Re: [tcpinc] Call for adoption of draft-rescorla-tcpinc-tls-option-05

[Matt Corallo <[e-mail]>]

I do not support adopting tcpinc-tls-option because:

* Using TLS (even a limited set of allowed options) as the tcpinc
mechanism loses the "defense in depth" property that tcpinc nicely
provides for some applications.
* I believe the extra round-trip for new connections to a new server
will significantly harm adoption of such a proposal. While it seems
reasonably possible to fix this issue, doing so brings tls-option
further from the stated(?) goal of re-using TLS and all of its many
years of study. This would seem to largely defeat the purpose.

Matt

On 10/20/15 16:49, Mirja Kühlewind wrote:
> Hi all,
> 
> please indicate if you support adoption of
> draft-rescorla-tcpinc-tls-option-05 as a tcpinc working group item, or
> not, by
> 
>     Monday, Nov 2, 2015.
> 
> draft-rescorla-tcpinc-tls-option is one candidate for tcpinc where the
> first version of this draft was proposed more than a year ago. Verison
> -04 was release about three weeks ago and specifies the TLS 1.3 profile
> as well as the use of draft-rescorla-tcpinc-tls-option with tcp-eno.
> Since then this draft received a lot of discussion. The lasted update
> was provided yesterday, but only changes a few minor fixes.
> 
> Similar as before, if you do not support adoption of this document
> because you think it is not in scope for the wg or has fundamental
> technicals flaws and would therefore harm the goals of the wg, it would
> be great if you could given some reasoning/explanation with your response.
> 
> This is solely an adoption call for draft-rescorla-tcpinc-tls-option
> independent of any other documents. If you have a personal preference
> for a different approach that should not be a reason to reject this
> adoption. Forcing the wg to make a decision has not worked previously,
> and even though both proposed approaches have evolved, I do not see any
> indication that the wg is now ready to make a decision. The goal of this
> adoption call is to figure out if there is enough interest and energy to
> further follow the approach as outlined in
> draft-rescorla-tcpinc-tls-option-05.
> 
> This process may lead to the situation where the wg will adopt and work
> on two solution approaches. This does not mean that the wg will publish
> two (incompatible) approaches, as this would not fulfill our charter. If
> we end up adopting more than one approach, I currently see three way to
> proceed:
> 
> 1) Both approaches (naturally) converge into one approach.
> 
> 2) We work on both approaches to get them into a (similar) state where
> the wg is able to make a decision (and withdraw the other doc).
> 
> 3) We publish both approaches as different 'versions' of tcpinc that can
> be negotiated in the tcp-eno handshake, where at least one of them is
> mandatory to support/implement.
> 
> Thanks!
> Mirja
> 
> _______________________________________________
> Tcpinc mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/tcpinc