All you've done is parallelize the problem. Each string is its own highly limited computer. You'll note that to scale to larger graphs, you need to scale the number of strings. That is, you kept the running time the same but had to increase the power of the computer in proportion to the number of edges. Each bead, as a place where forces are summed, also represents a limited power computer. Diskstra's algorithm runs in roughly O(V^2) time when E is comparable to V^2, or O(E*log(V)) time when it's much lower. Not coincidentally, your physical computer's computational power grows at comparable rates.
Physics doesn't make a particularly more or less efficient computer than a Turing machine; there's no good comparison. What it does do is provide ways to massively parallelize some problems. The shortest path algorithm can be done in O(edges in path) time on a conventional computer with unlimited processors and no communications bottlenecks, which is very similar to what you have described.
Especially since compiling the code yourself is completely sufficient to prevent security flaws. Erm. You were planning to audit it, right? Since everyone knows that's sufficient.
Computer security is hard. Doing it right is really hard.
While I'm sure everything you said is true, that doesn't make it non-obvious. That system, or something like it, is a fairly obvious use of computers as applied to banking. The only reason it hadn't appeared before was because of regulatory issues. Patents should be for new and innovative ideas, not really obvious ones that no one is doing because they're not allowed to.
Given that this is specifically about long-distance transfer of money, it would seem that the interstate commerce clause applies much more readily than it normally does. Not that I support this bill, it just seems to me it's merely idiotic, and not unconstitutional.
My complaint about units of work based schemes is this: how do they reward improvements in efficiency? Much of technological advancement has been about making it easier to make things, through improvements in tools and technique. That increased efficiency translates directly into higher standards of living for all concerned (either directly or indirectly). If you base pay on work required, there is greatly reduced incentive to improve efficiency. I have yet to see a scheme that was a labor theory of value that rewarded efficiency without turning into a more complicated version of capitalism.
And what makes you think we would do better the second time around? Is there any actual evidence that anyone knows how to do better (as in, backed by at least a modicum of data, not just a few papers and manifestos)? Is there any evidence that if they do, the people actually rebuilding the economy would pay any attention?
What makes you think that drastic economic changes would actually succeed? Do you know of any significant examples of replacing a capitalist system with something drastically different and having it work?
Re:This is What Freenet Was Made For
on
WikiLeaks Under Fire
·
· Score: 2, Informative
Freenet has improved remarkably. It's certainly not what you'd call fast, but for popular content or anything small (text documents, for example) it isn't bad at all. You'd probably end up waiting several minutes for a 1MiB chunk of text that wasn't overly popular, but that's hardly problematic for something like this.
My usual browsing experience is that Freesites load their text in somewhere between 10s and 60s, with the pictures loading over the course of the next 2-3 minutes. Some load instantly if they're popular enough for your node or one of its immediate neighbors to have a copy already. If you haven't tried it recently, check out the current build. Be patient, and give it a couple hours of uptime to get thoroughly integrated, but it's way better than it once was.
Re:This is What Freenet Was Made For
on
WikiLeaks Under Fire
·
· Score: 4, Informative
If you haven't used one of the recent builds, you should try it again. Currently I have it set to use a max of 18KiB/s outbound bandwidth (I'm on a somewhat slow connection), and CPU usage varies from about 5% to 25% on a 1.2GHz Athlon. Memory usage is under 100MB. Both CPU usage and memory will grow with increased bandwidth usage, but it's way way better than it used to be.
To good approximation, it also has very little penetrating power (a mm or two at most, I imagine, though it would depend on the frequency) -- the pulse is not long enough for anything to get out of the way. So it deposits a fair bit of energy in a very, very small volume, resulting in a tiny ball of very hot mostly steam, which might as well be an explosion. Heat transfer from that to the surrounding environment will be very poor (time scale is too short), but the blast effect will be significant. The blast would probably actually destroy a larger part of your finger than the energy could vaporize, but I don't know how to estimate the exact effects.
Windows "Just works" so that's what they'll go back to.
Oddly enough, my experience is the opposite. The last time I had to set up windows, I had to go find and install a decent web browser (as opposed to just typing "apt-get install firefox" and being done with it, I had to go find the web site, download the package, install the package...). I had to locate the correct printer driver, which involved moving furniture so I could read the exact model number off the back. It kept asking me whether I wanted to allow the program I had just clicked on to do the thing it is supposed to do. It would sit there waiting for ages doing simple thing like copying or deleting files. Those few settings I really wanted to change were in non-obvious places. Several reboots were involved in this process. And they call this a modern OS? I wasn't even doing a clean install, it came on the machine!
I suppose it's just what you're used to. Windows doesn't "just work" any more than Linux does. I honestly have no clue which one is closer to "just working", but in my experience, I find Linux simpler and easier to get to do what I want it to do. Yes, I'm aware that's because I'm used to it and I'm not used to Windows, but I think the reverse is true as well -- people only find Windows easy because they're used to it.
Got any idea what shape it would be? It's a micron-diameter spot, and it's too short to penetrate (the stuff doesn't get out of the way).
Assume a spherical cow...
I made a simplifying assumption to give readers a feel for how large a chunk of stuff we're talking about. And besides, quasi-spherical might actually be pretty close to reasonable...
17 joules is plenty of energy to detect without special instruments. It's enough to vaporize about 6 mg of water, or in other words it could blast a spherical hole about 1/16" diameter in your finger -- which I think you'd notice. Granted, it's not a lot of energy, but it's not a trivial amount, either.
That this is actually quite appropriate. Since he didn't have any fiduciary duty, the SEC shouldn't take his money away. That said, since it's profit from an illegal act, I would hope that the money would be taken away -- if and when he is convicted for the crime of stealing the data.
Too often in this country we seem to be throwing every law available at people and making up new ones to go with them, when the acts we're trying to punish are already illegal. If he didn't break securities laws, he shouldn't be punished under them. Since he did (we assume, but it hasn't yet been proven) break unauthorized access laws, he should be punished under those.
We don't need more laws against things that are already illegal, and we don't need to make a mockery of existing laws by applying them to things they don't apply to. On a related note, why do we need separate "identity theft" and "atm card fraud" laws, when anyone being charged with them is already also being charged with uttering false instruments and fraud? Our legal code needs to be smaller and simpler; making it so would make it more effective and efficient, not less.
As in, computers aren't involved in artificial intelligence? Or that they aren't computers as in Turing-equivalent machines? Or what? What's the alternative? I'm highly skeptical that there's anything magical about intelligence or our brains that can't be simulated by enough understanding of what's going on combined with raw cpu horsepower.
Do you think that it is possible to demonstrate intelligence through a relatively restrictive interface that computers can handle today -- for example, a purely text conversation? If so, what is required, both for intelligence and for something that counts as "a start"? Historically, a lot of AI detractors seem to play the moving goal posts game -- defining any problem that computers can't solve well as necessary for intelligence (though not necessarily sufficient). "Computers aren't really intelligent; they can't comprehend a game like chess." "Computers aren't intelligent, they can't write music." Yet computers get slowly better at those specific problems, even if one general program can't attack all of them. So what makes your position not of that genre? (I'm not trying to be accusatory, I'm just not seeing it -- I suspect you have and answer, and I'm curious.)
I would argue no - all the examples you've mentioned are solutions created by humans executed on computer hardware.
Take the case of a program that learns the rules of the game its playing based on feedback only, without external training. Is that not a case of a single program solving a class of problems, without specific instruction? If not, why not? If it is, how much broader does the class have to get before it counts as the start of a solution? Surely it doesn't have to solve *any* problem at all to count as a start on the path.
I'm inclined to agree that (at least) a few more major breakthroughs are needed. But at the same time, I think we've seen a few glimmers of intelligence already.
It seems obvious to me that you won't get anywhere, ever, if you attempt to tackle the entire problem at once. You have to start with a piece of it, and expand that.
We have computer programs that play chess well. We also have computer programs that can figure out the rules to the game they're playing without explicit teaching, and then play somewhere between horribly and well, depending on the details. That seems a step in the right direction -- go from a very specific case to a more general one.
We have computer programs that can check mathematical proofs, and there have been programs that created proofs from nothing on their own. Is that not a step in the right direction?
A lot of the role of the human in using the machine as a tool is in converting the outside world into a form the computer can interact with. If that's not a reasonable role for the human in developing the very beginnings of AI technology, I don't know what is.
I agree we're far, far away from where some people would like us to be or think we are, and that we have a long way to go. But I ask simply this: where would you start? And haven't a few of the very first baby steps been taken already? We're not far along, but I really don't think we're at zero either.
Actually, long-haul lines are all alumium these days. The resistivity is slightly higher, so the lines are thicker, but the aluminum costs less and weighs less for the same load capacity as the copper.
It's the frivolous stuff that actually hits a lot of people that gets a lot of people angry enough to do something. The big things that only hit a few people are easier to sell to the masses, since it's always someone else being affected. But when everyone is affected...
If horror movies is the only thing hit, I predict nothing will come of it. But if this is the start of a trend, then something very interesting might be about to happen.
Revolutions have been started over taxes on tea...
Interestingly enough, this is actually quite close to how Freenet works in its opennet mode. The turnover rate is probably rather low, but it has no non-encrypted protocol header and is constantly connecting to new nodes. With some tweaking it would be very hard to detect. IIRC it also already runs entirely over UDP, not TCP, which makes injecting RST packets impossible.
Except for two important points: first, what little data I've seen suggests such collisions won't change the orbit of either object or its remains much at all. If you have references, I'd love to see them -- they're hard to find. And secondly, without a circularization burn, any piece of debris will end up on an orbit that intersects the collision point. And since the collision point is very low, that means the orbit will decay rapidly, even if it is highly elliptical with a high-altitude apogee. So even in the worst case, it won't pose a hazard to other satellites for very long. The real problem with the Chinese test is that the debris isn't going away.
Did you pay any attention to the last 30 years or so of cryptography? Any peer-to-peer patch distribution system would use digital signatures that are difficult to fake. The corresponding public keys would be distributed with the OS install or through some other secure mechanism (SSL from the main update site or similar). Any attacker that can install their own key could install a worm through that route anyway.
P2P is quite good at solving intermittent high demand distribution problems, and is quite well matched to this.
Plenty of fuel-tank sized bits of satellites have survived uncontrolled reentry, when protected by the other bits of satellite. Propellant tanks are actually fairly good candidates for surviving largely intact -- ie, still in one piece and full of hydrazine, but broken enough to be leaking hydrazine everywhere. Also, it's surprisingly difficult to overstate the nastiness of hydrazine. It's flammable, explosive, corrosive, toxic, mutagenic, carcinogenic, and teratogenic -- all at very, very high potency. At least it's not radioactive. Monomethylhydrazine (which is widely used as a fuel in applications just like this; I don't know which variety of hydrazine is in use here) is one of the most potent carcinogens known -- a single drop applied to the skin of lab rats caused cancer in about 90% of cases in one study (more than that and the toxic effects dominated).
Seriously. The propellant tanks might or might not survive, but some large bits definitely will make it to the ground, and they might well contain hydrazine or have hydrazine on them from leaking / disintegrating tanks. If they do, you really, really, *really* wouldn't want to touch it. Souvenir gathering from crashed satellites (or Space Shuttles) can be very bad for your health.
To be fair, the space junk isn't equivalent -- the junk from a satellite that's about to reenter will also reenter promptly, whereas the junk from a satellite in a high orbit will remain in a high orbit. The impact won't actually alter the orbital parameters of the junk as much as you might expect; nearly all of it will reenter promptly, and I'd be surprised if any of it managed to get high enough to present a danger to other satellites (the satellite in question is well below normal operating altitudes).
Of course, I'm not trying to say the US isn't guilty of hypocrisy -- just that this case isn't as bad as you make it out to be.
Because that's a lot harder. If all they wanted was for it to deorbit, they wouldn't be doing anything -- the satellite in question never reached its final orbit, and is rapidly decaying. It will reenter fairly soon if left alone. Presumably they're trying to prevent it from reentering where Russia or China or someone might find the pieces and get clues about our capabilities -- it's pretty clearly a recent-generation spy satellite. If they are careful about how they shoot it down, they may get some control over where it reenters.
Currently, we don't have any (non-classified, anyway) capability to attach an unmanned booster to an arbitrary orbitting satellite. And there isn't time to develop a custom solution before it reenters as the current orbit decays.
Of course, the real place these documents should be available, is Freenet. I haven't yet checked to see if they're available, though, so I can't give you a link.
Slashdot Mirror
All you've done is parallelize the problem. Each string is its own highly limited computer. You'll note that to scale to larger graphs, you need to scale the number of strings. That is, you kept the running time the same but had to increase the power of the computer in proportion to the number of edges. Each bead, as a place where forces are summed, also represents a limited power computer. Diskstra's algorithm runs in roughly O(V^2) time when E is comparable to V^2, or O(E*log(V)) time when it's much lower. Not coincidentally, your physical computer's computational power grows at comparable rates.
Physics doesn't make a particularly more or less efficient computer than a Turing machine; there's no good comparison. What it does do is provide ways to massively parallelize some problems. The shortest path algorithm can be done in O(edges in path) time on a conventional computer with unlimited processors and no communications bottlenecks, which is very similar to what you have described.
Especially since compiling the code yourself is completely sufficient to prevent security flaws. Erm. You were planning to audit it, right? Since everyone knows that's sufficient.
Computer security is hard. Doing it right is really hard.
While I'm sure everything you said is true, that doesn't make it non-obvious. That system, or something like it, is a fairly obvious use of computers as applied to banking. The only reason it hadn't appeared before was because of regulatory issues. Patents should be for new and innovative ideas, not really obvious ones that no one is doing because they're not allowed to.
Given that this is specifically about long-distance transfer of money, it would seem that the interstate commerce clause applies much more readily than it normally does. Not that I support this bill, it just seems to me it's merely idiotic, and not unconstitutional.
My complaint about units of work based schemes is this: how do they reward improvements in efficiency? Much of technological advancement has been about making it easier to make things, through improvements in tools and technique. That increased efficiency translates directly into higher standards of living for all concerned (either directly or indirectly). If you base pay on work required, there is greatly reduced incentive to improve efficiency. I have yet to see a scheme that was a labor theory of value that rewarded efficiency without turning into a more complicated version of capitalism.
if we fix it the right way.
And what makes you think we would do better the second time around? Is there any actual evidence that anyone knows how to do better (as in, backed by at least a modicum of data, not just a few papers and manifestos)? Is there any evidence that if they do, the people actually rebuilding the economy would pay any attention?
What makes you think that drastic economic changes would actually succeed? Do you know of any significant examples of replacing a capitalist system with something drastically different and having it work?
Freenet has improved remarkably. It's certainly not what you'd call fast, but for popular content or anything small (text documents, for example) it isn't bad at all. You'd probably end up waiting several minutes for a 1MiB chunk of text that wasn't overly popular, but that's hardly problematic for something like this.
My usual browsing experience is that Freesites load their text in somewhere between 10s and 60s, with the pictures loading over the course of the next 2-3 minutes. Some load instantly if they're popular enough for your node or one of its immediate neighbors to have a copy already. If you haven't tried it recently, check out the current build. Be patient, and give it a couple hours of uptime to get thoroughly integrated, but it's way better than it once was.
If you haven't used one of the recent builds, you should try it again. Currently I have it set to use a max of 18KiB/s outbound bandwidth (I'm on a somewhat slow connection), and CPU usage varies from about 5% to 25% on a 1.2GHz Athlon. Memory usage is under 100MB. Both CPU usage and memory will grow with increased bandwidth usage, but it's way way better than it used to be.
To good approximation, it also has very little penetrating power (a mm or two at most, I imagine, though it would depend on the frequency) -- the pulse is not long enough for anything to get out of the way. So it deposits a fair bit of energy in a very, very small volume, resulting in a tiny ball of very hot mostly steam, which might as well be an explosion. Heat transfer from that to the surrounding environment will be very poor (time scale is too short), but the blast effect will be significant. The blast would probably actually destroy a larger part of your finger than the energy could vaporize, but I don't know how to estimate the exact effects.
Windows "Just works" so that's what they'll go back to.
Oddly enough, my experience is the opposite. The last time I had to set up windows, I had to go find and install a decent web browser (as opposed to just typing "apt-get install firefox" and being done with it, I had to go find the web site, download the package, install the package...). I had to locate the correct printer driver, which involved moving furniture so I could read the exact model number off the back. It kept asking me whether I wanted to allow the program I had just clicked on to do the thing it is supposed to do. It would sit there waiting for ages doing simple thing like copying or deleting files. Those few settings I really wanted to change were in non-obvious places. Several reboots were involved in this process. And they call this a modern OS? I wasn't even doing a clean install, it came on the machine!
I suppose it's just what you're used to. Windows doesn't "just work" any more than Linux does. I honestly have no clue which one is closer to "just working", but in my experience, I find Linux simpler and easier to get to do what I want it to do. Yes, I'm aware that's because I'm used to it and I'm not used to Windows, but I think the reverse is true as well -- people only find Windows easy because they're used to it.
Got any idea what shape it would be? It's a micron-diameter spot, and it's too short to penetrate (the stuff doesn't get out of the way).
Assume a spherical cow...
I made a simplifying assumption to give readers a feel for how large a chunk of stuff we're talking about. And besides, quasi-spherical might actually be pretty close to reasonable...
17 joules is plenty of energy to detect without special instruments. It's enough to vaporize about 6 mg of water, or in other words it could blast a spherical hole about 1/16" diameter in your finger -- which I think you'd notice. Granted, it's not a lot of energy, but it's not a trivial amount, either.
That this is actually quite appropriate. Since he didn't have any fiduciary duty, the SEC shouldn't take his money away. That said, since it's profit from an illegal act, I would hope that the money would be taken away -- if and when he is convicted for the crime of stealing the data.
Too often in this country we seem to be throwing every law available at people and making up new ones to go with them, when the acts we're trying to punish are already illegal. If he didn't break securities laws, he shouldn't be punished under them. Since he did (we assume, but it hasn't yet been proven) break unauthorized access laws, he should be punished under those.
We don't need more laws against things that are already illegal, and we don't need to make a mockery of existing laws by applying them to things they don't apply to. On a related note, why do we need separate "identity theft" and "atm card fraud" laws, when anyone being charged with them is already also being charged with uttering false instruments and fraud? Our legal code needs to be smaller and simpler; making it so would make it more effective and efficient, not less.
And, at least for protons (as opposed to larger atomic nuclei), something out there has anything we can build beat by a wide, wide margin...
As in, computers aren't involved in artificial intelligence? Or that they aren't computers as in Turing-equivalent machines? Or what? What's the alternative? I'm highly skeptical that there's anything magical about intelligence or our brains that can't be simulated by enough understanding of what's going on combined with raw cpu horsepower.
Do you think that it is possible to demonstrate intelligence through a relatively restrictive interface that computers can handle today -- for example, a purely text conversation? If so, what is required, both for intelligence and for something that counts as "a start"? Historically, a lot of AI detractors seem to play the moving goal posts game -- defining any problem that computers can't solve well as necessary for intelligence (though not necessarily sufficient). "Computers aren't really intelligent; they can't comprehend a game like chess." "Computers aren't intelligent, they can't write music." Yet computers get slowly better at those specific problems, even if one general program can't attack all of them. So what makes your position not of that genre? (I'm not trying to be accusatory, I'm just not seeing it -- I suspect you have and answer, and I'm curious.)
I would argue no - all the examples you've mentioned are solutions created by humans executed on computer hardware.
Take the case of a program that learns the rules of the game its playing based on feedback only, without external training. Is that not a case of a single program solving a class of problems, without specific instruction? If not, why not? If it is, how much broader does the class have to get before it counts as the start of a solution? Surely it doesn't have to solve *any* problem at all to count as a start on the path.
I'm inclined to agree that (at least) a few more major breakthroughs are needed. But at the same time, I think we've seen a few glimmers of intelligence already.
So where would you start?
It seems obvious to me that you won't get anywhere, ever, if you attempt to tackle the entire problem at once. You have to start with a piece of it, and expand that.
We have computer programs that play chess well. We also have computer programs that can figure out the rules to the game they're playing without explicit teaching, and then play somewhere between horribly and well, depending on the details. That seems a step in the right direction -- go from a very specific case to a more general one.
We have computer programs that can check mathematical proofs, and there have been programs that created proofs from nothing on their own. Is that not a step in the right direction?
A lot of the role of the human in using the machine as a tool is in converting the outside world into a form the computer can interact with. If that's not a reasonable role for the human in developing the very beginnings of AI technology, I don't know what is.
I agree we're far, far away from where some people would like us to be or think we are, and that we have a long way to go. But I ask simply this: where would you start? And haven't a few of the very first baby steps been taken already? We're not far along, but I really don't think we're at zero either.
Actually, long-haul lines are all alumium these days. The resistivity is slightly higher, so the lines are thicker, but the aluminum costs less and weighs less for the same load capacity as the copper.
Your point is still completely correct, though.
It's the frivolous stuff that actually hits a lot of people that gets a lot of people angry enough to do something. The big things that only hit a few people are easier to sell to the masses, since it's always someone else being affected. But when everyone is affected...
If horror movies is the only thing hit, I predict nothing will come of it. But if this is the start of a trend, then something very interesting might be about to happen.
Revolutions have been started over taxes on tea...
Interestingly enough, this is actually quite close to how Freenet works in its opennet mode. The turnover rate is probably rather low, but it has no non-encrypted protocol header and is constantly connecting to new nodes. With some tweaking it would be very hard to detect. IIRC it also already runs entirely over UDP, not TCP, which makes injecting RST packets impossible.
Except for two important points: first, what little data I've seen suggests such collisions won't change the orbit of either object or its remains much at all. If you have references, I'd love to see them -- they're hard to find. And secondly, without a circularization burn, any piece of debris will end up on an orbit that intersects the collision point. And since the collision point is very low, that means the orbit will decay rapidly, even if it is highly elliptical with a high-altitude apogee. So even in the worst case, it won't pose a hazard to other satellites for very long. The real problem with the Chinese test is that the debris isn't going away.
Did you pay any attention to the last 30 years or so of cryptography? Any peer-to-peer patch distribution system would use digital signatures that are difficult to fake. The corresponding public keys would be distributed with the OS install or through some other secure mechanism (SSL from the main update site or similar). Any attacker that can install their own key could install a worm through that route anyway.
P2P is quite good at solving intermittent high demand distribution problems, and is quite well matched to this.
Plenty of fuel-tank sized bits of satellites have survived uncontrolled reentry, when protected by the other bits of satellite. Propellant tanks are actually fairly good candidates for surviving largely intact -- ie, still in one piece and full of hydrazine, but broken enough to be leaking hydrazine everywhere. Also, it's surprisingly difficult to overstate the nastiness of hydrazine. It's flammable, explosive, corrosive, toxic, mutagenic, carcinogenic, and teratogenic -- all at very, very high potency. At least it's not radioactive. Monomethylhydrazine (which is widely used as a fuel in applications just like this; I don't know which variety of hydrazine is in use here) is one of the most potent carcinogens known -- a single drop applied to the skin of lab rats caused cancer in about 90% of cases in one study (more than that and the toxic effects dominated).
Seriously. The propellant tanks might or might not survive, but some large bits definitely will make it to the ground, and they might well contain hydrazine or have hydrazine on them from leaking / disintegrating tanks. If they do, you really, really, *really* wouldn't want to touch it. Souvenir gathering from crashed satellites (or Space Shuttles) can be very bad for your health.
To be fair, the space junk isn't equivalent -- the junk from a satellite that's about to reenter will also reenter promptly, whereas the junk from a satellite in a high orbit will remain in a high orbit. The impact won't actually alter the orbital parameters of the junk as much as you might expect; nearly all of it will reenter promptly, and I'd be surprised if any of it managed to get high enough to present a danger to other satellites (the satellite in question is well below normal operating altitudes).
Of course, I'm not trying to say the US isn't guilty of hypocrisy -- just that this case isn't as bad as you make it out to be.
Because that's a lot harder. If all they wanted was for it to deorbit, they wouldn't be doing anything -- the satellite in question never reached its final orbit, and is rapidly decaying. It will reenter fairly soon if left alone. Presumably they're trying to prevent it from reentering where Russia or China or someone might find the pieces and get clues about our capabilities -- it's pretty clearly a recent-generation spy satellite. If they are careful about how they shoot it down, they may get some control over where it reenters.
Currently, we don't have any (non-classified, anyway) capability to attach an unmanned booster to an arbitrary orbitting satellite. And there isn't time to develop a custom solution before it reenters as the current orbit decays.
Of course, the real place these documents should be available, is Freenet. I haven't yet checked to see if they're available, though, so I can't give you a link.