shule1

Nan, bower, and jmsieglaff,

This reminds me of how there are multiple kinds of random.

There's a kind which doesn't really seem to exist much, if at all, in nature (in the context I mean; it perhaps exists in other contexts). It's kind of like an even distribution with a randomized order. So, no number gets generated a disproportionate amount of times, and so basically all numbers eventually get generated. And you're very unlikely to ever see the same number 3+ times in a row (if it's even possible). I'm not sure why this methodology is considered more random than the next method, but it is. I can, however, see why it's popular (it makes fairness a lot easier, in many contexts), but to me, it's less random because it's more predictable (in that you can essentially rule out a number of outcomes); plus, it requires planning and memory (to know which things have already been generated).

Another method is basically the same one you get with a physical dice roll. It's technically possible to roll the same number an infinite number of times (were it possible to roll them infinite times), but the odds are pretty slim. So, with this method, you have a chance at getting the absolute best and worst outcomes. With the other method, you always get something in between, wherein you have more than a certain number of generations.