Thank you for the evaluation and info. The combination thing sounds familiar to me. I'm reminding that I've heard of that before in a crypto talk anywhere...
--- Ursprüngliche Nachricht --- Von: Christian Grothoff <[email protected]> Datum: 12.08.2018 23:33:29 An: [email protected] Betreff: Re: [GNUnet-developers] EcDSA signature scheme > This does not sound like a great idea, largely because the PQ algorithms > > are all a bit new and not nearly as well understood as classical crypto. > > A sane PQ implementation should _combine_ classical and PQ crypto, i.e. > sign/verify with both types of algorithms and for encryption use two > types of KX algorithms and then HKDF the results together. As they are > > not doing that (at least nothing in their documentation suggests this), > I would advise to stay away. > > Also, as far as GNUnet is concerned, Jeff is planning on putting some PQ > > crypto into the Lake design, and I'm don't see an urgent need to deploy > PQ elsewhere yet. But having good PQ crypto primitive implementations > out there would definitively be a good thing, but I'm not sure codecrypt > > is where I'd look. ;-) > > On 08/12/2018 06:44 PM, [email protected] wrote: > > News on the PQ site of things - at least worth it to keep an eye on: > > > Whonix includes Codecrypt by default now - > > https://www.whonix.org/wiki/PQCrypto#Use_Instructions > . > > Codecrypt is a GnuPG-like unix program for encryption and signing that > uses only quantum-computer-resistant algorithms. It's Free Software using > "GNU LGPLv3 or later" license, which is good. Codecrypt git: > https://gitea.blesmrt.net/exa/codecrypt > > > > > > _______________________________________________ > GNUnet-developers mailing list > [email protected] > https://lists.gnu.org/mailman/listinfo/gnunet-developers > _______________________________________________ GNUnet-developers mailing list [email protected] https://lists.gnu.org/mailman/listinfo/gnunet-developers
