The COSA model is a revolutionary new approach to software construction. It is based on the hypothesis that the reason that software is so unreliable is that it is algorithmic. Switch to a synchronous, signal-based model and the problem will disappear.
What happened to the days when people discussed real-life issues rather than masturbation like this?
IMO, you're probably a greedy SOB who hopes to make a killing by abusing an evil system that takes away the liberty of others. I hope some nasty-looking alien from Andromeda knocks on your door one day and demands payment for using the wheel which they had patented a billion years ago in their home world. Any display of public masturbation on your part won't be accepted as payment, sorry. ahahaha...
Figure out a way to get "chance" into that automatic-restitution scheme of yours, and you'll have a workable replacement. But if a crappy invention and a great invention both get rewarded the same, well, then you've just re-invented communism.
The restitution scheme should be adjusted to reflect the value of the invention to society. Value should be based on economic benefits, work done in developing the invention, use, etc... So, if you invent a shit simulator, you get nothing; but if you invent true AI, then society owes you a lot. The value should be computed (and retroactively applied, if necessary) by a formula that is refined as we gain more experience with the system so as to encourage creativity without stifling freedom.
Any inventor/writer/musician should be free to file a claim with the restitution bureau. This is a workable system and it's not communism. It's all about fairness and free market. Current IP laws are evil, in my opinion.
You don't know anything about the patent in question. How much did Iridian sink into developing it? How much have they made back from it? How much would they have made back if this flood of copycats came immediately after they announced their discovery?
IMO, companies should be reimbursed only for their costs by a properly instituted restitution system. Yes, we (the government) should all chip in to pay the cost out of our tax money. The important thing is that nobody should be refrained from using an idea. It is an artificial infringement on liberty and just plain stupid. More often than not, patents give the discoverer/holder an unfair advantage that gives them the power to gouge the public and make a killing. Witness Xerox, Microsoft, etc... Bill Gates does not deserve all this money for DOS and Windows.
As soon as Xerox's patents expired and competitors joined in, we saw an amazing expansion of copier technology. Xerox found it hard to compete on a level playing field. I am not shedding any tears for them.
this thing will be simulating about 10 thousand neurons
It will simulate many more than that, IMO. What most people fail to consider is that only a relatively small number of neurons are active at any one time. This is especially true in the motor cortex and sequence memory areas of the neocortex because we can only do and think about a few things at a time. In addition, if one considers that neurons are discrete signal processors (they generate discrete spikes) the performance requirements can be cut down by orders of magnitude overall.
You Can't Copy Consciousness
on
Download Your Brain
·
· Score: 1, Interesting
Conciousness is so poorly understood that I don't think you can even say that for sure.
Yep. What if you made a copy of yourself, which one is the real you? What if the copy decides you're the one who's is not the real you and should be destroyed?
This whole business of uploading the mind onto a computer is so much unmitigated crackpottery. Star-Trek voodoo science, that's all. You can copy the brain but you can't copy consciousness. For one, you don't know what it is. Until you do, you're up shit creek. And when you find out what it is, you'll realize that it can be neither copied, nor created, nor destroyed.
Having said that, if someone found a way to copy the brain and move your unique consciousness into the copy, now that would be cool!
There is something fundamentally wrong with the way we create software. Contrary to conventional wisdom, unreliability is not an essential characteristic of complex software programs. The solution to the unreliability problem will require a radical change in the way we program our computers. I argue that the main reason that software is so unreliable and so hard to develop has to do with a custom that is as old as the computer: the practice of using the algorithm as the basis of software construction. I argue further that moving to a signal-based, synchronous (**) software model will not only result in an improvement of several orders of magnitude in productivity, but also in programs that are guaranteed free of defects, regardless of their complexity. For additional info on why algorithmic software is the werewolf that must be killed with a silver bullet, go to link below.
What is the evolutionary advanage of being in love with music? Why is music so important since it is not needed for survival? Is there any remnant human species on earth who doesn't care about music? Did music-loving humans kill off the non-music loving humans at some distant time in the past?
I mean, where are the counter-examples? Where is the music-indifferent missing link, so to speak?
I guess Ray Kurzeil's predictions that computers will have the same power as the human brain by 2020 will not be met...
It can be met right now using clusters. The technology is here now. The problem is that we can't even make a machine as intelligent as a honey bee (only about 1 million neurons), what good would a system with a hundred billion neurons be other than to sit and vegetate?
This chip could be used in clusters like nobody's business. An array of 128 of these processors could simultaneously handle 8,192 active threads.
True. But the problem with clusters and multiprocessing in general is that, if the application requires a lot of interprocess communication (e.g., a highly connected neural network), performance suffers. The whole thing boils down to the von Neuman bottleneck. Until somebody comes up with a type of cheap memory that allows unlimited parallel access, we will continue to be hampered. When parallel memory arrives on the scene, processor speed will not matter much for most situations: just have a whole bunch of cheap processors access a single huge memory space and voila!
IMO, cheap non-von Neumann memory technology should be the main holy grail of fast computing, not CPU technology.
Death allows for greater variety and variety is one of the key factors in evolutionary success.
Something's weird here. The whole point of evolution is survival. Whether it is the survival of the individual or the gene makes no difference, IMO. One would think that a species that did not die, would be extremely successful, evolutionarily speaking. Why should variety cum death be considered more successful than a stable uniform species that lives forever?
And while we're at it, why no evolve a system whereby the organism adapts by changing itself as opposed to changing its offsprings through a complex process involving sex? And one more thing, what is so important about music to evolution that all known human races spend a good deal of time creating and listening to it? Some things to think about.
re: unlimited simultaneous memory access - it's called a crossbar switch.
Yes, but I was thinking of a new and more practical architecture, something revolutionary and cheap, maybe a new optical memory. This should be the holy grail of computing research, IMO.
How about a more intelligent (parallel) architecture to begin with?
Unless you have a way around the von Neumann bottleneck, what intelligent architecture are you thinking about? Adding multiple cores will eventually hit a wall because of memory bus contention. The only solution I see is for someone to create a memory architecture that permits unlimited simultaneous memory access. At that point, fast processors will not matter much. Just have a bunch of cheap processors share a single huge memory space.
Each neuron really does operate as a mini processor but it doesn't have access to ALL memory. It only has access to a small fraction of it.
I think you may be confusing connection with access. A neuron can potentially connect itself to any other neuron in the brain via an axonic fiber (no access restriction). Every neuron receives signals from thousands of other neurons on the average. Some neurons in the cerebellum are known to be connected to more than one hundred thousand others. Billions of neurons can process signals simultaneously. This is what unlimited simultaneous access means, IMO. We need a memory system for our computers that has no limitation on parallel/simultaneous access.
It has very slow(compared to transistors, gates, flip flops) individual processing and storage elements yet has incredible throughput and paralell processing power.
Indeed. There is no memory access bottleneck in the brain. Now imagine how much faster the brain would be if each individual processing element was running at 1 ghz or more! Imagine a system with huge numbers of small dedicated and superfast processors with simultaneous access to memory. Extremely fast computing will arrive soon after we get rid of the memory access bottleneck. Does anybody know of any ongoing research on this aspect of computing?
- Formal verification is probably easier to perform with hardware.
True. And this is the reason that we should be writing software pretty much the same way logic designers design logic circuits. That's the basic idea behind synchronous reactive programming languages like Esterel, Signal, Occam and others. Also check out Project COSA at the link below.
If it actually works, then there's little to complain about.
It can only work for so long. The biggest problem that is keeping performance down is not the processor but the memory retrieval and writing system: only one memory location can be accessed at any one time. This is also known as the von Neumann bottleneck. Not even clustering can get around this problem because there is a need for inter-process communication that slows things down. If someone could come up with a system that allows unlimited random and simultaneous memory access, the physical limit to processor speed would not be such a big deal anymore. We would have found the holy grail of fast computing.
It may irk you that some very smart people know and understand stuff you don't, that doesn't mean it should be dismissed.
Pomposity is a sure sign of cluelessness, IMO. There's plenty of it in the scientific community and, strangely enough, in religious circles. Anybody who thinks the fundamental is complex is either clueless or has redfined the meaning of the term to suit his or her own purpose. If it is not simple it is certainly not fundamental.
There is no reason at all to assume the rules the universe operates under actually are intelligible to humans at all (our brains evolved for survival, not science), let alone the average layperson.
I beg to differ. There is every reason to suppose that the fundamental is simple by virtue of being fundamental. As I wrote in another post, if you don't understand the fundamentals of something, you are as clueless as everybody else. Any claim to the contrary is just pomposity.
I took many university level physics and math courses, and I still have no idea of how to visualize fractional spin, for example. And believe me I've tried. These disciplines are truly humbling even to the most gifted individuals.
You make my point for me. Everything is simple at the fundamental level. Unless you understand the fundamentals of a phenomenon, you have no real understanding.
Just because you don't understand things, doesn't make them "Crackpottery" or "vodoo".
In my opinion, if a scientist comes up with a theory about a phenomenon and is unable to explain it in a simple manner that is intelligible to the average layperson, one can bet that said scientist is as clueless about the nature of the phenomenon as everyone else. Truth is, a million know-it-alls claiming otherwise notwithstanding, nobody really understands black holes, wormholes, the cause of gravity, or how a dimension can be compactified into a tiny ball. If it walks like a duck...
Crackpottery in high places is the most dangerous form of crackpottery because it can lead to a wild goose chase that lasts a thousand years. Lately, many of the mystical (unexplainable) stuff from the high priests of academia seem like voodoo science to me. It's the kind of stuff religions are made of. Black holes, wormholes, time travel, parallel universes, quantum computers, string theory are just a few that fall into this category. Just one man's opinion, of course.
Slashdot Mirror
However, after so long you're gonna start getting rehashed methods
Here is a revolutionary new programming model that addresses the biggest problem in computing, software unreliability. No rehashed methods.
Project COSA:
http://www.rebelscience.org/Cosas/COSA.htm
The COSA model is a revolutionary new approach to software construction. It is based on the hypothesis that the reason that software is so unreliable is that it is algorithmic. Switch to a synchronous, signal-based model and the problem will disappear.
What happened to the days when people discussed real-life issues rather than masturbation like this?
IMO, you're probably a greedy SOB who hopes to make a killing by abusing an evil system that takes away the liberty of others. I hope some nasty-looking alien from Andromeda knocks on your door one day and demands payment for using the wheel which they had patented a billion years ago in their home world. Any display of public masturbation on your part won't be accepted as payment, sorry. ahahaha...
Figure out a way to get "chance" into that automatic-restitution scheme of yours, and you'll have a workable replacement. But if a crappy invention and a great invention both get rewarded the same, well, then you've just re-invented communism.
The restitution scheme should be adjusted to reflect the value of the invention to society. Value should be based on economic benefits, work done in developing the invention, use, etc... So, if you invent a shit simulator, you get nothing; but if you invent true AI, then society owes you a lot. The value should be computed (and retroactively applied, if necessary) by a formula that is refined as we gain more experience with the system so as to encourage creativity without stifling freedom.
Any inventor/writer/musician should be free to file a claim with the restitution bureau. This is a workable system and it's not communism. It's all about fairness and free market. Current IP laws are evil, in my opinion.
You don't know anything about the patent in question. How much did Iridian sink into developing it? How much have they made back from it? How much would they have made back if this flood of copycats came immediately after they announced their discovery?
IMO, companies should be reimbursed only for their costs by a properly instituted restitution system. Yes, we (the government) should all chip in to pay the cost out of our tax money. The important thing is that nobody should be refrained from using an idea. It is an artificial infringement on liberty and just plain stupid. More often than not, patents give the discoverer/holder an unfair advantage that gives them the power to gouge the public and make a killing. Witness Xerox, Microsoft, etc... Bill Gates does not deserve all this money for DOS and Windows.
As soon as Xerox's patents expired and competitors joined in, we saw an amazing expansion of copier technology. Xerox found it hard to compete on a level playing field. I am not shedding any tears for them.
this thing will be simulating about 10 thousand neurons
It will simulate many more than that, IMO. What most people fail to consider is that only a relatively small number of neurons are active at any one time. This is especially true in the motor cortex and sequence memory areas of the neocortex because we can only do and think about a few things at a time. In addition, if one considers that neurons are discrete signal processors (they generate discrete spikes) the performance requirements can be cut down by orders of magnitude overall.
Conciousness is so poorly understood that I don't think you can even say that for sure.
Yep. What if you made a copy of yourself, which one is the real you? What if the copy decides you're the one who's is not the real you and should be destroyed?
This whole business of uploading the mind onto a computer is so much unmitigated crackpottery. Star-Trek voodoo science, that's all. You can copy the brain but you can't copy consciousness. For one, you don't know what it is. Until you do, you're up shit creek. And when you find out what it is, you'll realize that it can be neither copied, nor created, nor destroyed.
Having said that, if someone found a way to copy the brain and move your unique consciousness into the copy, now that would be cool!
There is something fundamentally wrong with the way we create software. Contrary to conventional wisdom, unreliability is not an essential characteristic of complex software programs. The solution to the unreliability problem will require a radical change in the way we program our computers. I argue that the main reason that software is so unreliable and so hard to develop has to do with a custom that is as old as the computer: the practice of using the algorithm as the basis of software construction. I argue further that moving to a signal-based, synchronous (**) software model will not only result in an improvement of several orders of magnitude in productivity, but also in programs that are guaranteed free of defects, regardless of their complexity. For additional info on why algorithmic software is the werewolf that must be killed with a silver bullet, go to link below.
What is the evolutionary advanage of being in love with music? Why is music so important since it is not needed for survival? Is there any remnant human species on earth who doesn't care about music? Did music-loving humans kill off the non-music loving humans at some distant time in the past?
I mean, where are the counter-examples? Where is the music-indifferent missing link, so to speak?
I got the perfect OS for the Cell Processor. I just need funding. :-D
The COSA Operatin System
See also the link below.
I guess Ray Kurzeil's predictions that computers will have the same power as the human brain by 2020 will not be met...
It can be met right now using clusters. The technology is here now. The problem is that we can't even make a machine as intelligent as a honey bee (only about 1 million neurons), what good would a system with a hundred billion neurons be other than to sit and vegetate?
This chip could be used in clusters like nobody's business. An array of 128 of these processors could simultaneously handle 8,192 active threads.
True. But the problem with clusters and multiprocessing in general is that, if the application requires a lot of interprocess communication (e.g., a highly connected neural network), performance suffers. The whole thing boils down to the von Neuman bottleneck. Until somebody comes up with a type of cheap memory that allows unlimited parallel access, we will continue to be hampered. When parallel memory arrives on the scene, processor speed will not matter much for most situations: just have a whole bunch of cheap processors access a single huge memory space and voila!
IMO, cheap non-von Neumann memory technology should be the main holy grail of fast computing, not CPU technology.
So they didn't measure anything at all at 600 GHz; they only found device behavior that indicated that there would be a response up to that frequency.
Thanks for RTFA. I'm too lazy right now. It's seems to be an indirect (i.e., inferred) measurement. I wonder how accurate that is.
the researchers have demonstrated a speed of 604 gigahertz - the fastest transistor operation to date.
How does one measure 604 gigahertz? Just asking.
Death allows for greater variety and variety is one of the key factors in evolutionary success.
Something's weird here. The whole point of evolution is survival. Whether it is the survival of the individual or the gene makes no difference, IMO. One would think that a species that did not die, would be extremely successful, evolutionarily speaking. Why should variety cum death be considered more successful than a stable uniform species that lives forever?
And while we're at it, why no evolve a system whereby the organism adapts by changing itself as opposed to changing its offsprings through a complex process involving sex? And one more thing, what is so important about music to evolution that all known human races spend a good deal of time creating and listening to it? Some things to think about.
re: unlimited simultaneous memory access - it's called a crossbar switch.
Yes, but I was thinking of a new and more practical architecture, something revolutionary and cheap, maybe a new optical memory. This should be the holy grail of computing research, IMO.
How about a more intelligent (parallel) architecture to begin with?
Unless you have a way around the von Neumann bottleneck, what intelligent architecture are you thinking about? Adding multiple cores will eventually hit a wall because of memory bus contention. The only solution I see is for someone to create a memory architecture that permits unlimited simultaneous memory access. At that point, fast processors will not matter much. Just have a bunch of cheap processors share a single huge memory space.
Each neuron really does operate as a mini processor but it doesn't have access to ALL memory. It only has access to a small fraction of it.
I think you may be confusing connection with access. A neuron can potentially connect itself to any other neuron in the brain via an axonic fiber (no access restriction). Every neuron receives signals from thousands of other neurons on the average. Some neurons in the cerebellum are known to be connected to more than one hundred thousand others. Billions of neurons can process signals simultaneously. This is what unlimited simultaneous access means, IMO. We need a memory system for our computers that has no limitation on parallel/simultaneous access.
It has very slow(compared to transistors, gates, flip flops) individual processing and storage elements yet has incredible throughput and paralell processing power.
Indeed. There is no memory access bottleneck in the brain. Now imagine how much faster the brain would be if each individual processing element was running at 1 ghz or more! Imagine a system with huge numbers of small dedicated and superfast processors with simultaneous access to memory. Extremely fast computing will arrive soon after we get rid of the memory access bottleneck. Does anybody know of any ongoing research on this aspect of computing?
- Formal verification is probably easier to perform with hardware.
True. And this is the reason that we should be writing software pretty much the same way logic designers design logic circuits. That's the basic idea behind synchronous reactive programming languages like Esterel, Signal, Occam and others. Also check out Project COSA at the link below.
If it actually works, then there's little to complain about.
It can only work for so long. The biggest problem that is keeping performance down is not the processor but the memory retrieval and writing system: only one memory location can be accessed at any one time. This is also known as the von Neumann bottleneck. Not even clustering can get around this problem because there is a need for inter-process communication that slows things down. If someone could come up with a system that allows unlimited random and simultaneous memory access, the physical limit to processor speed would not be such a big deal anymore. We would have found the holy grail of fast computing.
It may irk you that some very smart people know and understand stuff you don't, that doesn't mean it should be dismissed.
Pomposity is a sure sign of cluelessness, IMO. There's plenty of it in the scientific community and, strangely enough, in religious circles. Anybody who thinks the fundamental is complex is either clueless or has redfined the meaning of the term to suit his or her own purpose. If it is not simple it is certainly not fundamental.
There is no reason at all to assume the rules the universe operates under actually are intelligible to humans at all (our brains evolved for survival, not science), let alone the average layperson.
I beg to differ. There is every reason to suppose that the fundamental is simple by virtue of being fundamental. As I wrote in another post, if you don't understand the fundamentals of something, you are as clueless as everybody else. Any claim to the contrary is just pomposity.
I took many university level physics and math courses, and I still have no idea of how to visualize fractional spin, for example. And believe me I've tried. These disciplines are truly humbling even to the most gifted individuals.
You make my point for me. Everything is simple at the fundamental level. Unless you understand the fundamentals of a phenomenon, you have no real understanding.
Just because you don't understand things, doesn't make them "Crackpottery" or "vodoo".
In my opinion, if a scientist comes up with a theory about a phenomenon and is unable to explain it in a simple manner that is intelligible to the average layperson, one can bet that said scientist is as clueless about the nature of the phenomenon as everyone else. Truth is, a million know-it-alls claiming otherwise notwithstanding, nobody really understands black holes, wormholes, the cause of gravity, or how a dimension can be compactified into a tiny ball. If it walks like a duck...
Crackpottery in high places is the most dangerous form of crackpottery because it can lead to a wild goose chase that lasts a thousand years. Lately, many of the mystical (unexplainable) stuff from the high priests of academia seem like voodoo science to me. It's the kind of stuff religions are made of. Black holes, wormholes, time travel, parallel universes, quantum computers, string theory are just a few that fall into this category. Just one man's opinion, of course.