[Cryptography] Is it time for a revolution to replace TLS?

Sandy Harris sandyinchina at gmail.com
Sat Apr 19 12:31:06 EDT 2014


On Fri, Apr 18, 2014 at 1:22 PM, Tony Arcieri <bascule at gmail.com> wrote:

> We have at least 4 contests here, I think:
>
> 1) Better transport encryption (Tcpcrypt is already tackling this)

One criterion, I think, is that forward secrecy is a MUST.

I'd also have MUST support AES, SHOULD support the
other AES finalists with open licenses (Twofish, MARS &
Serpent).

> 2) Better key exchange (Tcpcrypt is also tackling this)

Is it perhaps time for another look at Photuris? That was a simpler
alternative to IPsec, might still have useful ideas to offer. There
are RFCs.

Definitely look at the JFK (Just Fast Keying) work, a
simpler alternative to IKE. There are academic papers
and were once Internet drafts; I do not know current
state of those.

> 3) A better certificate format

Can we get rid of certificates instead? FreeS/WAN did not
use them; it just put hex representations of raw RSA public
keys in DNS reverse maps. The current work on DANE
may make this approach more feasible and widespread
deployment of DNSsec would make it more secure.

> 4) A better system for authenticating/revoking keys (e.g. Convergence, Tack,
> CT)

Can we punt that over to DNSsec and DANE?


More information about the cryptography mailing list