1 and 2 when you're not prepared to substantiate them, borders on disruption. Sorry to speak so plainly, but that's what I mean by saying I see a "strange barrier" here. For the avoidance of doubt, I have seen many good contributions to this thread, and I thank everyone involved for their participation and patience. Cheers, Bovlb (talk) 02:33, 19 May 2008 (UTC)
 * 1) Is the BLP/3RR exception followed in practice?
 * 2) Is the BLP/3RR exception a net good, given its (potential) drawbacks?
 * 1) It's pointless to raise either of these questions because the BLP policy was established in its entirety by fiat.
 * 2) BLP is a special policy because it is prescriptive, therefore it is pointless to ask whether it is descriptive.
 * 3) It's pointless to ask whether any aspect of policy is beneficial or has consensus because changing it would require consensus to change.
 * 4) It's important to have a strong BLP policy, the BLP/3RR exception appears on its face to strengthen our BLP policy, therefore it's pointless to ask either question.


 * I see it completely different from you. The way I see it, in the opinion of many people, most of who regularly work on BLP issues, the 3RR exemption gives teeth to an important policy. This is difficult to quantify but at the very least they therefore oppose it's removal without GOOD evidence it's harmful. So far, all we have is a few examples of extreme cases where it didn't work, but no real evidence of significant harm from the 3RR policy (and indeed, most of these examples only came after editors began pointing out that there was little reason why those supporting the exemption had to provide all the examples that were being demanded from them). I'm sure editors currently opposed to the removal would be willing to discuss the issue, if we actually had something to discuss. But coming here, say 'I don't like it' or getting into long technical arguments isn't particularly constructive. The simple fact is, for better or worse on wikipedia the burden is always on those proposing a disputed change to show why it's necessary, and so far, (and I admit I haven't followed this discussion completely so forgive me if I missed something) we haven't see anything showing why it's necessary to remove the 3RR exemption and indeed, there doesn't even appear to be a great number of editors who do feel it's necessary to remove the exemption. Particularly given that there are plenty of people who do feel it's necessary, even if they can't provide strong evidence, it seems to me that it's resonable that they don't feel the need to discuss it further, until something new actually comes up. And that this lack of participation by those opposed to the change, does not imply consensus for a change, but simply a general feeling the discussion doesn't appear to be achieving anything. Nil Einne (talk) 06:37, 19 May 2008 (UTC)


 * I also have to take exception to Bovlb's statement. I tried a number of times to address the value of the 3RR exception and even proposed additional language that might clear up any confusion, but those comments were largely ignored. Instead the discussion revolved around a claim that policies on Wikipedia are descriptive, not prescriptive and therefore evidence the policy isn't being followed invalidates the policy.  I find this argument completely bogus. Policies and guidelines says up front "policies are considered a standard that all users should follow." That means they prescribe behavior. All our policies are ignored to varying degrees. There is no difficulty finding numerous examples of POV articles, original research, incivility,  etc.  A count of "citation needed" tags with dates over 3 months would run in the ten thousands, if not more. Should we drop NPOV, NOR, CIVIL and V? WP:BLP in particular is clearly states is purpose as changing behavior on Wikipedia. Until we put  stake in the heart of this spurious descriptive vs. prescriptive argument, no progress is possible.--agr (talk) 18:59, 19 May 2008 (UTC)


 * Arnold, I'm not sure where you're coming from. Policies and guidelines are descriptive, with a very few exceptions imposed by the Foundation or American law. That's how it's been since the beginning - editors do stuff, and eventually stuff gets written down in proposed guidelines, and as they go unchallenged, become guidelines and even policy. That's the whole point of the very first 'prescriptivist' policy ever written down: IAR. It's worth noting one of the things written on the archived page IAR links - "Rules are established according to the vigor of their enforcement."
 * Rules are as editors do. Form follows function. We're the OED, not the Academie Francaise. I remember back in '04, this was never a real debate - the prescriptivists were the odd men out, what with moaning about pseudonymous editing and how we had too many articles and too much vandalism (and, incidentally, pushing draconian policy suggestions reminiscent of BLP thanks to the Seigenthaler thing). Do you still remember the discussions back then in wikien-l and #wikipedia and the other places?
 * The point of policyspace has never been "we (the cabal) think editors are being naughty; what rule can we devise to stop this?" It's always been "People are doing this sort of stuff; let's write it down for convenience and for the newbies to read."
 * 'A count of "citation needed" tags with dates over 3 months would run in the ten thousands, if not more. Should we drop NPOV, NOR, CIVIL and V?'
 * I don't see what CIVIL has to do with that... But your example is a perfect descriptivist example. Standards change. We didn't use to be nearly so interested in referencing things, and the policies and guidelines were correspondingly lax. As editors convinced others that sourcing was important, other editors like Ta bu shi da yu developed tools - - or the s, or the various other templates and extensions we readily use these days. It wasn't that Jimbo said from on high that "we need more sourcing" and the policy gnomes scurried about to reinforce the language and summoned their legions to ride out into the Wikipedia countryside to slay and rape and pillage the unreferenced heathens, wasn't at all.
 * Until y'all finally put the stake into the heart of the idea that prescriptivism is or should be the way Wikipedia governs itself - and not by editors working hard, discussing intelligently & sensibly & in good faith, and reaching compromises (if I may paraphrase Jimbo), you're always going to be getting push back and disputes. --Gwern (contribs) 20:00 19 May 2008 (GMT)

First of all, I think civility suggests reading what the editor you are responding to actually wrote and not quoting them out of context. In particular please read the sentence before the two of mine you excerpted.

Second, we do have rules on Wikipedia and becoming an effective member of the community involves learning them and understanding which are flexible and which are not. That was true when I started in 2004 and it's even more true now. Pretending it isn't so helps no one. The two policies we are talking about, WP:BLP and WP:3RR set clear limits on what editors may do. 3RR starts out "An editor must not perform more than three reverts, in whole or in part, on a single page within a 24-hour period." (emphasis in original) How does IAR apply here? Well there are exceptions: clear vandalism for one; also copyright violations and unfree content. And violations of WP:BLP. What we are discussing here is getting rid of the exception for BLP, i.e. making 3RR even more prescriptive.

The BLP 3RR exception does not restrict newbies, it gives them freedom. The folks it restricts are admins. Admins are no different from other editors except that they are given certain tools that are particularly subject to abuse. Admins have a greater responsibility than other editors to follow the rules when they are exercising these powers. WP:ADMIN says: "When a policy or communal norm is clear that tools should not be used, then tools should not be used without an explanation that shows the matter has been considered and why a (rare) exception is genuinely considered reasonable." To have it any other way would create a situation where no matter what the community consensus was, admins would have the final say.--agr (talk) 22:14, 19 May 2008 (UTC)


 * Note for clarity: Kim has alerted me to this discussion; he probably won't like my answer. :-)
 * Given that I've had a (strong) hand in developing both BLP and, somewhat(!) further back, 3RR, I would like to help answer the question, shedding some light and avoid fruitless circular discussion.
 * Firstly, having read this talk page, much of what I see is a great deal of theorising as to the nature of policy - commendable, but surely Wikipedia talk:Policy would be a better venue?
 * To the meat of the matter, as I see it:
 * We (ArbCom/OTRS people/etc.) inserted an explanation of common sense (or, if you prefer, "exemption") around enforcing the BLP policy into the 3RR wording after a few complaints by people more willing to apply the "rules" as written rather than common sense, something which is often distressingly uncommon.
 * Kim opposes this extension; some significant time subsequently, and without notice to the drafters of the policy (or, indeed, anyone), Kim changed it to mean something rather different (raising the threshold to an unrealistic level, precisely against the BLP concept of being better-safe-than-sorry).
 * FCYTravis (and others) point out that BLP and the concomitant policy variations and exemptions that abound are not general policy, but "first-degree" level, on which it is rather inappropriate to Be Bold (which, as those with long memories can attest, was meant only to refer to fixing articles), and (correctly, IMO) reverts such a major, unannounced, and undiscussed change.
 * This page ensues. :-)
 * Certainly, we expect sysops to be rational and intelligent when applying policies (though this is perhaps an irrational desire), so to an extent, as encapsulated by IAR, all policies are descriptive rather than prescriptive. My favoured wording of this is that true policy is an unknowable thing which exists in both written and unwritten form, beyond specific wording; any particular instantiation of words in an attempt to encapsulate a policy will be necessarily flawed and imperfect.
 * However, I would strongly suggest that BLP (along with NPOV, NOR, and V) are such fundamental content policies that it would be better to consider them, if not set in stone, at least needing an exceptionally strong and well-supported argument to change.
 * James F. (talk) 22:48, 19 May 2008 (UTC)

Hmm. We seem to be talking past each other. Let me clear a straw man out of the way, and then I'll try to restate my questions as briefly and clearly as possible.

First the straw man: No-one (on this page) is saying that policy is purely descriptive. Clearly it exists to guide people into best practices and away from counter-productive behaviour. That does not mean, however, that policy stands entirely independent from what people do. It's informed and usually inspired by experience in the field.

So, to my questions: Should we be interested in the size of the gap between written policy and common practice? Should we be attempting to close that gap? If the answer to both is "yes", then let's get on with discussing that, with as little friction as possible. Bovlb (talk) 23:24, 19 May 2008 (UTC)


 * To the extent there is a significant gap between written policy and common practice with regard to 3RR and BLP, we should certainly try to close it. But it seems to me the first thing to try is starting a conversation between the admins involved in 3RR enforcement and the folks at the BLP noticeboard. They are the ones best in a position to work things out. This page is for discussion changes to the BLP 3RR policy and that should be the last resort. I think it's time to put this thread to bed for now.--agr (talk) 02:25, 20 May 2008 (UTC)
 * It seems that the people making the most outrageous claims have long since stopped bothering to defend their position here, so perhaps I'll take up your suggestion. Bovlb (talk) 03:30, 20 May 2008 (UTC)