What Randomness Really Means, and Why Computers Need Help
Published on
Random means the next value cannot be predicted from the previous ones. Rolling an honest die is random; counting one by one is not.
Computers are predictable machines: given the same instructions, they do the same thing. So almost everything a program calls a random number is actually pseudorandom. A formula turns a starting value into a sequence that looks random but is entirely determined by that starting value.
That starting value is the seed. If it is predictable, like the clock time, the entire sequence is too. This is why the quality of the seed matters more than the formula. Operating systems collect noise from hardware, such as timing jitter and device events, to produce unpredictable seeds. That is where the 32-byte seed of our draws comes from.
This is not a weakness, it is an advantage. With an unpredictable seed and a known method, the result is random to anyone who does not know the seed and perfectly reproducible for anyone who does. That combination is what allows a draw to be checked later.
For everyday purposes, think of it this way: random is what nobody can guess beforehand; verifiable is what anyone can redo afterward. A good draw needs both.
To see the idea in action, draw a few numbers and open the verification.