“There is a theory which states that if ever anyone discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarre and inexplicable.
There is another theory which states that this has already happened.”
Actually, a sim coming into the real world is trivial -- if the runners of the sim help. They just put you in a blank simulation and change the inputs to reflect 'real world' sensory data. Make a robot body or something. Maybe they'll have to project the sensory data down to a form you can perceive, or just put you in a blank room with a computer that can talk to the outside.
The simulation simulating a real person is much harder...
There are a few well-known issues: not publishing negative results, overinflating importance of your new method, and the drive to publish like crazy discourages collaboration to some extent.
In the first issue, most well-known conferences and journals have so many positive results to choose from (partly because of the second problem), that they just don't care about the negative results. Negative results also don't draw a lot of attention in CS, as they'd pertain to someone trying their own random idea. Some people might think the idea is dumb and would never work -- but who is to say? If it works, then it wasn't so dumb -- but then that's a positive result. If it doesn't work, I can have 1,000 other bridges to publish at your journal by next Tuesday... heh. I think it's a hard problem, but not unmanageable. Maybe some new journals dedicated to negative results (there are some already but aren't known for being very rigorous yet) that at least put your negative result through the gauntlet first.
The second issue is a consequence of a researcher not wanting to accept a negative result -- just find some data that works on your new method and act that's the first dataset you tried it on. This may horrify researchers in other fields, but it definitely happens here. What mitigates this problem, at least, is (a) a lot of research code is on GitHub now, and (b) you can tell what new research is useful whether or not people actually use it in the future. That's the good thing about machine learning, if it works, it'll actually be used somewhere. Basically, we have easy reproducibility.
The last issue is if you have a cool new idea, I've found you don't really want to tell anyone about it until (a) you find out it sucks, or (b) you have it accepted as a paper somewhere. THEN you can collaborate with others.
I agree -- but people like GP would argue that those services could be provided by the free market. I think a better argument against libertarianism (as one who has been tempted by it, for point 68 seconds), is just that it doesn't work in theory.
Start with the prisoner's dilemma (with heterogeneous agents). The optimal outcome requires two parties to agree not to backstab each other, and this outcome can only be enforced externally. Which means, government. Unfortunately, this problem is not simply theoretical, and it applies to such a huge variety of real-world issues (pollution, anti-competitive behaviors, subsidizing healthcare, etc), that you basically need a government to step in, quite frequently, and lay down rules.
One may argue that a government that only enforces contracts and laws could work, where the two prisoners would enter into contract not to backstab each other. But then you run into problems where, in some cases, the entire population "should" enter into a contract to do XYZ. And then that's basically the same thing as electing a regulatory body...
I think it's a prevailing attitude on Slashdot -- look at the comments of any article that has a positive outlook on future technology. But it's not just slashdot, it's true in most educated circles too -- general skepticism and cynicism. Most people's BS filters are turned to 100% -- which keeps them safe against the crazies, but saps the imagination. Even at work, doing natural language processing research, I find it a little depressing that the common view among my colleagues is essentially "human level intelligence will never happen".
I think that's actually an argument in favor of simulation, not the reverse. It usually takes less energy to do a simulation than it takes to actually do the thing, because simluations make approximations. Industrial processes simulate reactions and processes to see if they are feasible, before they actually invest in doing them. Which is cheaper, sending a million people into space, or having a million people play a space video game?
Go into Google Maps, click the menu bar, and then open up "Your timeline". It has a calendar display, and you can click on any date in the past year and a half (on mine, at least), and it pretty much tells you where you went, and how long you were there. It's spooky. I mean, we all knew the technology was here, but it's another thing to realize that they've actually done it. Even when you aren't navigating, even if you've disabled hi-res location services. Plus, they have a nice UI for telling you they did all this, implying that nothing's wrong with this picture.
I agree. I think free will just means you have the capacity to make a decision based on the sum of your experiences. Given those, principles, motivations, and environmental context, it should be deterministic. Predictable, theoretically. Practically not, because of chaotic effects, and the need for massive compute power.
Some people find it a matter of pride to not want to be deterministic -- but I don't see any shame in coming to the same conclusions, given the exact same data. (Unless we're stuck in one those dreaded time loops)
Sometimes a person just has a bad day... maybe their S.O. starts a heated argument via g-chat. Then they get distracted, make a mistake, and with the worst timing, the city bus that doesn't give a damn plows on through.
But the prevailing wisdom on/. is that hey, they're dumb and deserved it, right?
Some good coders who post good solutions may stop posting if ad-block is not allowed. The people reading the answers will probably still go, regardless of ads. So, it would hurt the community.
Not that all good coders block ads, or all people in search of answers don't mind ads... but there's a diversity of personality types required to make Stack Overflow work.
Slashdot Mirror
“There is a theory which states that if ever anyone discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarre and inexplicable. There is another theory which states that this has already happened.”
Actually, a sim coming into the real world is trivial -- if the runners of the sim help. They just put you in a blank simulation and change the inputs to reflect 'real world' sensory data. Make a robot body or something. Maybe they'll have to project the sensory data down to a form you can perceive, or just put you in a blank room with a computer that can talk to the outside. The simulation simulating a real person is much harder...
There are a few well-known issues: not publishing negative results, overinflating importance of your new method, and the drive to publish like crazy discourages collaboration to some extent.
In the first issue, most well-known conferences and journals have so many positive results to choose from (partly because of the second problem), that they just don't care about the negative results. Negative results also don't draw a lot of attention in CS, as they'd pertain to someone trying their own random idea. Some people might think the idea is dumb and would never work -- but who is to say? If it works, then it wasn't so dumb -- but then that's a positive result. If it doesn't work, I can have 1,000 other bridges to publish at your journal by next Tuesday... heh. I think it's a hard problem, but not unmanageable. Maybe some new journals dedicated to negative results (there are some already but aren't known for being very rigorous yet) that at least put your negative result through the gauntlet first.
The second issue is a consequence of a researcher not wanting to accept a negative result -- just find some data that works on your new method and act that's the first dataset you tried it on. This may horrify researchers in other fields, but it definitely happens here. What mitigates this problem, at least, is (a) a lot of research code is on GitHub now, and (b) you can tell what new research is useful whether or not people actually use it in the future. That's the good thing about machine learning, if it works, it'll actually be used somewhere. Basically, we have easy reproducibility.
The last issue is if you have a cool new idea, I've found you don't really want to tell anyone about it until (a) you find out it sucks, or (b) you have it accepted as a paper somewhere. THEN you can collaborate with others.
I agree -- but people like GP would argue that those services could be provided by the free market. I think a better argument against libertarianism (as one who has been tempted by it, for point 68 seconds), is just that it doesn't work in theory. Start with the prisoner's dilemma (with heterogeneous agents). The optimal outcome requires two parties to agree not to backstab each other, and this outcome can only be enforced externally. Which means, government. Unfortunately, this problem is not simply theoretical, and it applies to such a huge variety of real-world issues (pollution, anti-competitive behaviors, subsidizing healthcare, etc), that you basically need a government to step in, quite frequently, and lay down rules. One may argue that a government that only enforces contracts and laws could work, where the two prisoners would enter into contract not to backstab each other. But then you run into problems where, in some cases, the entire population "should" enter into a contract to do XYZ. And then that's basically the same thing as electing a regulatory body...
I think it's a prevailing attitude on Slashdot -- look at the comments of any article that has a positive outlook on future technology. But it's not just slashdot, it's true in most educated circles too -- general skepticism and cynicism. Most people's BS filters are turned to 100% -- which keeps them safe against the crazies, but saps the imagination. Even at work, doing natural language processing research, I find it a little depressing that the common view among my colleagues is essentially "human level intelligence will never happen".
I agree. I think the more the merrier when it comes to working on the big problems.
I think that's actually an argument in favor of simulation, not the reverse. It usually takes less energy to do a simulation than it takes to actually do the thing, because simluations make approximations. Industrial processes simulate reactions and processes to see if they are feasible, before they actually invest in doing them. Which is cheaper, sending a million people into space, or having a million people play a space video game?
Go into Google Maps, click the menu bar, and then open up "Your timeline". It has a calendar display, and you can click on any date in the past year and a half (on mine, at least), and it pretty much tells you where you went, and how long you were there. It's spooky. I mean, we all knew the technology was here, but it's another thing to realize that they've actually done it. Even when you aren't navigating, even if you've disabled hi-res location services. Plus, they have a nice UI for telling you they did all this, implying that nothing's wrong with this picture.
I agree. I think free will just means you have the capacity to make a decision based on the sum of your experiences. Given those, principles, motivations, and environmental context, it should be deterministic. Predictable, theoretically. Practically not, because of chaotic effects, and the need for massive compute power. Some people find it a matter of pride to not want to be deterministic -- but I don't see any shame in coming to the same conclusions, given the exact same data. (Unless we're stuck in one those dreaded time loops)
Sometimes a person just has a bad day... maybe their S.O. starts a heated argument via g-chat. Then they get distracted, make a mistake, and with the worst timing, the city bus that doesn't give a damn plows on through. But the prevailing wisdom on /. is that hey, they're dumb and deserved it, right?
Some good coders who post good solutions may stop posting if ad-block is not allowed. The people reading the answers will probably still go, regardless of ads. So, it would hurt the community. Not that all good coders block ads, or all people in search of answers don't mind ads... but there's a diversity of personality types required to make Stack Overflow work.
has done pretty well on up-time.