Wasn't there a post here a while back that talked about how anyone positing a confidence of 0.999 on something non-trivial was most likely to be suffering from their own cognitive biases?

my somewhat admittedly sketchy reasoning:

I go to the University of Washington where there is considerable interest in the case. Of the people who have only been marginally involved in the case, most believe that Amanda Knox is innocent. Of the people who are interested in the case, many believe she is guilty. There's an obvious hometown effect here which biases towards innocence so I'm assuming those who look into the case are taking that into account when and still reach a guilty verdict.

Therefore, I assign a 70% probability to Amanda Knox being guilty (+ or - 30%).

When you encounter a road block, you don't need to give up. You can simply emulate each other's intuitions and proceed with as a provisional argument (assuming your world view is true...).

As far as rationality is concerned, it's achieving the place of what I call "rational ignorance". An awareness of the limits of your rationality and how you overestimate how rational you are:

http://blog.bumblebeelabs.com/the-ego-dilemma/

In every debate I've heard of, the pro-evolution people believe that the evolution side soundly thrashed the creation side and the pro-creation people believe that the creation side thrashed the evolution side.

This subjectivity over even who won makes debates eminently pointless for convincing anyone of anything.

"All models are wrong, some are useful" - George Box

Where's the bias?

• If something occurred once after long time while it could it seems unlikely.
• If something occurred soon after prerequisites were met it seems likely.
• If something occurred multiple times independently it seems likely.

1 and 3 seem obviously true. There are multiple trials separated either by geography or time, and they have enough failures / successes to make our intuitions right. Anthropic principle doesn't get involved here in any way. If agriculture was invented 5 times independently it couldn't have possibly be the limiting unlikely step.

2 might be luck - something might have been extremely unlikely but just have happened (by anthropic principle). But anthropic principle doesn't really give any reasons why it should have happened quickly. Of course it's extremely naive to consider (like Robin's paper) time as a series of independent trials - maybe it was unlikely as in prerequisites were just in place, and it was either fast or never. That's why I seriously doubt physics-inspired modeling of such events.

Our observations are biased because anything that occurs multiple times is very easy to see but something that occurs only once could be completely missed as an essential step towards civilization because we assume it was inevitable.

There is no rational argument against quantum suicide and the truth of it easily tested. The longer you live without knowing about quantum suicide, the less optimal your life will turn out. At the same time, you cannot look to anyone else's success as social proof for you to do it, you have to be the first.