I believe true randomness can only be man made, and only in the context of the situation where you are trying to produce random results.

So then you agree that a computer program generates what some call "pseudo randomness" is no less random than a man flipping a coin?

The main point of the question is to compare the capacity for randomness from both examples. Either they are both random, or neither of them is random, but however you wanna define, the computer isn't less capable of random generation.

yep



They both have the ability to produce results that are essentially impossible to predict. Neither seems to be truly random though: we know exactly what the input was for the computer, and can show that it is in fact 100% deterministic. But this is only because computers are such simple and binary systems. We can only suspect the outcome of the flip coin isn't random either, since we can never know the exact input used; and therefore can never reproduce the process to insure we get the same result each time. If you could reproduce the input for the coin flip perfectly, we would know for sure.

This post was edited by Azrad on May 18 2013 06:29am

Hidden variables are not enough to save you from randomness, even with them there must be truly random events.

a coin flip is a chaotic event. Look up chaos theory.

there could be a difference in the degree of perceived randomness, but in practice, algorithm based random generators for the most part are complex enough to be perceived as random(-like).

Yes, I'll agree to that. The input can be modeled by an algorithm

This post was edited by saber_x3 on May 18 2013 02:07pm

The problem comes when you try to apply a more strict definition of random; For an event to be truly random, the probability of all the possible outcomes would have to be equal. In a deterministic universe (and much of what we see in our universe very much appears deterministic, even if it isn't "underneath"), there is only ever one possible outcome to an event. So randomness defined in this way doesn't even make sense. For all practical purposes, yes, computers can produce random numbers within a given range. Numbers always work in a predefined way and no number can be truly random ever, due to the nature of numbers and the numeral system. Numbers are just ideas, there is nothing to stop them from going on forever. Though at some point we wouldn't be able to count that high, mind or machine; as they are limited by their capacities

This post was edited by NatureNames on Jun 5 2013 06:59pm

the one who finds a computer algorithm that generates a random sequence that doesn't repeat after a certain amount of bits will be rich.
(it's also not possible due to simple logical rules regarding state machines.)

im pretty sure godels theorem prohibits truely random computer algorithms. could be wrong tho.

This is an interesting question, because one has to wonder if the basis of our reality outside of virtual reality is based on on/off, this way/that way, true/false with no inbetween. If that is the case, then yes, computers can POSSIBLY generate random numbers. The key to this would be fully understanding reality outside of virtual reality....but until that point is reached, we cannot generate true random numbers. "Random" numbers as we calculate them now are based on static seeds, or seeds that are based on some other static formula with static inputs.

If physics in nature is fundamentally more complicated than 2 values as is binary in computer science, then truly random numbers will be impossible unless we rebuild computer science from the ground up, which is not going to happen.

Bottom line is, we don't know enough yet to conclude with 100% confidence that this is possible with our current system and understanding.

I would also like to say that random is in essence a quality of which a pattern can not be derived. With any computer with any formula with any seed, in x iterations you can derive a pattern (and charted by a line that repeats its fluctuations at the same intervals), because the seed and the formula that uses the seed have some basis of pre-determination. That is the nature of having 1 and 0 as your fundamental building block of computer science. In nature, assuming the quality of random exists, your basis would be -∞ to ∞.

This post was edited by Asno on Jun 11 2013 06:54am