[jkeler]

> a probability for civilizations to form of one in 10 billion per planet was considered highly pessimistic

I think people underestimate how low probabilities can really be. According to wiki the shortest self-replicating RNA is 165-bases long. Even if you convert all observable universe in RNA this is still not enough for this RNA to form by chance (4^165 is much larger than number of atoms in observable universe).

[perlgeek]

> Even if you convert all observable universe in RNA this is still not enough for this RNA to form by chance (4^165 is much larger than number of atoms in observable universe).

Which kinda suggests there is either a mechanism that promotes that, or there are other, easier paths to get self replication. Otherwise we mostly likely wouldn't be here.

[jkeler]

For people who want an alternative view on this problem: http://biologydirect.biomedcentral.com/articles/10.1186/1745...

[tbrake]

Not if you only give it one chance. Those are incredibly grim odds just for life - nevermind civilization, which would be rarer still - somehow arising.

But aren't those chances constantly being created, on the scale of uncountable -illions of interactions of molecules that happen every second, across the billions of square and cubic kilometers of the surfaces and oceans of those billions of planets, across billions of years?

I mean, doesn't that improve the odds just a little bit?

[qaq]

Given enough time anything can form out of random fluctuations if it has >0 probability no matter how small.

[Houshalter]

"Enough time" can easily be longer than the age of the universe. A million monkeys typing on keyboards will eventually write shakespeare. But it will be longer than the universe before they even type a single sentence of it correctly.

[titzer]

You are postulating the existence of only a single universe, ours. But there could be any number of universes, and so far we have no scientific way of testing that. The anthropic principle and the possibility of multiple universes puts all probabilistic arguments to dust: anything with non-zero probability happens in an infinite number of universes.

> A million monkeys typing on keyboards will eventually write shakespeare. But it will be longer than the universe before they even type a single sentence of it correctly.

All it takes is _one_ tiny selection force to blow this argument out of the water. For example, if only _a single_ correct letter is retained from each attempt, the problem gets exponentially easier with every iteration, making the expected time to generate any work linear in its length.

[Houshalter]

I agree with you completely about the anthropic principle. But we are talking about the probability of aliens. If the anthropic principle is true, then we should expect aliens to be really really unlikely.

[qaq]

It sure can be, but as long as it is less than the expected lifetime of the universe it will still happen. A million monkeys - given that Hamlet contains such long sentences as "He." it will take far less then a minute for them to type a single sentence correctly.