It's a different paradigm for programming. Every paradigm opens up new ways of developing software that can be applied to other programming languages.
Even if you never write a single line of assembly in your life there are times when you might need to understand the concepts - for example you might need to embed an interpreter for a language in your software and use a virtual machine. Before you do that it might be a good idea to know how people have designed real machines.
You know, when you're trying to get a game out the door that runs at 30fps and your competitor has a similar game running at 60fps because they coded their inner loops in assembly you begin to realize why optimization is important after all.
Omega is like a little tiny postscript to Gödel's theorem, but Chaitin pushes it like it's something important. When someone does a great piece of mathematical work then usually a whole flurry of results follow that are completely obvious to anyone who happens to be there at the right time to hear the big result. That's what Chaitin's work in logic is. (OTOH His stuff on register allocation via graph coloring is quite neat.)
No, I still don't get it. I'd expect that doing X amount of damage would requite the same number of kilograms of -bot, nano- or not, and thus the same amount of power. I also don't see why nanobots would be efficient - I'd expect them to waste as much energy as anything else. What does size have to do with it? This is even crazier than people arguing that nuclear power gives us unlimited free power. We all know that in practice things are complicated. And in the case of nanobots we have no clue about what the technology might look like.
Why are nanobots supposed to have been a risk to the planet while just plain old bots aren't? What's so special about something small that there's a risk of exponential runaway which isn't there for macroscopic factories. They still require raw materials and energy like any other type of production.
There are many times when the compiler says exactly what the error is and says exactly what to do to correct it. But it's probably a good idea that the compiler doesn't actually make those changes. I think it's good to have just one person make the final decision about whether a change should be made.
The difference between hypothesizing about whether he would have proved it, or hypothesizing about whether he would have been able to have proved it, seems pretty minor to me.
But just for you, I'll rewrite my epitaph:
Here lies exp(pi*sqrt(163))
He died ere he could get laid by Charlize Theron
You need to get that one right. What you are probably referring to is something like the halting problem. You can't write a 100% reliable computer program that can take as input another program and determine, in a finite time, whether or not that program is guaranteed to terminate in a finite time. As we have no reason to believe humans can do this either it doesn't show anything.
A human can look at a program, and without compiling in his head, say what is wrong with it
Sometimes. Computers can do this too. Most compilers emit helpful error and warning messages all the time.
The human brain is good at what it does
Does that mean anything? If the human brain wasn't good at what it did what state of affairs would give?
Computers do not come up with insights.
They make some pretty insightful moves in chess, for example. I don't think it's fair to say they don't ever have insights.
Probably not in the medium term future. More likely it will give asymptotic formulae for how fast some algorithms are for cracking codes. At the moment you occasionally see papers saying things like an n-bit code of type X can be cracking in time O(sqrt(log n)*2^n) time if the RH is true otherwise in time O((log n)*2^n). That's a made up example but it's the right sort of thing.
But the Riemann hypothesis is hard. We can be pretty confident Riemann wouldn't have figured it out (whatever that means). He probably didn't even have the mathematical tools he needed available to him. The only way I would allow such a statement would be if he died and left a manuscript with a partial proof that could be extended to a full proof by a good mathematician in a reasonable time. We know that no such document exists.
The origins of the hypothesis date back to 1859, when mathematician Bernhard Riemann came up with a theory about how prime numbers were distributed, but he died in 1866, before he could conclusively prove it.
As he didn't prove the result, either before or after his death, how can it be said that he died before he proved it? Maybe the lives of great mathematicians form arcs in some abstract space that can be extrapolated beyond their death?
I think I might as well write my epitaph now:
Here lies exp(pi*sqrt(163))
He died before he could get laid by Charlize Theron
Name some things that are intelligent but not human. The meaning of 'intelligent' is pretty bound up up with the notion of human. It means being able to do the sorts of things humans do or try to do (though I'd have to tighten that up to exclude achievements like running fast). We simply have no other model for intelligence. Saying "I don't get why Turing thought that mimicking the human brain would be a step toward intelligence" is absurd, and I'd almost go as far as to say it doesn't really have a meaning. It does, however, carry a connotation. It makes me think that maybe you are a misanthrope.
Slashdot Mirror
I've nothing against assembly. I just think 3 stories about a not very significant programming language (HLA) is getting dangerously close to spam.
I can see you have much to learn about assembly.
Even if you never write a single line of assembly in your life there are times when you might need to understand the concepts - for example you might need to embed an interpreter for a language in your software and use a virtual machine. Before you do that it might be a good idea to know how people have designed real machines.
Game developers - especially console game developers.
And this is at least the third time this has been promoted on /.
You know, when you're trying to get a game out the door that runs at 30fps and your competitor has a similar game running at 60fps because they coded their inner loops in assembly you begin to realize why optimization is important after all.
Omega is like a little tiny postscript to Gödel's theorem, but Chaitin pushes it like it's something important. When someone does a great piece of mathematical work then usually a whole flurry of results follow that are completely obvious to anyone who happens to be there at the right time to hear the big result. That's what Chaitin's work in logic is. (OTOH His stuff on register allocation via graph coloring is quite neat.)
...Chaitin tell you his own work is the most important work in mathematics? If he does it less than 100 times in the book I may consider reading it.
No, I still don't get it. I'd expect that doing X amount of damage would requite the same number of kilograms of -bot, nano- or not, and thus the same amount of power. I also don't see why nanobots would be efficient - I'd expect them to waste as much energy as anything else. What does size have to do with it? This is even crazier than people arguing that nuclear power gives us unlimited free power. We all know that in practice things are complicated. And in the case of nanobots we have no clue about what the technology might look like.
Why are nanobots supposed to have been a risk to the planet while just plain old bots aren't? What's so special about something small that there's a risk of exponential runaway which isn't there for macroscopic factories. They still require raw materials and energy like any other type of production.
...play TuxRacer on my PC. Now it seems I didn't need to pay at all. Can someone tell me who to contact to get my money back?
Or even better, Blood Music by Greg Bear.
There are many times when the compiler says exactly what the error is and says exactly what to do to correct it. But it's probably a good idea that the compiler doesn't actually make those changes. I think it's good to have just one person make the final decision about whether a change should be made.
But just for you, I'll rewrite my epitaph:
Martin Gardner wrote about it in SciAm in 1975? Damn! Here I was thinking that it was at least a slightly obscure number.
The human brain is a crappy processor. If it was any good at processing we'd use brains instead of computers for the jobs we use computers for.
Probably not in the medium term future. More likely it will give asymptotic formulae for how fast some algorithms are for cracking codes. At the moment you occasionally see papers saying things like an n-bit code of type X can be cracking in time O(sqrt(log n)*2^n) time if the RH is true otherwise in time O((log n)*2^n). That's a made up example but it's the right sort of thing.
But the Riemann hypothesis is hard. We can be pretty confident Riemann wouldn't have figured it out (whatever that means). He probably didn't even have the mathematical tools he needed available to him. The only way I would allow such a statement would be if he died and left a manuscript with a partial proof that could be extended to a full proof by a good mathematician in a reasonable time. We know that no such document exists.
I think I might as well write my epitaph now:
Name some things that are intelligent but not human. The meaning of 'intelligent' is pretty bound up up with the notion of human. It means being able to do the sorts of things humans do or try to do (though I'd have to tighten that up to exclude achievements like running fast). We simply have no other model for intelligence. Saying "I don't get why Turing thought that mimicking the human brain would be a step toward intelligence" is absurd, and I'd almost go as far as to say it doesn't really have a meaning. It does, however, carry a connotation. It makes me think that maybe you are a misanthrope.
I know. Those hunchbacks are always cracking my system.
No authority. Argument by authority is usually fallacious and you need to give up the habit.
It's tempting to make a little bet with you over how large the sales of this thing will be.