It's sad that most people don't realize that bookmarking is like roulette - you will lose on average no matter how good (statistical) information about the winner you have. This isn't strictly true - by definition, if you have perfect statistical information, you win every bet and cannot possibly lose on average.
Extending this down to worse statistics, to win in the long run all you need to do is have sufficiently better information than the bookie to ensure that you can overcome the extra padding they give to their chosen set of odds, which is not impossible in principle.
Of course, doing such a thing in practice is an entirely different kettle of fish, which is why it's still much better to be the bookie, but that glimmer of hope is why people keep playing.
The reason the GP brings up noise in the context of the different games is an implicit assumption that the "noise" in results is such that it follows Poisson statistics - in general, if you assume you're sampling some "true" probability of scoring/winning, the odds of getting a result which significantly deviates from the true value is smaller if the number of samples is higher. Consider how the ratio of heads to tails in coin tosses converges towards the mean as you perform more tosses as an example. Thus the GP assumes that more points scored and more games played constitutes more samples, hence yielding a much more accurate measurement of the quality of the team than would be obtained from a game with relatively lower absolute scores and fewer matches like football.
Of course, I wouldn't necessarily be convinced that the statistics are that much better behaved in basketball as opposed to other games, due to the variety of confounding factors you mentioned (which are fairly constant between most sports) which may not necessarily be represented in a simple scoreline.
Maybe the inverse square law doesn't work this way It doesn't. The radius of either body has no influence on the types of orbits they have - it's solely a result of their mass, and the mass of the earth wouldn't change at all if it became a black hole.
Ah, you're quite right about the iodine decay - I had a quick hunt to see what the energy was before I posted, but just found a mention that the beta energy was 200keV, and the gamma was "weak". I had presumed the article meant weak in comparison to the beta energy, and thus still not particularly easy to detect. Digging up an actual paper on the subject indicates that it's actually several hundred keV, so that's much more detectable a distance from the cat, although I'm still somewhat dubious about this particular story because of the intensities involved.
I think your post is being a little sensationalist, because really, radiation just isn't as bad as you seem to want to make it out in small doses. Chernobyl is a case in point - many of the dire predictions were just simply wrong, based off incorrect assumptions or outright scaremongering. The clearly attributable death rate from Chernobyl was relatively small (a few thousand), and the rest is largely within statistical noise in the population. The land in the exclusion zone is still largely unsuitable for growing crops if you want to err on the side of caution, but it's a relatively small area which was covered by a tremendous amount of radioactive material.
Moreover, although your scenario of targeting food suppliers seems worrying, I doubt it would have much effect in any western country. Very few of us regularly eat from the same suppliers, which means that even if a whole batch of contaminated food somehow got into the general populace, the produce would likely end up being so dispersed that the net effect would be very small. While you want to suggest that even the tiniest amounts of radiation are likely to give us irreparable health effects, this is simply not the case. Humans have evolved on a radioactive planet, containing many more isotopes than the shortlist you produced (and indeed, we are a non-trivial source of radiation ourselves. A favourite point of mine in this respect is that it was calculated that here in Ireland, one person each year dies because of the radiation emitted by their partner while they sleep together). As a result, we have mechanisms to cope with low-level damage, and indeed humans can take hundreds of times background radiation from many types of source with little health effects, which means that delivering much more than that in a concentrated dose is very difficult.
While I do agree that these things should be screened for, having such a terror of dirty bombs is not so desirable - as you say, they primarily want to promote fear in the populace, giving into it is bound to be a little counterproductive.
Except there is - external beam irradiation is effectively gone as soon as the treatment is finished. Despite the stories you see popping up sometimes, you don't get residual radiation floating around inside you after normal radiotherapy, as there are no long-lived isotopes which can be created in a human body through these sorts of treatments, and thus any excitations decay back to the ground state effectively instantly when treatment stops. Thus it's impossible to detect whether someone recently had radiotherapy recently with a Geiger counter or the like.
Contrast this to a situation where you have an implanted radioisotope, where it is constantly producing radiation as it decays towards its ground state, which can often have a half-life of days or weeks, which means there will likely be a significant radiation signature for some considerable time to come.
I don't think I131, or any of the other traditional nuclear medicine isotopes can reasonably be the culprit here: they're typically (perhaps exclusively, but I'm sure there's an exception somewhere) chosen because they develop their radiation almost entirely locally. In the case of I131, it undergoes beta decay, and the electrons are unlikely to escape the patient (and certainly not going to make it through a car, or the 80 feet to the detector). There is a possibility of the production of characteristic line x-rays also being produced as a product of ionisation caused by these electrons (although they're more common in the lighter Iodine isotopes which decay through electron capture rather than beta decay), but these too have a short range in the body: the most energetic x-ray which can be produced through this would be about 33keV (the Iodine k-edge), which would have a half-length in the body on the order of cm.
Either way, the total signal to be picked up would be minuscule - I'm sure it could be picked up by a close-by detector, if you were looking for it, but through a car 80 feet away? Colour me skeptical.
Yeah, I am somewhat suspicious of this story too. Most radiotherapy isn't going to cause any lingering radiation effects: While it's possible for x-ray or proton therapy to in principle cause nuclear excitations, these are relatively rare and have lifetimes of tiny fractions of a second. The only sort of radiotherapy that should leave a measurable trace would be Brachytherapy where a radioactive source is implanted into the patient. The issue is, these sources are typically chosen specifically because they emit radiation through processes with a short half-length in tissue (alpha, beta, or electron capture decays, mostly) so as to give better localisation of dose in the tumour.
And even if we ignore that fact and assume that they have for some reason implanted a source which emits purely high-energy gamma rays, it would need to be ridiculously hot to be picked up clearly above background 80 feet away: Radioactive decay is effectively a random process, so it spreads out in R^3 (in principle). Lets say the intensity was only just equal to background at the cop (probably an underestimate, given the nature of the counting statistics making it hard to pick up such a small increase conclusively). That means that at the surface of the cat (assuming it's 1 foot from the source), the intensity would be 500,000 times background - a ridiculously high amount, and pushing up into the LD50 for humans in less than a day. Anyone driving around with a cat that radioactive should certainly have been pulled over, if for their own safety if nothing else.
Yes, the helpful editing has now made the article more wrong. The milky way, as a whole, is further from 12,000 light years thick than it is from 1,000 light years thick. The issue is that stating a single thickness for the Milky Way is pretty much impossible - it's much thicker in the center than it is on the periphery, and if you don't bother to illustrate this fact the article is always going to be flawed.
(And before anyone asks, the reason I don't fix it myself is because I don't really care).
I've read these comments, and I see two potential issues with them:
Firstly, he makes a comparison to Wikipedia on day one, and says how all anyone would see is a page with a "funny editing syntax". The problem is, from all reports I've seen, this new engine doesn't even have the equivalent - it's just a standard algorithmic search with a very limited site list at the moment. If it had the beginnings of a collaborative search engine it may be more interesting, but as far as I can see this is all "in the future", and right now its just a bad search engine gathering together tags from users.
Secondly, I think Jimmy has misjudged his market slightly - he seems to feel that Wikipedia was successful because of "openness, transparency, and participation". Personally, I would guess it was because Wikipedia was free, when most comparable knowledge sources were closed and expensive. Sure, the first few users were probably idealists to whom the above points were important, but it's the free aspect and the momentum they gained from that which has been the real source of Wikipedia's popularity, I suspect. Given they lack a similar advantage over the well-established search engines, I don't think Jimmy can get away with being so blase about not having any slick releases.
There is no good scientific evidence linking MMR to autism, or even decent circumstantial evidence supporting such a link. Most of the experiments claiming to indicate such a link have been thoroughly discredited, and the theories claiming to explain the possible cause and effect are mostly junk, too. Case in point: you posted a link to a video about thiomersal, a mercury-based compound used in some vaccines (presumably thiomersal, anyway, as I only read the blurb to verify it was the standard "evil mercury" story). Despite there being no convincing evidence of a link between thiomersal and autism either, is often implicated in the stories about how MMR causes autism. The problem is the MMR vaccine does not contain thiomersal, and indeed never has, so the attempts to explain one in terms of the other is baffling.
You've never heard of this until now, so you seem surprised other people know about it? An interesting outlook. The idea of using protons for radiotherapy is relatively old (4 decades or more of papers from a quick google scholar search), and this particular type of facility is also well-established: it's been in commercial production for a decade or more, so it's not surprising its available on standard programs now. Unfortunately both you and the NYT are just a little behind the times.
Based on a couple of assumptions*, the entire reason for making use of this therapy is to mitigate the side effects of traditional radiotherapy. In traditional x-ray based therapies, the energy from the beam is deposited nearly continuously along the beam length, giving a roughly exponential falloff (I say nearly, as there is an initial buildup at the surface as secondary particle counts build up, and it is from the peak slightly below the surface that the exponential falloff begins).
By contrast, accelerated protons deposit their energy almost evenly, at a relatively low rate, until they are slowed to a certain energy, at which point their deceleration rapidly increases, accompanied by a massive increase in linear energy deposition. This leads to the "Bragg Peak", which offers a much, much more accurately targeted beam than is possible with conventional sources. (See this illustration as an example - compare the red line (in this case, C12 ions, but a similar principle) to the green line (an 18MeV photon beam). By carefully tuning the beam energy and orientation this point can be scanned over the tumour volume, giving a very localised dose deposition.
What puzzles me is why this is news - I was under the impression that this concept is well-established, and has been fairly well verified already. Just some fluff to fill up the science and medicine section, maybe? Now if it was about the CERN anti-proton tests, that's certainly something with a more dubious cost/benefit analysis...
* - I say a few assumptions, these are basically the principle ones behind all radiotherapy - that is, that all dose at the end of track structures is created equal and all dose is bad according to the LNT. While these ideas may not be strictly true, it is unlikely for them to be so wrong that it would invalidate the treatment as a whole.
It does imply "hotter than", if you quote the full definition from the article you cite:
... but unlike in the Solar System, where Jupiter orbits at 5 AU, the planets referred to as hot Jupiters orbit within approximately 0.05 AU of their parent stars, about one eighth the distance that Mercury orbits the Sun. Being only 1% as far away from their parent star does imply they would be significantly hotter than Jupiter (I say imply because I can't be bothered to work out the exact numbers on whether it would be feasible for such a planet orbiting a very cold star to be colder than Jupiter. I doubt it, but don't want to go around throwing out absolutes without basis).
Pfft, revolution my ass. This is just a bunch of people who are cranky because they got speeding tickets (and/or wanted to avoid tickets in the future) and took it out on the machines. Not to mention it's been going on for years on a low level (random BBC news story from September 2006). They don't care about liberty or the like, which is demonstrated both by their actions - they aren't bothering to try and destroy any sort of CCTV which actually keeps track of people in public areas - and their actual manifesto - that is, that the cameras are just a money-making scheme by the government.
It just demonstrates that civil liberties are to these people, a rather lower concern than, say, 50 quid in fines.
Well, looking at the article itself (I know, I know, heresy), the point is that there are whole classes of attacks (specifically "Cross Site Request Forgery" attacks, the focus of this article) which require significant effort on the part of websites to defend against, but which are trivially defended against by having users make a point of not accessing secure and insecure sites at the same time.
It's in no way presented as a solution to all security on the internet, but a way of addressing one specific class of problems in a simple manner with a minimum of effort. Unfortunately there's plenty of sufficiently smug people on/. who will continue to repeat this idea in this discussion without even glancing at the article.
Despite the fact that I'm still doggedly sticking it out with the books (I need closure after all these years, dammit), I really have to agree that the power escalation has gotten out of hand. Or, rather, the power de-escalation of the foes.
In the early books, even a few Trollocs or single Myrrdraal was an issue. By the middle books, they were being beaten up by farmwives with kitchen implements. By the late books, hundreds of them aren't really a big deal. A shame really, as when you lose respect for the foes, the series loses a lot of depth.
There's no where to put the radioactive waste and without HUGE government subsidies (no private insurer will cover nuclear power plants) Part of the problem here is that we simply don't understand the interaction between radiation and patient health - at high acute doses it's well characterised by the LNT model, but at low doses and dose rates we just don't know. As a result, the allowable exposure limits for a storage location for radioactive materials is vanishingly small (something like 0.03mSv a year, I think?), as opposed to an annual average background dose of about 2.4mSv. However, there are places on earth where the natural annual background is 100 times that - without significant health impacts on the population.
And even if the LNT is valid, at low doses of radiation there just isn't that much risk. If it wasn't for the somewhat mindless terror associated with the specter of radioactivity, the problem would be much less significant and could be dealt with fairly easily (The amount of high-grade waste produced by most plants is small, in terms of volume, and because of its high activity has a relatively short "dangerous" period). The low grade waste, hell, pay a fraction of the fees these storage people charge and you can bury it wrapped in a bit of steel and concrete in my back yard.
Don't be stupid. We're not talking about a perpetual motion device here. You use it once. It's used. If you want to 'reinvigorate' it to the point that it's usable as fuel again, you have to put in more energy than you'd get out of it. Fission is not close to 100% efficient, and there is a non-trivial number of other isotopes (notably plutonium) which are produced in a reaction which can be extracted and re-used. Obviously the yield from the spent uranium is less than the original uranium, but there is a non-trivial fraction of the "waste" products which can be reused to produce more power. Not an endless source of energy, but more power per unit mass of end product, which is certainly a good thing.
Yes, in principle, a certain number of people in Ireland also get cancer from their own potassium content. That's the nature of the Linear No Threshold model: You work out how many people are exposed to radiation and what the dose is, you can (in principle) directly work out how many cancers result from it. They don't actually go out and measure the cancers though, as its basically impossible to measure something so close to statistical noise. Hence the problem with verification at the low-dose limit.
This isn't strictly true: there is no good evidence that a minute dose of radiation is harmful, as the risk factors are so small that it's simply unfeasible to get them above background in a controlled fashion (Random "minute dose of radiation" fact: One person in Ireland every year gets cancer because of the potassium content of the person they're sleeping with). What was done was that high dose measurements fitted a linear line well, and so this line was extrapolated down to zero. This has never been verified, and as you say its primarily a precautionary measure, rather than something which has been rigorously demonstrated.
The issue is that, as suggested by this article, there is a big step from large, acute radiation doses to much smaller dose rates spread over a long time, and many of the assumed risk factors don't seem to apply on scales which are on a similar magnitude to background.
Interestingly, this issue actually came up in the technology sections of the UK media a while back, but under a rather different guise - that of "toothing".
Specifically, people would use bluetooth to discover other people with active devices (on trains or what have you), and send the message "toothing?" as an invitation to have sex in a nearby bathroom or similar. The media of course lapped it up, and for a while there was quite a bit of talking about what exactly you could and couldn't do over bluetooth on a standard phone.
Of course, it eventually turned out the whole "toothing" thing was a hoax. But it wouldn't surprise me if there were a lot of very confused people on trains around the UK for a while.
Slashdot Mirror
Extending this down to worse statistics, to win in the long run all you need to do is have sufficiently better information than the bookie to ensure that you can overcome the extra padding they give to their chosen set of odds, which is not impossible in principle.
Of course, doing such a thing in practice is an entirely different kettle of fish, which is why it's still much better to be the bookie, but that glimmer of hope is why people keep playing.
Of course, I wouldn't necessarily be convinced that the statistics are that much better behaved in basketball as opposed to other games, due to the variety of confounding factors you mentioned (which are fairly constant between most sports) which may not necessarily be represented in a simple scoreline.
Ah, you're quite right about the iodine decay - I had a quick hunt to see what the energy was before I posted, but just found a mention that the beta energy was 200keV, and the gamma was "weak". I had presumed the article meant weak in comparison to the beta energy, and thus still not particularly easy to detect. Digging up an actual paper on the subject indicates that it's actually several hundred keV, so that's much more detectable a distance from the cat, although I'm still somewhat dubious about this particular story because of the intensities involved.
Moreover, although your scenario of targeting food suppliers seems worrying, I doubt it would have much effect in any western country. Very few of us regularly eat from the same suppliers, which means that even if a whole batch of contaminated food somehow got into the general populace, the produce would likely end up being so dispersed that the net effect would be very small. While you want to suggest that even the tiniest amounts of radiation are likely to give us irreparable health effects, this is simply not the case. Humans have evolved on a radioactive planet, containing many more isotopes than the shortlist you produced (and indeed, we are a non-trivial source of radiation ourselves. A favourite point of mine in this respect is that it was calculated that here in Ireland, one person each year dies because of the radiation emitted by their partner while they sleep together). As a result, we have mechanisms to cope with low-level damage, and indeed humans can take hundreds of times background radiation from many types of source with little health effects, which means that delivering much more than that in a concentrated dose is very difficult.
While I do agree that these things should be screened for, having such a terror of dirty bombs is not so desirable - as you say, they primarily want to promote fear in the populace, giving into it is bound to be a little counterproductive.
Contrast this to a situation where you have an implanted radioisotope, where it is constantly producing radiation as it decays towards its ground state, which can often have a half-life of days or weeks, which means there will likely be a significant radiation signature for some considerable time to come.
Either way, the total signal to be picked up would be minuscule - I'm sure it could be picked up by a close-by detector, if you were looking for it, but through a car 80 feet away? Colour me skeptical.
And even if we ignore that fact and assume that they have for some reason implanted a source which emits purely high-energy gamma rays, it would need to be ridiculously hot to be picked up clearly above background 80 feet away: Radioactive decay is effectively a random process, so it spreads out in R^3 (in principle). Lets say the intensity was only just equal to background at the cop (probably an underestimate, given the nature of the counting statistics making it hard to pick up such a small increase conclusively). That means that at the surface of the cat (assuming it's 1 foot from the source), the intensity would be 500,000 times background - a ridiculously high amount, and pushing up into the LD50 for humans in less than a day. Anyone driving around with a cat that radioactive should certainly have been pulled over, if for their own safety if nothing else.
(And before anyone asks, the reason I don't fix it myself is because I don't really care).
Firstly, he makes a comparison to Wikipedia on day one, and says how all anyone would see is a page with a "funny editing syntax". The problem is, from all reports I've seen, this new engine doesn't even have the equivalent - it's just a standard algorithmic search with a very limited site list at the moment. If it had the beginnings of a collaborative search engine it may be more interesting, but as far as I can see this is all "in the future", and right now its just a bad search engine gathering together tags from users.
Secondly, I think Jimmy has misjudged his market slightly - he seems to feel that Wikipedia was successful because of "openness, transparency, and participation". Personally, I would guess it was because Wikipedia was free, when most comparable knowledge sources were closed and expensive. Sure, the first few users were probably idealists to whom the above points were important, but it's the free aspect and the momentum they gained from that which has been the real source of Wikipedia's popularity, I suspect. Given they lack a similar advantage over the well-established search engines, I don't think Jimmy can get away with being so blase about not having any slick releases.
There is no good scientific evidence linking MMR to autism, or even decent circumstantial evidence supporting such a link. Most of the experiments claiming to indicate such a link have been thoroughly discredited, and the theories claiming to explain the possible cause and effect are mostly junk, too. Case in point: you posted a link to a video about thiomersal, a mercury-based compound used in some vaccines (presumably thiomersal, anyway, as I only read the blurb to verify it was the standard "evil mercury" story). Despite there being no convincing evidence of a link between thiomersal and autism either, is often implicated in the stories about how MMR causes autism. The problem is the MMR vaccine does not contain thiomersal, and indeed never has, so the attempts to explain one in terms of the other is baffling.
You've never heard of this until now, so you seem surprised other people know about it? An interesting outlook. The idea of using protons for radiotherapy is relatively old (4 decades or more of papers from a quick google scholar search), and this particular type of facility is also well-established: it's been in commercial production for a decade or more, so it's not surprising its available on standard programs now. Unfortunately both you and the NYT are just a little behind the times.
By contrast, accelerated protons deposit their energy almost evenly, at a relatively low rate, until they are slowed to a certain energy, at which point their deceleration rapidly increases, accompanied by a massive increase in linear energy deposition. This leads to the "Bragg Peak", which offers a much, much more accurately targeted beam than is possible with conventional sources. (See this illustration as an example - compare the red line (in this case, C12 ions, but a similar principle) to the green line (an 18MeV photon beam). By carefully tuning the beam energy and orientation this point can be scanned over the tumour volume, giving a very localised dose deposition.
What puzzles me is why this is news - I was under the impression that this concept is well-established, and has been fairly well verified already. Just some fluff to fill up the science and medicine section, maybe? Now if it was about the CERN anti-proton tests, that's certainly something with a more dubious cost/benefit analysis...
* - I say a few assumptions, these are basically the principle ones behind all radiotherapy - that is, that all dose at the end of track structures is created equal and all dose is bad according to the LNT. While these ideas may not be strictly true, it is unlikely for them to be so wrong that it would invalidate the treatment as a whole.
... but unlike in the Solar System, where Jupiter orbits at 5 AU, the planets referred to as hot Jupiters orbit within approximately 0.05 AU of their parent stars, about one eighth the distance that Mercury orbits the Sun. Being only 1% as far away from their parent star does imply they would be significantly hotter than Jupiter (I say imply because I can't be bothered to work out the exact numbers on whether it would be feasible for such a planet orbiting a very cold star to be colder than Jupiter. I doubt it, but don't want to go around throwing out absolutes without basis).It just demonstrates that civil liberties are to these people, a rather lower concern than, say, 50 quid in fines.
It's in no way presented as a solution to all security on the internet, but a way of addressing one specific class of problems in a simple manner with a minimum of effort. Unfortunately there's plenty of sufficiently smug people on /. who will continue to repeat this idea in this discussion without even glancing at the article.
In the early books, even a few Trollocs or single Myrrdraal was an issue. By the middle books, they were being beaten up by farmwives with kitchen implements. By the late books, hundreds of them aren't really a big deal. A shame really, as when you lose respect for the foes, the series loses a lot of depth.
And even if the LNT is valid, at low doses of radiation there just isn't that much risk. If it wasn't for the somewhat mindless terror associated with the specter of radioactivity, the problem would be much less significant and could be dealt with fairly easily (The amount of high-grade waste produced by most plants is small, in terms of volume, and because of its high activity has a relatively short "dangerous" period). The low grade waste, hell, pay a fraction of the fees these storage people charge and you can bury it wrapped in a bit of steel and concrete in my back yard.
Yes, in principle, a certain number of people in Ireland also get cancer from their own potassium content. That's the nature of the Linear No Threshold model: You work out how many people are exposed to radiation and what the dose is, you can (in principle) directly work out how many cancers result from it. They don't actually go out and measure the cancers though, as its basically impossible to measure something so close to statistical noise. Hence the problem with verification at the low-dose limit.
The issue is that, as suggested by this article, there is a big step from large, acute radiation doses to much smaller dose rates spread over a long time, and many of the assumed risk factors don't seem to apply on scales which are on a similar magnitude to background.
Specifically, people would use bluetooth to discover other people with active devices (on trains or what have you), and send the message "toothing?" as an invitation to have sex in a nearby bathroom or similar. The media of course lapped it up, and for a while there was quite a bit of talking about what exactly you could and couldn't do over bluetooth on a standard phone.
Of course, it eventually turned out the whole "toothing" thing was a hoax. But it wouldn't surprise me if there were a lot of very confused people on trains around the UK for a while.
It's a typo. From what I've seen on the internet, both the UK and US are chock full of them.