Whether or not the asteroids started the evolution of life on earth is hard to tell, but does it really matter?
First of all, "evolution" isn't the issue here, it's biogenesis. Different concepts and it's important to keep them straight. (If only to keep the Creationists from confusing the two more than they already do.)
Second, yes, it matters. If the argument is, "Hey, meteorites have delivered organics, but Earth already had plenty," fine, but a) That's not what people, especially researchers, keep saying. b) No one cares if there's no connection to the terrestrial biogenesis. (OK, not no one. It's an interesting datum, but it lacks the cache to get published in the popular press.)
If their argument is that early Earth wasn't conducive to water, it's not clear how bringing in organics and water would help. If you bring in organics to a hot planet, they'll break apart just as surely as if they had formed there, after all.
It's never been clear to me why this mechanism is any better than just forming the danged organics on Earth surface. The Urey-Miller experiments demonstrated nicely that you can form organics under a wide range of conditions. (Which ones correspond to early Earth is an outstanding question, but it doesn't appear to much matter, oddly.)
Come to all that, we don't know that these asteroids (assuming they are asteroids and not dead comets, which it kind of sounds like they may be...) had much in the way of organics 4 billion years ago or if the organics formed due to reactions since then.
Basically, I'm uncomfortable with how excited people seem to get about the idea that this might have delivered the "building blocks of life" to Earth. Possible, sure, but it's far from a strong case.
A not unreasonable number of comets would be required. The oceans aren't that extensive compared the bulk of the planet, after all. You'd need more asteroids, obviously, since the water content is lower. (But since we're not really sure how much water is in an asteroid, let alone was in them 4 billion years ago, it'd be difficult to come up with even a ballpark figure.)
Unfortunately, if they can't understand the consent form then they can't give informed consent, can they?
True, up to a point. How do you verify that they really understood all the ramifications of the agreement, including all of the things you assume are obvious? (Or that they assume you mean?) You could give a 2-hr test to every person, I suppose, to verify that they understand, but that seems a bit excessive.
In general, misunderstandings and failures to use the same underlying assumptions will always exist.
Your definition of x' is only an approximation and not useful for many purposes. As such, it doesn't really usefully expand the set of differentiable functions. In this case, x may be a wildly non-linear function and a small change in input may yield radically different results. As such, it's not differentiable in those areas.
Intuitively, you know your logic doesn't hold. If it did, no optimization problem would ever be unsolvable with the calculus of variations. That's clearly not the case, however.
As for your objection that a GA doesn't produce the best possible solution: so what? Most real world problems have so many uncertainties and other issue defining "best" that it's a moot point anyway. That, and often the fractional improvement between the true optimal solution and the best guess is so small as to not be worth pursuing.
All that said, I've repeatedly said that I don't think that the tour planners used a GA or any other such algorithm. From hearing them talk to us in meetings, it sounds like the manually build and optimize tours themselves.
No, this would be a great problem for a GA, in general. You can't do a variational calculus approach to tour selection because the parameter space is huge and not a differentiable function, in general. A GA would be fine if you had the time, I suspect. You just give it a goodness of fit metric (like meets so many priority goals of the science team) and let it loose.
In reality, this is kind of what they did, but they did it iteratively with the scientists themselves. I'm not sure how much they numerically optimized, although hearing them talk about it, mostly it seems like skill and hard work by the tour designers.
Mostly from what I saw of this process, they build trajectories using a toolkit of tricks to make certain specific objectives. (Objectives are things like, "fly by this moon" or "spend N hours above this latitude on the planet.") I don't know how much they can optimize within the vast parameter space of the orbital trajectories, mostly it seemed like a lot of man-hours and skill.
That's because its initial trajectory was unbelievably convoluted: the ship actually traveled to Venus first, got a gravity boost then traveled back out and used the Earth for its next boost.
Venus twice, then Earth. What's unbelievable about it? Gravity assists have been used on so many probes now that it's basically assumed for any target other than Venus or Mars. (They even used it to get to Mercury with MESSENGER.)
The plutonium scare was mostly an overreaction by people who like to overreact to whatever they can find. Many of them are now complaining about the LHC, I'm sure. (NASA designs the RTGs to be nearly indestructible and they biased the Cassini trajectory away from Earth on the flyby so an accident would send it away from us.
One of the key administrators involved has been answering all questions about the program by invoking the Fifth Amendment.
Which, to be fair, is entirely his or her right. Trying to infer guilt from this (tempting though it may be) violates what most of us stand for. Tossing that statement in at the end of the summary seems to be an attempt to imply guilt, though.
(Which isn't to say that I don't think this program was stupid and criminal.)
"We know people who think this is a nonsense subject. And we'll refer you to voluminous literature and facts about UFOs."
Seriously? These guys do understand that "voluminous" literature doesn't equate to "quality literature", right? There are tomes and tomes on dragons at your local library, but I don't think many of us would consider "Draconic Studies" a worth academic pursuit.
. The results, which he shared exclusively with TIME, represent the largest study of financial incentives in the classroom — and one of the more rigorous studies ever on anything in education policy.
So... these results, they are unvetted? Not published or peer-reviewed yet? In fact, it sounds rather like they aren't on their way to being peer-reviewed, even. Then again, either Fryer or the author has a fairly serious misunderstanding of what things were about, given the repeated statements to the effect of Fryer was setting things up "just like a real scientist." Either this is scientific or it isn't.
In many ways, this guy seems more interested in attention than in the validity of the results. (The article doesn't really help much. It's mostly a story told with Fryer as the hero and very little contrary view, although we're told he encountered resistance. Why? How much of that resistance was well-founded?)
The point isn't about third parties stealing the nuclear material, it's about having accidents with nukes in the first place. Even if they only partially detonating 1 time in 100, if you do it a bunch of times, you stand fair odds of accidentally nuking someone.
What the other replier said: aggregate stats. Your odds of an A are only 1 in 39 on any given letter, but when you're stamping out a thousand characters, odds go well above 99% of getting at least one. Similarly, odds may be only a few percent that a given accident with nuclear weapons would result in a real "nuclear accident", but if you have a few dozen accidents, eventually the really bad outcome is likely to occur.
I'm not so worried about the lost devices as the accidents at the moment that they happen, when accidental detonation should be most likely. (I have to imagine, anyway.) Those devices will be in good firing order, too, since that's their whole point and the point of maintaining them.
What would be interesting to see is if the old bombs that have been left around have maintained the perfect symmetry required to properly compress the plutonium and ignite the nuclear fire; otherwise the ensuing explosion will be weak compared to the optimum yield, if it can occur at all.
Absolutely true and likely. However, a "weak" nuclear blast in an inhabited area would still suck for those involved, and that's the thing to remember. "It could be worse" is well and good, but it often overlooks the fact that it's still bad.
You're assuming that this article is talking about lost (ie, stolen) materials. It's about accidents, so you're arguing the wrong case to start with. And yes, it's more likely that a device will be too damaged to properly explode in an accident, but given enough accidents, odds are pretty good that at least a partial nuclear detonation could occur. Failing that, a blast of the conventional explosives (which has happened) could scatter some rather nasty radioactive material about, possibly in a residential area.
It's pretty hard to split an atom, which is why we poured billions into learning how during the Manhattan Project.
True enough that there are way more ways for an accident to not result in a full detonation than to do so. But the above statement is a bit misleading: thanks to the Manhattan project, we now have devices lying around that are designed to split atoms. (Itself, not difficult. Nature does it every second of every day.) Comparing the probability of an accident yielding a nuclear explosion to the pre-Manhattan odds is dubious.
That's not evidence, that's accusing the other side of bias. (An accusation which, unsupported, cuts all ways.) And while I can't guarantee that none of the sources I read aren't biased, I do pay some attention to the research in the area and have found a fairly consistent view that no diet plan offered yet works universally.
Do you have any citations of clear studies that show that carbs are bad? Because I can point out quite a few human populations that eat a lot of carbs by any standard and are yet quite healthy. I'm not saying fat is the problem with our diet, but I am suggesting that many people seek out magic bullets to fix diets when in fact no such thing may exist and nutrition may require a much more individual arrangement. (In fact, the only people I find arguing for such magic bullets are either selling their diet plan or people without the expertise to judge who have bought into one of those people's plans.)
It depends on the state. There's often (usually?) an exception in place for parties within a given age gap. The only state I can recall for certain sets it at 18 months. A 17-year-old therefore is quite likely outside of that range. An 18-year-old almost certainly is. (Allowing for possible differences in when we've dated the ages of the parties involved.)
Also, you have to assume that it was consensual (the article is exceedingly sparse on those details) in order to bemoan the charge. You could just as well assume the opposite and condemn the young men right now, with what we know. Unless you know more than I've seen, it's premature to be making judgments either way.
Which studies are these? Last I checked, there is no consensus on what diets lead to weight loss consistently over all populations. In fact, there's a growing body of evidence that indicates that there may be a widely varying set of diets for people of different genetic backgrounds. Which would explain why so many cultures eat carb-heavy diets and are far healthier than Americans. (Not that there aren't other factors, of course.) Scientific American (I think) had an article on it a few months ago.
Also, physics *does*, broadly, explain the caloric in/out. If you eat fewer calories than you expend, you will lose weight because the energy has got to come from somewhere. The problem is in account for energy consumed and energy used. Different people apparently process food differently, so isn't as simple as figuring out the total energy freed up in oxidizing the food.
Yup. Actually, bottled water in general is a good counter example to the original post: tap water in the US is actually better regulated than bottled water (so the threshold to which it's clean is stricter, unless the bottler wants to be nicer than legally required), but with some snazzy marketing, people pay for bottled water thinking it's better than tap water.
Slashdot Mirror
Whether or not the asteroids started the evolution of life on earth is hard to tell, but does it really matter?
First of all, "evolution" isn't the issue here, it's biogenesis. Different concepts and it's important to keep them straight. (If only to keep the Creationists from confusing the two more than they already do.)
Second, yes, it matters. If the argument is, "Hey, meteorites have delivered organics, but Earth already had plenty," fine, but
a) That's not what people, especially researchers, keep saying.
b) No one cares if there's no connection to the terrestrial biogenesis. (OK, not no one. It's an interesting datum, but it lacks the cache to get published in the popular press.)
If their argument is that early Earth wasn't conducive to water, it's not clear how bringing in organics and water would help. If you bring in organics to a hot planet, they'll break apart just as surely as if they had formed there, after all.
It's never been clear to me why this mechanism is any better than just forming the danged organics on Earth surface. The Urey-Miller experiments demonstrated nicely that you can form organics under a wide range of conditions. (Which ones correspond to early Earth is an outstanding question, but it doesn't appear to much matter, oddly.)
Come to all that, we don't know that these asteroids (assuming they are asteroids and not dead comets, which it kind of sounds like they may be...) had much in the way of organics 4 billion years ago or if the organics formed due to reactions since then.
Basically, I'm uncomfortable with how excited people seem to get about the idea that this might have delivered the "building blocks of life" to Earth. Possible, sure, but it's far from a strong case.
A not unreasonable number of comets would be required. The oceans aren't that extensive compared the bulk of the planet, after all. You'd need more asteroids, obviously, since the water content is lower. (But since we're not really sure how much water is in an asteroid, let alone was in them 4 billion years ago, it'd be difficult to come up with even a ballpark figure.)
Unfortunately, if they can't understand the consent form then they can't give informed consent, can they?
True, up to a point. How do you verify that they really understood all the ramifications of the agreement, including all of the things you assume are obvious? (Or that they assume you mean?) You could give a 2-hr test to every person, I suppose, to verify that they understand, but that seems a bit excessive.
In general, misunderstandings and failures to use the same underlying assumptions will always exist.
Your definition of x' is only an approximation and not useful for many purposes. As such, it doesn't really usefully expand the set of differentiable functions. In this case, x may be a wildly non-linear function and a small change in input may yield radically different results. As such, it's not differentiable in those areas.
Intuitively, you know your logic doesn't hold. If it did, no optimization problem would ever be unsolvable with the calculus of variations. That's clearly not the case, however.
As for your objection that a GA doesn't produce the best possible solution: so what? Most real world problems have so many uncertainties and other issue defining "best" that it's a moot point anyway. That, and often the fractional improvement between the true optimal solution and the best guess is so small as to not be worth pursuing.
All that said, I've repeatedly said that I don't think that the tour planners used a GA or any other such algorithm. From hearing them talk to us in meetings, it sounds like the manually build and optimize tours themselves.
No, this would be a great problem for a GA, in general. You can't do a variational calculus approach to tour selection because the parameter space is huge and not a differentiable function, in general. A GA would be fine if you had the time, I suspect. You just give it a goodness of fit metric (like meets so many priority goals of the science team) and let it loose.
In reality, this is kind of what they did, but they did it iteratively with the scientists themselves. I'm not sure how much they numerically optimized, although hearing them talk about it, mostly it seems like skill and hard work by the tour designers.
Mostly from what I saw of this process, they build trajectories using a toolkit of tricks to make certain specific objectives. (Objectives are things like, "fly by this moon" or "spend N hours above this latitude on the planet.") I don't know how much they can optimize within the vast parameter space of the orbital trajectories, mostly it seemed like a lot of man-hours and skill.
That's because its initial trajectory was unbelievably convoluted: the ship actually traveled to Venus first, got a gravity boost then traveled back out and used the Earth for its next boost.
Venus twice, then Earth. What's unbelievable about it? Gravity assists have been used on so many probes now that it's basically assumed for any target other than Venus or Mars. (They even used it to get to Mercury with MESSENGER.)
The plutonium scare was mostly an overreaction by people who like to overreact to whatever they can find. Many of them are now complaining about the LHC, I'm sure. (NASA designs the RTGs to be nearly indestructible and they biased the Cassini trajectory away from Earth on the flyby so an accident would send it away from us.
Which is perfectly fair and sensible. But of course, adding that she's taking the fifth is superfluous, given that those facts exist.
One of the key administrators involved has been answering all questions about the program by invoking the Fifth Amendment.
Which, to be fair, is entirely his or her right. Trying to infer guilt from this (tempting though it may be) violates what most of us stand for. Tossing that statement in at the end of the summary seems to be an attempt to imply guilt, though.
(Which isn't to say that I don't think this program was stupid and criminal.)
You're joking, right? Right!?
Wire your tuition to the bank account I've set up in Nigeria and we'll be set!
"We know people who think this is a nonsense subject. And we'll refer you to voluminous literature and facts about UFOs."
Seriously? These guys do understand that "voluminous" literature doesn't equate to "quality literature", right? There are tomes and tomes on dragons at your local library, but I don't think many of us would consider "Draconic Studies" a worth academic pursuit.
I read on, but I really got skeptical at:
. The results, which he shared exclusively with TIME, represent the largest study of financial incentives in the classroom — and one of the more rigorous studies ever on anything in education policy.
So... these results, they are unvetted? Not published or peer-reviewed yet? In fact, it sounds rather like they aren't on their way to being peer-reviewed, even. Then again, either Fryer or the author has a fairly serious misunderstanding of what things were about, given the repeated statements to the effect of Fryer was setting things up "just like a real scientist." Either this is scientific or it isn't.
In many ways, this guy seems more interested in attention than in the validity of the results. (The article doesn't really help much. It's mostly a story told with Fryer as the hero and very little contrary view, although we're told he encountered resistance. Why? How much of that resistance was well-founded?)
The point isn't about third parties stealing the nuclear material, it's about having accidents with nukes in the first place. Even if they only partially detonating 1 time in 100, if you do it a bunch of times, you stand fair odds of accidentally nuking someone.
What the other replier said: aggregate stats. Your odds of an A are only 1 in 39 on any given letter, but when you're stamping out a thousand characters, odds go well above 99% of getting at least one. Similarly, odds may be only a few percent that a given accident with nuclear weapons would result in a real "nuclear accident", but if you have a few dozen accidents, eventually the really bad outcome is likely to occur.
You're quite right, but I don't know why you're replying to my post. Your point has nothing to do with what I said.
I'm not so worried about the lost devices as the accidents at the moment that they happen, when accidental detonation should be most likely. (I have to imagine, anyway.) Those devices will be in good firing order, too, since that's their whole point and the point of maintaining them.
What would be interesting to see is if the old bombs that have been left around have maintained the perfect symmetry required to properly compress the plutonium and ignite the nuclear fire; otherwise the ensuing explosion will be weak compared to the optimum yield, if it can occur at all.
Absolutely true and likely. However, a "weak" nuclear blast in an inhabited area would still suck for those involved, and that's the thing to remember. "It could be worse" is well and good, but it often overlooks the fact that it's still bad.
You're assuming that this article is talking about lost (ie, stolen) materials. It's about accidents, so you're arguing the wrong case to start with. And yes, it's more likely that a device will be too damaged to properly explode in an accident, but given enough accidents, odds are pretty good that at least a partial nuclear detonation could occur. Failing that, a blast of the conventional explosives (which has happened) could scatter some rather nasty radioactive material about, possibly in a residential area.
It's pretty hard to split an atom, which is why we poured billions into learning how during the Manhattan Project.
True enough that there are way more ways for an accident to not result in a full detonation than to do so. But the above statement is a bit misleading: thanks to the Manhattan project, we now have devices lying around that are designed to split atoms. (Itself, not difficult. Nature does it every second of every day.) Comparing the probability of an accident yielding a nuclear explosion to the pre-Manhattan odds is dubious.
That's not evidence, that's accusing the other side of bias. (An accusation which, unsupported, cuts all ways.) And while I can't guarantee that none of the sources I read aren't biased, I do pay some attention to the research in the area and have found a fairly consistent view that no diet plan offered yet works universally.
Do you have any citations of clear studies that show that carbs are bad? Because I can point out quite a few human populations that eat a lot of carbs by any standard and are yet quite healthy. I'm not saying fat is the problem with our diet, but I am suggesting that many people seek out magic bullets to fix diets when in fact no such thing may exist and nutrition may require a much more individual arrangement. (In fact, the only people I find arguing for such magic bullets are either selling their diet plan or people without the expertise to judge who have bought into one of those people's plans.)
It depends on the state. There's often (usually?) an exception in place for parties within a given age gap. The only state I can recall for certain sets it at 18 months. A 17-year-old therefore is quite likely outside of that range. An 18-year-old almost certainly is. (Allowing for possible differences in when we've dated the ages of the parties involved.)
Also, you have to assume that it was consensual (the article is exceedingly sparse on those details) in order to bemoan the charge. You could just as well assume the opposite and condemn the young men right now, with what we know. Unless you know more than I've seen, it's premature to be making judgments either way.
Which studies are these? Last I checked, there is no consensus on what diets lead to weight loss consistently over all populations. In fact, there's a growing body of evidence that indicates that there may be a widely varying set of diets for people of different genetic backgrounds. Which would explain why so many cultures eat carb-heavy diets and are far healthier than Americans. (Not that there aren't other factors, of course.) Scientific American (I think) had an article on it a few months ago.
Also, physics *does*, broadly, explain the caloric in/out. If you eat fewer calories than you expend, you will lose weight because the energy has got to come from somewhere. The problem is in account for energy consumed and energy used. Different people apparently process food differently, so isn't as simple as figuring out the total energy freed up in oxidizing the food.
Germans seem to be smarter than Americans, or at least less susceptible to advertising.
(I've never been to Germany, but in other European countries, I've often gotten funny looks for asking for tap water.)
Yup. Actually, bottled water in general is a good counter example to the original post: tap water in the US is actually better regulated than bottled water (so the threshold to which it's clean is stricter, unless the bottler wants to be nicer than legally required), but with some snazzy marketing, people pay for bottled water thinking it's better than tap water.