Re: [hybi] Why not just use ssh?

[Gabriel Montenegro <[e-mail]>]

> I claim that the TLS+NPN protocol that I sent to the list earlier resists cross-
> protocol attacks, which are a clear and plausible threat model.  The security
> argument is fairly straightforward.

My question was whether there was anything to substantiate such a claim. The proposal you sent did not have an analysis, from what I recall. However, I'm actually much more interested in your analysis of the HTTP Upgrade handshake (not the TLS-NPN version).  In particular, I'd be interested in your analysis of much simpler variants of the handshake (than the one in version -76) that would provide similar protection against cross-protocol attacks.

It would be good to hear your answer to Greg's last question here:
http://www.ietf.org/mail-archive/web/hybi/current/msg01269.html

I know this is not your main focus, but I'm hoping you recognize that the "HTTP Upgrade variant" is much closer to the WG position than a "TLS-NPN only" position, and that this will convince you to contribute there as well. [Of course, having an HTTP upgrade variant does not preclude a TLS-NPN variant from being pursued as well. ]

> As for complexity, the TLS-only model is less complex than the existing
> handshake, as evidenced by the fact that I would describe it completely simply
> by deleting text from the current draft.

Deletion of text upstairs only? Presumably, you're assuming some other text (and complexity) gets added elsewhere (TLS-NPN spec), implying changes to massively deployed software, acceptance in the TLS WG based on perceived tradeoff, etc.