Re: Signing HTTP (Was: New Version Notification for draft-thomson-http-encryption-00.txt)

Le Mer 13 mai 2015 01:38, Martin Thomson a écrit :

> The signing scheme in [1] is definitely a candidate here. But I think
> that it attempts to resolve what is a fundamental dichotomy in the
> protocol: intermediaries can and do change header fields and we can't
> predict which.

The fundamental dichotomy is that the spec mandates that intermediaries
piggy-back on http headers to communicate with web clients (auth…) but
https advocates want to cipher everything including those headers.

This can not work out.

No mater what scheme you choose, until you separate the proxy client com
from the server client com nothing will be resolved

Http2 would have been a golden opportunity to define separate hop by hop
frames and end the bickering once and for all.

-- 
Nicolas Mailhot

Received on Wednesday, 13 May 2015 12:47:36 UTC