My reaction was the opposite, that much of this stuff belongs in the realm of obvious truths that American society is too fucked up to admit. Idon't know what Americans think we are, but it sure bears little resemblance to real human beings. But the article describes is how actual humans tend to feel and act. Get over the bullshit and forget your idiotic college classes.
Since forcing people to buy new cartridges is environmentally wasteful, perhaps we should propose legislation that outlaws this sort of behavior on environmental grounds.
"When recycling is outlawed only outlaws will recycle."
Hard to tell from the some of those "papers" since they seem to be written for kindergarteners - or journalists. But with that much parallelism I'm guessing that these computers basically allow "dataflow" style programming, with a certain amount of automatic decomposition, similar to the way PC chips decompose assembly into a simpler language on-chip.
He's right as far as it goes that a multiple choice test where the recipients know almost none of the answers is not very accurate at measuring their marginal knowledge.
However in my experience, hard multiple choice tests have a different problem..
"hard" can mean that you compare against a curve that's known for that particular test and that the curve has a long enough upper tail to seem to measure something at the upper end. Ie, the last couple of questions as you approach 100% are worth more than the questions before them.
The problem with that is that it seems to me that a common way to make a test have that longer upper tail is to make some of the questions ambiguous bad questions. If there are 10 questions on a test that are poorly designed where a knowledgeable person is likely to pick a "wrong" answer, then you can count on it that VERY few people will get all of the "right" answers. Instant "hard" test!
Considering that we have not much better than "not a clue" how the brain actually associates the sound you hear to memory, I am skeptical that this is how their approach works.
But "not a clue" is exactly what executives, patent lawyers and patent judges know about how software and say, mathematics, work, so how is this any different? They wrote a patent on something they don't understand and will approve it without understanding it. They might as well be patenting life - oh wait they do that too,
When machines become capable, for the first time, of being social or moral each basic step toward that will be patentable as well. We will have a patents that covers not-being-evil and one on not-being-an-asshole.
It's a good thing we don't have that sort of problem with children, such that only one family can have children that, say, know the difference between right and wrong and since they patented that no one else is allowed. Or only one family that has children that have a sense of rhythm.
But as computers finally gain the abilities of people, patents become that absurd, they become limits on sentience, ability and personality.
The sort of processing this patent covers is something that hasn't been possible until recently, but I think, in principle, is something absolutely necessary for robust AI, and that is doing recognition simultaneously on both low level features and high level features of data and on intersections of the two.
By "high level" I mean things like word choice, language etc. By low level I imagine they mean things like the specific resonance characteristics of a voice. In voice there are intermediate levels of features too, such a the characteristics of phonemes.
The upshot of this is that just as algorithms and hardware begins to reach a level of power necessary to show intelligence, it will be impossible to do so without stepping on patents.
We will have patents on a machine not being stupid.
I have both windows XP Pro and Ubuntu on my laptop.
Ubuntu boots and shuts down much faster than Windows.
The wireless driver is more robust. I have a fairly poor wireless connection in my apartment, and under Windows the best setup I can manage still loses the connection for a few seconds every few minutes - and that screws youtube and all sorts of other programs.
There's no such problem in Ubuntu. The driver probably doesn't time out as quickly.
I could list other things like that where the Ubuntu drivers are more flexible than the windows ones or where there are more options available like traffic shaping etc...
But it does take more expertise to install some things or fix problems in Ubuntu.
To get a of the blue screen of death all you have to do is hit it with something. That would be a strange sort of reset.
Heh, that's one advantage I saw when playing with virtualization. I could blue-screen windows in my virtual machine and the host operating system (linux) would keep chugging away. It's a pretty good feeling when BSOD no longer requires a reboot.
In the "experimental view" slashdot cuts off the last two entries:
10. Databases (locking, rolling back etc). For those cases where you don't need speed, database technology solves most of the problems of multiple entities keeping the data constant more flexibly and easily than anything other than stopping the world. It could be useful to build some database technology into your product, even if only for inter-process communication.
11. The usual semaphores and mutexes. By the way there's a "many readers but only one writer" mutex pattern that's so useful that I wonder why it isn't more common. In any case you should all know to avoid mutual blocking etc etc...
I've worked on a high performance multiprocessor system.
There are a few principles and tools that make much of it manageable, and then there's some parts that are inherently nearly impossible.
A few principles:
1. Task switches are very very time consuming, at least under preemptive multitasking, and especially when that preemption (as it does in Windows) involves process switches. Since mutexes rely on task switches, they're very expensive, although exact implementation matters too.
2. It's possible to come up with non-blocking algorithms that require no mutexes (though you do need low level access to interlocked instructions, or barriers). Unfortunately that's where it gets nearly impossible because proving that a non-blocking algorithm is correct requires a combinatorial analysis. I could go into the some of the principles of non-blocking algorithms, but they're still very hard even when you know what you're doing. Still it's useful to understand some basic interlocked instructions because people often use mutexes when a simple interlocked instruction would do the same task infinitely quicker.
3. Having a few basic non-blocking queue algorithms (LIFO and FIFO) up your sleeve can suffice to allow rather separate tasks to communicate cheaply. There are even nonblocking queues that allow an unlimited number of processors to compete for each end of the queue simultaneously, and so you can have common banks of work to be balanced between processors, for instance.
4. It's often a win to give very separate tasks to each processor, and have a single process per processor. Put all IO on one processor, for instance and have it communicate with the other processors through non-blocking queues.
5. Even though fair preemptive multitasking is expensive (because of the task switches), in special cases you can use cooperative multitasking (fibers, green threading) and save the time. One advantage to cooperative multitasking is that you can control exactly when a fiber wakes up. In some benchmarks I've seen (producer-consumer), green threading outraces preemptive multitasking by over an order of magnitude.
6. If you can all tie all interrupts to a single processor (which is possible in Windows) and lock specific processes to each processor, then you can (almost) guarantee that certain processors won't interrupted, and thus those processors are the ones that are cheapest to use for any task that the other processors will have to wait for.
7. Aggregate. There are some tasks that can not be accomplished in parallel with the usual work. It's useful to save up a bunch of work like that so that you don't have to shut down the other tasks too often.
8. State machines with atomic arcs. This gets closer to the harder sort of algorithms, but its useful to talk about what's easy. If you need a bunch of processors to cooperate, it's useful to have a state machine description of the global state. If the state can be stored in a single double word, then transitions between states can be accomplished atomically with a CAS instruction (compare and swap), and then it doesn't matter which processor changes the state. They may all try, and one will succeed. If one of the states is too complicated to fit in a double word, then you can either have a spin-lock intermediate state or in some cases, the state double word can include a pointer (but this gets complicated).
9. Cache delays. Multiprocessor systems have to keep the caches coherent. Because of this situations like a CPU reading a value that's dirty in another processor's cache can bring all of the processors to a halt while they transfer data. I'm a few years out of date on what the newest processors do, but it's probably still true, and still true that these pauses are much longer on systems like the Sparc that allow hundreds of processors than it is in x86 machines that are more limited in the number of processors. In any case it's a win to keep data memory use separate between processors when possible.
I take the $10 bit to be exagguration that means "as cheap as possible".
So if the machine costs $35 at its cheapest point, the $10 promise will still have been useful as a lie to get the political machine moving... That's how this works. It's not quite a first world country we're talking about after all, and that's the whole point. We shouldn't expect Indian politics to be that precise and honest.
I just hope that any machine they come up with will have enough flash to hold Linux, a developement environment or two and some extra packages. It will be of much less use to the children if it doesn't.
"Walk a mile in the shoes you want other people to wear and I might believe you."
I might be nice if you tried to walk a mile in the shoes of children whose community can't afford $100 machines (or medicine or books or buildings for classrooms etc. etc.).
The $100 laptop was always a bit strange in that recognizing that current color displays were too short lived and too power hungry for the third world they decided to spend years inventing a new color screen technology instead of going with the long perfected, cheap (and much lower cost and power) reflective black and white LCD displays.
They obsessed over "electric paper" because they thought it "looked like paper". What nerds!!! Who gives a damn? Black and white reflective LCDs lower resolution, but they have very high contrast and are very readable. But not cool enough for trust fund kids and Negroponte's upper crust crowd.
I understand that children may be more enthusiastic to have color toys with sound and everything, but from the point of view of really poor places having three or four times as many machines is better. It's better to have something than nothing.
Slashdot Mirror
My reaction was the opposite, that much of this stuff belongs in the realm of obvious truths that American society is too fucked up to admit. Idon't know what Americans think we are, but it sure bears little resemblance to real human beings. But the article describes is how actual humans tend to feel and act. Get over the bullshit and forget your idiotic college classes.
Lt. Walter Haut had a great sense of humor.
Since forcing people to buy new cartridges is environmentally wasteful, perhaps we should propose legislation that outlaws this sort of behavior on environmental grounds.
"When recycling is outlawed only outlaws will recycle."
Hard to tell from the some of those "papers" since they seem to be written for kindergarteners - or journalists. But with that much parallelism I'm guessing that these computers basically allow "dataflow" style programming, with a certain amount of automatic decomposition, similar to the way PC chips decompose assembly into a simpler language on-chip.
Too bad there's no "moderate down - asshole"
He's right as far as it goes that a multiple choice test where the recipients know almost none of the answers is not very accurate at measuring their marginal knowledge.
However in my experience, hard multiple choice tests have a different problem..
"hard" can mean that you compare against a curve that's known for that particular test and that the curve has a long enough upper tail to seem to measure something at the upper end. Ie, the last couple of questions as you approach 100% are worth more than the questions before them.
The problem with that is that it seems to me that a common way to make a test have that longer upper tail is to make some of the questions ambiguous bad questions. If there are 10 questions on a test that are poorly designed where a knowledgeable person is likely to pick a "wrong" answer, then you can count on it that VERY few people will get all of the "right" answers. Instant "hard" test!
Considering that we have not much better than "not a clue" how the brain actually associates the sound you hear to memory, I am skeptical that this is how their approach works.
But "not a clue" is exactly what executives, patent lawyers and patent judges know about how software and say, mathematics, work, so how is this any different? They wrote a patent on something they don't understand and will approve it without understanding it. They might as well be patenting life - oh wait they do that too,
When machines become capable, for the first time, of being social or moral each basic step toward that will be patentable as well. We will have a patents that covers not-being-evil and one on not-being-an-asshole.
It's a good thing we don't have that sort of problem with children, such that only one family can have children that, say, know the difference between right and wrong and since they patented that no one else is allowed. Or only one family that has children that have a sense of rhythm.
But as computers finally gain the abilities of people, patents become that absurd, they become limits on sentience, ability and personality.
The sort of processing this patent covers is something that hasn't been possible until recently, but I think, in principle, is something absolutely necessary for robust AI, and that is doing recognition simultaneously on both low level features and high level features of data and on intersections of the two.
By "high level" I mean things like word choice, language etc. By low level I imagine they mean things like the specific resonance characteristics of a voice. In voice there are intermediate levels of features too, such a the characteristics of phonemes.
The upshot of this is that just as algorithms and hardware begins to reach a level of power necessary to show intelligence, it will be impossible to do so without stepping on patents.
We will have patents on a machine not being stupid.
You're so right. Why didn't it say "ejaculating into space?"
I'm looking to the satire loophole applying to software - that you can use code in order to make fun of it.
You're making Dan Quayle look good!
Cheer up, the rest of the world will still have freedom on the internet. It's just us Americans who will be regulated out of having any expression.
We'll still be able to listen to Russian stations.
Where's you're "In Soviet Russia..." joke now, bitches?
I want a medic robot covered with bloodstained fur.
Actually it's a good idea. The enemy will be laughing too hard to draw a bead.
It's robo care bear! Where's his rainbow stomach (or red cross or whatever)?
I have both windows XP Pro and Ubuntu on my laptop.
Ubuntu boots and shuts down much faster than Windows.
The wireless driver is more robust. I have a fairly poor wireless connection in my apartment, and under Windows the best setup I can manage still loses the connection for a few seconds every few minutes - and that screws youtube and all sorts of other programs.
There's no such problem in Ubuntu. The driver probably doesn't time out as quickly.
I could list other things like that where the Ubuntu drivers are more flexible than the windows ones or where there are more options available like traffic shaping etc...
But it does take more expertise to install some things or fix problems in Ubuntu.
You're an insane masochist.
If we has a "completely bozo insane" tag, would that add points or subract them?
To get a of the blue screen of death all you have to do is hit it with something. That would be a strange sort of reset.
Heh, that's one advantage I saw when playing with virtualization. I could blue-screen windows in my virtual machine and the host operating system (linux) would keep chugging away. It's a pretty good feeling when BSOD no longer requires a reboot.
In the "experimental view" slashdot cuts off the last two entries:
10. Databases (locking, rolling back etc). For those cases where you don't need speed, database technology solves most of the problems of multiple entities keeping the data constant more flexibly and easily than anything other than stopping the world. It could be useful to build some database technology into your product, even if only for inter-process communication.
11. The usual semaphores and mutexes. By the way there's a "many readers but only one writer" mutex pattern that's so useful that I wonder why it isn't more common. In any case you should all know to avoid mutual blocking etc etc...
I've worked on a high performance multiprocessor system.
There are a few principles and tools that make much of it manageable, and then there's some parts that are inherently nearly impossible.
A few principles:
1. Task switches are very very time consuming, at least under preemptive multitasking, and especially when that preemption (as it does in Windows) involves process switches. Since mutexes rely on task switches, they're very expensive, although exact implementation matters too.
2. It's possible to come up with non-blocking algorithms that require no mutexes (though you do need low level access to interlocked instructions, or barriers). Unfortunately that's where it gets nearly impossible because proving that a non-blocking algorithm is correct requires a combinatorial analysis. I could go into the some of the principles of non-blocking algorithms, but they're still very hard even when you know what you're doing. Still it's useful to understand some basic interlocked instructions because people often use mutexes when a simple interlocked instruction would do the same task infinitely quicker.
3. Having a few basic non-blocking queue algorithms (LIFO and FIFO) up your sleeve can suffice to allow rather separate tasks to communicate cheaply. There are even nonblocking queues that allow an unlimited number of processors to compete for each end of the queue simultaneously, and so you can have common banks of work to be balanced between processors, for instance.
4. It's often a win to give very separate tasks to each processor, and have a single process per processor. Put all IO on one processor, for instance and have it communicate with the other processors through non-blocking queues.
5. Even though fair preemptive multitasking is expensive (because of the task switches), in special cases you can use cooperative multitasking (fibers, green threading) and save the time. One advantage to cooperative multitasking is that you can control exactly when a fiber wakes up. In some benchmarks I've seen (producer-consumer), green threading outraces preemptive multitasking by over an order of magnitude.
6. If you can all tie all interrupts to a single processor (which is possible in Windows) and lock specific processes to each processor, then you can (almost) guarantee that certain processors won't interrupted, and thus those processors are the ones that are cheapest to use for any task that the other processors will have to wait for.
7. Aggregate. There are some tasks that can not be accomplished in parallel with the usual work. It's useful to save up a bunch of work like that so that you don't have to shut down the other tasks too often.
8. State machines with atomic arcs. This gets closer to the harder sort of algorithms, but its useful to talk about what's easy. If you need a bunch of processors to cooperate, it's useful to have a state machine description of the global state. If the state can be stored in a single double word, then transitions between states can be accomplished atomically with a CAS instruction (compare and swap), and then it doesn't matter which processor changes the state. They may all try, and one will succeed. If one of the states is too complicated to fit in a double word, then you can either have a spin-lock intermediate state or in some cases, the state double word can include a pointer (but this gets complicated).
9. Cache delays. Multiprocessor systems have to keep the caches coherent. Because of this situations like a CPU reading a value that's dirty in another processor's cache can bring all of the processors to a halt while they transfer data. I'm a few years out of date on what the newest processors do, but it's probably still true, and still true that these pauses are much longer on systems like the Sparc that allow hundreds of processors than it is in x86 machines that are more limited in the number of processors. In any case it's a win to keep data memory use separate between processors when possible.
I take the $10 bit to be exagguration that means "as cheap as possible".
So if the machine costs $35 at its cheapest point, the $10 promise will still have been useful as a lie to get the political machine moving... That's how this works. It's not quite a first world country we're talking about after all, and that's the whole point. We shouldn't expect Indian politics to be that precise and honest.
I just hope that any machine they come up with will have enough flash to hold Linux, a developement environment or two and some extra packages. It will be of much less use to the children if it doesn't.
"Walk a mile in the shoes you want other people to wear and I might believe you."
I might be nice if you tried to walk a mile in the shoes of children whose community can't afford $100 machines (or medicine or books or buildings for classrooms etc. etc.).
My ignorance it palpable? Jesus, not only don't you know what you're talking about, but your an ass about it too!
Remember the Tandy 102?
It had a reflective LCD screen. Low res but VERY easy to read. Very low power. Very cheap. 1983 technology! And that's what I was talking about.
I also used a sinclair Cambridge z-88. Once again, easy to read screen, if too small. I burned an eprom to make a dvorak keyboard for one of those.
The $100 laptop was always a bit strange in that recognizing that current color displays were too short lived and too power hungry for the third world they decided to spend years inventing a new color screen technology instead of going with the long perfected, cheap (and much lower cost and power) reflective black and white LCD displays.
They obsessed over "electric paper" because they thought it "looked like paper". What nerds!!! Who gives a damn? Black and white reflective LCDs lower resolution, but they have very high contrast and are very readable. But not cool enough for trust fund kids and Negroponte's upper crust crowd.
I understand that children may be more enthusiastic to have color toys with sound and everything, but from the point of view of really poor places having three or four times as many machines is better. It's better to have something than nothing.