Closures as life(tmsh.posterous.com)
tmsh.posterous.com
Closures as life
http://tmsh.posterous.com/closures-as-life
9 comments
I guess it's just a phase that people go through. Like writing their first confused monad tutorials / explanations.
I don't think this is complete nonsense. At the root of all this, he's weirded out by the concept of entropy. Entropy exists, yet the complexity on this infinitesimal speck of dust in the universe is increasing. Why is that?
tmsh, I wrote a equally disparaged blog entry a while back that you might be interested in: http://www.jierenchen.com/2009/10/memory-and-evolution.html
tmsh, I wrote a equally disparaged blog entry a while back that you might be interested in: http://www.jierenchen.com/2009/10/memory-and-evolution.html
"Why is that?"
Complex configurations have more entropy than simple configurations. Life causes entropy to increase more quickly. You can think of living beings as agents of entropy if you like, but that implies that entropy acts with purpose, which it doesn't.
I remember having an argument at college with someone who argued that the laws of thermodynamics conflicted with the theory of evolution. So asked him what entropy was. He gave a bunch of different layman's definitions, such as comparing it "disorder" or "chaos". I wrote down for him, "S = k log N" and told him that if he doesn't know Boltzmann's equation, he has no business lecturing me on thermodynamics. This person also said that there was no natural process that created information. I wanted to tell him that the laws of thermodynamics, combined with Boltzmann's equation, and used in combination with Shannon's definition of information, imply that the amount of information in a closed system must always increase. Interpreted physically, mutation increases the entropy/information in genetic code.
If you want to know more about entropy, talk to a chemist. Chemists have to know entropy or they can't do their jobs.
As for the linked blog post, the idea of applying theories from evolutionary biology to the study of information is usually attributed to Richard Dawkins in his 1976 book, "The Selfish Gene". This is the book that coined the term "meme", and I recommend it.
Complex configurations have more entropy than simple configurations. Life causes entropy to increase more quickly. You can think of living beings as agents of entropy if you like, but that implies that entropy acts with purpose, which it doesn't.
I remember having an argument at college with someone who argued that the laws of thermodynamics conflicted with the theory of evolution. So asked him what entropy was. He gave a bunch of different layman's definitions, such as comparing it "disorder" or "chaos". I wrote down for him, "S = k log N" and told him that if he doesn't know Boltzmann's equation, he has no business lecturing me on thermodynamics. This person also said that there was no natural process that created information. I wanted to tell him that the laws of thermodynamics, combined with Boltzmann's equation, and used in combination with Shannon's definition of information, imply that the amount of information in a closed system must always increase. Interpreted physically, mutation increases the entropy/information in genetic code.
If you want to know more about entropy, talk to a chemist. Chemists have to know entropy or they can't do their jobs.
As for the linked blog post, the idea of applying theories from evolutionary biology to the study of information is usually attributed to Richard Dawkins in his 1976 book, "The Selfish Gene". This is the book that coined the term "meme", and I recommend it.
In this vein, the integrated information theory of consciousness is very interesting.
I have little time right now, so I'll let the papers speak for themselves. The abstract of the first paper follows. The second one goes deeper in the same theory. The most interesting part of the theory is that it is quantitative and testable (see the last two papers). The full text of all papers should be freely available.
----
## An information integration theory of consciousness ##
Giulio Tononi
Background
Consciousness poses two main problems. The first is understanding the conditions that determine to what extent a system has conscious experience. For instance, why is our consciousness generated by certain parts of our brain, such as the thalamocortical system, and not by other parts, such as the cerebellum? And why are we conscious during wakefulness and much less so during dreamless sleep? The second problem is understanding the conditions that determine what kind of consciousness a system has. For example, why do specific parts of the brain contribute specific qualities to our conscious experience, such as vision and audition?
Presentation of the hypothesis
This paper presents a theory about what consciousness is and how it can be measured. According to the theory, consciousness corresponds to the capacity of a system to integrate information. This claim is motivated by two key phenomenological properties of consciousness: differentiation – the availability of a very large number of conscious experiences; and integration – the unity of each such experience. The theory states that the quantity of consciousness available to a system can be measured as the Φ value of a complex of elements. Φ is the amount of causally effective information that can be integrated across the informational weakest link of a subset of elements. A complex is a subset of elements with Φ>0 that is not part of a subset of higher Φ. The theory also claims that the quality of consciousness is determined by the informational relationships among the elements of a complex, which are specified by the values of effective information among them. Finally, each particular conscious experience is specified by the value, at any given time, of the variables mediating informational interactions among the elements of a complex.
Testing the hypothesis
The information integration theory accounts, in a principled manner, for several neurobiological observations concerning consciousness. As shown here, these include the association of consciousness with certain neural systems rather than with others; the fact that neural processes underlying consciousness can influence or be influenced by neural processes that remain unconscious; the reduction of consciousness during dreamless sleep and generalized seizures; and the time requirements on neural interactions that support consciousness.
Implications of the hypothesis
The theory entails that consciousness is a fundamental quantity, that it is graded, that it is present in infants and animals, and that it should be possible to build conscious artifacts.
----
[1] http://www.biomedcentral.com/1471-2202/5/42/
[2] http://www.biolbull.org/cgi/content/full/215/3/216?view=long...
[3] http://www.coma.ulg.ac.be/papers/vs/massimini_PBR_coma_scien...
[4] http://www.coma.ulg.ac.be/papers/vs/boly_PBR_coma_science_20...
I have little time right now, so I'll let the papers speak for themselves. The abstract of the first paper follows. The second one goes deeper in the same theory. The most interesting part of the theory is that it is quantitative and testable (see the last two papers). The full text of all papers should be freely available.
----
## An information integration theory of consciousness ##
Giulio Tononi
Background
Consciousness poses two main problems. The first is understanding the conditions that determine to what extent a system has conscious experience. For instance, why is our consciousness generated by certain parts of our brain, such as the thalamocortical system, and not by other parts, such as the cerebellum? And why are we conscious during wakefulness and much less so during dreamless sleep? The second problem is understanding the conditions that determine what kind of consciousness a system has. For example, why do specific parts of the brain contribute specific qualities to our conscious experience, such as vision and audition?
Presentation of the hypothesis
This paper presents a theory about what consciousness is and how it can be measured. According to the theory, consciousness corresponds to the capacity of a system to integrate information. This claim is motivated by two key phenomenological properties of consciousness: differentiation – the availability of a very large number of conscious experiences; and integration – the unity of each such experience. The theory states that the quantity of consciousness available to a system can be measured as the Φ value of a complex of elements. Φ is the amount of causally effective information that can be integrated across the informational weakest link of a subset of elements. A complex is a subset of elements with Φ>0 that is not part of a subset of higher Φ. The theory also claims that the quality of consciousness is determined by the informational relationships among the elements of a complex, which are specified by the values of effective information among them. Finally, each particular conscious experience is specified by the value, at any given time, of the variables mediating informational interactions among the elements of a complex.
Testing the hypothesis
The information integration theory accounts, in a principled manner, for several neurobiological observations concerning consciousness. As shown here, these include the association of consciousness with certain neural systems rather than with others; the fact that neural processes underlying consciousness can influence or be influenced by neural processes that remain unconscious; the reduction of consciousness during dreamless sleep and generalized seizures; and the time requirements on neural interactions that support consciousness.
Implications of the hypothesis
The theory entails that consciousness is a fundamental quantity, that it is graded, that it is present in infants and animals, and that it should be possible to build conscious artifacts.
----
[1] http://www.biomedcentral.com/1471-2202/5/42/
[2] http://www.biolbull.org/cgi/content/full/215/3/216?view=long...
[3] http://www.coma.ulg.ac.be/papers/vs/massimini_PBR_coma_scien...
[4] http://www.coma.ulg.ac.be/papers/vs/boly_PBR_coma_science_20...
This article vastly overstates the role of closures. The connection isn't that deep - it is a simple consequence of life being a super-complex software executed in a biochemical machinery. Thus, you can find any programming construct the process of life, and argue that life wouldn't work without that.
There are lots of much more important things than closures. One example is the ability be self-referential on all levels (not to be confused with plain recursion). This blurs not only the line between code and data, but also between software and hardware. And this happens in a much deeper way than we're able to do now with things like FPGA/CPLD or hardware virtualization.
If you are really interested in that topic, I recommend the book "Gödel, Escher, Bach" by Douglas Hofstadter. It is written very well and should be especially easy to understand by programmers:
http://en.wikipedia.org/wiki/G%C3%B6del,_Escher,_Bach
There are lots of much more important things than closures. One example is the ability be self-referential on all levels (not to be confused with plain recursion). This blurs not only the line between code and data, but also between software and hardware. And this happens in a much deeper way than we're able to do now with things like FPGA/CPLD or hardware virtualization.
If you are really interested in that topic, I recommend the book "Gödel, Escher, Bach" by Douglas Hofstadter. It is written very well and should be especially easy to understand by programmers:
http://en.wikipedia.org/wiki/G%C3%B6del,_Escher,_Bach
Do we really need `real' self-references, if we have things like quines? Or do you think quines are part of the magic? (Like Goedel's theorems are?)
the idea of equating software to life itself is quite limited ...
life is frequency based, software can only ever approximate life by representing it numerically, binarily (??)
take sound, real sound is something that has a frequency spectrum, with harmonics ... fourier analysis allows us to take a snapshot of the wave, which approximates the wave, yet never represents it's full frequency spectrum - because it's impossible to do that ...
to boil life itself down to something which can be represented numerically (or even verbally for that matter) kinda misses the point ... The essence of life will always be absent from such an approximation ...
life is frequency based, software can only ever approximate life by representing it numerically, binarily (??)
take sound, real sound is something that has a frequency spectrum, with harmonics ... fourier analysis allows us to take a snapshot of the wave, which approximates the wave, yet never represents it's full frequency spectrum - because it's impossible to do that ...
to boil life itself down to something which can be represented numerically (or even verbally for that matter) kinda misses the point ... The essence of life will always be absent from such an approximation ...
Once upon a time there were some molecules. Bumping into each other, these molecules formed into larger molecules. Some of these molecules bumped into other molecules in such a way that the other molecules formed into copies of the first molecules.
I don't see how this has anything to do with "periodic functions". And somehow closures "introduce this idea of 'time'." Time is merely the arrow that points towards entropy.
Conversations like these are why I transfered out of a certain college.