anotherblackhat comments on Welcome to Less Wrong! (2012) - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (1430)
O.k. suppose It's false. Rather than wasting time disproving the CRA, I simply act on my "false" belief and reject it out of hand. Since the CRA is invalid for many other reasons as well, I'm still right. Win.
Generalizing; Say I have an approximation that usually gives me the right answer, but on rare occasion gives a wrong one. If I work through a much more complicated method, I can arrive at the correct answer. I believe the approximation is correct. As long as;
effort involved in complicated method > cost of being wrong
I'm better off not using it. If I knew the truth, then I could still use the approximation, but I now have an extra step in my thinking. Instead of;
1. Approximate.
2. Reject.
it's
1. Approximate.
2. Ignore possibility of being wrong.
3. Reject.
Ah, I see what you mean. Sure, agreed: as long as the false beliefs I arrive at using method A, which I would have avoided using method B, cost me less to hold than the additional costs of B, I do better with method A despite holding more false beliefs. And, sure, if the majority of false-belief-generating methods have this property, then it follows that I do well to adopt false-belief-generating methods as a matter of policy.
I don't think that's true of the world, but I also don't think I can convince you of that if your experience of the world hasn't already done so.
I'm reminded of a girl I dated in college who had a favorite card trick: she would ask someone to pick a card, then say "Is your card the King of Clubs?" She was usually wrong, of course, but she figured that when she was right it would be really impressive.