Eric Corbett (talk | contribs) →Appreciate and support statement: and neither should you believe it |
|||
(One intermediate revision by the same user not shown) | |||
Line 441: | Line 441: | ||
:Peace is the last step, not the first. Let's not forget that there is no singular community, but there is a singular WMF. [[User:Eric Corbett| <span style="font-variant:small-caps;font-weight:900; color:green;">Eric</span>]] [[User talk:Eric Corbett|<span style="font-variant:small-caps;font-weight:500;color: green;">Corbett</span>]] 20:42, 28 August 2014 (UTC) |
:Peace is the last step, not the first. Let's not forget that there is no singular community, but there is a singular WMF. [[User:Eric Corbett| <span style="font-variant:small-caps;font-weight:900; color:green;">Eric</span>]] [[User talk:Eric Corbett|<span style="font-variant:small-caps;font-weight:500;color: green;">Corbett</span>]] 20:42, 28 August 2014 (UTC) |
||
::I hope that this is not the WMF view because, having seemingly got their "preferred version" and removing the superprotect at de-WP pretty much with the proviso that no-one upsets it, it is easy to imagine the organisation being keen to promote platitudes about peace and chilling. If it ''is'' the WMF view then we're in a bigger hole than I thought. And, no, I don't think MV can be separated from the prior "botched" rollouts because it, too, was botched. - [[User:Sitush|Sitush]] ([[User talk:Sitush|talk]]) 20:51, 28 August 2014 (UTC) |
|||
===Appreciate and support statement=== |
===Appreciate and support statement=== |
Revision as of 20:52, 28 August 2014
(Manual archive list) |
WMF plans for mathematics III
I wanted to thank you for your interest in and support for the proposals that some mathematics editors put to you as a result of your challenge to us. Indeed, you were kind enough to "personally recommend that we allocate resources to this". Unfortunately, Lila tells me that, to her regret, she is unable to do that [1], and indeed it seems that mathematics is not "on the roadmap for the foreseeable future" [2]. Of course I find that hugely disappointing, not to say incomprehensible, and indeed personally I think it's quite offensive, but that's no longer the point. The decision is made, and we must work through the consequences.
Given that mathematics is not on the roadmap, there must be a significant risk that mathematics rendering and editing will not be sustained: that changes made elsewhere in the software to support the changes towards the more modern user experience desired by the Board will degrade or just break mathematics editing or rendering unexpectedly and that resources will not be available to fix it. Of course, I don;t know that's going to happen, I just see no reason for it not to, so there's no way you or I can affect that at present.
Here is one question which you might be able to help with. An academic body is currently in preliminary discussions over whether to institute a Wikimedian-in-Residence programme, which would involve mathematics content. Before committing resources, a question is bound to be asked, and quite properly so, by a sceptical member of their board about the sustainability of Wikipedia in general and mathematics on Wikipedia in particular: in essence, will Wikipedia still "doing maths" in five years time? I don't see how I could honestly encourage them to allocate resources to a project requiring mathematics editing and rendering when there is no commitment to support for those on the part of WMF. What answer would you give to that board member if you were asked?
I'm trying to find the way forwards here, but allow me to express my personal view to you for the moment. Mathematics is one of the oldest continuing human intellectual traditions, with a connected history of over two thousand years. It is fundamental to the sciences in general: no technological endeavour could succeed without it. It is an indispensible part of the sum of human knowledge. It also happens to be the subject I have spent almost all of my career so far advancing in one way or another, including writing dozens of Wikipedia articles. To suggest that it isn't important enough to allocate resources to, when WMF has over 200 staff and tens of millions of dollars at its disposal, is a decision I find both incomprehensible and, personally, indigestible. It seems that WMF would rather allocate its resources to make it easier for readers to view the pornography on Commons than the formulae on Wikipedia. It certainly tells me that the views and values of WMF and myself are now radically divergent. I thought it was about the sum of human knowledge. Apparently that sum does not involve mathematics. Deltahedron (talk) 19:29, 24 August 2014 (UTC)
- I think the WMF take on this is basically that Math doesn't generate enough page views (and thus potential donations) to be worthwhile investing in more than they are already doing. I'm not sure if WMF-paid devs or other contributors are behind mw:Extension:Math. Suffice to say that most of the heavy-duty infrastructure under that like MathJax, dvipng, or Mathoid does not appear to be WMF-sponsored at all. The only thing that was Mediawiki-specific was texvc, a sanitization/adapter program basically; and I think even that was written by someone not paid by the WMF. Someone from the WMF will surely correct me if I missed giving them some due credit on some Math-related infrastructure. JMP EAX (talk) 21:22, 24 August 2014 (UTC)
- Ok, Mathoid was forked/written by a WMF employee [3]. JMP EAX (talk) 21:30, 24 August 2014 (UTC)
- Also this is of some relevance: Math typesetting over the web is still sucky in general. According to [4] none of the browsers makers actually paid their devs for any MathML-related work. So it's not only the WMF which doesn't find it worthwhile much... JMP EAX (talk) 21:50, 24 August 2014 (UTC)
- Maybe people would find this more worth their while if any of our mathematics articles were even remotely comprehensible, but I don't see that happening anytime soon either. The Blade of the Northern Lights (話して下さい) 02:19, 25 August 2014 (UTC)
- That would be an old chestnut even if it were true. Some of the articles are specialised and some are not. Like all topics of any intellectual depth, some of the articles are hard for non-specialists to read, and that is the way it should be. See Wikipedia talk:WikiProject Mathematics/FAQ for the standard answer. Deltahedron (talk) 08:39, 25 August 2014 (UTC)
- Take our article on Graham's number. There are explanations for its use that are much more approachable, and are quite readily available (check YouTube for a video explaining the concept). Why it needs to be written in impenetrable language exclusively using esoteric examples eludes me. I'm currently working on an article which contains some fairly dense linguistics, and it's certainly not the easiest thing to read, but it doesn't take an M.A. in linguistics to at least get a basic understanding of things. The Blade of the Northern Lights (話して下さい) 15:22, 25 August 2014 (UTC)
- The current article seems to me to be well within the range of an intelligent 18 year old mathematics A-level student. Deltahedron (talk) 15:29, 25 August 2014 (UTC)
- Umm. Thirty-odd years ago, I would have been the equivalent of "an intelligent 18 year old mathematics A-level student", and I can't understand much of this article. Its most pressing weakness is an explanation that is understandable by a more general readership; I don't have much of a problem with the more sophisticated information being included, but there's nothing there that helps someone with limited mathematical skilo understand the concept. Risker (talk) 15:40, 25 August 2014 (UTC)
- Then we must agree to disagree as far as that goes. Neither of us seems to have a view of this article consistent with the suggestion that none of our mathematics articles is even remotely comprehensible, or that the language of the article is impenetrable exclusively using esoteric examples. Indeed the first sentence begins "Graham's number, named after Ronald Graham, is a large number" and I think that's fairly clear. If you or Blade wishes to improve this article, then this is the encyclopaedia that anyone can edit. Deltahedron (talk) 15:50, 25 August 2014 (UTC)
- If linguistics is under discussion, let me take a look at Antisymmetry which I came across by chance today in the course of a discussion at WT:WPM. It begins "In linguistics, antisymmetry is a theory of syntactic linearization". How many non-experts have the faintest idea what syntactic linearisation is? Curiously, syntactic linearization is a red link, and linearization is a mathematics article. The only two occurences of the phase "syntactic linearization" are in this article, and in Dynamic antisymmetry which also uses it without further explanation. It appears to be connected with "surface linearization", also not explained. So which of these is "impenetrable language exclusively using esoteric examples", "syntactic linearization" or "a large number"? But I don't follow people writing articles on linguistics around complaing about how hard it is. Deltahedron (talk) 16:00, 25 August 2014 (UTC)
- Yes, our linguistics articles are far from ideal as well; I'm not trying to dispute that. But I hardly go around complaining about how hard math is; search my contributions for the last year and a half, it should be obvious where I spend my time. The Blade of the Northern Lights (話して下さい) 20:35, 25 August 2014 (UTC)
- Two points: I myself find linguistics articles to be worse written (more impenetrable) than math ones too (how much do you understand from dependency grammar for instance?), but it may be a case of expertise as well. As for the Graham number example: c'mon a lot of graduate students in computer science (a sufficiently connected field) have trouble grokking Ramsey theory (just ask anyone teaching TCS classes.) So I don't see why you expect that article to be easily digestible. I'm more curious how far into composition of functions the average reader can get. (That's a "level-4 vital article", for what that's worth.) I'm curious if you'd rather read abstract machine (also a "level-4 vital article"), which is a very bad article from a specialist perspective actually, but it may be easily understandable as is it's mostly a copy of the FOLDOC material. JMP EAX (talk) 22:44, 25 August 2014 (UTC)
- The article on Graham's number is badly written, but for a different reason: it waffles too much in the lead (which almost entirely a history section) and fails to state the problem there, which is amusingly relegated to a sections called "context" that actually doesn't provide anymore context but just blurts out the problem statement. JMP EAX (talk) 23:06, 25 August 2014 (UTC)
- By the way, this has everything to do with who watches/maintains such articles, a major issue that's being discussed elsewhere eon this page. See Talk:Graham's_number#.22all_sorts_of_poorly_chosen_words.22 for the concrete disputes. JMP EAX (talk) 23:40, 25 August 2014 (UTC)
- Also, it's not practical in an Wikipedia article to repeat the definitions from Knuth's up-arrow notation everywhere they are used, but that's clearly a prerequisite to read/grok Graham's number, even if you don't care about the Ramsey theory problem that led to it. And most people probably aren't familiar with Knuth's notation either. I should note that in this respect the Graham number article is compliant with Wikipedia:Make_technical_articles_understandable#Write_one_level_down. JMP EAX (talk) 23:30, 25 August 2014 (UTC)
- The article on Graham's number is badly written, but for a different reason: it waffles too much in the lead (which almost entirely a history section) and fails to state the problem there, which is amusingly relegated to a sections called "context" that actually doesn't provide anymore context but just blurts out the problem statement. JMP EAX (talk) 23:06, 25 August 2014 (UTC)
- Two points: I myself find linguistics articles to be worse written (more impenetrable) than math ones too (how much do you understand from dependency grammar for instance?), but it may be a case of expertise as well. As for the Graham number example: c'mon a lot of graduate students in computer science (a sufficiently connected field) have trouble grokking Ramsey theory (just ask anyone teaching TCS classes.) So I don't see why you expect that article to be easily digestible. I'm more curious how far into composition of functions the average reader can get. (That's a "level-4 vital article", for what that's worth.) I'm curious if you'd rather read abstract machine (also a "level-4 vital article"), which is a very bad article from a specialist perspective actually, but it may be easily understandable as is it's mostly a copy of the FOLDOC material. JMP EAX (talk) 22:44, 25 August 2014 (UTC)
- Yes, our linguistics articles are far from ideal as well; I'm not trying to dispute that. But I hardly go around complaining about how hard math is; search my contributions for the last year and a half, it should be obvious where I spend my time. The Blade of the Northern Lights (話して下さい) 20:35, 25 August 2014 (UTC)
- The current article seems to me to be well within the range of an intelligent 18 year old mathematics A-level student. Deltahedron (talk) 15:29, 25 August 2014 (UTC)
- Take our article on Graham's number. There are explanations for its use that are much more approachable, and are quite readily available (check YouTube for a video explaining the concept). Why it needs to be written in impenetrable language exclusively using esoteric examples eludes me. I'm currently working on an article which contains some fairly dense linguistics, and it's certainly not the easiest thing to read, but it doesn't take an M.A. in linguistics to at least get a basic understanding of things. The Blade of the Northern Lights (話して下さい) 15:22, 25 August 2014 (UTC)
- That would be an old chestnut even if it were true. Some of the articles are specialised and some are not. Like all topics of any intellectual depth, some of the articles are hard for non-specialists to read, and that is the way it should be. See Wikipedia talk:WikiProject Mathematics/FAQ for the standard answer. Deltahedron (talk) 08:39, 25 August 2014 (UTC)
- Maybe people would find this more worth their while if any of our mathematics articles were even remotely comprehensible, but I don't see that happening anytime soon either. The Blade of the Northern Lights (話して下さい) 02:19, 25 August 2014 (UTC)
What the WMF needs (for this to happen) is a Murray Sargent [5] or an Eliyezer Kohen [6]. JMP EAX (talk) 04:26, 25 August 2014 (UTC)
A lot of our math pages are actually very useful. One should not compare page views of math pages to page views of other pages. Rather one should look at links to our math pages from websites known for high quality academic content. Our math and physics pages are very often used as a de-facto reliable reference by stackexchange. Articles like Gaussian quadrature may look like one giant mess to people who are not into math, but they are very useful to people who need to look up something about the topic. Count Iblis (talk) 16:15, 25 August 2014 (UTC)
- I would not take the use on stackexchange as much of an endorsement. The unwritten but well known rule there is that whoever posts something first wins the most points. As such pasting stuff directly from Wikipedia is often practiced, even when it's not really (the most) appropriate answer to a particular question. JMP EAX (talk) 01:16, 26 August 2014 (UTC)
The trustworthiness of Wikipedia
Hi Jmmy.
The Wikimedia Foundation envisions a world in which everyone can freely share in the sum of all knowledge. [7] But "knowledge" of something implies well-founded confidence in its accuracy. While Wikipedia is untrustworthy, it is sharing something other than knowledge. This is a problem for the foundation, since it is failing to realise its vision, but also for humankind, who deserves an encyclopaedia it can trust.
At Wikimania 2014 you said, "We're trusted slightly more than the BBC. Now, that's a little scary, and probably inappropriate. ... We all know it's flawed. We all know we don't do as good a job as we wish we could do ... People trusted Encyclopedia Britannica - I think it was, like - 20 points ahead of us. ... I'm not going to rest until people trust us more than they ever trusted Encyclopedia Britannica in the past." [8]
Are you doing anything at the moment aimed at either improving the public's understanding of Wikipedia's reliability or improving its actual reliability? If so, would you like to share those initiatives here? --Anthonyhcole (talk · contribs · email) 11:14, 21 August 2014 (UTC)
- Everything I do related to Wikipedia is aimed at those things, so I'm afraid I don't quite understand the question. I think one of the biggest changes is the hiring of Lila as CEO coupled with her intention to radically increase investment in software development to help make it easier for us to get our jobs done. There are many ideas that have been floating around for years but we haven't had sufficient developer resources to do them. (It is my view that in the past 5 years we have significantly underinvested in engineering.)--Jimbo Wales (talk) 12:56, 21 August 2014 (UTC)
- Thanks. Could you please elaborate on how technological fixes will solve the unreliability problem? --Anthonyhcole (talk · contribs · email) 13:03, 21 August 2014 (UTC)
- I think there are many ways but let's just talk about one example. Watchlists are a very primitive way to surface new edits to experienced users. A more sophisticated "news feed" style watchlist could take into account a variety of factors to do a better job of showing us edits that need to be scrutinized and as well as new users who are writing in areas that we care about so that we can evaluate them, greet them, welcome them, coach them. Imagine if edits to articles that you personally care a lot about (medical articles say) were scanned and highlighted to people who are part of the relevant Wikiproject if they contain certain "likely problems" (for example, a link to a tabloid newspaper is probably not the best link in a serious medical article, but that's an editorial judgment that could, in some cases, make sense).
- That's just one example and a random brainstorm by just one person (me) on some ideas that people have proposed over the years. But I hope it is indicative of the kinds of things that I have in mind.
- Here's another one: I set up a link in my browser to go to a random female British author whenever I click it. I click it and see if there is some small thing I can improve. But this is very very primitive. What if, instead, the system could take me to an article which based on several factors is likely to need attention. (For example, if readers have expressed displeasure, or if someone recently posted on the talk page, or if an ip address recently edited it, or...)--Jimbo Wales (talk) 13:27, 21 August 2014 (UTC)
- It is worth noting that Wikipedia:Verifiability policy rates as "questionable" any sources that "lack meaningful editorial oversight", and that this class includes Wikipedia itself. Deltahedron (talk) 13:39, 21 August 2014 (UTC)
- Thanks. Could you please elaborate on how technological fixes will solve the unreliability problem? --Anthonyhcole (talk · contribs · email) 13:03, 21 August 2014 (UTC)
- Thanks. How do you feel about scholars reviewing our medical content for accuracy, and us putting a prominent badge at the top of those articles, linking the reader to the fact-checked version? --Anthonyhcole (talk · contribs · email) 13:41, 21 August 2014 (UTC)
- Reliability of the information found here has to hinge on the verifiability of that information. Reliable sources...references...all of that tedious, unexciting stuff. It's quite frustrating when you know that something is true - to be unable to say it here because you need to reference reliable sources. But that's the only way we've ever found to increase both our reliability - and the perception of our reliability. When people tell me that Wikipedia can't possibly be reliable because any idiot can come by and put falsehoods here - I always tell them that if some fact really matters to you, then don't believe what Wikipedia says, click on the little blue number and read the original source of where we got that fact from. If there isn't a blue number - then forget that we ever said that. HOWEVER, if you just need to know something trivial, for idle curiosity - then Wikipedia is pretty darned reliable (and we have studies that show that we're at least as reliable as Encyclopedia Brittannica.
- The problem is not our actual reliability (which seems to be pretty amazingly good) - but the perception of our reliability (which is kinda terrible). The bizarre part about that is that while our editor community is shrinking - our readership is climbing. That's odd...you'd think that if this perception of unreliability was prevalent, that fewer people would be reading Wikipedia. I suspect (without evidence) that convenience trumps perceived reliability for most people. If I want to know "Will there be another series of Crossbones (TV series)?" (darn...no!) then being able to tap the "W" icon on my phone and type one word into the interface to get the answer is ASTOUNDING. The fact that the answer might be wrong...0.1% of the time...is actually less critical. If I were instead researching how people used dockyard cranes in Medieval Europe (which I was actually doing last week) - then I still use Wikipedia - but not for the answers it contains, rather for the curated links it gives me to the source documents. In that regard, Wikipedia is more like a highly effective version of Google-search than it is an encyclopedia. It gives me the links to the source material, and functions kinda like the Brittannica "propedia" that summarizes and organizes the knowledge that can be found elsewhere.
- The general public are also unaware of the fact that for any common question you're likely to be searching for the answer to, there are likely to be dozens of reliable editors watching articles and deleting incorrect information added by random idiots within a very short span of time...and those same random idiots don't get much fun from putting garbage into very obscure subject matter where it might linger for a while before being removed. Even when you point out this undoubted truth to people, we're faced with the problem that people are not good at estimating probabilities and risk. If a junk edit happens in (say) Theory of relativity - then it's going to be fixed in a matter of minutes. Since those changes happen maybe weekly - the probability of you landing on that page while the information is incorrect - multiplied by the probability that the change actually affects you - is a very, very small number. But people are bad at estimating risk...so that's a hard line to sell to people.
- I'm not sure we need technological changes, or even changes to editing rules or habits to make us more reliable (although greater reliability is obviously desirable). Mostly we need public awareness of how reliable we already are - combined with education in how to use Wikipedia when the answer really matters...and when it doesn't.
- SteveBaker (talk) 13:56, 21 August 2014 (UTC)
- So your argument would be that we are right and everyone else, including all the people who use the encyclopaedia, are wrong? Deltahedron (talk) 14:05, 21 August 2014 (UTC)
- More seriously, any useful risk assessment should include an estimate of the impact as well as the probability. While the probability of being misinformed about medical information may be low (I couldn't say), the impact of that misinformation may be very high. We regard that as an acceptable risk simply because we don't have to take it. Deltahedron (talk) 14:08, 21 August 2014 (UTC)
- Sadly, Deltahedron, no one knows the probability of being misinformed about medical information. No one has done a rigorous systematic review of the various small studies into the reliability of our medical offering. I've looked at all of those studies, I think, and in my opinion, a rigorous systematic review is likely to conclude nothing can be inferred from them about the reliability of our medical content. Most have fatal design flaws including tiny sample size, dubious measure of reliability and opaque selection criteria. What's really needed is a large enough, well-enough designed study. --Anthonyhcole (talk · contribs · email) 14:21, 21 August 2014 (UTC)
- But that's my point. For something as important as medical advice, you really shouldn't trust Wikipedia - you should use it as a curated set of links to documents that are presumed trustworthy (or at least more trustworthy than Wikipedia). It's very easy for some random idiot to change the name of a drug as treatment for some condition - but much MUCH harder for them to point the references that back this up. We need to educate people that they really shouldn't take our information as "The Truth" in any situation where it deeply matters. So I'm quite prepared to take the risk that there really is a second season of Crossbones (TV show) in the pipeline (despite Wikipedia saying that there isn't) - because it's just not that important to me. In that situation, convenience trumps absolute reliability. But in deciding whether the drug my doctor just prescribed my kid has side-effects that might concern me, then I'm still going to go to Wikipedia - but I'll pretty much ignore what it says and follow the little blue numbers to the actual medical journals that report the studies done on the drug. SteveBaker (talk) 23:35, 21 August 2014 (UTC)
- Thanks. I'm asking you to imagine a Wikipedia you can trust. Actually, I'm asking Jimmy to imagine a Wikipedia we can trust. I'd really like to know where he stands on the question of a prominent link on our medical articles to versions of those articles that meet WP:MEDRS. --Anthonyhcole (talk · contribs · email) 02:02, 22 August 2014 (UTC)
- While I want Wikipedia to be as accurate as we can manage, I hope we never ask people to trust us. People shouldn't ever trust Wikipedia. They shouldn't ever trust Britannica, or the New York Times either; hopefully we can push that into the public consciousness as we try to make a point about how trustworthy (or not) we are. A stable version, which represents our best work? Perhaps a good idea. A version of Wikipedia we tell people they can trust? A truly awful idea. WilyD 13:19, 22 August 2014 (UTC)
- Are you against the concept of people trusting all possible sources of information? Or are there some you think people should trust? Deltahedron (talk) 14:16, 22 August 2014 (UTC)
- Well, okay we can have a long discussion of how we're talking about trust like it's a boolean quantity when it ain't. There are no sources of information people should trust completely, or distrust completely. But "Oh, you can trust this" reeks of "Oh, you can trust this completely", which is awful, yes. Of course, one could be using "trust/not trust" differently. WilyD 15:30, 22 August 2014 (UTC)
- The foundation is the one talking about sharing knowledge. And, yes, I agree, trust should not be blind. And it is a thing of degrees. But there are some sources about which it is possible to say "I trust this". We can and should be one of those. --Anthonyhcole (talk · contribs · email) 12:49, 23 August 2014 (UTC)
- Well, okay we can have a long discussion of how we're talking about trust like it's a boolean quantity when it ain't. There are no sources of information people should trust completely, or distrust completely. But "Oh, you can trust this" reeks of "Oh, you can trust this completely", which is awful, yes. Of course, one could be using "trust/not trust" differently. WilyD 15:30, 22 August 2014 (UTC)
- Are you against the concept of people trusting all possible sources of information? Or are there some you think people should trust? Deltahedron (talk) 14:16, 22 August 2014 (UTC)
- While I want Wikipedia to be as accurate as we can manage, I hope we never ask people to trust us. People shouldn't ever trust Wikipedia. They shouldn't ever trust Britannica, or the New York Times either; hopefully we can push that into the public consciousness as we try to make a point about how trustworthy (or not) we are. A stable version, which represents our best work? Perhaps a good idea. A version of Wikipedia we tell people they can trust? A truly awful idea. WilyD 13:19, 22 August 2014 (UTC)
- Thanks. I'm asking you to imagine a Wikipedia you can trust. Actually, I'm asking Jimmy to imagine a Wikipedia we can trust. I'd really like to know where he stands on the question of a prominent link on our medical articles to versions of those articles that meet WP:MEDRS. --Anthonyhcole (talk · contribs · email) 02:02, 22 August 2014 (UTC)
- But that's my point. For something as important as medical advice, you really shouldn't trust Wikipedia - you should use it as a curated set of links to documents that are presumed trustworthy (or at least more trustworthy than Wikipedia). It's very easy for some random idiot to change the name of a drug as treatment for some condition - but much MUCH harder for them to point the references that back this up. We need to educate people that they really shouldn't take our information as "The Truth" in any situation where it deeply matters. So I'm quite prepared to take the risk that there really is a second season of Crossbones (TV show) in the pipeline (despite Wikipedia saying that there isn't) - because it's just not that important to me. In that situation, convenience trumps absolute reliability. But in deciding whether the drug my doctor just prescribed my kid has side-effects that might concern me, then I'm still going to go to Wikipedia - but I'll pretty much ignore what it says and follow the little blue numbers to the actual medical journals that report the studies done on the drug. SteveBaker (talk) 23:35, 21 August 2014 (UTC)
- Sadly, Deltahedron, no one knows the probability of being misinformed about medical information. No one has done a rigorous systematic review of the various small studies into the reliability of our medical offering. I've looked at all of those studies, I think, and in my opinion, a rigorous systematic review is likely to conclude nothing can be inferred from them about the reliability of our medical content. Most have fatal design flaws including tiny sample size, dubious measure of reliability and opaque selection criteria. What's really needed is a large enough, well-enough designed study. --Anthonyhcole (talk · contribs · email) 14:21, 21 August 2014 (UTC)
- I'm sure you're right. However, even with the most perfect information about the risk, it would remain true that while we lay that risk off onto other people, it will give us little incentive to get things right. Deltahedron (talk) 14:25, 21 August 2014 (UTC)
- I think we agree, we have a moral responsibility to minimise that risk as much and as quickly as possible. --Anthonyhcole (talk · contribs · email) 14:36, 21 August 2014 (UTC)
- Yes, I think I was too compressed about risk. I'm not suggesting that WP or WMF accept legal responsibility for medical or other information in the same way as a doctor or othe rprofessional. What I am suggesting that that WP/WMF aspire to get things right and be reliable, say so in public and accept the reputational damage if we are not. If I publish an academic paper and it proves to be wrong, my reputation suffers, and I may fail to get a job or promotion. Fortunately in my area of mathematics, people will not be killed in the ruins of a collapsing theorem, but it could happen to others. If we absolve ourselves from all blame in advance by saying "you should have known how to use our encyclopaedia", then we insulate ourselves from all those real-world consequences. I would argue that WP's position should be: yes, we are an encyclopaedia, we aim and claim to be the best, most accurate and most reliable there is, and if we screw up then tell us so and shame on us. That way we take a risk and have an incentive to Get It Right. Technological tools are part of that; so are processes and culture. Have we got any of those right at the moment? Deltahedron (talk) 14:39, 21 August 2014 (UTC)
- In his state of the wiki address this year, Jimmy said with regard to our biographical assertions, we need to do more than the minimum required by law in portraying our subjects. I think that applies to the reliability question, too. The trustworthiness of Wikipedia is a moral vision thing, not a legalistic ass-covering maneuver.
- I'm working on a strategy for this, and I'm in email discussion with another user devising another strategy. Both of our strategies involve the "current" or "dynamic" Wikipedia article sporting a prominent badge, linking the reader to the reliable version. So, I'd like to know where Jimmy stands on that. --Anthonyhcole (talk · contribs · email) 15:00, 21 August 2014 (UTC)
- I agree on the moral point as well: we have an obligation to get things right. Deltahedron (talk) 15:16, 21 August 2014 (UTC)
- Yes, I think I was too compressed about risk. I'm not suggesting that WP or WMF accept legal responsibility for medical or other information in the same way as a doctor or othe rprofessional. What I am suggesting that that WP/WMF aspire to get things right and be reliable, say so in public and accept the reputational damage if we are not. If I publish an academic paper and it proves to be wrong, my reputation suffers, and I may fail to get a job or promotion. Fortunately in my area of mathematics, people will not be killed in the ruins of a collapsing theorem, but it could happen to others. If we absolve ourselves from all blame in advance by saying "you should have known how to use our encyclopaedia", then we insulate ourselves from all those real-world consequences. I would argue that WP's position should be: yes, we are an encyclopaedia, we aim and claim to be the best, most accurate and most reliable there is, and if we screw up then tell us so and shame on us. That way we take a risk and have an incentive to Get It Right. Technological tools are part of that; so are processes and culture. Have we got any of those right at the moment? Deltahedron (talk) 14:39, 21 August 2014 (UTC)
- I think we agree, we have a moral responsibility to minimise that risk as much and as quickly as possible. --Anthonyhcole (talk · contribs · email) 14:36, 21 August 2014 (UTC)
- I'm sure you're right. However, even with the most perfect information about the risk, it would remain true that while we lay that risk off onto other people, it will give us little incentive to get things right. Deltahedron (talk) 14:25, 21 August 2014 (UTC)
- A system like Watson (computer) could perhaps be trusted to not just highlight problems but also correct them. We would then only need to check if the correction made are appropriate. Count Iblis (talk) 14:56, 21 August 2014 (UTC)
- Think software is the least important factor in improving reliability. I would venture that reliability is a function of the degree of expertise of editors in the topic they are editing. There is no bot that can recognize an article that makes wrong or otherwise problematic statements, or uses bad sources. There is also no bot that point out missing content and add it. That makes the key problem finding out how to attract and retain expert editors in different topic areas. I personally think that the biggest obstacle to retaining expert editors is that they are not generally willing to monitor what they write indefinitely nor to defend it in endlessly recurring discussions with laypeople with all kinds of different personal or partial perspectives. The idea of identifying certain revisions as reliable would certainly be one way of making it easier for expert editors to justify spending their time on writing for wikipedia. User:Maunus ·ʍaunus·snunɐw· 20:12, 22 August 2014 (UTC)
- I endorse this view. Partially-informed enthusiasts will generally outlast more broadly informed laymen or experts in most topics, with the result that there's widespread content which sounds plausible and may appear to be properly-sourced but in fact reflects oversimplified, biased, or misinformed views of highly motivated but unqualified editors. This appears to occur irrespective of the underlying topics or categories. SPECIFICO talk 20:43, 22 August 2014 (UTC)
- A good example would be the the article Mass–energy equivalence. Which tells us in no uncertain terms that "mass cannot be converted to energy,[citation needed]". (I added the tag.) Of course the matter is actually of some fairly deep physical and philosophical debate as the SEP entry on the topic will show. JMP EAX (talk) 22:17, 22 August 2014 (UTC)
- Another example is Quantum teleportation which assures us that quantum computers are useless basically as " For example, a qubit cannot be used to encode a classical bit (this is the content of the no-communication theorem).[citation needed]" (I have added the tag.) As to the persistence of some: that text was added by an editor who is blocked indefinitely, after the block... JMP EAX (talk) 07:18, 23 August 2014 (UTC)
- Another example is just about any humanities article.User:Maunus ·ʍaunus·snunɐw· 20:21, 23 August 2014 (UTC)
- I endorse this view. Partially-informed enthusiasts will generally outlast more broadly informed laymen or experts in most topics, with the result that there's widespread content which sounds plausible and may appear to be properly-sourced but in fact reflects oversimplified, biased, or misinformed views of highly motivated but unqualified editors. This appears to occur irrespective of the underlying topics or categories. SPECIFICO talk 20:43, 22 August 2014 (UTC)
Place a prominent link on an article to the version that has passed rigorous independent expert review?
In my opinion, it all hangs on the quality of the reviewers and their rigor. If this goes ahead, we need to have the highest standards; higher than any existing encyclopedia, journal or textbook review process. We should be the benchmark against which those are measured.
I think we should reverse the usual process, and have anonymous writers and named reviewers. We have to have anonymous writers - it's the way of this wiki and that's not going to change. Naming the reviewers would offer the readers transparency, and (if our reviewed versions become the gold standard of reliability) offer the reviewers prominent kudos. --Anthonyhcole (talk · contribs · email) 02:24, 23 August 2014 (UTC)
- I'm curious how many experts you think it'll take to review 4.6 million articles, and how all those experts will be recruited and paid for. How strict would this review be? Would it be like GA? FA? Higher? And how would this system be integrated with the actual wiki? What happens to articles that don't pass this (presumably extremely stringent) review? --Jakob (talk) 13:04, 23 August 2014 (UTC)
- Jakob, I shall ask the relevant scholarly societies, ones with decades (some with centuries) of experience in peer review, to manage this independent external review process. And I'll ask medical charities with public education as part of their mission to fund it. The Wikimedia Foundation should not fund it, to avoid any semblance of undue influence over the process. Critical review of all our current best medical articles, and rigorous fact-checking of all our other medical articles should cost less (probably a lot less) than $10 million.
- I'm personally only interested in making our medical content reliable. The idea is scalable, though, to any field well-covered by serious scholarship.
- This won't affect any of our existing practice, but we will need to agree whose reviewed versions we should link to. That will require discussion (not here please) and a new policy or guideline governing that decision-making process.
- Articles that don't pass fact-check won't get a badge with a link to the fact-checked version, and errors will be listed on their talk pages. Class As and FAs that fail critical scholarly review won't get the badge linking to the critically-reviewed version until they pass a critical scholarly review. The societies managing the review process should have most say in how the review and fact-checking processes work. Wikipedia's role here is to put a prominent badge at the top of articles, linking the readers to the most recent rigorously fact-checked or critically reviewed versions.
- I'm curious to know Jimmy's position on that proposed practice. Obviously, it will affect my confidence going forward with this if Wikipedia's most prominent evangelist is enthusiastically supportive, indifferent or actively opposed to the proposal. --Anthonyhcole (talk · contribs · email) 14:18, 23 August 2014 (UTC)
- I would be interested in helping you discuss proposals with learned bodies in Mathematics. Perhaps you could email me? Deltahedron (talk) 10:30, 25 August 2014 (UTC)
- The "most read" ten thousand articles account for a huge percentage of usage, and we can remove "popular culture" as being intrinsically non-reviewable, alas. The 2010 list of "most read" pages shows the ones at the 10,000 level has only 2% of the readership of the top actual articles (about 90K down to under 2K views per day). I suspect that, barring web-crawlers, the top 10,000 actual articles account for well over 95% (possibly quite a bit more) of total page views. [9] confirms this a bit. Collect (talk) 13:36, 23 August 2014 (UTC)
- I agree, it would be easy to identify specific fields of knowledge that require expert level knowledge of somekind to assure reliability. And starting with the 10,000 most vital would be obvious.User:Maunus ·ʍaunus·snunɐw· 20:21, 23 August 2014 (UTC)
- The "most read" ten thousand articles account for a huge percentage of usage, and we can remove "popular culture" as being intrinsically non-reviewable, alas. The 2010 list of "most read" pages shows the ones at the 10,000 level has only 2% of the readership of the top actual articles (about 90K down to under 2K views per day). I suspect that, barring web-crawlers, the top 10,000 actual articles account for well over 95% (possibly quite a bit more) of total page views. [9] confirms this a bit. Collect (talk) 13:36, 23 August 2014 (UTC)
- The key to recruit expert editors would be to find some way to make it relevant for them as a career move, for example by having their name officially attached to specific articles so that they could put it on their CVs as a public outreach kind of thing.User:Maunus ·ʍaunus·snunɐw· 20:21, 23 August 2014 (UTC)
- I disagree. I think the key to recruiting expert editors is to ensure that they don't spend all of their time locked in circular arguments with ill-informed obsessives. (I know you've had this experience, on the race & intelligence articles). The Randy-from-Boise essay nailed it. Unfortunately, the problem has gotten worse over time, as we've steadily lost non-insane editors and maintained a constant number (and thus an increasing proportion) of obsessives and cranks. It's really dispiriting, and I'm speaking as someone who (let's say, for the sake of argument) has some real-life expertise but no longer writes much, if any, content. I don't know that c.v. credit will help; most of the people who'd be in a position to review my c.v. would laugh at me if I included my Wikipedia work on it. MastCell Talk 04:12, 25 August 2014 (UTC)
- MastCell, I don't see any way for Wikipedia writers to get public credit under the current editing model - and that model is unlikely to change. But if our reviewed versions were the gold standard of trustworthiness, public credit for reviewing a reliable version would be worth something to your academic reputation. If we begin by selecting only the masters/mistresses of their fields, then being chosen to review an article will be seen as a significant accolade. I should repeat: an essential element of this strategy is the cast-iron rigor of its review process. You and I are painfully aware of the curate's egg that passes for peer-review in the journals right now. Wikipedia's reviewed versions can be the standard against which the journals and textbooks are measured, absurd as that sounds in today's context.
- I disagree. I think the key to recruiting expert editors is to ensure that they don't spend all of their time locked in circular arguments with ill-informed obsessives. (I know you've had this experience, on the race & intelligence articles). The Randy-from-Boise essay nailed it. Unfortunately, the problem has gotten worse over time, as we've steadily lost non-insane editors and maintained a constant number (and thus an increasing proportion) of obsessives and cranks. It's really dispiriting, and I'm speaking as someone who (let's say, for the sake of argument) has some real-life expertise but no longer writes much, if any, content. I don't know that c.v. credit will help; most of the people who'd be in a position to review my c.v. would laugh at me if I included my Wikipedia work on it. MastCell Talk 04:12, 25 August 2014 (UTC)
- I'm imagining the next wave of innovation in the free distribution of knowledge, at least as important as the first, with Wikipedia and the academic communities driving it in concert.
- This may not be the final solution to the Randy problem, but knowing their work won't be washed away by some POV-pusher, troll or well-meaning amateur once it's passed expert review may encourage more experts to edit. I'm confident it will, actually, if, once the process is in place, we can get the word out loud and clear to the scholarly community through their journals and the public through public evangelism. --Anthonyhcole (talk · contribs · email) 04:38, 25 August 2014 (UTC)
- With regard to medical content, are the people commenting here aware of meta:Wiki Project Med and its project Wikipedia:WikiProject Medicine/Translation Task Force? But there are also several expertly edited free online sources, including the US National Institutes of Health's MedlinePlus, the Mayo Clinic's site,]the Professional and home editions of Merck Manual of Diagnosis and Therapy, and WebMD, -- all of which usually rank higher in a Google search than Wikipedia. They too have their limitations, but everything is not up to us alone.
- with respect to expert editing in general, in my opinion one of the many reasons for the relative failure of Citizendium was the difficulty in getting changes approved. The "expert" editors there (of which I was one) were much more interested in writing and approving new articles than the routine work of incorporating and improving changes. I put "expert" in quotes to indicate my opinion that another cause of failure was the difficulty (and sometimes errors) in actually selecting the volunteer experts. DGG ( talk ) 15:42, 25 August 2014 (UTC)
- A lot of it is up to us. Many of the obvious, broad topics are covered to some degree by sites such as those you mention, but most topics are not. We have more than 30,000 medical articles so far. I would estimate no more than 5,000 of those topics are covered by Mayo, NIH, etc. --Anthonyhcole (talk · contribs · email) 16:49, 25 August 2014 (UTC)
Although this documentary suggests that false medical information can be useful :) . Count Iblis (talk) 18:56, 25 August 2014 (UTC)
WP:MEDICAL. Now, please don't perform 5th grade surgery with an iPad next to you showing a Wikipedia article about how to cut your victim patient open and find the bullet that's somewhere in him. --k6ka (talk | contribs) 21:25, 25 August 2014 (UTC)
- The issue that is that the so-called reviews in Wikipedia which lead to GA/FA status are often conducted by people whose main qualification is that they know a lot of wikirules. The content is seldom actually reviewed. A good example is the extremely sophomoric Commutative property article. The reviwerer was of course extremely excited he could understand all of it: Talk:Commutative_property#GA_Review (unlike other math articles being discussed above...) JMP EAX (talk) 00:38, 26 August 2014 (UTC)
Since someone mentioned stackoverflow above, let me quote from a somewhat well known critique of that [10] a portion that applies to Wikipedia as well, in my opinion:
“ | Here is the recipe that all such "community-driven" approaches almost, but not quite, invariably follow:
This happened at Wikipedia and it's happened at StackOverflow. |
” |
It surely applies to the so-called reviews (GA/FA) in Wikipedia. JMP EAX (talk) 01:25, 26 August 2014 (UTC)
- Reviewing articles and forking Wikipedia are two commonly proposed solutions that are actually the same thing, or should be. As I started to suggest in m:Usenetpedia (I wasn't even the first to think of that, it turns out) if we have freely distributed versions under no central control, from which multiple authorities can choose "the latest version" independently of one another, we could abolish the authority games that go on here. True, that is in favor of authority games at the authorities, but if they have to compete for readers they will be subject to limitations. Wnt (talk) 11:01, 26 August 2014 (UTC)
- One reliable Wikipedia and one unreliable but possibly more up-to-date Wikipedia. I suppose you could call that "forking" but it's not forking. They'll both be Wikipedia. If Wikipedia refused to allow a badge at the top of the dynamic version linking to the reliable version, then we'd need to arrange a fork, but I'm hoping that won't be the case. --Anthonyhcole (talk · contribs · email) 11:51, 26 August 2014 (UTC)
- I fear I won't have time to engage in a very long discussion of this, at least not this week, but I wanted to make two quick points. First, I'm interested and generally positively inclined to agree that if we had a serious peer review process to run medical articles (or others, but medical is a good place to start) through some recognized expert process, then we should link prominently to that reviewed version. But second, and this is a matter for empirical study in the long haul, I think that within a very short period of time (shortness depending on the topic) the peer reviewed article will NOT be the better article because it will be out of date. The truth is, the majority of edits to most articles (particularly on serious topics) are improvements over what was there before, particularly if the edit isn't vandalism. What I'm saying is "the most recent version edited by Anthonyhcole (or some other trustworthy editor)" is likely to be better than a 3 month old peer reviewed version. That doesn't make peer review a useless exercise, though, not by a long shot! The point of the peer review is to bring in someone with specific expertise in some specific sub-topic to look for errors or subtle problems that even a highly experienced Wikipedia editor in the general field of medicine might miss.--Jimbo Wales (talk) 12:11, 26 August 2014 (UTC)
- One reliable Wikipedia and one unreliable but possibly more up-to-date Wikipedia. I suppose you could call that "forking" but it's not forking. They'll both be Wikipedia. If Wikipedia refused to allow a badge at the top of the dynamic version linking to the reliable version, then we'd need to arrange a fork, but I'm hoping that won't be the case. --Anthonyhcole (talk · contribs · email) 11:51, 26 August 2014 (UTC)
- Thanks. Yes, I think both versions - current and last reviewed - will be valuable to the reader. I hope readers will get into the habit of reading the reviewed version and viewing the diff to see what's new. Just for the record, good Wikipedia medical articles are very stable, and one review a year would be more than adequate for most. But frequency is probably best determined by the societies running the review process, and will vary from topic to topic. --Anthonyhcole (talk · contribs · email) 13:49, 26 August 2014 (UTC)
- Here, by "good" I mean WP:GA and WP:FA. --Anthonyhcole (talk · contribs · email) 23:47, 26 August 2014 (UTC)
- I think that's a great idea, Anthony. Too often I find myself having to go to the talk page, pulling down a menu or two, looking for the "milestones", before I can get the "official" FA version. Drmies (talk) 04:41, 27 August 2014 (UTC)
- I've long thought there should be a prominent link on our GA/FAs to the version that passed GA/FA. --Anthonyhcole (talk · contribs · email) 10:20, 27 August 2014 (UTC)
- I think that's a great idea, Anthony. Too often I find myself having to go to the talk page, pulling down a menu or two, looking for the "milestones", before I can get the "official" FA version. Drmies (talk) 04:41, 27 August 2014 (UTC)
- Here, by "good" I mean WP:GA and WP:FA. --Anthonyhcole (talk · contribs · email) 23:47, 26 August 2014 (UTC)
Medical warning
Hi Jimmy. At WikiProject Medicine we're trying to find a form of words for a prominent warning on our medical articles ... something that conveys that anyone can edit but that also conveys the inherent effect on trustworthiness, without scaring the readers unduly. The discussion has just started - but any suggestions you may have for concise, elegant language to that effect would be very welcome. This pertains to your comment quoted in the opening post, above, where you acknowledge our readership may have a poor understanding of our reliability. It is an attempt to plainly, straightforwardly appraise the readers of the inherent risk in the anyone-can-edit-live model. I think, if we acknowledge our UK readers have an inflated perception of our reliability, we have a duty to clearly, plainly inform them. I think you've spoken in favor of a prominent disclaimer on our medical content. Possibly on this page - I can't remember. --Anthonyhcole (talk · contribs · email) 09:55, 27 August 2014 (UTC)
The founder's opinions on Superprotect and MediaViewer
Dear Jimmy, since days now there are big discussions and huge irritations following the rollout of Superprotect and Media Viewer against the communities' will. I am interested in what your personal views of these happenings are. In my eyes they are clearly against your own original project principles (esp. #4). Did you change them, without publishing it? I didn't find any statemets by you on these issues yet. --Trofobi (talk) 07:02, 25 August 2014 (UTC)
- I share the concern about MV and SP, and I think this is a S
NAFU where input from Jimbo might be one of the few things that can have a positive influence on a very divisive issue. A significant part of this happens on the German Wikipedia, where there is serious talk about strikes, passive resistance, and even forks. I think the WMF has made a big mistake here, in trying to force issues via decrees and technical means that should have been addressed via communication and dialogue. --Stephan Schulz (talk) 07:26, 25 August 2014 (UTC)- Oh, it's here on en:wp for long, too:
- Wikipedia:Media Viewer/June 2014 RfC
- Special:Diff/616531136#Media viewer
- Wikipedia:Administrators' noticeboard/Archive263#Authoritative basis of threats to temporarily desysop as a "WMF action"
- Wikipedia:Arbitration/Requests/Case/Media Viewer RfC
- Wikipedia:Village pump (technical)#New superprotect protection level, coming to your wiki soon
- Wikipedia:Bureaucrats' noticeboard/Archive 30#Request for removal of 'expired' temporary administrative rights.
- User talk:Jimbo Wales/Archive 169#Super protect please help
- User talk:Jimbo Wales/Archive 170#WMF superblocks its community
- User talk:Phoebe#WMF superblocks it's community
- User talk:John Vandenberg#About the Superprotects rights
- Wikipedia:Wikipedia Signpost/2014-08-13/News and notes
- ...
- and on meta: m:Letter to Wikimedia Foundation: Superprotect and Media Viewer, and many more
- ... --Trofobi (talk) 07:59, 25 August 2014 (UTC)
- I never knew the MV caused such serious conflict between the community and the WMF until I read those archived discussions. I don't have any statistics to prove my point, but as a long term Wikimedian, I can sincerely say that the MV is just a waste of time, money and user's computer power. It is not useful to either the editor or reader. It's so silly (yet funny) to see how much WMF staffers try to sugarcoat this gadget). IMHO WMF/MediaWiki should instead improve the media description page to satisfy both the reader and editor. -- Sameboat - 同舟 (talk) 23:36, 25 August 2014 (UTC)
- Oh, it's here on en:wp for long, too:
I think Lila Tretikov has posted some conciliatory steps/moves on her meta-wmf talk page: [11] (although some have described them as "California-speak" IIRC.) JMP EAX (talk) 04:04, 26 August 2014 (UTC)
- Honestly I'm still a bit sore about the patronizing non-action on the solution we fucked-up en-wiki crackheads (don't you just love that?) came up with, and since then I've seen absolutely no change in attitude. It's somewhat tempting to spaz out (in the American English sense, I'm aware it carries a very different meaning in the UK) and unleash a torrent of profanity here, but instead I'll put it this way. Since the rollout of these two changes, I've noticed essentially no difference in my editing experience. If these were changes that were essential to keeping all WMF projects online, or something of equivalent importance, I could see why the WMF would push them this hard. Since they're not, though I'm failing to see how these changes were worth the amount of rancor they've generated. The Blade of the Northern Lights (話して下さい) 04:51, 26 August 2014 (UTC)
- As a small side note, it is unclear to me why a comment from someone who doesn't work at the WMF is something you are taking as a matter of attitude from the WMF. In terms of thinking about this issue and trying to understand and feel everyone's perspective clearly, understand that there are those who are just as bewildered that such a minor software change ("essentially no difference in my editing experience") is causing so many people to man their battle stations. It is my considered opinion that this fight (in the German Wikipedia and elsewhere) isn't really about the MediaViewer at all, except in the most superficial ways.--Jimbo Wales (talk) 12:00, 26 August 2014 (UTC)
- Partially because I'm a bit bitter about it, but mostly because their comments were, while not as explicit, along similar lines (I don't have it in me to track all the discussions down blow-by-blow, but they're there). I also see where you're coming from, in my view the reaction does seem to me to be at least a bit excessive. The Blade of the Northern Lights (話して下さい) 02:31, 27 August 2014 (UTC)
- The "excessive reaction" has been accumulated since the implementation of VisualEditor and Typography Refresh. MV itself isn't a deal breaker (if it came before VE), but Superprotect only worsens the mistrust between WMF and experienced Wikimedians. All in all, WMF should instantly retract the implementation of SP and politely dissuade German admins from reverting any change imposed by WMF. -- Sameboat - 同舟 (talk) 03:11, 27 August 2014 (UTC)
- Partially because I'm a bit bitter about it, but mostly because their comments were, while not as explicit, along similar lines (I don't have it in me to track all the discussions down blow-by-blow, but they're there). I also see where you're coming from, in my view the reaction does seem to me to be at least a bit excessive. The Blade of the Northern Lights (話して下さい) 02:31, 27 August 2014 (UTC)
- As a small side note, it is unclear to me why a comment from someone who doesn't work at the WMF is something you are taking as a matter of attitude from the WMF. In terms of thinking about this issue and trying to understand and feel everyone's perspective clearly, understand that there are those who are just as bewildered that such a minor software change ("essentially no difference in my editing experience") is causing so many people to man their battle stations. It is my considered opinion that this fight (in the German Wikipedia and elsewhere) isn't really about the MediaViewer at all, except in the most superficial ways.--Jimbo Wales (talk) 12:00, 26 August 2014 (UTC)
- I'm preparing a comment on this to be posted tomorrow at the earliest. I'm reading and studying all that I can in preparation for that, and I have a call scheduled with Lila this afternoon.--Jimbo Wales (talk) 11:56, 26 August 2014 (UTC)
REVIEW BY EXPERT MEMBERS OF THE PRE-RAPHAELITE BROTHERHOOD SOCIETY
This article and those of all the personalities mentioned are going to be checked for errors and omissions by the expert members of the Pre-Raphaelites Society, hopefully without editor interference. In the unlikely event of 'editorial interference', Jimmy Wales will be advised. This Article and several others look as though they need to be comprehensively checked.
2.30.208.46 (talk) 21:06, 26 August 2014 (UTC)
- Goodness me, Jimbo. What a dark horse! Who knew?? Martinevans123 (talk) 21:09, 26 August 2014 (UTC)
- Before they start complaining of "editorial interference", the members of the Pre-Raphaelites Society are respectfully requested to read Wikipedia:Ownership of articles. JohnCD (talk) 21:52, 27 August 2014 (UTC)
- According to one Martin Packer here, "Wikipedia has just been advised that all 66 articles are to be reviewed; I suggest that you post your additions/deletions/corrections here too? Wikipedia is easy to edit, you will soon get the hang of it, or post to me!". He seems to be quite a quirky character who enjoys writing in CAPITALS TO GET HIS MESSAGE ACROSS. I'm sure he'll soon learn our little ways. Ghmyrtle (talk) 22:07, 27 August 2014 (UTC)
- ... some more Pre-Raphaelite CAPITALS for you to enjoy ... Martinevans123 (talk) 22:21, 27 August 2014 (UTC)
- According to one Martin Packer here, "Wikipedia has just been advised that all 66 articles are to be reviewed; I suggest that you post your additions/deletions/corrections here too? Wikipedia is easy to edit, you will soon get the hang of it, or post to me!". He seems to be quite a quirky character who enjoys writing in CAPITALS TO GET HIS MESSAGE ACROSS. I'm sure he'll soon learn our little ways. Ghmyrtle (talk) 22:07, 27 August 2014 (UTC)
- Before they start complaining of "editorial interference", the members of the Pre-Raphaelites Society are respectfully requested to read Wikipedia:Ownership of articles. JohnCD (talk) 21:52, 27 August 2014 (UTC)
Status of Flagged Revisions in German Wikipedia?
Can someone tell me how flagged revisions is used in German Wikipedia?--Jimbo Wales (talk) 11:01, 27 August 2014 (UTC)
- According to de:Hilfe:Gesichtete und geprüfte Versionen and other relevant guidelines:
- General features
- is used to indicate that a revision is neither a sighted nor quality version.
- Trusted editors are grouped into two types of reviewers: Passive and Active sighters. The latter are allowed to review the edits of other users, while the former have their edits automatically marked as sighted but they do not have the right to review the work of others.
- Requirements
- To automatically attain the status of a passive sighter, a user account must be at least 30 days old and contain at least 150 unsighted edits or 50 sighted edits. Deleted edits and all edits made in the last two days are not taken into account. In addition, the user must not be blocked for any offense, he must make at least 7 edits over a period of at least three days, and he must edit at least 8 different articles. The edit summary must be used for at least 20 edits.
- To automatically attain the status of an active sighter, a user account must be at least 60 days old and contain at least 300 unsighted edits or 200 sighted edits. Deleted edits and all edits made in the last two days are not taken into account. In addition, the user must not be blocked for any offense, he must make at least 15 edits over a period of at least three days, and he must edit at least 14 different articles. The edit summary must be used for at least 30 edits. The proportion of reverted edits must not exceed 3% of the user's total edits.
- Alternatively, there is a noticeboard for users to manually apply for the right to sight articles. Passive sighting rights are granted to users who can demonstrate that they are willing to work constructively, while active sighting rights are generally, but not always, granted to users with Stimmberechtigung - which could be loosely translated into English as the right to participate in voting processes, such as WP:RFA and consensus discussions.
- All admins have the right to remove sighting rights, especially to prevent abuses.
- General sighting of articles
- The manual sighting of articles is similar to the reviewing of pending changes on semi-protected pages of the English Wikipedia. This can only be done by active sighters.
- The automatic sighting of articles takes place when an already sighted article is edited by a passive or active sighter. If a passive sighter edits an article with unsighted edits (e.g. from an IP address), the article will be marked with unsighted revisions until it is manually reviewed by an active sighter.
- The sighted revisions of an article can only be reverted by active sighters.
- Current discussion
- There is no consensus yet on how to implement the quality reviewing of articles by technically competent editors.
- It seems that they are still discussing how to certify a user as technically competent.
- According to de:Hilfe:Gesichtete_und_geprüfte_Versionen#Pr.C3.BCfen, the technical reviewing of articles by qualified editors may not be implemented after all. The page claims that the Wikimedia Foundation is refusing to allow the implemention of the sighting system for other Wikipedias (???)
- Disclosure: I have active sighting rights on the German Wikipedia, so I may be able to answer related questions from other editors, but I don't work behind the scenes and I can only translate what their guidelines say. I hope this was helpful.
- -A1candidate (talk) 16:07, 27 August 2014 (UTC)
- There are no current discussions on flagged revisions in German wikipedia.
- Sighted versions have been widely accepted and even appreciated for years.
- And for years, no one has been trying to launch Quality versions.
- Any further questions? --Niki.L (talk) 19:50, 27 August 2014 (UTC)
- And maybe it should be added that it is a widely accepted tool for the quality management and that a removal like indicated by the acting against the community consensus pro flagged revisions in the Norwegian Wikipedia would cause the next big disturbance between the WMF and the German Wikipedia. Sorry if it sounds like i assume bad faith, but honestly i have no reason anymore to assume good faith on the foundation's side. --Julius1990 (talk) 20:10, 27 August 2014 (UTC)
- For some discussion about not turning on flagged revisions in Norwegian see bugzilla:64726. (Personally I am opposed to flagged revisions of any kind based on wiki-ideological reasons, but do not have any data that supports either side in this argument). —Kusma (t·c) 07:36, 28 August 2014 (UTC)
- And maybe it should be added that it is a widely accepted tool for the quality management and that a removal like indicated by the acting against the community consensus pro flagged revisions in the Norwegian Wikipedia would cause the next big disturbance between the WMF and the German Wikipedia. Sorry if it sounds like i assume bad faith, but honestly i have no reason anymore to assume good faith on the foundation's side. --Julius1990 (talk) 20:10, 27 August 2014 (UTC)
- There is some analysis, stats, comparison and recommendations for flagged revisions at meta:Talk:Flagged Revisions#Comparison of some Wikipedias. See also de:user:Atlasowa/gesichtete Versionen. I think it's a successful feature on deWP (and plWP, fiWP), very very effective against vandalism, spam and the degradation of articles over time on a large scale. The careless attitude of WMF [12][13] is making me mad. --Atlasowa (talk) 09:29, 28 August 2014 (UTC)
- Maybe you are more interested in the cultural issues. Edits by anons and new users to deWP don't go live until an experienced editor looked at them and approved them as "not vandalism". Of course often the "sighter" can determine more about the quality of an edit than just "vandalism"/"non vandalism", but officially this is the primary purpose of the feature and anyone can and should "sight" every article, no matter if he or she is an expert on the topic. So we use it as first line of defense in our quality control, the second line would be the watchlist used by editors who see themselves as responsible for a given topic. deWP does not use bots to determine the quality of recent changes - or at least by far not to the degree of enWP. We use this feature a lot in explaining about quality in Wikipedia, and we get universal approval for it in the public. No one I ever talked to complained that he or she feels excluded because his or her edits do not go live immediately. --h-stt !? 12:20, 28 August 2014 (UTC)
Actually the flagged revisions have at least two points of benefits: First: If vandalism happens the read-only user does not see it. Second: Theoretically every edit is seen by at least one "sighter" so it is unlikely that vandalism is more unlikely missed at all. The feature helped to improve and the archieved quality level of our articles. What now was told to NO:WP is just again ridiculous. --Matthiasb (talk) 13:43, 28 August 2014 (UTC)
A few follow-up questions, if I may: (1) what percentage of articles fall into the sighted category? (2) Is it usually the case that sighted/accepted edits are acceptable, or do some unacceptable ones get overlooked? (3) What is the usual backlog? Do different types of articles have different rates of backlog? (4) Are there issues with edit conflicts, where a second editor edits text that has already been edited in a not-yet-accepted change, and how are those handled? Thanks, Newyorkbrad (talk) 14:11, 28 August 2014 (UTC)
- all - Flagged revisions are implemented as mandatory for the article edit>and category/edit space at deWP.
- Plain vandalism is reverted to 100%. Really bad edits that are not vandalism are detected almost all of the time, say 98%
- median or max? median a few hours, max 14 days.
- sure. all of the time. Experienced editors look for newish, not yet "sighted" revisions and flag them before they edit an article themselves. If several new editors edited "on top of" each other, I do a diff over all versions since my last time at the article and evaluate the diff. If there are valid and unwanted edits mixed together, I have to look for the one version that inserted the unwanted information (or deleted the good one) and revert just this edit. Or I edit the page manually neutralizing the offending edit without reverting it with the software feature. hth --h-stt !? 14:48, 28 August 2014 (UTC)
- Additional answer at @Newyorkbrad: (and Jimbo, off course): It is not clear what are the goals the WMF has set. Maybe the hope that Wikipedia draws more editors, which add information like that. However deciding wether this edit is helpful or not is a bit more complicated than recognizing vandalism in which the F-word is added or parts of the text are removed, and obviously the additional workload for regular users to verify such modification seems to be of no concern in the WMF strategy. --Matthiasb (talk) 18:01, 28 August 2014 (UTC)
- @Matthiasb:, you are completely mistaken about the concerns and strategy of the Wikimedia Foundation. The Foundation seeks to make the editing process easier for everyone - this is not trivial to do obviously and it is not helpful when people take a combative attitude when one is not warranted.--Jimbo Wales (talk) 19:43, 28 August 2014 (UTC)
What Wikipedia is in the Eyes of the World
This comment is the epitome of Wikipedia's philosophy:
what Wikipedia has become in the eyes of the world, is the place on the internet where the people who express opinions like the above about the "other guy's" religious beliefs, are encouraged to label all the practitioners whom they are persecuting as "cultic types, not very wikifriendly" etc in his appeal for like-minded souls. The whole world by now knows this is the kind of place wikipedia "Fringe Theory Noticeboard" openly is and that these are the very same editors who typically get backslapped, rewarded and promoted, because the range of permissible world views tolerated among wikipedia editors seems to be getting more and more restricted to only those who see the world your way. 71.246.152.192 (talk) 11:33, 27 August 2014 (UTC)
- I think it unfortunate to refer to other editors in that way, but it sounds like he's legitimately asking for help in improving some decidedly poor articles. If Wikipedia has a reputation for taking a pretty hard nosed approach to cults and fringe theories, then I am proud.--Jimbo Wales (talk) 11:56, 27 August 2014 (UTC)
- Most people see their beliefs as something very personal and private, nobody wants to see their own firm beliefs or religion wind up on the wikipedia official "fringe theory" list, and the origin of the term is from "lunatic fringe", is inherently pejorative, and should never have been embraced by a "neutral" or professional project. If editors want a blog where they can plot and scheme whom they are going to anathematize next, at least it should not be on-wiki with wiki endorsement. 71.246.152.192 (talk) 12:00, 27 August 2014 (UTC)
- You are mistaken. The term 'lunatic fringe' would be pejorative, of course. But 'fringe' is not inherently pejorative and conveys information that the reader needs. Do you have a specific page that you think has been categorized unfairly or incorrectly?--Jimbo Wales (talk) 12:05, 27 August 2014 (UTC)
- In fact the phrase "lunatic fringe" is far earlier, and the use of "fringe" in that sense is derived as an abbreviation of that expression. 71.246.152.192 (talk) 12:10, 27 August 2014 (UTC)
- Oxford Dictionaries: "The outer, marginal, or extreme part of an area, group, or sphere of activity:" [14]. Not necessarily pejorative in any way. Martinevans123 (talk) 12:23, 27 August 2014 (UTC)
- I'm sorry, when I said "in that sense" I meant in the sense of the phrase "fringe theory". The phrase "fringe theory" and "lunatic fringe theory" was unheard of before the 70s at the earliest and was slang avoided by balanced works; the earlier phrase it is derived from, "lunatic fringe", dates to Theodore Roosevelt. 71.246.152.192 (talk) 12:41, 27 August 2014 (UTC)
- Language evolves? We're not all stuck in the 1970s? Personally, I think Roosevelt was perfectly sane. Martinevans123 (talk) 12:44, 27 August 2014 (UTC)
- Pretending that a pejorative or disparaging term is not, isn't the same as language evolving. 71.246.152.192 (talk) 12:50, 27 August 2014 (UTC)
- You're quite right, it's not. But why should I pretend? Is this all full rubbish or lunatics too? Or is it just not on the main stage? Martinevans123 (talk) 13:23, 27 August 2014 (UTC)
- Our real problem here is that there any mild epithet like "fringe" or "lunatic" ought to pale in comparison to our epithet for people who take reliable, relevant sources out of Wikipedia articles. We try calling them deletionists, vandals, POV warriors, censors... it's all somewhat accurate but there's always some defect in the word that they use to say "oh, no, that's not us, and there really isn't such a thing!" We need to craft a decently offensive epithet that can lock onto that particular behavior like a missile. Wnt (talk) 13:50, 27 August 2014 (UTC)
- Luckily I am not Subud myself, or I would be much more incensed to see them brought up as "fringe", and I know little about them other than that the few Subud colleagues and friends I have ever known, never did me any wrong that I can recall. I hardly know what their beliefs are, but I doubt fringe theory is appropriate to characterize many things like this that it gets thrown around on too much. I took a look at the article and talkpage for talk:Subud now and see a low amount of discussion including about some critical source, but I am too uninformed on that one to have an opinion. 71.246.152.192 (talk) 15:15, 27 August 2014 (UTC)
- Edinburgh is definitely full of lunatics and rubbish. -mattbuck (Talk) 14:01, 27 August 2014 (UTC)
- I trust Alistair can count on your vote. Martinevans123 (talk) 14:12, 27 August 2014 (UTC)
- Our real problem here is that there any mild epithet like "fringe" or "lunatic" ought to pale in comparison to our epithet for people who take reliable, relevant sources out of Wikipedia articles. We try calling them deletionists, vandals, POV warriors, censors... it's all somewhat accurate but there's always some defect in the word that they use to say "oh, no, that's not us, and there really isn't such a thing!" We need to craft a decently offensive epithet that can lock onto that particular behavior like a missile. Wnt (talk) 13:50, 27 August 2014 (UTC)
- You're quite right, it's not. But why should I pretend? Is this all full rubbish or lunatics too? Or is it just not on the main stage? Martinevans123 (talk) 13:23, 27 August 2014 (UTC)
- Pretending that a pejorative or disparaging term is not, isn't the same as language evolving. 71.246.152.192 (talk) 12:50, 27 August 2014 (UTC)
- Language evolves? We're not all stuck in the 1970s? Personally, I think Roosevelt was perfectly sane. Martinevans123 (talk) 12:44, 27 August 2014 (UTC)
- I'm sorry, when I said "in that sense" I meant in the sense of the phrase "fringe theory". The phrase "fringe theory" and "lunatic fringe theory" was unheard of before the 70s at the earliest and was slang avoided by balanced works; the earlier phrase it is derived from, "lunatic fringe", dates to Theodore Roosevelt. 71.246.152.192 (talk) 12:41, 27 August 2014 (UTC)
- Oxford Dictionaries: "The outer, marginal, or extreme part of an area, group, or sphere of activity:" [14]. Not necessarily pejorative in any way. Martinevans123 (talk) 12:23, 27 August 2014 (UTC)
- In fact the phrase "lunatic fringe" is far earlier, and the use of "fringe" in that sense is derived as an abbreviation of that expression. 71.246.152.192 (talk) 12:10, 27 August 2014 (UTC)
- You are mistaken. The term 'lunatic fringe' would be pejorative, of course. But 'fringe' is not inherently pejorative and conveys information that the reader needs. Do you have a specific page that you think has been categorized unfairly or incorrectly?--Jimbo Wales (talk) 12:05, 27 August 2014 (UTC)
- Most people see their beliefs as something very personal and private, nobody wants to see their own firm beliefs or religion wind up on the wikipedia official "fringe theory" list, and the origin of the term is from "lunatic fringe", is inherently pejorative, and should never have been embraced by a "neutral" or professional project. If editors want a blog where they can plot and scheme whom they are going to anathematize next, at least it should not be on-wiki with wiki endorsement. 71.246.152.192 (talk) 12:00, 27 August 2014 (UTC)
- IMO WP:FRINGE is rather misused. I think it is used a lot to stifle other opinions and viewpoints by dismissing them as "fringe theories". Remember, there are real people out here who wholly believe in these beliefs. To them, they are not fringe theories. To them, other theories are fringe theories. "Fringe" and "Psuedoscience" are very nice buzzwords, but they are subjective and Wikipedia should not be the final arbiter of what is "normal" and what is not. KonveyorBelt 16:06, 27 August 2014 (UTC)
- Fringe to anything only means on the edge of the mainstream which in itself is not pejorative. As well, today's fringe may be tomorrow's Nobel Prize so we don't need to psychologically turn our noses up at fringe topics. However, fringe should not be used as a sledge hammer to exclude that which is notable enough for Wikipedia or content that is significant per sources for Wikipedia. Unfortunately fringe is used as an excuse to exclude legitimate content, and often hides bias.
- Cult is a word that pigeon holes often based on subjective interpretation. We might do well to follow the academics who are more likely to define aspects which contribute to something being non acceptable to majorities rather than use a word that has become a cliché driven word for orgs we don't understand or perhaps like. In an encyclopedia the academic route might be more dignified and well ... academic.(Littleolive oil (talk) 17:45, 27 August 2014 (UTC))
Biographies of politicians created as part of election campaigns
In the course of working at AfC I quite frequently see attempts to create advertorial biographies of politicians who are busy campaigning for election. Sometimes the article authors are quite blatant about their intentions to use WP as a campaign vehicle while others seem to try to do it "under the radar". Now AfC has the ability to filter out blatantly promotional articles but I presume many such articles may be created directly in mainspace too. So I'm thinking maybe we could institute a prohibition against the creation of new biographies of politicians while they are busy campaigning. Though some of the subjects might already be notable, such as a former large city mayor running for a state or national level office or their notability is for something else prior to their political career, some are first time candidates who would actually only become notable if they were to win the election, except that the media attention generated by the campaign itself sometimes does push them over the GNG threshold. If an prohibition against such article creation for the duration of electioneering is not acceptable to the community, perhaps we could devise a variant of the "COI editor" tag that says something like "This article may have been created in furtherance of the subject's election campaign" which stays on the page until the election is over. Roger (Dodger67) (talk) 14:09, 27 August 2014 (UTC)
- I'd go for the flip-side solution (although both options will require policy change) that a candidate running in an election that is considered notable get a page (at least a stub) and, while still requiring that sources meet RS for that area, and BLP, etc, have a tag placed on them that allows deletion discussion after the declaration of official results. After that 'election period', the 'candidate' notability ends and normal rules apply. Some unsuccessful candidates may have pages avoid deletition because they're engaged in the community, state/province etc, while 'winners' may still be deleted. But, for the election period, they should get a pass. Just my two cents... AnonNep (talk) 15:45, 27 August 2014 (UTC)
- I would vigorously oppose that, as the Wikipedia would get drawn in to debates on just what candidates would be eligible for this. Many U.S. political races have a dozen or more candidates from the legitimate parties all the way down to the...to be frank...kooks, who run on bizarre one-issue platforms. If we're going to say "no kooks", then we get into drawing an arbitrary "X is worthy and Y is worthy but Z isn't", just like the debate committee gets into with its "you must be above 20% (or whatever) in the opinion polls in order to be included" thing. I'd rather just use the existing notability rules on the books rather than create a new system for the campaign season. Tarc (talk) 16:53, 27 August 2014 (UTC)
- Democracy can be scary but the way I see it working, the constituency is Federal, state/province/council and local (for which we usually have pages), in English language countries, as represented by EN:WP (we don't have to host an English version of every possible candidate page for non-eng-wikis, or a contentious Puddley-on-Marshes Women's Institute AGM). All candidate pages (if anyone bothers creating them) get a banner with 'no delete' from start to end of polling period as agreed at a constituency level. They meet a 'local' RS level (shire/county newspapers etc without fear of delete because there's no 'Google Books' ref) but only for the voting period). Then the tag goes and they can be ProD'd like any page. As it is, existing notability policy has bias for any 'new' candidate who has a media record nationally - I suggest, given them all equal footing for the election period (which they've agreed to by becoming candidates). AnonNep (talk) 17:42, 27 August 2014 (UTC)
- If you're wanting an encyclopedia to be an up-to-the-minute source for elections, then you're really in the wrong business here. Steer people interested in this to Ballotpedia, where even the 5th-tier candidate for sewer commissioner in East Toad Strangle, Nevada gets a page. Tarc (talk) 17:53, 27 August 2014 (UTC)
- For US politics, nice aside ... but not for the world of en:WP in many other countries. As I said at the start, I responded to 'stop candidate pages' with an 'allow candidate pages' on en-wiki, by adopting a defined constituency base (which we have, pretty much, page/category wise) with an election period timeframe tag and more localised RS (all of which would need consensus, just like 'No candidate evaaa!!!'). We do have candidate pages for local elections if someone is wealthy enough/media savvy enough, so let's allow at least a time-limited stub for all, with usual rules applying to all (including the winner) at the end. No requirement for 'up-to-the-mintute', it is, what it is. AnonNep (talk) 18:08, 27 August 2014 (UTC)
- If a person cannot satisfy the general notability guideline, then they should not have an article. Period. This suggestion that a stub of an article should be allowed to remain in the project in contravention to existing notability policy, even on a limited-time basis) is not something I would ever be in favor of. Tarc (talk) 18:51, 27 August 2014 (UTC)
- Nice try but no Straw man, dude. I'm not suggesting 'a stub of an article should be allowed to remain in the project in contravention to existing notability policy' but that, given the reported infux of candidate articles during elections they should be allowed, during the election period, suitably templated, with local issues/sources prioritised. At the end of the election period delete as per standard policy - the result makes no difference. AnonNep (talk) 19:09, 27 August 2014 (UTC)
- There's no strawman here, I can't help it if you're tripping over your own back-pedalling words. It's very simple; create an article on a candidate; if it passes WP:GNG, WP:NPOL and the like, then it stays. If it doesn't, then it gets deleted or redirected to the article on the race. No grace period, no waiting til the election is over, no special protection. Clear? Tarc (talk) 19:25, 27 August 2014 (UTC)
- You're repeating the Straw man. All reasonable qualifications given above. AnonNep (talk) 19:37, 27 August 2014 (UTC)
- (ec) I'm pretty sure that's quite incorrect. Any article that passes GNG should be kept, but the bazillion special notability guidelines all set up "alternate criteria" for articles that fail GNG. I assume for example that we would want a candidate to have a page if his opponent has a page, even if only one passes GNG. Personally I think we should trash all the novelty notability guidelines and make a short paragraph in GNG about rounding out a complete set of individuals when most are notable and they are all most notable for the same reason. I realize that the novelty guidelines are frequently abused to claim an article fails even though it passes GNG but every time I've looked the guideline itself hasn't been what's to blame, it's the user misapplying it. But because they're more confusing than useful, we'd be better off without them. Wnt (talk) 19:40, 27 August 2014 (UTC)
- Something of substance to discuss, finally. There has been confusion for years as to exactly what the purpose and function of the "sub notability guides" actually is. At one time I did think that it was like a safety net that could "save" articles that failed the GNG, but as time has gone on I think they serve more as a complement. A subject still has to meet some form of general notability, the subs just provide a fine-tuned, subject-specific way to get there. Like the infamous WP:PORNBIO that provides passage for a person who wins a noted adult film award; that can get them in, but some reliable source somewhere still has to take note of the person for his/her article to bew truly safe. Tarc (talk) 20:15, 27 August 2014 (UTC)
- Funnily enough, a similar discussion arose out of WP:NJournals just now. I would put it this way. We are writing en encyclopaedia, and so to write an article on a subject it is necessary to have independent reliable sources. Verifiability tells us we cannot write an article any other way. However, there are a few things we can say without independent sources, although without those sources the article would be little more than an entry in some kind of directory. The supplementary guidelines are there to record decisions about what kind of directories we are willing to tolerate even when the entries are not truly be encyclopaedic. Some of those supplementary guidelines are, in my personal view, far too liberal/lax, and almost all of them seem subject to intense wikilayering even when they are self-proclaimed as supplementary guidelines, as essays or whatever. Perhaps the time has come to review and harmonise some of those supplements? Deltahedron (talk) 20:31, 27 August 2014 (UTC)
- Something of substance to discuss, finally. There has been confusion for years as to exactly what the purpose and function of the "sub notability guides" actually is. At one time I did think that it was like a safety net that could "save" articles that failed the GNG, but as time has gone on I think they serve more as a complement. A subject still has to meet some form of general notability, the subs just provide a fine-tuned, subject-specific way to get there. Like the infamous WP:PORNBIO that provides passage for a person who wins a noted adult film award; that can get them in, but some reliable source somewhere still has to take note of the person for his/her article to bew truly safe. Tarc (talk) 20:15, 27 August 2014 (UTC)
- There's no strawman here, I can't help it if you're tripping over your own back-pedalling words. It's very simple; create an article on a candidate; if it passes WP:GNG, WP:NPOL and the like, then it stays. If it doesn't, then it gets deleted or redirected to the article on the race. No grace period, no waiting til the election is over, no special protection. Clear? Tarc (talk) 19:25, 27 August 2014 (UTC)
- Nice try but no Straw man, dude. I'm not suggesting 'a stub of an article should be allowed to remain in the project in contravention to existing notability policy' but that, given the reported infux of candidate articles during elections they should be allowed, during the election period, suitably templated, with local issues/sources prioritised. At the end of the election period delete as per standard policy - the result makes no difference. AnonNep (talk) 19:09, 27 August 2014 (UTC)
- If a person cannot satisfy the general notability guideline, then they should not have an article. Period. This suggestion that a stub of an article should be allowed to remain in the project in contravention to existing notability policy, even on a limited-time basis) is not something I would ever be in favor of. Tarc (talk) 18:51, 27 August 2014 (UTC)
- For US politics, nice aside ... but not for the world of en:WP in many other countries. As I said at the start, I responded to 'stop candidate pages' with an 'allow candidate pages' on en-wiki, by adopting a defined constituency base (which we have, pretty much, page/category wise) with an election period timeframe tag and more localised RS (all of which would need consensus, just like 'No candidate evaaa!!!'). We do have candidate pages for local elections if someone is wealthy enough/media savvy enough, so let's allow at least a time-limited stub for all, with usual rules applying to all (including the winner) at the end. No requirement for 'up-to-the-mintute', it is, what it is. AnonNep (talk) 18:08, 27 August 2014 (UTC)
- If you're wanting an encyclopedia to be an up-to-the-minute source for elections, then you're really in the wrong business here. Steer people interested in this to Ballotpedia, where even the 5th-tier candidate for sewer commissioner in East Toad Strangle, Nevada gets a page. Tarc (talk) 17:53, 27 August 2014 (UTC)
- Democracy can be scary but the way I see it working, the constituency is Federal, state/province/council and local (for which we usually have pages), in English language countries, as represented by EN:WP (we don't have to host an English version of every possible candidate page for non-eng-wikis, or a contentious Puddley-on-Marshes Women's Institute AGM). All candidate pages (if anyone bothers creating them) get a banner with 'no delete' from start to end of polling period as agreed at a constituency level. They meet a 'local' RS level (shire/county newspapers etc without fear of delete because there's no 'Google Books' ref) but only for the voting period). Then the tag goes and they can be ProD'd like any page. As it is, existing notability policy has bias for any 'new' candidate who has a media record nationally - I suggest, given them all equal footing for the election period (which they've agreed to by becoming candidates). AnonNep (talk) 17:42, 27 August 2014 (UTC)
- I would vigorously oppose that, as the Wikipedia would get drawn in to debates on just what candidates would be eligible for this. Many U.S. political races have a dozen or more candidates from the legitimate parties all the way down to the...to be frank...kooks, who run on bizarre one-issue platforms. If we're going to say "no kooks", then we get into drawing an arbitrary "X is worthy and Y is worthy but Z isn't", just like the debate committee gets into with its "you must be above 20% (or whatever) in the opinion polls in order to be included" thing. I'd rather just use the existing notability rules on the books rather than create a new system for the campaign season. Tarc (talk) 16:53, 27 August 2014 (UTC)
- Campaign bios are an absolutely inevitable part of WP's future. Articles For Deletion actually treats politician biographies more harshly than any other category, I think; simple GNG sometimes doesn't cut it there for low level elected politicians covered in the media for their campaigns and ordinary job activities. That stuff just doesn't "count" towards GNG. High level elected politicians are auto-Kept, as they should be; unelected politicians are either deleted if not part of a major ongoing campaign, or redirected/merged to coverage of that campaign. In short, there are already tools to use to weed out much of the politicized POV campaign dreck... The idea of a new flag for campaign bios isn't a bad one. Maybe something more neutral, like "The subject of this article is currently participating in a political campaign" would be sufficient warning of possible (probable?!?!) content shenanigans. Carrite (talk) 16:29, 27 August 2014 (UTC)
- @Carrite I like the idea of the "currently campaigning" tag. It would be relevant to new as well as even long established articles of undoubtedly notable politicians which may be prone to shenanigans by supporters or opposers. Roger (Dodger67) (talk) 20:30, 27 August 2014 (UTC)
anent this topic: Wikipedia_talk:Biographies_of_living_persons#Categories.2C_lists_and_navigation_templates and Wikipedia:An#RfC for discussions concerning political issues, claims, BLPs and the like during "silly season". Collect (talk) 17:01, 27 August 2014 (UTC)
- If we can get a small percentage of the $6 billion, then we'll do all the necessary work to make room for such wiki pages, making sure these pages are consistent with all of our policies. Count Iblis (talk) 17:09, 27 August 2014 (UTC)
- I agree with Carrite that we handle politician biographies quite well at AfD, and have a pretty well established consensus of which biographies we keep, delete or merge. Bios of high ranking elected officials are kept, bios of unelected candidates for high office are redirected to articles about the campaign, which are easier to keep neutral, and bios of small town mayors and council members are usually deleted. By the way, many political biographies get very little disruptive editing. Cullen328 Let's discuss it 20:44, 27 August 2014 (UTC)
- At AfC we sometimes get campaign staff moaning at us for declining their advertorials because "it's really important that this article is accepted because my boss is standing for election / because his opponent has an article so he must have one too / because the voters need to know about her...blah blah blah. The standard reply is of course "try again after he's won". My concern is about articles that are not actually deletable - but are abused for campaign purposes by supporters as well as opposers. Roger (Dodger67) (talk) 21:44, 27 August 2014 (UTC)
- I think that creating a new warning flag as you suggest could be done BOLDly, could it not? Carrite (talk) 22:16, 27 August 2014 (UTC)
- I'm toying with a RfC regarding what we do with bio articles that fail both GNG and the narrower NPOL. I was astonished recently to see that one AfD for a general election candidate had resulted in a redirect when, at least in my limited experience of the things at AfD, they are usually deleted. I got nowhere at WP:RFD because of issues of what might be called policy compartmentalism - people dealing with narrow scopes instead of big pictures. To give an extreme example of the problem, in 1996 one constituency in India had over 1000 candidates. That's not a typo. - Sitush (talk) 13:36, 28 August 2014 (UTC)
- Actually, it looks like WP:GNG now makes really low-key reference to the subject-specific guidelines; it seems like maybe they are becoming deprecated. I think what we need to phase them out entirely is that in the GNG, we should say that if an article topic is
- part of a well-defined, objective set of topics (all Indian regional candidates, all British hereditary peers, all asteroids, all Irish soccer teams)
- in which most of the members pass GNG requirements (the Indian candidates might fail this from what you say)
- in which there is merit to complete and equitable coverage of the category (to have a complete catalogue of extrasolar planets, to ensure all candidates in a general election are treated equally)
- Then we should have an article on it. After that, we should treat the special notability guidelines like the archives of a noticeboard, as past decisions about how to implement specific distinctions. Maybe a new noticeboard should take over to replace their continued development and refinement. Wnt (talk) 15:38, 28 August 2014 (UTC)
- I think we're just leaving ourselves wide open to some serious BLP problems: politics is a rough-house environment at the best of times but the idea of monitoring thousands of additional, often transient, BLP subjects during an election campaign is pretty scary. There is a conflict between being neutral and being notable, sure, but Wikipedia is not a democracy and we do not have to follow democratic principles. In fact, I'm not convinced that a lot of so-called democracies follow democratic principles: for example, broadcast air-time given to politicians on public service radio and television varies widely. - Sitush (talk) 17:54, 28 August 2014 (UTC)
- Actually, it looks like WP:GNG now makes really low-key reference to the subject-specific guidelines; it seems like maybe they are becoming deprecated. I think what we need to phase them out entirely is that in the GNG, we should say that if an article topic is
- I'm toying with a RfC regarding what we do with bio articles that fail both GNG and the narrower NPOL. I was astonished recently to see that one AfD for a general election candidate had resulted in a redirect when, at least in my limited experience of the things at AfD, they are usually deleted. I got nowhere at WP:RFD because of issues of what might be called policy compartmentalism - people dealing with narrow scopes instead of big pictures. To give an extreme example of the problem, in 1996 one constituency in India had over 1000 candidates. That's not a typo. - Sitush (talk) 13:36, 28 August 2014 (UTC)
- I think that creating a new warning flag as you suggest could be done BOLDly, could it not? Carrite (talk) 22:16, 27 August 2014 (UTC)
Have in mind that an article of a certain politician, written during the elections, may not always be the result of someone trying to promote the candidate. It can also be written the usual way: someone noticed a topic (that man) with coverage in the press, and which is not included yet in Wikipedia, and so starts an article about him, reporting what do the newspapers say about him. It can also be a page written from the other side of things, a page written by the opposing party that tries to highlight the negative info about the candidate. Cambalachero (talk) 19:26, 28 August 2014 (UTC)
Sexism in drug testing and begging for your support dealing with Arbcom in another 6 months.
To the best of my knowledge and evidently Wikipedia's as well, sexism in drug testing isn't noted by any secondary medical studies, just the head of the NIH repeatedly and in recently passed legislation. However Binksternet is just deleting this every time I add it to any relevant article, even non medical ones. Correct me if I'm wrong but isn't Wikipedia trying to get more women to edit, talking about this might help a bit. Rationalwiki let me stay a systems operator, and several people there would vouch I'm more help than harm on their wiki, please let me return to editing the CensoredScribe accoun, I believe I've proven I can avoid mass categorizations and make rather unique edits to topics like Buddha and Plastic that may be of use to society. My problem area is fiction not non fiction. Exiled Encyclopedist (talk) 06:15, 28 August 2014 (UTC)
- FWIW, there is no law and it doesn't effect pharma companies. The editorial you keep citing is by [[Francis Collings, the Director of the National Institutes of Health (NIH), which funds basic research in universities. The NIH is saying it is going to require scientists that get funding from the NIH (which is basically every biomedical researcher that is not working in pharma in the US) to do experiments on male and female cells, animals, and to the extent they are working with human subjects (which is more rare for NIH-funded research), humans. Pharma doesn't use NIH funding - they use their own money. The NIH is entirely separate from the FDA, which is the agency that approves drugs and has leverage over pharma - in other words, could (within the law) require pharma to submit data from testing drug candidates on male and female cells, animals, and humans. There is no mention of new law in the editorial - the NIH will implement these changes in the contracts through which it funds research. The FDA is not affected by this, nor is pharma. Jytdog (talk) 10:28, 28 August 2014 (UTC)
- A Google Scholar search will find lots of secondary medical literature relating to the issue of sexism in drug testing, going back decades -- for example http://www.nejm.org/doi/full/10.1056/NEJM199307223290429 in the New England Journal of Medicine (open access). The basic policy point is that you shouldn't make universal generalizations unless you have sources that directly support those universal generalizations. You can't assert that something doesn't exist simply because you don't know of any examples of it. Looie496 (talk) 14:26, 28 August 2014 (UTC)
Levitin's statements about Wikipedia and expertise
I happened to catch the tail end of this WBEZ radio interview with neuroscientist Dan Levitin in a discussion related to overcoming information overload that I wanted folks to give a listen to. Here's an excerpt, but please check out the last 5-6 minutes of the interview for a little context:
...the problem is that the Wikipedia model, as it was stated by its founders, is that they don't consider that expertise exists, effectively. Anybody can edit. And an expert has no greater standing than a non-expert. So, the problem is that for any given entry, an 11-year-old can go in and change it. And it could be about something really important...
- ...now, Wikipedia says, "sooner or later, someone will set it right," but that's not always true. The idea is that the 11-year old [who has an imperfect understanding of the topic] may change the article to reflect their non-expert understanding. Eventually, an expert might come along and fix it, but if the 11-year-old is relentless enough with the edit key, the 11-year-old is going to win the battle of attrition because the expert will just give up...so, the problem is that information can be quite unreliable, and there's no way for the average user to know.
(I'm sure a lot of editors have heard concerns like this from somebody at one point or another. I don't know that it's a very common story here, but there it is nonetheless.)
Levitin goes on to explain that because unreliable information is so easy to access, it's important that as a society, we should reinforce ideas that "expertise exists," and that "we don't all have an equal voice in things like the truth and facts," because some folks have better access to information (like journalists), and that such expertise should be "spoken about" and "promoted."
There's a lot to talk about here, so I'll just say my piece. I don't actually believe that every expert feels this way about Wikipedia in terms of reliability, but I think experts do not contribute to Wikipedia for reasons not unlike what Levitin spoke about. I do wonder what we can do to make sure people with domain-specific expertise can be made more easily aware of guidelines like writing one level down and systems we have to tackle things like disruptive editing, so they can start editing with the right expectations about the project. I, JethroBT drop me a line 07:19, 28 August 2014 (UTC)
- I will try to contact Prof. Levitin to correct his misunderstanding. It is simply not true that I believe that expertise does not exist. And it is obvious that in Wikipedia that the model he describes of the editing process is wrong.--Jimbo Wales (talk) 13:12, 28 August 2014 (UTC)
- He's wrong to some degree, but not entirely. When we have a group of competent editors maintaining an article and watching it closely, then an "11 year old" won't be able to corrupt it. The problem comes in for articles that draw only limited interest from maintainers. In neuroscience (Levitin's area and also mine), we have articles such as brain fitness, brainwave entrainment, and binaural beats that contain content I don't think is justifiable, but when I have tried to clean them up, I ran into resistance from, um, enthusiasts, and was unable to recruit enough support to overcome it. (I'm not willing to take on 1-vs-1 contests, because in that situation a person who is willing to revert-war will always overcome a person who refuses to.) Bottom line: this problem is not as pervasive as the passage above makes it seem, but it does exist. Looie496 (talk) 13:39, 28 August 2014 (UTC)
- The problem raised by JethroBT should concern us all and the somehow caricatural editing model described by Prof. Levitin has a lot of truth in it. I have a couple of expertise fields myself. However very seldom I take the initiative of editing the articles on those fileds. The reason is I don't want to subject myself to any kind of edit war or negotiation with some very active editors who have distorted or popular ideas about the subjects. Yes, it is possible to circumvent these difficulties if you are patient enough, persistent enough and can get the support of other editors. However many of us (including me) can't spare the time and the effort to fight that kind of war. Just my 2 cents. Alvesgaspar (talk) 13:41, 28 August 2014 (UTC)
- In the more political areas persistent editors putting in garbage and removing RS -- including in BLPs -- is a constant struggle. It would be so nice to edit in areas 15 year olds don't care about and 35 year olds were editing collaboratively in. Carolmooredc (Talkie-Talkie) 13:44, 28 August 2014 (UTC)
- He is largely correct, but he does miss that experts have considerable leeway to publish. For example, the experts can create an accessible, approachable web site for the public that clearly states the principles and bears the imprimatur of their department or scientific organization, then cite it in the article. So while I sympathize with his perspective very much, I think the best place for him to fight for his POV is by creating an off-wiki resource. In general, where you don't want crowdsourcing, where you want one or a few named authors to have the final say, that's an off-wiki project. Wnt (talk) 14:16, 28 August 2014 (UTC)
- Indeed if the expert really is one, he/she will, almost unavoidably, have access to entire libraries of the highest quality independent reliable sources, thus "squashing" the hypothetical 11-year-old edit warrior's nonsense should be trivially easy. The problem with some experts that I've come across while working at AfC is that they simply don't grok that "because I say so" is not a valid reference. If experts would simply learn to cite instead of assert they'd have a far easier time here. "Citation, not assertion" could be a variant of VNT for experts. Roger (Dodger67) (talk) 16:36, 28 August 2014 (UTC)
- The counter-argument to this is laid out with terrifying plausibility at User:Jnc/Astronomer vs Amateur. Deltahedron (talk) 18:20, 28 August 2014 (UTC)
- Speaking to that point about access to sources, at the end of that interview, Prof. Levitin mentions the need to educate people on the difference between "valid and invalid sources," which is definitely important whether you are an editor or a reader. I'd also argue that the average Wikipedian who has made some nontrivial amount of contributions probably knows more about how to distinguish sources that than the average adult. I, JethroBT drop me a line 18:13, 28 August 2014 (UTC)
- Indeed if the expert really is one, he/she will, almost unavoidably, have access to entire libraries of the highest quality independent reliable sources, thus "squashing" the hypothetical 11-year-old edit warrior's nonsense should be trivially easy. The problem with some experts that I've come across while working at AfC is that they simply don't grok that "because I say so" is not a valid reference. If experts would simply learn to cite instead of assert they'd have a far easier time here. "Citation, not assertion" could be a variant of VNT for experts. Roger (Dodger67) (talk) 16:36, 28 August 2014 (UTC)
@Jimbo Wales:: I am not sure if Mr. Levitin discriminates between the WMF board and you, and recent developments support very indeed that the Board of Trustees really believes in what Mr. Levitin is critizising. It seems that the whole strategy of the WMF is based on quantity instead of quality. It might be necessary to forward the statement It is simply not true that I believe that expertise does not exist. And it is obvious that in Wikipedia that the model he describes of the editing process is wrong. above to the board members. Maybe they should write it down 100 times – by hand of course, not by copy and paste ;-) --Matthiasb (talk) 18:11, 28 August 2014 (UTC)
- Indeed, I too am looking forward to hearing more from the WMF Board chair on his opinion that Our entire approach on knowledge dissemination is based on the western idea of an encylopedia and referencing other written sources in order to back up articles. Yet a lot of cultures around the world have a different way of disseminating (and consuming) knowledge. We should be able to get adapt our model to these cultures as well (celebrating the diversity and seeing new opportunities) [15]. Deltahedron (talk) 18:44, 28 August 2014 (UTC)
A barnstar for you!
The Real Life Barnstar | |
Good Job for making wikipedia. Bobherry talk 13:41, 28 August 2014 (UTC) |
Some thoughts about MediaViewer, my Statement of Principles, and the community's relationship with the foundation
One of the key statements that has been made is that the Wikimedia Foundation is in violation of #4 of my well-known Statement of Principles (User:Jimbo Wales/Statement of principles) so I want to spend a little bit of time specifically discussing that issue.
Here is my original: "4. Any changes to the software must be gradual and reversible. We need to make sure that any changes contribute positively to the community, as ultimately determined by me, in full consultation with the community consensus."
And here is how it is stated today: "4. Any changes to the software must be gradual and reversible. We need to make sure that any changes contribute positively to the community, as ultimately determined by the Wikimedia Foundation, in full consultation with the community consensus."
I must say that I am surprised and pleased to see how well the principle has held up over time and how clearly it still expresses some important ideas.
It is worth noting first and foremost what it does not say. It does not say that software changes must be approved by community vote. It does not say that community consensus is the primary principle for deciding whether some feature should be implemented or not, but rather that changes contribute positively to the community, as decided by the Wikimedia Foundation. And finally it says that the Wikimedia Foundation should make that decision in full consultation with the community consensus.
What should that consultation look like? It should look a fair amount like what we have seen in the past few weeks but without the wheel-warring and drama. Remember that this feature, which can be changed easily, has already been improved to overcome sensible objections and resolve the sorts of issues that are normally caught by live deployments. I look forward to the Foundation's plan to have an incremental rollout process to reduce drama around this sort of improvement.
Here's a view of the future that I think is a disaster for the community: suppose we adopt as a new policy, which has never existed in any formal way, that every community votes (looking for majority levels of support) on every new feature and whether they want to turn it on or off by default. The result is that the software development gets even slower and we fall further behind than where we are today because it becomes impossible for the developers to have a clear view of how it works in all the different environments. The amount of effort that would need to go into addressing every feature variation on each of hundreds of wikis would be exponentially higher as each of them will needed to be identified, monitored, tested and coded.
And here is a view of the future that I think is fantastic for the community: the WMF invests a lot more resources in engineering and product including building a proper consultative process with the community, and introducing incremental roll-outs (to 1% of the editors, then 2%, then 5%, or similar) so that problems can be identified and fixed before we have a huge drama. In this vision we don't have a set of features that are voted on to be turned on and off, we have a dynamic and ongoing healthy conversation about how to improve things.
I have personally been frustrated in the past many times with the disastrous product roll-outs that we've seen (I am not talking about MV, but I'm sure we all remember Flagged Revisions and the Visual Editor). And I want that to change. By hiring Lila, we have committed to making that change and she's investing in building up capacity to get things done in a better way. And we in the community need to support that and call people out on some particularly unhelpful and false attitudes (boiled down to the essence: the WMF is against editors - there are many variants of this claim, all false).
Has the Foundation screwed up? Yeah, sure, lots of times. Has the community screwed up? Yeah, sure, lots of times. Is there a better way? Yes.
What I'm asking people to do is, as I used to say, "relax a notch or two." Let's calm things down for a couple of months. It seems that the Foundation is about to remove (or has just removed) the superprotection of a javascript file in German Wikipedia, and I beg the German Wikipedians to work to reduce tension by not implementing the controversial javascript hack again. And then let's have a real conversation about what improvements need to be made to the MediaViewer and expect that the Foundation will indeed make those improvements.
And then the more important task - let's talk about and help the Foundation design a sane process for community consultation on developing software. And let's not do this in the sense of a political battle or power struggle but rather "Assume Good Faith," and understand that software decisions made by committee or community vote is not a functional process (and indeed, has gotten us to the sorry state we are in today) but that equally, software created by developers who have a poor understanding of our real needs doesn't work either. Let's work on a better way that is both efficient in terms of getting software that works produced and effective in terms of meeting our real needs as editors.
Peace is the first step, so let's chill.--Jimbo Wales (talk) 19:32, 28 August 2014 (UTC)
- Peace is the last step, not the first. Let's not forget that there is no singular community, but there is a singular WMF. Eric Corbett 20:42, 28 August 2014 (UTC)
- I hope that this is not the WMF view because, having seemingly got their "preferred version" and removing the superprotect at de-WP pretty much with the proviso that no-one upsets it, it is easy to imagine the organisation being keen to promote platitudes about peace and chilling. If it is the WMF view then we're in a bigger hole than I thought. And, no, I don't think MV can be separated from the prior "botched" rollouts because it, too, was botched. - Sitush (talk) 20:51, 28 August 2014 (UTC)
Appreciate and support statement
- I want to thank you Mr. Wales for giving us the above statement, and say that I fully support all of the points it clarified.—John Cline (talk) 20:05, 28 August 2014 (UTC)
- no more power struggle? to be clear: no new superpowers for the staff? really? I can't believe it! thanks =) --Sargoth (talk) 20:39, 28 August 2014 (UTC)
A candid question
- I have a single question, the same I've been asking from the very beggining without getting an answer. Knowing that giving up (while the product is being developped) implementing MV as the default viewer for everyone would considerably deflate the present tension between WMF and the community, why that step was not taken already? Considering the present situation of conflict it would certainly be more important to deflate the tension than to stick to some obscure operational agenda. If you want to chill, that is certainly a good step to begin with. Alvesgaspar (talk) 20:36, 28 August 2014 (UTC)