Why Quantum Cryptanalysis Is Bollocks [pdf]

cs.auckland.ac.nz

4 pointsdweekly3 comments

As long as the useful, non-nutty research is supported, this isn't something to worry about.

But...is it? And how to find out? Like, afaik in 2014 OpenBSD was going to shut down build farm for non-mainstream architectures, that did not happen only thanks to last minute donations.

the presentation/article point is exactly the opposite is/will happen

Oh, it already happened.

Case in point: it is possible to build cheap and fully auditable harware RNG, with bandwidth fully sufficient to replace /dev/random on most servers. I have one. But apparently noone wants such a thing, or if they even consider it, they insist on adding unneeded "feel-good" complexity like whitening.

But...somehow the sky has not fallen. Trying to understand why. Maybe it isn't really a problem?

Reply on news.ycombinator.com