Re: [TLS] Publication of draft-rhrd-tls-tls13-visibility-00

[Stephen Farrell <[e-mail]>]

(With the obvious caveat that I hate the whole
idea... :-)

On 12/10/17 13:57, Hubert Kario wrote:
>>
>> Anyway, I think key life length could be addressed in later drafts, but the
>> inability of the client to identify (and possibly reject) the tapping third
>> party is a "no go" for me...
> yes, a three-way DH with two certificates (one IPS one EE server) does seem 
> like a much better approach, especially if IPS certs need to have special 
> flags that make them useless for anything else and cannot be set by CAs in 
> public CA programs.
> 

But... even if one did add names/certs for the wiretapper
or IPS or whatever then there could be more than one of
those who'd like to be able to interfere with the TLS
session, and there's no way I can see that makes sense for
the client and server to negotiate which wiretappers they
allow/like.

This is all just squaring-the-circle, TLS is meant to be
a 2-party protocol (ignoring CAs) and I don't see any way
to make TLS a multi-party protocol that works.

S.