5 ms·

"the next atom of information can be (physically) more or less distant from the previous one" Not true, at least with something like DNA. When something like R

"the next atom of information can be (physically) more or less distant from the previous one" Not true, at least with something like DNA. When something like RNA polymerase goes through DNA, it always goes to the physically closest nucleotide. Sure there is a "link" between bases in the form of molecular bonds, but this bond only works between nucleotides right next to each other. No nucleotide can bond to a nucleotide farther away, and have a polymerase follow that bond. If DNA were a linked list, it would basically be one where every element is adjacent to the next one in memory. The whole point of a linked list is that they don't have to be like that, one element can point to another 4, 12, 32, etc bytes away. That way insertions and deletions are much easier. DNA works much more like an array. The "index" for a gene is the promoter sequence at its beginning. A protein complex forms there and polymerase runs down sequentially from nucleotide to the next physically closest nucleotide. Also, why do you say computers will never substitute a brain when linked lists can quite easily represent a neural network?

polos 17y ago
Computer memory is always aligned in a one-dimensional, (relatively) huge array. All of artificial intelligence is based on a one-dimensional representation of a 3 dimensional fact (our physical existence). DNA, our brain, every bit of information in Nature, always, exclusively, uses all 3 physical dimensions present to store its data, so we never will be able to directly compete with Nature in that fact. Because every dimension present has its own role, and how would you represent that in a simple computer memory (leaving out the fact that computer memory is really huge and really slow compared to the memory systems of Nature). Do you (personally) know anything about how Nature uses all of the 3 dimensions present, to align its data in a really ingenious and unique way? Do you? Physically near has a special meaning in our human brain, and influences all the bits of information near the stored information. So, you see, Nature is able to use all of the 3 dimensions of physical reality to represent its data, while computers (and, consequently, all of AI) seems to be condemned to use a 1 dimensional, slow, physically huge, representation... So, please please, don't directly compare Nature with human investigation -- they are as far from each other as our planet is from sun... The more you observe Nature (and I really do!), the more you experience yourself really ignorant about the Intelligence hiding behind...
- Scriptor 17y ago
  I suggest you download FoldIt (http://fold.it/portal/ http://fold.it/portal/), it does a pretty good job at modeling the 3d interactions in proteins using your computer, which I'm pretty sure also has that "1 dimensional" memory. Do you know how nature uses all 3 dimensions to store information, do you even know what you're talking about? DNA, no matter how "ingenious" its structure, ends up just being a series of bases. Each base in DNA will lead to a particular base in the resulting RNA, this is nothing more than a 1-to-1 mapping, a hash. Ever three bases of RNA (a codon) lead to a particular anti-codon which is mapped to an amino acid. Again, nothing complicated conceptually. You use a series of mappings to translate one series of molecules into another. It can be easily done with a computer. It's only complex in nature because it needs to be complex in nature, otherwise it wouldn't work. And even if 1 dimensional memory is limiting, how long until computers themselves use molecules to read and store data? Your argument is effectively that of someone in the 19th century saying the flight is impossible because current transportation can only move on land. Also, electronic systems are much faster than biological ones, compare the speed of an electric signal in a wire to that of a nerve impulse.
  - polos 17y ago
    In our brain, certain impulses are carried via electricity, so they are as fast as possible (inside their structure). Other, much more complex processes, don't involve electricity directly, but they are still fast enough -- but given their complexity, they are still really fast, relatively speaking, and they need quite no space (more important than everything else...). Time is not a measure for intelligence. If it were so, humans could be defined as some of the least intelligent beings on our planet (as they need many many years to finally be adults; in fact, we need to learn so many things; most animals are quite ready when they are born -- ask somebody who knows animals). As for your speculations of human future: are we talking about science or fiction? Although humans are really good in inventing really fantastical future scenes, that has nothing to do with science -- so let's call it fiction for now...
    - Scriptor 17y ago
      "In our brain, certain impulses are carried via electricity, so they are as fast as possible (inside their structure)" No, electric impulses in your brain are not carried in the same way as electricity in wires. Please read up on basic neuroscience before you talk about the nervous system. Nerve signals are passed by having one section of a nerve be depolarized (lose polarity, the difference in charge from inside and outside the nerve wall), which causes the next section to depolarize, and so on. Your basic electric wire uses a difference in electric potential, or voltage, between one end and the other, which creates an electromagnetic field that moves electrons. You completely misunderstood the importance of time. I'm not talking about long-term development (and even then your average human infant of even 6 months is far more intelligent than most animals on Earth). I meant the speed at which the nervous system works as compared to a computer. If you read my previous paragraph, you can see that a computer is still much faster than the human brain. The brain's advantage is the complex networking of all the neurons. Science has already gone a fair way in understanding the brain, we can even decipher signals to let paralyzed people control computers. How much longer until we can start modeling simple brains, and then further? And as for my speculations, look at what you're doing right now. You, and I, can instantly communicate over a host of different ways even if we are on opposite sides of the Earth. If we wanted to talk face-to-face, we could even see digitized versions of each other in video chat. If we wanted to meet in person, it's only a matter of ours for one of us to go to the other. What do you think people from a hundred years ago would say when you told them such things would be possible? It may take a while, dozens, hundreds, or even thousands of years, but your speculations of the uniqueness of nature are narrow-minded. You say that modern computers can never match nature because they use 1-dimensional data storage. I've already shown you how such storage can model 3d interactions (it's not about the data, it's about the way it's processed). Maybe we should call it fiction for now, but remember that the very idea of having a machine calculate an integral, much less solve protein structures, would have been laughable in the past.

polos 17y ago

Computer memory is always aligned in a one-dimensional, (relatively) huge array. All of artificial intelligence is based on a one-dimensional representation of a 3 dimensional fact (our physical existence). DNA, our brain, every bit of information in Nature, always, exclusively, uses all 3 physical dimensions present to store its data, so we never will be able to directly compete with Nature in that fact. Because every dimension present has its own role, and how would you represent that in a simple computer memory (leaving out the fact that computer memory is really huge and really slow compared to the memory systems of Nature). Do you (personally) know anything about how Nature uses all of the 3 dimensions present, to align its data in a really ingenious and unique way? Do you? Physically near has a special meaning in our human brain, and influences all the bits of information near the stored information. So, you see, Nature is able to use all of the 3 dimensions of physical reality to represent its data, while computers (and, consequently, all of AI) seems to be condemned to use a 1 dimensional, slow, physically huge, representation... So, please please, don't directly compare Nature with human investigation -- they are as far from each other as our planet is from sun... The more you observe Nature (and I really do!), the more you experience yourself really ignorant about the Intelligence hiding behind...

Scriptor 17y ago
I suggest you download FoldIt (http://fold.it/portal/ http://fold.it/portal/), it does a pretty good job at modeling the 3d interactions in proteins using your computer, which I'm pretty sure also has that "1 dimensional" memory. Do you know how nature uses all 3 dimensions to store information, do you even know what you're talking about? DNA, no matter how "ingenious" its structure, ends up just being a series of bases. Each base in DNA will lead to a particular base in the resulting RNA, this is nothing more than a 1-to-1 mapping, a hash. Ever three bases of RNA (a codon) lead to a particular anti-codon which is mapped to an amino acid. Again, nothing complicated conceptually. You use a series of mappings to translate one series of molecules into another. It can be easily done with a computer. It's only complex in nature because it needs to be complex in nature, otherwise it wouldn't work. And even if 1 dimensional memory is limiting, how long until computers themselves use molecules to read and store data? Your argument is effectively that of someone in the 19th century saying the flight is impossible because current transportation can only move on land. Also, electronic systems are much faster than biological ones, compare the speed of an electric signal in a wire to that of a nerve impulse.
- polos 17y ago
  In our brain, certain impulses are carried via electricity, so they are as fast as possible (inside their structure). Other, much more complex processes, don't involve electricity directly, but they are still fast enough -- but given their complexity, they are still really fast, relatively speaking, and they need quite no space (more important than everything else...). Time is not a measure for intelligence. If it were so, humans could be defined as some of the least intelligent beings on our planet (as they need many many years to finally be adults; in fact, we need to learn so many things; most animals are quite ready when they are born -- ask somebody who knows animals). As for your speculations of human future: are we talking about science or fiction? Although humans are really good in inventing really fantastical future scenes, that has nothing to do with science -- so let's call it fiction for now...
  - Scriptor 17y ago
    "In our brain, certain impulses are carried via electricity, so they are as fast as possible (inside their structure)" No, electric impulses in your brain are not carried in the same way as electricity in wires. Please read up on basic neuroscience before you talk about the nervous system. Nerve signals are passed by having one section of a nerve be depolarized (lose polarity, the difference in charge from inside and outside the nerve wall), which causes the next section to depolarize, and so on. Your basic electric wire uses a difference in electric potential, or voltage, between one end and the other, which creates an electromagnetic field that moves electrons. You completely misunderstood the importance of time. I'm not talking about long-term development (and even then your average human infant of even 6 months is far more intelligent than most animals on Earth). I meant the speed at which the nervous system works as compared to a computer. If you read my previous paragraph, you can see that a computer is still much faster than the human brain. The brain's advantage is the complex networking of all the neurons. Science has already gone a fair way in understanding the brain, we can even decipher signals to let paralyzed people control computers. How much longer until we can start modeling simple brains, and then further? And as for my speculations, look at what you're doing right now. You, and I, can instantly communicate over a host of different ways even if we are on opposite sides of the Earth. If we wanted to talk face-to-face, we could even see digitized versions of each other in video chat. If we wanted to meet in person, it's only a matter of ours for one of us to go to the other. What do you think people from a hundred years ago would say when you told them such things would be possible? It may take a while, dozens, hundreds, or even thousands of years, but your speculations of the uniqueness of nature are narrow-minded. You say that modern computers can never match nature because they use 1-dimensional data storage. I've already shown you how such storage can model 3d interactions (it's not about the data, it's about the way it's processed). Maybe we should call it fiction for now, but remember that the very idea of having a machine calculate an integral, much less solve protein structures, would have been laughable in the past.