Conversation

Notices

1. Embed this notice
   *(*(void **(*)()) lizzy)() :v_trans: :v_bi: (fleckenstein@social.lizzy.rs)'s status on Wednesday, 19-Jun-2024 22:52:12 JST *(*(void **(*)()) lizzy)() :v_trans: :v_bi:

   • anna

   @navi i think that phrase applies to implementations as well. but i disagree strongly...

   • Haelwenn /элвэн/ :triskell: likes this.
   • Embed this notice
     Haelwenn /элвэн/ :triskell: (lanodan@queer.hacktivis.me)'s status on Wednesday, 19-Jun-2024 22:57:43 JST Haelwenn /элвэн/ :triskell:

     in reply to

     • anna
     @fleckenstein @navi Same, although some implementations got glaring design faults but I'd say most are due to people not evaluating their dependencies and a culture of cowboy coding.
     
     So far the cryptographic algorithms I've seen got pretty good test vectors so you're unlikely to screw up the math. The rest is purely programming flaws and those can happen both in libraries and applications (and libraries can't protect you much).