Occam's razor is also highly suspect. There's the issue of cultural bias when counting assumptions.
The way Hutter uses it is in the sense of Kolmogorov complexity.
Let's say you have a series of inputs 1 2 3 4 5 6 7 8 and want to predict how it will continue from there.
Kolmogorv complexity says that the most likely explanation for this series is given by the shortest universal Turing machine program that will produce that series.
Sure. If there's a game of A vs B, in which it's important to know your opponent's intentions (like rock paper scissors), then when player A can predict how play B will play while B can't do the same, A will always have an advantage. However, that says nothing about how good the AI of B is.
In a game like chess, an optimal move is an optimal move, even if the opponent knows that it's going to be played.
In general, Hutter aims for optimal performance given all the available information (and not just for games - for any situation). If your opponent has better information (like your source code), he can still be better in a game like poker, but that cannot be helped.
I bet that Hutter's AI will crack a human's "randomness" rather quickly, by the way. We're not good at being random. Meanwhile, you still don't know the machine's cards.
I'd say a career change is certainly possible and not such a big deal, but I don't think you should go to college for a second time.
The most important things you get from college (some maturity, the ability to digest hard books on your own, to finish a large project) are things you hopefully already have after the first time. College is good, but it's not the most efficient way to get specific knowledge on a subject; if your first degree was decent, it'd be a waste of time.
So learn to program, get into the habit of reading hard books on software regularly, and go find a job in the field. Added plus is that you'll find out a bit sooner whether you actually like the new direction.
It is just greed. I mean, does a 10 byte text message actually cost 10 cents? If that were the case, then it should be costing me $30.00 per minute to talk to people. It's just plain old greed.
That's a bit naive. A large telco has a lot of costs that aren't directly related to sending a single SMS message, that still need to be covered by things like SMS messages. Plus it's perfectly normal for them to want to make a profit on top of it all.
They are to blame. It's a DNS server, it should function as a DNS server. If you do a request to it for a non-existing domain, it should report that, because the protocol says so.
This sort of thing assumes that 1) the WWW is the only thing that exists on the Internet 2) that people asking for a non-existing.cm domain must have made a typo, and 3) that there is no such thing as a DNS standard that a DNS server has to adhere to. They're all wrong.
This breaks all sorts of technology, for instance spam filters that try to check if an originating domain actually exists. According to Cameroon, every possible.cm domain, no matter what the context, exists. That's wrong.
It reminds me of the Belgian "Garden Gnome Liberation Front" or something like that a few years ago, that would steal gnomes from gardens and set them free in the wild. At least that was what police assumed when they found about 50 stolen gnomes together in a forest. (This is a vague recollection of some small news item some years back, no accuracy at all is claimed)
But there is definitely an assumption among a lot of Linux users that if it's available on Linux the course code has got to be made available or else it doesn't belong on Linux, like it's some kind of plague.
And it's correct! The whole point of the FSF and the GNU operating system (usually running on a Linux kernel) is to make an entirely Free operating system, with Free applications. The whole point of this entire exercise is to make closed source software superfluous, not just Windows. That's the main difference with BSD.
NVidia's efforts are counterproductive, since they don't actually contribute to the body of Free software, but they do make it less likely that someone does the work independently.
But there is a company behind Ubuntu - Canonical. They offer professional support for those who want it. Of course, Red Hat is much larger, more entrenched and more experienced, but I think that outside of the US the situation isn't as clear cut.
I'm sorry, I thought your post was good, but I was just trying to score points with the opposite, a quick funny... seem to have failed completely. Ah well:-)
Well, it's simply that I believe Linus has pretty much proven that his project leadership works for Linux. You don't have to agree with everything he does, of course anybody can criticize him on details or specific issues... but to say he isn't suited for the job, you're going to have to have some serious authority on the subject before that can be taken seriously.
If they deliver the code and can keep the project together with their style, then can act how they want. Linus, RMS, Theo de Raadt - they have their quirks (Linus much less than the others), but the code speaks for itself.
Since you're going as far as to question Linus suitability for his job (architecting his hobby project that conquered the world) - where's your wildly succesful open source project?
Actually, the result of genetical engineering (or at least, good old selective breeding). Wikipedia says:
At one time, most canned and fresh pineapples were produced on Smooth Cayenne plants. Since about 2000, the most common fresh pineapple fruit found in U.S. and European supermarkets is a low-acid hybrid that was developed in Hawaii in the early 1970s.
Look for brand names like 'Del Monte Extra Sweet Pineapple Gold' and you know you have the sweet version.
For one thing, look at the times. That is news created live in an extremely chaotic situation. That's how it goes, mistakes happen.
For another, he was bleeding profusely after being hit by a bit of merchandise held by a stupid spectator while cycling 60+ km/h, he was in a terrible state (needed lots of stitches), but he did not actually crash, he managed to stay on his bike. It's easy to see how that little confusion happened.
From what I see of the case, it doesn't look like they're getting as far as having a burden of proof - they're having difficulty pointing out what exactly they're actually accusing IBM of. After years.
As tradition requires it, allow me to point out that Emacs is at least equally great at running in a bunch of terminal windows for quick editing, plus it comes with several carefully selected types of kitchen sink:-)
Slashdot Mirror
Occam's razor is also highly suspect. There's the issue of cultural bias when counting assumptions.
The way Hutter uses it is in the sense of Kolmogorov complexity.
Let's say you have a series of inputs 1 2 3 4 5 6 7 8 and want to predict how it will continue from there.
Kolmogorv complexity says that the most likely explanation for this series is given by the shortest universal Turing machine program that will produce that series.
Sure. If there's a game of A vs B, in which it's important to know your opponent's intentions (like rock paper scissors), then when player A can predict how play B will play while B can't do the same, A will always have an advantage. However, that says nothing about how good the AI of B is.
In a game like chess, an optimal move is an optimal move, even if the opponent knows that it's going to be played.
In general, Hutter aims for optimal performance given all the available information (and not just for games - for any situation). If your opponent has better information (like your source code), he can still be better in a game like poker, but that cannot be helped.
I bet that Hutter's AI will crack a human's "randomness" rather quickly, by the way. We're not good at being random. Meanwhile, you still don't know the machine's cards.
I'd say a career change is certainly possible and not such a big deal, but I don't think you should go to college for a second time.
The most important things you get from college (some maturity, the ability to digest hard books on your own, to finish a large project) are things you hopefully already have after the first time. College is good, but it's not the most efficient way to get specific knowledge on a subject; if your first degree was decent, it'd be a waste of time.
So learn to program, get into the habit of reading hard books on software regularly, and go find a job in the field. Added plus is that you'll find out a bit sooner whether you actually like the new direction.
Apparently they call it "Magyarország" themselves. Making the difference between Hungaria and Hungary pretty much moot.
What gets me going though is what is outside of those 180 billion light years of width?
There is no outside.
It is just greed. I mean, does a 10 byte text message actually cost 10 cents? If that were the case, then it should be costing me $30.00 per minute to talk to people. It's just plain old greed.
That's a bit naive. A large telco has a lot of costs that aren't directly related to sending a single SMS message, that still need to be covered by things like SMS messages. Plus it's perfectly normal for them to want to make a profit on top of it all.
They are to blame. It's a DNS server, it should function as a DNS server. If you do a request to it for a non-existing domain, it should report that, because the protocol says so.
This sort of thing assumes that 1) the WWW is the only thing that exists on the Internet 2) that people asking for a non-existing .cm domain must have made a typo, and 3) that there is no such thing as a DNS standard that a DNS server has to adhere to. They're all wrong.
This breaks all sorts of technology, for instance spam filters that try to check if an originating domain actually exists. According to Cameroon, every possible .cm domain, no matter what the context, exists. That's wrong.
Show it. Show us a completely usable system, running a Linux kernel and no GNU stuff. I don't think it can be done.
On the other hand, Debian with a BSD kernel and Debian with a Hurd kernel are both available.
It reminds me of the Belgian "Garden Gnome Liberation Front" or something like that a few years ago, that would steal gnomes from gardens and set them free in the wild. At least that was what police assumed when they found about 50 stolen gnomes together in a forest. (This is a vague recollection of some small news item some years back, no accuracy at all is claimed)
Ahh - the tried and tested way of getting support for Linux: claim something is impossible / doesn't exist :-).
My personal favorite simple graphical FTP client is Krusader. Your distro probably has it.
But there is definitely an assumption among a lot of Linux users that if it's available on Linux the course code has got to be made available or else it doesn't belong on Linux, like it's some kind of plague.
And it's correct! The whole point of the FSF and the GNU operating system (usually running on a Linux kernel) is to make an entirely Free operating system, with Free applications. The whole point of this entire exercise is to make closed source software superfluous, not just Windows. That's the main difference with BSD.
NVidia's efforts are counterproductive, since they don't actually contribute to the body of Free software, but they do make it less likely that someone does the work independently.
Seems I'm not the only one to recently re-read Snow Crash :-)
But there is a company behind Ubuntu - Canonical. They offer professional support for those who want it. Of course, Red Hat is much larger, more entrenched and more experienced, but I think that outside of the US the situation isn't as clear cut.
I'm sorry, I thought your post was good, but I was just trying to score points with the opposite, a quick funny... seem to have failed completely. Ah well :-)
PPS: if this post doesn't deserve a modding up - I don't know what the hell does.
In the current Slashdot climate, what you need is FACTS backed up by well researched and thought out opinion.
Well, it's simply that I believe Linus has pretty much proven that his project leadership works for Linux. You don't have to agree with everything he does, of course anybody can criticize him on details or specific issues... but to say he isn't suited for the job, you're going to have to have some serious authority on the subject before that can be taken seriously.
If they deliver the code and can keep the project together with their style, then can act how they want. Linus, RMS, Theo de Raadt - they have their quirks (Linus much less than the others), but the code speaks for itself.
Since you're going as far as to question Linus suitability for his job (architecting his hobby project that conquered the world) - where's your wildly succesful open source project?
Actually, the result of genetical engineering (or at least, good old selective breeding). Wikipedia says:
At one time, most canned and fresh pineapples were produced on Smooth Cayenne plants. Since about 2000, the most common fresh pineapple fruit found in U.S. and European supermarkets is a low-acid hybrid that was developed in Hawaii in the early 1970s.
Look for brand names like 'Del Monte Extra Sweet Pineapple Gold' and you know you have the sweet version.
It's the name of the town. He didn't say the companies were named that (though I had to check Google first to realize that...)
For one thing, look at the times. That is news created live in an extremely chaotic situation. That's how it goes, mistakes happen.
For another, he was bleeding profusely after being hit by a bit of merchandise held by a stupid spectator while cycling 60+ km/h, he was in a terrible state (needed lots of stitches), but he did not actually crash, he managed to stay on his bike. It's easy to see how that little confusion happened.
From what I see of the case, it doesn't look like they're getting as far as having a burden of proof - they're having difficulty pointing out what exactly they're actually accusing IBM of. After years.
But if the net profit is 0.6%, there's no way the dividend is going to be more than 0.6%.
As tradition requires it, allow me to point out that Emacs is at least equally great at running in a bunch of terminal windows for quick editing, plus it comes with several carefully selected types of kitchen sink :-)
But how's life in Beirut? Business as usual too?
This is as absurd as saying that Microsoft never pays any profit taxes, because the government is too timid to come collect it. Please...