> On 13 Jul 2018, at 22:37, Bernd Fix <[email protected]> wrote:
> And maybe even a third one: I stumbled across an approach to use
> Curve25519 keypairs for both ECDH and Ed25519 signatures
> [https://moderncrypto.org/mail-archive/curves/2014/000293.html].

I don’t think it breaks Taler per se, but it’s needlessly complex.. and it 
damages the deterministic signatures property of Ed25519.

Also, I’m not 100% sure that NaCL based libraries lack a suitable Edwards 
scalar multiplication.  They may not expose it, but Ed25519 signature 
verification involves a variable-time double scalar multiplication.   This 
variable-time operation suffices, except that it enables javascript side 
channel attacks.  You could prevent those using key splitting.  Ain’t pretty 
obviously.  :)

Jeff


Attachment: signature.asc
Description: Message signed with OpenPGP

_______________________________________________
GNUnet-developers mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/gnunet-developers

Reply via email to