[Cryptography] Debunking the "SMTP TLS "s a mess" myth.

Viktor Dukhovni cryptography at dukhovni.org
Tue Oct 4 17:58:18 EDT 2016


> On Oct 4, 2016, at 2:10 PM, Phillip Hallam-Baker <phill at hallambaker.com> wrote:
> 
> ​The way to improve upon STARTTLS isn't actually to use DANE.
> 
> It would be to develop an infrastructure in which the active attacker doesn't ​know if their attack is going to be detected or not until after they have committed.
> 
> Something like 'pinning with fangs'.

Pinning fails in the face of MX indirection and STARTTLS
stripping.  MTA-to-MTA SMTP is not HTTP, the security model
does not carry over:

   https://tools.ietf.org/html/rfc7672#section-1.3

Real alternatives take the form of a detailed specification,
and implementation in a couple of MTAs...

-- 
	Viktor.



More information about the cryptography mailing list