Don't take this the wrong way, but you have a pretty weird view of the US vs. the rest of the world.
You are basing the entire argument on the premise that the US and/or its citizens are somehow different from the rest of the world. You may not be aware of it, but most intellectual property is produced outside of the US. Android itself is based on Linux which originated in Finland. Korea, where HTC is from, files 114 patents per $billion GDP compared to 18,6 for the US. (Yes, that's more than six times as much!)
If anything is holding America back, it is that obsolete we-and-them kind of attitude that stipulates that manufacturing cannot be done in the US, and no one else can produce as much or as high quality intellectual property.
Linux, like Android, is an excellent piece of software, but you vastly overestimate the draw for 'open' to the typical end user.
But the thing is that "open" is important to industry. What makes Android successful is that it provides an even playing field for all device manufacturers, since Google really isn't making devices (the Nexus One is a rebranded HTC). Symbian is controlled by Nokia, iOS is not available to anyone but Apple and Windows has failed on mobile devices. Android, on the other hand, provides a standardized platform with a large user base and rapidly growing application ecosystem, it is owned by a fairly neutral party and it is open.
It looks very much like decompiled code. If I recall correctly, all member variable names would be available in the compiled Java class file, while the method argument names would not. In the Android version, all these arguments have generic names derived from the type (for instance: Set set, Set set1), as would be expected if the code were auto-generated from a compiled file.
Or if I put it like this: The Android version matches exactly what I would expect a decompiler to generate, given a compiled version of the Java code.
I'm not sure how large databases these bots use, but a human brain has been primed with an extreme amount of information too, throughout the person's life. Every conversation you have ever heard may have influence in your ability to formulate new statements.
Nevertheless, why should the competition rules limit the allowed solutions to emulating of the human brain's methods of achieving intelligence? If a computer can repeatedly pass the Turing test and convince other humans that it is human even when they try their hardest to expose it, should it then not be considered intelligent, no matter how the intelligent behavior is implemented?
If on the other hand the ability to conceptualize (whatever that really is) is important for intelligence, then surely we should be able to ask the computer questions that requires it to conceptualize, and in that tell it apart from humans.
The point is that it doesn't matter how the bot fools the judges, as long as it does. If repeating data from a huge database is enough to seem so intelligent that it's impossible to discern from "real" intelligence, then it is by definition intelligence. And the fact is that we don't know how the human brain represents understanding of concepts, or what it means to understand something. Perhaps we too just repeat aggregates from a huge database of previous experiences.
The British (or Western Europe rather, if we are talking about a civilization) never really fell with the loss of their colonies. Instead, you should look at examples such as the Hittite Empire, the Maya civilization, the Mycenaean Greece, the Khmer Empire or the Western Roman Empire.
You deliberately cut the sentence in half. Increasing demand results in increasing production in the long term (supply goes up to meet the demand), which means more investments and research in the solar cell industry, which will lead to cheaper production technology. Look at any high volume semi-conductor industry and you will see examples of this. Producing a single computer monitor is very expensive but producing one million is not that expensive, per monitor. Volume is the key here.
While technology might very well "save us" once again, it's a bit audacious to assume that it always will in the future. Civilizations have fallen before, and all of them could probably have argued in a similar way before the end: It has worked fine up until now, so why shouldn't it continue to?
I actually think energy is one of the easier problems to solve -- solar cells will drop in price as demand increases and technology advances, and the sun provides orders of magnitude more power than we have use for at the moment. But if you look at almost any other natural resource, demands are increasing at an exponential rate. Since resources are limited, it is impossible for this to continue for very long. I have no doubt that society will adapt, the question is how disruptive the changes will be. At the moment, it appears that some prominent economies think that even reducing oil consumption is out of the question due to the economical effects it would have.
...or it means that we are living off our "savings" at the moment: cutting down forests faster than we plant new ones, using up ground water reserves, depleting farmland soil of nutrients and so on. The fact that we are surviving at this moment does not mean that the current situation is sustainable.
I'm not familiar with the details of the patent system of the USA, but normally the requirement is that it shouldn't be obvious the someone "skilled in the art" and having access to all available knowledge in the field. For instance, if the invention offers a solution to a problem that skilled people have been trying to solve for some time, it cannot be considered obvious. In practice, however, I think the requirements are quite a bit lower.
Thinking about it, I wonder what the consequences would be if that specific requirement were added: The problem that the invention solves must have been known for at least X years.
I'm not sure I'm following you. The point of the heat pump is that, instead of simply heating the inside of the building, you use the energy to increase the temperature difference between the exterior and interior of the building. By doing this, it is possible to get a bigger increase in interior temperature than by converting the energy directly to heat, for the same amount of input energy. This is because the average temperature of the combined system (the house and its surroundings) doesn't have to increase as much, only its entropy has to decrease. The "price" you pay is that the exterior temperature decreases slightly, but since this is usually either the ground or the surrounding atmosphere, depending on the kind of heat pump, you will never notice the difference.
Electric heating is often less efficient than for instance district heating, since the latter doesn't involve conversion from heat to electricity and back to heat again.
And in fact, an electric radiator (or a light bulb) is not the most efficient form of electric heating either. An electric heat pump is more efficient than turning the electricity straight into heat, as it exploits the inherent order ("lack of" entropy) in the electric energy. That is, you can in a sense get more than 100% efficiency by increasing the temperature difference between the exterior and interior of your house, rather than just increasing the temperature of the interior.
Yes, the contrast between the systematic, sometimes almost mechanical, way of seeing things in Japanese budo tradition as opposed to the more organic and paradigm based Chinese approach is quite interesting, and very obvious if you ever come into contact with both cultures. Not only in the ranking system (or lack thereof).
Where a Chinese would perhaps tell you to, let's say, move your arm like a serpent, a Japanese instructor would deconstruct the motion into steps and tell you the how many degrees you should bend you elbow at certain points.
I'm not sure how you define a traditional system, but out of the five Japanese martial arts disciplines that I have practiced none that I can recall had 5 kyu (colored blets). I would say 6 kyu is the most common, but 10 kyu systems also exist. But you could be right about american dojos treating martial arts as a business rather than an art. In Europe, and as far as I know, Japan, trainers are almost always teaching without any compensation at all. And this includes even the most proficient masters. You teach because it means doing a service to the art -- some even travel to other countries for a few years to establish dojos and try to spread the art. Such efforts are often considered, apart from mental and physical skill, when handing out the higher ranked black belts.
What you describe might once have been true, and might to an extent still be true, but things are changing rapidly. The general attitude in China nowadays is very optimistic and forward-looking. They want to get back to the former glory and importance China once had, and want to show that they can match the western world.
This kind of thinking that we are somehow superior is a bit dangerous, and our attempts to create international legal frameworks preventing the developing world from using our ideas are in the long run mostly going to prevent us from using the ideas by the remaining 90% of the world's population.
What you describe I think is also the reason why the USA in particular and the western world in general are so focused on intellectual property protection at the political level. It is based on the notion that the western world has knowledge and ideas that are somehow inaccessible to the rest of the world unless they get it from us, and that we must protect them from leaking out to prevent the rest of the world to catch up and compete with us.
The reality however is that new inventions and ideas are generated every day at an amazing pace, and I wager to say that most of them appear outside of the western world. Hint: China graduates something like 300 000 new engineers each year, and they are no less intelligent than we are. Whatever exclusive knowledge we have today will be commonly available to the entire world tomorrow. The biggest losers when we try to inhibit the free exchange of knowledge and technology is not the rest of the world -- they can carry on perfectly well without our help -- the biggest losers are going to be ourselves.
I do not see an AI problem (unless you would want to remove the pilot entirely, which is a bit extreme to say the least) -- what I see is a (potential) information system and/or procedural problem.
From what you describe, I get the impression that we have created a less than ideal situation for commercial pilots, with regard to handling diverse information in-flight. I appreciate what you are saying, but I haven't yet been convinced that there is nothing that can be done technically to improve situation awareness and decision support. Obviously, there are all kinds of complicated judgments that only a human can make, but the technological system should support the pilot in making them. I can imagine ways in which almost all you the considerations you mentioned might be simplified with the right information, computer systems or possibly regulation changes.
Non-pilots think the work required is simple control system theory, just need a fancier autopilot. Can't you replace that whole paragraph about with a simple linear equation or something?
I'm not sure which non-pilots you are talking about. I certainly do not think that, and I didn't try to argue that a pilot can be replaced by a computer.
The OP was discussing how a co-pilot is required, if not for redundancy, simply to handle the sheer amount of information and communication surrounding a flight landing or take-off. My question was: Is this apparent information overload an inherent property of flying, or could it be alleviated by using modern technology?
Since I indeed am no pilot, it is hard for me to argue with you (and in fact, I wasn't try to argue in the first place -- I was posing an honest question), but just to take an example of what seems strange to an outsider: Why should there normally need to be any doubt about what the NOTAM means? Couldn't common events be standardized, and perhaps integrated into the avionics? Why do you even need to understand English to interpret it? Couldn't it just pop up graphically on a map somewhere, maybe automatically alerting you if it interferes with any of your potential routes?
Or am I totally off, and it really fundamentally needs to be as bad as you describe?
[...] it's the sheer number of tasks at hand -- between monitoring a zillion instruments and talking to approach, then the tower, then the ground -- that you just need a second person there.
Out of curiosity: Could some of these tasks and procedures be simplified, perhaps with the help of technology? For instance, exactly what information does the pilot need from/provide to the approach, tower and ground? Couldn't any of this be sent automatically by computers?
I realize it would not be possible over-night, but in security oriented and strictly regulated contexts, I often get the feeling that things are done the way they are mostly out of historical or conservative reasons, rather than it being the optimal way work with current technology. I suspect it has to do in part with the fact that if you change a security procedure and there is an accident because of it, you are likely to get the blame, but if security increases slightly nobody notices much. So there is really no incentive to change, unless there is evidence that change is needed (e.g. an accident that could have been prevented if procedures had been different). And of course, too much change in a short period of time can be dangerous, if it conflicts with old habits or requires new training.
The existence of the human brain shows that physics allows for three dimensional computing structures with very high processing power and low energy use and consequently low heat dissipation to exist. In other words, efficient computation (by todays standards) is not fundamentally limited to flat objects, if you can exploit parallelism.
Slashdot Mirror
Don't take this the wrong way, but you have a pretty weird view of the US vs. the rest of the world.
You are basing the entire argument on the premise that the US and/or its citizens are somehow different from the rest of the world. You may not be aware of it, but most intellectual property is produced outside of the US. Android itself is based on Linux which originated in Finland. Korea, where HTC is from, files 114 patents per $billion GDP compared to 18,6 for the US. (Yes, that's more than six times as much!)
Notwithstanding the fact that Google isn't at all doing badly "giving away" stuff.
If anything is holding America back, it is that obsolete we-and-them kind of attitude that stipulates that manufacturing cannot be done in the US, and no one else can produce as much or as high quality intellectual property.
Linux, like Android, is an excellent piece of software, but you vastly overestimate the draw for 'open' to the typical end user.
But the thing is that "open" is important to industry. What makes Android successful is that it provides an even playing field for all device manufacturers, since Google really isn't making devices (the Nexus One is a rebranded HTC). Symbian is controlled by Nokia, iOS is not available to anyone but Apple and Windows has failed on mobile devices. Android, on the other hand, provides a standardized platform with a large user base and rapidly growing application ecosystem, it is owned by a fairly neutral party and it is open.
It looks very much like decompiled code. If I recall correctly, all member variable names would be available in the compiled Java class file, while the method argument names would not. In the Android version, all these arguments have generic names derived from the type (for instance: Set set, Set set1), as would be expected if the code were auto-generated from a compiled file.
Or if I put it like this: The Android version matches exactly what I would expect a decompiler to generate, given a compiled version of the Java code.
I'm not sure how large databases these bots use, but a human brain has been primed with an extreme amount of information too, throughout the person's life. Every conversation you have ever heard may have influence in your ability to formulate new statements.
Nevertheless, why should the competition rules limit the allowed solutions to emulating of the human brain's methods of achieving intelligence? If a computer can repeatedly pass the Turing test and convince other humans that it is human even when they try their hardest to expose it, should it then not be considered intelligent, no matter how the intelligent behavior is implemented?
If on the other hand the ability to conceptualize (whatever that really is) is important for intelligence, then surely we should be able to ask the computer questions that requires it to conceptualize, and in that tell it apart from humans.
The point is that it doesn't matter how the bot fools the judges, as long as it does. If repeating data from a huge database is enough to seem so intelligent that it's impossible to discern from "real" intelligence, then it is by definition intelligence. And the fact is that we don't know how the human brain represents understanding of concepts, or what it means to understand something. Perhaps we too just repeat aggregates from a huge database of previous experiences.
The British (or Western Europe rather, if we are talking about a civilization) never really fell with the loss of their colonies. Instead, you should look at examples such as the Hittite Empire, the Maya civilization, the Mycenaean Greece, the Khmer Empire or the Western Roman Empire.
Do you think the daily life of the average person is independent of a functioning society and commercial markets?
You deliberately cut the sentence in half. Increasing demand results in increasing production in the long term (supply goes up to meet the demand), which means more investments and research in the solar cell industry, which will lead to cheaper production technology. Look at any high volume semi-conductor industry and you will see examples of this. Producing a single computer monitor is very expensive but producing one million is not that expensive, per monitor. Volume is the key here.
While technology might very well "save us" once again, it's a bit audacious to assume that it always will in the future. Civilizations have fallen before, and all of them could probably have argued in a similar way before the end: It has worked fine up until now, so why shouldn't it continue to?
I actually think energy is one of the easier problems to solve -- solar cells will drop in price as demand increases and technology advances, and the sun provides orders of magnitude more power than we have use for at the moment. But if you look at almost any other natural resource, demands are increasing at an exponential rate. Since resources are limited, it is impossible for this to continue for very long. I have no doubt that society will adapt, the question is how disruptive the changes will be. At the moment, it appears that some prominent economies think that even reducing oil consumption is out of the question due to the economical effects it would have.
...or it means that we are living off our "savings" at the moment: cutting down forests faster than we plant new ones, using up ground water reserves, depleting farmland soil of nutrients and so on. The fact that we are surviving at this moment does not mean that the current situation is sustainable.
Obviously, but that would have spoiled the humor of it.
"The price may be too high and too hidden, but it's not that that doesn't make the value provided any less meaningless."?
Fixed the negation parity for you.
There is this saying, which I think is very true: "Children do not do as you say -- they do what you do."
Kids learn by copying their parents. You may get short term obedience with words, but in the long run it's your own behavior that matters.
I'm not familiar with the details of the patent system of the USA, but normally the requirement is that it shouldn't be obvious the someone "skilled in the art" and having access to all available knowledge in the field. For instance, if the invention offers a solution to a problem that skilled people have been trying to solve for some time, it cannot be considered obvious. In practice, however, I think the requirements are quite a bit lower.
Thinking about it, I wonder what the consequences would be if that specific requirement were added: The problem that the invention solves must have been known for at least X years.
I'm not sure I'm following you. The point of the heat pump is that, instead of simply heating the inside of the building, you use the energy to increase the temperature difference between the exterior and interior of the building. By doing this, it is possible to get a bigger increase in interior temperature than by converting the energy directly to heat, for the same amount of input energy. This is because the average temperature of the combined system (the house and its surroundings) doesn't have to increase as much, only its entropy has to decrease. The "price" you pay is that the exterior temperature decreases slightly, but since this is usually either the ground or the surrounding atmosphere, depending on the kind of heat pump, you will never notice the difference.
Electric heating is often less efficient than for instance district heating, since the latter doesn't involve conversion from heat to electricity and back to heat again.
And in fact, an electric radiator (or a light bulb) is not the most efficient form of electric heating either. An electric heat pump is more efficient than turning the electricity straight into heat, as it exploits the inherent order ("lack of" entropy) in the electric energy. That is, you can in a sense get more than 100% efficiency by increasing the temperature difference between the exterior and interior of your house, rather than just increasing the temperature of the interior.
[...] and people are already suing the manufacturers, e.g. Toyota, claiming that those systems malfunctioned after a crash.
Well, obviously the systems are gonna stop working if you crash the car!
Yes, the contrast between the systematic, sometimes almost mechanical, way of seeing things in Japanese budo tradition as opposed to the more organic and paradigm based Chinese approach is quite interesting, and very obvious if you ever come into contact with both cultures. Not only in the ranking system (or lack thereof).
Where a Chinese would perhaps tell you to, let's say, move your arm like a serpent, a Japanese instructor would deconstruct the motion into steps and tell you the how many degrees you should bend you elbow at certain points.
I'm not sure how you define a traditional system, but out of the five Japanese martial arts disciplines that I have practiced none that I can recall had 5 kyu (colored blets). I would say 6 kyu is the most common, but 10 kyu systems also exist. But you could be right about american dojos treating martial arts as a business rather than an art. In Europe, and as far as I know, Japan, trainers are almost always teaching without any compensation at all. And this includes even the most proficient masters. You teach because it means doing a service to the art -- some even travel to other countries for a few years to establish dojos and try to spread the art. Such efforts are often considered, apart from mental and physical skill, when handing out the higher ranked black belts.
What you describe might once have been true, and might to an extent still be true, but things are changing rapidly. The general attitude in China nowadays is very optimistic and forward-looking. They want to get back to the former glory and importance China once had, and want to show that they can match the western world.
This kind of thinking that we are somehow superior is a bit dangerous, and our attempts to create international legal frameworks preventing the developing world from using our ideas are in the long run mostly going to prevent us from using the ideas by the remaining 90% of the world's population.
What you describe I think is also the reason why the USA in particular and the western world in general are so focused on intellectual property protection at the political level. It is based on the notion that the western world has knowledge and ideas that are somehow inaccessible to the rest of the world unless they get it from us, and that we must protect them from leaking out to prevent the rest of the world to catch up and compete with us.
The reality however is that new inventions and ideas are generated every day at an amazing pace, and I wager to say that most of them appear outside of the western world. Hint: China graduates something like 300 000 new engineers each year, and they are no less intelligent than we are. Whatever exclusive knowledge we have today will be commonly available to the entire world tomorrow. The biggest losers when we try to inhibit the free exchange of knowledge and technology is not the rest of the world -- they can carry on perfectly well without our help -- the biggest losers are going to be ourselves.
I do not see an AI problem (unless you would want to remove the pilot entirely, which is a bit extreme to say the least) -- what I see is a (potential) information system and/or procedural problem.
From what you describe, I get the impression that we have created a less than ideal situation for commercial pilots, with regard to handling diverse information in-flight. I appreciate what you are saying, but I haven't yet been convinced that there is nothing that can be done technically to improve situation awareness and decision support. Obviously, there are all kinds of complicated judgments that only a human can make, but the technological system should support the pilot in making them. I can imagine ways in which almost all you the considerations you mentioned might be simplified with the right information, computer systems or possibly regulation changes.
Non-pilots think the work required is simple control system theory, just need a fancier autopilot. Can't you replace that whole paragraph about with a simple linear equation or something?
I'm not sure which non-pilots you are talking about. I certainly do not think that, and I didn't try to argue that a pilot can be replaced by a computer.
The OP was discussing how a co-pilot is required, if not for redundancy, simply to handle the sheer amount of information and communication surrounding a flight landing or take-off. My question was: Is this apparent information overload an inherent property of flying, or could it be alleviated by using modern technology?
Since I indeed am no pilot, it is hard for me to argue with you (and in fact, I wasn't try to argue in the first place -- I was posing an honest question), but just to take an example of what seems strange to an outsider: Why should there normally need to be any doubt about what the NOTAM means? Couldn't common events be standardized, and perhaps integrated into the avionics? Why do you even need to understand English to interpret it? Couldn't it just pop up graphically on a map somewhere, maybe automatically alerting you if it interferes with any of your potential routes?
Or am I totally off, and it really fundamentally needs to be as bad as you describe?
[...] it's the sheer number of tasks at hand -- between monitoring a zillion instruments and talking to approach, then the tower, then the ground -- that you just need a second person there.
Out of curiosity: Could some of these tasks and procedures be simplified, perhaps with the help of technology? For instance, exactly what information does the pilot need from/provide to the approach, tower and ground? Couldn't any of this be sent automatically by computers?
I realize it would not be possible over-night, but in security oriented and strictly regulated contexts, I often get the feeling that things are done the way they are mostly out of historical or conservative reasons, rather than it being the optimal way work with current technology. I suspect it has to do in part with the fact that if you change a security procedure and there is an accident because of it, you are likely to get the blame, but if security increases slightly nobody notices much. So there is really no incentive to change, unless there is evidence that change is needed (e.g. an accident that could have been prevented if procedures had been different). And of course, too much change in a short period of time can be dangerous, if it conflicts with old habits or requires new training.
The existence of the human brain shows that physics allows for three dimensional computing structures with very high processing power and low energy use and consequently low heat dissipation to exist. In other words, efficient computation (by todays standards) is not fundamentally limited to flat objects, if you can exploit parallelism.