answering some questions that requires thinking involves cognitive reasoning. If answering questions doesn't involve cognitive reasons, we will not be answering questions in schools, we will be writing essays for every class.
Their model is very flawed and so is the neural net model that a lot of AI books like to preach. They are not the key to true AI, sure enough CycCorp and neural net can be used to design systems that are "intelligent" in specialized area. But to build a system that can achieve general intelligence in all areas? Nope! I am not a skeptic, I know it can be done, not just how they are doing it.
I hate IM for this reason, I usually tell people I don't have any IM account till I make sure they are not part of the "i am often bored" group. Else, I tend to give them only email. It's amazing how people who can't find the strength to write a 2 line email can write 5000 lines over IM.
Whever I look at those automatons, I can't help but think robotics and programming. The drums encoded the music and all. I wondered how much this inspired automata theory in computer science... I wonder if learning about automatons and watch making will give one an edge in robotics making today.
I don't know about that. I can count how many PGP/GPG encrypted emails I have received. Don't underestimate the power of laziness. So long as it is done that it takes 1 minute to set up. generate key, upload key and not automated. Forget it. Only a few would.
yeah, lots of people said that about japanese companies. but what does it matter, it is said that imitaton is the most sincere form of flattery, and it is said that the best ideas are often stolen.
Now, your rebate should be $21.25 What you forget is that say 100,000 customers apply for said rebate, the company temporarily has $2,125,000 for say 30 days. Now the company invests it and earns 4% in that 30 days period. Or $85,000.
when I was running linux on a 486 100mhz in 96. I had doom, and gameboy emulator. my sister was 10 at that time. i taught her quickly how to login to the system, and some basic unix commands, and she would get on to play with gameboy emulator. Really, there are lots of quality games with emulators...
commanding intelligence for the most part is almost independent of college. By high school, those that have commanding intellignece have pretty much established the fact that they do. college is about exposure. It is not just about exposure to professors and their experiences but real knowledge. A high school dropout in the real world is usually limited to silly books at borders. The sheer amount of books and journals that I obtain from my science and engineering library is definitely something to envy for a curious mind without access to them.
in short, when i meet young people who are so cocky. i throw a few challenges their way.
show them some algorithms and ask them to calculate their complexity.
show them a simple function/loop and ask them to prove the correctness.
ask them to write a parser, small compiler or interpreter.
ask them to write a simple device driver.
ask them to port their code to run parallel on machines.
ask them to have their code embed another language as a scripting engine.
ask them to implement a server/client according to some RFC specs.
I dropped out of college to make money, and the best decision I have made is to go back! On pure coding, I can outcode alot of my professors, so I use to think I was smarter, but when it boils down to experience and intution, they smack me down. Case in point, I had a very complex problem bothering me for a commerical project I am currently undertaking. So I stop one of my professors who's focus is in Application Research. I spend 6 minutes describing said problem, when I was done. He continues, "Oh, that's easy, I will do foo and bar, I am teaching a class next semester on it, perhaps you should come take it!" I was enlightened! I ran to the library, grabbed some books on said topics and immediately realized a way to take a crack at my problem. Without an experienced mind like that, I could have wasted a lot of time writing code that does nothing or at best reinventing the wheel.
when i discovered linux in 94 at 10th grade, it had nothing to do with games. It was sheer curiosity. A lot of windows game geeks do nothing but play games, period. they are not interesting in exploring and learning, all they are interested is simply playing games.
a young person discovering linux today, most likely will have to do so by theirself of through friends because their parents are definitely not going to be the source of introduction...
some of the "best" programmers and folks you know. maybe they are "best" to you, cuz they are better than you.
i dropped out of college to work full time, in my circles of friends and coworkers, i was the computer guru in pretty much anything, unix, coding, hardware, whatever. but i decided to go back to school, and when i started taking high level classes, i began to realize how clueless i was. sure enough, you can find a lot of non college educated programmers who understand generic data structures and algorithms and can pull off a lot of web app development. But when it comes to development of new efficient algorithms and technologies, they are lacking.
It is not the case that IBM is lacking in funds to do it. It is the case that students who tend to think out of the box might hold the key to better solutions than industry vets.
whoever that comes after you will not sleep better at night. if a script is several thousand lines of code. then a local unix shell is not the answer. it's time for you to migrate to python or perl.
I don't think CPU should be designed for AI-style algorithms, when the said algorithms have not been proven. Assume we finally suceed in implementing the Holy Grail of AI right, then we can seek out ways to optimize and make it fast, thus custom CPUs will come in. Right now, most of the algorithms are a joke.
He is wrong. Intelligence is not about speed. I have met people who are very very smart, but they think very slowly. You ask questions, and the I too knows (ITKs) will blurt out an answer so damn fast, but mr smarty pant will think and think, and you would think they are clueless, but when they final answer, you can't tear apart their answer.
We can build a machine that has human intelligence and run it on a 2ghz process. The only issue is that instead of answering a question in a second. Perhaps it will take 1 or 2 hours to deliver an intelligence reply. But it should be able to pass a turing test with time thrown at the window.
Go read what 3D researchers said about graphics in the 70's. I bet they believed a 10ghz was good enough for real life 3D graphics.
What is hindering us is not speed, but our approach to AI research.
How the hell does an AI malfunction? Is it a mechanical hardware? That's like asking, what if an algorithm malfunctions. Somethings don't just malfunction, it may contain a logic error, the hardware may malfunction, but software doesn't malfunction, it always does what it is told to do, the way it is told to do so.
I very well agree with your thoughts on the limited set of coding languages for most.
I believe every programmer should know at least a low level language C, C++ or Asm.
A dynamic language like Perl, Python or Ruby.
Our modern cobols, Java or.NET
A functional language, Haskell, ML or even Lisp...
I have never met a programmer who knows at least 1 of each from those categories who is a bad programmer. But I rarely trust programmers who can't code in more than one or two languages.
Slashdot Mirror
answering some questions that requires thinking involves cognitive reasoning. If answering questions doesn't involve cognitive reasons, we will not be answering questions in schools, we will be writing essays for every class.
Their model is very flawed and so is the neural net model that a lot of AI books like to preach. They are not the key to true AI, sure enough CycCorp and neural net can be used to design systems that are "intelligent" in specialized area. But to build a system that can achieve general intelligence in all areas? Nope! I am not a skeptic, I know it can be done, not just how they are doing it.
More seriously, an answering machine would be cool.
I hate IM for this reason, I usually tell people I don't have any IM account till I make sure they are not part of the "i am often bored" group. Else, I tend to give them only email. It's amazing how people who can't find the strength to write a 2 line email can write 5000 lines over IM.
Whever I look at those automatons, I can't help but think robotics and programming. The drums encoded the music and all. I wondered how much this inspired automata theory in computer science... I wonder if learning about automatons and watch making will give one an edge in robotics making today.
rubbish. do you think third world countries care? the black market doesn't exist in US for such products but in africa, asia and some parts of europe.
I don't know about that. I can count how many PGP/GPG encrypted emails I have received. Don't underestimate the power of laziness. So long as it is done that it takes 1 minute to set up. generate key, upload key and not automated. Forget it. Only a few would.
yeah, lots of people said that about japanese companies. but what does it matter, it is said that imitaton is the most sincere form of flattery, and it is said that the best ideas are often stolen.
Now, your rebate should be $21.25 What you forget is that say 100,000 customers apply for said rebate, the company temporarily has $2,125,000 for say 30 days. Now the company invests it and earns 4% in that 30 days period. Or $85,000.
Nope! It as much a computer program as clocks are!
when I was running linux on a 486 100mhz in 96. I had doom, and gameboy emulator. my sister was 10 at that time. i taught her quickly how to login to the system, and some basic unix commands, and she would get on to play with gameboy emulator. Really, there are lots of quality games with emulators...
commanding intelligence for the most part is almost independent of college. By high school, those that have commanding intellignece have pretty much established the fact that they do. college is about exposure. It is not just about exposure to professors and their experiences but real knowledge. A high school dropout in the real world is usually limited to silly books at borders. The sheer amount of books and journals that I obtain from my science and engineering library is definitely something to envy for a curious mind without access to them.
ditto!
in short, when i meet young people who are so cocky. i throw a few challenges their way.
show them some algorithms and ask them to calculate their complexity.
show them a simple function/loop and ask them to prove the correctness.
ask them to write a parser, small compiler or interpreter.
ask them to write a simple device driver.
ask them to port their code to run parallel on machines.
ask them to have their code embed another language as a scripting engine.
ask them to implement a server/client according to some RFC specs.
I dropped out of college to make money, and the best decision I have made is to go back! On pure coding, I can outcode alot of my professors, so I use to think I was smarter, but when it boils down to experience and intution, they smack me down. Case in point, I had a very complex problem bothering me for a commerical project I am currently undertaking. So I stop one of my professors who's focus is in Application Research. I spend 6 minutes describing said problem, when I was done. He continues, "Oh, that's easy, I will do foo and bar, I am teaching a class next semester on it, perhaps you should come take it!" I was enlightened! I ran to the library, grabbed some books on said topics and immediately realized a way to take a crack at my problem. Without an experienced mind like that, I could have wasted a lot of time writing code that does nothing or at best reinventing the wheel.
rubbish.
when i discovered linux in 94 at 10th grade, it had nothing to do with games. It was sheer curiosity. A lot of windows game geeks do nothing but play games, period. they are not interesting in exploring and learning, all they are interested is simply playing games.
a young person discovering linux today, most likely will have to do so by theirself of through friends because their parents are definitely not going to be the source of introduction...
some of the "best" programmers and folks you know. maybe they are "best" to you, cuz they are better than you.
i dropped out of college to work full time, in my circles of friends and coworkers, i was the computer guru in pretty much anything, unix, coding, hardware, whatever. but i decided to go back to school, and when i started taking high level classes, i began to realize how clueless i was. sure enough, you can find a lot of non college educated programmers who understand generic data structures and algorithms and can pull off a lot of web app development. But when it comes to development of new efficient algorithms and technologies, they are lacking.
It is not the case that IBM is lacking in funds to do it. It is the case that students who tend to think out of the box might hold the key to better solutions than industry vets.
whoever that comes after you will not sleep better at night. if a script is several thousand lines of code. then a local unix shell is not the answer. it's time for you to migrate to python or perl.
Everything works in theory, but not pratice.
I don't think CPU should be designed for AI-style algorithms, when the said algorithms have not been proven. Assume we finally suceed in implementing the Holy Grail of AI right, then we can seek out ways to optimize and make it fast, thus custom CPUs will come in. Right now, most of the algorithms are a joke.
He is wrong. Intelligence is not about speed. I have met people who are very very smart, but they think very slowly. You ask questions, and the I too knows (ITKs) will blurt out an answer so damn fast, but mr smarty pant will think and think, and you would think they are clueless, but when they final answer, you can't tear apart their answer.
We can build a machine that has human intelligence and run it on a 2ghz process. The only issue is that instead of answering a question in a second. Perhaps it will take 1 or 2 hours to deliver an intelligence reply. But it should be able to pass a turing test with time thrown at the window.
Go read what 3D researchers said about graphics in the 70's. I bet they believed a 10ghz was good enough for real life 3D graphics.
What is hindering us is not speed, but our approach to AI research.
basically, it's possible to have any at 1, and the rest huge.
ie friends and fun = 1, food = 1000
is equal to friends = 10, fun = 10, food = 10
i think not!
My machines, I will not describe as ancient. But, the are 400mhz and 450mhz, and GAIM runs awesome on them. One is OpenBSD, the other is Gentoo Linux.
*shrugs*
How the hell does an AI malfunction? Is it a mechanical hardware? That's like asking, what if an algorithm malfunctions. Somethings don't just malfunction, it may contain a logic error, the hardware may malfunction, but software doesn't malfunction, it always does what it is told to do, the way it is told to do so.
if the laptop has no cdrom or network card. one can still install linux via serial cable.
I very well agree with your thoughts on the limited set of coding languages for most.
.NET
I believe every programmer should know at least a low level language C, C++ or Asm.
A dynamic language like Perl, Python or Ruby.
Our modern cobols, Java or
A functional language, Haskell, ML or even Lisp...
I have never met a programmer who knows at least 1 of each from those categories who is a bad programmer. But I rarely trust programmers who can't code in more than one or two languages.