Well, the way for a correlation to exist and not be due to A->B or B->A is if there is some third thing C such that C->A, C->B. So if you aren't aware of C, all you see is a correlation between A and B.
However, in the case in which you can control the situation, you can try to eliminate the possibility of there being a C. With aspirin, you can take two groups of people with headaches, and give aspirin to people in one group and not in the other, and compare, keeping other variables the same.
Of course, there is error in this, but as you increase the group size, the error decreases, so pick a sufficiently large group and the chance of some third factor coming into play approaches zero.
This also leads to a deeper question: what aspect of taking aspirin causes the headache to go away (if it does)? It might be a psychological effect from taking medicine and believing your headache will go away as a result, but you can test that by giving the control group a placebo that contains the same ingredients except for the active ingredient you're testing.... And so on and so on. The point being, by using a control of some sort, you can try eliminate possible external causes. Also, usually one wouldn't be satisfied with a result in the sense of the result 'proving' something (which it can't really do, it just supports it) without having some model for how the thing they're testing works. And models like that can make predictions, which you can then test. If you can predict something correctly based on the assumption of a certain model, then that lends support to your model. The next step would be to look for other possible models that make the same prediction, and find the simplest (Occam's razor, which is no guarantee of correctness). Or, better yet, split your models into groups which are each internally mathematically equivalent (which you can't do anything about), and then find out where the models give differing predictions, and test those predictions.
You're never totally certain, but you can do better than just finding a correlation by controlling the parameters and requiring that your results show some sort of internal consistency and predictive ability.
It doesn't necessarily follow that this problem was solved now because of lack of funding. Rather, I'd say it's more likely that it means that the difficulty of the two problems is about equal.
We still can't solve the three-body problem analytically (except for some special cases), and thats been around for 400 years. And its not for lack of trying.
However, only within the last 50 years or so could we make approximations to the solution that work for long enough to be interesting and give insight into the problem. It's the availability of computers that makes it possible.
Fluid dynamics is a hot topic in astrophysics right now (simulating stars, gravitational collapse of nebulae, accretion discs and jets around blackholes,...), and there's a lot of consideration being given to tricks to solve Navier-Stokes (and other more complicated models that include the fluid being conducting or charged, or in some GR framework). So it's reasonable to expect that with new algorithms popping up, and refinements on the old ones, suddenly some intractable problems become accessible.
So I don't think that this was a 'problem left behind', as much as a problem which is just now becoming solvable. (Part of) the reason we spend billions on particle physics and not on this sort of problem is that the minimal 'thing' to advance the science in particle physics costs billions, whereas nowadays one can run fairly large-scale simulations (of classical systems) on a $2000 laptop: the biggest cost for those problems is hiring students/postdocs/professors to work on them. So really there what funding enables is diversity in the problems being tackled (how many laptops can you afford? how many grad students?), rather than the speed at which any one particular problem is solved.
Of course, this isn't true of some problems (quantum systems) which you really do need 1000 cutting edge systems all networked together to solve even a simple problem. In that case, you're going to have to be willing to throw a fair amount of money at the problem before you can see any progress.
Ice floats because its less dense than water. If you had ice which was frozen onto the bottom of the seabed, then melting that ice would cause a drop in sea level (because the water takes up less space as a liquid than a solid). If you have ice which is floating in the ocean, then melting that ice has no effect on ocean level (an ice block displaces an amount of water with mass equal to the total mass of the ice block). If you have ice which is anchored to land which is above sea level, and it melts, then it can cause an increase in sea level.
So the question is, what fraction of ice is in the form of icebergs, what fraction in submerged structures, and what fraction fixed to land above sea level.
You can disable cookies, delete cookies, etc, and thus show up as a totally new person. This is part of the base feature-set of most browsers, even if few people actually take advantage of it.
With a machine-unique ID, in order to get around it as a unique identifier, you have to fight against what your own software is doing (i.e. trick it into sending a different ID). So that makes it so that people who aren't previously aware of the issue won't even know that they CAN disable it. It takes securing your privacy from a legitimate feature to something shady that only 'hackers' do.
It doesn't take great powers of judgement, all it takes is a special interest. View it as a particular case of a more general phenomenon. If you have 'set of things you value' then you will be oversensitive to any perceived threat on 'set of things you value'. Saying that hackers are able to detect totalitarianism is no different than saying that mothers are able to detect threats against their children or if you want a negative example, saying that extreme racists are able to detect whenever the targets of their hatred are making progress in society.
The point isn't that this is some magical special power that hackers and hackers alone have. It's that the ability to perceive threats to specific things depends on how much you care about said things, and that hackers as a fringe group are more likely to have something they consider their right taken away than a group that is not on the fringe.
Well, I could justify that with other analogies, but it wouldn't really be a good argument.
Instead, I'll propose that perhaps what happens is that people who interact meaningfully with humans come to equilibrium with the rest of society much more quickly. So if there's some process by which people are being made to believe that certain infractions of their rights are okay, or even desireable, people who interact strongly with society will be more susceptible to picking up the same reasoning than people who are isolated from it.
To put it another way, people who feel that they need to sustain themselves independantly from society will be more concerned with their personal freedoms than society's goals, whereas people who feel they draw more support from society will be less concerned with the freedom to act in a way perpendicular to the majority of society, but will be more concerned with how society as a whole behaves. So someone who defines the important thing in their life to be how they interact with society won't necessarily care as much if the particular form of the interaction changes (you can do this and that, but not the other thing) so long as the interaction itself remains. On the other hand, someone isolated from society will care more about the specific nature of actions, rather than doing said actions in groups.
Given how abstract the preceeding paragraphs were, maybe I should've stuck with analogies: 'It's easier to spot the shape of a forest from the outside than from the inside'.
If we try to extend our supplies using conservation, then no, we mignt not last 100 years. Lets say we have enough for 10 years right now, erring on the side of caution. Then to last 100 years we need to make the majority of our power drains 10 times as efficient. Given that most thermodynamic processes for converting one form of energy into another have a maximum efficiency of around 70% at the temperature differences we can safely work at (Carnot efficiency), that would mean that we'd have to be using processes that are now 7% or less efficient. This is very old science, and was a big concern back in the days of steam engines. Right now, we're (mostly) doing better than 7%.
The point isn't that we can't be more efficient. It's that we can't be sufficiently more efficient to make much of a difference compared to what we can gain by taking advantage of nuclear power.
The big problem with nuclear power isn't that it produces waste. Everything produces waste. Nor is it the danger of meltdown or incorrect storage of fuel. Those things are very local risks, and statistically are sufficiently infrequent that the total 'cost' including lives and property damage, is still much much smaller than coal or oil.
The big problem with nuclear power is human psychology. People see something that they know was once used to kill millions, and are acutely aware of the times in which there have been nuclear accidents, and then immediately in their minds assume that every nuclear plant will fail, and that it will fail catastrophically. If you were to ask people (who do not live near a plant of any sort) whether they'd rather live next to a nuclear plant or a coal plant, I have to wonder what they'd say, compared to people who actually do live near either structure. People who live near a nuclear plant are going to have evidence which to them suggests that it is perfectly safe: the fact that they haven't experienced a meltdown or other disaster. Whereas people who have not done so are going to extrapolate based on the few cases they are aware of, which are entirely of the 'bad' variety (since who would make a news report that a nuclear plant operated perfectly this week?).
It's the same way with any negative event. People very quickly become afraid of it, and give it much higher relevance than it, statistically, deserves. We see this in all sorts of things: people's reaction to terrorist attacks, disasters in the space program, plane crashes (goodbye Concorde) and so on. The lifetime of fission byproducts is a bad number to use to estimate the price of nuclear fission. In practice, the price is almost negligible, considering that we can reprocess fuels to reduce the amount of fuel and its lifespan. The main thing stopping that is the whole fear of nuclear weapons, which is a bit silly since anyone who wants to develop the technology to extract particular isotopes can probably manage it in relatively short time anyways. It's hard, but not so hard that we should believe that it will never be done again. If we were to dump our waste in a suitable location, sealed, etc, then neither you, nor your kids, nor your grandkids, nor THEIR kids would ever see an effect. So the storage can't last for 10000 years, we're talking about running out of fuel in the next 100! Is it really that hard to build another layer of concrete around the waste dump every millenium or so? I mean, right now, we've got raw waste in barrels, which has a lifetime-of-isolation of essentially zero, and we can upgrade that to 10000 by various means.
As for 'we're going to fast, slow down, etc' the simplest thing I can say is, 'you're living twice as long as you would have 200 years ago'. Despite all of the doomsaying, predictions of environmental disaster, carcinogens, dangers of technology and so on, you're still easily outliving your recent ancestors. The fear of change stems from evolution: the species would continue to exist even if we had never progressed beyond the stage of Neandrethals. As far as survival of the fittest, we
Conservation won't do much in the long run though. If you set as the constraint that the things we do with the power can't decrease; that is that after all conservation efforts, we should still be using our computers, transporting ourselves to work and on vacations over the same distances, and so on, then you're limited by how much you can increase the efficiency of things. There will be a lower limit to how much power you can use to still be able to do the things you want to do, and once we hit that limit we can expect no more help from conservation.
Conservation takes our current effective fuel supply (that could either be how much oil and gas we can extract, or how much we can afford to burn and release) and multiplies it by a factor of two or three. Maybe for some applications as high as ten. If our fuel supply is 'small' (on whatever scale you want to consider), then the gain will be small, and as we are currently facing shortage, it stands to reason that with respect to our technological level, our reserves are quite small.
On the other hand, exploring alternate fuels is an additive process. By taking advantage of nuclear fuels, we would add a large number to our effective fuel supply. In terms of multiplication, that could be a factor of 100 or 1000 or more. By figuring out how to effectively do fusion, we add an even larger number. Other things such as solar or wind or geothermal will add smaller contributions. Solar isn't necessarily a small contribution (if you calculate the total power reaching the Earth's surface its something on the order of 10^23 watts) but you have to consider that a large portion of that has to go to maintaining biological growths, keeping the surface temperature livable, etc, and also that our current solar technology is pretty costly to construct and pretty inefficient at extracting energy. So it seems to me that the sensible way to proceed would be to first go to nuclear fission processes, continue fusion research and try to make it a reality within, say, the next 100 years (being a bit conservative here perhaps, since the usual crystal ball gazing for fusion is now+20 years). Meanwhile we should be using any and all independant methods of energy production that we can (by independant, I mean such that increasing production by one method will not decrease production by another method for whatever reason).
The thing is, if we want to advance, our power consumption is going to on the net increase. Even with improvements in efficiency, there will be more things that we want to do with our technology, and such technology will become commonplace. Simply putting a cap on our advancement and saying 'this is enough' isn't in my eyes a reasonable thing to do.
As I see it, we either get over our fears of things going wrong and accept the risk, or we will end up just burying our heads in the sand and the 20th century will end up being the pinnacle of human advancement.
The benefits are there, but maybe they require a deeper look. With GM, we can create alternate production paths for biological compounds that would otherwise be wasteful or impossible to produce. We can make crops that produce human insulin for diabetics, crops that are higher in concentrations of certain necessary nutrients that are otherwise hard to get in some situations (the 'low tech' version of this is enriched flour), and even some really crazy stuff like making crops that extract gold from the soil and bacteria that eat up oil spills.
Plus we can also be more efficient at producing food, so if for whatever reason we suffer a major reduction in our ability to produce, it won't hit us as hard.
Plus there's also the benefit that when we mess around, we learn very quickly what changes give what results. That knowledge by no means needs to be restricted to the original problem of engineering crops.
The cost is mostly in building several million/billion/trillion copies of the equipment to run simultaneously, to pay for the energy to power those machines, to pay for the liquid helium needed to cool the superconductors to contain the antimatter, etc.
If you can make a single particle per minute, and you need to spend a million dollars to make a single machine that can operate continuously for 20 years making a particle a minute, your end result is that if you want a gram of this stuff (10^20 particles, if they were antiprotons. More for positrons) you end up spending 5*10^17 dollars (500 million billion if you like). So its 'cheap' to make a couple dozen of these things, or even a hundred thousand, but when you try to make macroscopic quantities you get hit by the huge difference in orders of magnitude between the macroscopic world and the world of nuclear interactions. I haven't done the calculation, but I wouldn't be surprised that if you want to make a gram of antimatter within a human lifespan, you'd have to cover every square centimeter of the earth with particle accelerators and have them run continuously.
You might say 'why can we only do 1 a minute when we can do nuclear reactions that give us macroscopic quantities of some byproduct?' and the answer is roughly 'the strength of the interactions we can use to make antimatter are ridiculously small compared to that'. So sure, if you could find a process with a higher amplitude that produces antimatter, you might be able to step up the production. However, thats more of a fundamental physics problem than an engineering one.
No neutrons. Matter and antimatter are opposite in all of the conserved (well, jointly conserved in some cases) quantities except mass-energy and spin, so all that you're left with its a pair of photons that have the net energy, momentum, and angular momentum of the original particles. To get neutrons you'd need to have a net baryon number, which is positive for 'normal' matter and negative for antimatter. Neutrons have positive baryon number and have an antiparticle version (antineutrons).
You could of course get more complicated results, but you'd need to put in extra energy to start with so that you can create other particle/antiparticle pairs (since all the numbers have to balance out in the end), and since we're usually talking about something like electron-positron collisions, there's not much out there with lower mass-energy to be formed.
The perfect annihilation is why anything involving antimatter is so attractive from an energetic point of view: its a 100% efficient (since we're talking about individual annihilations, not an ensemble of them with someone trying to extract energy from the whole mess, we're not violating any thermodynamic laws) process that converts one form of energy (that is, the mass-energy of the particle-antiparticle pair) to another (the resultant photons). And it has no byproducts.
Of course, you could never use it as an energy 'generation' scheme since there's not really any antimatter out there to go and harvest thanks to that weird breaking of the symmetry between matter and antimatter. There are processes that do not have that symmetry (i.e. CP violation,...), but to the best of my recollection they've been shown to be too weak to produce the current state of the universe (i.e. the particular ratios of matter/energy we have). At best it'd be an extremely efficient form of storage (a major leap from those hydrogen fuel-cells).
As a consequence, it's also the classic bomb-like weapon: 'pack a lot of metastable or unstable energy into something and drop it on your target'.
However, given the difficulty involved in making macroscopic quantities of this stuff, we could probably make a couple of those carbon nanotube space ladders with the budget to build a single significantly-sized bomb.
I'm not totally up on it, but I know that there's a professor here (University of Illinois at UC) who studies the statistics of earthquakes and models them as an avalanche process.
There's a theory called 'self-organized criticality' which models systems in which you have the natural evolution of an unstable state, which can then be set off by relatively small perturbations, so large events which would be statistically unlikely in an uncorrelated system (exponentially unlikely as a function of magnitude) become far more likely (power law distribution).
This is a good strategy for raising the percentage of the US population below the poverty line, and making the US less competitive in world markets. I guess if the economy isn't so great, it won't be that much of a loss to shoot ourselves in the foot on top of it. Or maybe we're hoping to become a place with small cost of living and a mostly poor population that other countries outsource their labor to.
Now, I can understand letting some random selection of people in because you're concerned about fairness - its not as optimal as selecting for those 'elite' you're talking about, but you can take a moral stand on it, or say that there's always a chance that there are hidden geniuses in that group who wouldn't have the opportunity to use their brilliance in their home country. But what you're saying sounds like intentionally preventing other not-hidden geniuses to enter because you personally feel threatened.
There's a program called Praat that does this. However, you need a medical degree, or at least working knowledge of the muscles of the human vocal tract and what positions the must be in to produce certain sounds, in order to get any use out of it. After about 5 hours of playing with the parameters, I got it to say 'e'.
Now, if someone were to make a program that generated coordinates for the muscles that corresponded to going between different uterances, we'd be in business.
It only undermines the GPL if you're being hypocritical. The GPL acts as a defensive measure to prevent code which is currently open from becoming restricted (i.e. company takes it, makes a product using it, sues the original programmers for illegal distribution). If there were no copyrights, the GPL would be unnecessary as a company wouldn't be able to restrict the code in the first place. At worst they could release a binary (which you'd be free to redistribute) without the changes they made to the source. So calling for the defense of copyrights to maintain the strength of the GPL is somewhat silly, since you're arguing to strengthen the offense which the GPL defends against at the same time as strengthening the defense the GPL provides, which gives you a net change of zero.
For a random decay process occuring on N objects, the deviation in number decayed over some amount of time goes as sqrt(N), so if you look at the ratio of the deviation to the number, you get a relative deviation that goes as 1/sqrt(N). For something like this, with N=10^20, that means for something with a projected halflife of 1 year, the 1-sigma deviation in that halflife is going to be about 3 milliseconds. So I wouldn't worry about it too much.
I vaguely remember hearing that one of the helium superfluid/normal fluid mixtures (I don't remember if it was He-3 or He-4) at very low temperature has a lower entropy than the corresponding solid phase, since the superfluid component carries no entropy (its all in the ground state...), but the solid phase can have small excitations and has a nonzero entropy at that same temperature. So conceivably you'd see the same sort of behavior in that form of helium for a range of temperatures (i.e. it'd solidify when heated, and then liquefy again once you're past the lambda transition). Unfortunately I haven't been able to find a reference to this, so I may just be misremembering.
Not only does it solidify, but at sufficient pressures I believe it is supposed to become a metal (this has to do with there being an odd net number of electrons in a unit cell of the crystal lattice that forms). I don't know whether that particular configuration has been observed experimentally though.
If you're making macroscopic quantities of nanotubes, a microscopic inspection probably isn't practical, so I'd imagine that you'd also need to come up with a new way of checking them for defects (try to run a small current through and see if the resistance jumps at a certain point?)
Well, this is the difference between slip and no-slip boundary conditions, which can make a difference. The question isn't really 'how short are the windmills', its 'what portion of the surface airflow do they block?'. If it's a fraction of a fraction of a percent at the surface, then probably no big deal.
To give some idea of the effect, lets say we have a system of two infinite plates separated by a distance L where the upper surface is free to slip, but there is no flow through it (v.n = 0) and the lower surface has the same conditions. The fluid has some non-zero viscosity, which we'll label N.
The equation that governs the time evolution of the velocity field for non-turbulent incompressible flow (which is a good model for water but not so good for air) is:
dv/dt = (N/density)*laplacian(v)
If the entire bulk of the fluid is moving at some initial velocity v0, the velocity as a function of time will be constant (it slips freely so no problem). Of course, there's no perfect slip-free system, so eventually that drag is felt and the velocity decays. However, the timescale on which it decays depends on the degree of friction at the boundaries.
If the conditions at y=0 are non-slip, then the velocity profile is:
The modes for large n decay very quickly. At very long times, only n=0 remains, which is stationary at the bottom and moving at v0 at the surface. The next-lowest mode persists for a time that goes as L^2*rho/(N). If we plug in some numbers for air (rho ~= 1.1x10^-6 kg/m^3, N ~= 1.8x10^-5 kg/m*s [CRC Handbook 77th ed.]) and the depth of the 'ocean' you listed then we get the decay time of the longest-lived transient mode to be: 5.8x10^8 seconds, which is about two years.
Now, granted, I was using the numbers at sea level for both density and viscosity. Both decrease as you go higher in the atmosphere. Also, the entire surface of the earth would not be covered with these structures: the surface of the ocean will probably remain free, which is about 70% of the surface. Also, the structures will not be perfect dampers, and we already have some damping from geographical features, so the change won't be as large as this prediction. Still, it shows that the timescales on which something like this can affect the system are on the order of a human lifespan (unless I made a stupid error somewhere, which is quite possible).
Depends what you read. There's lots of sci-fi out there which is heavy on new science (or at least taking existing science and adding more resources than you can shake a stick at and seeing what becomes possible). Stuff by Stephen Baxter, Robert Forward, and Charles Sheffield for example.
Well, it could just be that the drive to do math, or whatever, is a subtle emergent thing, so when a stronger pull exists, like the time requirements due to a family, the drive towards academics becomes diluted. Plus, theres the peace and quiet of no kids/spouse running around, which is much more conducive to spending time thinking about a hard problem than constant ruckus.
I think you strongly overestimate the strength of this effect (if it exists, etc). Since they're measuring the neutron flux from the thing, they can pretty much tell you how likely they'd be to drop dead from radiation poisoning, but considering how finnicky this experiment seems to be I'd be surprised if the surplus of neutrons is more than a couple percent above the background radiation, otherwise we could just throw out all the expensive lab equipment and use a canary to see if something is happening.
Well, integer math in very much connected with quantum mechanics in some ways, so that aspect at least should probably show up in other cultures that understand quantum mechanics, and that could readily lead to some of their scientists pursuing the properties of integers in hope of making the QM problems easier. If you want to do spectroscopy (mass spectroscopy, UV,...), you're dealing with discrete values. Also, integers are natural for any culture that ever had need to count things and compare totals (which wouldn't necessarily apply to some arbitrary form of life though). Integers also are very important towards an understanding of chemistry: the realization that when you take some compound in bulk and break it up into components, those components have ratios that are integers or simple fractions leads to understanding of atoms, and allows a methodical approach to studying chemistry.
I suppose its more of an issue that 'discrete mathematics' is important in many things, rather than that integers themselves are. I think though that its important to realize that as much as an alien culture would be, well, alien in the things it finds important, most scientific cultures would have members who would try different approaches, since there are often many ways of solving problems, and its always useful to find the easier or more natural ways (you can do quantum mechanics with linear algebra, or a purely differential-equation approach, or an integral approach (Feynman path-integrals) or even in most cases you can just apply symmetry and vastly simplify the problem, which when formalized gives you group theory). So just as the majority of people on earth may not care about some particular esoteric field, the variety in our population means that there is likely someone interested in it somewhere in the world. And I'd say with some confidence that the same thing will be true of alien civilizations (at least, those produced by some evolution process. All bets are off if we talk about 'created' civilizations like those sci-fi stories about computer civilizations that survived the death of their organic progenitors or whatever).
Slashdot Mirror
Well, the way for a correlation to exist and not be due to A->B or B->A is if there is some third thing C such that C->A, C->B. So if you aren't aware of C, all you see is a correlation between A and B.
... And so on and so on. The point being, by using a control of some sort, you can try eliminate possible external causes. Also, usually one wouldn't be satisfied with a result in the sense of the result 'proving' something (which it can't really do, it just supports it) without having some model for how the thing they're testing works. And models like that can make predictions, which you can then test. If you can predict something correctly based on the assumption of a certain model, then that lends support to your model. The next step would be to look for other possible models that make the same prediction, and find the simplest (Occam's razor, which is no guarantee of correctness). Or, better yet, split your models into groups which are each internally mathematically equivalent (which you can't do anything about), and then find out where the models give differing predictions, and test those predictions.
However, in the case in which you can control the situation, you can try to eliminate the possibility of there being a C. With aspirin, you can take two groups of people with headaches, and give aspirin to people in one group and not in the other, and compare, keeping other variables the same.
Of course, there is error in this, but as you increase the group size, the error decreases, so pick a sufficiently large group and the chance of some third factor coming into play approaches zero.
This also leads to a deeper question: what aspect of taking aspirin causes the headache to go away (if it does)? It might be a psychological effect from taking medicine and believing your headache will go away as a result, but you can test that by giving the control group a placebo that contains the same ingredients except for the active ingredient you're testing.
You're never totally certain, but you can do better than just finding a correlation by controlling the parameters and requiring that your results show some sort of internal consistency and predictive ability.
It doesn't necessarily follow that this problem was solved now because of lack of funding. Rather, I'd say it's more likely that it means that the difficulty of the two problems is about equal.
...), and there's a lot of consideration being given to tricks to solve Navier-Stokes (and other more complicated models that include the fluid being conducting or charged, or in some GR framework). So it's reasonable to expect that with new algorithms popping up, and refinements on the old ones, suddenly some intractable problems become accessible.
We still can't solve the three-body problem analytically (except for some special cases), and thats been around for 400 years. And its not for lack of trying.
However, only within the last 50 years or so could we make approximations to the solution that work for long enough to be interesting and give insight into the problem. It's the availability of computers that makes it possible.
Fluid dynamics is a hot topic in astrophysics right now (simulating stars, gravitational collapse of nebulae, accretion discs and jets around blackholes,
So I don't think that this was a 'problem left behind', as much as a problem which is just now becoming solvable. (Part of) the reason we spend billions on particle physics and not on this sort of problem is that the minimal 'thing' to advance the science in particle physics costs billions, whereas nowadays one can run fairly large-scale simulations (of classical systems) on a $2000 laptop: the biggest cost for those problems is hiring students/postdocs/professors to work on them. So really there what funding enables is diversity in the problems being tackled (how many laptops can you afford? how many grad students?), rather than the speed at which any one particular problem is solved.
Of course, this isn't true of some problems (quantum systems) which you really do need 1000 cutting edge systems all networked together to solve even a simple problem. In that case, you're going to have to be willing to throw a fair amount of money at the problem before you can see any progress.
Ice floats because its less dense than water. If you had ice which was frozen onto the bottom of the seabed, then melting that ice would cause a drop in sea level (because the water takes up less space as a liquid than a solid). If you have ice which is floating in the ocean, then melting that ice has no effect on ocean level (an ice block displaces an amount of water with mass equal to the total mass of the ice block). If you have ice which is anchored to land which is above sea level, and it melts, then it can cause an increase in sea level.
So the question is, what fraction of ice is in the form of icebergs, what fraction in submerged structures, and what fraction fixed to land above sea level.
You can disable cookies, delete cookies, etc, and thus show up as a totally new person. This is part of the base feature-set of most browsers, even if few people actually take advantage of it.
With a machine-unique ID, in order to get around it as a unique identifier, you have to fight against what your own software is doing (i.e. trick it into sending a different ID). So that makes it so that people who aren't previously aware of the issue won't even know that they CAN disable it. It takes securing your privacy from a legitimate feature to something shady that only 'hackers' do.
It doesn't take great powers of judgement, all it takes is a special interest. View it as a particular case of a more general phenomenon. If you have 'set of things you value' then you will be oversensitive to any perceived threat on 'set of things you value'. Saying that hackers are able to detect totalitarianism is no different than saying that mothers are able to detect threats against their children or if you want a negative example, saying that extreme racists are able to detect whenever the targets of their hatred are making progress in society.
The point isn't that this is some magical special power that hackers and hackers alone have. It's that the ability to perceive threats to specific things depends on how much you care about said things, and that hackers as a fringe group are more likely to have something they consider their right taken away
than a group that is not on the fringe.
Well, I could justify that with other analogies, but it wouldn't really be a good argument.
Instead, I'll propose that perhaps what happens is that people who interact meaningfully with humans come to equilibrium with the rest of society much more quickly. So if there's some process by which people are being made to believe that certain infractions of their rights are okay, or even desireable, people who interact strongly with society will be more susceptible to picking up the same reasoning than people who are isolated from it.
To put it another way, people who feel that they need to sustain themselves independantly from society will be more concerned with their personal freedoms than society's goals, whereas people who feel they draw more support from society will be less concerned with the freedom to act in a way perpendicular to the majority of society, but will be more concerned with how society as a whole behaves. So someone who defines the important thing in their life to be how they interact with society won't necessarily care as much if the particular form of the interaction changes (you can do this and that, but not the other thing) so long as the interaction itself remains.
On the other hand, someone isolated from society will care more about the specific nature of actions, rather than doing said actions in groups.
Given how abstract the preceeding paragraphs were, maybe I should've stuck with analogies: 'It's easier to spot the shape of a forest from the outside than from the inside'.
If we try to extend our supplies using conservation, then no, we mignt not last 100 years. Lets say we have enough for 10 years right now, erring on the side of caution. Then to last 100 years we need to make the majority of our power drains 10 times as efficient. Given that most thermodynamic processes for converting one form of energy into another have a maximum efficiency of around 70% at the temperature differences we can safely work at (Carnot efficiency), that would mean that we'd have to be using processes that are now 7% or less efficient. This is very old science, and was a big concern back in the days of steam engines. Right now, we're (mostly) doing better than 7%.
The point isn't that we can't be more efficient. It's that we can't be sufficiently more efficient to make much of a difference compared to what we can gain by taking advantage of nuclear power.
The big problem with nuclear power isn't that it produces waste. Everything produces waste. Nor is it the danger of meltdown or incorrect storage of fuel. Those things are very local risks, and statistically are sufficiently infrequent that the total 'cost' including lives and property damage, is still much much smaller than coal or oil.
The big problem with nuclear power is human psychology. People see something that they know was once used to kill millions, and are acutely aware of the times in which there have been nuclear accidents, and then immediately in their minds assume that every nuclear plant will fail, and that it will fail catastrophically. If you were to ask people (who do not live near a plant of any sort) whether they'd rather live next to a nuclear plant or a coal plant, I have to wonder what they'd say, compared to people who actually do live near either structure. People who live near a nuclear plant are going to have evidence which to them suggests that it is perfectly safe: the fact that they haven't experienced a meltdown or other disaster. Whereas people who have not done so are going to extrapolate based on the few cases they are aware of, which are entirely of the 'bad' variety (since who would make a news report that a nuclear plant operated perfectly this week?).
It's the same way with any negative event. People very quickly become afraid of it, and give it much higher relevance than it, statistically, deserves. We see this in all sorts of things: people's reaction to terrorist attacks, disasters in the space program, plane crashes (goodbye Concorde) and so on.
The lifetime of fission byproducts is a bad number to use to estimate the price of nuclear fission. In practice, the price is almost negligible, considering that we can reprocess fuels to reduce the amount of fuel and its lifespan. The main thing stopping that is the whole fear of nuclear weapons, which is a bit silly since anyone who wants to develop the technology to extract particular isotopes can probably manage it in relatively short time anyways. It's hard, but not so hard that we should believe that it will never be done again.
If we were to dump our waste in a suitable location, sealed, etc, then neither you, nor your kids, nor your grandkids, nor THEIR kids would ever see an effect. So the storage can't last for 10000 years, we're talking about running out of fuel in the next 100! Is it really that hard to build another layer of concrete around the waste dump every millenium or so? I mean, right now, we've got raw waste in barrels, which has a lifetime-of-isolation of essentially zero, and we can upgrade that to 10000 by various means.
As for 'we're going to fast, slow down, etc' the simplest thing I can say is, 'you're living twice as long as you would have 200 years ago'. Despite all of the doomsaying, predictions of environmental disaster, carcinogens, dangers of technology and so on, you're still easily outliving your recent ancestors.
The fear of change stems from evolution: the species would continue to exist even if we had never progressed beyond the stage of Neandrethals. As far as survival of the fittest, we
Conservation won't do much in the long run though. If you set as the constraint that the things we do with the power can't decrease; that is that after all conservation efforts, we should still be using our computers, transporting ourselves to work and on vacations over the same distances, and so on, then you're limited by how much you can increase the efficiency of things. There will be a lower limit to how much power you can use to still be able to do the things you want to do, and once we hit that limit we can expect no more help from conservation.
Conservation takes our current effective fuel supply (that could either be how much oil and gas we can extract, or how much we can afford to burn and release) and multiplies it by a factor of two or three. Maybe for some applications as high as ten. If our fuel supply is 'small' (on whatever scale you want to consider), then the gain will be small, and as we are currently facing shortage, it stands to reason that with respect to our technological level, our reserves are quite small.
On the other hand, exploring alternate fuels is an additive process. By taking advantage of nuclear fuels, we would add a large number to our effective fuel supply. In terms of multiplication, that could be a factor of 100 or 1000 or more.
By figuring out how to effectively do fusion, we add an even larger number. Other things such as solar or wind or geothermal will add smaller contributions.
Solar isn't necessarily a small contribution (if you calculate the total power reaching the Earth's surface its something on the order of 10^23 watts) but you have to consider that a large portion of that has to go to maintaining biological growths, keeping the surface temperature livable, etc, and also that our current solar technology is pretty costly to construct and pretty inefficient at extracting energy.
So it seems to me that the sensible way to proceed would be to first go to nuclear fission processes, continue fusion research and try to make it a reality within, say, the next 100 years (being a bit conservative here perhaps, since the usual crystal ball gazing for fusion is now+20 years). Meanwhile we should be using any and all independant methods of energy production that we can (by independant, I mean such that increasing production by one method will not decrease production by another method for whatever reason).
The thing is, if we want to advance, our power consumption is going to on the net increase. Even with improvements in efficiency, there will be more things that we want to do with our technology, and such technology will become commonplace. Simply putting a cap on our advancement and saying 'this is enough' isn't in my eyes a reasonable thing to do.
As I see it, we either get over our fears of things going wrong and accept the risk, or we will end up just burying our heads in the sand and the 20th century will end up being the pinnacle of human advancement.
The benefits are there, but maybe they require a deeper look. With GM, we can create alternate production paths for biological compounds that would otherwise be wasteful or impossible to produce. We can make crops that produce human insulin for diabetics, crops that are higher in concentrations of certain necessary nutrients that are otherwise hard to get in some situations (the 'low tech' version of this is enriched flour), and even some really crazy stuff like making crops that extract gold from the soil and bacteria that eat up oil spills.
Plus we can also be more efficient at producing food, so if for whatever reason we suffer a major reduction in our ability to produce, it won't hit us as hard.
Plus there's also the benefit that when we mess around, we learn very quickly what changes give what results. That knowledge by no means needs to be restricted to the original problem of engineering crops.
The cost is mostly in building several million/billion/trillion copies of the equipment to run simultaneously, to pay for the energy to power those machines, to pay for the liquid helium needed to cool the superconductors to contain the antimatter, etc.
If you can make a single particle per minute, and you need to spend a million dollars to make a single machine that can operate continuously for 20 years making a particle a minute, your end result is that if you want a gram of this stuff (10^20 particles, if they were antiprotons. More for positrons) you end up spending 5*10^17 dollars (500 million billion if you like). So its 'cheap' to make a couple dozen of these things, or even a hundred thousand, but when you try to make macroscopic quantities you get hit by the huge difference in orders of magnitude between the macroscopic world and the world of nuclear interactions. I haven't done the calculation, but I wouldn't be surprised that if you want to make a gram of antimatter within a human lifespan, you'd have to cover every square centimeter of the earth with particle accelerators and have them run continuously.
You might say 'why can we only do 1 a minute when we can do nuclear reactions that give us macroscopic quantities of some byproduct?' and the answer is roughly 'the strength of the interactions we can use to make antimatter are ridiculously small compared to that'. So sure, if you could find a process with a higher amplitude that produces antimatter, you might be able to step up the production. However, thats more of a fundamental physics problem than an engineering one.
No neutrons. Matter and antimatter are opposite in all of the conserved (well, jointly conserved in some cases) quantities except mass-energy and spin, so all that you're left with its a pair of photons that have the net energy, momentum, and angular momentum of the original particles. To get neutrons you'd need to have a net baryon number, which is positive for 'normal' matter and negative for antimatter. Neutrons have positive baryon number and have an antiparticle version (antineutrons).
...), but to the best of my recollection they've been shown to be too weak to produce the current state of the universe (i.e. the particular ratios of matter/energy we have). At best it'd be an extremely efficient form of storage (a major leap from those hydrogen fuel-cells).
You could of course get more complicated results, but you'd need to put in extra energy to start with so that you can create other particle/antiparticle pairs (since all the numbers have to balance out in the end), and since we're usually talking about something like electron-positron collisions, there's not much out there with lower mass-energy to be formed.
The perfect annihilation is why anything involving antimatter is so attractive from an energetic point of view: its a 100% efficient (since we're talking about individual annihilations, not an ensemble of them with someone trying to extract energy from the whole mess, we're not violating any thermodynamic laws) process that converts one form of energy (that is, the mass-energy of the particle-antiparticle pair) to another (the resultant photons). And it has no byproducts.
Of course, you could never use it as an energy 'generation' scheme since there's not really any antimatter out there to go and harvest thanks to that weird breaking of the symmetry between matter and antimatter. There are processes that do not have that symmetry (i.e. CP violation,
As a consequence, it's also the classic bomb-like weapon: 'pack a lot of metastable or unstable energy into something and drop it on your target'.
However, given the difficulty involved in making macroscopic quantities of this stuff, we could probably make a couple of those carbon nanotube space ladders with the budget to build a single significantly-sized bomb.
I'm not totally up on it, but I know that there's a professor here (University of Illinois at UC) who studies the statistics of earthquakes and models them as an avalanche process.
There's a theory called 'self-organized criticality' which models systems in which you have the natural evolution of an unstable state, which can then be set off by relatively small perturbations, so large events which would be statistically unlikely in an uncorrelated system (exponentially unlikely as a function of magnitude) become far more likely (power law distribution).
Here's a random search result that explains it: Self-Organized Criticality and Earthquakes
This is a good strategy for raising the percentage of the US population below the poverty line, and making the US less competitive in world markets. I guess if the economy isn't so great, it won't be that much of a loss to shoot ourselves in the foot on top of it. Or maybe we're hoping to become a place with small cost of living and a mostly poor population that other countries outsource their labor to.
Now, I can understand letting some random selection of people in because you're concerned about fairness - its not as optimal as selecting for those 'elite' you're talking about, but you can take a moral stand on it, or say that there's always a chance that there are hidden geniuses in that group who wouldn't have the opportunity to use their brilliance in their home country. But what you're saying sounds like intentionally preventing other not-hidden geniuses to enter because you personally feel threatened.
There's a program called Praat that does this. However, you need a medical degree, or at least working knowledge of the muscles of the human vocal tract and what positions the must be in to produce certain sounds, in order to get any use out of it. After about 5 hours of playing with the parameters, I got it to say 'e'.
Now, if someone were to make a program that generated coordinates for the muscles that corresponded to going between different uterances, we'd be in business.
It only undermines the GPL if you're being hypocritical. The GPL acts as a defensive measure to prevent code which is currently open from becoming restricted (i.e. company takes it, makes a product using it, sues the original programmers for illegal distribution). If there were no copyrights, the GPL would be unnecessary as a company wouldn't be able to restrict the code in the first place. At worst they could release a binary (which you'd be free to redistribute) without the changes they made to the source.
So calling for the defense of copyrights to maintain the strength of the GPL is somewhat silly, since you're arguing to strengthen the offense which the GPL defends against at the same time as strengthening the defense the GPL provides, which gives you a net change of zero.
For a random decay process occuring on N objects, the deviation in number decayed over some amount of time goes as sqrt(N), so if you look at the ratio of the deviation to the number, you get a relative deviation that goes as 1/sqrt(N). For something like this, with N=10^20, that means for something with a projected halflife of 1 year, the 1-sigma deviation in that halflife is going to be about 3 milliseconds. So I wouldn't worry about it too much.
I vaguely remember hearing that one of the helium superfluid/normal fluid mixtures (I don't remember if it was He-3 or He-4) at very low temperature has a lower entropy than the corresponding solid phase, since the superfluid component carries no entropy (its all in the ground state...), but the solid phase can have small excitations and has a nonzero entropy at that same temperature. So conceivably you'd see the same sort of behavior in that form of helium for a range of temperatures (i.e. it'd solidify when heated, and then liquefy again once you're past the lambda transition).
Unfortunately I haven't been able to find a reference to this, so I may just be misremembering.
Not only does it solidify, but at sufficient pressures I believe it is supposed to become a metal (this has to do with there being an odd net number of electrons in a unit cell of the crystal lattice that forms). I don't know whether that particular configuration has been observed experimentally though.
If you're making macroscopic quantities of nanotubes, a microscopic inspection probably isn't practical, so I'd imagine that you'd also need to come up with a new way of checking them for defects (try to run a small current through and see if the resistance jumps at a certain point?)
Well, this is the difference between slip and no-slip boundary conditions, which can make a difference. The question isn't really 'how short are the windmills', its 'what portion of the surface airflow do they block?'. If it's a fraction of a fraction of a percent at the surface, then probably no big deal.
( 4/pi)/(2n+1)}
To give some idea of the effect, lets say we have a system of two infinite plates separated by a distance L where the upper surface is free to slip, but there is no flow through it (v.n = 0) and the lower surface has the same conditions. The fluid has some non-zero viscosity, which we'll label N.
The equation that governs the time evolution of the velocity field for non-turbulent incompressible flow (which is a good model for water but not so good for air) is:
dv/dt = (N/density)*laplacian(v)
If the entire bulk of the fluid is moving at some initial velocity v0, the velocity as a function of time will be constant (it slips freely so no problem). Of course, there's no perfect slip-free system, so eventually that drag is felt and the velocity decays. However, the timescale on which it decays depends on the degree of friction at the boundaries.
If the conditions at y=0 are non-slip, then the velocity profile is:
v(y,t) = sum[n=0,1,...] { v0*sin((n+1/2)*pi*y/L)*exp(-n^2*(N/(rho*L^2))*t)*
The modes for large n decay very quickly. At very long times, only n=0 remains, which is stationary at the bottom and moving at v0 at the surface. The next-lowest mode persists for a time that goes as L^2*rho/(N). If we plug in some numbers for air (rho ~= 1.1x10^-6 kg/m^3, N ~= 1.8x10^-5 kg/m*s [CRC Handbook 77th ed.]) and the depth of the 'ocean' you listed then we get the decay time of the longest-lived transient mode to be: 5.8x10^8 seconds, which is about two years.
Now, granted, I was using the numbers at sea level for both density and viscosity. Both decrease as you go higher in the atmosphere. Also, the entire surface of the earth would not be covered with these structures: the surface of the ocean will probably remain free, which is about 70% of the surface. Also, the structures will not be perfect dampers, and we already have some damping from geographical features, so the change won't be as large as this prediction. Still, it shows that the timescales on which something like this can affect the system are on the order of a human lifespan (unless I made a stupid error somewhere, which is quite possible).
Depends what you read. There's lots of sci-fi out there which is heavy on new science (or at least taking existing science and adding more resources than you can shake a stick at and seeing what becomes possible). Stuff by Stephen Baxter, Robert Forward, and Charles Sheffield for example.
Well, it could just be that the drive to do math, or whatever, is a subtle emergent thing, so when a stronger pull exists, like the time requirements due to a family, the drive towards academics becomes diluted. Plus, theres the peace and quiet of no kids/spouse running around, which is much more conducive to spending time thinking about a hard problem than constant ruckus.
Since when is deuterium toxic?
I think you strongly overestimate the strength of this effect (if it exists, etc). Since they're measuring the neutron flux from the thing, they can pretty much tell you how likely they'd be to drop dead from radiation poisoning, but considering how finnicky this experiment seems to be I'd be surprised if the surplus of neutrons is more than a couple percent above the background radiation, otherwise we could just throw out all the expensive lab equipment and use a canary to see if something is happening.
Well, integer math in very much connected with quantum mechanics in some ways, so that aspect at least should probably show up in other cultures that understand quantum mechanics, and that could readily lead to some of their scientists pursuing the properties of integers in hope of making the QM problems easier. If you want to do spectroscopy (mass spectroscopy, UV, ...), you're dealing with discrete values. Also, integers are natural for any culture that ever had need to count things and compare totals (which wouldn't necessarily apply to some arbitrary form of life though).
Integers also are very important towards an understanding of chemistry: the realization that when you take some compound in bulk and break it up into components, those components have ratios that are integers or simple fractions leads to understanding of atoms, and allows a methodical approach to studying chemistry.
I suppose its more of an issue that 'discrete mathematics' is important in many things, rather than that integers themselves are. I think though that its important to realize that as much as an alien culture would be, well, alien in the things it finds important, most scientific cultures would have members who would try different approaches, since there are often many ways of solving problems, and its always useful to find the easier or more natural ways (you can do quantum mechanics with linear algebra, or a purely differential-equation approach, or an integral approach (Feynman path-integrals) or even in most cases you can just apply symmetry and vastly simplify the problem, which when formalized gives you group theory). So just as the majority of people on earth may not care about some particular esoteric field, the variety in our population means that there is likely someone interested in it somewhere in the world. And I'd say with some confidence that the same thing will be true of alien civilizations (at least, those produced by some evolution process. All bets are off if we talk about 'created' civilizations like those sci-fi stories about computer civilizations that survived the death of their organic progenitors or whatever).