That makes sense. I guess I know that there are long-winded proofs, and shorter more elegant ones. It must also depend on how difficult the problem is, and how many pieces of the puzzle (intermediate results, notations) have been provided.
I know a guy who taught computational complexity, and had to mark students code written in a lisp-like language. He was half-seriously thinking about writing an automated prover to test their code for correctness. It might be a hard problem, but testing proofs/code for isomorphic solutions seems possible:)
Will the solution to a problem often have only one series of steps, in one order? As in - is it possible to tell that the (correct) answer a student gives is his own, or from a third party?
I guess the only experience I have of this is implementations of algorithms in code. It is usually possible to tell who wrote a piece of code by the style (whitespace, variable naming, etc) but for mathematical proofs, I'm not so sure. I guess the names of the mathematical objects used could be a giveaway, like if half the students start the answer with "let Zeta0 = {z0, z1,..., zr}" and everyone else used different letters. Hmm...
I wonder how much the spread of 'gifting' (pyramid) schemes is like the spread of an infectious disease.
People must get 'inoculated' against them (by losing their money), but depending on how weak their mental immune system is - how stupid they are - they can catch the disease again.
There could also be various strategies in a population of people who are joining and leaving schemes at various times. This should be a research project; along the lines of research into iterated spatial prisoner's dilemma models.
In order to balance out this comment, I will add a youtube-style comment:
"OMG! THZ GIFTING PROGRAM IS LIEK TOTALLY COOOL! CAN I GET MY LAMBORGINI NOW?!!?"
They (http://morepypy.blogspot.com/) have noticed the project, it seems.
We were a bit confused about usage of the term JIT, because as far as we understood, it's going to be upfront compilation into LLVM. In the past we have looked into LLVM - at one point PyPy extensively use it but it wasn't clear how we could make good use to it.
Without realising it! I was trying to draw a diagram similar to the one in the article (a mixture of circles of different sizes, within a larger circle).
In my case, it was to represent the sizes of subsets of an isomer space, where each subset shares the same predicted 13C-NMR spectrum.
This makes me feel slightly better about doing so badly at it (using adaptive simulated annealing)
There is an excellent book by Alexander Graham Cairns-Smith called "Seven Clues to the Origin of Life" that talks about such self-replicating clay
The main feature of his argument is that the clay surfaces could serve as templates for catalysis of polynucleotides (RNA, probably). These, then would form the first RNA world.
He uses the metaphor of a rope, where no strand goes from one end to the other - the rope is time, and strands within it are clayworld, rna world, dna world...
Exactly right. Carbon rich molecules are more diverse and larger than any other sort.
You can form chains or rings of around 6 sulphurs (with oxygen), but carbon can be found in chains of 30+ atoms and in multiple ring systems.
It's very difficult to grasp how large the isomer spaces are - and how quickly they grow, but a recent guestimate I made was that if a program (molgen) can enumerate all possible C10H16 molecules in 2 seconds, and all C13H22 in 2 minutes, then it would take 2 days for C18H36 and 1 billion years for C36H72...
Also, there are 25,000 C10s and 9 million C15s. So the sheer number of possible carbon compounds argues that carbon is the only likely candidate.
Even if it was just as good as DNA at some point, evolution is a historical science. An arbitrary 'choice' made in the past can steer the future of a system away from what would be a more fit state.
The choice between left- and right- handed amino acids was one such decision that was fixed by the system freezing into using one handedness over the other.
A slight difference in the proportion of DNA:PNA could have been amplified by feedback until only one survived.
As far as I understand, the only valid point to the Anthropic Principle is that it is a default answer to a question of "why this". So, (to stretch the lottery analogy), if you are asking "Why do I sleep on a big pile of money with two beautiful women?" the answer is "Because I won the lottery".
That is; Q : "Why does the Universe have solid matter?"; A : "If it did not, we would not be around to see it". And so on.
..the fact that life exists in this universe can be seen as simply a curious coincidence..
I know this might seem pedantic, but isn't "coincidence" when two or more things happen. So, if my friend and I turn up at the same place at the same time, without planning to do so, that's coincidence.
So, our existance in the Universe is merely "incidence". It is not 'co-' with anything else.
Yes, that's why my link is about a biological "optical-fibre" example:) In short, it's about the 'Sea Mouse' which has chitin spines that are photonic band-gap crystals (I read - I'm a biologist, too, not a physicist).
I'd go as far as to rule out any other chemical basis for life.
Agreed. I would go even further and suggest that DNA/RNA/protein are the only possibilities for living systems.
Of course, chemical space is very large, but there is a relatively small subspace that is (bio)synthetically accessible. Further, there is an even smaller region that is self-synthetically accessible. Even more, there is a tiny part that can form spontaneously on a planet, self-synthesise, and evolve.
And I have spent 5 years of my life on the topology of proteins. It is not quite true to refer to "knots" when talking about proteins, as Professor Taylor has shown (http://www.nature.com/nature/journal/v406/n6798/full/406916a0.html) that only a few proteins are actually 'knotted'.
However, mathematical theory of tangled strings is as important as simulations. Estimating the total number of folds, for example. More than just a fancy excursion - but maybe not to your taste?
You are right, that is the idea. From Behe's book "Darwin's Black Box" (a pretty stupid book) the 'problem' is that systems can be "irreducibly complex". That is, like the mousetrap - remove or change any part and it stops working.
The problem again (and since Behe's is a biochemist he is either stupid or lying if he doesn't understand this) is that nature builds her mousetraps in a very different way.
All previous 'versions' of any particular mousetrap (or other design) HAD to work. Small changes to them, including replacing parts or modifying parts were made, and those mousetraps that failed to catch any mice were rejected (died off).
This is only possible with systems whose parts can be replaced with slightly similar ones, and still sort of work. Evolutionary systems have evolved to be evolvable.
So, it's not the self-assembly, but the mutability of natural systems that is under dispute. Most biologists understand that natural systems can change quite radically - species evolution - while a few ID-ers just don't get it. Their loss; natural systems are truly astonishing.
Very good point, but the attack on Intelligent Design doesn't _quite follow (sadly!).
Not that I am a fan of ID. Not on the same page at all. Perhaps not even in the same library. But there is nothing illogical about considering organisms to be 'designed' the same way that technology is.
I think it is because complex design is so familiar to us that it is easy for some people to assume that's how biological systems came about.
In fact, biological structures are very different to most human-designed artifacts. They self-assemble, are wet, made from millions to trillions of tiny parts, exploit chemistry to an amazing degree, and so on.
Ironically, it is only now we are starting to design systems that are nearly as complex as biological ones that most people are cut off from the design process and many are probably unable to understand modern high tech.
So, yes the ID advocate in a high-tech SUV is ridiculous. What he sees is the large-scale parts (wheels, body, seats) and fails to notice the tiny parts in the airbag trigger.
Is organic chemistry grimly humerus? Or twisted out of shape - perhaps into a boat or a chair?
Is the skew E- or Z-? D- or L-? That's the important thing.
Slashdot Mirror
The FIRST WORD of the story is 'Dr', closely followed by his name. Did that not clue you into the fact that he already has a PhD?
That makes sense. I guess I know that there are long-winded proofs, and shorter more elegant ones. It must also depend on how difficult the problem is, and how many pieces of the puzzle (intermediate results, notations) have been provided.
I know a guy who taught computational complexity, and had to mark students code written in a lisp-like language. He was half-seriously thinking about writing an automated prover to test their code for correctness. It might be a hard problem, but testing proofs/code for isomorphic solutions seems possible :)
Will the solution to a problem often have only one series of steps, in one order? As in - is it possible to tell that the (correct) answer a student gives is his own, or from a third party?
I guess the only experience I have of this is implementations of algorithms in code. It is usually possible to tell who wrote a piece of code by the style (whitespace, variable naming, etc) but for mathematical proofs, I'm not so sure. I guess the names of the mathematical objects used could be a giveaway, like if half the students start the answer with "let Zeta0 = {z0, z1, ..., zr}" and everyone else used different letters. Hmm...
I wonder how much the spread of 'gifting' (pyramid) schemes is like the spread of an infectious disease.
People must get 'inoculated' against them (by losing their money), but depending on how weak their mental immune system is - how stupid they are - they can catch the disease again.
There could also be various strategies in a population of people who are joining and leaving schemes at various times. This should be a research project; along the lines of research into iterated spatial prisoner's dilemma models.
In order to balance out this comment, I will add a youtube-style comment:
"OMG! THZ GIFTING PROGRAM IS LIEK TOTALLY COOOL! CAN I GET MY LAMBORGINI NOW?!!?"
Well, given an alphabet of just 'a-z' + ' ', that would be 26 ^ 140 or:
1, 248, 155, 560, 712, 888, 693, 721, 116, 035, 178, 646, 463, 649, 590, 092, 724, 076, 699, 557, 919, 198, 775, 318, 840, 655, 335, 967, 337, 203, 969, 601, 545, 498, 350, 937, 608, 330, 255, 529, 112, 180, 176, 094, 892, 997, 792, 623, 787, 890, 917, 357, 870, 916, 489, 701, 094, 150, 005, 153, 729, 071, 148, 146, 282, 725, 376
which is quite a few possibilities.
They (http://morepypy.blogspot.com/) have noticed the project, it seems.
We were a bit confused about usage of the term JIT, because as far as we understood, it's going to be upfront compilation into LLVM. In the past we have looked into LLVM - at one point PyPy extensively use it but it wasn't clear how we could make good use to it.
They seem a bit sceptical.
Then you have to genetically engineer lava-sharks to put your lasers on.
Without realising it! I was trying to draw a diagram similar to the one in the article (a mixture of circles of different sizes, within a larger circle).
In my case, it was to represent the sizes of subsets of an isomer space, where each subset shares the same predicted 13C-NMR spectrum.
This makes me feel slightly better about doing so badly at it (using adaptive simulated annealing)
(Quote from the arxivblog post)
Is this some physics jargon for lab assistant? Or is it just a clockwork orange referenced insult?
There is an excellent book by Alexander Graham Cairns-Smith called "Seven Clues to the Origin of Life" that talks about such self-replicating clay
The main feature of his argument is that the clay surfaces could serve as templates for catalysis of polynucleotides (RNA, probably). These, then would form the first RNA world.
He uses the metaphor of a rope, where no strand goes from one end to the other - the rope is time, and strands within it are clayworld, rna world, dna world...
You can form chains or rings of around 6 sulphurs (with oxygen), but carbon can be found in chains of 30+ atoms and in multiple ring systems.
It's very difficult to grasp how large the isomer spaces are - and how quickly they grow, but a recent guestimate I made was that if a program (molgen) can enumerate all possible C10H16 molecules in 2 seconds, and all C13H22 in 2 minutes, then it would take 2 days for C18H36 and 1 billion years for C36H72...
Also, there are 25,000 C10s and 9 million C15s. So the sheer number of possible carbon compounds argues that carbon is the only likely candidate.
GP was quoting. It was -15 on the 14th Jan 300 years ago.
I agree ... in principle. The picture is of Gates with AN OPEN JAR - which is not proof of him opening a jar full of mosquitoes.
Indeed, the picture could be of him at any time, on a stage, with an open jar in his hand! :)
"At least one side of one sheep..."
Even if it was just as good as DNA at some point, evolution is a historical science. An arbitrary 'choice' made in the past can steer the future of a system away from what would be a more fit state.
The choice between left- and right- handed amino acids was one such decision that was fixed by the system freezing into using one handedness over the other.
A slight difference in the proportion of DNA:PNA could have been amplified by feedback until only one survived.
Ha! That's exactly right.
As far as I understand, the only valid point to the Anthropic Principle is that it is a default answer to a question of "why this". So, (to stretch the lottery analogy), if you are asking "Why do I sleep on a big pile of money with two beautiful women?" the answer is "Because I won the lottery".
That is; Q : "Why does the Universe have solid matter?"; A : "If it did not, we would not be around to see it". And so on.
..the fact that life exists in this universe can be seen as simply a curious coincidence..
I know this might seem pedantic, but isn't "coincidence" when two or more things happen. So, if my friend and I turn up at the same place at the same time, without planning to do so, that's coincidence.
So, our existance in the Universe is merely "incidence". It is not 'co-' with anything else.
The explanation that has been given in the Bible is the only one whose source is decidedly not human. [It is] an answered question.
To clarify: yes, you are a creationist.
If you believe that an Intelligent Designer created life AND that the science is looking at an "answered" question, then you are a creationist.
If you are uncomfortable with the company you keep, of Biblical literalists, then that is your problem...
Yes, that's why my link is about a biological "optical-fibre" example :) In short, it's about the 'Sea Mouse' which has chitin spines that are photonic band-gap crystals (I read - I'm a biologist, too, not a physicist).
There are examples of biological optics:
A small node on one example
I'd go as far as to rule out any other chemical basis for life.
Agreed. I would go even further and suggest that DNA/RNA/protein are the only possibilities for living systems.
Of course, chemical space is very large, but there is a relatively small subspace that is (bio)synthetically accessible. Further, there is an even smaller region that is self-synthetically accessible. Even more, there is a tiny part that can form spontaneously on a planet, self-synthesise, and evolve.
That's also what's most confusing. Their database is, at heart, a list of books.
How can they imagine that they could prevent other people from making a similar list?
And I have spent 5 years of my life on the topology of proteins. It is not quite true to refer to "knots" when talking about proteins, as Professor Taylor has shown (http://www.nature.com/nature/journal/v406/n6798/full/406916a0.html) that only a few proteins are actually 'knotted'.
However, mathematical theory of tangled strings is as important as simulations. Estimating the total number of folds, for example. More than just a fancy excursion - but maybe not to your taste?
You are right, that is the idea. From Behe's book "Darwin's Black Box" (a pretty stupid book) the 'problem' is that systems can be "irreducibly complex". That is, like the mousetrap - remove or change any part and it stops working.
The problem again (and since Behe's is a biochemist he is either stupid or lying if he doesn't understand this) is that nature builds her mousetraps in a very different way.
All previous 'versions' of any particular mousetrap (or other design) HAD to work. Small changes to them, including replacing parts or modifying parts were made, and those mousetraps that failed to catch any mice were rejected (died off).
This is only possible with systems whose parts can be replaced with slightly similar ones, and still sort of work. Evolutionary systems have evolved to be evolvable.
So, it's not the self-assembly, but the mutability of natural systems that is under dispute. Most biologists understand that natural systems can change quite radically - species evolution - while a few ID-ers just don't get it. Their loss; natural systems are truly astonishing.
Very good point, but the attack on Intelligent Design doesn't _quite follow (sadly!).
Not that I am a fan of ID. Not on the same page at all. Perhaps not even in the same library. But there is nothing illogical about considering organisms to be 'designed' the same way that technology is.
I think it is because complex design is so familiar to us that it is easy for some people to assume that's how biological systems came about.
In fact, biological structures are very different to most human-designed artifacts. They self-assemble, are wet, made from millions to trillions of tiny parts, exploit chemistry to an amazing degree, and so on.
Ironically, it is only now we are starting to design systems that are nearly as complex as biological ones that most people are cut off from the design process and many are probably unable to understand modern high tech.
So, yes the ID advocate in a high-tech SUV is ridiculous. What he sees is the large-scale parts (wheels, body, seats) and fails to notice the tiny parts in the airbag trigger.
Is organic chemistry grimly humerus? Or twisted out of shape - perhaps into a boat or a chair? Is the skew E- or Z-? D- or L-? That's the important thing.