3 ms·

>if SHA256 proves to be insecure slight pedantism here, but "when" not "if"

• marcan_42 5y ago
  That's not a given. As far as I know, there is no proof of the security or lack thereof of this kind of function. Increases in computing power alone can't render SHA256 insecure; that can be proven by thermodynamic/size-of-the-universe arguments. You need an actual cryptographic weakness.
  • newby 5y ago
    > Increases in computing power alone can't render SHA256 insecure; that can be proven by thermodynamic/size-of-the-universe arguments. Fascinating. Could you please recommend a source where a curious layman could learn more about this?
    • marcan_42 5y ago
      I posted a link and some further analysis in this comment: https://news.ycombinator.com/item?id=29883965 https://news.ycombinator.com/item?id=29883965
• hannob 5y ago
  There's no indication whatsoever of this happening. In the time since SHA1 came under attack, a lot more research has happened on hash function security. If this has shown anything, it's that SHA256 is more secure than people used to think. People seem to have this widely believed idea that "all crypto will be broken eventually". But in pretty much all cases you can trace back that broken crypto was known to be weak for a long time. SHA256 is not known to be weak (with the slight caveat that it doesn't protect against length extension attacks, but this is a known property that matters only in very rare circumstances).