Maybe not so funny. The main engine for "truth" at Wikipedia is citation of a verifiable reference. Allowing for deductive logic would at least complicate the job for many editors, who would call it "Original Research" (which is disallowed). In the cast of the Totient Function proofs example, those proofs are all in books, some can be cited, so the only issue is notability. But it introduces the possibility of establishing the truth of something by giving an explicit proof, more plausible in math than most disciplines. Many editors wouldn't want the responsibility of validating a proof and would object to it as O.R., but in fact we all make reasonable inferences when we compose prose glosses of things for articles. It's a grey area with lots of room for debate. --yoof
I'd never heard of Durova until I looked into this on account of the Slashdotting. It's a mess, but it's not her doing and while there are in fact conspiracies, this isn't particularly one of them.
1. A user was banned temporarily for edit-warring, which is normal. The target of the war was "Irreducible Complexity" (which sounds scientific but...) which is about "Creation Science" (so consensus is nearly impossible).
2. The edit-warring was pretty well documented. But there was also an issue of "sockpuppetry" (some people who get banned, log in with new accounts to continue fighting for their cause). Since personal info is involved, Durova sent it to a group of permed admins; that's normal too. They had not yet reviewed it when the last ban occurred.
3. The result was the impression that a user had been banned for "secret" reasons, when actually a user had been banned for normal reasons, and the part that we all want kept confidential, was. It did turn out that this particular user was a legitimate additional account so no ruling was made on Sockpuppetry, but that's OK too, it's why we have permed admins who can review such stuff. The evidence for sockpuppetry was, I think, inconclusive but certainly suggestive so it was right to submit that material for examination.
Wiki is a mess, but all organic, living, complex things are messes. Take a look inside your spleen someday:-) I have my own edit war going (some anti-intellectuals deleted the "Erdos Number Category", and that's a mess because it's not so easy to explain why anyone cares about Erdos Numbers) but even the editor I would like to strangle honestly believes she's protecting the best interest of the encyclopedia (at the cost, I claim, of the Encyclopedia's content). We aren't monolithic, and we shouldn't be, because neither are scholarship or humanity. --yoof
OLPC isn't "a rich man's idea of what a poor man needs"; it's an educated man's idea of what an uneducated, and poor, man needs. In the U.S. we spend on the order of $10,000/yr on a child's education; OLPC is on the order of $10 to $100/yr. If I could buy a needy child $200 worth of books, it would be one laptop with a hotlink to Wikipedia; he could explore for a lifetime, at least until he is big enough to labor in the fields and have no more time for either study or play. Meanwhile he may have learned about low-tech things he can use, like crop-rotation and archimedian irrigation and boiling water.
Bantying metaphors is fun. I suggest: sell a beatup old pickup to a farmer in a region with no paved roads. It may still take him days to get his produce to market, but at least now he can get his produce to market, and it will help his region become prosperous enough to someday pave their roads. The point is bang for the buck, and this is a lot of bang, if you appreciate the positive effect of literacy and communication, which in the West we take for granted. -- yoof
The gloss misquotes the article in exactly the way to cause the effect we are talking about: the gloss says "scientists are more worried than the general public..." but the article's title is "scientists are more concerned...". The "King Kong" theme is that we tend to fear what we don't understand, but that's just what scientists don't do: they study what they don't understand. The survey merely established that most scientists haven't thought about the safety issues much yet. No reason to; it's like worrying about a rocket exploding at liftoff, when you are working on the aerodynamics of a single fin in a wind-tunnel. When you are designing for a prototype that will actually burn fuel, then is a good time to worry about exploding on lift-off. --yoof
I don't see the bias. I just read the first page of the survey. The first question gets how many client/workstations are "in your organization"; the second question is, what percentage of those is each of a short list of OS families (linux, "Windows", BSD, Mac). Presumably Dermoth666 answered "none" (or say 1-5%) for Linux and BSD and "more than 50%" (the high end) for "Windows". So right off the bat, the survey knows that his is not a unix workstation house. The next question asks about how linux is used (he would answer something like "servers"). Questions in a survey can't be taken usefully out of context. None of the later questions can be misleading about workstation linux use if it's already established that there is none at his organization. Good data can be analyzed badly, always, but in this case, competent reporting would be results like "30% of respondents reporting more than 20% linux workstation at organizations with more than 100 workstations said..." etc. -- yoof
There is precedent for one player knowing more about the other's preparation; for example, Fischer made extraordinary effort to develop a new opening repetoire for his match with Spassky in 72. Besides the Najdorf he played Alekhine's and the Pirc, neither of which had been seen at that level of play in a long time. And Fischer played queen pawn openings (mostly by transposition from English openings) which nobody had seen him do before. The effort to reinvent himself like that was superhuman; most of us lose many games the first time we try a new opening in tournaments, and he came up with several in a World Championship, and overtaxed Spassky's preparation. So if humans can do it maybe computers should be permitted also:-)
On the second point, of the surprising move: In the 80's Belle computed hundreds of thousands of positions per second to come up with her move. I don't know the numbers for Deep Blue, but that's obviously more analysis than a PR guy at IBM can synopsyse for you. There just aren't always simple explanations for the results of deductions that required millions of calculations. Playing a computer, you should expect to be surprised.
Slashdot Mirror
Maybe not so funny. The main engine for "truth" at Wikipedia is citation of a verifiable reference. Allowing for deductive logic would at least complicate the job for many editors, who would call it "Original Research" (which is disallowed). In the cast of the Totient Function proofs example, those proofs are all in books, some can be cited, so the only issue is notability. But it introduces the possibility of establishing the truth of something by giving an explicit proof, more plausible in math than most disciplines. Many editors wouldn't want the responsibility of validating a proof and would object to it as O.R., but in fact we all make reasonable inferences when we compose prose glosses of things for articles. It's a grey area with lots of room for debate. --yoof
I'd never heard of Durova until I looked into this on account of the Slashdotting. It's a mess, but it's not her doing and while there are in fact conspiracies, this isn't particularly one of them. :-) I have my own edit war going (some anti-intellectuals deleted the "Erdos Number Category", and that's a mess because it's not so easy to explain why anyone cares about Erdos Numbers) but even the editor I would like to strangle honestly believes she's protecting the best interest of the encyclopedia (at the cost, I claim, of the Encyclopedia's content). We aren't monolithic, and we shouldn't be, because neither are scholarship or humanity. --yoof
1. A user was banned temporarily for edit-warring, which is normal. The target of the war was "Irreducible Complexity" (which sounds scientific but...) which is about "Creation Science" (so consensus is nearly impossible).
2. The edit-warring was pretty well documented. But there was also an issue of "sockpuppetry" (some people who get banned, log in with new accounts to continue fighting for their cause). Since personal info is involved, Durova sent it to a group of permed admins; that's normal too. They had not yet reviewed it when the last ban occurred.
3. The result was the impression that a user had been banned for "secret" reasons, when actually a user had been banned for normal reasons, and the part that we all want kept confidential, was. It did turn out that this particular user was a legitimate additional account so no ruling was made on Sockpuppetry, but that's OK too, it's why we have permed admins who can review such stuff. The evidence for sockpuppetry was, I think, inconclusive but certainly suggestive so it was right to submit that material for examination.
Wiki is a mess, but all organic, living, complex things are messes. Take a look inside your spleen someday
OLPC isn't "a rich man's idea of what a poor man needs"; it's an educated man's idea of what an uneducated, and poor, man needs. In the U.S. we spend on the order of $10,000/yr on a child's education; OLPC is on the order of $10 to $100 /yr. If I could buy a needy child $200 worth of books, it would be one laptop with a hotlink to Wikipedia; he could explore for a lifetime, at least until he is big enough to labor in the fields and have no more time for either study or play. Meanwhile he may have learned about low-tech things he can use, like crop-rotation and archimedian irrigation and boiling water.
Bantying metaphors is fun. I suggest: sell a beatup old pickup to a farmer in a region with no paved roads. It may still take him days to get his produce to market, but at least now he can get his produce to market, and it will help his region become prosperous enough to someday pave their roads. The point is bang for the buck, and this is a lot of bang, if you appreciate the positive effect of literacy and communication, which in the West we take for granted. -- yoof
The gloss misquotes the article in exactly the way to cause the effect we are talking about: the gloss says "scientists are more worried than the general public..." but the article's title is "scientists are more concerned...". The "King Kong" theme is that we tend to fear what we don't understand, but that's just what scientists don't do: they study what they don't understand. The survey merely established that most scientists haven't thought about the safety issues much yet. No reason to; it's like worrying about a rocket exploding at liftoff, when you are working on the aerodynamics of a single fin in a wind-tunnel. When you are designing for a prototype that will actually burn fuel, then is a good time to worry about exploding on lift-off. --yoof
I don't see the bias. I just read the first page of the survey. The first question gets how many client/workstations are "in your organization"; the second question is, what percentage of those is each of a short list of OS families (linux, "Windows", BSD, Mac). Presumably Dermoth666 answered "none" (or say 1-5%) for Linux and BSD and "more than 50%" (the high end) for "Windows". So right off the bat, the survey knows that his is not a unix workstation house. The next question asks about how linux is used (he would answer something like "servers"). Questions in a survey can't be taken usefully out of context. None of the later questions can be misleading about workstation linux use if it's already established that there is none at his organization. Good data can be analyzed badly, always, but in this case, competent reporting would be results like "30% of respondents reporting more than 20% linux workstation at organizations with more than 100 workstations said..." etc. -- yoof
On the second point, of the surprising move: In the 80's Belle computed hundreds of thousands of positions per second to come up with her move. I don't know the numbers for Deep Blue, but that's obviously more analysis than a PR guy at IBM can synopsyse for you. There just aren't always simple explanations for the results of deductions that required millions of calculations. Playing a computer, you should expect to be surprised.