This is epigenetic. That is, the genes aren't being altered. What is being altered is the pattern of which genes are on and off--which proteins are being expressed. What's impressive here is that the change is so big and capable of reversals.

You seem to be saying that computers get Go, but there's this little problem that prevents them from doing well. I'm sorry to inform you, but "know[ing] which moves are most advantageous in the long run" is the entirety of understanding a game.

The only way I can see computers as basically "getting" Go is that Go is a turn-taking, deterministic game of perfect information, so in theory, all you have to do is solve the game tree and tada, winning strategy. Wikipedia gives the "game tree complexity" of Go (number of nodes in the tree?) as being on the order of 10^360. With maybe 10^80 fundamental particles in the known universe (rough estimate, sourced from Wikipedia also), I don't think Moore's law is gonna help a lot here.

(Sorry to jump into the action so late, but I greatly sympathize w/the "Computer Science has figured a lot of your shit out in practice, Philosophers" idea.)

Here, for you, is a concrete example:

I was taking a data mining class. My friend was taking a religion/philosophy kind of a class, don't remember the topic. In one particular week, I was studying the original Google algorithm (PageRank), which, if you're unfamiliar w/it, works like this (Though there are other equivalent ways to describe it.):

• Each page has its own quantity, called 'rank'.
• Pages 'vote' for each other by linking.
• If a page has n links and rank r, then each of its votes has a value of r/n.
• A page's rank is the sum of the votes for it.
• Note that this problem is circular, in that the value of one's votes is determined by ones rank, which is determined by the values of the votes for you, which is determined by the rank of those who voted for you, which is....
• Do math. Solve the problem.

In my friend's class, on the other hand, they were discussing nominalism, and, from the sounds of it, they were... going in circles. Of course I only have the 2nd hand version of it, but I heard a thought experiment along the lines of:

• What does a word mean?
• Well, look it up in the dictionary.
• What do those words mean?
• Look them up in the dictionary.
• ZOMG IT GOES IN CIRCLES. HOW DEEP IS THAT?!

My friend's mind was totally boggled that in my class we had taken a similar (but much more concrete problem), and by formalizing it and applying linear algebra, we got rid of the circularity and came up with a useful answer.

So, there you go, a concrete example. Here are some caveats to head off certain responses:

• Finding the rank of a page was more concrete and amenable to formalization than finding the meaning of a word. Yes, I noticed
• My friend could've just had a crappy teacher or been missing something important that he was supposed to be learning. Who knows, maybe they had way more insightful thought experiments on the topic.
• I don't know if this is an "open problem" in philosophy, but clearly the teacher felt it was still relevant.

So, in conclusion, I think the point to be made is not that philosophy is worthless. I think what ShakaUVM was getting at (certainly the point I would be trying to make) is that people should not sit on their asses marveling at the intractability of an abstract problem when other people are solving concrete cases of that problem IRL. Because at that point, the only thing preventing you from resolving your argument is an unwillingness to get down to brass tacks. Bitches.

This is actually a common misconception about Chinese, and it has obscured a lot of debate about writing systems. The truth is that all forms of written language are based on sounds, and Chinese is no exception (although it's a much bigger pain in the ass than an alphabetic system). I highly recommend "Visible Speech" by John DeFrancis which propounds and defends this thesis.

There already exists an online meeting place between (angry teenage) people in the US and (angry teenage) people in the Middle East. It's called CounterStrike--The main difference is that instead of discussing politics, you act them out by shooting one another.

(For anyone who has played the HL2 mod Insurgency, you know what I mean. The English/Arabic ratio of players is about 2:1, and everyone takes turns yelling Allahu Akbar! and shooting each other with rocket launchers and M-16s and whatnot. Kind of fucked up, when you think about it, but it's also loads of fun.)

I agree, Synthesis was a much bigger achievement with a much grander vision.

The article is interesting, but Schneier is not the first person to consider such questions. Last year (I think?), Ron Rivest gave a couple talks at my school on the subject of voting. One of them was about auditing, and the other was about using crypto to achieve safer e-voting. You can see something similar to what he said here: http://people.csail.mit.edu/rivest/RivestSmith-ThreeVotingProtocolsThreeBallotVAVAndTwin.pdf Some of the comments here have been arguing over the relative merits of verifiability and secrecy (as in having voting receipts or whatever). Cryptographic methods can be used to partly reconcile those ostensibly contradictory goals. Anyhows, have fun reading.