Quote from: o_e_l_e_o on Today at 07:55:50 AM

...which is why whenever the topic of manually generating entropy comes up, I always suggest von Neumann's coin flips to simply, quickly, and most importantly verifiably generate 128 or 256 bits of provably unbiased entropy.

I am not familiar with it, but I wouldn't trust myself with coin flips and I will tell you why. If I throw a pen in the air and tell myself I am going to catch it by the tip, guess what, more often than not, I manage to do that. That's not a random throw because I figured out how high to throw it, how many times it's going to spin before it lands in my hands, etc. Obviously, a coin is different, but the point is we aren't good sources of entropy if our subconscious tells us to manipulate the result to see if it works.

If my coin flip would land on tails twice, for instance, I can't trust myself not trying to land another tails just to see if I can. You might say you can do that with dice as well. Sure, but a die has 6 possible results, while a coin flip only has two. I hope you understand what I am trying to say. Does von Neumann's system account for that, and in what way?