4 ms·

What? Your theorem doesn't even match with the basic prisoner's dilemma. True, in the classic version of this game there's no communication... but the point is

What? Your theorem doesn't even match with the basic prisoner's dilemma. True, in the classic version of this game there's no communication... but the point is the same - just because an outcome is Pareto efficient doesn't mean that it's a stable equilibrium - people's rational self interest can result in worse outcomes. I don't know what kind of economics you're talking about. Most of the absurd simplifying assumptions really have to do with the idea that people have perfect foresight and their statistical guesses are unbiased - and that we perfectly understand variance/risk and all that jazz. It doesn't have as much to do with how we communicate with one another, rather that we are super awesome flawless statistical machines when in reality we all suck at statistics (have all kinds of biases and heuristics to rely on) in every day life. And just because information is cheaper doesn't mean it's free. At the very least it costs us time - at least in the current form of the internet it's not like there's simultaneous, effortless, instant communication between every human being on Earth. Now that might change things. But I don't think you fully made the case that the internet is fundamentally different than a "better" or more effective printing press. My point, I guess, is that maybe we approach that zero cost world, but will never reach it. And that there's a fundamental distinction between being at zero cost and being at any positive cost... so that the internet is more like a printing press rather than a zero cost communication mechanism.

ewzimm 13y ago
Users are the most important part of any information processing system, and any Pareto-optimal communication configuration requires that perfection extend to the users and their decision-making process. You could model this with the Prisoner's Dilemma by playing it this way: in your system of perfect communication, both players can know each other's decision before they commit and both can respond to any change in the other player's position. This would lead to always choosing the mutually beneficial position. Any change to the rules would make at least one individual worse off because they would lose information. This situation requires that both players give up the freedom to conceal their decision from the other. Privacy is at odds with optimal communication. In the move toward perfect communication, there will be changes that damage the positions and advantages of people involved in the system. It's clear that existing power differences are enhanced by rapid growth. Although perfection might not be a reachable goal, I believe there's evidence that approaching it will lead to a massive long-term benefit for everyone involved. There's no other way to overcome the mistakes we all make by being constantly misinformed.
vbuterin 13y ago
> What? Your theorem doesn't even match with the basic prisoner's dilemma. Yes it does. > True, in the classic version of this game there's no communication That's why. The infinitely iterated prisoner's dillema the theorem works with just fine because there's perfect communication (or rather, the imperfection in communication drops to 1/∞), any finite prisoner's dilemma, on the other hand, relies on both parties making their decisions in secret. > I don't know what kind of economics you're talking about. I would recommend anything by Ronald Coase or David Friedman. > And just because information is cheaper doesn't mean it's free. Correct. We will never have a perfect economy of any sort. We can only get better. But that doesn't change the fact that the internet is vastly better than anything we've had before in a number of very significant ways.

ewzimm 13y ago

Users are the most important part of any information processing system, and any Pareto-optimal communication configuration requires that perfection extend to the users and their decision-making process. You could model this with the Prisoner's Dilemma by playing it this way: in your system of perfect communication, both players can know each other's decision before they commit and both can respond to any change in the other player's position. This would lead to always choosing the mutually beneficial position. Any change to the rules would make at least one individual worse off because they would lose information. This situation requires that both players give up the freedom to conceal their decision from the other. Privacy is at odds with optimal communication. In the move toward perfect communication, there will be changes that damage the positions and advantages of people involved in the system. It's clear that existing power differences are enhanced by rapid growth. Although perfection might not be a reachable goal, I believe there's evidence that approaching it will lead to a massive long-term benefit for everyone involved. There's no other way to overcome the mistakes we all make by being constantly misinformed.

vbuterin 13y ago

> What? Your theorem doesn't even match with the basic prisoner's dilemma. Yes it does. > True, in the classic version of this game there's no communication That's why. The infinitely iterated prisoner's dillema the theorem works with just fine because there's perfect communication (or rather, the imperfection in communication drops to 1/∞), any finite prisoner's dilemma, on the other hand, relies on both parties making their decisions in secret. > I don't know what kind of economics you're talking about. I would recommend anything by Ronald Coase or David Friedman. > And just because information is cheaper doesn't mean it's free. Correct. We will never have a perfect economy of any sort. We can only get better. But that doesn't change the fact that the internet is vastly better than anything we've had before in a number of very significant ways.