Re: [TLS] RFC 5878 - why?

[Ben Laurie <[e-mail]>]

On 19 September 2013 09:55, Peter Gutmann <[e-mail]> wrote:
> Trevor Perrin <[e-mail]> writes:
>
>>I assume that's why no-one reviewed it carefully.
>
> [...]
>
>>The "Experimental" marking wasn't enough to prevent people like CT, TACK,
>>DTCP, and OpenSSL wasting a lot of time evaluating this and trying to use it
>>in the last couple of years.
>
> This seems to combine the worst of both worlds, no-one was really interested
> in it so it passed without review, but then there was enough interest in it
> later for people to try and implement it.  Can anyone who tried to implement
> it explain why they did so?

In CT's case it was because it seemed convenient at the time - we need
to send authz data during the TLS handshake, and here was a way
already defined to do it. We debated allocating a new extension for
it, but decided against it. In the end, because 5878 had errors that
we did not have the power to correct, we decided to abandon it.

I notice that the authors have not responded to the corrections,
either. So I guess they've abandoned it, too.