This also has the advantage of needing no new code.

Nonjudgemental people may help you socially with trying new ideas, but will not help you epistemically with finding the correct ones. You will have to reinvent every wheel alone. If you have unlimited time, go ahead. Otherwise, it is better to find people who are a bit judgemental -- who have a preference for correct beliefs.

Of course, replacing judgemental people with a preference for incorrect beliefs by nonjudgemental people is an improvement. But sometimes you can do much better than this.

Maybe this is what happens to many people: they replace the judgemental people with incorrect beliefs by nonjudgemental people, they realize the improvement... but then they can't improve further because their heuristics says that "nonjudgemental is best". Noise can be better than active misinformation, but signal can be even better than noise. But when most of your experience is with active misinformation, all signals seem dangerous.

I always thought that there are bits of useful insight scattered among all the nonsense of self-help books and that reading lots of them and then letting your mind sort through it could lead to those bits being assembled into a useful whole. But I was never able to get myself to actually do it (is there a self-help book on reading self-help books?)

The problem is that I can't really tolerate the idea that I'm doing something mostly useless in the hope that some benefits will nevertheless accrue, invisibly, in my subconscious. I feel the need to judge whatever I'm reading and either seriously engage with it or discard it. And self-help books are not only full of crap, but are also designed to push all your enthusiasm buttons. For me, it's hard to read through bazillions of 'simple insights that will change my life' without slipping into excessive enthusiasm or outright contempt.

So, I think, do it, if you can tolerate the tedium.

This sounds like a great idea.

But, how will you know if you managed to replicate EURISKO's success? Are people still playing this? Or is there some archive of tournament entries available so you could see how your program's fleet compares against them?

I really enjoyed this article, and I can see how many of my own behaviors map onto this cue-routine-reward structure. I've been wanting to read this book, but now I don't feel that I need to.

I would appreciate it if another reader could try to explain how rumination (focused attention on the symptoms of one's distress, and on its possible causes and consequences, as opposed to its solutions) would fit into this framework. Here's my attempt:

Let's say you were fired from a job you liked, and you ruminate on the loss of the job.

The cue: seeing a former co-worker The routine: think of everything that went wrong on the job leading to your firing, the possible changes you could have made but didn't, feeling stupid The reward: to discipline yourself?

I feel that the cue and routine are well described, but I'm not sure about how the reward.

Nice.

Bug reports:

Note that even though t is 1, we'll end up with a unit error if we don't carry it around.

I think you meant r here.

And you have flipped the sign of \Delta U_{linear}.

Question: when people say that two weeks on dry ice is suicide, is the 'two weeks' part significant? Are two weeks on dry ice worse than two days and better than two months on dry ice or is it that the moment you get frozen without cryoprotectant, you're done?

Now we just need to figure out how to make people more competitive in the areas they care about...

I'm not sure that replacing intrinsic motivation with the motivation to prove yourself better than others is always a good idea. It will help some but it might also hinder.

If I'm feeling competitive, why would I ever skip the chores and lose?

I could skip the chores and lose because I'm feeling the wrong sort of competitiveness, I tied my feelings of self-worth to the outcome and avoiding thinking of the issue altogether is preferable to facing the prospect of loss.

And then when you join, it will display a message to all your contacts that you are "following their answers", of course without telling you anything.