Re:

Dear Jeffrey,

thanks for the Skinner article. I did not know that he was aware in 1981 that computers can simulate evolutionary processes. While I think that it is terrible how Skinner devastated the field of psychology, that is probably not his own fault but those of stupid acolytes that were unable to step out of the shadow of the giant. (I think that happens quite often: Maturana's followers did cybernetics in, Luhmann's followers killed systemic sociology. Artificial Intelligence was lucky that Minsky, who apparently has a similarly strong personality, was constantly on the move, so nobody could stay in his shadow for long.) So Skinner heralded evolutionary psychology, hm?

Today, it seems to be fashionable again to scoff at evolutionary arguments in psychology, usually on the grounds that they are working post-hoc instead of predictive. (At least for the Axelrod simulations, that certainly does not apply.) In Al, evolution is mainly conceived of as a method for constructive problem solving: a blind search, and very few people, like Aaron Sloman, have been clamoring for looking at evolutionary trajectories when trying to understand intelligence.

By the way, I have looked at Bill Gates' AMA on Reddit today. http://www.reddit.com/r/IAmA/comments/18bhme/im_bill_gates_cochair_of_the_bill_melinda_gates/? limit=500

I am still thinking about what I would want to pitch to him. Naturally, I would go for AI, because I am invested into the idea that understanding the nature of the mind is the most fascinating puzzle there is. But if I look at his current mindset and guess what could genuinely interest him, I would go for large scale economic simulation. I think that the world needs a joint effort on building the economic equivalent of global climate models. Next to understanding finance and economy, and experimenting with the results of short-term and long-term interventions on such models, we might use them to raise the economic understanding of our political class.

I think that one of the reasons that the nordic nations do so well has to do with their small population size and relatively high integration between milieus. This enables short and and at least somewhat efficient feedback loops for law-making, perhaps most visible in the way Iceland has handled the banking crisis. (Of course, the relative abundance of natural resources per capita helps.) Being able to quantify, predict and understand the economic and social impact of regulations might go a long way towards improving feedback loops in other societies, too.

Cheers,
Joscha

I would like you to skim roger shanks book on script and goal s for ai. You roger and I will Skype