Neural networks are certainly not new, or groundbreaking. We already know their strengths and weaknesses, and they aren't a universal solution to every AI problem. First of all, while they have been inspired by the brain, they don't "mimic" it. Neural networks are based on some neurons having negative weights, reversing the polarity of the signal, which doesn't happen in the brain. They are also linear, which bears similarities to some simple parts of the brain, but are very far from modeling its complex nonlinear processing. Neural networks are useful AI tools, but aren't brain models. Second neural networks are only good at things when they have to immediately react to an input. Originally, neural networks didn't have memory, and while it's possible to add it, it doesn't fit right into the system and is hard to work with. While neural networks make good reflex machines, even simple stateful tasks like a linear or cyclic multi-step motion are nontrivial to implement in them. Which is why they are most effective in combination with other methods, instead of declared a universal solution.
Yes, there are albums that are composed in a way that adds to the value of the individual songs (although concept albums are a cheap talentless way of doing that, a good musician can create an atmosphere from the songs without filler bullshit inbetween). But even in those rare cases, the tradition to make a composition of songs album-length is an unnecessary constraint. For better or worse, music consumers are shifting towards individual songs, and no amount of concept albums is going to stop them. Back to the original point, if there's a well-composed album, that means it has more than two good songs, so torrenters may actually buy it.
Errors per lines of code may give you a hard number, but that number has nothing to do with the quality of code. It only takes one well-placed error to ruin a piece of software.
The problem does exist. Traffic jams are real, and flying in a straight line could save a lot of fuel. There is plenty of demand for flying cars, the problems have always been legislatory.
To be honest I doubt we can engineer a carbon sink that could rival the capacity of the oceans. Most of the plankton from the blooms, for example, gets devoured by marine wildlife fairly quickly, so it only removes carbon temporarily. To be able to sequestrate a reasonable amount of CO2 we would have to use up energy in the order of magnitude of the energy gained from fossil fuels. Stuff like reforestration could help, but that's also a slow process.
There was a small dip in '09 in CO2 emissions but we have quickly recovered from it. Mind you, this is CO2 concentration, not production, so it's the log of the integral of production (that's why it's linear).
You can't make predictions when a human factor is involved. If we were all controlled by a hivemind and made CO2 decrease our highest priority, stopping CO2 production could be done in 20 years (there's enough nuclear power to sustain our needs, and we have electric vehicles - technologically it's already possible). After that, the oceans would absorb the excess CO2 and bring it below 300ppm in about 300 years (according to a study I sadly can't find now). So absent some miracle like fusion reactors even in a best case scenario it would take at least 150 years to get below 350.
But the most important thing is the human factor which depends on our decisions, and can make that time much much longer.
It's a plastic toy that's shaped like a gun, but I don't believe it can be fired. The trigger looks already broken on the picture, imagine how reliable the other parts of the gun are.
Slashdot Mirror
No, they are counting on revenue from the application fees these 80k idiots have just paid. It's not like they really plan to send people to space.
Saddam Hussein definitely had chemical weapons, he just used them all up on his own people by that time.
Or decoys misdirecting attention from the real ones.
Neural networks are certainly not new, or groundbreaking. We already know their strengths and weaknesses, and they aren't a universal solution to every AI problem.
First of all, while they have been inspired by the brain, they don't "mimic" it. Neural networks are based on some neurons having negative weights, reversing the polarity of the signal, which doesn't happen in the brain. They are also linear, which bears similarities to some simple parts of the brain, but are very far from modeling its complex nonlinear processing. Neural networks are useful AI tools, but aren't brain models.
Second neural networks are only good at things when they have to immediately react to an input. Originally, neural networks didn't have memory, and while it's possible to add it, it doesn't fit right into the system and is hard to work with. While neural networks make good reflex machines, even simple stateful tasks like a linear or cyclic multi-step motion are nontrivial to implement in them. Which is why they are most effective in combination with other methods, instead of declared a universal solution.
Yes, there are albums that are composed in a way that adds to the value of the individual songs (although concept albums are a cheap talentless way of doing that, a good musician can create an atmosphere from the songs without filler bullshit inbetween). But even in those rare cases, the tradition to make a composition of songs album-length is an unnecessary constraint. For better or worse, music consumers are shifting towards individual songs, and no amount of concept albums is going to stop them.
Back to the original point, if there's a well-composed album, that means it has more than two good songs, so torrenters may actually buy it.
Underestimating the artists themselves as most don't have more than 2-3 good songs an album anyways.
To be fair, with nobody buying physical CDs, organizing songs into albums is only a tradition now.
It don't know if it's the best, but it definitely tells more about coding abilities than questions like 'what is your biggest weakness?'.
Job interviews also take time.
No it isn't. In this case, you have to ask the subjective but professional opinion of developers.
Errors per lines of code may give you a hard number, but that number has nothing to do with the quality of code. It only takes one well-placed error to ruin a piece of software.
But at least they only have access to what you allow them.
The problem is that flying cars can only take off and land on airfields, which makes them impractical for transportation.
The problem does exist. Traffic jams are real, and flying in a straight line could save a lot of fuel. There is plenty of demand for flying cars, the problems have always been legislatory.
Doesn't that increase the risk of an embolism?
What these critics all miss is that Microsoft is now betting on the tablet market, and doesn't give a damn what its PC users think.
Of the four test shots it worked twice, misfired once and exploded once. Not exactly something I would want to rely upon.
To be honest I doubt we can engineer a carbon sink that could rival the capacity of the oceans. Most of the plankton from the blooms, for example, gets devoured by marine wildlife fairly quickly, so it only removes carbon temporarily. To be able to sequestrate a reasonable amount of CO2 we would have to use up energy in the order of magnitude of the energy gained from fossil fuels. Stuff like reforestration could help, but that's also a slow process.
We already have glasses and hearing aides that don't make you look like a complete dork.
At least not plaintext email. I guess Google staff is knowledgable enough to encrypt their sensitive messages.
There was a small dip in '09 in CO2 emissions but we have quickly recovered from it. Mind you, this is CO2 concentration, not production, so it's the log of the integral of production (that's why it's linear).
You can't make predictions when a human factor is involved. If we were all controlled by a hivemind and made CO2 decrease our highest priority, stopping CO2 production could be done in 20 years (there's enough nuclear power to sustain our needs, and we have electric vehicles - technologically it's already possible). After that, the oceans would absorb the excess CO2 and bring it below 300ppm in about 300 years (according to a study I sadly can't find now). So absent some miracle like fusion reactors even in a best case scenario it would take at least 150 years to get below 350.
But the most important thing is the human factor which depends on our decisions, and can make that time much much longer.
That's an overgeneralization. Iran's nuclear program doesn't become good just because it annoys Israel.
It's a plastic toy that's shaped like a gun, but I don't believe it can be fired. The trigger looks already broken on the picture, imagine how reliable the other parts of the gun are.
This is the third time this question comes up.