[noise] "have one joint and keep it oiled"

Alexey Ermishkin scratch.net at gmail.com
Thu Dec 28 08:08:50 PST 2017


I don’t think we need canonicalization (yet) because it’s basically a write-once operation.

Might be useful with some kind of certificates, but even then it’s not a big deal. We can just treat whatever message as an array of bytes and forbid editing.

 

BTW looks like Google uses some homegrown protocol instead of TLS inside their infrastructure. Seems like a good place for NoiseSocket to fit in 😊

From: Noise [mailto:noise-bounces at moderncrypto.org] On Behalf Of Tony Arcieri
Sent: Thursday, December 28, 2017 8:48 PM
To: David Wong <davidwong.crypto at gmail.com>
Cc: noise <noise at moderncrypto.org>
Subject: Re: [noise] "have one joint and keep it oiled"

 

On Wed, Dec 27, 2017 at 8:53 AM, David Wong <davidwong.crypto at gmail.com <mailto:davidwong.crypto at gmail.com> > wrote:

I think I can see from that why Trevor wants to use something
flexible/extensible like protobuf in the payload parts of handshake
patterns. Is there more discussions on this? I've only found this:
https://moderncrypto.org/mail-archive/noise/2017/001373.html

 

Serialization is certainly an interest area of mine.

 

On that front, DeepMind (where Ben Laurie works inside of Alphabet) just put up a project for computing structured/nested content hashes of protos (as an alternative to canonicalization):

 

https://github.com/deepmind/objecthash-proto

 

-- 

Tony Arcieri

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://moderncrypto.org/mail-archive/noise/attachments/20171228/e3d659c5/attachment.html>


More information about the Noise mailing list