you're right, of course. COMPUTERS are not intelligent. However, some algorithms are based on models of intelligent behaviour. Whether or not you are using hardware that can rewire itself, as long as the code can rewrite itself you are dealing with some type of intelligence, simulated by a "stupid" computer. I can't really speak about this "Watson" thing, because I didn't read the article. But when your algorithm uses neural networks that have to be trained, you simulate a system that changes its behaviour according to a system of values (while training); it is true that once trained, if it can no longer rewrite itself, it's just a function taking an array of bits and returning another array of bits.
What we need is a proper definition of the term "intelligence". I agree that, like in computational physics, we are currently just making simulations of intelligent systems. In physics, it can be determined when an experiment would have the same outcome as a simulation; in the same way, algorithms can be "intelligent".
And I wasn't exactly trying to say they should be beggars. I was trying to say that they can avoid going through the middlemen (Apple in this case). If they're good enough, then the "consumers" will pay so that they keep doing what they're doing. Otherwise, they have to go through the middlemen, but then they can't really complain if a middleman refuses to distribute something.
I've already told someone in this thread: I'm a researcher. A PhD student to be exact; my advisor pays me to solve some problems. Once I have a solution to a problem, I am free to write any number of papers explaining it, and distribute some of them freely (he does need to have something in a peer-reviewed journal, because he gets his funding based on that). Am I a beggar? I don't care. Society created a system where I am being paid to generate something that I can then freely distribute.
With art, as you say, things are trickier. Because you can't objectively determine which artist should "have his grant renewed". But having your distributor say "I don't think your art is appropriate" when it costs nothing to distribute has to feel really weird.
I'm not doing this just to contradict you. But seriously, what kind of "art" is the art that people won't pay you to keep doing? From prehistoric times people only become artists if people payed to see their stuff. or if they had a rich uncle that put up with their s...stuff.
I'm a researcher. I give the results of my research away for free (arXiv, personal webpage). Yes, I submit it to journals that cost money, but you can get the results without going to the journals.
If you're an artist, give your stuff away for free. If you're good enough, people will make donations. If not, then what's the point of being an artist? If you give it away for free, then people are free to make txt/pdf/avi/ogg or whatever, that can be passed around with no problems.
it will never really work properly. it has been said before: a good enough mathematician can "gamble" on the stock market for a short time, but once he starts moving large enough sums around, the market will react, and the model will become obsolete. The only possible use for such a program is to bet on sporting events. In such a way that the players don't know how much you bet on them. Once you start betting on a hundred games a week, you can use statistics. On the other hand, doing this will make the odds for sportbets be computed with a bit more care.
I don't have a car, and I walk to the university, because it only takes me 20 minutes. When I lived too far, I took the subway. I don't print anything unless someone else requires me to, I have the cheapest cellphone I could buy (and I use it once every 2 weeks maybe), I don't play games on my computer (I do have "play time" on my computer, I don't deny that), me and my wife do not throw away food, and I only buy clothes when she gets angry with me because stuff gets torn (this goes for shoes too).
I'm sorry if I came out as a rant. I just thought that PhongUK was wrong, and I tried to explain why; I wasn't trying to tell him how to live, I was just showing an alternative to his somewhat sarcastic proposal. And I do try to live by my own standards.
Wrong. We need to stop using cellphones unless we actully need them, we need to restart gathering for telling stories in the evening in stead of watching TV, we need to forbid any kind of personal cars and force people to use bikes (what, are you really going to tell me that a car is faster than a bike in a big city? be serious), we need to stop wasting paper for stupid bureaucrats and use computers for what they were built for, we need to stop playing chess and other cardgames online and restart doing it with neighbours, we need to stop throwing away food, we need to stop buying clothes that we don't wear. And there are a whole lot of other things that we could do, but I'm bored now.
further on, once we have a good idea about what happened in the universe, we can start checking that against the various sets of "laws of the universe" that we can generate, and decide which are valid (i.e. "is string theory ok, or do we need something else?"). Afterwards, we can try to use the "correct" model of the universe to generate cool stuff (like quantum physics was used to properly describe semiconductors, and we got miniaturized electronics).
I wasn't trying to exagerate. Usually in science, each geek gets excited about a different thing; some years later, a less geeky person tries to do something practical, but is geeky enough to understand what the other geeks were doing, and succeeds in putting together all the information to come up with something useful. Generally, we have to try to finance all the excited geeks, because we can't properly decide which of them will come up with something useful.
that is not a "CON". there is an explanation: statistically, that is the class that the image belongs to. consider this: infants have a lot of inborn reflexes. and trust me, they don't understand why they do the things they do anymore than you would understand why your foot jumps when the doctor pokes your knee (I don't know the name of that reflex). there is absolutely no difference between using a trained human brain and using a trained simulated neural net.
It is really a journal. Once an arXiv paper is read by a specialist, they can spend half an hour writing a small text with his comments. If you'd have the option to see the paper with all it's comments, then that would basically mean that the process of peer review is reproduced; even better, you get to see all the comments of the reviewers. Anyway, I think this is the way to go. In the internet age, there is really no point to have publishers handle the work of scientists. We just have to figure out a way to reach each other.
In all seriousness, this is a sign that something is wrong with the human race. But no, it is not a sign that things are gonna happen. I was born in Romania in 1982, and my grandfather had been collecting a magazine since the '50s. This is the kind of news that was presented in those magazines in the '50s, and we still had a totalitarian regime 30 years later.
Things are only gonna happen if the civilized world wakes up and realizes that one of the leading causes of conflict is the fact that some peoples are kept stupid enough to go to war.
my mother is a medical doctor, and she also learned acupuncture some time after medical school. I felt acupuncture (actually, pressupuncture --- no needles, just finger poking) work on me. After more than 20 years experience of sometimes using acupuncture on her patients (she generally relies on what she learned in medschool), she says that there is stuff where acupuncture really works. She told me that she knows the "theory" is most likely bogus, but what she cares about is that if she follows the rules, good things happen to her patients. I do realize that with acupuncture there will be a lot of people who say it works for everything (my mother says it doesn't), just so that they can make more money. On the other hand, I also know that an animal is a very complicated physical system, and I wouldn't be surprised to find that there are still things we don't know about it (like for instance if someone pokes you with needles with the intent to heal, it actually heals you). I've heard of sick people swimming with dolphins and feeling better afterwards; I've also heard about people having different brainscans after swimming with dolphins.
I think it's worth investigating (1) the exact level of control that the brain can exert on the body and (2) the full scope of possible interactions with the brain/nervous system. What if the whole "qi line" thing is just an average model of some superposition of minuscule coherent electromagnetic fields of the body that do interact with the nervous system in some way?
Personally, I'm a physicist, and I know that with nonlinear systems, tiny fluctuations can do many things.
one religion asks that you dismiss "observed truth" (science) in favor of "divine inspiration". others do not. others don't even care about divine inspiration and observed truth, because they are about something else entirely (how can you possibly disprove reincarnation? you can show that knowledge about pastlives is equivalent to knowledge from dreams and pure imagination, but that is not a proof, and if a mind believes something, it can make it true in some ways).
This argument by Epicurus is void for several reasons. The simplest: why assume what you consider evil is the same as what god considers evil? maybe we have no idea what that kind of evil is, because we've never actually seen it.
seriously now. haven't you ever heard of artificial intelligence experiments? If I make a simulation of a world, with a bunch of intelligent agents that evolve according to some rules, am I not being God? I can in principle always alter the simulation any way I please, and if in fact the intelligent agents develop some kind of morality, it is possible that I will in fact choose as "winners" those that do not follow that morality (because my purpose is something else entirely).
I'm not doing this just to contradict you. It's just that I believe what I'm saying, and these things influence the way we interact with the rest of the world. My interest is not that you start believing in God, my interest is that you truly apply the concepts of mathematical logic to your reasoning. As a physicist, I see many models created to describe observed phenomena. However, it is a mathematical certainty that a very large (in Cantor's sense) set of models can be used, with the same results for observed phenomena, but with completely different fundamentals. We just try to find the simplest models, and the ones that are the easiest to work with.
We shouldn't be agressive to religions. We should focus on determining when religion is imposed rather than embraced, and try to forbid that. And similar things (like acting when there is fear, opression, stupid religious rules that forbid some child from being subjected to a needed medical procedure, etc). However, these are hard things to do (for instance, the whole "ban the burka" issue in the European Union. maybe some women where happy wearing it. is it right to stop someone because you think they're stupid? how is wearing a burka different, on the stupid level, from smoking?).
An earlier post says you cannot disprove the existence of an omnipotent god. By definition of "omnipotent". Personally, I doubt there is an omnipotent god, and that science does find truths. But you are wrong in saying that science can prove there are no omnipotent gods. Frist of all, we can't even prove that we are not living in a computer simulation http://xkcd.com/505/ (ok, here's the good link http://www.simulation-argument.com/). The basic fact is that if there is a middleman attack in your experiment (omnipotent god changing your data), than the resulting "science" is wrong. And you can't disprove that, you can just have faith that the data is real. I agree that it's best to trust the data, but we should be honest with ourselves: we BELIEVE the data is real, we have no proof of it. And yes, I agree that after they burned people for saying the earth circles the sun we need to be agressive against any organized religion that requires its followers to convert all other humans. But that kind of makes me someone who wants to convert others, so maybe we should instead agree to Gandhi's nonviolent noncooperation and stick to that.
I think that if a scientists says roughly "i'm an atheist, but I consider myself spiritual", it means that he doesn't see the world as being meaningless. most religions place the human "soul" outside of direct observation, so outside science. The sentence is probably trying to say that these atheist scientists still think of their mind as being something special. Personally, I have to agree. I see the mind as "meta-software" in a human body ("software" would be the one responsible for breath, digestion and so on), that is indepenedent of the human body in the sense that it could be copied in another adequate "computer". This basically means I consider the human mind to be constructed by the material world, but independent in the end; I might say "the body is the connection between reality and the mind". And this sentence sounds a lot like religion, hence "I consider myself spiritual".
i think history proves that open discussion of religion is actually the surest path to war. for scientists, that would mean a bunch of harmless geeks calling each other idiots, but it would still be an endless and fruitless discussion. you have to realize that, of all the people in the world, scientists are the ones who actually have proof that there is a higher order. And yet, in the interest of objectivity, they try to separate their science from their personal views, because experience shows that unless your hypothesis is directly verifiable, there will be others with an opposite view (think of the wave-particle view of light; Newton said particle with no real proof). And whenever there are people with opposing views, everything gets more complicated --- in science, unlike politics, we can't afford to sit around talking about our feelings; medicine is always needed "yesterday", as are more efficient machines and so on. scientists did not become atheists when they discovered they had opposing views about metaphysics. they just realized that in order to work together, they shouldn't take religion into account.
reporters had the purpose of gathering information from witnesses to various events, and redistributing that information. today, some bloggers can read a few other blogs from witnesses, and redistribute that information. yes, it will take some time to get this new system to work properly, but advertisers will give money to bloggers instead of reporters, so...
i don't know. maybe some bloggers will be good enough so that readers will prefer to pay them directly instead of seeing adverts on the blog, but i really don't see where a big media mogul fits into all this.
And now they want money from all the people using semiconductors. Seriously now, society advances because people can communicate ideas, not because the greedy get to keep all the money. Instead of being thankful that communication is almost free... It really is sad. no sarcasm. it's sad.
Well, you have to realize that there are only two classes of scientists that work on the edge: emergency room doctors and the people at NASA. Any other scientist can get back to the drawing board, and it's no big deal, but NASA people have only one chance at getting it right. Offtopic: that's probably why real scientists are surprised that is such a thing as a "software bug" --- they don't really expect you to say a program works unless it actually works.
to tell you the truth, i took my QFT course four or five years ago, and we didn't discuss rigurously the problem of virtual particles. When I think of virtual particles, I allways think of this Feynman diagram http://en.wikipedia.org/wiki/File:Vacuum_polarization.svg . Is this wrong? If it's wrong, what would be a correct picture?
The uncertainty principle imposes that if you look at some region of space for very short time periods, the energy contained in that region of space is not well defined (it can have many values, with a minimum inversely proportional to the time interval). this energy must be associated (in quantum field theory, the model that must be used to describe the phenomena) to some field that satisfies certain constraints. for instance, it could be associated to a pairing of a particle and its antiparticle (or a pairing of two photons). these pairs are pairs of virtual particles, because they only exist for the small time interval, during which the unvertainty principle allows the existence of the corresponding energy (actually, it allows for an "error" in the amount of energy contained in the region of space). So if you look at a region in space for some macroscopic time, you don't actually see anything, but as you start to look for smaller and smaller amounts of time, things start to happen (this was tested experimentally, see http://en.wikipedia.org/wiki/Casimir_effect). Hawking radiation is a very special occurence of this pair production. Supposedly, close to a black hole, it is possible that when a pair of virtual particles is created, one is eaten up by the black hole before they have a chance to anihilate one another, and the second particle manages to get away. At least, this is how I understand the idea, because I never had the chance to actually study it.
no, this is doesn't fit the physics i know of. in quantum field theory, you can describe the phenomena of a photon splitting into a particle-antiparticle pair that then anihilates to recreate the initial photon. these are the pairs that appear and disappear all the time (because of virtual photons that appear and disappear). However, a photon splitting into a particle-particle pair doesn't fit QFT.
Slashdot Mirror
yours is funny, but i liked it more when Sheldon's mom told him that "evolution is fact" was his opinion.
you're right, of course. COMPUTERS are not intelligent. However, some algorithms are based on models of intelligent behaviour. Whether or not you are using hardware that can rewire itself, as long as the code can rewrite itself you are dealing with some type of intelligence, simulated by a "stupid" computer. I can't really speak about this "Watson" thing, because I didn't read the article. But when your algorithm uses neural networks that have to be trained, you simulate a system that changes its behaviour according to a system of values (while training); it is true that once trained, if it can no longer rewrite itself, it's just a function taking an array of bits and returning another array of bits.
What we need is a proper definition of the term "intelligence". I agree that, like in computational physics, we are currently just making simulations of intelligent systems. In physics, it can be determined when an experiment would have the same outcome as a simulation; in the same way, algorithms can be "intelligent".
I'm the GP :)
And I wasn't exactly trying to say they should be beggars. I was trying to say that they can avoid going through the middlemen (Apple in this case).
If they're good enough, then the "consumers" will pay so that they keep doing what they're doing.
Otherwise, they have to go through the middlemen, but then they can't really complain if a middleman refuses to distribute something.
I've already told someone in this thread: I'm a researcher. A PhD student to be exact; my advisor pays me to solve some problems. Once I have a solution to a problem, I am free to write any number of papers explaining it, and distribute some of them freely (he does need to have something in a peer-reviewed journal, because he gets his funding based on that). Am I a beggar? I don't care. Society created a system where I am being paid to generate something that I can then freely distribute.
With art, as you say, things are trickier. Because you can't objectively determine which artist should "have his grant renewed". But having your distributor say "I don't think your art is appropriate" when it costs nothing to distribute has to feel really weird.
http://www.xkcd.com/
I'm not doing this just to contradict you. But seriously, what kind of "art" is the art that people won't pay you to keep doing? From prehistoric times people only become artists if people payed to see their stuff. or if they had a rich uncle that put up with their s...stuff.
I'm a researcher. I give the results of my research away for free (arXiv, personal webpage).
Yes, I submit it to journals that cost money, but you can get the results without going to the journals.
If you're an artist, give your stuff away for free. If you're good enough, people will make donations. If not, then what's the point of being an artist?
If you give it away for free, then people are free to make txt/pdf/avi/ogg or whatever, that can be passed around with no problems.
it will never really work properly. it has been said before: a good enough mathematician can "gamble" on the stock market for a short time, but once he starts moving large enough sums around, the market will react, and the model will become obsolete.
The only possible use for such a program is to bet on sporting events. In such a way that the players don't know how much you bet on them. Once you start betting on a hundred games a week, you can use statistics. On the other hand, doing this will make the odds for sportbets be computed with a bit more care.
I don't have a car, and I walk to the university, because it only takes me 20 minutes. When I lived too far, I took the subway. I don't print anything unless someone else requires me to, I have the cheapest cellphone I could buy (and I use it once every 2 weeks maybe), I don't play games on my computer (I do have "play time" on my computer, I don't deny that), me and my wife do not throw away food, and I only buy clothes when she gets angry with me because stuff gets torn (this goes for shoes too).
I'm sorry if I came out as a rant. I just thought that PhongUK was wrong, and I tried to explain why; I wasn't trying to tell him how to live, I was just showing an alternative to his somewhat sarcastic proposal. And I do try to live by my own standards.
Wrong.
We need to stop using cellphones unless we actully need them, we need to restart gathering for telling stories in the evening in stead of watching TV, we need to forbid any kind of personal cars and force people to use bikes (what, are you really going to tell me that a car is faster than a bike in a big city? be serious), we need to stop wasting paper for stupid bureaucrats and use computers for what they were built for, we need to stop playing chess and other cardgames online and restart doing it with neighbours, we need to stop throwing away food, we need to stop buying clothes that we don't wear.
And there are a whole lot of other things that we could do, but I'm bored now.
further on, once we have a good idea about what happened in the universe, we can start checking that against the various sets of "laws of the universe" that we can generate, and decide which are valid (i.e. "is string theory ok, or do we need something else?").
Afterwards, we can try to use the "correct" model of the universe to generate cool stuff (like quantum physics was used to properly describe semiconductors, and we got miniaturized electronics).
I wasn't trying to exagerate. Usually in science, each geek gets excited about a different thing; some years later, a less geeky person tries to do something practical, but is geeky enough to understand what the other geeks were doing, and succeeds in putting together all the information to come up with something useful. Generally, we have to try to finance all the excited geeks, because we can't properly decide which of them will come up with something useful.
that is not a "CON". there is an explanation: statistically, that is the class that the image belongs to.
consider this: infants have a lot of inborn reflexes. and trust me, they don't understand why they do the things they do anymore than you would understand why your foot jumps when the doctor pokes your knee (I don't know the name of that reflex).
there is absolutely no difference between using a trained human brain and using a trained simulated neural net.
It is really a journal. Once an arXiv paper is read by a specialist, they can spend half an hour writing a small text with his comments. If you'd have the option to see the paper with all it's comments, then that would basically mean that the process of peer review is reproduced; even better, you get to see all the comments of the reviewers.
Anyway, I think this is the way to go. In the internet age, there is really no point to have publishers handle the work of scientists. We just have to figure out a way to reach each other.
In all seriousness, this is a sign that something is wrong with the human race.
But no, it is not a sign that things are gonna happen. I was born in Romania in 1982, and my grandfather had been collecting a magazine since the '50s. This is the kind of news that was presented in those magazines in the '50s, and we still had a totalitarian regime 30 years later.
Things are only gonna happen if the civilized world wakes up and realizes that one of the leading causes of conflict is the fact that some peoples are kept stupid enough to go to war.
my mother is a medical doctor, and she also learned acupuncture some time after medical school.
I felt acupuncture (actually, pressupuncture --- no needles, just finger poking) work on me.
After more than 20 years experience of sometimes using acupuncture on her patients (she generally relies on what she learned in medschool), she says that there is stuff where acupuncture really works. She told me that she knows the "theory" is most likely bogus, but what she cares about is that if she follows the rules, good things happen to her patients.
I do realize that with acupuncture there will be a lot of people who say it works for everything (my mother says it doesn't), just so that they can make more money.
On the other hand, I also know that an animal is a very complicated physical system, and I wouldn't be surprised to find that there are still things we don't know about it (like for instance if someone pokes you with needles with the intent to heal, it actually heals you). I've heard of sick people swimming with dolphins and feeling better afterwards; I've also heard about people having different brainscans after swimming with dolphins.
I think it's worth investigating (1) the exact level of control that the brain can exert on the body and (2) the full scope of possible interactions with the brain/nervous system. What if the whole "qi line" thing is just an average model of some superposition of minuscule coherent electromagnetic fields of the body that do interact with the nervous system in some way?
Personally, I'm a physicist, and I know that with nonlinear systems, tiny fluctuations can do many things.
that's not a valid argument.
one religion asks that you dismiss "observed truth" (science) in favor of "divine inspiration". others do not. others don't even care about divine inspiration and observed truth, because they are about something else entirely (how can you possibly disprove reincarnation? you can show that knowledge about pastlives is equivalent to knowledge from dreams and pure imagination, but that is not a proof, and if a mind believes something, it can make it true in some ways).
This argument by Epicurus is void for several reasons. The simplest: why assume what you consider evil is the same as what god considers evil? maybe we have no idea what that kind of evil is, because we've never actually seen it.
seriously now. haven't you ever heard of artificial intelligence experiments? If I make a simulation of a world, with a bunch of intelligent agents that evolve according to some rules, am I not being God? I can in principle always alter the simulation any way I please, and if in fact the intelligent agents develop some kind of morality, it is possible that I will in fact choose as "winners" those that do not follow that morality (because my purpose is something else entirely).
I'm not doing this just to contradict you. It's just that I believe what I'm saying, and these things influence the way we interact with the rest of the world. My interest is not that you start believing in God, my interest is that you truly apply the concepts of mathematical logic to your reasoning.
As a physicist, I see many models created to describe observed phenomena. However, it is a mathematical certainty that a very large (in Cantor's sense) set of models can be used, with the same results for observed phenomena, but with completely different fundamentals. We just try to find the simplest models, and the ones that are the easiest to work with.
We shouldn't be agressive to religions. We should focus on determining when religion is imposed rather than embraced, and try to forbid that. And similar things (like acting when there is fear, opression, stupid religious rules that forbid some child from being subjected to a needed medical procedure, etc). However, these are hard things to do (for instance, the whole "ban the burka" issue in the European Union. maybe some women where happy wearing it. is it right to stop someone because you think they're stupid? how is wearing a burka different, on the stupid level, from smoking?).
An earlier post says you cannot disprove the existence of an omnipotent god. By definition of "omnipotent".
Personally, I doubt there is an omnipotent god, and that science does find truths. But you are wrong in saying that science can prove there are no omnipotent gods. Frist of all, we can't even prove that we are not living in a computer simulation http://xkcd.com/505/ (ok, here's the good link http://www.simulation-argument.com/). The basic fact is that if there is a middleman attack in your experiment (omnipotent god changing your data), than the resulting "science" is wrong. And you can't disprove that, you can just have faith that the data is real.
I agree that it's best to trust the data, but we should be honest with ourselves: we BELIEVE the data is real, we have no proof of it.
And yes, I agree that after they burned people for saying the earth circles the sun we need to be agressive against any organized religion that requires its followers to convert all other humans. But that kind of makes me someone who wants to convert others, so maybe we should instead agree to Gandhi's nonviolent noncooperation and stick to that.
I think that if a scientists says roughly "i'm an atheist, but I consider myself spiritual", it means that he doesn't see the world as being meaningless. most religions place the human "soul" outside of direct observation, so outside science. The sentence is probably trying to say that these atheist scientists still think of their mind as being something special.
Personally, I have to agree. I see the mind as "meta-software" in a human body ("software" would be the one responsible for breath, digestion and so on), that is indepenedent of the human body in the sense that it could be copied in another adequate "computer". This basically means I consider the human mind to be constructed by the material world, but independent in the end; I might say "the body is the connection between reality and the mind". And this sentence sounds a lot like religion, hence "I consider myself spiritual".
i think history proves that open discussion of religion is actually the surest path to war. for scientists, that would mean a bunch of harmless geeks calling each other idiots, but it would still be an endless and fruitless discussion. you have to realize that, of all the people in the world, scientists are the ones who actually have proof that there is a higher order. And yet, in the interest of objectivity, they try to separate their science from their personal views, because experience shows that unless your hypothesis is directly verifiable, there will be others with an opposite view (think of the wave-particle view of light; Newton said particle with no real proof). And whenever there are people with opposing views, everything gets more complicated --- in science, unlike politics, we can't afford to sit around talking about our feelings; medicine is always needed "yesterday", as are more efficient machines and so on.
scientists did not become atheists when they discovered they had opposing views about metaphysics. they just realized that in order to work together, they shouldn't take religion into account.
reporters had the purpose of gathering information from witnesses to various events, and redistributing that information. ...
today, some bloggers can read a few other blogs from witnesses, and redistribute that information.
yes, it will take some time to get this new system to work properly, but advertisers will give money to bloggers instead of reporters, so
i don't know. maybe some bloggers will be good enough so that readers will prefer to pay them directly instead of seeing adverts on the blog, but i really don't see where a big media mogul fits into all this.
And now they want money from all the people using semiconductors. Seriously now, society advances because people can communicate ideas, not because the greedy get to keep all the money.
Instead of being thankful that communication is almost free...
It really is sad. no sarcasm. it's sad.
Well, you have to realize that there are only two classes of scientists that work on the edge: emergency room doctors and the people at NASA. Any other scientist can get back to the drawing board, and it's no big deal, but NASA people have only one chance at getting it right.
Offtopic: that's probably why real scientists are surprised that is such a thing as a "software bug" --- they don't really expect you to say a program works unless it actually works.
to tell you the truth, i took my QFT course four or five years ago, and we didn't discuss rigurously the problem of virtual particles.
When I think of virtual particles, I allways think of this Feynman diagram http://en.wikipedia.org/wiki/File:Vacuum_polarization.svg . Is this wrong? If it's wrong, what would be a correct picture?
You're right. It's beginning to look a lot like "the boat that rocked" (fun movie, if you haven't seen it).
The uncertainty principle imposes that if you look at some region of space for very short time periods, the energy contained in that region of space is not well defined (it can have many values, with a minimum inversely proportional to the time interval). this energy must be associated (in quantum field theory, the model that must be used to describe the phenomena) to some field that satisfies certain constraints. for instance, it could be associated to a pairing of a particle and its antiparticle (or a pairing of two photons). these pairs are pairs of virtual particles, because they only exist for the small time interval, during which the unvertainty principle allows the existence of the corresponding energy (actually, it allows for an "error" in the amount of energy contained in the region of space).
So if you look at a region in space for some macroscopic time, you don't actually see anything, but as you start to look for smaller and smaller amounts of time, things start to happen (this was tested experimentally, see http://en.wikipedia.org/wiki/Casimir_effect).
Hawking radiation is a very special occurence of this pair production. Supposedly, close to a black hole, it is possible that when a pair of virtual particles is created, one is eaten up by the black hole before they have a chance to anihilate one another, and the second particle manages to get away. At least, this is how I understand the idea, because I never had the chance to actually study it.
no, this is doesn't fit the physics i know of.
in quantum field theory, you can describe the phenomena of a photon splitting into a particle-antiparticle pair that then anihilates to recreate the initial photon. these are the pairs that appear and disappear all the time (because of virtual photons that appear and disappear). However, a photon splitting into a particle-particle pair doesn't fit QFT.