I don't know what the legal situation in Canada is like, but it's entirely possible that a reasonable standard of evidence will end up being required in the U.S., too. While I know we have tons of settled suits and paid-off threats, there is but one actual case that's gone to trial. My rough review of the evidence in that case is that it was circumstantial, but fairly good -- that is, it seemed highly probable she actually downloaded and made available copyrighted music. (I'm sure there's no evidence indicating that someone else actually downloaded them from her, and "making available" is on shaky ground.) However, a single case doesn't define a policy -- only rulings by appellate courts count for that, and the trial is currently under appeal.
We'll hope for the best. Everyone's favourite video-game-hating lawyer can tell you what can happen if you keep filing junk lawsuits in the U.S.
It's common in physics for people to produce papers of no value and essentially troll for a response. This isn't what happens in your typical university research lab, no, but there are countless papers produced by independent researchers on cold fusion, string theory, quantum mechanics, and relativity that have absolutely no scientific merit. They get mailed to legitimate researchers hoping for their support in getting published.
Indeed. A lot of people didn't trust him at the time, and a lot of people disagreed with going to war before we even did so. Now, when politicians start up with "if we knew then what we know now..." business, I can't help but think that apparently a lot of people are more well-informed than they.
It's a whole different ball of wax to talk about humans defining intelligence in a context that is distinctly human. The Turing test certainly applies a particular slant to "artificial intelligence", as opposed to "artificial machines that do useful computations" or "artificial life".
What are you objecting to? Claiming that there's otherwise zero cost to submission? Creating a paper is time-consuming, but submitting it is not. There are a ton of really worthless papers out there and plenty of people whose time is cheap. Without some cost (whether a transfer of money or some other impediment) associated with submitting a paper, each one of those worthless papers will be submitted to every possible venue, resulting in prohibitively high costs for the journal.
As an interesting anecdote, I once asked a German classics journal I'd published in for permission to include the work in my thesis. They were quite amused, since apparently they hadn't required any copyright transfer or publishing exclusivity.
It provides motivation to not submit worthless articles. If there's zero cost for submission, then tons of completely useless articles would be submitted, and the cost for going through all of them would be a problem.
Not that, as an author, I particularly liked the charges for submitting (or the insane charges for subscription), but there is reasonable motivation behind it.
You pointed out pretty well earlier how it's practically necessary to pass a Turing test. In order to convincingly act human, an AI has to "know" quite a lot about humans that we get over the course of years of experience. (In this case, it seems they're modeling the AI's "character" on a single human, which cuts down on the amount of story-invention the AI requires.)
Processing enormous amounts of information from different sources that are conflicting and generating a single set of beliefs is a widely-researched problem, since it's terribly useful.
All it really needs to do is deduce that these concepts exist, and then ask other people about them in the training phase. Given the number of books that provide answers to all three of those questions, creating good answers is certainly possible.
I think you underestimate the capabilities of a good liar.
There have been a few very well-organized, very successful attacks on U.S. military organizations (though the only ones there are any details about are commercial members of the military-industrial complex) from unidentified outside sources. Far from idiots using well-published exploits.
There's undoubtedly a silent assumption of using a real testing process. That is, attempting this using many humans as the tester and as the computer's "competition".
While you may berate the intelligence of others, it's unlikely you actually thought they were computers very often.
"Not held as priori fact, but it is held as a firm belief of many scientists."
This is true, but you're straying far afield. If they are proper scientists, their belief that such a model exists does not influence their work or behavior.
"If quantum entanglement can be explained in a way that makes sense then why did the great minds of the 20th century try so hard to find a way of getting quantum mechanics to work without it (EPR)."
I didn't say it was easy to explain; I said you weren't explaining it properly. It's counterintuitive, but it is by no means some Zen-like belief. If you want to criticize those "great minds", you might say that the fact that it's counterintuitive led to them having a poor understanding of it. Certainly quantum mechanics is held as difficult to properly understand. Or, you could simply say that healthy scientific criticism is standard for scientists. At the time, quantum mechanics was new and presented a model that functioned very differently from previously-understood models; as such, it's natural to question it.
The EPR paradox you refer to, though, doesn't really encompass "trying so hard to find a way of getting quantum mechanics to work without it". In fact, the EPR paradox rather quickly led to a falsifiable test to determine if quantum entanglement is valid. (Bell's theorem, which is confirmed, demonstrates that either quantum entanglement is valid or there are nonlocal hidden variables. It's been further shown that quantum entanglement cannot be used for faster-than-light communication, which would violate causality -- one of the original concerns of EPR. The existence of nonlocal hidden variables, however, would violate causality.)
It doesn't, like I said. The comment was that Newton was not a scientist, which is true (at least for much of his life). His thinking was nonscientific: there is a difference between developing models through experimentation and searching for experimental confirmation of preestablished theories. However, his unscientific obsessions has nothing to do with that fraction of his work that was empirically supported.
It's no tenet of science that they necessarily must obey the same set of laws at different scales. It's well known that our model for gravity (general relativity) is incompatible with our current model of quantum mechanics at QM scales. On the assumption that there should be a single, reasonable model where both gravity and quantum mechanics coexist at all scales, people search for such a model. It is not, however, held as an a priori fact that we must be able to find such a model.
The point re: reasonableness is that your unavoidably biased view of whether a proposition "looks reasonable" is completely unscientific. You're exposing exactly what I mean by pointedly separating whether something "looks reasonable" from its empirical backing and correctness.
It is in fact proper scientific thinking to not take as correct a proposition until it is proven correct. Astrology fails to make falsifiable claims that are not shown to be false; as such, it is not correct.
Choosing not to learn programming on high cosmic-ray days a) indicates you're not familiar with the probability a cosmic ray will cause a computer error on Earth b) is clearly, from the definition you copied, not astrology. It is not a belief or tradition, it's not based on the position of celestial bodies or a related detail, and has nothing to do with personality or human affairs. The only way it can be astrology is if you choose to read the definition of astrology as "any human behavior dependent on extraterrestrial conditions and events", which is at odds with the common definition.
Zen is not a theory, it's a belief system. Your explanation of quantum entanglement shows that, at least, you don't understand quantum entanglement. Many things, particularly quantum mechanics, seem strange, mysterious, and unlikely if incorrectly described.
Computing the potential effects of distant stars' and planets' gravitational fields on events on Earth is an introductory physics problem.
The "belief" that it's possible to model the behavior of the universe based on empirical observation is, appropriately, one that's quite explainable -- it's borne out by empirical testing.
Thinking that astrology (which has no physical basis or empirical support) looks "normal" compared to quantum entanglement (which has plenty of empirical support) just shows that either you're not a scientist (so you consider things based on how much "sense they make" to the logical system you've constructed based on everyday experiences) or you don't understand quantum mechanics. The former is common; the latter, more so.
Newton had his moments. He was an excellent analyst and, for a time, a very good empiricist. He was also quite mad and, for most of his life, not at all a good scientist. While still dedicated to empirical testing, he was obsessed with empirically verifying beliefs he had constructed, which is not at all scientific.
Science isn't a popularity contest: you don't actually have to be a great scientist to produce great scientific results.
Slashdot Mirror
That's not true either, it's just a convenient analogy to make sense of a double-slit interference experiment.
I don't know what the legal situation in Canada is like, but it's entirely possible that a reasonable standard of evidence will end up being required in the U.S., too. While I know we have tons of settled suits and paid-off threats, there is but one actual case that's gone to trial. My rough review of the evidence in that case is that it was circumstantial, but fairly good -- that is, it seemed highly probable she actually downloaded and made available copyrighted music. (I'm sure there's no evidence indicating that someone else actually downloaded them from her, and "making available" is on shaky ground.) However, a single case doesn't define a policy -- only rulings by appellate courts count for that, and the trial is currently under appeal.
We'll hope for the best. Everyone's favourite video-game-hating lawyer can tell you what can happen if you keep filing junk lawsuits in the U.S.
Whether they tax blank CDs or not has no bearing on whether or not distributing copyrighted content without the copyright holder's consent is legal.
In Canada, for example, it's not: http://grep.law.harvard.edu/articles/03/08/22/1655233.shtml
Who said I don't have any empirical evidence?
It's common in physics for people to produce papers of no value and essentially troll for a response. This isn't what happens in your typical university research lab, no, but there are countless papers produced by independent researchers on cold fusion, string theory, quantum mechanics, and relativity that have absolutely no scientific merit. They get mailed to legitimate researchers hoping for their support in getting published.
Indeed. A lot of people didn't trust him at the time, and a lot of people disagreed with going to war before we even did so. Now, when politicians start up with "if we knew then what we know now..." business, I can't help but think that apparently a lot of people are more well-informed than they.
It won't be replaced for the same reason as ipv4, though.
It's a whole different ball of wax to talk about humans defining intelligence in a context that is distinctly human. The Turing test certainly applies a particular slant to "artificial intelligence", as opposed to "artificial machines that do useful computations" or "artificial life".
Well, if the journal has access to a large pool of qualified reviewers that they're not paying, the "cost" of taking submissions is pretty low.
What are you objecting to? Claiming that there's otherwise zero cost to submission? Creating a paper is time-consuming, but submitting it is not. There are a ton of really worthless papers out there and plenty of people whose time is cheap. Without some cost (whether a transfer of money or some other impediment) associated with submitting a paper, each one of those worthless papers will be submitted to every possible venue, resulting in prohibitively high costs for the journal.
That's not tolerated in these physics journals.
As an interesting anecdote, I once asked a German classics journal I'd published in for permission to include the work in my thesis. They were quite amused, since apparently they hadn't required any copyright transfer or publishing exclusivity.
Oh, I agree there are certainly more reasonable schemes.
:-)
Although, with financial incentive to get a paper accepted on top of the academic incentive, I fear for the grad students.
Most of the physics journals I've submitted to have required copyright transfer or, at the very least, an exclusive publishing agreement.
It provides motivation to not submit worthless articles. If there's zero cost for submission, then tons of completely useless articles would be submitted, and the cost for going through all of them would be a problem.
Not that, as an author, I particularly liked the charges for submitting (or the insane charges for subscription), but there is reasonable motivation behind it.
You pointed out pretty well earlier how it's practically necessary to pass a Turing test. In order to convincingly act human, an AI has to "know" quite a lot about humans that we get over the course of years of experience. (In this case, it seems they're modeling the AI's "character" on a single human, which cuts down on the amount of story-invention the AI requires.)
Processing enormous amounts of information from different sources that are conflicting and generating a single set of beliefs is a widely-researched problem, since it's terribly useful.
All it really needs to do is deduce that these concepts exist, and then ask other people about them in the training phase. Given the number of books that provide answers to all three of those questions, creating good answers is certainly possible.
I think you underestimate the capabilities of a good liar.
There have been a few very well-organized, very successful attacks on U.S. military organizations (though the only ones there are any details about are commercial members of the military-industrial complex) from unidentified outside sources. Far from idiots using well-published exploits.
There's undoubtedly a silent assumption of using a real testing process. That is, attempting this using many humans as the tester and as the computer's "competition".
While you may berate the intelligence of others, it's unlikely you actually thought they were computers very often.
"Not held as priori fact, but it is held as a firm belief of many scientists."
This is true, but you're straying far afield. If they are proper scientists, their belief that such a model exists does not influence their work or behavior.
"If quantum entanglement can be explained in a way that makes sense then why did the great minds of the 20th century try so hard to find a way of getting quantum mechanics to work without it (EPR)."
I didn't say it was easy to explain; I said you weren't explaining it properly. It's counterintuitive, but it is by no means some Zen-like belief. If you want to criticize those "great minds", you might say that the fact that it's counterintuitive led to them having a poor understanding of it. Certainly quantum mechanics is held as difficult to properly understand. Or, you could simply say that healthy scientific criticism is standard for scientists. At the time, quantum mechanics was new and presented a model that functioned very differently from previously-understood models; as such, it's natural to question it.
The EPR paradox you refer to, though, doesn't really encompass "trying so hard to find a way of getting quantum mechanics to work without it". In fact, the EPR paradox rather quickly led to a falsifiable test to determine if quantum entanglement is valid. (Bell's theorem, which is confirmed, demonstrates that either quantum entanglement is valid or there are nonlocal hidden variables. It's been further shown that quantum entanglement cannot be used for faster-than-light communication, which would violate causality -- one of the original concerns of EPR. The existence of nonlocal hidden variables, however, would violate causality.)
It doesn't, like I said. The comment was that Newton was not a scientist, which is true (at least for much of his life). His thinking was nonscientific: there is a difference between developing models through experimentation and searching for experimental confirmation of preestablished theories. However, his unscientific obsessions has nothing to do with that fraction of his work that was empirically supported.
It's no tenet of science that they necessarily must obey the same set of laws at different scales. It's well known that our model for gravity (general relativity) is incompatible with our current model of quantum mechanics at QM scales. On the assumption that there should be a single, reasonable model where both gravity and quantum mechanics coexist at all scales, people search for such a model. It is not, however, held as an a priori fact that we must be able to find such a model.
The point re: reasonableness is that your unavoidably biased view of whether a proposition "looks reasonable" is completely unscientific. You're exposing exactly what I mean by pointedly separating whether something "looks reasonable" from its empirical backing and correctness.
It is in fact proper scientific thinking to not take as correct a proposition until it is proven correct. Astrology fails to make falsifiable claims that are not shown to be false; as such, it is not correct.
Choosing not to learn programming on high cosmic-ray days a) indicates you're not familiar with the probability a cosmic ray will cause a computer error on Earth b) is clearly, from the definition you copied, not astrology. It is not a belief or tradition, it's not based on the position of celestial bodies or a related detail, and has nothing to do with personality or human affairs. The only way it can be astrology is if you choose to read the definition of astrology as "any human behavior dependent on extraterrestrial conditions and events", which is at odds with the common definition.
Zen is not a theory, it's a belief system. Your explanation of quantum entanglement shows that, at least, you don't understand quantum entanglement. Many things, particularly quantum mechanics, seem strange, mysterious, and unlikely if incorrectly described.
Computing the potential effects of distant stars' and planets' gravitational fields on events on Earth is an introductory physics problem.
The "belief" that it's possible to model the behavior of the universe based on empirical observation is, appropriately, one that's quite explainable -- it's borne out by empirical testing.
Thinking that astrology (which has no physical basis or empirical support) looks "normal" compared to quantum entanglement (which has plenty of empirical support) just shows that either you're not a scientist (so you consider things based on how much "sense they make" to the logical system you've constructed based on everyday experiences) or you don't understand quantum mechanics. The former is common; the latter, more so.
Newton had his moments. He was an excellent analyst and, for a time, a very good empiricist. He was also quite mad and, for most of his life, not at all a good scientist. While still dedicated to empirical testing, he was obsessed with empirically verifying beliefs he had constructed, which is not at all scientific.
Science isn't a popularity contest: you don't actually have to be a great scientist to produce great scientific results.
No, it's arrogant.
Do you have a reference for how a typical day of movie shooting goes, or are you making it up to support your own presumptions?