Some things to try with brain farms...

1. Connect the brains to virtual worlds and try to get them to learn, grow, and become intelligent. They may need participation of grown people in the virtual worlds to teach them.

2. Speed up the physics in the virtual world to see how this affects intelligence.

3. Try to make a brain intelligent enough in a virtual world to educate other new brains that are connected to the virtual world.

4. Put a new brain that is to be educated in a virtual world in a virtual world that is embedded in the virtual world of a brain that was educated in that virtual world. Increase the simulation speed (and/or vary other parameters) in each level of the hierarchy of virtual worlds. Oversight by brains whose access spans multiple levels of virtual worlds might be needed.

5. Connect new brains in different ways at different times in their development. For example, connect the output of one brain's visual cortex past the basic image processing functions to the input of the visual cortex of a new brain so that higher levels of abstaction may be formed. This could also be done in the code of a virtual world. Or, try to connect multiple brains to one agent in a virtual world.

6. Vary the physics of the virtual worlds. For example, make random things less random, make everything squishier, make physics non-local, or omit conservation of mass and energy.

7. Vary the basis of "reward". Make the reward mechanism be based on arrangement of symbols that are valid syntax of a programming language with increasing rewards for code that calculates elementary mathematical functions, generates random numbers, or has an output of high Kolmogorov complexity.

Some Things I'm Not Completely Sure About...

1.. P=NP is either a number theory problem (2-SAT and 3-SAT could be converted into encodings for different types of numbers?) or a physics problem (a question regarding the nature of spacetime).

2. Fractal-like systems and/or systems with very simple rules can resolve inconsistencies between quantum physics and the general theory of relativity and may be important for explaining the genesis of galaxies, Wolfram style.

3. Computing that is more "reconfigurable" (like FPGAs, evolving processor architectures, programming for many-core systems) in increasingly higher levels of abstraction will bring big results in machine intelligence.

4. The first three items are related.

http://andyslife.org/games/sand.php

Evolution of Evolution
Molecules were some of the first things to use evolution to do what it does. Random permutations of interactions between atoms found stable molecules and they stuck around because they could. By this same process, molecules evolved that could group together and be more stable than they were on their own. By this same process, molecules evolved that could help make stable ones and form complex structures and mixtures that could make more of itself, which led to first living stuff. At each step in the process, a higher level abstraction formed, and this seems to be the theme of evolution in the universe. Our brains evolved in an analogous way. At first they formed low-level abstractions using sensory information that controlled basic behavior in a somewhat direct manner. The brain probably came about because it could support a wide variety of these low-level abstractions using one medium: neurons. Human brains still have this low-level machinery going on, and it's things like edge detection, basic shape detection, automatic pupil dilation, and so on. We now abstract these thing at a much higher level to invariant classification of objects, invariant classification of actions and the many other classes in our language, the self, and all of these things, for the most part were not evolved in the low-level machinery of the brain but instead evolved in the domain of society. A society spends years preparing a child's brain to be a part of it, and now society is in the process of evolving other mechanisms by which these abstractions can be built by way of overlapping work in biology, chemistry, neuroscience, psychology, physics, electrical engineering, materials science, and computer science. Both separately and in combination with our biologically evolved abstractions the evolution of these abstractions are also happening in the machine. At each step in the process, a higher level abstraction forms, but it seems particularly interesting from our point of view now because it seems that from the level of abstraction on which we operate and create new abstractions, we are continually aware of all the lower levels and the higher levels we create.

Justin Curry pointed me to an interesting book by R. Buckminster Fuller called Operating Manual for Spaceship Earth, which is a sort of guide for how we might want to direct our evolving efforts. Buckminster suggests that much of what has evolved in society and education makes us too specialized and gives a convincing account involving kings, pirates, big strong dudes, and trade of how this came to be. He suggests that specialized education came about as a means for control and for the few to maintain rule over the masses, and this extends to our whole education system now.

Humans that have a nose that works can smell fear, and Jupiter ate a bunch of moons it used to have.

readysetgo
You're right, our friends love sex. A random stranger might place a random group of us in the counter-culture/hipster/alternative category, and more often than not, that would be mostly accurate for many intents and purposes. (For all intensive purposes!) I think one of the eight reasons for this is that many of us sort of did as the Romans were doing in the parts of MIT's campus we ended up. A similar thing happened with most of those "Romans" when they first arrived. Among the living groups on the eastern side of campus, interest in the history and lore varies. I think the living groups' levels of interest in their history have remained fairly constant. As for my closest friends, which includes a very small subset of some sort of reasonably large counter-culture portion of MIT: not very many of us are careful. We're not cautious. It's not that we "ain't skeered"; most of us are probably super afraid of one thing or another. Here's a trubie: we are aware of the fact that a decent portion of what is widely considered dangerous is safer than a good portion of what is widely considered safe or normal. Besides those "built-in" fears (snakes, heights, moths), we probably have somewhat bizarre fears. But I could be wrong, especially about that. If we could come to a consensus on the ideal society, I think at first it would be too different for some people, too sinful for some people, too ambitious for some, too arbitrary for others, not structured enough for some folks, simultaneously too primitive and modern for some people, but I am very confident that an insane amount of people would be happy and come to appreciate its design. My friends are meticulous in curious directions. But we'll never come to a consensus. We won't even try. Can societies be designed? Are they designed? Do the protozoans think they designed us? I really do wish that there was something that all of us agreed requires all of our focused efforts for some extended period of time. About those Romans...I think they've been on to something for many decades. Will it ever take hold? I don't know. I know I'm an optimist, but I actually feel quite optimistic about it. It's a real competition, these idea battles that seem to be spiraling out of control, spewing pointless wars, objective opinions, and contextual truths. It looks a lot like the evolution of the universe:

BIG BANG! -> pure, uniform energy -> PARTICLES BANG TO MAKE ATOMS! -> atoms keep to themselves -> ATOMS BANG TO MAKE MOLECULES/PLANETS/STARS/BLACKHOLES -> everyone spins around

I could be wrong, but it's probably one of those things that isn't right or wrong in any reasonable context, like, "God is watching us." or "All men are created equal." or "Machines will never achieve true intelligence."

I see something sparkly. What's that noise? What's inside it? What is communicating with it? What is communicating with what it's communicating with?
<./digressivedigression>

From the Mathematica Audio Package Documentation on Frequency Modulation:
Frequency modulation (FM) is a technique used at major universities on workstations and in suburban homes on synthesizers.

Some evolutionary classes of conjectures:
1. Those which we can turn into theorems.
2. Those which we can prove undecidable or independent of the axiomatic system in question.
3. Those which we can prove decidable but can only approach theorem status.
4. Those which we can only approach proving decidable (or equivalently, undecidable).

I think this ad is funny, and I see it everywhere. They must be spending/making a lot of money. It looks like a stretch mark changed colors.