3 ms·

A useful article, originally from 2014 but updated recently. Previous HN discussions: https://news.ycombinator.com/item?id=7359992 https://news.ycombinator.co

• djsumdog 10y ago
  I remember seeing it before. It's a good post. What I think is really interesting is the question: how do you prove something is random. If you take a chunk of data that uses strong encryption, without headers or other identifiers, it should look like it has a random distribution. That's one of the reasons for the great concern around the Snowden leaks that implied some manufactures had compromised their hardware AES chips: https://theintercept.com/2016/01/04/a-redaction-re-visited-nsa-targeted-the-two-leading-encryption-chips/ https://theintercept.com/2016/01/04/a-redaction-re-visited-n...