Yeah, my favorite laptop is my 701c butterfly by IBM. runs linux well enough to ssh to my real machine and it's folding keyboacd never fails to impress. I would really like to see something like it made again, but with modern materials and batteries (all used to make it even smaller and lighter than before!)
Why do you assume they didn't do a computer simulation?
I don't know about that group in particular, but they most likely developed on a simulator, then implemented it real robots once they had a potential algorithm. These robots are also not very likely to be single purpose, They will most likely be used many times by other projects who were developing potential software with their simulators. Also, whoever got to build the robots got to learn about building robots and mechanical systems in general, likely as part of their course (much better than building things that will never be used).
I say potential because simulations are never perfect, and neither is the hardware (but in different ways) so they are only good up to a point.
the site is slashdotted, so I don't really know this for sure.. but it seems likely if this is similar to other university research projects and even ones in the commercial world as far as simulators are concerned.
Is that this is one of the first time (or perhaps the first time?) they were able to recover a meteorite this deep, while preserving and studying the soil above it and the track it made in the earth around it. Usually, one finds meteorites when their plow runs into one, causing the impact zone to be disturbed.
in practice, a space program is nothing like a broken window, the technology and science it produces well exceeds the cost, but the act of spending the money alone isn't where the economy derives its benefits.
meteoroids are often going many kilometers a second. In order to maintain steady planetary orbit, everything up there must be traveling really very fast compared to the earth. However, chances are that if you hit them, they are not in the exact same orbit as you, meaning that no matter what, the relative speed difference between you two is going to be huge.
a meteoroid in space that is moving relatively slow compared to you would actually be a very novel thing to encounter. probablistically, if you are gonna be hit by anything unexpected, it is going to be something moving really really really fast from your perspective.
Don't forget 'Primer' http://imdb.com/title/tt0390384/ made for $7,000 and is excellent. The only movie I ever immediatly rewound and watched again. Also one of the few good time travel movies out there that doesn't gloss over the messy details of causality, but rather revels in them.
So I wonder why they haven't done it yet. They've got plenty of opportunity.
Well, the obvious conclusion is that they arn't nearly as prevalant as one thinks. Not to say terrorists don't exist. A lot of people hate one another for various reasons. but for the most part, people are not willing to kill inoccent bystanders or anyone actually. People willing to do so are few and far between.
Because for the most part, humans are decent folks to each other. We evolved as social creatures, we naturally follow social norms like not killing each other.
Things are patented for a specific purpose. So, his patent wouldn't disallow anyone else from producing the chemical for other reasons. But it would disallow anyone else selling it as mosquito repellant. It is not the chemical he is patenting, it is the idea of using it as mosquito repellant.
Rather than require source code be released, which might be tricky legally to define. I would prefer to see a law saying that the interface and specifications of hardware devices must be made available on request to anyone that purchases said device. You should be able to know what you are getting when you purchase something.
not true. with _credit cards_ only, specifically not debit cards, the law is on your side. The burden of proof is on them to show you bought something. This is not true with debit cards. no matter what guarentees banks say they give with them, the law is still on their side and they can legally tell you to go away and suck it up. With credit cards you can do the same to them.
the reason is that with debit cards, it is your money, losing it is the same as if someone swiped your wallet. No one refunds you if you can't catch the thief. With credit cards, it is not your money, it is the credit card companies money, so they are stealing from the credit card company, not you, so it is their burden to recover the money and they can't make you pay.
The main question is whether they use actual gyros. While accelerometer chips are fairly inexpensive, gyroscopes on a chip are still quite expensive. Accelerometers are fast and cheap but they only detect change, you have to integrate over time to get location which causes an ever growing loss of precision that builds up very quickly. I think but am not positive that this is what the PS3 uses. accelerometers are nice, better than nothing, but without an _absolute_ reference like a gyroscope to constantly recalibrate your accelerometers, you can't detect position or orientation. so you can jerk the controller around to move something, but you can't hold it still to position things, there is no way for the controler to tell whether you are holding it at an angle relative to gravity or twisting it (or some combination). The lag noticed on the PS3 demos is what makes me think it is accelerometer based, a standard way to figure out the difference between gravity and rotation is to wait and look at the signal and use a heuristic algorithm to determine what was happening after the fact, this introduces an inherent lag. you can't decide what was done with the controler without examining some future data.
Now, the Wii also has accelerometers I assume, but in addition it has a stationary reference on top of the tv, if it uses this as its absolute positioning reference rather than a gyro, then they can cut a _substantial_ cost out of the per controller price, and get true 3d positioning to boot. as well as the advantage of no-lag interpretation of the accelerometer data.
Of course, economy of scale might have gotten nintendo a great deal on some real gyros. in which case I am going to pick up some extra controllers and canibalize them for parts:)
This is mainly speculation. On one hand, I have to hand it to nintendo for cleverness if they can get true good orientation data without gyros, on the other RC helicopter enthusiasts would love cheap accelerometer/gyro modules at consumer prices.
now the feature I really hope nintendo implemented is _directional rumble_ mounted on a gimbal so the rumble weight can be positioned inside the controller and give a kick in an arbitrary direction. imagine a sword fighting game where the rumble kick was always in the correct direction relative to what direction you are hitting the other guy (or his sword) from. that would be spectacular.
I was wondering about that too. My best therories as to the wacky coloring are that either someone found an old pillow case to make the covering out of, or they were studing the "twist" of the arm with computer vision, so having the multi-colored stripes along the length of the tentacle let the computer (or even the people) easily see how much the tentacle was being torqued and twisted around its axis. Of course, this is just speculation. I'll check out their papers if available to see if there is any insight there, the technology is quite interesting.
Just like the crazy idea of creating a protocol to allow networking a bunch of computers at separate locations together they funded a while ago. I am sure many didn't see the point then either. As we learn to swim the deep-end doesn't seem so deep any more. Plus, selection bias, DARPA funds a huge number of projects, only the ones with a certain type of appeal end up on the front page of slashdot. It's like the news, if you thought it was representative of a random sampling of society you would think the world was a pretty darn scary place, but they don't report on all the people getting along just fine having a long normal life because that is the norm.
google doesn't censor anything themselves, they simply run the chineese webcrawler from within the great firewall. the problem is with the great firewall not being too bright about mispellings, not anything google does. it's not googles job to make sure china gets their software right.
Yes, but with a credit card, the law guarentees you don't have liability. the thief is stealing from the bank. not you. With a debit card, the bank may have a policy of limiting your liability, but that is purely part of their policy and who knows how they interpret it and I bet they have clauses in their policy making appeal of their decision quite difficult. With a debit card the law says the money was stolen from you, so it is ultimately your responsibility to recover your money. with a credit card, the money was stolen from the bank so it is the banks responsibility.
the interesting thing is that quantum systems scale superlinearly with the number of qubits relative to standard computers.
so if we assume vanilla computers double in power every year, in order for quantum computers to keep up, one needs only to produce a quantum computer with _one_ more qubit every year.
of course, there is some subtly in the meaning of 'keeping up' since the problems practically solvable by each are different so you can't compare them directly, but you can compare their growth in power over time, which is what moores law is about.
Actually, the other choice is not that it is quantum random, it is that _it doesn't matter_. Science is only concerned with things that are testable and observable. Science is agnostic (or even apatheist), but not inherently athiest. (though scientists themselves come in those and many othen varieties.)
Also, there are many TOEs that don't involve randomness FYI. Randomness is an observed characteristic of quantum mechanics, the book is still out as to whether it is a defining one.
It is because apple doesn't have an effective monopoly and use it to obtain monopolies in other markets. The laws are not about bundling software, they are about unfair business practices.
admittedly, those are subject to interpretation, as is the term monopoly but that is why they don't go after apple. Microsoft does have more money to milk, but this is a case of coorelation not causation, the monopoly holder is the one making the money off their unfair business practices. Some practices are only unfair if you hold a huge majority of the market as you are free to manipulate it unfairly in ways other companies cannot.
This is a common misconception of what Occam's Razor is. It has absolutly nothing to do with "simplicity". It's exact wording is
one should not increase, beyond what is necessary, the number of entities required to explain anything.
An example would be saying that "The man died because of his heart attack and he was wearing blue pants". the "blue pants" are not needed to explain the fact he died, so can be dropped from the assertion without changing its validity. The application to the ID situation is a whole lot more obvious now and doesn't use overloaded terms like 'simplicity'.
Just because you say a building is 100 meters tall, that doesn't mean you took a meter stick and used it stepwise all the way up to determine how tall it was. you could have done many things, measured a line away from it and figure out the exact angle you look up at to see the top. You could know the inclination of the sun and measure the length of the shadow.
likewise, when you say something in 5 lightyears away, that doesn't mean you measured it by bouncing a signal off it and it took exactly 10 years to get back. just as there are lots of ways to measure things on earth, there are lots of ways to measure them in space.
A common one is parallax, point a telescope tangent to earths orbit. wait half a year and do the same thing (backwards, so you are looking in the same direction), based on the difference in the two pictures (closer things will look like they moved more) you can determine the distance because you know how big the earths orbit is.
The point is not to solve the energy problem, it is to decouple the energy production method from its consumption method. If cars run on gasoline then you can't power them with solar, nuclear or whatever other power because gasoline is well, refined oil. it doesn't matter that we don't have a hydrogen source now, but at some point in the future, we _will_ run out of gas or it will become so expensive that running cars off it is will cost much more than it is worth. when that does happen, we don't want the entire transportation infrastructure to collapse, if we are already using hydrogen cars, it doesn't matter, because the method of hydrogen production will just change to whatever the best method of producing energy is at the time.
switching to hydrogen is about protecting the country from a (nearish) future collapse, not about solving problems that exist now (not that we shouldn't solve those, they are being worked on too). also, the difference between the energy cost of a gallon of oil and the equivalant cost of energy from some other source is wasted money due only to the artifical demand for oil created by the ability of most cars to only run off of it. that is money flowing out of the country needlessly, which is bad for the economy.
I mean, it is not a good thing that we are so dependent on our cars, but it is the case, and mitigating that dependence is being worked on, but the dependence on oil specifically is a pressing concern for the near future.
Slashdot Mirror
Yeah, my favorite laptop is my 701c butterfly by IBM. runs linux well enough to ssh to my real machine and it's folding keyboacd never fails to impress. I would really like to see something like it made again, but with modern materials and batteries (all used to make it even smaller and lighter than before!)
Why do you assume they didn't do a computer simulation?
I don't know about that group in particular, but they most likely developed on a simulator, then implemented it real robots once they had a potential algorithm. These robots are also not very likely to be single purpose, They will most likely be used many times by other projects who were developing potential software with their simulators. Also, whoever got to build the robots got to learn about building robots and mechanical systems in general, likely as part of their course (much better than building things that will never be used).
I say potential because simulations are never perfect, and neither is the hardware (but in different ways) so they are only good up to a point.
the site is slashdotted, so I don't really know this for sure.. but it seems likely if this is similar to other university research projects and even ones in the commercial world as far as simulators are concerned.
Is that this is one of the first time (or perhaps the first time?) they were able to recover a meteorite this deep, while preserving and studying the soil above it and the track it made in the earth around it. Usually, one finds meteorites when their plow runs into one, causing the impact zone to be disturbed.
Don't get me wrong, I think the scientific space program is great, well worth the money, with many tangible benefits.
y
But the logic of increasing the size of our economy is flawed, see the following for a treatment of the issue.
http://en.wikipedia.org/wiki/Broken_window_fallac
in practice, a space program is nothing like a broken window, the technology and science it produces well exceeds the cost, but the act of spending the money alone isn't where the economy derives its benefits.
meteoroids are often going many kilometers a second. In order to maintain steady planetary orbit, everything up there must be traveling really very fast compared to the earth. However, chances are that if you hit them, they are not in the exact same orbit as you, meaning that no matter what, the relative speed difference between you two is going to be huge.
a meteoroid in space that is moving relatively slow compared to you would actually be a very novel thing to encounter. probablistically, if you are gonna be hit by anything unexpected, it is going to be something moving really really really fast from your perspective.
Don't forget 'Primer' http://imdb.com/title/tt0390384/ made for $7,000 and is excellent. The only movie I ever immediatly rewound and watched again. Also one of the few good time travel movies out there that doesn't gloss over the messy details of causality, but rather revels in them.
So I wonder why they haven't done it yet. They've got plenty of opportunity.
Well, the obvious conclusion is that they arn't nearly as prevalant as one thinks. Not to say terrorists don't exist. A lot of people hate one another for various reasons. but for the most part, people are not willing to kill inoccent bystanders or anyone actually. People willing to do so are few and far between.
Because for the most part, humans are decent folks to each other. We evolved as social creatures, we naturally follow social norms like not killing each other.
Things are patented for a specific purpose. So, his patent wouldn't disallow anyone else from producing the chemical for other reasons. But it would disallow anyone else selling it as mosquito repellant. It is not the chemical he is patenting, it is the idea of using it as mosquito repellant.
Rather than require source code be released, which might be tricky legally to define. I would prefer to see a law saying that the interface and specifications of hardware devices must be made available on request to anyone that purchases said device. You should be able to know what you are getting when you purchase something.
not true. with _credit cards_ only, specifically not debit cards, the law is on your side. The burden of proof is on them to show you bought something. This is not true with debit cards. no matter what guarentees banks say they give with them, the law is still on their side and they can legally tell you to go away and suck it up. With credit cards you can do the same to them.
the reason is that with debit cards, it is your money, losing it is the same as if someone swiped your wallet. No one refunds you if you can't catch the thief. With credit cards, it is not your money, it is the credit card companies money, so they are stealing from the credit card company, not you, so it is their burden to recover the money and they can't make you pay.
The main question is whether they use actual gyros. While accelerometer chips are fairly inexpensive, gyroscopes on a chip are still quite expensive. Accelerometers are fast and cheap but they only detect change, you have to integrate over time to get location which causes an ever growing loss of precision that builds up very quickly. I think but am not positive that this is what the PS3 uses. accelerometers are nice, better than nothing, but without an _absolute_ reference like a gyroscope to constantly recalibrate your accelerometers, you can't detect position or orientation. so you can jerk the controller around to move something, but you can't hold it still to position things, there is no way for the controler to tell whether you are holding it at an angle relative to gravity or twisting it (or some combination). The lag noticed on the PS3 demos is what makes me think it is accelerometer based, a standard way to figure out the difference between gravity and rotation is to wait and look at the signal and use a heuristic algorithm to determine what was happening after the fact, this introduces an inherent lag. you can't decide what was done with the controler without examining some future data.
:)
Now, the Wii also has accelerometers I assume, but in addition it has a stationary reference on top of the tv, if it uses this as its absolute positioning reference rather than a gyro, then they can cut a _substantial_ cost out of the per controller price, and get true 3d positioning to boot. as well as the advantage of no-lag interpretation of the accelerometer data.
Of course, economy of scale might have gotten nintendo a great deal on some real gyros. in which case I am going to pick up some extra controllers and canibalize them for parts
This is mainly speculation. On one hand, I have to hand it to nintendo for cleverness if they can get true good orientation data without gyros, on the other RC helicopter enthusiasts would love cheap accelerometer/gyro modules at consumer prices.
now the feature I really hope nintendo implemented is _directional rumble_ mounted on a gimbal so the rumble weight can be positioned inside the controller and give a kick in an arbitrary direction. imagine a sword fighting game where the rumble kick was always in the correct direction relative to what direction you are hitting the other guy (or his sword) from. that would be spectacular.
I was wondering about that too. My best therories as to the wacky coloring are that either someone found an old pillow case to make the covering out of, or they were studing the "twist" of the arm with computer vision, so having the multi-colored stripes along the length of the tentacle let the computer (or even the people) easily see how much the tentacle was being torqued and twisted around its axis. Of course, this is just speculation. I'll check out their papers if available to see if there is any insight there, the technology is quite interesting.
Just like the crazy idea of creating a protocol to allow networking a bunch of computers at separate locations together they funded a while ago. I am sure many didn't see the point then either. As we learn to swim the deep-end doesn't seem so deep any more. Plus, selection bias, DARPA funds a huge number of projects, only the ones with a certain type of appeal end up on the front page of slashdot. It's like the news, if you thought it was representative of a random sampling of society you would think the world was a pretty darn scary place, but they don't report on all the people getting along just fine having a long normal life because that is the norm.
Sounds like the perfect job for this really neat stuff (it goes by polymorph in europe)
http://www.shapelock.com/
I would say "The glass is twice as big as it needs to be"
google doesn't censor anything themselves, they simply run the chineese webcrawler from within the great firewall. the problem is with the great firewall not being too bright about mispellings, not anything google does. it's not googles job to make sure china gets their software right.
Just be prepared to wake up in a Death Cube and then wake up again, and again, and again...
Yes, but with a credit card, the law guarentees you don't have liability. the thief is stealing from the bank. not you. With a debit card, the bank may have a policy of limiting your liability, but that is purely part of their policy and who knows how they interpret it and I bet they have clauses in their policy making appeal of their decision quite difficult. With a debit card the law says the money was stolen from you, so it is ultimately your responsibility to recover your money. with a credit card, the money was stolen from the bank so it is the banks responsibility.
the interesting thing is that quantum systems scale superlinearly with the number of qubits relative to standard computers.
so if we assume vanilla computers double in power every year, in order for quantum computers to keep up, one needs only to produce a quantum computer with _one_ more qubit every year.
of course, there is some subtly in the meaning of 'keeping up' since the problems practically solvable by each are different so you can't compare them directly, but you can compare their growth in power over time, which is what moores law is about.
Actually, the other choice is not that it is quantum random, it is that _it doesn't matter_. Science is only concerned with things that are testable and observable. Science is agnostic (or even apatheist), but not inherently athiest. (though scientists themselves come in those and many othen varieties.)
Also, there are many TOEs that don't involve randomness FYI. Randomness is an observed characteristic of quantum mechanics, the book is still out as to whether it is a defining one.
It is because apple doesn't have an effective monopoly and use it to obtain monopolies in other markets. The laws are not about bundling software, they are about unfair business practices.
admittedly, those are subject to interpretation, as is the term monopoly but that is why they don't go after apple. Microsoft does have more money to milk, but this is a case of coorelation not causation, the monopoly holder is the one making the money off their unfair business practices. Some practices are only unfair if you hold a huge majority of the market as you are free to manipulate it unfairly in ways other companies cannot.
but day trading does, and people do that, often loosing a large amount of their investment. So this guy fared pretty well.
This is a common misconception of what Occam's Razor is. It has absolutly nothing to do with "simplicity". It's exact wording is
one should not increase, beyond what is necessary, the number of entities required to explain anything.
An example would be saying that "The man died because of his heart attack and he was wearing blue pants". the "blue pants" are not needed to explain the fact he died, so can be dropped from the assertion without changing its validity. The application to the ID situation is a whole lot more obvious now and doesn't use overloaded terms like 'simplicity'.
Just because you say a building is 100 meters tall, that doesn't mean you took a meter stick and used it stepwise all the way up to determine how tall it was. you could have done many things, measured a line away from it and figure out the exact angle you look up at to see the top. You could know the inclination of the sun and measure the length of the shadow.
likewise, when you say something in 5 lightyears away, that doesn't mean you measured it by bouncing a signal off it and it took exactly 10 years to get back. just as there are lots of ways to measure things on earth, there are lots of ways to measure them in space.
A common one is parallax, point a telescope tangent to earths orbit. wait half a year and do the same thing (backwards, so you are looking in the same direction), based on the difference in the two pictures (closer things will look like they moved more) you can determine the distance because you know how big the earths orbit is.
The point is not to solve the energy problem, it is to decouple the energy production method from its consumption method. If cars run on gasoline then you can't power them with solar, nuclear or whatever other power because gasoline is well, refined oil. it doesn't matter that we don't have a hydrogen source now, but at some point in the future, we _will_ run out of gas or it will become so expensive that running cars off it is will cost much more than it is worth. when that does happen, we don't want the entire transportation infrastructure to collapse, if we are already using hydrogen cars, it doesn't matter, because the method of hydrogen production will just change to whatever the best method of producing energy is at the time.
switching to hydrogen is about protecting the country from a (nearish) future collapse, not about solving problems that exist now (not that we shouldn't solve those, they are being worked on too). also, the difference between the energy cost of a gallon of oil and the equivalant cost of energy from some other source is wasted money due only to the artifical demand for oil created by the ability of most cars to only run off of it. that is money flowing out of the country needlessly, which is bad for the economy.
I mean, it is not a good thing that we are so dependent on our cars, but it is the case, and mitigating that dependence is being worked on, but the dependence on oil specifically is a pressing concern for the near future.