I don't owe him an apology. I am a physicist, and my field of expertise is foundations of quantum mechanics. I did not say that this idea is insane as a knee-jerk reaction. I have read his paper, and I've been to a presentation by the men himself.
I find superdeterminism isane because it is too strong a hyphotesis: it can explain any experimental result. It is, therefore, incompatible with the very idea of doing science. Let me give you an example. Suppose that you want to measure some property of a particle, say, the polarization of a photon. In a superdeterministic theory, the photon's state could depend on which basis you chose to do the measurement, so you could conclude nothing at all from the result. Even the idea the a particle has some property independ of your choice of measurement does not apply.
And it's also deeply troubling how he arrived at superdeterminism. He wanted to do a deterministic model of quantum theory, but there is these pesky Bell inequalities tell you it's impossible. So what does he do? Simply assumes superdeterminism to brush off the issue.
It's not as if he was building a superdeterministic model per se, it is just an ad hoc hyphotesis that he added to get rid of Bell inequalities.
This kind of article bothers me immensely. It treats Einstein as the God of Science, and uses the fact the he worked on something as evidence that this idea is no crackpottery. Well, guess what, Einstein also shat, farted, pissed, had bad ideas, and even commited mathematical mistakes.
And one should never evaluate a scientific idea based on who's working on it. The Steady-State model of the universe is not a crackpot idea, simply because it is consistent with the laws of GR and (superficially) consistent with observational evidence. Philosophically, thought, it does seem quite silly, and I myself would never have regarded it as more than a mathematical curiosity, had it not been already falsified when I was born.
A more modern example would be 't Hooft's work on superdeterminisc models for quantum theory. The guy is obviously a genius, but this idea is pure insanity, and it saddens me to see people taking it seriously just because a Nobel prize is working on it.
That was a sarcastic answer to GP; I'm not one of these people, I think Sweden is a great country, and I'm fact gonna visit there in two weeks.
I'm aware that yours is the "correct" definition of socialism, but you should also taking into account that the definition is very much culture-dependent. People in Europe usually refer to Sweden as socialist, and I'm happy to go along, since the old form of socialsm doesn't exist anymore in Europe.
You should do some research on how Cuba was before the revolution. People don't revolt for nothing, you know? Although it's undeniable that today Cuba is quite poor (partly because of their own economic mismanagement, partly because of the US embargo), it is still in a better shape than it was under Fulgencio Batista. At least the people now have universal access to health and education.
Come on, Minecraft does not pretend the implement classical physics.
The problem isn't lack of rigour, is that it gives you the wrong intuition. For example, the essential feature that distinguishes superposition from a classical mixture is that there is a basis in which the result of a measurement is deterministic, and from what I have seen this is not the case in the mod.
One way that it could be done: a "superposition" block, that can be prepared in the states |0>, |1>, |+>, or |->. If you look at it vertically, you measure in the Z basis, and if you continue measuring in the Z basis, the result doesn't change. If you then measure in the X basis, it's gonna collapse randomly to either |+> or |->, and it's gonna keep being this one as long as you keep looking at it vertically.
And this gives you also complementarity, which I think is one of the fundamental quantum concepts. The way the "observation" block was implemented simply makes no sense whatsoever. It's just a block that is different depending on the way you look. There is no complementarity between the aspects of the block, or anything to deal with the randomness of the quantum observation.
I don't understand what are you talking about teleportation. It doesn't sound to me very much related to real quantum teleportation. But I think the trouble with teleportation is that it is actually quite boring, and would serve no purpose in a game.
But it would be nice if we had some "quantum interaction" thingie (a CNOT) that would take two "superpositions" in different basis to an entangled state, and with this entangled state we could run the chsh game or ek91 qcrypto protocol.
Well, you could say that doing a PhD in physics is playing such a game, and, sure enough, most physicists have a working intuition on how quantum mechanics work -- more on the level of given a description of a situation, they can tell more or less what will happen without doing the calculatios. Doing that part is not hard at all. Heck, this kind of intuition mathematicians develop about the most abstract and artificial objects.
But on a more fundamental level, I don't think it is possible to develop a good intuition, because our brains process information classically; we need well defined bits to reason about. But it might not even make sense to be able to reason quantumly, as the essential feature would be preserving the superposition of the systems we interact with. But to preserve the superposition is to have no memory of the interaction, so... it would be very weird indeed.
Try to imagine what a quantum computer would feel as it processes quantum information: it can only apply transformations to its information blindly, without ever reading out what its input actually is, until the very end of the calculation -- and even this final measurement only because we, humans, want it, it's perfectly legitimate to never have a final measurement at all.
I am a physicist, but these are only drunken speculations...
At first I was quite excited with the idea that someone was able to use quantum mechanical elements in a game. But of course, they were not able to do this. They just created a mod vaguely inspired by quantum mechanics, that helps to perpetuate the myths so beloved by the lay media.
The video linked just shows a dude running around, nothing very interesting. If you search youtube a bit, you can find videos talking about the mechanics they implemented. I found this one, about the basic elements -- observation, superposition, and entanglement --, and this one, with the extremely exciting title "quantum computers and teleportation".
Of course, what they call observation and superposition have nothing to do with the quantum concepts, they are just blocks that are different depending on which direction you look at them, and the "entanglement" block is just a glorifed telephone. Their quantum computer doesn't seem to do anything besides teleportation, which is Star Trek teleportation instead of quantum teleportation.
Admitedly, these guys set out do to a terribly difficult task: quantum mechanics is a bit subtle, and quite far from games. The only ones I can remember off the top of my mind are the CHSH game, which is about as exciting as tic-tac-toe, and a quantum strategy to cheat at bridge, which requires you to do a nontrivial amount of maths (and is actually unpublished research =).
They have, nevertheless, failed. The mod looks cool as a game, though.
This is completely beside the point (although true). The problem is that the voting system allows for such a thing to happen. Ever wondered why no European democracy has a two-party state? Well, they have sane voting laws. First past the post system without runnofs is just insane. Gerrymandering, electoral college, come on. Once the US was an inspiration to every democracy in the world. Now it has become a laughing matter.
As far as I know the Bristol folks do integrated optics, so it is in fact a chip; not as small as a classical chip, but about the size of a fingernail.
As a side note, it makes me sad that such an awesome project had to taint itself with the mention of Google and NASA waste of millions of dollars to D-Wave. What the Bristol people are doing is quantum computing. What D-Wave is doing isn't.
I don't think the analogy holds. Certainly knowing the details of quantum optics is quite labourious, but doing quantum programming is already some kind of quantum mechanics, that does not reference any particular physical system. Just like a Turing Machine uses Boolean algebra, but does not reference any particular machine to implement it.
Also, it's true that the algorithms are described in purely classical terms, but they probably sound like gibberish if you are not familiar with the concepts. As an example, let me describe you one of the simplest quantum algorithms known, the single-qubit version of Deutsch's algorithm, which decides whether a (classical) single-bit function f is such that f(0)=f(1) or f(0) != f(1):
Initialize your qubit in the state |0> Apply the Hadamard gate to it Apply the oracle that calculates f to it; it is described by the operator e^(i pi f(0)) |0>1| Apply the Hadamard gate to it again Measure in the computational basis.
I don't think it is particularly hard to understand it, though; if you want to do it seriously, there's a quite good book about the subject: it's called "Quantum Computation and Quantum Information", by Michael Nielsen and Isaac Chuang.
Can you elaborate on that? My laymen impression is that prosecutorial discretion allows you to at any time ignore any law you wish. But to ignore a specific law all the time seems a bit of an exaggeration.
So you'd rather trust a 89-year physicist with no training in climate science than the ones who have actually learned the subject? That's very enlightened of you.
I don't know what you mean by "very skeptical", but wikipedia says he does believe that the planet is warming and it is our fault; he seems to distrust the mathematical models we used to predict the climate.
For whatever it's worth, I am a physicist, and I have never met (personally) a physicist that is a climate change denier.
This sounds cool, but actually doesn't make any sense. The problem is not big endian versus litlle endian. Microsoft could very well use com.microsoft.xbox.360, but then somebody could squat com.xbox.360. Or microsoft could use com.xbox.360, and somebody would try to squat com.xbox360. The advantage of USENET is that its hierarchical structure was more or less well defined, while in the WWW it is completely arbitrary what you put before the.com part.
How the source adds nothing to the discussion? I've seen this on slashdot when the story first appeared, and the media coverage was uniformly believing CERN's hype, without a shred of scepticism. This blog has a simple, correct explanation of the physics involved, and its interpretation of the data pretty much agrees with I've been hearing at the University.
I did just that. Watched "the wrath of khan" for the first time this wednesday and then watched "into darkness" yesterday. Damm sure this made me dislike "into darkness" more. It's just a rehatch of the "emotional" scenes of "the wrath of khan" without any coherent plot to link them together. While "the wrath of khan" had a plot that made sense, and was thought provoking, this one had... nothing.
I wouldn't mind to see good visual effects and some action scenes in a Trek movie. But why can't they also use a decent script? It's not difficult to find somebody capable of writing a better story then the one that was used.
I don't mean to insult you, but you seem to be talking without having read TFP, which I did. It is here. (Incidentally, this paper just repeats the arguments first published here, but with a better experimental setup.)
I'm saying this because, as the paper says explicitly, the number 10,000 c has nothing to do with resolution of the clocks, but with their choice of preferred reference frame.
And the experiment I had in mind was quantum teleportation, where it is a little bit less insane talking about the "speed of quantum information".
So what you do is to send a quantum state from Alice to Bob. If you do the calculations, they show that quantum teleportation instantly transfers the state from Alice to Bob. But Bob is only capable of acessing the state after A also sends him some auxiliary classical information (two bits, to be precise). That's why I said that the only scientific thing that you can claim is that the information propagates at lightspeed, because before receiving also the classical information no measurement that Bob can do can say whether he received the state or not.
About the subjectivity, I know this is weird, but there dosn't seem to be any way around it. The state that the calculations show changing instantaneously is Alice's knowledge of the spin, which in fact does change as soon as she makes the measurement. But it makes no sense to say that this is also Bob's knowledge about the state of the particle, since he has no idea about what's going on.
This is bullshit. The scientific content behind this claim is that "nonlocal realistic models that reproduce the results of quantum mechanics must have speed of communication at least 10,000 faster than the speed of light in some arbitrary ference frame that we've chosen".
This means that this number is completely irrelevant, i.e., does not measure anyhting related to the real world.
What can be said, scientifically, about the speed of this channel is that it is the speed of light, because we can only actually measure the presence of the information on the other side after a light signal is sent from one party to the other.
The fact that it looks instantaneous is more of an artifact of our mathematical formalism, and a common philosophical misunderstanding about the nature of the quantum state (i.e., people regard it as objective rather than subjective).
Jackpot! This is the whole issue. Just observing these correlations means nothing, as you said, the same data comes out of our friends with the cards.
But the reason we are fascinated by entanglement is that there is more to it. We can actually prove that no theory in which the contents of the letters were predetermined (and FTL communication does not exist) can reproduce the correlations that can be produced via entangled states. This is the famous Bell's theorem.
Sorry, this is not the case. Again, I'm not an expert on this, but the best evidence I now that dark matter isn't just normal matter that happens to be dark is the Bullet Cluster. The thing is that, even if we can't see the normal matter, we know that it interacts with normal matter, well, normally. So in a collision of galactical clusters, we expect even this dark gas to interact a bit and get left behind, while the interactionless dark matter passes straight throught, and this is what we have observed in th Bullet Cluster.
If the effects we attribute to dark matter could be explained by known matter I don't think anybody would be excited about it =)
The sentences that could have given you warning were "An excess of antimatter within the cosmic ray flux was first observed around two decades ago.", i.e., the effect was already well-known, and "the AMS measurement can not yet rule out the alternative explanation that the positrons originate from pulsars distributed around the galactic plane. Supersymmetry theories also predict a cut-off at higher energies above the mass range of dark matter particles, and this has not yet been observed.", i.e., boring explanations were not ruled out, and the "smoking gun" evidence for dark matter was not yet found.
As usual, this is just a press release full of hype.
They didn't discover dark matter. They measured, with higher precision than ever, the excess in the positron fraction coming from cosmic rays. The existence of this effect, however, was already well-established. The question that was open, and still is, is which is the origin of this effect. One of the possible answers is dark matter. The problem with this answer is that we have to assume a discredited theory -- supersymmetry, and even within this theory a very artificial model of dark matter annihilation. The higher precision of the current measurements does not credence to this answer, nor does it discard more boring answers (i.e. coming from astrophysical processes that do not involve new physics). If you want to understand more about it, please read it from an actual particle physicist. I am a physicist, but not an astrophysicist nor a particle physicist.
Please keep in mind that I'm not criticising the AMS experiment itself: its job was to measure this excess with high precision, and this it did quite well. What I'm criticising is the people who have published this irresponsible press release.
Hmm, not in my definition of "few hundred". The calculation is actually easy to make:
The earth is about 1,5E11 m away from the Sun, let's say that 1% is the variation that we want, so we get it to 1,515E11 m. So the difference in energy that we need is GMm(1/R1-1/R2) \approx 5E31 J; quite a lot.
The best (or worst, depending on your point of view) nuke we ever exploded is the Tsar Bomba, which was 57 megatons or better 2,4E17 J.
So if we managed to use this energy with 100% efficiency (which we obviously can't) to move the Earth, we would need 10^14 nukes. Well, guess we're stuck here.
Have you ever succeeded in changing someone's beliefs in pseudoscience? Do you think that it is possible to do so in a large scale, to move humanity towards a more rational way of thinking?
Sorry for the down tone, but I have plenty of experience in failing to convince people of the falsehood in astrology, homeopathy, acupunture, etc., and very little in succeeding.
Slashdot Mirror
I don't owe him an apology. I am a physicist, and my field of expertise is foundations of quantum mechanics. I did not say that this idea is insane as a knee-jerk reaction. I have read his paper, and I've been to a presentation by the men himself.
I find superdeterminism isane because it is too strong a hyphotesis: it can explain any experimental result. It is, therefore, incompatible with the very idea of doing science. Let me give you an example. Suppose that you want to measure some property of a particle, say, the polarization of a photon. In a superdeterministic theory, the photon's state could depend on which basis you chose to do the measurement, so you could conclude nothing at all from the result. Even the idea the a particle has some property independ of your choice of measurement does not apply.
And it's also deeply troubling how he arrived at superdeterminism. He wanted to do a deterministic model of quantum theory, but there is these pesky Bell inequalities tell you it's impossible. So what does he do? Simply assumes superdeterminism to brush off the issue.
It's not as if he was building a superdeterministic model per se, it is just an ad hoc hyphotesis that he added to get rid of Bell inequalities.
This kind of article bothers me immensely. It treats Einstein as the God of Science, and uses the fact the he worked on something as evidence that this idea is no crackpottery. Well, guess what, Einstein also shat, farted, pissed, had bad ideas, and even commited mathematical mistakes.
And one should never evaluate a scientific idea based on who's working on it. The Steady-State model of the universe is not a crackpot idea, simply because it is consistent with the laws of GR and (superficially) consistent with observational evidence. Philosophically, thought, it does seem quite silly, and I myself would never have regarded it as more than a mathematical curiosity, had it not been already falsified when I was born.
A more modern example would be 't Hooft's work on superdeterminisc models for quantum theory. The guy is obviously a genius, but this idea is pure insanity, and it saddens me to see people taking it seriously just because a Nobel prize is working on it.
That was a sarcastic answer to GP; I'm not one of these people, I think Sweden is a great country, and I'm fact gonna visit there in two weeks.
I'm aware that yours is the "correct" definition of socialism, but you should also taking into account that the definition is very much culture-dependent. People in Europe usually refer to Sweden as socialist, and I'm happy to go along, since the old form of socialsm doesn't exist anymore in Europe.
You should do some research on how Cuba was before the revolution. People don't revolt for nothing, you know? Although it's undeniable that today Cuba is quite poor (partly because of their own economic mismanagement, partly because of the US embargo), it is still in a better shape than it was under Fulgencio Batista. At least the people now have universal access to health and education.
Sweden also chose socialism. Look at the shithole it has become.
Come on, Minecraft does not pretend the implement classical physics.
The problem isn't lack of rigour, is that it gives you the wrong intuition. For example, the essential feature that distinguishes superposition from a classical mixture is that there is a basis in which the result of a measurement is deterministic, and from what I have seen this is not the case in the mod.
One way that it could be done: a "superposition" block, that can be prepared in the states |0>, |1>, |+>, or |->. If you look at it vertically, you measure in the Z basis, and if you continue measuring in the Z basis, the result doesn't change. If you then measure in the X basis, it's gonna collapse randomly to either |+> or |->, and it's gonna keep being this one as long as you keep looking at it vertically.
And this gives you also complementarity, which I think is one of the fundamental quantum concepts. The way the "observation" block was implemented simply makes no sense whatsoever. It's just a block that is different depending on the way you look. There is no complementarity between the aspects of the block, or anything to deal with the randomness of the quantum observation.
I don't understand what are you talking about teleportation. It doesn't sound to me very much related to real quantum teleportation. But I think the trouble with teleportation is that it is actually quite boring, and would serve no purpose in a game.
But it would be nice if we had some "quantum interaction" thingie (a CNOT) that would take two "superpositions" in different basis to an entangled state, and with this entangled state we could run the chsh game or ek91 qcrypto protocol.
Well, you could say that doing a PhD in physics is playing such a game, and, sure enough, most physicists have a working intuition on how quantum mechanics work -- more on the level of given a description of a situation, they can tell more or less what will happen without doing the calculatios. Doing that part is not hard at all. Heck, this kind of intuition mathematicians develop about the most abstract and artificial objects.
But on a more fundamental level, I don't think it is possible to develop a good intuition, because our brains process information classically; we need well defined bits to reason about. But it might not even make sense to be able to reason quantumly, as the essential feature would be preserving the superposition of the systems we interact with. But to preserve the superposition is to have no memory of the interaction, so... it would be very weird indeed.
Try to imagine what a quantum computer would feel as it processes quantum information: it can only apply transformations to its information blindly, without ever reading out what its input actually is, until the very end of the calculation -- and even this final measurement only because we, humans, want it, it's perfectly legitimate to never have a final measurement at all.
I am a physicist, but these are only drunken speculations...
At first I was quite excited with the idea that someone was able to use quantum mechanical elements in a game. But of course, they were not able to do this. They just created a mod vaguely inspired by quantum mechanics, that helps to perpetuate the myths so beloved by the lay media.
The video linked just shows a dude running around, nothing very interesting. If you search youtube a bit, you can find videos talking about the mechanics they implemented. I found this one, about the basic elements -- observation, superposition, and entanglement --, and this one, with the extremely exciting title "quantum computers and teleportation".
Of course, what they call observation and superposition have nothing to do with the quantum concepts, they are just blocks that are different depending on which direction you look at them, and the "entanglement" block is just a glorifed telephone. Their quantum computer doesn't seem to do anything besides teleportation, which is Star Trek teleportation instead of quantum teleportation.
Admitedly, these guys set out do to a terribly difficult task: quantum mechanics is a bit subtle, and quite far from games. The only ones I can remember off the top of my mind are the CHSH game, which is about as exciting as tic-tac-toe, and a quantum strategy to cheat at bridge, which requires you to do a nontrivial amount of maths (and is actually unpublished research =).
They have, nevertheless, failed. The mod looks cool as a game, though.
That's the kind of mentality that leads Firefox to shipping with a different interface every version.
This is completely beside the point (although true). The problem is that the voting system allows for such a thing to happen. Ever wondered why no European democracy has a two-party state? Well, they have sane voting laws. First past the post system without runnofs is just insane. Gerrymandering, electoral college, come on. Once the US was an inspiration to every democracy in the world. Now it has become a laughing matter.
As far as I know the Bristol folks do integrated optics, so it is in fact a chip; not as small as a classical chip, but about the size of a fingernail.
As a side note, it makes me sad that such an awesome project had to taint itself with the mention of Google and NASA waste of millions of dollars to D-Wave. What the Bristol people are doing is quantum computing. What D-Wave is doing isn't.
I don't think the analogy holds. Certainly knowing the details of quantum optics is quite labourious, but doing quantum programming is already some kind of quantum mechanics, that does not reference any particular physical system. Just like a Turing Machine uses Boolean algebra, but does not reference any particular machine to implement it.
Also, it's true that the algorithms are described in purely classical terms, but they probably sound like gibberish if you are not familiar with the concepts. As an example, let me describe you one of the simplest quantum algorithms known, the single-qubit version of Deutsch's algorithm, which decides whether a (classical) single-bit function f is such that f(0)=f(1) or f(0) != f(1):
Initialize your qubit in the state |0>
Apply the Hadamard gate to it
Apply the oracle that calculates f to it; it is described by the operator e^(i pi f(0)) |0>1|
Apply the Hadamard gate to it again
Measure in the computational basis.
I don't think it is particularly hard to understand it, though; if you want to do it seriously, there's a quite good book about the subject: it's called "Quantum Computation and Quantum Information", by Michael Nielsen and Isaac Chuang.
Can you elaborate on that? My laymen impression is that prosecutorial discretion allows you to at any time ignore any law you wish. But to ignore a specific law all the time seems a bit of an exaggeration.
So you'd rather trust a 89-year physicist with no training in climate science than the ones who have actually learned the subject? That's very enlightened of you.
I don't know what you mean by "very skeptical", but wikipedia says he does believe that the planet is warming and it is our fault; he seems to distrust the mathematical models we used to predict the climate.
For whatever it's worth, I am a physicist, and I have never met (personally) a physicist that is a climate change denier.
This sounds cool, but actually doesn't make any sense. The problem is not big endian versus litlle endian. Microsoft could very well use com.microsoft.xbox.360, but then somebody could squat com.xbox.360. Or microsoft could use com.xbox.360, and somebody would try to squat com.xbox360. The advantage of USENET is that its hierarchical structure was more or less well defined, while in the WWW it is completely arbitrary what you put before the .com part.
How the source adds nothing to the discussion? I've seen this on slashdot when the story first appeared, and the media coverage was uniformly believing CERN's hype, without a shred of scepticism. This blog has a simple, correct explanation of the physics involved, and its interpretation of the data pretty much agrees with I've been hearing at the University.
I was not expecting it to be like old stark trek. I was just hoping that this time the plot would make any sense. Clearly I was mistaken.
I did just that. Watched "the wrath of khan" for the first time this wednesday and then watched "into darkness" yesterday. Damm sure this made me dislike "into darkness" more. It's just a rehatch of the "emotional" scenes of "the wrath of khan" without any coherent plot to link them together. While "the wrath of khan" had a plot that made sense, and was thought provoking, this one had... nothing.
I wouldn't mind to see good visual effects and some action scenes in a Trek movie. But why can't they also use a decent script? It's not difficult to find somebody capable of writing a better story then the one that was used.
I don't mean to insult you, but you seem to be talking without having read TFP, which I did. It is here. (Incidentally, this paper just repeats the arguments first published here, but with a better experimental setup.)
I'm saying this because, as the paper says explicitly, the number 10,000 c has nothing to do with resolution of the clocks, but with their choice of preferred reference frame.
And the experiment I had in mind was quantum teleportation, where it is a little bit less insane talking about the "speed of quantum information".
So what you do is to send a quantum state from Alice to Bob. If you do the calculations, they show that quantum teleportation instantly transfers the state from Alice to Bob. But Bob is only capable of acessing the state after A also sends him some auxiliary classical information (two bits, to be precise). That's why I said that the only scientific thing that you can claim is that the information propagates at lightspeed, because before receiving also the classical information no measurement that Bob can do can say whether he received the state or not.
About the subjectivity, I know this is weird, but there dosn't seem to be any way around it. The state that the calculations show changing instantaneously is Alice's knowledge of the spin, which in fact does change as soon as she makes the measurement. But it makes no sense to say that this is also Bob's knowledge about the state of the particle, since he has no idea about what's going on.
This is bullshit. The scientific content behind this claim is that "nonlocal realistic models that reproduce the results of quantum mechanics must have speed of communication at least 10,000 faster than the speed of light in some arbitrary ference frame that we've chosen".
This means that this number is completely irrelevant, i.e., does not measure anyhting related to the real world.
What can be said, scientifically, about the speed of this channel is that it is the speed of light, because we can only actually measure the presence of the information on the other side after a light signal is sent from one party to the other.
The fact that it looks instantaneous is more of an artifact of our mathematical formalism, and a common philosophical misunderstanding about the nature of the quantum state (i.e., people regard it as objective rather than subjective).
Jackpot! This is the whole issue. Just observing these correlations means nothing, as you said, the same data comes out of our friends with the cards.
But the reason we are fascinated by entanglement is that there is more to it. We can actually prove that no theory in which the contents of the letters were predetermined (and FTL communication does not exist) can reproduce the correlations that can be produced via entangled states. This is the famous Bell's theorem.
Sorry, this is not the case. Again, I'm not an expert on this, but the best evidence I now that dark matter isn't just normal matter that happens to be dark is the Bullet Cluster. The thing is that, even if we can't see the normal matter, we know that it interacts with normal matter, well, normally. So in a collision of galactical clusters, we expect even this dark gas to interact a bit and get left behind, while the interactionless dark matter passes straight throught, and this is what we have observed in th Bullet Cluster.
If the effects we attribute to dark matter could be explained by known matter I don't think anybody would be excited about it =)
The sentences that could have given you warning were "An excess of antimatter within the cosmic ray flux was first observed around two decades ago.", i.e., the effect was already well-known, and "the AMS measurement can not yet rule out the alternative explanation that the positrons originate from pulsars distributed around the galactic plane. Supersymmetry theories also predict a cut-off at higher energies above the mass range of dark matter particles, and this has not yet been observed.", i.e., boring explanations were not ruled out, and the "smoking gun" evidence for dark matter was not yet found.
As usual, this is just a press release full of hype.
They didn't discover dark matter. They measured, with higher precision than ever, the excess in the positron fraction coming from cosmic rays. The existence of this effect, however, was already well-established. The question that was open, and still is, is which is the origin of this effect. One of the possible answers is dark matter. The problem with this answer is that we have to assume a discredited theory -- supersymmetry, and even within this theory a very artificial model of dark matter annihilation. The higher precision of the current measurements does not credence to this answer, nor does it discard more boring answers (i.e. coming from astrophysical processes that do not involve new physics). If you want to understand more about it, please read it from an actual particle physicist. I am a physicist, but not an astrophysicist nor a particle physicist.
Please keep in mind that I'm not criticising the AMS experiment itself: its job was to measure this excess with high precision, and this it did quite well. What I'm criticising is the people who have published this irresponsible press release.
Hmm, not in my definition of "few hundred". The calculation is actually easy to make:
The earth is about 1,5E11 m away from the Sun, let's say that 1% is the variation that we want, so we get it to 1,515E11 m. So the difference in energy that we need is GMm(1/R1-1/R2) \approx 5E31 J; quite a lot.
The best (or worst, depending on your point of view) nuke we ever exploded is the Tsar Bomba, which was 57 megatons or better 2,4E17 J.
So if we managed to use this energy with 100% efficiency (which we obviously can't) to move the Earth, we would need 10^14 nukes. Well, guess we're stuck here.
Have you ever succeeded in changing someone's beliefs in pseudoscience? Do you think that it is possible to do so in a large scale, to move humanity towards a more rational way of thinking?
Sorry for the down tone, but I have plenty of experience in failing to convince people of the falsehood in astrology, homeopathy, acupunture, etc., and very little in succeeding.