Adam has posted about the nonsense that is the new politically motivated holiday in honor of Reagan, and about how he doesn't like Reagan in general.
I admit that I don't know enough detail of Reagan's history to validate everything he says (though given his propensity for facts, I have faith he is ludicrously accurate, if not somewhat biased). However, it is refreshing to see someone applying some solid analysis to the holiday-on-a-week's-notice mania that has overtaken this country.
Let's hope that when Adam leaves MIT, he'll re-incarnate his blog with some software that lets people post comments on his site directly, so people like me can praise him more directly.
As I understand it, when Darwin proposed the theory of evolution, it appeared to be at odds with the second law of thermodynamics. The second law essentially says that "the entropy of the universe increases during any spontaneous process". Essentially, the notion that organisms evolve and therefore get more organized seems to contradict the notion that particles, when left to their own devices, tend towards less organization. I know creationists who have argued that the second law of thermodynamics proves that the theory of evolution is false.
The most satisfying answer I have come across for this connundrum is the idea that organisms are open systems. Why is this important? It turns out that a lot of physics (correct me if I'm wrong) relies on the theoretical construct of a "closed system". This doesn't necessarily mean that whatever you are studying needs to be locked in some kind of a box, but it does mean that you have to draw some boundaries around that which you are studying in order for your equations to work. Within the construct of a closed system, the second law of thermodynamics makes sense. Once there is a finite amount of energy inside a system, and you have chemical processes that require energy to function, then it is only a matter of time until those chemical processes consume all the available energy and eventually cease. If those chemical processes were reducing entropy as a side effect of functioning, then after they stop functioning, entropy will again increase. If those chemical processes were your body, and you were locked in a room with no air and no food, you would die, decompose, and the particles that make up your body would become less organized.
However, let's take the notion of an open system. An open system differs from a closed system because its boundaries are not fixed. Unlike the closed system, we might say that energy is infinite. Particles may leave the system, particles may enter the system. Consider the particles in your body as an open system. In order to maintain their organization and low entropy, we must consume food and extract energy from it. We must breathe oxygen in order to aid in the extraction of energy from our food. We must also eliminate wastes from our bodies. As far as our bodies are concerned, it is a system that constantly has energy flowing through it, brought in from the outside. This constant flow of particles and of energy into and out of our bodies makes us seem a lot more like an open system than a closed system.
So how does this resolve the conflict between evolution and the second law of thermodynamics? It seems that trying to apply the second law to the particles in one's body is to imply that our bodies are closed systems rather than open systems. The systems that evolution has produced are not closed at all. As a pre-requisite of their existance, organisms have to be capable of constantly extracting energy from their environments. Fortunately, life on earth has had the advantage of the sun to provide ubiquitous amounts of energy. Organisms that don't get their energy directly from the sun feed off organisms that do, or organisims that feed off organisms that do. The point is that both ideas, evolution and the second law of thermodynamics are not in conflict if you understand the difference between closed systems and open systems that have access to infinite energy supplies.
The thing about open systems that is really interesting is this: how do we understand open systems? By their definition their boundaries are fuzzy; what is inside them and what is outside them is not fixed. Its the sort of thing that makes a physicist cringe. The abstract nature of the open system is likely one reason why the idea is poorly understood.
It appears that the best we can do is to find ways of describing open systems in terms of their organization: how particles relate to one another while they are part of the system. We might say, then, that when oxygen is being used by my blood cells, that it is part of the system that is my body, but not before.
If this sort of explanation seems unsatisfying, it should. One of the reasons we are likely to have difficulties in this kind of field is because by nature, humans want to draw boundaries around things in order to understand them: that's how reductionism works. However, what we may be approaching in science is a limit of reductionism as we know it.
Understanding open systems and arguing that they are important to understand is the subject of a book called Life Itself by Robert Rosen. If you find this sort of thing interesting, I highly recommend it.
I am preoccupied with the idea of strong A.I. This is the notion that it is possible to construct a machine which is capable of thinking the way a human can. Some call it "true artificial intelligence".
There are a myriad of things that are interesting about this topic to me, some of which are if it is possible to create, why people think it is possible or not, how the academic community may or may not have the right approach to acheiving it, the impact of a more captialisitc research community on this pursuit, etc. I may touch on those topics later.
For now, I wanted to make a case about why it would be a worthy goal to accomplish if we could.
The first reason ties into a general theme of human accomplishment that we have seen throughout history. Humans have a desire to create, and in particular to create things in their own image. Consider some great works of antiquity: Michaelangelo's David, Da Vinci's Mona Lisa, etc. Creating a machine that could think the way we do would be a further monument to human ingenuity and mastery of our environment.
The second reason focuses on what we would learn from such a creation. Assuming that such an accomplishment could only arise from a complete understanding of cognition, we would potentially have at our fingertips a mirror into the human experience unlike any ever before. Not only would we have greater insight into why people act the way they do, we would also have greater insight into why humanity looks at the world and universe the way we do. Providing greater transparency into the foundations of thought processes would undoubtably reveal how much of the world we experience is a product of our minds. Many different landscapes of thought, philosophy and religion, would need to be revised. Our ideas of eduction would likely change and improve. Our understandings of motiviations and actions would improve, potentially leading to revisions in political thought. It is difficult to think of a field of study that would not somehow be impacted.
The third reason is the practical utility of intelligent entities. Artificially intelligent entities could do work we could not do or do not want to. This is one of the more controversial benefits of strong A.I. because of the many depictions in popular media of artificially intelligent entites running amok. I tend to agree that it would be very problematic to turn over control of any societal functions to artificially intelligent entities for a long time after perfecting them, simply because there would be a host of issues to work out first. Much caution would need to be used. It might be impossible to manage effectively. Nevertheless, as pessimistic as I sound, it is not too difficult to imagine the economic benefits that automation with truly intelligent entities would provide.