Let's make this simple. Don't use webmail. Don't use Yahoo.com, Gmail.com, Hotmail.com, squirrelmail, etc. There are SO many vulnerable access points between the web application and your email that it is almost impossible to secure the entire stack.
The use of Ajax alone (like most major webmail vendors) increases your vulnerability by huge amounts. SOP (same origin policy) is broken. A combination of a reflected XSS attack (which are everywhere http://blogs.zdnet.com/Google/?p=451 ) and a stored XSS attack can completely compromise your session.
Forgetting about XSS, there's still CSRF, injections, RFI, information leakage, broken authentication/session management, insecure url access, etc.
So seriously - unless you trust that your email server has secured every possible hole in every possible layer of their stack, stick to TLS/SSL encrypted imap/pop3/smtp. Now, I'm not saying these are perfect, but email protocols are just simpler. There will always be fewer places to attack and thus the chances of your email being compromised are just smaller.
I ran a high volume, 24/7 uptime file server on FreeBSD for 6 months. I had 2 GB of ram, an Intel Core 2 Quad, 2 GB of RAM, and 5 SATA drives in a 3 GB RAIDZ with close to a terabyte of daily throughput. I would have a kernel panic every two weeks.
The worst bug is this:
Heavy IO activity between ZFS and another file system (like rsyncing between ZFS and UFS or between ZFS and NFS) may result in a deadlock. Symptoms: processes wanting to do IO on a ZFS file system get stuck forever in "zfs" state (WCHAN), other file systems (e.g. UFS) are still working.
I could only handle about 3 days without having this lockup. There is no known fix. ZFS in FreeBSD is EXPERIMENTAL and should not be used for high availability filesystems.
I take two steps forward I take two steps back We come together cuz opposites attract And you know: It ain't fiction Just a natural fact We come together cuz opposites attract
Idk if your doors are the same, but for most of the doors in my house you can actually just pull the pins out of the hinges with a flathead screwdriver and a pair of pliers - takes half the time compared to unscrewing the hinges.
I'm a cryptography student so I'm not well versed in Mersenne prime generation and testing but I seem to remember that there was an optimized FFT called the Discrete Weighted Transform, but that it had issues with actually being implemented in GPUs because of floating point precision, coupled with the memory requirements.
What's really interesting is that I saw something floating around last year by a mathematician, Fuhrer I think? Before the multiplication algorithm with the best algorithmic time was Schoenhage-Strassen, but I think it managed to beat that. It wasn't by much, something like log log n - 2^log*n, so I'm not sure if it would make much of a difference, even with 10 million digit numbers. I'd do the math and/or find the paper for you, but its like 3:30 and I'm goin to sleep.
Wrong. The Internet is federally regulated. Ever heard of the FCC?
You can't have a state controlled power grid. Ever. Eventually you need to cross state lines for something (even if it isn't necessarily power), and there you get Congressional dominion.
. JPEG and PNG will be readable by the included reader device but if it fails irreparably, some future digital archeologists could have a hard time decoding it (JPEG is actually a very clever and CPU-intensive format; we don't notice it now because hardware is so fast)
What? Jesus, I wrote the DCT/iDCT and Huffman coding transformations for a primitive JPEG library when I was still a senior in high school. It's elementary calculus. Unless you foresee some kind of nuclear holocaust (in which case his data is probably useless anyway), go ahead and use JPEG. The only harm with compression is the possibility of excess data loss when cylinders die, etc.
If you really can't figure out JPEG in 25 years, come and look me up - I should still be alive (in my 40s)
Doesn't matter - free will is a human based philosophical concept. It has no meaning in the context of determinism. Whether or not our "free will" is a result of the nondeterministic nature of particles is irrelevant.
Right - but our understanding of quantum mechanics is so rudimentary as to make this discussion worthless. Nondeterminism is the best bet according to our current Schroedinger model, which is the most widely accepted theory.
There is no practical difference between true randomness and sufficiently-pseudo-random deterministic chaos, so this whole free-will / predetermination argument is mutual intellectual masturbation. It doesn't make a difference to us, honestly, so let's stop pretending it important, shall we?
Can't agree more - this discussion is just stupid. We don't know nearly enough to make any debate on determinism vs nondeterminism.
However, the most widely accepted model of QM is the Schroedinger model, which predicts nondeterminism and, until proven otherwise, I'll go by that.
Hey, I'm not saying your arguments don't have philosophic merit, I'm definitely not qualified to evaluate them. In fact, I agree with a lot of what you say. I am only saying that the article is a discussion about determinism and quantum mechanics, not philosophy.
No. You are doing the same thing that I spoke about in my original post - you are using some weird made up semi-scientific and semi-philosophical definition of free will. Free will is a term that has very specific philosophical and ontological definition and it is NOT what we are talking about. Determinism, by definition, states that the observer would not affect the state of the system. Quantum mechanics states the opposite, which is why it is very unlikely that the universe (and, by extension, the human mind) is deterministic.
I completely agree! My position was that the article is not about the philosophical concept of "free will," but of the physics concept of determinism. I only have the barest undergraduate understanding of philosophy but from what I can tell there haven't been many great posts on that, either.
The very description of particles having "free will" underlies my point - the phisophical concept of free will applied to elementary particles wouldn't make any sense, while the question of whether or not a subatomic particle can lie within a deterministic system makes perfect sense (even though it's basically unprovable either way unless you accept the uncertainty principle).
And true random numbers can't be created by a P Turing machine - that doesn't mean PRNGs aren't useful. Many experiments have been used in the real world to test the Uncertainty Principle, which is essentially the basis of the theories of nondeterminism.
Statistics and mathematics allows us to pinpoint measurement and user errors instead of systematic errors, which allows us to present fairly conclusive evidence of determinism or nondeterminism.
Fortunately this debate is worthless because the best explanation has already been supported by a number experiments (Uncertainty Principle).
Mod parent up. Finally someone who knows what they are talking about.
The buzzword "free will" is bringing out the idiots with no science education. This discussion simplifies to one thing - if, given all the requisite variables in a system, one can predict the next infinite states of that system, that system is deterministic. Id est, if, ignoring the cloning theorem and other QM restraints, one knew the exact state of every particle in the human body and one could predict the next infinite states of that system (the body), then that system would be deterministic (have no "free will"). If, on the other hand, the human body (more precisely, the mind) could be proven to have a finite number of predictable states, then the underlying physical systems must therefore also have a finite number of predictable states (be unpredictable).
Now, QM predicts that subatomic particles are unpredictable. Technically, that would make our minds unpredictable HOWEVER - unpredictable is defined precisely as being unable to predict an infinite number of states in the system. A finite (even large) number may still be possible. This would the generalization of a large number of unpredictable subsystems in the system used to approximate the future states. As we see with Newtonian physics, this method can be fairly accurate.
The only way that humans could be proven to be completely predictable would be to disprove the tenets of quantum mechanics. Until then, humans have "free will."
You have to recognize the difference between determinism and extreme variable interference (chaos theory). "Unpredictable determinism" means that, on the large scale, there are too many variables to gain an accurate prediction of the outcome. However, one can always just control a miniature system with all independent variables held constant.
If the smallest possible system with all possible variables held constant and you still can't accurately produce an outcome, well, thats a nondeterministic system.
You are mistaking chaos theory for determinism. Simply because we are unable to produce an accurate mathematical model for something does not mean it is not deterministic. Economics is a profoundly bad example because 99% of the mathematics in practical economics is based on faulty assumptions and insufficient data. Physics and chemistry are much more useful in quantum analysis.
Specifically, I will speak to quantum particles as the GP did - on the subatomic level we can only produce probabilities for particle motion and position.
Mathematically (I wish Slashdot had LaTeX, there is no reason why it shouldn't)
(delta)x * (delta)p ⥠hbar/(4Ï) = hbar/2
while we can cannot exactly place a subatomic particles motion (where it will be in the next second), we can place the motion of a larger object. If your statement "the sum or average of a set of random variables is also a random variable" this would not be possible.
Fortunately, it is, and using statistics, we can use the heisenberg uncertainty of elementary particles in order to satisfy Newtonian physics - which is itself an approximation of particle movement (correct on neither the large scale, nor the small scale).
Oh, and humans may or may not have "free will." Particles never have free will. Particles may or may not have heisenbergian uncertainty, but never free will.
Anita Bean, sports nutritionist and author of Food for Fitness, finds it slightly hard to believe even Phelps can be expending quite as many calories as that. "Say he's doing about four miles a session, and a couple of sessions a day," she says, "plus his land training - I'd say he's burning maybe 5,000 calories in training, and maybe 2,500 simply to sustain himself. Something like 8,000 a day in all? Mind you, he is a very big bloke. I haven't looked at his schedules, but 12,000 seems a lot."
It doesn't take a nutritionist to work the First Law of Thermodynamics - if he were taking in more calories than he burned in one day than his weight (internal energy of the system) would have to increase proportionately to the unused Calories. As that obviously isn't happening, the work done by the system must approximately equal the energy introduced. As a trained nutritionist, she should probably know this... I wonder who she thinks is lying to her...
Honestly, I don't play WoW and don't care about what happens in regard to it, but I will say that Blizzard cannot do 100% authentication on their server. That's because they are not trying to prevent illegal commands from players, but trying to prevent people from having their computer run legitimate commands for them (bots). You can never prevent against a bot using server commands since the most clever way to construct the bot would simply to read and rewrite the memory of the client. Since the client would still be delivering the commands, there is no authentication method on the server that prevents interference.
Similarly, given enough time, even the most secure client can be cracked as the cracker has complete access to the allocated memory of the client program.
HVDC is interesting, but costly. The problem is that we continue to use AC power. While inverters can be built, they are complex and inefficient, often losing up to 7.5% of the power transmitted on the line.
Path 66 and path 15 are deceptive, as they appear to show large scale power transfer. However, the maximum load on the power lines is a little over 5000 Megawatts on a good day. According to the Energy Information Administration, Oregon, which is ranked 30th in the US in energy usage, requires 12,333 megawatts. San Francisco alone can require 950 megawatts on a summer night.
Solar power is a great advancement, but it simply cannot provide the energy needs of the country at this time. Supplemental use is good but the infrastructure needed to support wide scale solar use is simply too inefficient and expensive.
Slashdot Mirror
Let's make this simple. Don't use webmail. Don't use Yahoo.com, Gmail.com, Hotmail.com, squirrelmail, etc. There are SO many vulnerable access points between the web application and your email that it is almost impossible to secure the entire stack.
The use of Ajax alone (like most major webmail vendors) increases your vulnerability by huge amounts. SOP (same origin policy) is broken. A combination of a reflected XSS attack (which are everywhere http://blogs.zdnet.com/Google/?p=451 ) and a stored XSS attack can completely compromise your session.
Forgetting about XSS, there's still CSRF, injections, RFI, information leakage, broken authentication/session management, insecure url access, etc.
So seriously - unless you trust that your email server has secured every possible hole in every possible layer of their stack, stick to TLS/SSL encrypted imap/pop3/smtp. Now, I'm not saying these are perfect, but email protocols are just simpler. There will always be fewer places to attack and thus the chances of your email being compromised are just smaller.
I ran a high volume, 24/7 uptime file server on FreeBSD for 6 months. I had 2 GB of ram, an Intel Core 2 Quad, 2 GB of RAM, and 5 SATA drives in a 3 GB RAIDZ with close to a terabyte of daily throughput. I would have a kernel panic every two weeks.
The worst bug is this:
Heavy IO activity between ZFS and another file system (like rsyncing between ZFS and UFS or between ZFS and NFS) may result in a deadlock. Symptoms: processes wanting to do IO on a ZFS file system get stuck forever in "zfs" state (WCHAN), other file systems (e.g. UFS) are still working.
I could only handle about 3 days without having this lockup. There is no known fix. ZFS in FreeBSD is EXPERIMENTAL and should not be used for high availability filesystems.
I take two steps forward
I take two steps back
We come together cuz opposites attract
And you know: It ain't fiction
Just a natural fact
We come together cuz opposites attract
Sorry. Had to do it.
Heute gehÃrt uns Deutschland und morgen die ganze Welt?
Und nein, die meisten Amerikaner studieren Deutsch nicht.
Idk if your doors are the same, but for most of the doors in my house you can actually just pull the pins out of the hinges with a flathead screwdriver and a pair of pliers - takes half the time compared to unscrewing the hinges.
I'm a cryptography student so I'm not well versed in Mersenne prime generation and testing but I seem to remember that there was an optimized FFT called the Discrete Weighted Transform, but that it had issues with actually being implemented in GPUs because of floating point precision, coupled with the memory requirements.
What's really interesting is that I saw something floating around last year by a mathematician, Fuhrer I think? Before the multiplication algorithm with the best algorithmic time was Schoenhage-Strassen, but I think it managed to beat that. It wasn't by much, something like log log n - 2^log*n, so I'm not sure if it would make much of a difference, even with 10 million digit numbers. I'd do the math and/or find the paper for you, but its like 3:30 and I'm goin to sleep.
Wrong. The Internet is federally regulated. Ever heard of the FCC?
You can't have a state controlled power grid. Ever. Eventually you need to cross state lines for something (even if it isn't necessarily power), and there you get Congressional dominion.
. JPEG and PNG will be readable by the included reader device but if it fails irreparably, some future digital archeologists could have a hard time decoding it (JPEG is actually a very clever and CPU-intensive format; we don't notice it now because hardware is so fast)
What? Jesus, I wrote the DCT/iDCT and Huffman coding transformations for a primitive JPEG library when I was still a senior in high school. It's elementary calculus. Unless you foresee some kind of nuclear holocaust (in which case his data is probably useless anyway), go ahead and use JPEG. The only harm with compression is the possibility of excess data loss when cylinders die, etc.
If you really can't figure out JPEG in 25 years, come and look me up - I should still be alive (in my 40s)
25 years:
For high capacity? SCSI (not quite 25 years yet, but i dont predict it dying any time soon)
So SATA is probably your best bet.
The standard disk interfaces 25 years ago? ST-506, ESDI and SCSI.
You just proved yourself wrong. SCSI is still used everywhere in server systems - SATA is the only thing that comes close to usurping it.
Global warming is a misnomer anyway - it should be called, "global climate instability."
Doesn't matter - free will is a human based philosophical concept. It has no meaning in the context of determinism. Whether or not our "free will" is a result of the nondeterministic nature of particles is irrelevant.
Right - but our understanding of quantum mechanics is so rudimentary as to make this discussion worthless. Nondeterminism is the best bet according to our current Schroedinger model, which is the most widely accepted theory.
There is no practical difference between true randomness and sufficiently-pseudo-random deterministic chaos, so this whole free-will / predetermination argument is mutual intellectual masturbation. It doesn't make a difference to us, honestly, so let's stop pretending it important, shall we?
Can't agree more - this discussion is just stupid. We don't know nearly enough to make any debate on determinism vs nondeterminism.
However, the most widely accepted model of QM is the Schroedinger model, which predicts nondeterminism and, until proven otherwise, I'll go by that.
Hey, I'm not saying your arguments don't have philosophic merit, I'm definitely not qualified to evaluate them. In fact, I agree with a lot of what you say. I am only saying that the article is a discussion about determinism and quantum mechanics, not philosophy.
No. You are doing the same thing that I spoke about in my original post - you are using some weird made up semi-scientific and semi-philosophical definition of free will. Free will is a term that has very specific philosophical and ontological definition and it is NOT what we are talking about. Determinism, by definition, states that the observer would not affect the state of the system. Quantum mechanics states the opposite, which is why it is very unlikely that the universe (and, by extension, the human mind) is deterministic.
I completely agree! My position was that the article is not about the philosophical concept of "free will," but of the physics concept of determinism. I only have the barest undergraduate understanding of philosophy but from what I can tell there haven't been many great posts on that, either.
The very description of particles having "free will" underlies my point - the phisophical concept of free will applied to elementary particles wouldn't make any sense, while the question of whether or not a subatomic particle can lie within a deterministic system makes perfect sense (even though it's basically unprovable either way unless you accept the uncertainty principle).
Isolating a system is entirely impossible
And true random numbers can't be created by a P Turing machine - that doesn't mean PRNGs aren't useful. Many experiments have been used in the real world to test the Uncertainty Principle, which is essentially the basis of the theories of nondeterminism.
Statistics and mathematics allows us to pinpoint measurement and user errors instead of systematic errors, which allows us to present fairly conclusive evidence of determinism or nondeterminism.
Fortunately this debate is worthless because the best explanation has already been supported by a number experiments (Uncertainty Principle).
Mod parent up. Finally someone who knows what they are talking about.
The buzzword "free will" is bringing out the idiots with no science education. This discussion simplifies to one thing - if, given all the requisite variables in a system, one can predict the next infinite states of that system, that system is deterministic. Id est, if, ignoring the cloning theorem and other QM restraints, one knew the exact state of every particle in the human body and one could predict the next infinite states of that system (the body), then that system would be deterministic (have no "free will"). If, on the other hand, the human body (more precisely, the mind) could be proven to have a finite number of predictable states, then the underlying physical systems must therefore also have a finite number of predictable states (be unpredictable).
Now, QM predicts that subatomic particles are unpredictable. Technically, that would make our minds unpredictable HOWEVER - unpredictable is defined precisely as being unable to predict an infinite number of states in the system. A finite (even large) number may still be possible. This would the generalization of a large number of unpredictable subsystems in the system used to approximate the future states. As we see with Newtonian physics, this method can be fairly accurate.
The only way that humans could be proven to be completely predictable would be to disprove the tenets of quantum mechanics. Until then, humans have "free will."
"Free will" and religion have nothing to do with each other.
Read about the Heisenberg uncertainty principle.
You have to recognize the difference between determinism and extreme variable interference (chaos theory). "Unpredictable determinism" means that, on the large scale, there are too many variables to gain an accurate prediction of the outcome. However, one can always just control a miniature system with all independent variables held constant.
If the smallest possible system with all possible variables held constant and you still can't accurately produce an outcome, well, thats a nondeterministic system.
You are mistaking chaos theory for determinism. Simply because we are unable to produce an accurate mathematical model for something does not mean it is not deterministic. Economics is a profoundly bad example because 99% of the mathematics in practical economics is based on faulty assumptions and insufficient data. Physics and chemistry are much more useful in quantum analysis.
Specifically, I will speak to quantum particles as the GP did - on the subatomic level we can only produce probabilities for particle motion and position.
Mathematically (I wish Slashdot had LaTeX, there is no reason why it shouldn't)
(delta)x * (delta)p ⥠hbar/(4Ï) = hbar/2
while we can cannot exactly place a subatomic particles motion (where it will be in the next second), we can place the motion of a larger object. If your statement "the sum or average of a set of random variables is also a random variable" this would not be possible.
Fortunately, it is, and using statistics, we can use the heisenberg uncertainty of elementary particles in order to satisfy Newtonian physics - which is itself an approximation of particle movement (correct on neither the large scale, nor the small scale).
Oh, and humans may or may not have "free will." Particles never have free will. Particles may or may not have heisenbergian uncertainty, but never free will.
I like this quote:
Anita Bean, sports nutritionist and author of Food for Fitness, finds it slightly hard to believe even Phelps can be expending quite as many calories as that. "Say he's doing about four miles a session, and a couple of sessions a day," she says, "plus his land training - I'd say he's burning maybe 5,000 calories in training, and maybe 2,500 simply to sustain himself. Something like 8,000 a day in all? Mind you, he is a very big bloke. I haven't looked at his schedules, but 12,000 seems a lot."
It doesn't take a nutritionist to work the First Law of Thermodynamics - if he were taking in more calories than he burned in one day than his weight (internal energy of the system) would have to increase proportionately to the unused Calories. As that obviously isn't happening, the work done by the system must approximately equal the energy introduced. As a trained nutritionist, she should probably know this... I wonder who she thinks is lying to her...
Honestly, I don't play WoW and don't care about what happens in regard to it, but I will say that Blizzard cannot do 100% authentication on their server. That's because they are not trying to prevent illegal commands from players, but trying to prevent people from having their computer run legitimate commands for them (bots). You can never prevent against a bot using server commands since the most clever way to construct the bot would simply to read and rewrite the memory of the client. Since the client would still be delivering the commands, there is no authentication method on the server that prevents interference.
Similarly, given enough time, even the most secure client can be cracked as the cracker has complete access to the allocated memory of the client program.
HVDC is interesting, but costly. The problem is that we continue to use AC power. While inverters can be built, they are complex and inefficient, often losing up to 7.5% of the power transmitted on the line.
Path 66 and path 15 are deceptive, as they appear to show large scale power transfer. However, the maximum load on the power lines is a little over 5000 Megawatts on a good day. According to the Energy Information Administration, Oregon, which is ranked 30th in the US in energy usage, requires 12,333 megawatts. San Francisco alone can require 950 megawatts on a summer night.
Solar power is a great advancement, but it simply cannot provide the energy needs of the country at this time. Supplemental use is good but the infrastructure needed to support wide scale solar use is simply too inefficient and expensive.