This message can _not_ be deciphered if you make sure that you never reuse the same random noise.
Right - assuming the noise is random. Your suggested SETI data isn't entirely random, and of course most convenient (pseudo)random number generators (e.g./dev/urandom on Linux, rand() and random() in libc) aren't really random. Just because something looks random doesn't mean it is.
Even if you reuse it it is hard.
Define "hard". If you reuse your random numbers, can't an attacker just take two of your encoded messages, XOR them together (which removes the random numbers from the equation), and start searching the result for XORed combinations of shifted dictionary words?
You said "RPM sucks ass", but you seem to be comparing RPM (a package format and low level package tool) with ubuntu's high level package tools (synaptic and apt, IIRC?). I've been pretty happy using synaptic and apt on top of RPM, and I'm told that yum and up2date aren't bad either (never tried urpmi). I'm sure.deb files would suck ass too if all you had to install them was dpkg.
The speed of sound is the speed at which infinitesimal shockwaves (like sound waves) propagate. However, if a compression shockwave is strong enough that the fluid properties behind it are significantly different from the fluid properties in front of it, then it will propagate (relative to stationary fluid) faster than the speed of sound in front of it. That's why a shockwave doesn't have to be attached to a moving vehicle creating it, even if that vehicle is supersonic - the vehicle's motion influences the fluid between it and the shockwave (not very far, admittedly); it just can't influence fluid in front of the shock.
You're quoting a 6 year old memo; how about pointing to some actual instances of MS using their patents offensively?
Didn't you read my reply the last time you insinuated that Microsoft wouldn't use patents offensively? They've tried to do so at least twice already, once successfully to prevent other programs (even other Windows programs!) from using "their" patented file format, and once unsuccessfully (although that hasn't got them to take the threat off their webpage yet) to try and squeeze money from anyone who wants to format a Windows-compatible filesystem with long filenames.
Why would you ever use a signed int in a voting machine?
Because you suspect another programmer is going to try to commit fraud and you want that fraud to be more obvious to outside observers? That's what I guessed when Travis County's (Austin, TX) preliminary results included negative 2 votes for a writein Presidential candidate, anyway.
Otherwise, what's to stop a hacker from "doubling up" multiple voters onto a single index? You'd have to pair votes with a hash whose source included voter-specific information (full name and exact time of vote would probably be sufficient) along with a random number long enough to prevent anyone malicious from brute forcing it to find out who you voted for.
Any voter can verify that his vote was counted by looking it up with his index, and can prove his vote to a third party by using the signed copy
One word: cameraphone. It's no longer very expensive or obtrusive to take a short video of yourself casting your vote. Blackmailing or bribing someone into recording their vote isn't as obvious (or as cheap) a hole as getting them to reveal their receipt and key, but it's already there.
Isn't that exactly the opposite of Libertarian philosophy? If you don't think men are noble enough to govern their own lives, how can you possibly expect them to successfully govern each other?
The 15000 pounds figure is actually the low end of the estimate for one of their three operation modes - the engine can also be used just to supply electrical power for ion drives (much lower thrust, more efficient) or with an "afterburner" of LOX added to the hydrogen (45,000-60,000 pounds thrust, less efficient).
60,000 pounds thrust isn't nearly enough to be useful as a launch vehicle engine either, though. The only possible use in ground-to-orbit missions might be as an orbital tug: a chemical rocket puts your satellite in LEO; the nuclear rocket docks with it, moves it to GEO (or whereever else you want), then comes back for the next satellite. Even that's kind of a stretch.
I must have a very dull imagination, though: since there's already uranium scattered over Texas, the most exciting thing I can think of is that some uranium fuel might fall on someone and smash their house. I can't even figure out how that one would happen, either, since nobody would be bringing the fuel back down (when the reentry vehicle might fail over populated areas), they'd just be sending it up (when the launch vehicle might fail over ocean).
Not disagreeing with what you're saying though -- the equation is fundamental mathematics, independent of the physical universe, it doesn't make sense to imagine an "alternative universe" where it doesn't apply.
You can imagine an "alternative universe" where the math is true, but doesn't correspond to anything physical. In our universe, for example, if you're traveling 2 km/h and you throw a ball ahead of you at 2 km/h, relativity says the ball will be moving at 3.9999 (add another couple dozen 9's) miles per hour. Make those effects more obvious and make all the other laws of physics that weird and perhaps you'd get a universe whose inhabitants think of "2 + 2 = 4" as just a neat theoretical abstraction.
It's just the consequence of the way that exponentiation on complex numbers is defined.
Exponentiation on complex numbers is defined to be the analytic extension of exponentiation on real numbers. The power series definition looks arbitrary, but it isn't - any different definition of complex exponentiation either wouldn't be differentiable everywhere or wouldn't be the same as real exponentiation.
Systems like Condorcet's Method voting are technically superior but use a lot of math and are complicated to explain. If you can't explain it in a thirty second sound bite you won't get able to get enough popular support to get it passed.
Rank your candidates in order of preference, just like IRV. You are allowed to have ties.
If a candidate would beat any other candidate in a one-on-one race, that candidate will win.
If there is a group of candidates such that any candidate in the group would beat any candidate outside the group in a one-on-one race, then a candidate in that group will win.
That's about 20 seconds. (10 seconds if you leave out the last sentence).
I agree that IRV should make the process of switching to Condorcet simpler, though, and at least it's better than plurality.
First, voting for a Republican or Democrat candidate when you're not in a swing state is really a wasted vote, because your vote won't change anything. Any Texan who votes for Bush or Kerry won't make any difference in the world, because Bush will carry Texas.
As a fellow Texan voting for Badnarik, I've couldn't agree more.
If there are enough liberals who don't think the Democrat party isn't liberal enough, they will vote for Nader, possibly forcing the Democrats to lose. That's what happened in Florida in 2000. In theory, the Democrats will then be forced to become more liberal, i.e., "truer" to their party platform.
What theory is this? The only game theory scenarios I've seen predict the opposite:
Suppose the "liberalness" of people and candidates was ranked on a percentile scale from 1-100, and each voter votes for a candidate who comes closest to himself on the scale. If the members of the Democratic party occupy the 50-100th percentiles on that scale, then they'll want to nominate a 75th percentile candidate to optimize the happiness of their own party, but to prevent that candidate from being defeated by a Republican they'll have to nominate a 51st percentile candidate instead - whoever gets closest to 50th percentile wins.
If a 95th percentile Green candidate enters the picture, though, he leaves the Democrats unable to nominate anyone who won't be beaten by either the Green or by even a relatively non-moderate Republican. If everyone "votes their conscience", then the Republicans can now nominate a 30th percentile candidate and be certain that he'll win. Even if only the 94th through 100th percentiles vote Green, the Republicans can nominate a 47th percentile candidate and be certain he'll win - and the Democrats would have to nominate someone equally conservative (i.e. more conservative than they could have gotten away with without a Green in the race) to even have a chance.
Now, this is obviously oversimplified (I really should know better than to try and map political beliefs on a 1-D scale in front of another libertarian), but more complete models (so long as they use plurality voting) don't make things look much better. If a third party takes a few votes away from your extremes, your possible reactions are to become more extreme (and get a few extreme votes back at the cost of a lot more moderate votes) or become more moderate (exactly the opposite of what the third party would have wanted).
The best way third parties can currently use their leverage to pull the major parties in correct directions is if they dangle "we'll drop out of the race" (or "we'll tell our supporters to vote for you") carrots in exchange for campaign promises or party platform concessions. Telling the major parties "become more like us and you might get a few percent of our voters" just isn't a good deal; only telling them "become more like us and you might get all our voters" might be attractive enough to work.
Saying so leaves a nasty taste in my mouth, though. If I might put in a word for my own idealism: the correct solution to this problem isn't for third parties to figure out how to work around a broken voting system, it's for third parties (and non-zealots from the major parties) to replace that system with something (Condorcet or Approval voting) that isn't so broken.
We can get almost all of the practical advantages of Condorcet with approval voting, or do you have solid counter-examples that show a clearly unacceptable result under AV that would have been avoided with Condorcet?
Here's a clear counterexample: If voter preferences are:
97% A,B,C 2% C,B,A 1% A,C,B
And if every voter votes as you suggested to me (approval for their first and second favorite candidates), then candidate B wins with 99% approval (vs. 98% approval for candidate A) despite the fact that 98% of the public would have preferred A to B. That's a counter-example (and a pretty strong one IMHO) for AV vs. IRV, AV vs. Condorcet, and AV vs. plurality.
Question: which weighs more, the heat shielding and structure required to survive re-entry at orbital velocities or the fuel required to brake then re-enter at a low velocity?
The fuel weighs more, by far. You'd need as much fuel to get out of orbit without aerobraking as you needed to get into orbit in the first place.
(Another question, at high altitudes, does the atmosphere rotate with the Earth?)
Any given acceleration requires the same amount of force no matter how fast you're going. F = ma.
That would be true if it wasn't for gravity and aerodynamics adding to F. Imagine a spaceplane with wings and with engines that can indefinitely deliver 1G of acceleration: If it tried to launch straight up, it would never make it off a launch pad, but taking off from a runway it could reach orbit, because it's lift to drag ratio (even hypersonically) could be much larger than 1.
This doesn't apply to the shuttle, though; the shuttle's L/D ratio is larger than 1, but the L/D for the stack as a whole is pretty much zero.
Don't be too hard on Gates. There will always be people whose goal in life is to make more money, by any means that works. The problem is that our society has lost the checks and balances that used to constrain people like him.
Your post sounds like part of the problem, then - I thought being hard on people like him was one of those checks and balances!
If you vote approval for Nader, disapproval for Gore and Bush, then your vote has no effect on the election between Gore and Bush, despite the fact that you have a preference for Gore. If you vote approval for Nader and Gore, disapproval for Bush, then your vote has no effect on the election between Nader and Gore, despite the fact that you have a preference for Gore. Either way, you're throwing away part of your vote. You're not forced to throw away as much as with a plurality system, but you're still unable to express all your preferences, and that inability can change the election outcome for the worse.
If your preferences are really as simple as "I approve of these candidates, and disapprove of the rest" then you can express that in a Condorcet vote by giving the approved set a tie for first place. If your preferences aren't that simple, then with Condorcet you don't have to worry about how to simplify them.
Approval voting is certainly a step up from what we've got now, but to many voters ranked voting would be quite natural (who's your favorite, second favorite, third favorite, etc. are easy questions to answer) whereas approval voting would require the additional consideration of "Where in that rank do I draw the line between approval and disapproval?" and would still encourage tactical voting: I disapprove of both Kerry and Bush, but in a close race between them I'd probably approve of Kerry to avoid "throwing my vote away", which would be just as bad as the current system.
Finally, it's perverse that the LP would be complaining about this stuff. You can expect more of this type of thing to result from the deregulation and property rights the LP stands for. In their ideal world, the owners of property and capital get to make all kinds of decisions that affect all of us, including which political viewpoints to write about.
Is the Libertarian Party asking the government to step in and tell the media how to allocate their political coverage? Libertarianism doesn't mean that consumers of a bad product (including bad news) have no recourse, it just means that they have to get producers to change via market pressure instead of government laws. Trying to encourage that market pressure by publicizing and complaining about flaws is a perfectly libertarian thing to do.
And you don't even have to rank all of them precisely: you could vote for Badnarik as your first place choice, Kerry tied with Cobb as your second, Nader as your third, and everybody else tied for last.
Slashdot Mirror
If it were possible to factor any prime in one month
Any prime's factors are itself and one.
</smartass>
This message can _not_ be deciphered if you make sure that you never reuse the same random noise.
/dev/urandom on Linux, rand() and random() in libc) aren't really random. Just because something looks random doesn't mean it is.
Right - assuming the noise is random. Your suggested SETI data isn't entirely random, and of course most convenient (pseudo)random number generators (e.g.
Even if you reuse it it is hard.
Define "hard". If you reuse your random numbers, can't an attacker just take two of your encoded messages, XOR them together (which removes the random numbers from the equation), and start searching the result for XORed combinations of shifted dictionary words?
You said "RPM sucks ass", but you seem to be comparing RPM (a package format and low level package tool) with ubuntu's high level package tools (synaptic and apt, IIRC?). I've been pretty happy using synaptic and apt on top of RPM, and I'm told that yum and up2date aren't bad either (never tried urpmi). I'm sure .deb files would suck ass too if all you had to install them was dpkg.
One correction:
The speed of sound is the speed at which infinitesimal shockwaves (like sound waves) propagate. However, if a compression shockwave is strong enough that the fluid properties behind it are significantly different from the fluid properties in front of it, then it will propagate (relative to stationary fluid) faster than the speed of sound in front of it. That's why a shockwave doesn't have to be attached to a moving vehicle creating it, even if that vehicle is supersonic - the vehicle's motion influences the fluid between it and the shockwave (not very far, admittedly); it just can't influence fluid in front of the shock.
You're quoting a 6 year old memo; how about pointing to some actual instances of MS using their patents offensively?
Didn't you read my reply the last time you insinuated that Microsoft wouldn't use patents offensively? They've tried to do so at least twice already, once successfully to prevent other programs (even other Windows programs!) from using "their" patented file format, and once unsuccessfully (although that hasn't got them to take the threat off their webpage yet) to try and squeeze money from anyone who wants to format a Windows-compatible filesystem with long filenames.
Why would you ever use a signed int in a voting machine?
Because you suspect another programmer is going to try to commit fraud and you want that fraud to be more obvious to outside observers? That's what I guessed when Travis County's (Austin, TX) preliminary results included negative 2 votes for a writein Presidential candidate, anyway.
Otherwise, what's to stop a hacker from "doubling up" multiple voters onto a single index? You'd have to pair votes with a hash whose source included voter-specific information (full name and exact time of vote would probably be sufficient) along with a random number long enough to prevent anyone malicious from brute forcing it to find out who you voted for.
Any voter can verify that his vote was counted by looking it up with his index, and can prove his vote to a third party by using the signed copy
One word: cameraphone. It's no longer very expensive or obtrusive to take a short video of yourself casting your vote. Blackmailing or bribing someone into recording their vote isn't as obvious (or as cheap) a hole as getting them to reveal their receipt and key, but it's already there.
But you know, if I was one of the Ohio voters who got challenged into submitting a "provisional" vote yesterday, I would probably be pissed right now.
"Badnarik '04: Men Are Angels!"
Isn't that exactly the opposite of Libertarian philosophy? If you don't think men are noble enough to govern their own lives, how can you possibly expect them to successfully govern each other?
The 15000 pounds figure is actually the low end of the estimate for one of their three operation modes - the engine can also be used just to supply electrical power for ion drives (much lower thrust, more efficient) or with an "afterburner" of LOX added to the hydrogen (45,000-60,000 pounds thrust, less efficient).
60,000 pounds thrust isn't nearly enough to be useful as a launch vehicle engine either, though. The only possible use in ground-to-orbit missions might be as an orbital tug: a chemical rocket puts your satellite in LEO; the nuclear rocket docks with it, moves it to GEO (or whereever else you want), then comes back for the next satellite. Even that's kind of a stretch.
I must have a very dull imagination, though: since there's already uranium scattered over Texas, the most exciting thing I can think of is that some uranium fuel might fall on someone and smash their house. I can't even figure out how that one would happen, either, since nobody would be bringing the fuel back down (when the reentry vehicle might fail over populated areas), they'd just be sending it up (when the launch vehicle might fail over ocean).
Not disagreeing with what you're saying though -- the equation is fundamental mathematics, independent of the physical universe, it doesn't make sense to imagine an "alternative universe" where it doesn't apply.
You can imagine an "alternative universe" where the math is true, but doesn't correspond to anything physical. In our universe, for example, if you're traveling 2 km/h and you throw a ball ahead of you at 2 km/h, relativity says the ball will be moving at 3.9999 (add another couple dozen 9's) miles per hour. Make those effects more obvious and make all the other laws of physics that weird and perhaps you'd get a universe whose inhabitants think of "2 + 2 = 4" as just a neat theoretical abstraction.
It's just the consequence of the way that exponentiation on complex numbers is defined.
Exponentiation on complex numbers is defined to be the analytic extension of exponentiation on real numbers. The power series definition looks arbitrary, but it isn't - any different definition of complex exponentiation either wouldn't be differentiable everywhere or wouldn't be the same as real exponentiation.
For instance, the developers of "Deus Ex: Invisible War" managed to overhype its quality and exaggerate its features by first releasing "Deus Ex". ;-)
Systems like Condorcet's Method voting are technically superior but use a lot of math and are complicated to explain. If you can't explain it in a thirty second sound bite you won't get able to get enough popular support to get it passed.
Rank your candidates in order of preference, just like IRV. You are allowed to have ties.
If a candidate would beat any other candidate in a one-on-one race, that candidate will win.
If there is a group of candidates such that any candidate in the group would beat any candidate outside the group in a one-on-one race, then a candidate in that group will win.
That's about 20 seconds. (10 seconds if you leave out the last sentence).
I agree that IRV should make the process of switching to Condorcet simpler, though, and at least it's better than plurality.
First, voting for a Republican or Democrat candidate when you're not in a swing state is really a wasted vote, because your vote won't change anything. Any Texan who votes for Bush or Kerry won't make any difference in the world, because Bush will carry Texas.
As a fellow Texan voting for Badnarik, I've couldn't agree more.
If there are enough liberals who don't think the Democrat party isn't liberal enough, they will vote for Nader, possibly forcing the Democrats to lose. That's what happened in Florida in 2000. In theory, the Democrats will then be forced to become more liberal, i.e., "truer" to their party platform.
What theory is this? The only game theory scenarios I've seen predict the opposite:
Suppose the "liberalness" of people and candidates was ranked on a percentile scale from 1-100, and each voter votes for a candidate who comes closest to himself on the scale. If the members of the Democratic party occupy the 50-100th percentiles on that scale, then they'll want to nominate a 75th percentile candidate to optimize the happiness of their own party, but to prevent that candidate from being defeated by a Republican they'll have to nominate a 51st percentile candidate instead - whoever gets closest to 50th percentile wins.
If a 95th percentile Green candidate enters the picture, though, he leaves the Democrats unable to nominate anyone who won't be beaten by either the Green or by even a relatively non-moderate Republican. If everyone "votes their conscience", then the Republicans can now nominate a 30th percentile candidate and be certain that he'll win. Even if only the 94th through 100th percentiles vote Green, the Republicans can nominate a 47th percentile candidate and be certain he'll win - and the Democrats would have to nominate someone equally conservative (i.e. more conservative than they could have gotten away with without a Green in the race) to even have a chance.
Now, this is obviously oversimplified (I really should know better than to try and map political beliefs on a 1-D scale in front of another libertarian), but more complete models (so long as they use plurality voting) don't make things look much better. If a third party takes a few votes away from your extremes, your possible reactions are to become more extreme (and get a few extreme votes back at the cost of a lot more moderate votes) or become more moderate (exactly the opposite of what the third party would have wanted).
The best way third parties can currently use their leverage to pull the major parties in correct directions is if they dangle "we'll drop out of the race" (or "we'll tell our supporters to vote for you") carrots in exchange for campaign promises or party platform concessions. Telling the major parties "become more like us and you might get a few percent of our voters" just isn't a good deal; only telling them "become more like us and you might get all our voters" might be attractive enough to work.
Saying so leaves a nasty taste in my mouth, though. If I might put in a word for my own idealism: the correct solution to this problem isn't for third parties to figure out how to work around a broken voting system, it's for third parties (and non-zealots from the major parties) to replace that system with something (Condorcet or Approval voting) that isn't so broken.
If it ever does become sentient, it'll probably classify it's own creators as a terrorist organization and end up working with the good guys.
We can get almost all of the practical advantages of Condorcet with approval voting, or do you have solid counter-examples that show a clearly unacceptable result under AV that would have been avoided with Condorcet?
Here's a clear counterexample: If voter preferences are:
97% A,B,C
2% C,B,A
1% A,C,B
And if every voter votes as you suggested to me (approval for their first and second favorite candidates), then candidate B wins with 99% approval (vs. 98% approval for candidate A) despite the fact that 98% of the public would have preferred A to B. That's a counter-example (and a pretty strong one IMHO) for AV vs. IRV, AV vs. Condorcet, and AV vs. plurality.
Question: which weighs more, the heat shielding and structure required to survive re-entry at orbital velocities or the fuel required to brake then re-enter at a low velocity?
The fuel weighs more, by far. You'd need as much fuel to get out of orbit without aerobraking as you needed to get into orbit in the first place.
(Another question, at high altitudes, does the atmosphere rotate with the Earth?)
Yes.
Any given acceleration requires the same amount of force no matter how fast you're going. F = ma.
That would be true if it wasn't for gravity and aerodynamics adding to F. Imagine a spaceplane with wings and with engines that can indefinitely deliver 1G of acceleration: If it tried to launch straight up, it would never make it off a launch pad, but taking off from a runway it could reach orbit, because it's lift to drag ratio (even hypersonically) could be much larger than 1.
This doesn't apply to the shuttle, though; the shuttle's L/D ratio is larger than 1, but the L/D for the stack as a whole is pretty much zero.
Don't be too hard on Gates. There will always be people whose goal in life is to make more money, by any means that works. The problem is that our society has lost the checks and balances that used to constrain people like him.
Your post sounds like part of the problem, then - I thought being hard on people like him was one of those checks and balances!
If you vote approval for Nader, disapproval for Gore and Bush, then your vote has no effect on the election between Gore and Bush, despite the fact that you have a preference for Gore. If you vote approval for Nader and Gore, disapproval for Bush, then your vote has no effect on the election between Nader and Gore, despite the fact that you have a preference for Gore. Either way, you're throwing away part of your vote. You're not forced to throw away as much as with a plurality system, but you're still unable to express all your preferences, and that inability can change the election outcome for the worse.
If your preferences are really as simple as "I approve of these candidates, and disapprove of the rest" then you can express that in a Condorcet vote by giving the approved set a tie for first place. If your preferences aren't that simple, then with Condorcet you don't have to worry about how to simplify them.
Approval voting is certainly a step up from what we've got now, but to many voters ranked voting would be quite natural (who's your favorite, second favorite, third favorite, etc. are easy questions to answer) whereas approval voting would require the additional consideration of "Where in that rank do I draw the line between approval and disapproval?" and would still encourage tactical voting: I disapprove of both Kerry and Bush, but in a close race between them I'd probably approve of Kerry to avoid "throwing my vote away", which would be just as bad as the current system.
Finally, it's perverse that the LP would be complaining about this stuff. You can expect more of this type of thing to result from the deregulation and property rights the LP stands for. In their ideal world, the owners of property and capital get to make all kinds of decisions that affect all of us, including which political viewpoints to write about.
Is the Libertarian Party asking the government to step in and tell the media how to allocate their political coverage? Libertarianism doesn't mean that consumers of a bad product (including bad news) have no recourse, it just means that they have to get producers to change via market pressure instead of government laws. Trying to encourage that market pressure by publicizing and complaining about flaws is a perfectly libertarian thing to do.
And you don't even have to rank all of them precisely: you could vote for Badnarik as your first place choice, Kerry tied with Cobb as your second, Nader as your third, and everybody else tied for last.