The wiki page is slightly sparse, but links to this site and several others with more information are in the external links section. There are some really cool animations of the thing in action here, especially this one.
Or my favorite, the F1/F0 ATP Synthase, which is literally a proton-powered turbine which inter-converts chemical and mechanical energy with ~97% efficiency.
Thanks, I had no idea. I will definitely be watching that one soon. Also, I thought about it some more and now I think that absolutely every problem with a polygraph is present with fMRI.
The MRI will need to be interpreted (like a polygraph) because 1) every brain is a little different, 2) there is still no "unit of lying" (i.e. no single objective output), 3) brain activity correlated with lying will also correlate with other thoughts or activities and 4) impossible to really link transient brain activity to transient speech (i.e. brain activity occurs before, during and after speech, thus can't link given word uttered now to brain activity observed now / in near future). As with the polygraph, the "expert" interpretation cannot be more than opinion.
So yeah, short of the type of actual telepathy in Bester's The Demolished Man, no lie detection.
Supposedly LastPass never transmits the master key over the network, so actually they do not have this capability.
The real security benefit of a password manager is in using automatically generated passwords. I use 16-25 characters, allowing alphanumerics and special characters and requiring each type.
That's not how LastPass works. All of the credentials are packed into a single binary blob which is encrypted locally using a key which itself is never sent over the network. They actually have a very good security model. In comparison, Chrome doesn't even bother to encrypt stored passwords.
The weakness is the same as with any piece of software: unless you personally read every line of code, you don't know for sure what it will do.
Much better to use LastPass or whathaveyou instead of the Chrome keystore, IMHO. For one thing, you're right about separating that from your user account keystore, but also the Chrome keystore is pretty insecure. LastPass makes a point of this during installation, once you've OK'd the install it's able to silently access all your passwords.
It's actually a hodgepodge across different jurisdictions. It seems like for the most part it's very difficult to use polygraph evidence in court, but the government does employ them generally. A fair number of states have severely restricted use.
Source
There's a huge difference between a pressure sensor applied to oil and a polygraph applied to lie detection.
In the first case, calibrated measurements are made in a standard, objectively defined unit by taking advantage of a law of physics. 1 kPa is 1 kPa is 1 kPa.
In the second, a bunch of graphs are written out based on physiological measurements, then "interpreted" by a supposed polygraph "expert." There is no objective standard or unit of "lying," and different experts will come up with different interpretations. Indeed, the US Supreme Court ruled that unlike DNA or fingerprint evidence, polygraph evidence is nothing more than the opinions of the examiners.
On the one hand, yes. On the other hand, pursuing these guys makes it seem like the test could work if you don't know how to fool it - which is exactly what they want, since the polygraph is only used to elicit confessions from chumps.
No, the polygraph doesn't work at all (US courts don't accept polygraph evidence).
They just set up this big machine and go through the whole ritual in order to trick people into confessing. They have to stamp down on the guys selling ways to "fool" the fake test in order to maintain the illusion that the test works.
It doesn't really matter if it deters others. The idea is just to convince people that the polygraph has some validity (or else it couldn't be fooled!) so that they can keep using them to elicit confessions from chumps.
A polygraph is absolutely not a "lie detector" with high false negative and false positive rates. Polygraphy is a pseudo-science and as such has no consistent FNR / FPR when turned to "lie detection."
The only use of the polygraph machine is to elicit a confession by trickery. And that is exactly why the government is so desperate to crush the guys who teach people how to "evade" the fake test: the belief that the "test" can possibly be fooled is enough to break the psychology of the elicited confessions.
Fool proof anti-polygraph method: don't worry about it and lie anyway.
When do you think Li depletion will limit battery production? If the answer is a decade or more, even with significant increases in output, then I suspect TFA will prove wrong about a near-term price spike due to production shortfall.
The are no "design flaws" or "design objectives" of any kind, because there is no design.
Susceptibility to deleterious mutations does reduce fitness. Some species, such as humans and rats are quite susceptible somatic cell mutations, leading to a variety of cancers. Others, such as sharks, do not have this particular genetic weakness. On the other hand, death after reproduction has no effect on individual fitness whatsoever, although it could increase the fitness of the species as a whole by reducing resource contention.
You might bother to apply at least a little though exploring simple corollaries of evolution by natural selection before making up nonsense - or at least you might if you weren't an arational, religious wacko.
Affirmative action means that the kid with brown skin has a slightly higher chance of getting into college than the kid with the pink skin
Not really. Constitutional affirmative action systems allow admission committees to consider race as an additional factor among the many factors contributing to admittance or rejection (typically including grades, extracurriculars, income level, sometimes legacy status, etc. though not appearance). Ultimately these committees make admissions decisions based on their collective, subjective judgments as opposed to objective scoring functions or something.
I'm not aware of any cases where the total effect on by-race admission probabilities is actually to give applicants of the targeted ethnicity a higher admission probability than "white" applicants. Such applicants just have a better chance than they would have had otherwise.
It's almost like you didn't even read what he wrote. If you come over here to Oakland and start following people around to "see if they're up to no good" you better believe that you just might end up at Highland giving a statement to some very unsympathetic police officers.
That's h4rr4r's point, and it's a valid one: if you create a situation where you might reasonably appear threatening, in a local where it is legal to retaliate against reasonable threats, then you should accept at least some responsibility for the consequences.
You are claiming that a single scientist using a single word to express surprise to a science journalist is somehow supportive of the intelligent design fallacy - irrationality which is well matched to your rambling prose.
The "proto-echolocation" behavior is regular mammalian hearing, so in that sense you are exactly right. However the convergent evolution being discussed did occur in parallel in different kinds of bats as well as whales, after the species split but before either had any pre-echolocation abilities other than regular hearing.
I agree completely, and I have multiple degrees in biochemistry and molecular biology. You do get totally different sequences which end up coding for the same protein structures, but like you seem to understand there are strict physical limits on what structures can actually perform a particular function, so especially for complex functions involving hundreds of coding sequences you might expect significant convergence.
Slashdot Mirror
The wiki page is slightly sparse, but links to this site and several others with more information are in the external links section. There are some really cool animations of the thing in action here, especially this one.
Or my favorite, the F1/F0 ATP Synthase, which is literally a proton-powered turbine which inter-converts chemical and mechanical energy with ~97% efficiency.
borrowed search algorithms
Uh, you mean the Perron-Frobenius theorem? I'm pretty sure there was no NSA in 1912.
Thanks, I had no idea. I will definitely be watching that one soon. Also, I thought about it some more and now I think that absolutely every problem with a polygraph is present with fMRI.
The MRI will need to be interpreted (like a polygraph) because 1) every brain is a little different, 2) there is still no "unit of lying" (i.e. no single objective output), 3) brain activity correlated with lying will also correlate with other thoughts or activities and 4) impossible to really link transient brain activity to transient speech (i.e. brain activity occurs before, during and after speech, thus can't link given word uttered now to brain activity observed now / in near future). As with the polygraph, the "expert" interpretation cannot be more than opinion.
So yeah, short of the type of actual telepathy in Bester's The Demolished Man, no lie detection.
Supposedly LastPass never transmits the master key over the network, so actually they do not have this capability.
The real security benefit of a password manager is in using automatically generated passwords. I use 16-25 characters, allowing alphanumerics and special characters and requiring each type.
That's not how LastPass works. All of the credentials are packed into a single binary blob which is encrypted locally using a key which itself is never sent over the network. They actually have a very good security model. In comparison, Chrome doesn't even bother to encrypt stored passwords.
The weakness is the same as with any piece of software: unless you personally read every line of code, you don't know for sure what it will do.
Much better to use LastPass or whathaveyou instead of the Chrome keystore, IMHO. For one thing, you're right about separating that from your user account keystore, but also the Chrome keystore is pretty insecure. LastPass makes a point of this during installation, once you've OK'd the install it's able to silently access all your passwords.
The House and Senate do have oversight of the NSA
Hmm, I guess that's why after Congress voted down the Clipper Chip, the NSA gave up on all its plans to backdoor domestic encryption software.
Oh wait...
the 5-15% where it fails are going to be the times when you most need it to be correct.
You need it most then, because it is far worse to imprison an innocent than to let a guilty person go free. I salute you.
It's actually a hodgepodge across different jurisdictions. It seems like for the most part it's very difficult to use polygraph evidence in court, but the government does employ them generally. A fair number of states have severely restricted use. Source
There's a huge difference between a pressure sensor applied to oil and a polygraph applied to lie detection.
In the first case, calibrated measurements are made in a standard, objectively defined unit by taking advantage of a law of physics. 1 kPa is 1 kPa is 1 kPa.
In the second, a bunch of graphs are written out based on physiological measurements, then "interpreted" by a supposed polygraph "expert." There is no objective standard or unit of "lying," and different experts will come up with different interpretations. Indeed, the US Supreme Court ruled that unlike DNA or fingerprint evidence, polygraph evidence is nothing more than the opinions of the examiners.
On the one hand, yes. On the other hand, pursuing these guys makes it seem like the test could work if you don't know how to fool it - which is exactly what they want, since the polygraph is only used to elicit confessions from chumps.
No, the polygraph doesn't work at all (US courts don't accept polygraph evidence).
They just set up this big machine and go through the whole ritual in order to trick people into confessing. They have to stamp down on the guys selling ways to "fool" the fake test in order to maintain the illusion that the test works.
It doesn't really matter if it deters others. The idea is just to convince people that the polygraph has some validity (or else it couldn't be fooled!) so that they can keep using them to elicit confessions from chumps.
It doesn't matter what you do, the whole rigmarole is just used to elicit confessions.
The only "lie detector" that really works is fMRI.
A polygraph is absolutely not a "lie detector" with high false negative and false positive rates. Polygraphy is a pseudo-science and as such has no consistent FNR / FPR when turned to "lie detection."
The only use of the polygraph machine is to elicit a confession by trickery. And that is exactly why the government is so desperate to crush the guys who teach people how to "evade" the fake test: the belief that the "test" can possibly be fooled is enough to break the psychology of the elicited confessions.
Fool proof anti-polygraph method: don't worry about it and lie anyway.
When do you think Li depletion will limit battery production? If the answer is a decade or more, even with significant increases in output, then I suspect TFA will prove wrong about a near-term price spike due to production shortfall.
The are no "design flaws" or "design objectives" of any kind, because there is no design.
Susceptibility to deleterious mutations does reduce fitness. Some species, such as humans and rats are quite susceptible somatic cell mutations, leading to a variety of cancers. Others, such as sharks, do not have this particular genetic weakness. On the other hand, death after reproduction has no effect on individual fitness whatsoever, although it could increase the fitness of the species as a whole by reducing resource contention.
You might bother to apply at least a little though exploring simple corollaries of evolution by natural selection before making up nonsense - or at least you might if you weren't an arational, religious wacko.
Affirmative action means that the kid with brown skin has a slightly higher chance of getting into college than the kid with the pink skin
Not really. Constitutional affirmative action systems allow admission committees to consider race as an additional factor among the many factors contributing to admittance or rejection (typically including grades, extracurriculars, income level, sometimes legacy status, etc. though not appearance). Ultimately these committees make admissions decisions based on their collective, subjective judgments as opposed to objective scoring functions or something.
I'm not aware of any cases where the total effect on by-race admission probabilities is actually to give applicants of the targeted ethnicity a higher admission probability than "white" applicants. Such applicants just have a better chance than they would have had otherwise.
Isn't the whole (fraudulent) concept of a "white" race based on skin color only and not ancestry?
It's almost like you didn't even read what he wrote. If you come over here to Oakland and start following people around to "see if they're up to no good" you better believe that you just might end up at Highland giving a statement to some very unsympathetic police officers.
That's h4rr4r's point, and it's a valid one: if you create a situation where you might reasonably appear threatening, in a local where it is legal to retaliate against reasonable threats, then you should accept at least some responsibility for the consequences.
You are claiming that a single scientist using a single word to express surprise to a science journalist is somehow supportive of the intelligent design fallacy - irrationality which is well matched to your rambling prose.
The "proto-echolocation" behavior is regular mammalian hearing, so in that sense you are exactly right. However the convergent evolution being discussed did occur in parallel in different kinds of bats as well as whales, after the species split but before either had any pre-echolocation abilities other than regular hearing.
I agree completely, and I have multiple degrees in biochemistry and molecular biology. You do get totally different sequences which end up coding for the same protein structures, but like you seem to understand there are strict physical limits on what structures can actually perform a particular function, so especially for complex functions involving hundreds of coding sequences you might expect significant convergence.
Also what is that T, "theorist?"
Only it's entirely credible. That is entire premise of a peer-reviewed publication.
You know your argument is worthless when it hinges entirely on nitpicking common expressions.