5 ms·

Patented, naturally.

• xenophonf 14y ago
  Crypto is probably one of the few types of software patents that I'd support. It's non-obvious in the extreme, and patenting a cryptosystem might be a good way to open it for public review.
  • DennisP 14y ago
    Bruce Schneier has written that patents for crypto algorithms are useless. A crypto algorithm is only trustworthy if it attracts the attention of a lot of cryptographers trying to break it, and for the most part they don't bother with patented algorithms. There's no worry about opening algorithms for public review. Nobody with a clue is willing to use a proprietary algorithm in the first place.
  • betterunix 14y ago
    When last I checked, we were not supposed to have patents on math. Now, we can argue endlessly about whether or not that covers all software, but if public key crypto is not math then I am not really sure what is...
    • DennisP 14y ago
      Sadly, RSA was patented. The patent expired in 2000.
      • betterunix 14y ago
        I know, and it is worse than that. Over a hundred patents on ECC techniques, at least five on identity based encryption, and even patents on fully homomorphic encryption despite the fact that FHE is still years from practicality (including at least one patent on FHE that was granted before Gentry even showed that FHE is possible). It is, simply put, a monumental failure on the part of patent examiners.