So someone in Ukraine uses a carded CC number to buy his game, then cracks and uploads it to TPB. The developer (more likely, the publisher) downloads. "Gotcha, Sergei Konovalov! You evildoer!" they say. But Sergei either doesn't exist at all, or merely lost his CC in the latest round of phishing.
Oh, and since what you're proposing is essentially a watermark, do note that watermarks are vulnerable to the collusion attack, wherein the attackers "average out" the watermark by looking at the differences between multiple copies to discern the structure of the watermark, then apply noise (in your scheme, shuffling the functions around) to remove the watermark itself.
You have to have a run of fiber directly between hither and yon for communications to be secure.
One can use quantum teleportation to build quantum repeaters. Add an out of band signal for addressing, and it shouldn't be that hard to make a quantum encrypted network - direct connection not required. If men in the middle rearrange the addressing data, the only thing that happens is that the signal doesn't arrive at the correct destination; the adversaries still can't clone the particle, so they still can't break quantum crypto.
A lumen is a candela per steradian, so highly directional lights would have lower lumen ratings. However, lumens say nothing about the color distribution of the light source. It's weighted so that the light that's easiest to see gets counted to a greater degree than light that's not so easy to see, but in the extreme, that means the most efficient type of light would emit 555 nm monochromatic light - hardly a pleasing light source.
Amazon et al should use a trust metric, preferably one that deals gracefully with attempts to manipulate it. Perhaps something like Advogato's metric could be used, or the manipulation-resistant versions of EigenTrust. What metric one may use, it would help decreasing spammers' powers, since they would presumably not be able to integrate themselves as thoroughly into the system, and definitely not do so in the kind of en masse, flooding, way that traditional spammers make use of.
Chernobyl didn't have a containment dome. It suffered Xenon poisoning because of the low power levels the test demanded, the technicians turned all the safety measures off to run their test, they pulled the control rods out, and it had a positive void coefficient so that the process fed back on itself. See the Hyperphysics page here. To quote from that page: "It was like airplane pilots experimenting with the engines in flight" (Legasov).
Okay, so human error will always happen, right? That's why newer plants fail safe. All western plants have containment domes, and newer plant types (such as pebble bed reactors or Generation III reactors like the AP1000) are passively safe, which means that even if all the coolant is removed, nothing happens. Also, most plants automatically scram (insert control rods to maximum positions) on error.
Now let's look at Three Mile Island. A problem in an unrelated system caused the primary feedwater pump to fail. The reactor automatically went to scram as designed (thus showing what I said in the last paragraph). However, a valve that would vent steam caused by heating the water by decay heat (since radioactive decay still happens even when no fission is going on), failed to close, and the monitoring systems did not show clearly enough that it was indeed open long after it should be.
The result was that the reactor in question (TMI-2) was severely damaged and some radioactive Krypton was released. What danger did this entail? To quote the Merck manual, "the Three Mile Island accident did not result in major radiation exposure; in fact, anyone living within 1 mile of the plant received only about 0.08 mSv additional radiation". As a comparison, a chest x-ray is between 0.05 and 0.1 mSv.
Solar might be more safe, but it also occupies a great deal of space and is much more expensive. Fossil fuel plants pollute and for coal in particular, there are mining accidents; since a given amount of coal provides much less energy than a given amount of uranium, a lot more has to be mined for the same amount of energy. Chernobyl was the Bhopal of nuclear power, but we don't stop making pesticides just because of Bhopal, and so we shouldn't stop nuclear power just because of Chernobyl either, but instead take the proper precautions and engineer the systems to be safe.
They can also be hacked, which is also pretty neat if you're the hacker, but not if you're trying to build an infrastructure based on the cards.
Come to think of it, Chaum's electronic money (digital cash), especially the off-line anonymous variants, would be very well suited to the kind of mobile payments discussed in the article; and such a solution would preserve all the important properties of "ordinary" cash.
It's no surprise. If you need to show that the market is the Supreme Ultimate form of economic organization, and first mover effect stands in your way, well guess what... you try to that the first mover effect is false, whether that be Dvorak or MS-DOS. The actual examples are just means to an end.
Not if it's an Energy Amplifier. Most likely, these are not, though, so what the summary intended to say was "prompt critical", not plain old critical.
That err, "feature", is called the User Operation Prohibition flag. Some DVD players can be patched to disregard the UOP, others disregard the UOP by default. Do a web search if you're interested... I note it's also considered DRM, which just shows exactly whose "rights" are being preserved here.
It shouldn't be that difficult to make a JavaScript version of ADVENT ("Colossal Cave Adventure"). An Inform source to JS compiler... that might be harder, but not impossible.
If anything is simple enough for formal verification to work, yet important enough that formal verification should be used, surely voting machines must be it. Of course, if they're really doing this out of a sinister motive, then (to them) there's no point.
Go ahead, make pirated versions even Better Than Original. If you can flip a JNE to JE and bypass protection, it should be no problem to just jump over the "render ads to screen" or "download ads from server and save to file" function.
True, modern DRM is a bit more difficult than flipping JNE to JE, but that just goes in the favor of the pirates; the ad-download function can't be more difficult than the DRM, and they're already quite able to remove the DRM... So, yeah, publishers, go ahead and compete yourself out of the market.
In order to be a simulation of everything, it has to contain a simulation of itself, too; and then it has to contain a simulation of a simulation of itself... infinite descent! Only Microsoft would think they could manage that in finite space:-)
It's not that hard to make the GA construct a program that always compiles; you just have to use a functional approach. Construct a program as a tree of functions. A function takes a certain number of parameters and gives an output (and ignores excess parameters). If you want modularization, you could have an alias operator, where if a function has an alias root, it's not called, and the main tree substitutes that function whenever it sees an alias reference operator among its nodes.
For more information, see the chromosome representation section of the Wikipedia article on genetic programming, here. If you know Standard ML, ADATE may also be of interest.
For any complex problem, the devil's in the details. Do you use tournament selection, roulette wheel, or stochastic sampling? Objective GA or Pareto coevolution? One-point, two-point, or random crossover? Continuous-valued or binary genes? And so on... and we haven't even got to the population/mutation parameters.
"Exploding a series of bombs under the rocket" actually works. It's called Project Orion, and you feasibly could send city-sized ships to Saturn with it. Later proposals used conventional explosives for the first blast, so there would be little fallout. Unfortunately, the thing's illegal (by the Partial Test Ban Treaty of 1963). The fallout would also kill ten people (globally) per launch, on average.
So that leaves internal engines. Yes, it's possible to make a nuclear rocket. It can either be solid core, liquid core, or gas core.
Solid core is your basic nuclear reactor mounted to a rocket. Heat the reactor up, pass hydrogen (or some other gas) through it, it heats up and goes faster out the other end. The problem is abrasion, and there's also a temperature limitation - if you run it too hot, the reactor melts and that's no good. The temperature limitation means that the rocket's limited to about 1000 seconds of specific impulse.
A liquid core reactor takes that lemon and makes lemonade. The reactor can't melt because it's already molten. With some sophistication, it's possible to make the reactor stable (spinning the mass in some way; I know too little about the details). Put hydrogen in one end, get superheated hydrogen out the other. But now that hydrogen drags with it some uranium too, and you get fallout galore.
The gas core rocket exists in two forms. The open gas core reactor is like the liquid core reactor, except now the uranium is gaseous. Even more uranium mixes with the hydrogen. This leads us to the closed gas core reactor, where the nuclear inferno is trapped inside a "light bulb" of quartz, which is almost transparent to the blackbody radiation in the temperature range in question. Still, that reintroduces temperature limits, but they're far more generous than solid core limits. Hydrogen (or the propellant, whatever it is) is introduced so that it flows on the outside of the quartz, gathering heat and speeding up.
According to this nuclear engine page, closed gas core reactors have specific impulse ranges of 1500 to 3000 seconds, and open gas core reactors have specific impulse ranges of 1800 to 7000s. In contrast, the Space Shuttle has a specific impulse of 450 seconds, and a jet about 2000 seconds. Note that I've only mentioned high-thrust engines; it's possible to make ion engines with extremely high specific impulse, but they'll be no good getting you to orbit, and it takes forever to accelerate in space using such engines.
So price floors are good, but price ceilings are bad? As we all know, "only commies allow price ceilings", so this sounds a lot like socialism for the rich and capitalism for the poor.
Slashdot Mirror
So someone in Ukraine uses a carded CC number to buy his game, then cracks and uploads it to TPB. The developer (more likely, the publisher) downloads. "Gotcha, Sergei Konovalov! You evildoer!" they say. But Sergei either doesn't exist at all, or merely lost his CC in the latest round of phishing.
Oh, and since what you're proposing is essentially a watermark, do note that watermarks are vulnerable to the collusion attack, wherein the attackers "average out" the watermark by looking at the differences between multiple copies to discern the structure of the watermark, then apply noise (in your scheme, shuffling the functions around) to remove the watermark itself.
Since it's still DRM, at some point or other the single-player verification will boil down to:
CALL check_for_legitimacy
TEST EAX, EAX
JNE bad_boy_you_cant_play
In other words, it won't deter crackers, and thus, pirates will still get Better Than Original.
No, that's IF. For that, you're going to need Gargoyle or Frotz.
You have to have a run of fiber directly between hither and yon for communications to be secure.
One can use quantum teleportation to build quantum repeaters. Add an out of band signal for addressing, and it shouldn't be that hard to make a quantum encrypted network - direct connection not required. If men in the middle rearrange the addressing data, the only thing that happens is that the signal doesn't arrive at the correct destination; the adversaries still can't clone the particle, so they still can't break quantum crypto.
As would MIRIAM ABACHA, WIFE OF GENERAL ABACHA, and countless other scamsters.
(Don't use so many caps, it's like yelling. Don't use so many caps, it's like yelling. Blah blah.)
Yeah, and in the middle there's a machine that can turn you into a god... if you can get around Plato's tenfold error!
Storing hydrogen is a pain. How about that wonder metal (metal?) discussed in another article, Boron?
See, this is why Windows is never going to rule the desktop. It doesn't even have a package manager!
If that's true, why don't they release a DRM removal patch a week (or month or whatever) after release?
A lumen is a candela per steradian, so highly directional lights would have lower lumen ratings. However, lumens say nothing about the color distribution of the light source. It's weighted so that the light that's easiest to see gets counted to a greater degree than light that's not so easy to see, but in the extreme, that means the most efficient type of light would emit 555 nm monochromatic light - hardly a pleasing light source.
Amazon et al should use a trust metric, preferably one that deals gracefully with attempts to manipulate it. Perhaps something like Advogato's metric could be used, or the manipulation-resistant versions of EigenTrust. What metric one may use, it would help decreasing spammers' powers, since they would presumably not be able to integrate themselves as thoroughly into the system, and definitely not do so in the kind of en masse, flooding, way that traditional spammers make use of.
Chernobyl didn't have a containment dome. It suffered Xenon poisoning because of the low power levels the test demanded, the technicians turned all the safety measures off to run their test, they pulled the control rods out, and it had a positive void coefficient so that the process fed back on itself. See the Hyperphysics page here. To quote from that page: "It was like airplane pilots experimenting with the engines in flight" (Legasov).
Okay, so human error will always happen, right? That's why newer plants fail safe. All western plants have containment domes, and newer plant types (such as pebble bed reactors or Generation III reactors like the AP1000) are passively safe, which means that even if all the coolant is removed, nothing happens. Also, most plants automatically scram (insert control rods to maximum positions) on error.
Now let's look at Three Mile Island. A problem in an unrelated system caused the primary feedwater pump to fail. The reactor automatically went to scram as designed (thus showing what I said in the last paragraph). However, a valve that would vent steam caused by heating the water by decay heat (since radioactive decay still happens even when no fission is going on), failed to close, and the monitoring systems did not show clearly enough that it was indeed open long after it should be.
The result was that the reactor in question (TMI-2) was severely damaged and some radioactive Krypton was released. What danger did this entail? To quote the Merck manual, "the Three Mile Island accident did not result in major radiation exposure; in fact, anyone living within 1 mile of the plant received only about 0.08 mSv additional radiation". As a comparison, a chest x-ray is between 0.05 and 0.1 mSv.
Solar might be more safe, but it also occupies a great deal of space and is much more expensive. Fossil fuel plants pollute and for coal in particular, there are mining accidents; since a given amount of coal provides much less energy than a given amount of uranium, a lot more has to be mined for the same amount of energy. Chernobyl was the Bhopal of nuclear power, but we don't stop making pesticides just because of Bhopal, and so we shouldn't stop nuclear power just because of Chernobyl either, but instead take the proper precautions and engineer the systems to be safe.
They can also be hacked, which is also pretty neat if you're the hacker, but not if you're trying to build an infrastructure based on the cards.
Come to think of it, Chaum's electronic money (digital cash), especially the off-line anonymous variants, would be very well suited to the kind of mobile payments discussed in the article; and such a solution would preserve all the important properties of "ordinary" cash.
It's no surprise. If you need to show that the market is the Supreme Ultimate form of economic organization, and first mover effect stands in your way, well guess what... you try to that the first mover effect is false, whether that be Dvorak or MS-DOS. The actual examples are just means to an end.
Not if it's an Energy Amplifier. Most likely, these are not, though, so what the summary intended to say was "prompt critical", not plain old critical.
That err, "feature", is called the User Operation Prohibition flag. Some DVD players can be patched to disregard the UOP, others disregard the UOP by default. Do a web search if you're interested... I note it's also considered DRM, which just shows exactly whose "rights" are being preserved here.
It shouldn't be that difficult to make a JavaScript version of ADVENT ("Colossal Cave Adventure"). An Inform source to JS compiler... that might be harder, but not impossible.
If anything is simple enough for formal verification to work, yet important enough that formal verification should be used, surely voting machines must be it. Of course, if they're really doing this out of a sinister motive, then (to them) there's no point.
Go ahead, make pirated versions even Better Than Original. If you can flip a JNE to JE and bypass protection, it should be no problem to just jump over the "render ads to screen" or "download ads from server and save to file" function.
True, modern DRM is a bit more difficult than flipping JNE to JE, but that just goes in the favor of the pirates; the ad-download function can't be more difficult than the DRM, and they're already quite able to remove the DRM... So, yeah, publishers, go ahead and compete yourself out of the market.
In order to be a simulation of everything, it has to contain a simulation of itself, too; and then it has to contain a simulation of a simulation of itself... infinite descent! Only Microsoft would think they could manage that in finite space :-)
It's not that hard to make the GA construct a program that always compiles; you just have to use a functional approach. Construct a program as a tree of functions. A function takes a certain number of parameters and gives an output (and ignores excess parameters). If you want modularization, you could have an alias operator, where if a function has an alias root, it's not called, and the main tree substitutes that function whenever it sees an alias reference operator among its nodes.
For more information, see the chromosome representation section of the Wikipedia article on genetic programming, here. If you know Standard ML, ADATE may also be of interest.
For any complex problem, the devil's in the details. Do you use tournament selection, roulette wheel, or stochastic sampling? Objective GA or Pareto coevolution? One-point, two-point, or random crossover? Continuous-valued or binary genes? And so on... and we haven't even got to the population/mutation parameters.
Their system only lets them vote for Kodos or Kang.
"Exploding a series of bombs under the rocket" actually works. It's called Project Orion, and you feasibly could send city-sized ships to Saturn with it. Later proposals used conventional explosives for the first blast, so there would be little fallout. Unfortunately, the thing's illegal (by the Partial Test Ban Treaty of 1963). The fallout would also kill ten people (globally) per launch, on average.
So that leaves internal engines. Yes, it's possible to make a nuclear rocket. It can either be solid core, liquid core, or gas core.
Solid core is your basic nuclear reactor mounted to a rocket. Heat the reactor up, pass hydrogen (or some other gas) through it, it heats up and goes faster out the other end. The problem is abrasion, and there's also a temperature limitation - if you run it too hot, the reactor melts and that's no good. The temperature limitation means that the rocket's limited to about 1000 seconds of specific impulse.
A liquid core reactor takes that lemon and makes lemonade. The reactor can't melt because it's already molten. With some sophistication, it's possible to make the reactor stable (spinning the mass in some way; I know too little about the details). Put hydrogen in one end, get superheated hydrogen out the other. But now that hydrogen drags with it some uranium too, and you get fallout galore.
The gas core rocket exists in two forms. The open gas core reactor is like the liquid core reactor, except now the uranium is gaseous. Even more uranium mixes with the hydrogen. This leads us to the closed gas core reactor, where the nuclear inferno is trapped inside a "light bulb" of quartz, which is almost transparent to the blackbody radiation in the temperature range in question. Still, that reintroduces temperature limits, but they're far more generous than solid core limits. Hydrogen (or the propellant, whatever it is) is introduced so that it flows on the outside of the quartz, gathering heat and speeding up.
According to this nuclear engine page, closed gas core reactors have specific impulse ranges of 1500 to 3000 seconds, and open gas core reactors have specific impulse ranges of 1800 to 7000s. In contrast, the Space Shuttle has a specific impulse of 450 seconds, and a jet about 2000 seconds. Note that I've only mentioned high-thrust engines; it's possible to make ion engines with extremely high specific impulse, but they'll be no good getting you to orbit, and it takes forever to accelerate in space using such engines.
So price floors are good, but price ceilings are bad? As we all know, "only commies allow price ceilings", so this sounds a lot like socialism for the rich and capitalism for the poor.