Plus the pesky old Flynn effect. At this point all of our children are above average, at least relative to 1900.
Average intelligence is more than sufficient to provide common sense, by the way. The difficulty isn't that the entire population doesn't consist of rocket scientists. The real problems are poverty, poor education leading to ignorance, trauma associated with things like poverty and military experience, mental illness in general, and above all religion. A person of average intelligence or even slightly lower than average intelligence as measured by IQ tests per se can still be plenty clever and sensible enough to come in out of the rain; if they have a weakness it is that they may lack the critical thinking skills to reject religious indoctrination and separate fantasy that they have been taught as fact from the real thing. They may also be more susceptible (on the same basis) to a cleverly constructed lie.
But there are plenty of people with I.Q. above 110 who have the exact same problems, or who have their own special set of problems. Sanity, as opposed to knowledge or problem solving capability, is distributed fairly sparsely across any population because our brains and perceptions do not work perfectly and because it is pretty much a truism that we are blind to our own failures. I am nominally pretty bright, myself, and teach physics and higher math to one of the most highly selected subgroups of human on the planet, and am confronted with my own failures in ratiocination, those of my also very bright colleagues, and those of the enormously bright students we teach on literally a daily basis.
This truly is a case where one should worry about the splinter in one's own eye (metaphorically) before worrying about that in everybody else's. Who knows what the real motivation is behind the idiocy posted above (or even what the real truth is, as the article doesn't reveal who had the energy to push for the ban or what their real motivation is). This is a country where a substantial number of people still think that high voltage power lines cause leukemia and that cell phones cause brain damage when you hold them next to your ear, and not just in North Carolina. It is a country where (according to at least some polls) roughly half the population still believe that the world is less than 10,000 years old and that Noah really did build a species-preserving ark:
Again, they don't all live in rurlal NC. Every state has its share of them, and if you pick the right community in almost any state or interview the right members of that community you can support the assertion that any state has small towns that are insane. There are members of congress that believe this tripe. Some of the readers of this reply on/. believe some fraction of tripe. Even as a theoretical physicist and hence an ultimate rationalist, I probably believe something that is tripe, or at least I almost certainly believe some things that are objectively false. A degree of humility is very definitely in order, even among the best and brightest of us, for the scale of our errors is determined as much by power and persuasiveness of the mind making the error as it is by the error itself, and the brightest of us often have the strongest opinions and can be quite persuasive of our errors.
There are solar farms all over, with more being built pretty steadily. Much of the state is well-suited for solar. The local electrical grid is supplied by Shearon-Harris, a nuclear plant about 15 miles away from where I type this. This is a nearly ideal mix -- solar reduces demand on the nuke when the sun shines, but there is plenty of capacity for days and times it doesn't.
Interestingly, it has achieved second place ranking even though electricity is cheaper than the national average at just over 0.09/kw-hr, where CA is upper-middle at just over $0.15/kw-hr. They have six more cents per kw-hr to use to amortize the cost of solar panels and regulators. I've looked at doing our house in NC, and for better or worse after I already replaced all of our furnaces and air conditioning units with super-high-efficiency units, installed low-E super-insulating glass windows, refinished the attic so that it acts as a thermal buffer in the summer with two layers of insulation (ceiling at R-40 and floor at R-20), and put in all CF and now gradually LED lighting -- there is just nothing left to use to amortize solar -- my monthly electrical bill is around $140/month for over 3000 hsf including summer air conditioning to a level of complete comfort, and I can barely break even on a 5 KW roof installation over 15 or so years.
However, I would prefer that you do judge NC by these rubes. Please! Stay away! You don't want to live in NC! These are not the droids you are looking for! California is definitely the state you want to move to. Or Florida -- think of all that sun! Or you can go work on civilizing Texas -- if enough people of sense move in, maybe they'll start to believe in things like evolution. And you really don't want to live in the fully integrated communities of Durham, and have to worry about deer eating your Hastas inside the city limits and have to worry about your kids drowning in the area lakes or getting lost in the local museums or galleries. And who really wants to see shows in venues like DPAC or the Carolina Theater, or experience world-class college basketball or even (gasp) football or soccer firsthand? See? I hear that Northern Virginia is very nice, and the suburbs of Detroit can probably hold a few more souls.
Look deeply into my eyes. North Carolina is a b-a-a-a-a-a-d place to live. You do not want to come here, even for a visit. You just want to buy our craft brews and cigarettes and feel superior right where you are... You may now wake up, and will not remember anything about this, but if anyone mentions the opportunities and advantages of living in North Carolina -- one of which is its relatively low population density and the charm of its small and heavily forested cities -- you will experience a feeling of panic that will only go away if you think about the joy of moving to rural Texas or Chicago.
I agree that this is controversial, but it is by no means a slam dunk that it isn't necessary. And if it isn't, it makes it more likely that a moon-forming atmosphere stripping collision is needed (or else bombardment by comets or whatever you think is needed for an ocean). And then there are the issues associated with radiation. The point is that there may be some subtleties associated with the requirements for life that mean that high radiation zones (sustained) are indeed inimical to life. Or (sure) not. It will likely be a decade or three before we even have a good enough theory (backed by at least SOME observations) to resolve these questions.
Building stargates aside, the problem with radiation is tied to the problem with magnetic moments. The radiation from the local sun alone is enough to make life anywhere from pretty unlikely at all to pretty unlikely to evolve into a rich ecology if the planet in question hasn't got enough of a magnetic field to provide a radiation shield from both cosmic rays and the solar wind. Without it, atmosphere is just blown away by the solar wind and cosmic rays reach the surface in abundance. The solar magnetic field may also be important, or rather the coupled geosolar field may be key. Oceans won't last without a stable atmosphere for the billions of years (apparently) necessary to get "interesting" life forms.
Another issue that I think is very much up in the air is how important the moon is to evolution. We have a very limited sample in our own solar system, but there are three planets that "should" be in a generically habitable zone. Earth and Venus are almost the same size, but Venus has no moon. As a consequence (?) it has an incredibly dense atmosphere (where I'm speculating that the collision that resulted in the moon blew off a substantial fraction of the Earth's early atmosphere and altered the composition of that which remains). Mars is perhaps too small to bind an atmosphere of the right composition, but in any event it lacks enough of a magnetic field to keep the solar wind from stripping off the atmosphere it has/had. It may have life. It may have HAD life. But it is very unlikely to be a place where life can develop into a complex ecology, let along intelligence.
It could well be that intelligent life requires several more "ands" of additional conditions: a planet with the "right" magnetic field in the habitable zone for the star in question, that has to have the "right" magnetic field as well, with at least one largish moon close enough to churn up its atmosphere (or worse, a moon that "has" to be formed by a collision between two proto-planets that strips the crust/atmosphere of the larger survivor), with the "right" mix of leftover heavy elements from the previous supernovae that created the stardust from which the whole system condensed to keep the core molten and magnetized, with the "right" mix of lighter crustal material to provide the needed carbon, oxygen, nitrogen, and hydrogen, especially oceans of water and at least eventually an atmosphere with a substantial proportion of oxygen.
But there are almost certainly more than Avogadro's number of planets out there, which is a lot of times to roll the dice for what is likely a fairly finite set of conditions. There are almost certainly billions of planets in just the visible part of the Universe alone, tens or hundreds or thousands per galaxy, that have all or most of the conditions needed for basic "life", and some fraction of them the conditions needed for the eventual evolution of intelligent life. However, we have no reliable way of computing the probability of life given those conditions, or intelligence given life. If either one requires the moon-forming collision, life could be quite rare as that probably doesn't happen (given all the other conditions) very often.
Except that the consideration of uniformity at and up to the boundaries of sight around 13 billion LY away implies that the visible cosmos is only a small fraction of the whole thing, which is at least 100-200 times larger (length scale) that the visible portion. Which means that you are underestimating everything by at LEAST a millionfold. To quote Adams:
Space is big. Really big. You just won't believe how vastly, hugely, mind-bogglingly big it is. I mean, you may think it's a long way down the road to the chemist, but that's just peanuts to space.
So let's shoot for order of 100 quadrillion galaxies in the material Universe, each with order of 100 billion stars, with order of 10 to the twenty-seventh planets. Yes, most of that is out beyond the surface horizon of the Big Bang and nearly all of it is as fundamentally disconnected from us by time and space as if it didn't exist at all (and to a logical positivist, perhaps it does not, showing the idiocy of the position that meaning derives from empirical testability) but the point is that your estimate of 4 times 10 to the twenty second is very conservative, and the actual number is likely many orders of magnitude larger. And even the estimate of 100 times larger in all directions is a lower bound -- there is no defensible upper bound AFAIK so far, and there may not be an upper bound if space is actually infinite.
I think you are mistaken. I think they cannot force you to cough up the password without a warrant or court order. I think this because I once gave an invited talk on this very subject to a law enforcement symposium, and an FBI agent supported by his pals in both the feeb and regular law enforcement told me this. Now, I'll bet he is a more authoritative voice than you are. If anything, I imagine it has gotten easier to force a password since then by legal coercion.
To cite a crude analogy, if the DEA shows up at your doorstep and says "Please sir, we would like to come in and see if you have a meth lab in your basement." you have a perfect right to say now, constitutionally guaranteed. If they show up at the same doorstep and meet you outside and show you a warrant and ask you politely to unlock the door to let them in, the fact that you do have a meth lab in your basement and will "incriminate yourself" by opening the door will not save you in court. If your house were unopenable (those six foot thick concrete walls) and could only easily/cheaply be opened by means of a combination lock, you bet your sweet bippy that the court could order you to divulge the password or rot eternally in jail for contempt.
It's the warrantless search that is at issue here. They want to be able to search your house with or without your presence or knowledge, with or without a key, and with or without a warrant being presented to you or any of your legal representatives. We can only wonder about whether or not a warrant will be required at all, and we don't HAVE to wonder that if all locks have to be openable with some sort of universal combination or key, it won't be long before our house can be entered at will by individuals who aren't even nominally members of the government or any official agency, who might USE our basement to make meth in, without our knowledge.
... force you to decrypt any encrypted document with a court order. In fact, the law is so broad that if you go into court and the judge says "please give these nice officers the encryption keys for your hard drive" and you say "no", they can say "OK, I'll just put you in jail for contempt of court, without bail, until you do." Which can literally be forever. There are no limits that I know of for jail time for contempt of court for an ongoing refusal to comply with a court order. So it can literally be life not even in prison, in JAIL, until you do.
If th issue is terrorism, the powers are even broader and can involve you being sent to a concentration ca -- I mean "federal jail on a remote island" until you cough up a lot more than just the keys.
What they want is the power to read dynamical communication streams in real-time, because decrypting them is often too difficult even for the NSA and because a lot of them are encrypted with one-time or digitally saved keys so that a user CAN'T just cough them up. If my ssh private keys went away, do you think I have them memorized? NOBODY could decrypt my old network traffic, not even me!
Now we just have to wait a bit for the legislative branch to realize that a) we lack the theorems needed to make their nifty idea work; and b) any end user can trivially work around it by simply exchanging keys for one of the known secure algorithms; c) it isn't necessary for any saved, recorded data; and d) it isn't constitutional. It's exactly like trying to pass legislature that would require all house keys to be "registered" and constructed in such a way that a master key in the possession of the police would open them. Good luck with that one.
Besides, they already can. The key is called a "brute force", and if they use it, yeah, they have to go up against the effort the householder put in to stopping brute force entry. If their "house" is a repurposed bomb shelter with six foot thick concrete walls, good luck to them.
I suspect that it is every bit as much used for its high thermal conductivity. Sure, less strain on the motor and if there are lots and lots of platters that could be an issue. But it also helps to keep the whole thing cool.
Sure, bit it isn't just "repeat itself". It is "show incorrect correlations within" across all possible separations. Where "incorrect" is a pretty delicate mathematical specification and in many cases virtually impossible to resolve empirically. For all of the "good" PRNGs, it IS more or less impossible to resolve their non-randomness empirically, even though it is known that somewhere out there, they aren't.
I see. So alcohol is related to the issue of whether or not 3 colors in quantum chromodynamics is enough to be considered "infinite"? Or are you trying to say that alcohol is normal. If the latter, I'm willing to believe it, because a D1 (as I understand the "n" in Dn) is a one sided die, which is sort of like flipping a mobius strip as a one-sided coin, and if I try to visualize that I'm gonna need a drink to soften the headache. But the resulting distribution isn't very bi. Or normal.
I don't know why you say "you crazy physicists" you crazy applied mathematician, you...;-)
Besides, I agree with everything you say. A random number generator is an oxymoron, which is why the "P", right? Pseudo random number generators. I'm not even convinced that quantum processes are "truly random" unless and until they are subjected to far more rigorous tests than they have been so far, and besides, if one believes quantum theory itself one has to believe the Nakajima-Zwanzig construction and approach to the generalized master equation, where in it is revealed that in quantum mechanics "random" just means "in an incompletely specified state". Which makes perfect sense, as quantum theory is no more random than classical theory -- arguably LESS random in some ways.
However, it is possible to subject any proposed "random number generator" -- or at least any (hypothetically) random uniform N-bit integer generator to a hypothesis test, with the null hypothesis "this is a perfect random number generator". One then can generate any statistic you like that should have some known/computable distribution and that has an equally computable cumulative distribution function, invert the result, and convert it into a p-value. One can repeat this process indefinitely to see if the distribution of p-values one obtains is itself a uniform distribution from 0 to 1 (p itself should be a uniform deviate for a perfect RNG for any hypothesis test of this sort).
Then one is studying the hypothesis the way physicists and maybe even some applied mathematicians study nature. You can never prove a (null) hypothesis true, but you can accumulate sufficient evidence to be pretty sure it is false, with patience. Or not. Failing to falsify the null hypothesis doesn't prove it true, it proves that we can't prove it false yet.
We have a pretty good pile of PRNGs that we cannot prove to be non-random by reasonable empirical tests, even ones that consume rather large numbers of random numbers from the generator. Some of them have computable periods, so we know they aren't "really random" a priori, but they might well be indistinguishable from random for the entire length of the period, or not. Since we can't come close to running the period of any of the long period generators (and since it is easy to extend the period if we should ever get close) we may never find out.
But yeah, absolutely. Unpredictable is good. A "good" PRNG is more than unpredictable, though. Unpredictable is pretty easy. A good PRNG has no discernible correlations across very long intervals, where "discernible" involves using subtle tests at both very small and very large scales. It is >>uniform, in just the right way. And in the end, we can't find a problem where we can compute the answer using theory and compute the same answer via some sort of Markov chain or Monte Carlo process based on supposedly random numbers where the latter gives us a significantly wrong answer.
I think the number of rolls were totally adequate... should be followed by "to resolve deviations from perfect dice as large as X with a p-value for the null hypothesis less than Y" (where ideally, Y isn't something moronic like "0.05"). Maybe something like 0.000001.
And damn skippy, you really do want to sample multiple runs of multiple dice from multiple sources per manufacturer, because one thing you are testing is the quality of their production line, the FREQUENCY with which their dice turn out to be worse than X per p Y given N samples in a run.
Amen, brother. Or rather, it might be enough. Depends on how different P(i) is from 1/N for the ith side of N-sided dice.
To put it another way, to get to a p-value on the null-hypothesis "this is a perfect die" that I'd be willing to accept as "certain" evidence that the die is bad is determined by the binomial distribution of the actual distribution of P(i) compared to the "ideal" distribution P(i) = 1/N. For some distributions ten rolls is plenty (if you get the same number 10x in a row on a 20 sided die, 20^10 is a pretty big, pretty unlikely number...;-)
OTOH a PERFECT 20 sided die isn't going to equally populate 20 bins filled with 3000 rolls, 150 each. At least, it will happen so rarely that I'd be suspicious of THAT in turn.
Absolutely dead on the money. I can't mod you up because I commented myself on the thread, but as the primary author of dieharder, I couldn't have said it any better. Detecting non-randomness in crypto-PRNGs implicitly comes with a door prize -- being able to become fabulously wealthy overnight, either semi-legitimately (letting the world's financial system bribe you with a few billion a year NOT to reveal how to crack internet security, for example) or by just cracking internet security and stealing lots of money or by selling the tools to the NSA (risky, as they might "pay" you by incarcerating you for life or terminating your existence as a national security risk once they had the tool in hand).
But there are a number of NON-crypto-grade PRNGS are both fast and good enough that it is essentially impossible to detect a lack of randomness, certainly not unless you process a few terabytes of random numbers output by them. And even stock Unix PRNGs are good enough for nearly all games. The days of using really, really bad RNGs in distribution libraries is mostly over, even though one does not to use a bit of common sense about it. Dieharder lets you test the speed of the GSL implementation of a small mountain of generators including several system generators and get a feel for optimizing the speed/randomness question.
I'm very curious. Why would you say that pseudorandom number generators are't fit for "competition use"? Specifically, I'd be happy to provide you with a small stack of generators that I cannot make fail any of a rather demanding set of tests in dieharder. I'd also be very interested in any test you have that they would consistently, and correctly, fail, or that would fail any set of them but not fail a "true" RNG if one could find such a thing.
This isn't just the sort of thing you can just state as if it is true. You need to have some research or counterexamples showing it to be true. An AES-based generator isn't going to fail any simple/trivial test that I know of. In fact, it won't fail any DIFFICULT test that I know of, certainly no test I've managed to implement so far. A Mersenne-Twister in principle can be made to fail, but in practice it is not simple to demonstrate any particular failure. Any of the crypto generators are almost be definition difficult to fail. Marsaglia's KISS-class generators are very difficult to fail. Permuting a number of comparatively poor generators -- e.g. linear congruential generators with known hyperplanar correlations -- makes them much more difficult to fail, as does XOR-ing generators with different kinds of weaknesses. Some of these things do increase the time required to compute them, but there are a number of them that are both fast and still damn good.
So what evidence do you have that ALL PRNGs -- for you haven't bothered to exempt any PRNGs including crypto-class and e.g. BBS -- aren't fit for competition use, and what exactly does that even mean? If you mean fit for use in gaming, you are insanely incorrect, as nearly any of the BUILT IN generators in modern Unices are fit for gaming, which is not a particularly demanding application set and requires only uniformity and a lack of any compelling short range correlations to be sufficient for most purposes. Is there some other "randomness competition" other than passing suites of tests of randomness or the ultimately practical test, being able to crack internet security and destroying the world of e-commerce overnight? Because I don't think this is correct.
I remember, back when I was taking a quick basic class (LOL!), that I noticed an anomaly in how the random number generator produces numbers.
Specifically (well, as specifically as I can recall without digging out old source code) I noted that the output of the RNG favored multiples of 4, after having some "Difficulties" with random numbers not being random enough in one of my programs.
I wanted to test that notion, so I created a small program that "should" have painted the screen with random colored fuzz, using a random walk. (EG, the X coord, the Y coord, and the color value are all based on "random" picks from the RNG)
Imagine how much I laughed when I saw a diagonal banding pattern appear instead of random fuzz, out of the random walk.
computed random sources: Random enough for some applications, but caveat emptor!
Well, you could have read Knuth's Art of Computer Programming, vol 2 and learned a bit about the many possible flaws one can find in pseudo-RNGs, where clustering on hyperplanes at some dimensionality is one of the simplest. Or, you could use e.g. diehard or dieharder to test the RNG a bit more scientifically than with outputting a 2d screen of fuzz (which will work, but only for a really truly terrible generator). Nowadays, though, there are many very, very good pseudo-random number generators out there. Mersenne Twisters, for example, have correlations only in some gawdawful high dimensionality (over 600) so you Will Not See Bands in 2D fuzz. AES-based routines, or threefish based routines, will give you cryptographic quality RNGs (except for the bit about choosing the seeds, if you repurpose them as PRNGs). Variations of Marsaglia's KISS generator are Very Good Indeed.
You can get some of these in the Gnu Scientific Library or built into R. Any of these would be random enough for almost any purpose requiring random numbers, since so far it is pretty damn difficult to come up with a test for randomness that they fail, and any purpose for which they fail would in principle be a good test. If you find one, let me know and I'll stick it into dieharder.
I find a lot of people who are agnostic or atheist have actually made science their religion. Most aren't even practicing scientists, and instead of looking to the scientific method to teach them new ideas, they have "faith" in theories despite science not yet having proven or disproven them. They use science as their religion not to further science, but to attack religion. Your comments are pretty close to putting you in this bucket.
Excuse number #912 -- "atheism is a religion too". Or better yet #912.A -- "science is a religion". Please. If you know anything at all about science, its purpose, and how it works, you know that it is not a religion. It is a way of figuring out what it is best to believe about the real world in a systematic and improvable way. Note well the two essential components -- "about the real world", and "in a systematic and improvable way". It addresses the real world, not a fantasy world, and the standard for truth is thus this objective world itself, not what people have said about it or believe about it or wrote about it in an ancient book long before we had anything vaguely approaching a science. But the second part is just as important. If I make a claim about some systematic organization supposed to hold in th real world, it is possible to accumulate evidence that supports the claim, refutes the claim, or is neutral towards the claim. Over time, more evidence and better methods of looking generally result in claims that we believe very, very strongly to be either true or very close to true, claims that we believe not to be true, and claims that cannot be decided by the evidence at hand. In all cases the standard of truth is correspondence of the assertion with reality itself, not with argumentation about reality, although the reasoning process is Bayesian and hence one isn't building up evidence-supported beliefs in isolation.
"Science" is not a religion, it is the set of interlocking assertions that have the strongest, mutually supporting evidentiary support. It is literally what it is best to believe about the real world according to an actual standard. It does not assert perfect truth, it asserts probable truth, provisional probable truth at that. If you want to actually learn something about the reasoning process involved, I would recommend E. T. Jaynes' "Probability Theory, the Logic of Science". You might also want to peruse Richard Cox's monograph "The Algebra of Probable Inference". The difference between a religious text and these two works is so profound, so obvious, so glaring, that perhaps you will reconsider your rash statement that science is a religion. These books establish, via a minimal set of axioms, a direct connection between evidence and networks of probable beliefs -- they provided mathematical support and a proof of sorts that it is better to believe things given evidence than to believe any random notion that is asserted by anyone, anywhere, for whatever reason that is not supported by evidence, that contradicts beliefs that are supported by strong evidence, or that is contradicted by the evidence itself directly.
I would offer examples -- but is there really any point? There are a near infinity of possible religions. There are quite a few actual religions, religions that contradict one another on numerous points, and the number swells to a really large number if one allows (as one should) all of the religions ever believed by any vagrant tribe throughout history, and all of the named variations on religions loosely shared between tribes.
For starters, probability theory would dictate that even if one knew that precisely one of these variants was precisely true, it is rather improbable that your particular beliefs out of this set of possibilities is correct. It isn't even probable that your beliefs come from a major family that could be correct. The odds are against any given religion being correct before you examine evidence. Without evidence
Seriously? The moon is gravitationally locked now, sure, but the Earth (and moon) still being liquid/hot when it slowed to a lock? I don't think so. For this to be plausible the moon would have had to coalesce, in an orbit, with nearly zero spin angular momentum, which seems absurdly unlikely. Otherwise, like a bird on a rotisserie, it would have been "roasted" pretty much equally on both sides. So maybe, but I doubt it.
The census in prose with the lovely insertions of slaughtering the Midianite captives including women and children EXCEPT for the young virginal girls who were given to Moses' soldiers to -- wait for it -- rape and enslave, followed by a half page of tallying up the loot of a genocidally slaughtered civilization and recording how much of it Moses, the priests, and the war captains got. If one rewrote it and published it as an account of what the ISIS "caliph" did with, say, Beirut or Damascus, the world would be shocked and cry out everything from genocide to infanticide, but because, well, it's MOSES, well then it is OK. Even Jesus loved Moses, and entertained the man himself during his transfiguration.
Come to think of it, a lot of what ISIS has done does fit right in with Old Testament reports of God's Own People.
This is the hardcore Salafi doctrine by which people like Daeesh operate by and it's theologically valid, though batshit insane.
HOWEVER, only about 70-80% of Muslims are Sunnis and of those maybe 10-15% are Salafis.
Oooo, let's do the arithmetic. There are roughly 1.6 billion Muslims (sources: multiple, although many would go as high as 1.9 billion and rising). Let's be conservative and go with 70% of the smaller number -- 0.7 x 1.6 = 1.12 billion are Sunnis. Of these (again, being conservative) 10% are Salafis. That means 112 million Muslims are Salafis! Or as many as 200 million, of one uses the higher end of that projection. Of the Salafis, roughly 10 million are estimated to be Salafi Jihadists. So sure, less than one percent, but the original numbers are so large that the actual numbers are still huge, if you view them as a potential army and recruiting ground for violent Jihad.
So what you are saying is that the Daeesh are a huge bunch of psychopathic assholes led by an eloquent sociopath from Baghdad, because while it is true that all Salafis, no matter how batshit insane the theology to which they subscribe, are not Daeesh Jihadists, they are at the very least a fertile ground for recruitment and have the potential to become a bandwagon several hundred million strong if ISIS appears to be succeeding, and it also strongly suggests that they aren't all foreign psychopathic assholes. There is a rather large pool of the domestic home-grown type.
All of which was true for Al-Queda as well, as it isn't just Salafi Jihadists. Jihad is comparatively ecumenical in Islam, and while it is not necessarily violent, it can easily become violent, often quite rapidly. There is a large disaffected population of young, volatile Muslims worldwide. In some, but not all, non-Muslim countries they feel highly marginalized. In many Muslim countries the governments are so corrupt that even though Muslims per se aren't marginalized, life is worse than it is under the foreign non-Muslim governments, marginalization or not. Then there is the antique Sunni vs Shia problem -- a division that perpetuates Muslim on Muslim violence and weakens Islam immeasurably as its adherents perpetually discover that the enemy of my enemy is a better friend, Muslim or not. Far too much of Islam is tribal culture still living in the 8th or 9th century, but dreaming of the 12th and 13th centuries when it was a, perhaps the, dominant world power. Young Muslims see this division and are easy meat for any charismatic religious leader who promises to end the rift and restore Islam to its "rightful" status as a, no, as the dominant world culture.
At the end of the day, though, there is the Quran. And the Quran is a stupendously violent document. It was a revolutionary document back when it was written, designed to generate an us vs them mentality and create a social and religious identity to support the establishment of a violently conquered tribal empire. It succeeded. Its memes are strong. This makes the followers of any sort of literalist interpretation of the Quran enormously dangerous to non-believers, because it quite literally makes converting or killing or enslaving them a holy duty, especially if they in any way interfere with or refuse to live under the religious laws of Islam, believer or not. This isn't really arguable -- one can just read the damn thing and see for yourself (but as usual with holy scriptures, nobody does, they just prefer to quote what somebody else says about them in a pious way). Being a liberal Muslim is even more difficult than it is to be a liberal Jew or Christian, and that really isn't easy if one reads, say Numbers 31 in the OT to see how God really feels about the Midianites and non-Jews in general, or reads Leviticus to see how he really feels about slavery and women. In all cases one has to sort of elide all of the passages that don't agree with what you want to believe is The Good, an
Sigh. I pitty me, since you are obviously clueless about more things than I can easily fix.
Gravity can add energy to things with "100% efficiency", but only after you've done the work of raising them up in a gravitational field. Pretty much the same thing is true for electrostatic energy. Magnetic fields do no work (seriously, and don't argue with me as I'm teaching electrodynamics at this very moment and You Will Be Wrong of you say otherwise and I will cheerfully prove that:
as an identity (d\vec{l}/dt = \vec{v}, and A dot (A times B) is zero). I suppose one can do no work with 100% efficiency, but your entire comment is three lines of pointless.
Now, I could wax poetical about the second law of thermodynamics and the presence of irreversible losses in even simple things like dropping something macroscopic in an actual gravitational field that cause some energy to be diverted to heat, causing a loss in thermodynamic efficiency, which I actually can define as well, but you are quite right. I was referring to the horrendous inefficiency of adding somewhere between 32 and 64 MJ/kg to an object lifting it with rockets. You know why? Because (and follow this argument carefully, because it is pretty important):
That's the only way we have to lift an object to Earth orbit, be it low or geosynchronous.
Sure, there are fantasies about alternatives. For example, a great Heinleinian favorite is the linear accelerator (and yes, I've read all of Heinlein's works, many of them too many times to count). Now, go out there and actually cost one out, and figure out how you are going to solive the many, many engineering problems associated with accelerating an object to order of 11.2 km/sec plus whatever you need to punch through the atmosphere, without it burning to a crisp and without thermodynamic losses at every step along the way -- from the power plant that makes the electricity to the wires that carry the power to the currently imaginary rings aligned along some currently imaginary mountain ridge that will fire an undesigned and improbable payload container along a track some hundreds of unflattened kilometers long (at least if you want to contemplate firing humans inside).
Then there is everybody's SciFi favorite, the Space Elevator. In this fantasy, somebody puts something nice and massive in geosync orbit above a land point on the equator. Then they imagine -- and I do mean imagine -- building this really, really long cable made out of Imaginarium, an imaginary material that is so strong that it will not snap under the stress of its own weight when dangling from 5 earth Radii out back to the surface, and that can (what the heck, imagination is free!) carry a payload as well! And then, it is flexible enough to be run in a continuous belt around a spindle in the asteroid, under tension, without doing bad things to the asteroid's "orbit", which would not any longer be at 5R_e because, well, you are pulling down on it with the tension in the belt.
Now the idea is to turn a crank and lift things up the "elevator" to not-exactly orbit! As the crank turns, one "car" would ascend while the other descends, so you'd only have to lift the "weight" of the payload, at least if we ignore friction and the like in the sprockets under enough tension to lift the weight of 20 to 30 thousand miles of cable plus the payload.
Now the owners of this particular fantasy do tend to leave out a handful of, um, "obstacles" to their pretty little scheme. For example, the fact that building it would cost a few hundred trillion dollars, the fact that it isn't even clear that materials that could support their own weight from the pseudorbit to the ground could even theoretically exist, given that the strength of molecular bonds is capped at something like eV, the molecules themselves have a minimum mass and volume, and I suspect (but haven't proven) that a simple scaling
Slashdot Mirror
Plus the pesky old Flynn effect. At this point all of our children are above average, at least relative to 1900.
Average intelligence is more than sufficient to provide common sense, by the way. The difficulty isn't that the entire population doesn't consist of rocket scientists. The real problems are poverty, poor education leading to ignorance, trauma associated with things like poverty and military experience, mental illness in general, and above all religion. A person of average intelligence or even slightly lower than average intelligence as measured by IQ tests per se can still be plenty clever and sensible enough to come in out of the rain; if they have a weakness it is that they may lack the critical thinking skills to reject religious indoctrination and separate fantasy that they have been taught as fact from the real thing. They may also be more susceptible (on the same basis) to a cleverly constructed lie.
But there are plenty of people with I.Q. above 110 who have the exact same problems, or who have their own special set of problems. Sanity, as opposed to knowledge or problem solving capability, is distributed fairly sparsely across any population because our brains and perceptions do not work perfectly and because it is pretty much a truism that we are blind to our own failures. I am nominally pretty bright, myself, and teach physics and higher math to one of the most highly selected subgroups of human on the planet, and am confronted with my own failures in ratiocination, those of my also very bright colleagues, and those of the enormously bright students we teach on literally a daily basis.
This truly is a case where one should worry about the splinter in one's own eye (metaphorically) before worrying about that in everybody else's. Who knows what the real motivation is behind the idiocy posted above (or even what the real truth is, as the article doesn't reveal who had the energy to push for the ban or what their real motivation is). This is a country where a substantial number of people still think that high voltage power lines cause leukemia and that cell phones cause brain damage when you hold them next to your ear, and not just in North Carolina. It is a country where (according to at least some polls) roughly half the population still believe that the world is less than 10,000 years old and that Noah really did build a species-preserving ark:
http://blogs.telegraph.co.uk/n...
Again, they don't all live in rurlal NC. Every state has its share of them, and if you pick the right community in almost any state or interview the right members of that community you can support the assertion that any state has small towns that are insane. There are members of congress that believe this tripe. Some of the readers of this reply on /. believe some fraction of tripe. Even as a theoretical physicist and hence an ultimate rationalist, I probably believe something that is tripe, or at least I almost certainly believe some things that are objectively false. A degree of humility is very definitely in order, even among the best and brightest of us, for the scale of our errors is determined as much by power and persuasiveness of the mind making the error as it is by the error itself, and the brightest of us often have the strongest opinions and can be quite persuasive of our errors.
rgb
More to the point, we are ranked second in the country as far as installing solar power is concerned anyway, behind only California:
http://www.seia.org/research-r...
There are solar farms all over, with more being built pretty steadily. Much of the state is well-suited for solar. The local electrical grid is supplied by Shearon-Harris, a nuclear plant about 15 miles away from where I type this. This is a nearly ideal mix -- solar reduces demand on the nuke when the sun shines, but there is plenty of capacity for days and times it doesn't.
Interestingly, it has achieved second place ranking even though electricity is cheaper than the national average at just over 0.09/kw-hr, where CA is upper-middle at just over $0.15/kw-hr. They have six more cents per kw-hr to use to amortize the cost of solar panels and regulators. I've looked at doing our house in NC, and for better or worse after I already replaced all of our furnaces and air conditioning units with super-high-efficiency units, installed low-E super-insulating glass windows, refinished the attic so that it acts as a thermal buffer in the summer with two layers of insulation (ceiling at R-40 and floor at R-20), and put in all CF and now gradually LED lighting -- there is just nothing left to use to amortize solar -- my monthly electrical bill is around $140/month for over 3000 hsf including summer air conditioning to a level of complete comfort, and I can barely break even on a 5 KW roof installation over 15 or so years.
However, I would prefer that you do judge NC by these rubes. Please! Stay away! You don't want to live in NC! These are not the droids you are looking for! California is definitely the state you want to move to. Or Florida -- think of all that sun! Or you can go work on civilizing Texas -- if enough people of sense move in, maybe they'll start to believe in things like evolution. And you really don't want to live in the fully integrated communities of Durham, and have to worry about deer eating your Hastas inside the city limits and have to worry about your kids drowning in the area lakes or getting lost in the local museums or galleries. And who really wants to see shows in venues like DPAC or the Carolina Theater, or experience world-class college basketball or even (gasp) football or soccer firsthand? See? I hear that Northern Virginia is very nice, and the suburbs of Detroit can probably hold a few more souls.
Look deeply into my eyes. North Carolina is a b-a-a-a-a-a-d place to live. You do not want to come here, even for a visit. You just want to buy our craft brews and cigarettes and feel superior right where you are... You may now wake up, and will not remember anything about this, but if anyone mentions the opportunities and advantages of living in North Carolina -- one of which is its relatively low population density and the charm of its small and heavily forested cities -- you will experience a feeling of panic that will only go away if you think about the joy of moving to rural Texas or Chicago.
rgb
I agree that this is controversial, but it is by no means a slam dunk that it isn't necessary. And if it isn't, it makes it more likely that a moon-forming atmosphere stripping collision is needed (or else bombardment by comets or whatever you think is needed for an ocean). And then there are the issues associated with radiation. The point is that there may be some subtleties associated with the requirements for life that mean that high radiation zones (sustained) are indeed inimical to life. Or (sure) not. It will likely be a decade or three before we even have a good enough theory (backed by at least SOME observations) to resolve these questions.
Building stargates aside, the problem with radiation is tied to the problem with magnetic moments. The radiation from the local sun alone is enough to make life anywhere from pretty unlikely at all to pretty unlikely to evolve into a rich ecology if the planet in question hasn't got enough of a magnetic field to provide a radiation shield from both cosmic rays and the solar wind. Without it, atmosphere is just blown away by the solar wind and cosmic rays reach the surface in abundance. The solar magnetic field may also be important, or rather the coupled geosolar field may be key. Oceans won't last without a stable atmosphere for the billions of years (apparently) necessary to get "interesting" life forms.
Another issue that I think is very much up in the air is how important the moon is to evolution. We have a very limited sample in our own solar system, but there are three planets that "should" be in a generically habitable zone. Earth and Venus are almost the same size, but Venus has no moon. As a consequence (?) it has an incredibly dense atmosphere (where I'm speculating that the collision that resulted in the moon blew off a substantial fraction of the Earth's early atmosphere and altered the composition of that which remains). Mars is perhaps too small to bind an atmosphere of the right composition, but in any event it lacks enough of a magnetic field to keep the solar wind from stripping off the atmosphere it has/had. It may have life. It may have HAD life. But it is very unlikely to be a place where life can develop into a complex ecology, let along intelligence.
It could well be that intelligent life requires several more "ands" of additional conditions: a planet with the "right" magnetic field in the habitable zone for the star in question, that has to have the "right" magnetic field as well, with at least one largish moon close enough to churn up its atmosphere (or worse, a moon that "has" to be formed by a collision between two proto-planets that strips the crust/atmosphere of the larger survivor), with the "right" mix of leftover heavy elements from the previous supernovae that created the stardust from which the whole system condensed to keep the core molten and magnetized, with the "right" mix of lighter crustal material to provide the needed carbon, oxygen, nitrogen, and hydrogen, especially oceans of water and at least eventually an atmosphere with a substantial proportion of oxygen.
But there are almost certainly more than Avogadro's number of planets out there, which is a lot of times to roll the dice for what is likely a fairly finite set of conditions. There are almost certainly billions of planets in just the visible part of the Universe alone, tens or hundreds or thousands per galaxy, that have all or most of the conditions needed for basic "life", and some fraction of them the conditions needed for the eventual evolution of intelligent life. However, we have no reliable way of computing the probability of life given those conditions, or intelligence given life. If either one requires the moon-forming collision, life could be quite rare as that probably doesn't happen (given all the other conditions) very often.
rgb
Except that the consideration of uniformity at and up to the boundaries of sight around 13 billion LY away implies that the visible cosmos is only a small fraction of the whole thing, which is at least 100-200 times larger (length scale) that the visible portion. Which means that you are underestimating everything by at LEAST a millionfold. To quote Adams:
Space is big. Really big. You just won't believe how vastly, hugely, mind-bogglingly big it is. I mean, you may think it's a long way down the road to the chemist, but that's just peanuts to space.
So let's shoot for order of 100 quadrillion galaxies in the material Universe, each with order of 100 billion stars, with order of 10 to the twenty-seventh planets. Yes, most of that is out beyond the surface horizon of the Big Bang and nearly all of it is as fundamentally disconnected from us by time and space as if it didn't exist at all (and to a logical positivist, perhaps it does not, showing the idiocy of the position that meaning derives from empirical testability) but the point is that your estimate of 4 times 10 to the twenty second is very conservative, and the actual number is likely many orders of magnitude larger. And even the estimate of 100 times larger in all directions is a lower bound -- there is no defensible upper bound AFAIK so far, and there may not be an upper bound if space is actually infinite.
rgb
I think you are mistaken. I think they cannot force you to cough up the password without a warrant or court order. I think this because I once gave an invited talk on this very subject to a law enforcement symposium, and an FBI agent supported by his pals in both the feeb and regular law enforcement told me this. Now, I'll bet he is a more authoritative voice than you are. If anything, I imagine it has gotten easier to force a password since then by legal coercion.
To cite a crude analogy, if the DEA shows up at your doorstep and says "Please sir, we would like to come in and see if you have a meth lab in your basement." you have a perfect right to say now, constitutionally guaranteed. If they show up at the same doorstep and meet you outside and show you a warrant and ask you politely to unlock the door to let them in, the fact that you do have a meth lab in your basement and will "incriminate yourself" by opening the door will not save you in court. If your house were unopenable (those six foot thick concrete walls) and could only easily/cheaply be opened by means of a combination lock, you bet your sweet bippy that the court could order you to divulge the password or rot eternally in jail for contempt.
It's the warrantless search that is at issue here. They want to be able to search your house with or without your presence or knowledge, with or without a key, and with or without a warrant being presented to you or any of your legal representatives. We can only wonder about whether or not a warrant will be required at all, and we don't HAVE to wonder that if all locks have to be openable with some sort of universal combination or key, it won't be long before our house can be entered at will by individuals who aren't even nominally members of the government or any official agency, who might USE our basement to make meth in, without our knowledge.
rgb
Hmmm, is the origin of the phrase "Bob's your uncle"?
Robert G. Brown (a.k.a. "Uncle Bob" to the offspring of his brothers and sisters...)
(Well, actually not. Actually Uncle Rob. Sorry for the confusion.)
... force you to decrypt any encrypted document with a court order. In fact, the law is so broad that if you go into court and the judge says "please give these nice officers the encryption keys for your hard drive" and you say "no", they can say "OK, I'll just put you in jail for contempt of court, without bail, until you do." Which can literally be forever. There are no limits that I know of for jail time for contempt of court for an ongoing refusal to comply with a court order. So it can literally be life not even in prison, in JAIL, until you do.
If th issue is terrorism, the powers are even broader and can involve you being sent to a concentration ca -- I mean "federal jail on a remote island" until you cough up a lot more than just the keys.
What they want is the power to read dynamical communication streams in real-time, because decrypting them is often too difficult even for the NSA and because a lot of them are encrypted with one-time or digitally saved keys so that a user CAN'T just cough them up. If my ssh private keys went away, do you think I have them memorized? NOBODY could decrypt my old network traffic, not even me!
Now we just have to wait a bit for the legislative branch to realize that a) we lack the theorems needed to make their nifty idea work; and b) any end user can trivially work around it by simply exchanging keys for one of the known secure algorithms; c) it isn't necessary for any saved, recorded data; and d) it isn't constitutional. It's exactly like trying to pass legislature that would require all house keys to be "registered" and constructed in such a way that a master key in the possession of the police would open them. Good luck with that one.
Besides, they already can. The key is called a "brute force", and if they use it, yeah, they have to go up against the effort the householder put in to stopping brute force entry. If their "house" is a repurposed bomb shelter with six foot thick concrete walls, good luck to them.
rgb
I suspect that it is every bit as much used for its high thermal conductivity. Sure, less strain on the motor and if there are lots and lots of platters that could be an issue. But it also helps to keep the whole thing cool.
Sure, bit it isn't just "repeat itself". It is "show incorrect correlations within" across all possible separations. Where "incorrect" is a pretty delicate mathematical specification and in many cases virtually impossible to resolve empirically. For all of the "good" PRNGs, it IS more or less impossible to resolve their non-randomness empirically, even though it is known that somewhere out there, they aren't.
I see. So alcohol is related to the issue of whether or not 3 colors in quantum chromodynamics is enough to be considered "infinite"? Or are you trying to say that alcohol is normal. If the latter, I'm willing to believe it, because a D1 (as I understand the "n" in Dn) is a one sided die, which is sort of like flipping a mobius strip as a one-sided coin, and if I try to visualize that I'm gonna need a drink to soften the headache. But the resulting distribution isn't very bi. Or normal.
I don't know why you say "you crazy physicists" you crazy applied mathematician, you...;-)
Besides, I agree with everything you say. A random number generator is an oxymoron, which is why the "P", right? Pseudo random number generators. I'm not even convinced that quantum processes are "truly random" unless and until they are subjected to far more rigorous tests than they have been so far, and besides, if one believes quantum theory itself one has to believe the Nakajima-Zwanzig construction and approach to the generalized master equation, where in it is revealed that in quantum mechanics "random" just means "in an incompletely specified state". Which makes perfect sense, as quantum theory is no more random than classical theory -- arguably LESS random in some ways.
However, it is possible to subject any proposed "random number generator" -- or at least any (hypothetically) random uniform N-bit integer generator to a hypothesis test, with the null hypothesis "this is a perfect random number generator". One then can generate any statistic you like that should have some known/computable distribution and that has an equally computable cumulative distribution function, invert the result, and convert it into a p-value. One can repeat this process indefinitely to see if the distribution of p-values one obtains is itself a uniform distribution from 0 to 1 (p itself should be a uniform deviate for a perfect RNG for any hypothesis test of this sort).
Then one is studying the hypothesis the way physicists and maybe even some applied mathematicians study nature. You can never prove a (null) hypothesis true, but you can accumulate sufficient evidence to be pretty sure it is false, with patience. Or not. Failing to falsify the null hypothesis doesn't prove it true, it proves that we can't prove it false yet.
We have a pretty good pile of PRNGs that we cannot prove to be non-random by reasonable empirical tests, even ones that consume rather large numbers of random numbers from the generator. Some of them have computable periods, so we know they aren't "really random" a priori, but they might well be indistinguishable from random for the entire length of the period, or not. Since we can't come close to running the period of any of the long period generators (and since it is easy to extend the period if we should ever get close) we may never find out.
But yeah, absolutely. Unpredictable is good. A "good" PRNG is more than unpredictable, though. Unpredictable is pretty easy. A good PRNG has no discernible correlations across very long intervals, where "discernible" involves using subtle tests at both very small and very large scales. It is >>uniform, in just the right way. And in the end, we can't find a problem where we can compute the answer using theory and compute the same answer via some sort of Markov chain or Monte Carlo process based on supposedly random numbers where the latter gives us a significantly wrong answer.
Evidence? Seriously dude... 30%? That's laughably incorrect.
I think the number of rolls were totally adequate... should be followed by "to resolve deviations from perfect dice as large as X with a p-value for the null hypothesis less than Y" (where ideally, Y isn't something moronic like "0.05"). Maybe something like 0.000001.
And damn skippy, you really do want to sample multiple runs of multiple dice from multiple sources per manufacturer, because one thing you are testing is the quality of their production line, the FREQUENCY with which their dice turn out to be worse than X per p Y given N samples in a run.
This isn't really that easy.
rgb
3000 rolls isn't anywhere near enough.
Amen, brother. Or rather, it might be enough. Depends on how different P(i) is from 1/N for the ith side of N-sided dice.
To put it another way, to get to a p-value on the null-hypothesis "this is a perfect die" that I'd be willing to accept as "certain" evidence that the die is bad is determined by the binomial distribution of the actual distribution of P(i) compared to the "ideal" distribution P(i) = 1/N. For some distributions ten rolls is plenty (if you get the same number 10x in a row on a 20 sided die, 20^10 is a pretty big, pretty unlikely number...;-)
OTOH a PERFECT 20 sided die isn't going to equally populate 20 bins filled with 3000 rolls, 150 each. At least, it will happen so rarely that I'd be suspicious of THAT in turn.
Damn. I thought it equaled 10.
Absolutely dead on the money. I can't mod you up because I commented myself on the thread, but as the primary author of dieharder, I couldn't have said it any better. Detecting non-randomness in crypto-PRNGs implicitly comes with a door prize -- being able to become fabulously wealthy overnight, either semi-legitimately (letting the world's financial system bribe you with a few billion a year NOT to reveal how to crack internet security, for example) or by just cracking internet security and stealing lots of money or by selling the tools to the NSA (risky, as they might "pay" you by incarcerating you for life or terminating your existence as a national security risk once they had the tool in hand).
But there are a number of NON-crypto-grade PRNGS are both fast and good enough that it is essentially impossible to detect a lack of randomness, certainly not unless you process a few terabytes of random numbers output by them. And even stock Unix PRNGs are good enough for nearly all games. The days of using really, really bad RNGs in distribution libraries is mostly over, even though one does not to use a bit of common sense about it. Dieharder lets you test the speed of the GSL implementation of a small mountain of generators including several system generators and get a feel for optimizing the speed/randomness question.
rgb
I'm very curious. Why would you say that pseudorandom number generators are't fit for "competition use"? Specifically, I'd be happy to provide you with a small stack of generators that I cannot make fail any of a rather demanding set of tests in dieharder. I'd also be very interested in any test you have that they would consistently, and correctly, fail, or that would fail any set of them but not fail a "true" RNG if one could find such a thing.
This isn't just the sort of thing you can just state as if it is true. You need to have some research or counterexamples showing it to be true. An AES-based generator isn't going to fail any simple/trivial test that I know of. In fact, it won't fail any DIFFICULT test that I know of, certainly no test I've managed to implement so far. A Mersenne-Twister in principle can be made to fail, but in practice it is not simple to demonstrate any particular failure. Any of the crypto generators are almost be definition difficult to fail. Marsaglia's KISS-class generators are very difficult to fail. Permuting a number of comparatively poor generators -- e.g. linear congruential generators with known hyperplanar correlations -- makes them much more difficult to fail, as does XOR-ing generators with different kinds of weaknesses. Some of these things do increase the time required to compute them, but there are a number of them that are both fast and still damn good.
So what evidence do you have that ALL PRNGs -- for you haven't bothered to exempt any PRNGs including crypto-class and e.g. BBS -- aren't fit for competition use, and what exactly does that even mean? If you mean fit for use in gaming, you are insanely incorrect, as nearly any of the BUILT IN generators in modern Unices are fit for gaming, which is not a particularly demanding application set and requires only uniformity and a lack of any compelling short range correlations to be sufficient for most purposes. Is there some other "randomness competition" other than passing suites of tests of randomness or the ultimately practical test, being able to crack internet security and destroying the world of e-commerce overnight? Because I don't think this is correct.
rgb
I remember, back when I was taking a quick basic class (LOL!), that I noticed an anomaly in how the random number generator produces numbers.
Specifically (well, as specifically as I can recall without digging out old source code) I noted that the output of the RNG favored multiples of 4, after having some "Difficulties" with random numbers not being random enough in one of my programs.
I wanted to test that notion, so I created a small program that "should" have painted the screen with random colored fuzz, using a random walk. (EG, the X coord, the Y coord, and the color value are all based on "random" picks from the RNG)
Imagine how much I laughed when I saw a diagonal banding pattern appear instead of random fuzz, out of the random walk.
computed random sources: Random enough for some applications, but caveat emptor!
Well, you could have read Knuth's Art of Computer Programming, vol 2 and learned a bit about the many possible flaws one can find in pseudo-RNGs, where clustering on hyperplanes at some dimensionality is one of the simplest. Or, you could use e.g. diehard or dieharder to test the RNG a bit more scientifically than with outputting a 2d screen of fuzz (which will work, but only for a really truly terrible generator). Nowadays, though, there are many very, very good pseudo-random number generators out there. Mersenne Twisters, for example, have correlations only in some gawdawful high dimensionality (over 600) so you Will Not See Bands in 2D fuzz. AES-based routines, or threefish based routines, will give you cryptographic quality RNGs (except for the bit about choosing the seeds, if you repurpose them as PRNGs). Variations of Marsaglia's KISS generator are Very Good Indeed.
You can get some of these in the Gnu Scientific Library or built into R. Any of these would be random enough for almost any purpose requiring random numbers, since so far it is pretty damn difficult to come up with a test for randomness that they fail, and any purpose for which they fail would in principle be a good test. If you find one, let me know and I'll stick it into dieharder.
rgb
Really? I would have thought it would have been completely binomial and only somewhat normal.
I find a lot of people who are agnostic or atheist have actually made science their religion. Most aren't even practicing scientists, and instead of looking to the scientific method to teach them new ideas, they have "faith" in theories despite science not yet having proven or disproven them. They use science as their religion not to further science, but to attack religion. Your comments are pretty close to putting you in this bucket.
Excuse number #912 -- "atheism is a religion too". Or better yet #912.A -- "science is a religion". Please. If you know anything at all about science, its purpose, and how it works, you know that it is not a religion. It is a way of figuring out what it is best to believe about the real world in a systematic and improvable way. Note well the two essential components -- "about the real world", and "in a systematic and improvable way". It addresses the real world, not a fantasy world, and the standard for truth is thus this objective world itself, not what people have said about it or believe about it or wrote about it in an ancient book long before we had anything vaguely approaching a science. But the second part is just as important. If I make a claim about some systematic organization supposed to hold in th real world, it is possible to accumulate evidence that supports the claim, refutes the claim, or is neutral towards the claim. Over time, more evidence and better methods of looking generally result in claims that we believe very, very strongly to be either true or very close to true, claims that we believe not to be true, and claims that cannot be decided by the evidence at hand. In all cases the standard of truth is correspondence of the assertion with reality itself, not with argumentation about reality, although the reasoning process is Bayesian and hence one isn't building up evidence-supported beliefs in isolation.
"Science" is not a religion, it is the set of interlocking assertions that have the strongest, mutually supporting evidentiary support. It is literally what it is best to believe about the real world according to an actual standard. It does not assert perfect truth, it asserts probable truth, provisional probable truth at that. If you want to actually learn something about the reasoning process involved, I would recommend E. T. Jaynes' "Probability Theory, the Logic of Science". You might also want to peruse Richard Cox's monograph "The Algebra of Probable Inference". The difference between a religious text and these two works is so profound, so obvious, so glaring, that perhaps you will reconsider your rash statement that science is a religion. These books establish, via a minimal set of axioms, a direct connection between evidence and networks of probable beliefs -- they provided mathematical support and a proof of sorts that it is better to believe things given evidence than to believe any random notion that is asserted by anyone, anywhere, for whatever reason that is not supported by evidence, that contradicts beliefs that are supported by strong evidence, or that is contradicted by the evidence itself directly.
I would offer examples -- but is there really any point? There are a near infinity of possible religions. There are quite a few actual religions, religions that contradict one another on numerous points, and the number swells to a really large number if one allows (as one should) all of the religions ever believed by any vagrant tribe throughout history, and all of the named variations on religions loosely shared between tribes.
For starters, probability theory would dictate that even if one knew that precisely one of these variants was precisely true, it is rather improbable that your particular beliefs out of this set of possibilities is correct. It isn't even probable that your beliefs come from a major family that could be correct. The odds are against any given religion being correct before you examine evidence. Without evidence
Seriously? The moon is gravitationally locked now, sure, but the Earth (and moon) still being liquid/hot when it slowed to a lock? I don't think so. For this to be plausible the moon would have had to coalesce, in an orbit, with nearly zero spin angular momentum, which seems absurdly unlikely. Otherwise, like a bird on a rotisserie, it would have been "roasted" pretty much equally on both sides. So maybe, but I doubt it.
The census in prose with the lovely insertions of slaughtering the Midianite captives including women and children EXCEPT for the young virginal girls who were given to Moses' soldiers to -- wait for it -- rape and enslave, followed by a half page of tallying up the loot of a genocidally slaughtered civilization and recording how much of it Moses, the priests, and the war captains got. If one rewrote it and published it as an account of what the ISIS "caliph" did with, say, Beirut or Damascus, the world would be shocked and cry out everything from genocide to infanticide, but because, well, it's MOSES, well then it is OK. Even Jesus loved Moses, and entertained the man himself during his transfiguration.
Come to think of it, a lot of what ISIS has done does fit right in with Old Testament reports of God's Own People.
rgb
This is the hardcore Salafi doctrine by which people like Daeesh operate by and it's theologically valid, though batshit insane.
HOWEVER, only about 70-80% of Muslims are Sunnis and of those maybe 10-15% are Salafis.
Oooo, let's do the arithmetic. There are roughly 1.6 billion Muslims (sources: multiple, although many would go as high as 1.9 billion and rising). Let's be conservative and go with 70% of the smaller number -- 0.7 x 1.6 = 1.12 billion are Sunnis. Of these (again, being conservative) 10% are Salafis. That means 112 million Muslims are Salafis! Or as many as 200 million, of one uses the higher end of that projection. Of the Salafis, roughly 10 million are estimated to be Salafi Jihadists. So sure, less than one percent, but the original numbers are so large that the actual numbers are still huge, if you view them as a potential army and recruiting ground for violent Jihad.
So what you are saying is that the Daeesh are a huge bunch of psychopathic assholes led by an eloquent sociopath from Baghdad, because while it is true that all Salafis, no matter how batshit insane the theology to which they subscribe, are not Daeesh Jihadists, they are at the very least a fertile ground for recruitment and have the potential to become a bandwagon several hundred million strong if ISIS appears to be succeeding, and it also strongly suggests that they aren't all foreign psychopathic assholes. There is a rather large pool of the domestic home-grown type.
All of which was true for Al-Queda as well, as it isn't just Salafi Jihadists. Jihad is comparatively ecumenical in Islam, and while it is not necessarily violent, it can easily become violent, often quite rapidly. There is a large disaffected population of young, volatile Muslims worldwide. In some, but not all, non-Muslim countries they feel highly marginalized. In many Muslim countries the governments are so corrupt that even though Muslims per se aren't marginalized, life is worse than it is under the foreign non-Muslim governments, marginalization or not. Then there is the antique Sunni vs Shia problem -- a division that perpetuates Muslim on Muslim violence and weakens Islam immeasurably as its adherents perpetually discover that the enemy of my enemy is a better friend, Muslim or not. Far too much of Islam is tribal culture still living in the 8th or 9th century, but dreaming of the 12th and 13th centuries when it was a, perhaps the, dominant world power. Young Muslims see this division and are easy meat for any charismatic religious leader who promises to end the rift and restore Islam to its "rightful" status as a, no, as the dominant world culture.
At the end of the day, though, there is the Quran. And the Quran is a stupendously violent document. It was a revolutionary document back when it was written, designed to generate an us vs them mentality and create a social and religious identity to support the establishment of a violently conquered tribal empire. It succeeded. Its memes are strong. This makes the followers of any sort of literalist interpretation of the Quran enormously dangerous to non-believers, because it quite literally makes converting or killing or enslaving them a holy duty, especially if they in any way interfere with or refuse to live under the religious laws of Islam, believer or not. This isn't really arguable -- one can just read the damn thing and see for yourself (but as usual with holy scriptures, nobody does, they just prefer to quote what somebody else says about them in a pious way). Being a liberal Muslim is even more difficult than it is to be a liberal Jew or Christian, and that really isn't easy if one reads, say Numbers 31 in the OT to see how God really feels about the Midianites and non-Jews in general, or reads Leviticus to see how he really feels about slavery and women. In all cases one has to sort of elide all of the passages that don't agree with what you want to believe is The Good, an
Sigh. I pitty me, since you are obviously clueless about more things than I can easily fix.
Gravity can add energy to things with "100% efficiency", but only after you've done the work of raising them up in a gravitational field. Pretty much the same thing is true for electrostatic energy. Magnetic fields do no work (seriously, and don't argue with me as I'm teaching electrodynamics at this very moment and You Will Be Wrong of you say otherwise and I will cheerfully prove that:
dW/dt = q(\vec{v} \times \vec{B}) \cdot d\vec{l}/dt = 0
as an identity (d\vec{l}/dt = \vec{v}, and A dot (A times B) is zero). I suppose one can do no work with 100% efficiency, but your entire comment is three lines of pointless.
Now, I could wax poetical about the second law of thermodynamics and the presence of irreversible losses in even simple things like dropping something macroscopic in an actual gravitational field that cause some energy to be diverted to heat, causing a loss in thermodynamic efficiency, which I actually can define as well, but you are quite right. I was referring to the horrendous inefficiency of adding somewhere between 32 and 64 MJ/kg to an object lifting it with rockets. You know why? Because (and follow this argument carefully, because it is pretty important):
That's the only way we have to lift an object to Earth orbit, be it low or geosynchronous.
Sure, there are fantasies about alternatives. For example, a great Heinleinian favorite is the linear accelerator (and yes, I've read all of Heinlein's works, many of them too many times to count). Now, go out there and actually cost one out, and figure out how you are going to solive the many, many engineering problems associated with accelerating an object to order of 11.2 km/sec plus whatever you need to punch through the atmosphere, without it burning to a crisp and without thermodynamic losses at every step along the way -- from the power plant that makes the electricity to the wires that carry the power to the currently imaginary rings aligned along some currently imaginary mountain ridge that will fire an undesigned and improbable payload container along a track some hundreds of unflattened kilometers long (at least if you want to contemplate firing humans inside).
Then there is everybody's SciFi favorite, the Space Elevator. In this fantasy, somebody puts something nice and massive in geosync orbit above a land point on the equator. Then they imagine -- and I do mean imagine -- building this really, really long cable made out of Imaginarium, an imaginary material that is so strong that it will not snap under the stress of its own weight when dangling from 5 earth Radii out back to the surface, and that can (what the heck, imagination is free!) carry a payload as well! And then, it is flexible enough to be run in a continuous belt around a spindle in the asteroid, under tension, without doing bad things to the asteroid's "orbit", which would not any longer be at 5R_e because, well, you are pulling down on it with the tension in the belt.
Now the idea is to turn a crank and lift things up the "elevator" to not-exactly orbit! As the crank turns, one "car" would ascend while the other descends, so you'd only have to lift the "weight" of the payload, at least if we ignore friction and the like in the sprockets under enough tension to lift the weight of 20 to 30 thousand miles of cable plus the payload.
Now the owners of this particular fantasy do tend to leave out a handful of, um, "obstacles" to their pretty little scheme. For example, the fact that building it would cost a few hundred trillion dollars, the fact that it isn't even clear that materials that could support their own weight from the pseudorbit to the ground could even theoretically exist, given that the strength of molecular bonds is capped at something like eV, the molecules themselves have a minimum mass and volume, and I suspect (but haven't proven) that a simple scaling