Re: WGLC comment on draft-ietf-httpbis-encryption-encoding-03, was: Encryption content coding simplification

On 2016-10-13 19:46, Mike Bishop wrote:
> Well, I can point at one, though it's not exactly a model of perfect HTTP C-E integration design....  Looking at[MS-PCCRTP].pdf, there are additional headers that carry the client's parameters (which the server will need if it chooses that coding) and then carry the server's selections back.
> Most notable (and probably the worst choice :-) ) is that rather than defining a new C-E value for v2, the parameters include the client's min/max supported versions, and the server tells the client which version it used in a response header.

...and SDCH will also need a way to send compression related parameters, 

Best regards, Julian

Received on Thursday, 13 October 2016 18:09:46 UTC