User talk:Doc glasgow/The BLP problem

Another reason Wikipedia struggles with BLPs

I think this is a great page - kudos for it, Doc. Two things I'd add under "Why Wikipedia struggles with BLPs" (either together, separately, or as part of the existing points):

• Many BLPs are not monitored by regular editors (while all articles are on somebody's watchlist, a given article may only be on the watchlist of retired, malicious, or infrequent editors), which means that the things under "Ways in which a biographical article can be harmful to the subject" can persist for extended periods of time. Even for those articles being nominally watched by frequent editors prepared to faithfully apply WP:BLP, subtler disruption can sometimes sneak by.
• It's nobody's job to fix BLP problems, which can cause a breakdown of accountability (i.e. if a false allegation stays in an article for three months, it's nobody's fault (except the vandal, from whom Wikipedia can't reasonably expect any accountability in the first place). In the real world, if somebody goes on a crime spree, primary responsibility rests with them, but the public can also expect accountability from the police (broadly defined) who failed to stop the crime spree once it starts. Sarcasticidealist (talk) 19:45, 31 March 2008 (UTC)

Some comments

One thing to note is that false allegations can easily start off as vandalism. AFAIK the Seigthaler controversy was one example (admitedly it wasn't that sneaky) and I'm sure most of us have come across (not necessarily in BLPs) examples of sneaky vandalism where the person changed or added information which may have seemed like a constructive edit but was not. Also, with wikipedia increasingly being used as a source, there is a strong risk for use inventing trivial stuff like real names, birth dates etc. For example it seems to remain unclear if Peaches Geldof full name is really her full name (see the article talk page). While this isn't quite as harmful as inventing stuff like XYZ is a terrorist who was convincted of raping 100 women and men, it's still not good Nil Einne (talk) 19:48, 31 March 2008 (UTC)

Vandalism is an unhelpful category in this regard, as it focuses on damage to wikipedia, not to the subject. That's why I wanted to differentiate between routine vandalism (whether sneeky or not) and credible and damaging mis-information about living people.--Doc^g 19:53, 31 March 2008 (UTC)

I think one of the problems is you risk creating a distinction that doesn't necessarily exist. While there is a difference at the extremes, there is a very, very large grey area. For example, if I say "Britney Spears is a terrorist who supports Al Qaeda" (or GWB, Blair, John Howard) it probably falls into the mostly harmless vandalism category but sadly if I were to add this to anyone who either is, or may appear to be a Muslim or Arab, it's far more likely to fall into the harmful misinformation category. (And to some extent, it also depends on how well known the person is). In other words, while I don't see anything wrong with making the distinction, I think it's important that you make it clear that while there is a difference, editors still need to be aware that stuff which seems absurd to you, may not seem absurd to all readers and that a lot of false allegations are intended as vandalism, in other words to make it clear that vandalism of BLPs can be incredibly harmful even though the vandal probably thinks it's 'funny'. I'll try to do this a bit Nil Einne (talk) 20:10, 31 March 2008 (UTC)

Possible, but not really. Wikipedia is over obsessed with the vandalism problem. 95% of it isn't a significant problem.--Doc^g 20:22, 31 March 2008 (UTC)

Also another thing it isn't just about small town gazettes and poor fact checking, I think a lot of editors struggle with the idea or why even though something may have been published in some source and even though it may be true, it's not always necessary to include it in the article and even if it is, it doesn't necessarily merit prominent, extensive mention. Wikipedia is increasingly the top or at least very near the top in most internet searches and is probably one of the first points of call for most people looking for information on something including a person. So for a lot of people the fact that you may be able to find something out about them if you search hard enough isn't really an issue. The fact that if you read the wikipedia article, the first thing it tells you is 'XYZ is/did/whatevered' is much more of an issue. I think the local level part in particular is problematic since some stuff may not be so much local as special interest. For example, if someone does an interview with some small time perhaps special interest magazine, they may disclose some stuff which while true, is not necessarily something they're expecting people to read in a biographical article, particularly given extreme prominence. Nil Einne (talk) 20:10, 31 March 2008 (UTC)

1. By local I meant non-prominent, whether that's geographical or special-interest it amounts to the same.--Doc^g 20:22, 31 March 2008 (UTC)

copied from my talk page

Thanks very much for your piece on the BLP problem.

Question: You wrote "There is an imbalance between a hard-pressed OTRS volunteer, trying to be helpful, but dealing with dozens of articles per day," Are there literally dozens of OTRS issues per day?!

I want to take more time to think over what you wrote.

One thing that bothers me re BLP articles is what I see as an "attitude of entitlement", on the part of Wikipedia editors, to publish whatever we want as long as it is not specifically illegal or against Wikipedia policy.

To me, this "entitlement" came through strongly in the recent discussion of a BLP article titled Merle Terlesky. This case by the way is also an example where the article subject tried to correct what they saw as a problem by doing some Wiki editing, and was abused in return.

Best wishes, Wanderer57 (talk) 20:23, 31 March 2008 (UTC)

PS If/when I'm going through 'The BLP Problem' again, is it a problem if I fix some typos on the way through?

By all means fix typos, please. I'll think a bit about your other comments. I intend to add to this essay as comment some in.--Doc^g 20:26, 31 March 2008 (UTC)

Entitlement is arguably a problem throughout modern western society, and amongst young people who haven't been tempered by experience but your "attitude of entitlement" description, Wanderer, is the best summing up of BLP problems I have seen, hence my commenting. Thanks, SqueakBox 20:27, 31 March 2008 (UTC)

Blimey

I wasn't aware here were 127,000 blp articles with referencing problems. I guess I kind of figured that someone else was looking after them. I guess a lot of us did. Hmmm. Shame we can't lock the database for a month and strip those articles back to the bare minimum. We really do need to consider a way of getting on top of the mess we have under us. Maybe a database lock for a month is too much, but could we get a semi-protection lock for a month? Maybe if we gathered up a few editors and brainstormed, we could work up some tools to get the whole database better evaluated and tagged? Or, how feasible would it be to get a database dump of the blp articles, work on them on a mirror and then copy them back and semi-protect them. Just thinking out loud. 127,000 articles. Hmm. Even at 100 each that's over a 1000 editors. Those numbers are frightening. Can we jettison them? Accidental server crash? :P We're really under performing. Hiding T 22:23, 31 March 2008 (UTC)

• Another thought. Like how we had the great image re-organisation, could we not work on getting the blp articles without references into some sort of category structure and templated to the point that if they don't get referenced within say a month of being tagged, they will be deleted. A kind of prod system. I guess there are a few issues there, mostly with people removing the tags and upsetting the category structure without doing the leg work, but I would suggest that could fall under disrupting Wikipedia. Hiding T 08:21, 1 April 2008 (UTC)
  • We already have Category:Articles with topics of unclear notability. See also this listing. Just make "unreferenced biography" a valid criterion for deletion and sorting will be much faster than it is now. (If you can get consensus for this criterion, that is.) --B. Wolterding (talk) 20:40, 11 April 2008 (UTC)

Comment: fine and insightful

This is one of the best essay pages I've seen in recent memory. Not sure what the answer is, but indeed defining the problem, and identifying the underlying issues, are the places to start.

One of the problems -- I'm a heretic now, and admitting it, for this is a thoroughly unpopular opinion here -- is that Wikipedia does not have enough adults. That attitude of entitlement which you mention, along with its associated irresponsibility, are adornments of youth; live a few more decades, and you acquire an appreciation for just how damaging the unsupervised practice of "free speech" can actually be.

Only noticed this because of the thread at WR. Funny how that is. Continue writing; I'm happy to see a good critical mind at work. Cheers, Antandrus (talk) 03:23, 1 April 2008 (UTC)

Great stuff....

I'm really pleased and impressed reading this - so thanks Doc! - there's a small danger that the 'breaking it down' misses the fact that the overall impact of an article's existence is greater than the sum of the parts you describe - but that's worth putting to one side for the sake of debate.

There's a kinda similar chat over at Lar's blog here (pitched in terms of a discussion of the sustainability of anonymity - but related) - I think there may also be merit in creating a 'problem in a nutshell' section - for clarity maybe? ...generally just well done and thanks! - will we see some ideas for next steps at some point? Privatemusings (talk) 05:15, 1 April 2008 (UTC)

Point of View?

Quoting the draft article: "On the whole Wikipedia has been reluctant to compromise secondary principles to reduce biographical damage. Even the limited rules of the biography policy have been controversial and are still widely disliked. However, Wikipedia has compromised its freedom for the sake of biographical damage in the past. In response to the Seigenthaler incident, Wikipedia removed the ability of non-logged in users to create new articles. Whether that limitation was effective is another judgement to make. Further, that controversial move was made by the authority of Jimbo Wales rather than by community agreement, whether Wikipedia's Godking has the muscle to make such decrees today is questionable, and whether the community could ever reach consensus for such a bold move even more questionable."

After reading this paragraph, I'm wondering from what point of view this article is being written.

Removing the ability of non-logged in users to create new articles doesn't seem to me "such a bold move".

I guess I'm way out of step with Wikipedia.  ? Wanderer57 (talk) 15:56, 1 April 2008 (UTC)

It isn't draconian. But compared to Wikipedia's normal ponderous conservative caution, it was a quick, decisive change. I suspect with little actual impact.--Doc^g 16:01, 1 April 2008 (UTC)

Minor points of clarity

Hello Doc:

These are a few minor questions that came to mind while reading the article. Answering them might help clarify.

"Although it does raise the question, is there a price for subjects which is simply too high?" I can guess a meaning but it's not clear.

"Angela Beeselys" OR "Angela Beesleys"?

"The loss of the Brandts, Beeselys and Murphys is low," I'm guessing this means 'under the present system, there are only a few losses of this type' but it's not really clear.

"Subject's 'opt-out' II (only for lower notables) Ease of implementation: Higher. " "Higher" meaning easier to implement?? I think having to draw a line between notable and less notable every time there is an opt-out request would be a lot more work.

Thank you. Wanderer57 (talk) 16:28, 1 April 2008 (UTC)

WP:BLP

Yes, this is why we have the WP:BLP policy and why it should be improved to say that semi-protecting BLPs should be standard procedure and why semi-protection itself should be strengthened (increase the number of days and number of edits required). WAS 4.250 (talk) 18:00, 1 April 2008 (UTC)

Agreed on the first point (largely neutral on the second). At what point and in what venue do we propose this change, though? Sorry if this is too off-topic from your essay, Doc. Sarcasticidealist (talk) 18:02, 1 April 2008 (UTC)

In my experience with wikipedia policy, change only happens slowly and must be driven by actual problems to get traction. So I recommend talking about this until the next BLP crisis and then amending WP:BLP to say something like "Semi-protecting one or more BLP articles for long periods is allowed whenever any admin thinks it will help to 'do no harm'." I would wait for the crisis after that to strengthen semi-protection; but that may not be needed, if by then we have implemented stable (flagged) versions. WAS 4.250 (talk) 16:25, 2 April 2008 (UTC)

I am not convinced semi protection will make much impact on the hatchet jobs.--Doc^g 16:51, 2 April 2008 (UTC)

Per discussion over at (link to BADSITE redacted) I think it will help quite a bit, although it obviously won't solve the problem (people can still register and wait a few days, people can still hatchet living people in articles whose subjects aren't living people, etc.). But I think by pointing out ways that vandals and other troublemakers could circumvent our protections, we're probably overestimating their motivation. In most cases - some troublemakers are obviously absurdly motivated. Sarcasticidealist (talk) 18:06, 2 April 2008 (UTC)

I've gone ahead and proposed it at WT:BLP. Sarcasticidealist (talk) 22:54, 2 April 2008 (UTC)

I am in favour of permanently semi-protecting all BLPs. Wikipedia cannot afford to allow anyonymous and new editors who lack a history of positive contributions to make changes to BLPs, which are not to be taken lightly according to policy. We don't know whether these people are editing in good faith, or whether they think they are improving the article but are clueless about how Wikipedia really works on a wider scale, and their good faith edits are actually harmful. If implemented, we should add a special "BLP semi-protect" message, explaining how this works and asking anonymous and new users to discuss changes they want to be made on the talk page - good idea perhaps?--h i s ^{s p a c e} _{r e s e a r c h} 17:18, 3 April 2008 (UTC)

Mitigate/militate

Pardon me for stumbling over this in the headline: something can mitigate a problem (reduce its severity), or it can militate against a problem (counteract it); but it cannot mitigate against a problem. Jayen466 22:11, 2 April 2008 (UTC)

Lack of responsibility

Suggest to add point (c) based on Wikipedia:NPOV#Neutrality_and_verifiability

c. asserting that because a fact is both verifiable and cited, should therefore be included. An attitude that dismisses concerns related to undue weight, non-neutral fact selection and wording. It is important to note that verifiability lives alongside neutrality, it does not override it. Good editorial judgment is always required, in particular in BLPs.

≈ jossi ≈ (talk)

Yes. Thanks, SqueakBox

The other side of the coin

Excellent essay, thank you very much. The issues of "who is responsible for fixing it?" and "low-notability subjects dominated by a highly movitated but small group of editors" are critical.

One thing is that the problems you discuss can work both ways. While you discuss mostly negative BLP articles, these issues can result in overly positive BLPs for similar reason, with the added problem that a tendentious editor can claim BLP justification to edit-war in defense of the subject, since they are exempt from 3RR. Also, the flip side of "who is responsible?" is ownership, which can also be a problem. Those close to the subject will rightly feel they know the subject better, and have a duty to block any criticism (no matter how neutral or NPOV). Of course, they probably feel that they are being fair and those seeking more criticism are POV-pushing, which is kind of the problem. Msalt (talk) 23:17, 2 April 2008 (UTC)

Totally agree with the above. BLP is often used as a club to excuse hagiography (I think the state of the Prem Rawat article at the time of the Register piece was awful). Any changes to WP:BLP need to ensure that genuine balance - properly referenced and relevant critical information - is still possible. But the magnitude of the damage is much, much less for hagiographies than for hatchet jobs, making preventing the latter much more critical than preventing the former. Sarcasticidealist (talk) 23:22, 2 April 2008 (UTC)

It is not necessarily the case that the magnitude of the damage is much, much less for hagiographies than for hatchet jobs. If a conman or other charlatan points someone to wikipedia in order to present his "credentials" and he or his accomplices have managed to keep the wikipedia article about him "clean" then he can use wikipedia as an instrument in his crimes or unethical practices in order to damage others, usually with a financial element. So this thing cuts both ways. Wikipedia BLP policies should not be capable of being used to bias articles in favour of people who are trying to hide truths about themselves for nefarious purposes. Where this is the case wikipedia can ethically perform the public service of publicising whistleblowers (obviously provided standards of verifiability, etc. are met).Matt Stan (talk) 10:04, 3 April 2008 (UTC)

I am an ex-follower of Prem Rawat, and have just read this generally excellent paper, which summarises the potential harm that can be done to minor notable people, but fails to address the harm that can be done to victims of the actions of some such people. My immediate thought when I read the principle 'do no harm', was exactly what you have said, Matt, so I came here to see if anyone else had raised it. So thank you. Prem Rawat has done harm to many people and my only interest at present in Wikipedia is to do what little I can to stop Wikipedia being an avenue for Rawat to do harm to more people. Even now, Rawat expects from his students a life long commitment, and hence dependence, in which no student ever graduates.--John Brauns (talk) 19:08, 3 April 2008 (UTC)

Yes, there are isolated cases in which hagiographies can do serious harm. But every single hatchet job does serious harm, so I guess maybe when I say that the magnitude of the damage is much greater for hatchet jobs, I'm speaking in aggregate. But I do still agree that BLP should not be used as a club to remove legitimate and properly-sourced criticism or other "negative" information. Sarcasticidealist (talk) 17:46, 3 April 2008 (UTC)

Good essay. Another, more extreme, action could be to block editors who create bios whose intention is negative against the subject. Let fans or neutrals start them and then critics can have their say too. Steve Dufour (talk) 00:12, 3 April 2008 (UTC)

I can see enforcement troubles with that. But regardless, I think the great problem facing Wikipedia where BLPs are concerned is not solving problems, but preventing them; Wikipedia has become so large that, even if problems are easily solved once detected, the window between when they are created and when they are detected is far too large. Sarcasticidealist (talk) 17:46, 3 April 2008 (UTC)

Inform the Subject

One point raised recently is that many subjects do not know when a Wikipedia article on them is created.

If true, a important control on BLP articles is being lost. If the subject knows about an article, I imagine that in many cases they will check the article or have someone do so.

I suggest a policy of informing a subject of an article that is scheduled to be posted, and sending an advance copy.

If there is no response in a certain timeframe, or a positive or neutral response, then the article is posted.

If there is a "negative" response, eg, "this is a crock of......", "I never did that", "that was ten years ago and he was over 18", etc etc. then the editors have some specifics to look at to improve the article. This is not, repeat not, to suggest that anything to which a subject objects should automatically be changed. But the feedback would serve to identify parts of the article deserving of closer scrutiny.

I think this would increase the feedback to Wikipedia about errors, and bring a stronger level of quality control. It might also decrease Wikipedia's liability for errors.

Wanderer57 (talk) 18:51, 3 April 2008 (UTC)

Vague, rambling comments

Would what happened in the Seigenthaler incident be unreverted nonsense? That said, I'm not sure why he was as upset as he was - was he worried that people would think that he shot two politicians, or was it because their murders were traumatic to him.

One problem that can occur with hatchet jobs is that it can start off as poorly written screed, but thanks to well-intentioned copyediting by others, end up as well-written but just biased, which can make it more harmful.

Should OTRS requests be added to the BLP noticeboard? That way, several people can ensure it gets looked after, possibly by people interested in the subject.

Maybe permanent protection should be introduced for BLPs that have had complaints for the author. Any changes could use {{editprotected}}.

One advantage to having subject opt-out is that it means that individuals don't need(?) to resort to legal or death threats in order to have a page removed.

One problem with semi-protecting BLPs is that it'd disable the subject from editing the article when they find out about it.

Another possible solution is to have some sort of template in the article itself about how to contact wikipedia about libel. Not one I'd like, but still. Talking of which, Wikipedia:BLP#Designated_agent seems to be slightly different from Wikipedia:Libel.

Also, will flagged revisions help the problem? Andjam (talk) 05:27, 5 April 2008 (UTC)

Dealing with gross address violations

Is there a point at which a privacy violation is so odious that the ISP or the police should be contacted, such as posting the address of a Muslim family who murdered their daughter (a case I dealt with recently)? Andjam (talk) 05:36, 5 April 2008 (UTC)

Complicated

BLPs are more complicated than they appear at first, and the more you know, the more complicated they become. For example, a BLP I am involved with has an author who has coauthored 10 books with his wife. He has announced this in interviews and in his autobiography, and this can be readily verified by looking at any bookseller like Amazon that he has written books with his wife. We stated this in his BLP on WP, but someone purporting to be him (are we even sure? I doubt it) contacted OTRS to complain about this, and threatened to sue us unless we stated in his biography that his wife was not a coauthor on any of his books. So in this case, I would ask:

• Are we sure that the email we received was from the subject? Can we ever be sure, even if we contact him, or his publishing company or agent?
• How certain are we that the subject is even sane or rational or honest (a nontrivial question, and the issue of mental instability becomes more prominent as one knows more about the subject, and a difficult circumstance becomes all the more probable the more one knows, although none of the information leading to this line of reasoning is in the public domain or our article, nor would I want it to be).
• How do we know that if we change our article to supposedly suit the subject by removing the description of his wife as a coauthor, we will not offend the wife, and cause further trouble?
• Why would we take an unverified email as a source that conflicts with 10 or more other reliable sources?

So in this case, almost any decision is "wrong" in one way or another; removal of the biography, stating his wife is a coauthor, stating his wife is not a coauthor, following the sources, not following the sources, and so on. It is impossible to satisfy everyone and every rule, since each have conflicting requirements.--Filll (talk) 04:09, 7 April 2008 (UTC)

Wikipedia:OptOut

Having been aware of some of these issued for some time, and having read with great interest the comments above, as well as Doc's work, I felt that a concrete proposal for discussion is timely... (actually, part of me feels like I'm rushing too much, and part of me feels this is way overdue, which I guess, on average, equates to timely!) - take a look, and all thoughts most welcome... Privatemusings (talk) 01:05, 10 April 2008 (UTC)

Statistics

Doc asked for some number crunching regarding the maintenance tags on BLP articles. Well, here are the top 10 maintenance categories that BLP articles fall into. (All data as of March 12, 2008.)

Maintenance category	# articles	percentage
Total BLP articles (Category:Living people)	259210	100.00%
Category:Articles lacking sources	13908	5.37%
Category:Articles with unsourced statements	13740	5.30%
Category:Articles needing additional references	5475	2.11%
Category:Orphaned articles	3157	1.22%
Category:Articles to be expanded	2511	0.97%
Category:Articles with topics of unclear notability	1971	0.76%
Category:Articles lacking reliable references	1918	0.74%
Category:Articles with trivia sections	1510	0.58%
Category:Wikipedia articles needing style editing	1420	0.55%
Category:Articles lacking in-text citations	860	0.33%

Note that some articles may be listed in more than one of these categories.

These numbers are slightly above the average of all Wikipedia articles. For example, 4.46% of all articles fall into "Articles lacking sources".

Based on these figures, I think one can get a better impression whether our BLPs are "maintained" or "maintainable". --B. Wolterding (talk) 20:34, 10 April 2008 (UTC)

Thanks for the figures, but the most crucial one is missing. What percentage of the existing BLPS are disputed enough for a drastic rewrite or deletion debate? Though a better question might be which of the BLPS have been checked and reviewed? (Probably very little, though the assumption is that someone read them at some point - maybe stats on pageviews or numbers of BLPs that have been New Pages Patrolled would also help). This is a problem very similar to the one about non-free images, in that there is a volume problem, a lack of clear statistics, a lack of management of the problem, and a similar solution in use (for disputed non-free images, the claim is sometimes made that images can be taken down on request - this is similar to letting OTRS deal with BLPS - dealing with problems as they arise). There is also the similarity of the strong community feelings this issue raises. My recent attempts at stats in the non-free image area can be seen at WP:NFCC-C, Wikipedia talk:WikiProject Image Monitoring Group#Responses and Wikipedia:VPT#Image stats. The currently tracked free and non-free images totals are 360,125 free images and 282,264 non-free images. There are a further 130,370 untracked media files, bringing the total to 772,759. This is of a similar order of magnitude to BLPs (259,210), which is why I said the same "volume" problems apply. The other point that should be made is that articles about people probably make up about 1 in 5 of all articles (either go to Special:MostLinkedCategories, and add together the "listas" categories, or go to Category:Biography articles by quality, where you see that the total of articles with {{WPBiography}} on their talk page recently broke the half-million barrier - currently 507124, though you need to take off an unknown number of music band articles to get the "single-person" biography figures, though of course music band articles count as BLPs. While you are at Special:MostLinkedCategories, see if you can work out why there is a discrepancy of over 30,000 between Category:Biography articles of living people (229,027 articles), and Category:Living people (265,878 articles)?). That half-million figure includes dead people as well, though at the rate we are going, most of the people currently in the BLP category will be dead before their articles are adequately reviewed and edited (presuming, that is, they are notable enough that it is easy to find out they have died...) - Category:Articles that were created while the subject was alive anyone? Seriously. Carcharoth (talk) 07:53, 11 April 2008 (UTC)

Of course, the numbers only reflect the problems that have been tagged, not all those that actually exist; and "disputed enough for a rewrite" can't be measured in this way. I also agree with you that the numbers rather indicate that the current volume of BLPs is unmaintainable (but wanted to keep the presentation of the figures neutral). As a comparison: The largest process on Wikipedia, WP:AFD, handles about 30.000-40.000 articles per year. So even using all these resources for BLP sourcing (which seems unrealistic), we would need half a year to clean up all BLPs in Category:Articles lacking sources. Another comparison: The notability backlog is about 15.000 articles long, going back in time about 1 year, and capacities barely suffice to deal with it. --B. Wolterding (talk) 08:14, 11 April 2008 (UTC)

One more thing. I have lots of questions where I would like to get actual concrete numbers, but it is very hard to actually find people willing to gather these sort of statistics. Wikipedia:Statistics has some stats, but not nearly enough, and Wikipedia:Statistics Department is inactive at the moment, but I would like to reactivate it to list people willing to gather and analyse statistics. Useful ones of course. Carcharoth (talk) 09:30, 11 April 2008 (UTC)

• One more point. I suspect that much of the BLP backlog coincides with the notability backlog. Lots of obscure borderline vanity articles that people haven't got around to AfD'ing yet. If they were written by the subject themselves, not a problem. It's the borderline obscure, subtle attack pages that need to be found. Carcharoth (talk) 09:54, 11 April 2008 (UTC)

Levels of notability

I kind of like the concept of "lesser notable" people that the text mentions; but it is not clearly defined as of now. Perhaps the following ideas can help.

One might introduce a notion of strict notability (choose any other term you prefer), stronger than what WP:BIO currently amounts to. I'd say a person is strictly notable if there are several sources (at least two) about him or her which meet all of the following requirements.

• They are biographical. That is, their primary topic is the biography of the person in question, his life, his achievements, etc. This would exclude interviews with the person on a non-biographical topic, coverage of an event in which the person participated, etc.
• They are independent of the person.
• They are reputable. Reputable sources in this context would include: mainstream national newspapers (but not the yellow press); books published by major publishing houses; mainstream print encyclopaedias; international scholarly journals. It would exclude the local press, tabloids, vanity publishers, and publication by "special interest groups" (political parties, religious organizations and so forth). The purpose is to single out sources which have a reputation of neutrality and proper fact-checking, and also to set good minimal standards for inclusion.
• They are substantial. The cutoff to be chosen here is necessarily somewhat arbitrary. I would suggest to choose something that can easily and objectively be measured; say the source must be at least 1000 words in length.

These sources should be identified in the article, and moreover, they should be the base of the article. Fulfilling the above, they are so substantial that they clearly allow to write a decent biography for Wikipedia. In case of any disputes regarding what to include in the article, we then have an easy solution: Restrict to what is mentioned in those sources.

The question is what to do with the "lesser notable" biographies, which do not meet the above criteria, but pass the current version of WP:BIO. One might argue that we should not have such articles in the first place; but I doubt that meets wide consensus at this time. Another option might be: In case of any disputes, complaints, etc. by whomever, we would remove such a "lesser" biography from main space, say to user space or to a separate "staging" namespace, where it can be improved until it meets the stricter criteria. (Note that I refer to "disputes" here, not "consensus" - if there are any concerns about a biography of lesser notability, the default should be to move the article out of main space, in order to protect the subject.) --B. Wolterding (talk) 17:21, 11 April 2008 (UTC)

I've said something like this in the past. The idea is that truly balanced biographies can only truly be written when you get some distance from a person's life and times. For example, the Oxford Dictionary of National Biography only writes about people who are dead. But for living people, the simple thing to do is only allow extensive articles for those who have had biographies written and published them (official or otherwise) by a reliable source. It is interesting to compare type and length of biographical material on a living person, compared to historical figures. It becomes quite clear that some living people are really only news stories. Carcharoth (talk) 17:29, 11 April 2008 (UTC)

A reputable source may be hard to agree upon. Even the New York Times has been accused of yellow journalism, and Damian Thompson's Counterknowledge argues that reputable publishers are nowadays willing to publish pseudoscience and pseudohistory. Andjam (talk) 13:34, 12 April 2008 (UTC)

• Yes, that would need some discussion; perhaps it would be best to set up a "white list" and discuss every entry individually. But I'm quite confident that we could agree on a reasonable core list. Most people would agree that biographical coverage in the New York Times is (in nearly all cases) free of libel, exaggeration and unfair bias. If our article is nothing more than a summary of this New York Times coverage, the typical BLP problems should not arise, except for very special cases. The current problem is rather that we do not have such sources for most BLPs. --B. Wolterding (talk) 22:18, 12 April 2008 (UTC)

Streisand effect

Is it worth incorporating this into this somehow? It occured to me that one of the issues is the Streisand effect (thinking beyond lawsuits here too) and the, dare I say it, 'immaturity' of some wikipedians. Ignoring the complexity of wikipedia rules, someone with a reasonable understanding of the internet probably knows they have to be rather careful in how they deal with things. Even with a legitimate issue, they will easily find themselves made of fun of and may easily encounter people who hassle them on (e.g. by doing their utmost to find a way to include whatever negative material they can) and off wiki (while not related to wikipedia, think of the way Lawrence Cooper was attacked over the Mass Effect controversy including attacks on her book by people who had never read it or for that matter, probably never even having heard anything about it [yes I get the irony here of her attacking Mass Effect without having played the game although she apparently relied on what she had been told by Fox News]). I mean 'immaturity' here because unfortunately often people who do these things aren't trying to be malicious, they simply somehow see it as to stop 'censorship' and think anyone who complains must be up to no good even when the community recognises the issue as legitimate. (There are IMHO far to many people who don't or worse, are unable to appreciate the spirit behind BLP, simply thinking it's to stop us being sued or because we're trying to be a serious encylopaedia or something of that sort) Because these people usually don't cross the line, they're unlikely to be blocked either even if their behaviour is unfair to the person they're targetting. Even if they are libeled by a wikipedian perhaps more then once, they may know it will almost definitely be worse for them to try legal action or to push this issue through other means due to the anticensorship brigade and seem to enjoy attacking other people for no good reason. In other words, it's not just that they have to avoid getting offside with the community, they have to avoid getting offside period and they have to avoid drawing attention to a minor thing. Nil Einne (talk) 20:16, 14 April 2008 (UTC)

Mirrors

You do mention mirrors briefly in this essay, but I think it deserves another mention in the section on eventualism. It is one thing to say, "Enh, it will get fixed eventually," when the "it" in that sentence is a single article on Wikipedia. It is another thing entirely when the "it" is dozens of mirrors, not all of which update regularly. --Jaysweet (talk) 20:21, 14 April 2008 (UTC)

Shooting George Bush

Honestly, Doc,

if you shoot George Bush on Monday, and by Tuesday there is no substantial biographical coverage about you in the mainstream press - would in not be better for Wikipedia to wait another day? Why should we, on sensitive issues, strive to be faster in research than the New York Times? --B. Wolterding (talk) 21:05, 14 April 2008 (UTC)

That's not the point. It would take at least six weeks before I'd make any "dead tree" standard (ie printed works other than newspapers). Any any standard that said being in this week's papers was sufficient to be a public figure - would really let anything in. But hey, why don't a buy a gun and experiment ;). --Doc^g 21:11, 14 April 2008 (UTC)

Another solution to consider

Rather than "delete on request of subject," allow the subject to open a "request for privacy" which will do the following:

• Determine if the subject is not notable, merely notable, or actually famous.
• If he is not notable, the article is deleted and and article creation blocked. The deletion log will indicate how to request an un-block of article creation if and when this person becomes notable.
• If he is famous, then the process stops and the subject has no special rights to control the article. He can still participate as any other editor, and can request oversight of publication of non-published information like social security numbers or his bank account, but then again, so can any other editor.
• If the subject is notable but not famous, he can request that the article be permanently fully protected, that uncited material be removed, that cited but harmful material be removed, and any edits that contain such material be WP:OVERSIGHTed. As a matter of convenience, his article will probably be added to a category containing all such articles. The permanent protection will be removed when he dies, becomes famous, or gives permission for the protection to be removed.— Preceding unsigned comment added by [[User:{{{1}}}|{{{1}}}]] ([[User talk:{{{1}}}|talk]] • [[Special:Contributions/{{{1}}}|contribs]])

Define "not notable, merely notable, or actually famous"? And really, you want to require a subject who wants privacy to have to come to an open wiki and publicly argue that he's a publicity-shy nonentity? Wow.--Doc^g 17:39, 21 April 2008 (UTC)

Ok, substitute "ask the Wikimedia foundation to look into the issue" for "open a 'request for privacy'." If the subject of the article isn't willing to take some step, I'm not willing to change anything for that article just because the person might be too shy to ask for a change.

As for not notable vs notable, that would be the same as today's standards. As for notable vs. famous, that will be a community decision and it boils down to "would locking the article against non-administrative edits as new news events happen do more harm than good." Perhaps rather than making the standard "famous" the standard should be "newsworthy and likely to be newsworthy in the future," i.e. new encyclopedic information is likely to be created and should be added without having to wait for an administrator.

If a newsworthy subject's article continues to be edited in an unfair way, then full page protection can be added to it on a case-by-case basis.

davidwr/(talk)/(contribs)/(e-mail) 18:15, 21 April 2008 (UTC)

The foundation will not involve themselves in content, period.--Doc^g 18:20, 21 April 2008 (UTC)

Ok, fine, replace with a mail-drop that is monitored by someone within the English Wikipedia who is in a position to take action. Create that office if necessary.

The details can be worked out later. The important part is finding some way to prevent harmful-to-the-subject content without negatively impacting Wikipedia's ability to serve as an up-to-date repository of encyclopedic knowledge. davidwr/(talk)/(contribs)/(e-mail) 18:57, 21 April 2008 (UTC)

True and False

"Untrue but credible allegations..." I had to stop right there. There is no way that we as editors of sources can determine whether an allegation is "true" or "false". Rather what we determine is, "Does the allegation come from a reliable source", "Can the allegation be cited and quoted to a reliable source" and so on. I would suggest changing this section to more closely follow policy-language so readers are not confused thinking that they individually must or should decide on veracity. Wjhonson (talk) 20:17, 21 April 2008 (UTC)

Not sure what you mean. I'm referring to a libel that's not obvious. "Bob Dylan was arrested four times for shooting the whole Kennedy clan with an Ozzie whist sleepwalking" is untrue but incredible enough not to reflect badly on the subject "John Seigenthaler was a suspect in the Kennedy assassination" is untrure but credible to someone who knows little about it - thus damaging. Or more extreme "Bob Smith is a an evil Martian" is untrue but incredible enough not to damage a reputation, but "Bob Smith filed for bankruptcy" - even if untrue is credible. I'm differentiating libels from patent nonsense.--Doc^g 20:26, 21 April 2008 (UTC)

However, if any of those statements appeared in a reliable source such as the New York Times as fact, they would be okay on Wikipedia until they were either retracted or otherwise demonstrated false. None of these statements, true or false, belongs in a BLP article or for that matter anywhere in Wikipedia without a suitable source. davidwr/(talk)/(contribs)/(e-mail) 22:26, 21 April 2008 (UTC)

Another type of harm

I don't remember the details of this, but there was a case where some scientist (archeologist, paleontologist, or something similar; I said I don't remember the details) had the same name as a game designer who happened to design games about a similar area of science. The result was that the scientist couldn't get a job, because everyone would find the Wikipedia page and assume he was a game designer who thinks that designing games gives him scientific expertise.

That doesn't seem to fall into any of the categories of possible BLP harm listed here, yet it's obviously harm. Ken Arromdee (talk) 21:24, 21 April 2008 (UTC)

True, but remarkably rare cases like that will not well illustrate the general problem. Further, if the game designer had his own website, the same effect might happen. Obviously that example is unfortunate - but I'd say very rare that a name confusion wouldn't be spotted by a same employer.--Doc^g 21:47, 21 April 2008 (UTC)

The best solution is to put a disambig tag at the top of the article. It could be worse: He could have the same name as someone on the United States No-Fly list. davidwr/(talk)/(contribs)/(e-mail) 22:28, 21 April 2008 (UTC)

This issue has been mentioned a bit at Wikipedia:Biographies_of_living_persons/Noticeboard#BLP_subject_ambiguity, for example with Russell Bishop (sex offender). Andjam (talk) 04:30, 22 April 2008 (UTC)

Your assessment of semi-ing all BLPs

Hi Doc - I disagree with your essay's verdict on semi-protecting all BLPs. Here's my reasoning; I'd really appreciate it if you're able to point out which of my assumptions you disagree with:

1. What you call "false allegations" are one of the most severe classes of BLP violation. I presume you agree with this, given your verdict of it as "highly harmful". Personally, I think it's actually the most harmful, taking into account severity and scope, but one needn't agree with that to follow the rest of my point.
2. A large majority of false allegations are inserted by IPs and new users. I have done no research to substantiate this, so you're welcome to disagree with me here, but based only on my own experience it seems self-evident.
3. A large portion of the people inserting false allegations are idiots with free time rather than malicious and devoted defamers. This can't really be more than a hypothesis, but I have little doubt that it's true, though I don't know whether this "large portion" is 50% or 90%.
4. Idiots with free time, in contrast with devoted defamers, are unlikely to jump through hoops required to bypass semi-protection.
5. Therefore, semi-protecting BLPs is likely to prevent a large number of false allegations from being added to biographies of living persons.

Thoughts? Critiques? Sarcasticidealist (talk) 22:18, 21 April 2008 (UTC)

1. "One of" perhaps - but my experience from OTRS was that the highly biased hatchet job was actually the more difficult to deal with.
2. Unsure - and depends what we mean by "new users". Remember that semi-protection is only good for brand new accounts, they can edit after a day or two, and again the hatchet jobber is often a slightly older account, as they tend to maintain their handiwork for a bit
3. Maybe. But the hatchet jobber is more determined, and certainly motivated if not actually malicious
4. Sure, but not sure idiots are the only problem at all
5. Maybe. But it won't stop the determined libeller or character assassin.

--Doc^g 23:30, 21 April 2008 (UTC)

Thanks for your response. I quite agree that semi-protection isn't a panacea, but I don't think anything is (including your proposal). I just think semi-protection would, in one fell swoop, eliminate a big chunk of the problem, probably (my hypothesis only) cutting "false allegations" down by at least half (and unreverted nonsense down by even more, although as your essay correctly states that's the least of our problems), and that the cost of implementation is likely to be negligible. I gather you disagree, though, so I'll stop trying to convince you. Sarcasticidealist (talk) 23:37, 21 April 2008 (UTC)

I'm not necessarily opposed to it - it will do some good. But I just see those that are obsessed with the dangers of "simple vandalism" using it to demand semi-protection of all our A-list celebs and world leaders, which would substantially compromise the "anyone can edit" on major articles with no gain to the subject. However, I'd certainly support the notion that any BLP which has been the subject of real (non-simple vandalism) problems should be semi-protected.--Doc^g 23:51, 21 April 2008 (UTC)

I should clarify: I support pre-emptive semi-protection of all BLPs, not for reasons of simple vandalism but because of the possibility of seriously malicious vandalism. I'm quite willing to sacrifice IPs' ability to edit A-list celebs and world leaders (although for such heavily watched articles, we could maybe deal with them through some kind of "requests for unprotection" policy or something similarly bureuacratic). I consider "Anyone can edit most of our articles immediately and without registering, and anybody can edit our BLPs after registering and waiting a few days" to be a subset of "anyone can edit", so I'm not concerned from that angle. In any event, thanks again for your response. Sarcasticidealist (talk) 23:57, 21 April 2008 (UTC)

Flagged revisions

Once we have flagged revisions, I'd hope that the problem would be reduced. For the sake of argument, I'll pretend that anonymous users will only see the last sighted version (if any), and that the sighting power (power of sight?) is handed out enough to have a significant effect, but not enough to have significant abuse. This ought to not only reduce the number of BLP violation, but also lower our liability for them, because the vast majority of readers will only see sighted versions. Then the question will be: How many BLPs have no sighted version? That we could organise a taskforce for. Bovlb (talk) 04:22, 23 April 2008 (UTC) Typos fixed. Bovlb (talk) 16:05, 23 April 2008 (UTC)

Unfortunately, I have being hearing the "one we have flagged revisions" move for 2 years now. I no longer believe it.--Doc^g 08:37, 23 April 2008 (UTC)

I hear you. Two things give me hope: an implementation does seem to be appearing, and the increasing pressure on BLP issues gives a clear motivation for deployment. Bovlb (talk) 16:05, 23 April 2008 (UTC)

The FlaggedRevs demo wiki has been open for some time. All that remains is to agree on how to use flagged revisions. Some people imagine there could be as many as 20,000 surveyors. If we can educate that many people about BLP issues that might solve part of the problem being discussed here. See Wikipedia:Flagged revisions for some policy discussion. Logged-in users would continue to see the most recent version of each article regardless of sighting, so they could still see nonsense. EdJohnston (talk) 16:54, 2 May 2008 (UTC)

This essay is about the Wikipedia problem, not the BLP problem

Most of the problems / concerns / mitigating factors / potential harms stated in this essay equally apply to many other types of less well covered articles, not just living persons, including dead people, products, companies, scientific theories, academic topics, publications, concepts, icons, deities, charities, fringe groups etc etc etc. I don't see anybody intending to address the same problems wikipedia has with these articles as it allegedly has with BLPs. Why not? I don't see why, if any ground breaking change in policy does occur from this essay, it doesn't/wouldn't/shouldn't apply to these other articles on wikipedia as a whole. MickMacNee (talk) 18:26, 24 April 2008 (UTC)

Well I must be talking complete bollocks then. MickMacNee (talk) 03:10, 27 April 2008 (UTC)

Do you not see any reasons why articles about living persons need special attention?

There is a well-known tendency of people to develop likes and dislikes of other people and to edit biographical articles to express those likes and dislikes. This means the likelihood of error and bias is greater in blp articles.

There is also the not uncommon situation where someone takes offense at something that is said about them and raises serious protests or takes a case to court.

There is also the point that, during the lifetime of person X, the person who is by far the best informed on the subject of person X is (in most cases) person X. Thus anyone else writing an article about person X is at a great disadvantage in terms of knowledge and needs to be especially careful. Wanderer57 (talk) 16:04, 2 May 2008 (UTC)

Have a quick look at the contents of the article. Is there any "problem" there that can not equally be said to apply to companies, political partys, or even events? For convenience I list them here: Un-reverted nonsense, False allegations, Biased editing, Privacy violation, Undue broadcasting, Eventualism, Lack of eyes, Anonymous (and pseudonymous) editing, Difficulty with OTRS, Difficulty with wiki involvement, The Need for permanent vigilance, Lack of responsibility, Privileging of wiki-norms over real-norms, Maintainability, "This article can be fixed" syndrom.

BLP is not a special case I am afraid. The point is, if you allow anyone to edit, you will get malicious, biased and just plain wrong information. If you don't, however, then you have lost what is special about wikipedia. Try Citizendium instead. Thehalfone (talk) 22:01, 28 January 2009 (UTC)

You're correct that these problems apply across Wikipedia. The difference with BLP is that misinformation in BLPs threatens to hurt real people (who are innocent bystanders, making this different from the case of the reader who uncritically relies on false information in a non-BLP and is somehow harmed as a result). As for what makes Wikipedia special, I'm going to make a radical suggestion: what is valuable about Wikipedia is not that it allows anybody to edit, but that it is an enormous collection of free (as in air and beer) information that is readily accessible to anybody with an internet connection. No doubt some of this is the result of the open-editing model, but that doesn't justify treating the open-editing model as a first principle. Sarcasticidealist (talk) 22:07, 28 January 2009 (UTC)

Which roughly translates as "screw living people" we want no constraints on our wiki.--Scott Mac (Doc) 22:06, 28 January 2009 (UTC)

"Everyone with sense knows the problem"

I could not disagree more. See http://en.wikipedia.org/wiki/Wikipedia_talk:Biographies_of_living_persons#the_slippery_slope_of_having_a_.22distress.22_policy_over_and_above_a_reliable_source_policy

NPOV and RS cannot be compromised.Bdell555 (talk) 20:26, 24 April 2008 (UTC)

Did you read the essay? The problem identified is that NPOV and V are not being universally applied to BLPs. Sarcasticidealist (talk) 16:33, 2 May 2008 (UTC)

(Essay has been archived: current link.) — Athaenara ✉ 22:45, 1 June 2008 (UTC)

a request for some eyes and ears.....

User:Privatemusings/OptOutNoticeboard - hopefully some here might be interested in taking a look - all thoughts most appreciated! - cheers, Privatemusings (talk) 11:49, 2 May 2008 (UTC)

Increased threshhold

I found the article well-thought out on the whole, though I noticed you didn't really address "Increase the notability threshold for BLPs". I'm not sure, but I think this option holds significant potential. Joe Schmoe wouldn't have to worry about prank or vandalized articles, and in general there would just be fewer articles to deal with. At the same time, it avoids the problems of opting out and similar options. Thus, I took the liberty of filling it out the section on your essay. Of course, this is your userspace, so feel free to remove it if you prefer. Superm401 - Talk 17:30, 13 June 2008 (UTC)

Lack of Eyes is key

Excellent essay across the board, many good points in the project space and the talk page. I think that Lack of Eyes is key. Changes in policy which don't increase the surveillance of neglected articles may be rendered impotent by lack of effective enforcement; on the other hand, increased scrutiny of problem articles may reduce the need for tougher measures, minimizing collateral damage. Therefore, I'd highlight:

• "3. The desirability of decreasing the number of under-watched articles (either by increasing the watching, or decreasing the articles)"

as our most important proximate goal. We should develop community incentives for watching neglected articles. More relevant to the remedies proposed in this essay, we can use the amount of surveillance an article receives as a litmus test for whether stricter rules (from semi-protection up to article deletion) should apply. Here are some suggestions for measuring how neglected an article is; feel free to comment on them, suggest alternatives, or discuss the broader idea if none of these specific implementation ideas hold water (I'm not technically proficient enough to gauge the feasibility of these suggestions, so think of them as brainstorms rather than firm proposals).

• Watchlists: Articles that aren't on the watchlists of enough responsible editors are vulnerable and bad edits there may not be reverted for a long time. BLPs could be semi-protected, subjected to stricter rules or deleted outright if they aren't watchlisted by enough editors, or by enough auto-confirmed editors. How many is enough? I don't know, largely because I don't how many watchlists such articles are commonly on now.

• Recent edit count: If an article is infrequently edited, that may mean there are few people viewing it who feel a responsibility to maintain the article.

• Talk page edit count: Potentially a better measure than article edit count, as constructive editors are more likely than vandals to engage in discussion on the talk page.

• Incoming links: the more incoming links an article has, the more likely it is that outside editors will find and maintain it. Orphaned articles probably aren't being watched by anybody except their creators, who are more likely to be biased.

• Page views or Google/Alexa rank: Measure how many eyes are watching the page in the most direct way possible.

Note that if any of these approaches proved feasible in BLP articles, they could be applied to unwatched articles in general; Bio articles are a particularly serious issue, but BLP problem material and other bad edits exist outside of biographies. Baileypalblue (talk) 23:21, 19 February 2009 (UTC)

Minor point (not a big deal if newbies can't edit high profile articles)

High profile articles should have strong content which requires strong writers. Allowing newbie editing of high profile topics (like the President) can even be counterproductive as they get reverted so fast (even for innocent mistakes like duplicating info that is in a different section). Note also that allowing screwing up important topics, even important topics that are FAs, dissuades writers from writing good articles on those topics. I think better to ditch the mystique of open-ness and have more of a heirarchy based on capability. For similar reasons, newbies should not make new pages (we already disallow it for IPs). TCO (talk) 17:43, 28 December 2011 (UTC)