[Cryptography] A seemingly simple question ...

Thierry Moreau thierry.moreau at connotech.com
Wed Feb 27 10:32:29 EST 2019


On 20/02/19 10:38 PM, Alfie John wrote:
> On Wed, Feb 20, 2019 at 02:55:05PM +0000, Thierry Moreau wrote:
>
> However, Noise and Wireguard have
> recently caught my attention:
>
>    http://noiseprotocol.org/noise.html
>    https://www.wireguard.com/
>
>
> It's 2019... let's take the footguns away from users and make safe defaults!
>

Thank you for this reminder. I found it difficult to get the core lesson 
from the Noise protocol framework document.

Here is my summary: despite giving credit to SIGMA (an important source 
of inspiration for IpSEC IKEv2) and H*MQV, the noise protocol usage of 
public key crypto is basically a (flexible) combination of ephemeral 
Diffie-Hellman and static (i.e. authenticating) Diffie-Hellman.

There are many other aspects ... The above is just the core arrangement 
for a secure key exchange handshake. Is it good or bad? No opinion from 
my part except that this arrangement has been (implicitly) in the 
literature for a long time (too simple to attract attention of 
theoreticians).

- Thierry



More information about the cryptography mailing list