I sure hope this isn't really true. If mathematicans aren't really interested in helping understand the world, why should society fund them? I certainly know that a major motivation for my career in science is that understanding the world through science will help people, cure diseases, etc.
This is a pretty trollish statement that overlooks a few obvious facts about "society". First, society, by which I mean the American federal government, funds all sorts of things that don't help to understand the world. For instance, the federal government gives nonprofit groups, primarily churchs, tax-exemption. These groups certainly do not help to understand the world, in a scientific sense. The government also funds public television, public radio, and arts, via the NEA. These activities don't help to understand the world. Also, the government funds a substantial amount of research in applied science via the defense department and the various branches of the military. Though much of this research is "basic", the obvious hope is that this research will pay-off in advanced military technology, not "understanding the world"
Second, I think that mathematics alone is the one science that contributes the most to our understanding of the world. There are the obvious, innumerable applications of mathematics in science and engineering. But applied mathematics presupposes the availability of a bulk set of theorems, proofs, definitions, etc. from pure mathematics. Pure mathematics is the only scientific venture that contributes timeless, unchanging truth. Mathematical facts are not based on empirical evidence, which is mutable, but on immutable logical truth. Mathematical truths are erected upon rigorous logic, not the faulty and error-prone business of empirical scientific observation. These truths are true in a real and abstract sense; they never fade with time. If the discovery of permanent, immutable truth is not a contribution to our understanding of the world, then I don't know what is.
Marvin Minsky is an idiot
on
AI Going Nowhere?
·
· Score: 4, Insightful
Let's not forget that Minsky (along with Papert) declared that single-layer perceptrons were basically useless and there was no reason to suspect that multilayer varieties would be any better. That little, wrong statement basically shut down all neural networks research for ~20 years. Research didn't recover until the mid 1980's with the discovery of the backpropagation algorithm, along with other neural architectures such as the Hopfield network, ART, and the self-organizing map. Advances in neural modeling have helped theoretical neuroscientists begin to understand the nature of neural processing in the brain, no thanks to Minsky.
Minksy belongs to the old school of AI thinking. These guys believe that it is possible to make statements about intelligence itself, without considering the interactions of the organism/agent with its environment or the underlying architecture of the brain/CPU. I think that the total failure of this style of thinking to produce anything interesting in 50 years proves that this approach is sterile. Minsky laments the fact that graduate students build robots, but this activity exposes students to the challenges of constructing a device that must actually interact with the environment. It is ridiculous to assume that you could design a system capable of intelligent behavior without ever confronting the problems of sensors and actuators. Almost every part of the brain is devoted to processing raw sensory input or generating motor output. One cannot simply design an intelligent system without worrying about sensory input and behavioral output. The CYC project of Lenat has the laudable goal of teaching a machine "common sense" by hard coding a vast database of simple statements like "Trees cannot walk". This is a totally wrongheaded approach to learning and reasoning, and is typical of old school, hard AI.
We will only make progress in engineering intelligent, adaptive systems by studying actual examples of intelligent, adaptive systems, namely animals. Neuroscientists and psychologists are beginning to embrace the tools of mathematical modeling and simulation wo help explain nervous system structure and function. Computer scientists would do well to similarly embrace the work of experimental neuroscience.
Super Mario Bros. Think about it: Mario looks exactly like Ron Jeremy in suspenders. He's a plumber, and what do plumbers do? They "lay pipe". He runs around eating mushrooms that make him grow in "stature". So basically you play a drug-addled, size-enhanced porn star in suspenders who chases after a girl. Pretty disturbing stuff, if you ask me.
People that make assertions like this grossly overestimate the current level of knowledge about the brain, especially about the way(s) information is processed there. For starters we don't know what the units of information are in the brain. Are they spikes or spike rates? Are synapses the basis loci of computation or are neurons or supraneuronal structures such as columns? Is temporal synchrony important and if so, how does the brain achieve it? Is it even meaningful to talk about the brain this way? No one in neuroscience has answered these questions satisfactorily, nor is anyone seriously working on providing answers. Until neuroscience matures significantly and addresses these basic questions, the brain will remain largely mysterious and theoretical neuroscience will remain a speculative endeavour.
Talk of brains interfacing with computers in some sort of direct manner is basically sci-fi/pseudoscientific junk. We can't do anything like that today, and we are probably several decades away from being able to even try.
First, the "technical" books I'm most interested in are math books, which don't suffer the troubling problem of being out of date the minute they hit the press. I am also a programmer, so I like and need books on central programming topics, such as the C++ standard library, numerical algorithms, and computational geometry. These, also, are not "bleeding edge" topics and so are likely to be in my library.
Second, my library is not the local neighborhood library, but the Science and Engineering Library at Boston University (I'm a grad student here). It takes me all of thirty seconds to punch in the author into the library's search website to find out if the book has been checked out already. If it has and I need it badly, I can just put in a recall on the book. Also, as a graduate student, I get to check out books for 19 weeks. After 3 weeks, they are subject to recall, but that is not a big deal. Since this is a science and engineering library at a major university, it is relatively well-stocked with many recent CS/tech/programming books.
As for searching, most books have indices and tables of content. These radical advances in information management work surprisingly well - try them sometime!
Finally, I agree that if I were a professional coder or administrator working in IT, I would probably have a need for the lastest book on sendmail or MySQL or buzzword X. If that were the case, Safari might be for me. Since that is not the case, I'll pass on Safari. Besides, as other posters have noted, personal dead-tree libraries can be quite impressive. I hope mine is one day.
I'm sorry, but reading a book on a computer just doesn't cut it for me. If I'm serious about a book, then I'll shell out the bucks and buy the damn thing. Otherwise, I'll hoof it down to the library and check it out. Libraries are cheaper than this Safari system and have the added benefit of not ruining your eyes and/or fraying your nerves by making you read a friggin book on your computer screen. Maybe one day I'll be more convinced by the concept of e-books, but until then, I'll stick to the dead-tree variety.
The biggest selling point for me is D's direct, language-level support of complex and imaginary number data types. The C/C++ treatment of this issue is a hack at best. Delphi uses variants, which I find distasteful, and Java ignores the issue altogether. So for me, a developer who knows C++ and wants to write scientific applications, it looks like D has just the features I want.
I happened to notice that Orbitz.com apparently has a no linking policy. Which is OK, considering that nearly every major website out there vomits forth an Orbitz.com pop-up. You can't link to them, but they can bombard you with pop-ups. Jerks.
Actually, a FUD is a statement designed to instill fear, uncertainty, and doubt. According to the above quote, our friend Bill told an outright lie. Nice LIE Bill.
Screw the Oscars and screw Hollywood. Let's not forget that the fucking bastards that won't (ever) give Best Picture to a film like LotR are the same ones that are pushing SSSCA (or CBDTPA) through Congress right now. Boycott Hollywood!
The DOJ wants to target five types of people: drug offenders, legalization advocates, anarchists, and free-speech pushers. I only count four types of people there. Who are the mysterious fifth target group? Presumably terrorists, but that's hard to infer from the tone of this article. It seems that the DOJ thinks that ravers and First Amendment advocates are a bigger threat to America than Al-Qaida. Something's fucked up here.
This review mentions that Queen Amidala and Jar-Jar represent their respective planets in the senate. Which is interesting, considering that they are both from the same planet, at least in the mediocre Episode I that I saw. I'm looking forward to the scene in Episode II when Naboo is split in two by some cataclysmic force. Or maybe this so-called reviewer is just an idiot. You be the judge.
Slashdot Mirror
This is a pretty trollish statement that overlooks a few obvious facts about "society". First, society, by which I mean the American federal government, funds all sorts of things that don't help to understand the world. For instance, the federal government gives nonprofit groups, primarily churchs, tax-exemption. These groups certainly do not help to understand the world, in a scientific sense. The government also funds public television, public radio, and arts, via the NEA. These activities don't help to understand the world. Also, the government funds a substantial amount of research in applied science via the defense department and the various branches of the military. Though much of this research is "basic", the obvious hope is that this research will pay-off in advanced military technology, not "understanding the world"
Second, I think that mathematics alone is the one science that contributes the most to our understanding of the world. There are the obvious, innumerable applications of mathematics in science and engineering. But applied mathematics presupposes the availability of a bulk set of theorems, proofs, definitions, etc. from pure mathematics. Pure mathematics is the only scientific venture that contributes timeless, unchanging truth. Mathematical facts are not based on empirical evidence, which is mutable, but on immutable logical truth. Mathematical truths are erected upon rigorous logic, not the faulty and error-prone business of empirical scientific observation. These truths are true in a real and abstract sense; they never fade with time. If the discovery of permanent, immutable truth is not a contribution to our understanding of the world, then I don't know what is.
Minksy belongs to the old school of AI thinking. These guys believe that it is possible to make statements about intelligence itself, without considering the interactions of the organism/agent with its environment or the underlying architecture of the brain/CPU. I think that the total failure of this style of thinking to produce anything interesting in 50 years proves that this approach is sterile. Minsky laments the fact that graduate students build robots, but this activity exposes students to the challenges of constructing a device that must actually interact with the environment. It is ridiculous to assume that you could design a system capable of intelligent behavior without ever confronting the problems of sensors and actuators. Almost every part of the brain is devoted to processing raw sensory input or generating motor output. One cannot simply design an intelligent system without worrying about sensory input and behavioral output. The CYC project of Lenat has the laudable goal of teaching a machine "common sense" by hard coding a vast database of simple statements like "Trees cannot walk". This is a totally wrongheaded approach to learning and reasoning, and is typical of old school, hard AI.
We will only make progress in engineering intelligent, adaptive systems by studying actual examples of intelligent, adaptive systems, namely animals. Neuroscientists and psychologists are beginning to embrace the tools of mathematical modeling and simulation wo help explain nervous system structure and function. Computer scientists would do well to similarly embrace the work of experimental neuroscience.
Minsky is a dinosaur.
This is a fine example of thinking inside the box. You need to get out more.
Super Mario Bros. Think about it: Mario looks exactly like Ron Jeremy in suspenders. He's a plumber, and what do plumbers do? They "lay pipe". He runs around eating mushrooms that make him grow in "stature". So basically you play a drug-addled, size-enhanced porn star in suspenders who chases after a girl. Pretty disturbing stuff, if you ask me.
People that make assertions like this grossly overestimate the current level of knowledge about the brain, especially about the way(s) information is processed there. For starters we don't know what the units of information are in the brain. Are they spikes or spike rates? Are synapses the basis loci of computation or are neurons or supraneuronal structures such as columns? Is temporal synchrony important and if so, how does the brain achieve it? Is it even meaningful to talk about the brain this way? No one in neuroscience has answered these questions satisfactorily, nor is anyone seriously working on providing answers. Until neuroscience matures significantly and addresses these basic questions, the brain will remain largely mysterious and theoretical neuroscience will remain a speculative endeavour.
Talk of brains interfacing with computers in some sort of direct manner is basically sci-fi/pseudoscientific junk. We can't do anything like that today, and we are probably several decades away from being able to even try.
First, the "technical" books I'm most interested in are math books, which don't suffer the troubling problem of being out of date the minute they hit the press. I am also a programmer, so I like and need books on central programming topics, such as the C++ standard library, numerical algorithms, and computational geometry. These, also, are not "bleeding edge" topics and so are likely to be in my library.
Second, my library is not the local neighborhood library, but the Science and Engineering Library at Boston University (I'm a grad student here). It takes me all of thirty seconds to punch in the author into the library's search website to find out if the book has been checked out already. If it has and I need it badly, I can just put in a recall on the book. Also, as a graduate student, I get to check out books for 19 weeks. After 3 weeks, they are subject to recall, but that is not a big deal. Since this is a science and engineering library at a major university, it is relatively well-stocked with many recent CS/tech/programming books.
As for searching, most books have indices and tables of content. These radical advances in information management work surprisingly well - try them sometime!
Finally, I agree that if I were a professional coder or administrator working in IT, I would probably have a need for the lastest book on sendmail or MySQL or buzzword X. If that were the case, Safari might be for me. Since that is not the case, I'll pass on Safari. Besides, as other posters have noted, personal dead-tree libraries can be quite impressive. I hope mine is one day.
I'm sorry, but reading a book on a computer just doesn't cut it for me. If I'm serious about a book, then I'll shell out the bucks and buy the damn thing. Otherwise, I'll hoof it down to the library and check it out. Libraries are cheaper than this Safari system and have the added benefit of not ruining your eyes and/or fraying your nerves by making you read a friggin book on your computer screen. Maybe one day I'll be more convinced by the concept of e-books, but until then, I'll stick to the dead-tree variety.
Damn - I knew that one of these days clicking through those EULAs was going to bite me in the ass!
The biggest selling point for me is D's direct, language-level support of complex and imaginary number data types. The C/C++ treatment of this issue is a hack at best. Delphi uses variants, which I find distasteful, and Java ignores the issue altogether. So for me, a developer who knows C++ and wants to write scientific applications, it looks like D has just the features I want.
I happened to notice that Orbitz.com apparently has a no linking policy. Which is OK, considering that nearly every major website out there vomits forth an Orbitz.com pop-up. You can't link to them, but they can bombard you with pop-ups. Jerks.
Actually, a FUD is a statement designed to instill fear, uncertainty, and doubt. According to the above quote, our friend Bill told an outright lie. Nice LIE Bill.
I'm running Opera identifying as IE.
Screw the Oscars and screw Hollywood. Let's not forget that the fucking bastards that won't (ever) give Best Picture to a film like LotR are the same ones that are pushing SSSCA (or CBDTPA) through Congress right now. Boycott Hollywood!
The DOJ wants to target five types of people: drug offenders, legalization advocates, anarchists, and free-speech pushers. I only count four types of people there. Who are the mysterious fifth target group? Presumably terrorists, but that's hard to infer from the tone of this article. It seems that the DOJ thinks that ravers and First Amendment advocates are a bigger threat to America than Al-Qaida. Something's fucked up here.
This review mentions that Queen Amidala and Jar-Jar represent their respective planets in the senate. Which is interesting, considering that they are both from the same planet, at least in the mediocre Episode I that I saw. I'm looking forward to the scene in Episode II when Naboo is split in two by some cataclysmic force. Or maybe this so-called reviewer is just an idiot. You be the judge.