ThrustVectoring comments on 2013 Less Wrong Census/Survey - Less Wrong

78 Post author: Yvain 22 November 2013 09:26AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (616)

You are viewing a single comment's thread. Show more comments above.

Comment author: ThrustVectoring 22 November 2013 03:43:57PM *  3 points [-]

Cooperating for reasons other than "I expect cooperating to make other people cooperate" gives people a reason to defect and make the total (and your expected) reward lower.

I've done the math elsewhere in this thread, and if at least a third of all respondents decide to cooperate no matter what, the optimal solution is to just defect and take their money.

Comment author: Tyrrell_McAllister 22 November 2013 03:57:07PM *  7 points [-]

Cooperating for reasons other than "I expect cooperating to make other people cooperate" gives people a reason to defect and make the total (and your expected) reward lower.

Yes. And I did cooperate because I expected that it would make other people cooperate (acausally). I was explaining why I wanted more people to cooperate, even though it would mean that Yvain would lose more money.

I've done the math elsewhere in this thread, and if at least a third of all respondents decide to cooperate no matter what, the optimal solution is to just defect and take their money.

Good. Then a defector has been enticed to take the survey.