Re: [TLS] Confirming consensus about one draft-ietf-tls-renegotiation detail

[Bill Frantz <[e-mail]>]

[e-mail] (Paul Hoffman) on Thursday, January 28, 2010 wrote:

>The metadiscussion is raging (in both senses of the word), but the number of people expressing an opinion has dropped to near zero. 
>This is just a gentle nudge to those who care but have not responded to do so.

Let me cast a weak vote for:

   (1) I prefer publishing the specification as-is.

My real interest is getting on to the next version of TLS. I hope it
includes changes designed to:

(1) Limit the effect of bugs in a particular feature to those who actually
use the feature. Very few applications actually used renegotiation, but all
were affected. The use of PKI for peer authentication might fall in this
category.

(2) Eliminate features which can reasonably be handled at the application
layer. To my mind, renegotiation for key/certificate change on long-lasting
connections falls in this category.

(3) Enable future security bug fixes to be handled by the industry standard
mantra, "Just upgrade to the latest version", instead of requiring hacks to
be compatible with versions back as far as SSL2. (I realize this is more of
a social than a technical goal.)

Cheers - Bill

---------------------------------------------------------------------------
Bill Frantz        |"After all, if the conventional wisdom was working, the
408-356-8506       | rate of systems being compromised would be going down,
www.periwinkle.com | wouldn't it?" -- Marcus Ranum