More Random Randomness

jfleck writes "According to the American Institute of Physics' Physics News Update, Kent State physicist James Gleeson has developed a technique for generating numbers approaching true randomness. His trick is to shine light through a liquid crystal, taking advantage of its turbulence and avoiding the inevitable risk of predictability in deterministic random number number algorithms."

Comments

Re:silly

[GMontag451]

There are two different meanings of the phrase "random number". One is the coin-flipping variety, where any number within the range is possible, but is only as likely as every other number in the range. AFAICT, this isn't the kind of random numbers the article is talking about.

The other kind of random number is a number that is entirely entropy. In other words, an uncompressible number. This type of number is extremely hard to generate, and by definition, cannot be generated by an algorithm shorter than the number because this would be a form of compression.

Re:silly

[GMontag451]

If your random number generator is incapable of generating a string of 256 0's in a row, there is something wrong with it.

That is true for the first kind of random number generator that I talked about, not for the second kind.

Maybe I can describe what I'm talking about better. There are two different types of random number generators. One kind is a generator that randomly produces numbers, which is what you are talking about. The other kind produces *random numbers*, which is something completely different. It is the difference between the generator being random, or the number itself being random no matter how it is generated.

Randomly produced numbers are equidistributed along the range of the generator, usually from 0 to 1. Random numbers are numbers that are impossible to compress. A number is random no matter what the next result is, but is only randomly produced with respect to the numbers produced before and after it with the same algorithm.