The Council of Elders has declared with enthuisiasm our intention to obliterate the creatures from the blue planet in person.
"For to long have these pathetic monsters hidden in the safety of their hellish atmosphere, while their mechanical agents attacked our world," announced K'breel, speaker for the Council. "We shall have revenge for the unprovoked attacks of the past twenty-two years. Most of all we shall have revenge for the Life Day transmission."
When a junior intelligence officer declined to comment, K'breel had him nailed to a yeast-tree by his gelsacs for being a smartass.
(I'm no good, but I do it for the sake of tradition!)
Great Internet Mersenne Prime Search. No link; it's been hardly a month since I last slashdotted a Free software site. You can google it if you're serious.
True, it doesn't do anything spectacularly useful. But it's not useless and there are few things better for testing the stability of your CPU.
No, you weren't trolling. And if I'd seen this ten seconds earlier I'd have had points for you. But tone it down a little, okay? It's better to rant after the trolling starts.
(I sincerely believe that Eric Raymond is mistaken with this whole "Linus's law" thing. It applies to some types of bugs, but not to all.)
Please, please can we not mention religion on Slashdot?
It's always the same. Religious people flaming atheists, atheists flaming religious people and agnostics flaming both sides. The universal argument? "I'm right because it's obvious and you're stupid for not agreeing".
Let's say it's simpler than most earlier estimates have said, say an average of 7 inputs and 1 output per neuron, each firing up to 100 times per second with about 3 bits of phase information encoded in each spike. So that's 300bps of bandwidth out and 2100bps in, so 300-2400bps/neuron depending on how it's implemented. Times 20-100 billion neurons that's still some serious bandwidth, (6 -240Tbps) even if it is nearly all local.
I won't insult you by checking your calculations but your number is too large for your assumptions. Remember that when simulating a brain each processor won't compute for a single cell, but a local group of cells . The vast majority of the bandwidth use would therefore be between each CPU and its memory rather than between different CPUs. Each CPU would than have connections with those CPUs whose local groups of cells its own communicates with. The less local you get, the lower the total bandwidth requirements.
Still, a lot of non-local communication takes place as shown by the "connectome" data and the emergent synchrony of firing across the brain
Yes, as I said, there is non-local communication but it's mostly local, and furthermore the non-local connections are fixed (for the purpose of short-term simulations). In a simulation you need relatively few connections between distant CPUs and those connections don't carry as much data as connections between nearby CPUs.
Just about anything is "a lot" at the scale at which the brain operates. It just isn't as complex as it could conceivably have been.
I am not a neuroscientist. As a grad student I do study artificial neural networks, which means that I must also have a little knowledge of neuroscience.
The brain is not a fully connected network. It is divided into many sub-networks. I think it's estimated at about 500k, but don't quote me on that number. These sub-networks are often layered, so if you have a three-layer feed-forward sub-network of 5 cells in each layer, each of these cells has only 5 inputs except for the 5 nodes in the input layer, which connects to other sub-networks. (If there are connections from later layers back to earlier layers, the network is said to be a 'feedback' rather than feed-forward network.) These sorts of networks can be simulated very efficiently on parallel hardware, as a cell mostly gets information from the cells that are close to it.
In short, your suspicion is entirely correct. Moreover, you not only don't need fast connections between many of your processing nodes, most of them don't need to be connected to each other at all.
This is the reason why neural networks are interesting in the first place: that they can be simulated on parallel hardware when we don't know a good parallel algorithm with conventional computing techniques. (If it interests you: another name for neural networks is 'parallel distributed computing'.)
There is a hard limit on the 'order' (think of it as function complexity) of functions that can be computed with a given network. To compute a function beyond that limit, you need to have a larger number of inputs to some cells, thereby increasing the order of the network but making it less parallel. Most everyday things are in fact of surprisingly low order. Fukushima's neucognitron can perform tasks like handwriting recognition with only highly local information.
The second instance suggest that a plaintiff can cheat a man through a lawsuit, i.e. ruin a man with a lawsuit that has no merit.
???
A man can be ruined by a lawsuit in the sense that he can be cheated out of everything and unable to recover it in court.
This means that 1) Someone cheats you out of everything you own 2) You, the plaintiff, can't recover said things in a court (and additionally end up with legal fees). I.e., you are ruined by the lawsuit that you filed.
Perhaps I could have written more clearly, but it most certainly does mean what I said it does.
What Columbus was fighting against is that people though the Earth was (correctly) approximately 40,000 km in circumference (The Greeks had measured it fairly accurately) while Columbus thought (incorrectly) it was much smaller and that it was practical to sail west to India. Columbus lucked out in that America was in the way, otherwise he would have been a footnote in history as the leader of an expedition of 3 ships that sailed west never to be seen again.
Both in the sense that a plaintiff can take everything he owns and in the sense that he can be cheated out of everything and unable to recover it in court
I know this guy is a liar. That's not the point. A plaintiff should have rights too. It's common enough for crooks to cheat people, knowing they won't be able to take the matter to court.
(I don't really know your legal system, but I think you're right.)
Morally speaking, I find it abhorrent. A man can be ruined by a lawsuit. Both in the sense that a plaintiff can take everything he owns and in the sense that he can be cheated out of everything and unable to recover it in court. One should have rights in a civil case, as one does in a criminal case.
Whistleblowers are often protected from civil lawsuits, too, by whistleblower-protection laws. (Governments don't hate true whistleblowers! That's just a myth that the neo-anarchist movement has invented to justify their hatred of governments.)
They say they don't. You know, every whistleblower I've ever heard of had the system come after him.
We had a prison warden here in ZA -- a few years ago -- who let some prisoners take a camera and film the widespread corruption. (Buying a loaded gun from a guard, that sort of thing.) He was fired.
A little more recently, we had a teacher report fraudulent matric results (your final marks for high school). Got ignored, went to the media, government acts all outraged and launches investigations and all that stuff. Oh, the teacher just happened to get fired afterwards.
Then there's Manning in the USA. Leaking information about, you know, serious and violent crimes.
Please, feel free to provide some concrete examples of whistleblowers who were protected by the law. By the way -- I'm not an anarchist.
Despite it now being free, I will neither play Team fortress nor use the SDK. I would do so, even at twice the old price, if I could run it natively on Linux.
To Valve, and all the other game developers: if you deprive us of your games, you deprive yourselves of our money.
Slashdot Mirror
The schools are responsible for sending students home.
They me be in your part of the world. They're not in mine. We also didn't have any late bus.
If Google doesn't remove them from its searches, they demand money on the basis of ridiculous copyright claims.
If Google does remove them, they demand money on the basis of Google abusing its monopoly to punish them.
I know it doesn't make sense if you're sane, but that's how these sorts of people reason.
The Council of Elders has declared with enthuisiasm our intention to obliterate the creatures from the blue planet in person.
"For to long have these pathetic monsters hidden in the safety of their hellish atmosphere, while their mechanical agents attacked our world," announced K'breel, speaker for the Council. "We shall have revenge for the unprovoked attacks of the past twenty-two years. Most of all we shall have revenge for the Life Day transmission."
When a junior intelligence officer declined to comment, K'breel had him nailed to a yeast-tree by his gelsacs for being a smartass.
(I'm no good, but I do it for the sake of tradition!)
Great Internet Mersenne Prime Search. No link; it's been hardly a month since I last slashdotted a Free software site. You can google it if you're serious.
True, it doesn't do anything spectacularly useful. But it's not useless and there are few things better for testing the stability of your CPU.
That's a common myth.
People can and do die in their dreams and suffer no ill effect (personal experience).
No, you weren't trolling. And if I'd seen this ten seconds earlier I'd have had points for you. But tone it down a little, okay? It's better to rant after the trolling starts.
(I sincerely believe that Eric Raymond is mistaken with this whole "Linus's law" thing. It applies to some types of bugs, but not to all.)
Then we agree.
Please, please can we not mention religion on Slashdot?
It's always the same. Religious people flaming atheists, atheists flaming religious people and agnostics flaming both sides. The universal argument? "I'm right because it's obvious and you're stupid for not agreeing".
Small neutral particles of about the same size as the electrons (neutrons, neutrinos, etc.)
Let us do a quick Google search on that.
Neutrinos and electrons are regarded as fundamental particles with zero volume -- which may not be correct -- (http://en.wikipedia.org/wiki/Elementary_particle), so they would have the same size. Neutrons have measurable volume (http://en.wikipedia.org/wiki/Neutron), so "about the same" is entirely wrong.
If we suppose you mean mass, then we get a rest mass of about 10^-30 kg for the electron and at most 10^-36 kg for the neutrino (http://physicsworld.com/cws/article/news/32861) and around 10^-27 kg for the neutron (http://en.wikipedia.org/wiki/Neutron), making you off by many orders of magnitude.
I'm sorry. I just couldn't resist your sig.
Let's say it's simpler than most earlier estimates have said, say an average of 7 inputs and 1 output per neuron, each firing up to 100 times per second with about 3 bits of phase information encoded in each spike. So that's 300bps of bandwidth out and 2100bps in, so 300-2400bps/neuron depending on how it's implemented. Times 20-100 billion neurons that's still some serious bandwidth, (6 -240Tbps) even if it is nearly all local.
I won't insult you by checking your calculations but your number is too large for your assumptions. Remember that when simulating a brain each processor won't compute for a single cell, but a local group of cells . The vast majority of the bandwidth use would therefore be between each CPU and its memory rather than between different CPUs. Each CPU would than have connections with those CPUs whose local groups of cells its own communicates with. The less local you get, the lower the total bandwidth requirements.
Still, a lot of non-local communication takes place as shown by the "connectome" data and the emergent synchrony of firing across the brain
Yes, as I said, there is non-local communication but it's mostly local, and furthermore the non-local connections are fixed (for the purpose of short-term simulations). In a simulation you need relatively few connections between distant CPUs and those connections don't carry as much data as connections between nearby CPUs.
Just about anything is "a lot" at the scale at which the brain operates. It just isn't as complex as it could conceivably have been.
I am not a neuroscientist. As a grad student I do study artificial neural networks, which means that I must also have a little knowledge of neuroscience.
The brain is not a fully connected network. It is divided into many sub-networks. I think it's estimated at about 500k, but don't quote me on that number. These sub-networks are often layered, so if you have a three-layer feed-forward sub-network of 5 cells in each layer, each of these cells has only 5 inputs except for the 5 nodes in the input layer, which connects to other sub-networks. (If there are connections from later layers back to earlier layers, the network is said to be a 'feedback' rather than feed-forward network.) These sorts of networks can be simulated very efficiently on parallel hardware, as a cell mostly gets information from the cells that are close to it.
In short, your suspicion is entirely correct. Moreover, you not only don't need fast connections between many of your processing nodes, most of them don't need to be connected to each other at all.
This is the reason why neural networks are interesting in the first place: that they can be simulated on parallel hardware when we don't know a good parallel algorithm with conventional computing techniques. (If it interests you: another name for neural networks is 'parallel distributed computing'.)
There is a hard limit on the 'order' (think of it as function complexity) of functions that can be computed with a given network. To compute a function beyond that limit, you need to have a larger number of inputs to some cells, thereby increasing the order of the network but making it less parallel. Most everyday things are in fact of surprisingly low order. Fukushima's neucognitron can perform tasks like handwriting recognition with only highly local information.
Of course; I do apologise for the poor sentence.
http://en.wikipedia.org/wiki/Crucifixion_of_Jesus#Last_words_of_Jesus
The second instance suggest that a plaintiff can cheat a man through a lawsuit, i.e. ruin a man with a lawsuit that has no merit.
???
A man can be ruined by a lawsuit in the sense that he can be cheated out of everything and unable to recover it in court.
This means that 1) Someone cheats you out of everything you own 2) You, the plaintiff, can't recover said things in a court (and additionally end up with legal fees). I.e., you are ruined by the lawsuit that you filed.
Perhaps I could have written more clearly, but it most certainly does mean what I said it does.
What Columbus was fighting against is that people though the Earth was (correctly) approximately 40,000 km in circumference (The Greeks had measured it fairly accurately) while Columbus thought (incorrectly) it was much smaller and that it was practical to sail west to India. Columbus lucked out in that America was in the way, otherwise he would have been a footnote in history as the leader of an expedition of 3 ships that sailed west never to be seen again.
FTFY
I don't disagree with you, I was responding to the AC who said this:
This is a civil litigation case, he doesn't have the "right" to a lawyer
I do find that abhorrent.
Both in the sense that a plaintiff can take everything he owns and in the sense that he can be cheated out of everything and unable to recover it in court
I know this guy is a liar. That's not the point. A plaintiff should have rights too. It's common enough for crooks to cheat people, knowing they won't be able to take the matter to court.
(I don't really know your legal system, but I think you're right.)
Morally speaking, I find it abhorrent. A man can be ruined by a lawsuit. Both in the sense that a plaintiff can take everything he owns and in the sense that he can be cheated out of everything and unable to recover it in court. One should have rights in a civil case, as one does in a criminal case.
Sigh. I know what my government is like. You didn't supply what I asked for: proof that any government is better.
But you're probably just trolling anyway.
Whistleblowers are often protected from civil lawsuits, too, by whistleblower-protection laws. (Governments don't hate true whistleblowers! That's just a myth that the neo-anarchist movement has invented to justify their hatred of governments.)
They say they don't. You know, every whistleblower I've ever heard of had the system come after him.
We had a prison warden here in ZA -- a few years ago -- who let some prisoners take a camera and film the widespread corruption. (Buying a loaded gun from a guard, that sort of thing.) He was fired.
A little more recently, we had a teacher report fraudulent matric results (your final marks for high school). Got ignored, went to the media, government acts all outraged and launches investigations and all that stuff. Oh, the teacher just happened to get fired afterwards.
Then there's Manning in the USA. Leaking information about, you know, serious and violent crimes.
Please, feel free to provide some concrete examples of whistleblowers who were protected by the law. By the way -- I'm not an anarchist.
I don't think your average CEO cares about that.
As long as the six-month bottom line makes it look like the next guy's screw-up, they'd be perfectly happy to run the company into the ground.
Lucky I'm fluent in AC. ;)
Despite it now being free, I will neither play Team fortress nor use the SDK. I would do so, even at twice the old price, if I could run it natively on Linux.
To Valve, and all the other game developers: if you deprive us of your games, you deprive yourselves of our money.
Nope. Unconstitutional and counterproductive laws survive legal challenge all the time.
Haven't you been reading Slashdot?
Even if you are in the right, going to court is not easy and cheap.
You're also not guaranteed to win.
I suck.