Re: [TLS] In support of encrypting SNI (off-topic)

[S Moonesamy <[e-mail]>]

Hi Michael,
At 12:54 17-05-2014, Michael Carbone wrote:
>I'm not exactly sure what you are suggesting, but that's why Access is
>here -- to help highlight the importance of this proposal to the folks
>who are actively involved in the discussion and consideration of
>proposals for the specification.
>
>Waiting and assuming that an encrypted TLS handshake by default will
>appear in the TLS 1.3 specification seems like a non-ideal approach,
>given that without discussion, a proposed implementation, and support,
>it will likely not get there. We recognize honing a proposed
>implementation is a lot of work and we want to support those who are
>currently discussing and drafting one. It is worthwhile, and the
>real-world impact will be tangible.

I agree that waiting and assuming that the outcome of a specification 
would be favorable is a non-ideal approach.

>It's difficult for these communities to ensure their needs are taken
>into account given that they are non-technical, unaware of the IETF, and
>obviously not part of the process. We want to make sure such users are

Yes.

Although it might not be the intent the message came out as designing 
a TLS specification to bypass censorship for political 
reasons.  There were two organizations [1] which posted comments 
along those lines.  I suggested framing the comments as technical 
input and avoid the political advocacy angle.

Note that I am okay with encrypting the SNI as the average user would 
expect that the information being sent can only read by the endpoint 
on the other side.

Regards,
S. Moonesamy

1. I am aware of how affiliation works in the IETF.  I used the word 
in my explanation for clarity.