2 ms·
This is on of those articles where they have someone who doesn't really know anything about math or computers interview somebody technical in the field, I thin
by bee_rider 5y ago
This is on of those articles where they have someone who doesn't really know anything about math or computers interview somebody technical in the field, I think. So most of what the journalist gets, unfortunately, is that PRNGs are a thing but not perfect.
I'm pretty sure the project is looking at physical devices to add to the entropy pool, and they talked about two examples (which do seem pretty neat).
> One relies on the patterns magnetic films make when disturbed, the other on how electrons travel through the barrier of a quantum-tunnelling diode. Both of these things are truly random.
The use of the Von Neumann quote, "Anyone who considers arithmetical methods of producing random digits is, of course, in a state of sin” is pretty annoying, given that he said it in defense of his PRNG. Sort of a tongue-in-cheek "yes this is fundamentally wrong (sinful) but I'm going to do it" sort of thing.