Physics has lots to say about causality. Special relativity describes how, from the perspective of any observer at a particular point and time, the rest of the universe is consistently partitioned into past, future and the rest. It is however symmetrical between past and future.
Quantum mechanics is also almost past/future symmetric, although to get exact symmetry you have to also flip left and right and replace particles by their anti-particles.
Statistical mechanics is the place where the arrow of time really shows itself. From a statistical perspective many features of the current state of the universe are unlikely (for instance almost all the matter being hydrogen, rather than having fused to become rather hotter helium). These unlikely features tend to fade away with time, simply because they are unlikely, in the process creating all sorts of transient complexity (life, and sprial galaxies, for instance). This process is basically one-way, because getting back into the more ordered state is very unlikely.
Thus mechanisms by which events at different times can be related, which one might term causaility are nicely described by physics.
The don't explain why things fall, but they explain with superb beauty and conciseness how things fall. Asking why things happen is verging into the realm of philosophy, rather than physics.
Surely the necessary information can be recovered from the internet end. The VoIP provider certainly knows your IP number, which will typically track you back to your ISP. They know what their NAT boxes are doing, and should be able to track you to a specific dial-up line or wireless access point. In almost all cases, this should correspond to a reasonably specific location.
Protocols could be put in place to allow automated recovery of this information. Privacy freaks could evade it pretty easily by going through a non-cooperating packet forwarder between their phone and the VoIP provider.
Not really. I think what they're actually proposing is a separate "management layer" overlaid on the internet, either on separate fibres or VLANs or.... which would operate rather higher standards for connection (see the rules for connecting University MANs to JANET for an example) than the internet and provide secure and reliable exchange of management information (traffic patterns, for instance) to allow things like prompt detection of worms.
Makes sense as a way to go from where we are, even if it wouldn't necessarily be what you'd design from here.
Internet 2 is quite different -- it's a high-capacity backbone combined with a testbed for some new protocols.
Getting it away from Earth means that you can get long exposure more easily because Earth doesn't keep getting in the way. You can also keep things cool more easily, because you don't have a huge great IR source filling half your field of view. Finally tidal forces and variations in incoming sunlight are smaller, making fine pointing easy and saving on the batteries. The distance suggests it's going to one of the Earth-Sun Lagrange points, either between the Earth and the Sun or further away from the Sun (beyond the end of Earth's shadow cone).
As for using a filled image plane, rather than synthesising it by sweeping a sensor array across it, this is surely about how many low-brightness stars you can image and locate per hour. Given that the sources are dim (and/or you want to do spectrometry, so you are spreading the photons out according to wavelength), you don't want to waste any photons that come into the telescope if you can avoid it.
a 1GP image at (say) 16 bits of grayscale per pixel is 2GB. Say we take these at 1000s exposure (quite short for serious astronomy, I would think), then the raw data stream is only 2MB/s. Lossless compression will reduce that pretty dramatically, although error correction and engineering data will bump it up. Plus some of the time not all of the field of view will be of interest.
Ok. So now you have bearings and a motor. You have to run your oxygen, hydrogen and water hoses and power leads through the bearing somehow, and you need to make sure that the whole thing stays watertight even when it stops spinning and doesn't make too much noise.
As the man from Pentagon said after their ballistic interceptor test failed again, "this is rocket science".
I'm not saying it's impossible, just that it isn't simple.
A lot of work is done on proof-checking in AI, but even so, writing up your proof with enough detail and formality to be machine checked typically increases its length by a factor of 10, sometimes more. A very complex program can then convert this into a MUCH longer, but really fully detailed proof. A rather simple program can then check this final proof. Then, if the premises and conclusions are correct, and you believe the simple program then you really have checked your theorem.
Also the neutrinos get out of the Supernova core instantly, whereas the energy that is going to come out in photons has to fight its way up through the upper layers of the star.
General relativity gives a clear answer to the "speed of gravity" question. Essentially gravity does travel at the speed of light. However, it is not quite a simple "inverse-squared" attraction. Small corrections which you can think of as the properties of gravity waves mostly cancel out the "attracted to where the Sun was 8 minutes ago" effect, although a very tiny correction remains, so that the orbit of Mercury, for instance, is not exactly as predicted by Newton's laws -- although you need a good telescope and plenty of time to tell the difference.
The radiation question is a bit subtle in this particular case. There are in principle, three ways in which two deuterium nuclei can fuse:
1. to give a helium 4 nucleus and nothing else (except maybe some gamma rays) 2. to give a helium 3 nucleus and a neutron 3. to give a tritium nucleus and a proton
In a plasma 1 is very unlikely, because it's hard for a helium 4 nucleus and gamma rays to carry off exactly the right amount of energy and momentum with no other particles produced. 2 is somewhat more common than 3, but I read somewhere that this can be controlled by controlling the spins of the deuterium nuclei (or some such).
Anyway, the radiation that everyone is looking for (and not finding) is the neutrons from 2. They have a characteristic energy and a reasonable range through palladium and air. It is, at least to me, conceivable that if something is somehow enabling fusion inside the palladium, it is preferentially enabling reaction 3 (although there's no mention of detecting tritium) or reaction 1, presumbly dumping the energy and momentum into the Palladium lattice somehow.
Anyone know better than me whether this is remotely plausible?
Why can't we tap into the energy that is comeing off of the highly radioactive concentrated fission products, in a nuclear battery contained in a very thick concrete and lead shield- so that at least while we're storing it we can still use it to create new electricity?
You could. Indeed, in a sense NASA do this to power some long-distance space probes (having purified a very specific isotope from the waste that gives them the half-life and radiation profile they want. For most purposes however the heat is too low-grade to be economic. Serious commercial power generation needs steam at 1000C or more and lots of it to run the turbines efficiently enough. Nuclear waste storage typically gives you a tank of water that stays at 80C for decades.
Re:Nuclear energy works!
on
China Goes Nuclear
·
· Score: 2, Informative
The basic reaction in a standard reactor turns Uranium into highly radioactive isotopes of various much lighter elements (fission products). There also loads of parasitic reactions that make Plutonium, turn bits of the reactor into radioactice isotopes of this and that, and so on.
Once about 3% of the Uranium has fissioned, the fission products and the things they have decayed into become a problem -- they absorb neutrons when you don't want them to and generally mess up the chain reaction. The build-up of other radioactive isotopes is also a bit of a problem -- they can affect the structural and chemical stability of the reactor.
So, anyway you have to pull the fuel elements once they reach this state. Then you have two options. Firstly, you can write them off, and just try and keep them cool while the more radioactive elements decay (10-20 years) and then look for a way to get rid of them. Alternatively you can chemically separate our all the various elements present. This is a somewhat tricky and hazardous process, on its track record, but produces recovered unburnt uranium, some plutonium, various inert things and a relatively small quantity of highly radioactive concentrated fission products, which you then have to store or dispose of.
Depends if you reprocess, or even better (from this perspective at least) run breeder reactors. Also, we have not put a fraction of the effort into looking for uranium that we have put into looking for oil.
I, broadly speaking, support nuclear energy, but "As for the nuclear waste generated aftewards there are a number of clever idea's about how to deal with it including one which disposes of it in the giant fusion reaction that is our Sun" is not helpful.
There are plenty of clever ideas, of which dumping it into the Sun is frankly not one, but no tried and tested engineering solutions. If you don't reprocess, you have quite a lot of moderately high level waste, incidentally containing quite a lot of unburnt uranium and brand new plutonium. It is quite literally hot, and needs active cooling for several decades, to stop it melting or burning. Then, if you're going to be even vaguely responsible, you need to put it somewhere where it's pretty likely to stay undisturbed for 50000 years or thereabouts, which, while probably possible, is not all that easy, especially with the special handling it needs at every stage. If you do reprocess, you get some reusable fuel, which is nice, some plutonium, which you could use as fuel, but which has some security issues, and a rather smaller amount of seriously nasty high level waste, which has the same problems as the original waste only more so (it's hot, radioactive, corrosive, and needs to be stored for a long time).
So, not easy, and not to be dismissed, but probably doable with care, and probably a win over CO2 pollution. Energy efficiency is the place to start though -- if you can use less energy in the first place then you need less of everyting.
Planets much larger than Earth will inevitably be either
a) much hotter than Earth (which is the case with these ones, I think)
or
b) mostly made of hydrogen and helium, like Jupiter, Saturn, Uranus and Neptune
At our temperatues, a massive planet would captuse lots of hydrogen and helium from the initial nebula and never lose them.
In either case, no life remotely like us could exist. Of course one cannot rule out life based on some exotic chemistry, but the absence of evident life on Mercury or any of our own gas giants is a small piece of negative evidence.
QP, the class of problems solvable by a quantum computer in polynomial time is known to contain P (polynomial time on a Turing machine) and be contained in NP (non-deterministic polynomial time). Now, we don't know that P =/= NP, but assuming that it isn't, then, I think we know that QP NP, so there are NP problems NOT solvable in polynomial time on quantum computers. We don't know, even if P NP, that P QP, although there are problems currently known to be in QP, and not in P, such as integer factorisation (which is why RSA falls to quantum computing).
Assuming QP NP that tells you that no NP-Complete problem can be in QP, so the way to protect against quantum computer-equipped eavesdroppers is to find a crytpo-system such that you need to solve an NP-Complete problem to break it.
First a basic quantum system: a simple quantum system has some observation that you can make with two possible outcomes (for instance a photon polarised up-down or left-right). Typically, an actual photon is "in" a state with some mixture of the two results before you observe it, giving rise to probabilities of the two outcomes. Once you observe and get an answer the photon behaves forever more as if it had been in that state all along.
Now entanglement. Consider two photons created together in such a way that they have to have the same polarization. Not any particular one, but definitely the same one. Now the measurements still produce both results with equal probability, but measurements of the two photons will always produce the same result. The photons are called entangled. Notice that you can't transmit any information this way, but you can get the same "random" bit-stream delivered securely to two places at the same time.
There are more complicated forms of entanglement. For instance three photons whose polarizations will always be one up-down and two left-right. So a single measurement of one of them may or may not fix all the other measurements. The new result is of this kind. By entangling five quantum systems in just the right way you can observe one of them and observe any errors in the data without observing the actual data. This fixes the error and might allow you to correct it or at least to be sure that some results were free of single-qubit errors.
The teleportation thing is quite different. You don't really transport any object. What you do is to transfer the exact, unobserved state of one quantum system onto another one in a different place, using an entangled pair of systems transported in advance and a couple of bits of classical information. The innovation in this new work is to have a bunch of systems set up in different places and decide where you want the transferred state to end up in a more flexible way. Any practical quantum computer is going to need lots of tricks of this sort to get the qubits you want to operate on into the right place at the right time.
I would guess this will be deployed in the NHS administrative structures, rather than in hospitals or GP surgeries. There are loads of parts of the NHS where they need lots of seats to run a few specific applications rather than general computing:
* the NHS direct call centre operation * the huge adminstration that tracks monitors and pays for
all non-hospital NHS prescriptions * central and regional management and support -- allocating money
This really doesn't happen. A bomb near the bottom does very little, because the bottom of the elevator isn't in tension. Also early models will not carry passengers so bomb checks will be pretty easy.
A bomb higher up will cause quite a lot of the elevator to come down, if you can find a way to get your bomb to geostationary orbit and explode it close enough to cut the really very strong cable. However: (a) it comes down really slowly, over the best part of a day and (b) atmospheric resistance will, depending on how far it has fallen when it hits the atmosphere, either burn it up or slow it to a gentle flutter well before it hits the ground.
Terrorism is a risk in the sense that you might lose a $5bn asset, but even then most of it is in space and so pretty hard to reach. It is essentially no more risk than that.
The killer issue on cheap (ie PC) architectures is that the DMA controllers must compete with the CPU for memory access bandwidth. So, if a disk controller is DMAing a chunk of data into main memory and the CPU needs some instructions or data, they fight.
Mainframe architectures have multiple banks of memory on independent busses and some very expensive cross-bar switches connecting them, so that the DMA controller and the CPU will only fight if they are unlucky enough to need data in the same memory bank. Some mainframes also use expensive dual-port memory in large quantities, so that even this is not a problem.
Steve
Fusion power generation without a tokamak
on
Odds-on Science
·
· Score: 1
Disregarding the not inconsiderable ethical and sanity questions for a moment, could you achieve nuclear fusion power generation with energy breakeven using bombs?
I'm thinking a big underground cavern, a shaft with suitable twists and turns and some big blast doors and a lot of thermal power generation kit buried in the rock around the cavern deep enough not to get smashed by the blast.
Now open the blast doors, drop in the large fusion bomb (with small fission trigger) shut the blast doors quickly, detonate the bomb, DO NOT RINSE, repeat. Generate power from the hot rocks.
Could you achieve breakeven over the energy investment of making the bombs and drilling the cavern?
Realistically, there is no practical short-term ( 20 years) say benefit likely to arise from physics at these energies.
Short term benefits come in spin-offs from the technologies used to do the physics -- better magnets, materials, cryogenic technology, computing, microwave technologies, and loads more. Experience suggests that spending a proportion of your science budget trying to "push the limits of the possible" with some inspiring, but not necessarily directly useful project, such as landing a man on the Moon or understanding the physics of elementary particales at TeV energies, pays off. Obviously you don't spend all your science budget on this kind of thing, or even most of it.
Looking further down the line, particle physics and accelerator technologies developed 20 or 30 years ago are finding applications in medicine (cancer therapies, X-ray sources) materials science (muon and neutron beams to study superconductors) and doubtless many other areas I can't think of. These applications could not have been predicted then.
Slashdot Mirror
Physics has lots to say about causality. Special relativity describes how, from the perspective of any observer at a particular point and time, the rest of the universe is consistently partitioned into past, future and the rest. It is however symmetrical between past and future.
Quantum mechanics is also almost past/future symmetric, although to get exact symmetry you have to also flip left and right and replace particles by their anti-particles.
Statistical mechanics is the place where the arrow of time really shows itself. From a statistical perspective many features of the current state of the universe are unlikely (for instance almost all the matter being hydrogen, rather than having fused to become rather hotter helium). These unlikely features tend to fade away with time, simply because they are unlikely, in the process creating all sorts of transient complexity (life, and sprial galaxies, for instance). This process is basically one-way, because getting back into the more ordered state is very unlikely.
Thus mechanisms by which events at different times can be related, which one might term causaility are nicely described by physics.
In my view though this is still "how", not "why".
Steve
The don't explain why things fall, but they explain with superb beauty and conciseness how things fall. Asking why things happen is verging into the realm of philosophy, rather than physics.
Surely the necessary information can be recovered from the internet end. The VoIP provider certainly knows your IP number, which will typically track you back to your ISP. They know what their NAT boxes are doing, and should be able to track you to a specific dial-up line or wireless access point. In almost all cases, this should correspond to a reasonably specific location.
Protocols could be put in place to allow automated recovery of this information. Privacy freaks could evade it pretty easily by going through a non-cooperating packet forwarder between their phone and the VoIP provider.
Takes a bit of doing, but not basically hard.
Not really. I think what they're actually proposing is a separate "management layer" overlaid on the internet, either on separate fibres or VLANs or .... which would operate rather higher standards for connection (see the rules for connecting University MANs to JANET for an example) than the internet and provide secure and reliable exchange of management information (traffic patterns, for instance) to allow things like prompt detection of worms.
Makes sense as a way to go from where we are, even if it wouldn't necessarily be what you'd design from here.
Internet 2 is quite different -- it's a high-capacity backbone combined with a testbed for some new protocols.
Getting it away from Earth means that you can get long exposure more easily because Earth doesn't keep getting in the way. You can also keep things cool more easily, because you don't have a huge great IR source filling half your field of view. Finally tidal forces and variations in incoming sunlight are smaller, making fine pointing easy and saving on the batteries. The distance suggests it's going to one of the Earth-Sun Lagrange points, either between the Earth and the Sun or further away from the Sun (beyond the end of Earth's shadow cone).
As for using a filled image plane, rather than synthesising it by sweeping a sensor array across it, this is surely about how many low-brightness stars you can image and locate per hour. Given that the sources are dim (and/or you want to do spectrometry, so you are spreading the photons out according to wavelength), you don't want to waste any photons that come into the telescope if you can avoid it.
a 1GP image at (say) 16 bits of grayscale per pixel is 2GB. Say we take these at
1000s exposure (quite short for serious astronomy, I would think), then the raw data stream is only 2MB/s. Lossless compression will reduce that pretty dramatically, although error correction and engineering data will bump it up. Plus some of the time not all of the field of view will be of interest.
Perfectly within the compass of microwave links.
Ok. So now you have bearings and a motor. You have to run your oxygen, hydrogen and water hoses and power leads through the bearing somehow, and you need to make sure that the whole thing stays watertight even when it stops spinning and doesn't make too much noise.
As the man from Pentagon said after their ballistic interceptor test failed again, "this is rocket science".
I'm not saying it's impossible, just that it isn't simple.
In zero-gee the gasses don't bubble from the plates, they just sit there. This makes life much more complicated.
A lot of work is done on proof-checking in AI, but even so, writing up your proof with enough detail and formality to be machine checked typically increases its length by a factor of 10, sometimes more. A very complex program can then convert this into a MUCH longer, but really fully detailed proof. A rather simple program can then check this final proof. Then, if the premises and conclusions are correct, and you believe the simple program then you really have checked your theorem.
Also the neutrinos get out of the Supernova core instantly, whereas the energy that is going to come out in photons has to fight its way up through the upper layers of the star.
General relativity gives a clear answer to the "speed of gravity" question. Essentially gravity does travel at the speed of light. However, it is not quite a simple "inverse-squared" attraction. Small corrections which you can think of as the properties of gravity waves mostly cancel out the "attracted to where the Sun was 8 minutes ago" effect, although a very tiny correction remains, so that the orbit of Mercury, for instance, is not exactly as predicted by Newton's laws -- although you need a good telescope and plenty of time to tell the difference.
The radiation question is a bit subtle in this particular case. There are in principle, three ways in which two deuterium nuclei can fuse:
1. to give a helium 4 nucleus and nothing else (except maybe some gamma rays)
2. to give a helium 3 nucleus and a neutron
3. to give a tritium nucleus and a proton
In a plasma 1 is very unlikely, because it's hard for a helium 4 nucleus and gamma rays to carry off exactly the right amount of energy and momentum with no other particles produced. 2 is somewhat more common than 3, but I read somewhere that this can be controlled by controlling the spins of the deuterium nuclei (or some such).
Anyway, the radiation that everyone is looking for (and not finding) is the neutrons from 2. They have a characteristic energy and a reasonable range through palladium and air. It is, at least to me, conceivable that if something is somehow enabling fusion inside the palladium, it is preferentially enabling reaction 3 (although there's no mention of detecting tritium) or reaction 1, presumbly dumping the energy and momentum into the Palladium lattice somehow.
Anyone know better than me whether this is remotely plausible?
Steve
Why can't we tap into the energy that is comeing off of the highly radioactive concentrated fission products, in a nuclear battery contained in a very thick concrete and lead shield- so that at least while we're storing it we can still use it to create new electricity?
You could. Indeed, in a sense NASA do this to power some long-distance space probes (having purified a very specific isotope from the waste that gives them the half-life and radiation profile they want. For most purposes however the heat is too low-grade to be economic. Serious commercial power generation needs steam at 1000C or more and lots of it to run the turbines efficiently enough. Nuclear waste storage typically gives you a tank of water that stays at 80C for decades.
The basic reaction in a standard reactor turns Uranium into highly radioactive isotopes of various much lighter elements (fission products). There also loads of parasitic reactions that make Plutonium, turn bits of the reactor into radioactice isotopes of this and that, and so on.
Once about 3% of the Uranium has fissioned, the fission products and the things they have decayed into become a problem -- they absorb neutrons when you don't want them to and generally mess up the chain reaction. The build-up of other radioactive isotopes is also a bit of a problem -- they can affect the structural and chemical stability of the reactor.
So, anyway you have to pull the fuel elements once they reach this state. Then you have two options. Firstly, you can write them off, and just try and keep them cool while the more radioactive elements decay (10-20 years) and then look for a way to get rid of them. Alternatively you can chemically separate our all the various elements present. This is a somewhat tricky and hazardous process, on its track record, but produces recovered unburnt uranium, some plutonium, various inert things and a relatively small quantity of highly radioactive concentrated fission products, which you then have to store or dispose of.
Depends if you reprocess, or even better (from this perspective at least) run breeder reactors. Also, we have not put a fraction of the effort into looking for uranium that we have put into looking for oil.
I, broadly speaking, support nuclear energy, but "As for the nuclear waste generated aftewards there are a number of clever idea's about how to deal with it including one which disposes of it in the giant fusion reaction that is our Sun" is not helpful.
There are plenty of clever ideas, of which dumping it into the Sun is frankly not one, but no tried and tested engineering solutions. If you don't reprocess, you have quite a lot of moderately high level waste, incidentally containing quite a lot of unburnt uranium and brand new plutonium. It is quite literally hot, and needs active cooling for several decades, to stop it melting or burning. Then, if you're going to be even vaguely responsible, you need to put it somewhere where it's pretty likely to stay undisturbed for 50000 years or thereabouts, which, while probably possible, is not all that easy, especially with the special handling it needs at every stage. If you do reprocess, you get some reusable fuel, which is nice,
some plutonium, which you could use as fuel, but which has some security issues, and a rather smaller amount of seriously nasty high level waste, which has the same problems as the original waste only more so (it's hot, radioactive, corrosive, and needs to be stored for a long time).
So, not easy, and not to be dismissed, but probably doable with care, and probably a win over CO2 pollution. Energy efficiency is the place to start though -- if you can use less energy in the first place then you need less of everyting.
Planets much larger than Earth will inevitably be either
a) much hotter than Earth (which is the case with these ones, I think)
or
b) mostly made of hydrogen and helium, like Jupiter, Saturn, Uranus and Neptune
At our temperatues, a massive planet would captuse lots of hydrogen and helium from the initial nebula and never lose them.
In either case, no life remotely like us could exist. Of course one cannot rule out life based on some exotic chemistry, but the absence of evident life on Mercury or any of our own gas giants is a small piece of negative evidence.
You'll need to be a bit more specific. I'm using =/= for "not equal to" and for "is strictly contained in" does that help?
This is not quite right.
QP, the class of problems solvable by a quantum computer in polynomial time is known to contain P (polynomial time on a Turing machine) and be contained in NP (non-deterministic polynomial time). Now, we don't know that P =/= NP, but assuming that it isn't, then, I think we know that QP NP, so there are NP problems NOT solvable in polynomial time on quantum computers. We don't know, even if P NP, that P QP, although there are problems currently known to be in QP, and not in P, such as integer factorisation (which is why RSA falls to quantum computing).
Assuming QP NP that tells you that no NP-Complete problem can be in QP, so the way to protect against quantum computer-equipped eavesdroppers is to find a crytpo-system such that you need to solve an NP-Complete problem to break it.
Probably so, but it's not that hard.
First a basic quantum system: a simple quantum system has some observation that you can make with two possible outcomes (for instance a photon polarised up-down or left-right). Typically, an actual photon is "in" a state with some mixture of the two results before you observe it, giving rise to probabilities of the two outcomes. Once you observe and get an answer the photon behaves forever more as if it had been in that state all along.
Now entanglement. Consider two photons created together in such a way that they have to have the same polarization. Not any particular one, but definitely the same one. Now the measurements still produce both results with equal probability, but measurements of the two photons will always produce the same result. The photons are called entangled. Notice that you can't transmit any information this way, but you can get the same "random" bit-stream delivered securely to two places at the same time.
There are more complicated forms of entanglement. For instance three photons whose polarizations will always be one up-down and two left-right. So a single measurement of one of them may or may not fix all the other measurements. The new result is of this kind. By entangling five quantum systems in just the right way you can observe one of them and observe any errors in the data without observing the actual data. This fixes the error and might allow you to correct it or at least to be sure that some results were free of single-qubit errors.
The teleportation thing is quite different. You don't really transport any object. What you do is to transfer the exact, unobserved state of one quantum system onto another one in a different place, using an entangled pair of systems transported in advance and a couple of bits of classical information. The innovation in this new work is to have a bunch of systems set up in different places and decide where you want the transferred state to end up in a more flexible way. Any practical quantum computer is going to need lots of tricks of this sort to get the qubits you want to operate on into the right place at the right time.
I would guess this will be deployed in the NHS administrative structures, rather than in hospitals or GP surgeries. There are loads of parts of the NHS where they need lots of seats to run a few specific applications rather than general computing:
* the NHS direct call centre operation
* the huge adminstration that tracks monitors and pays for
all non-hospital NHS prescriptions
* central and regional management and support -- allocating money
This really doesn't happen. A bomb near the bottom does very little, because the bottom of the elevator isn't in tension. Also early models will not carry passengers so bomb checks will be pretty easy.
A bomb higher up will cause quite a lot of the elevator to come down, if you can find a way to get your bomb to geostationary orbit and explode it close enough to cut the really very strong cable. However: (a) it comes down really slowly, over the best part of a day and (b) atmospheric resistance will, depending on how far it has fallen when it hits the atmosphere, either burn it up or slow it to a gentle flutter well before it hits the ground.
Terrorism is a risk in the sense that you might lose a $5bn asset, but even then most of it is in space and so pretty hard to reach. It is essentially no more risk than that.
The killer issue on cheap (ie PC) architectures is that the DMA controllers must compete with the CPU for memory access bandwidth. So, if a disk controller is DMAing a chunk of data into main memory and the CPU needs some instructions or data, they fight.
Mainframe architectures have multiple banks of memory on independent busses and some very expensive cross-bar switches connecting them, so that the DMA controller and the CPU will only fight if they are unlucky enough to need data in the same memory bank. Some mainframes also use expensive dual-port memory in large quantities, so that even this is not a problem.
Steve
Disregarding the not inconsiderable ethical and sanity questions for a moment, could you achieve nuclear fusion power generation with energy breakeven using bombs?
I'm thinking a big underground cavern, a shaft with suitable twists and turns and some big blast doors and a lot of thermal power generation kit buried in the rock around the cavern deep enough not to get smashed by the blast.
Now open the blast doors, drop in the large fusion bomb (with small fission trigger) shut the blast doors quickly, detonate the bomb, DO NOT RINSE, repeat. Generate power from the hot rocks.
Could you achieve breakeven over the energy investment of making the bombs and drilling the cavern?
Steve
Realistically, there is no practical short-term ( 20 years) say benefit likely to arise from physics at these energies.
Short term benefits come in spin-offs from the technologies used to do the physics -- better magnets, materials, cryogenic technology, computing, microwave technologies, and loads more. Experience suggests that spending a proportion of your science budget trying to "push the limits of the possible" with some inspiring, but not necessarily directly useful project, such as landing a man on the Moon or understanding the physics of elementary particales at TeV energies, pays off. Obviously you don't spend all your science budget on this kind of thing, or even most of it.
Looking further down the line, particle physics and accelerator technologies developed 20 or 30 years ago are finding applications in medicine (cancer therapies, X-ray sources) materials science (muon and neutron beams to study superconductors) and doubtless many other areas I can't think of. These applications could not have been predicted then.