[noise] Protocol Names

Trevor Perrin trevp at trevp.net
Fri Jun 16 13:14:30 PDT 2017


On Fri, Jun 16, 2017 at 6:59 PM, Jake McGinty <me at jake.su> wrote:
> Based off of the input from this conversation, I wrote up a draft PR for
> more explicit (but backwards-compatible) definitions for how a protocol name
> is constructed and parsed.
>
> https://github.com/noiseprotocol/noise_spec/pull/34
>
> Does this reflect the general consensus so far?


Thanks, I think that captures what we've discussed so far:

Noise_BASEmod+mod+mod_Multiple+DH+Functions_Cipher_Multiple+Hash

Thinking about the future, a couple more ideas:

 * Maybe allow "/" in addition to alphanumerics in primitive names?
Thinking of cases like:

SHA3/256 or SHA3/512 (versus SHA3256, SHA3512)
SHA512/256
Salsa20/12
ChaCha20/Poly1305

 * I could also imagine parenthesized primitive names to allow arguments, e.g.:

Noise_XX+sig_25519+(RSA_PKCS1_2048)_AESGCM_SHA512

But I guess we could consider that later?


Trevor


More information about the Noise mailing list