Re: SPDY Header Frames

Hi Poul-Henning,

On Fri, Jul 13, 2012 at 10:27:22PM +0000, Poul-Henning Kamp wrote:
> In message <CABP7RbepWH4ahSPHDU_M_w0tRVz_RRm1FV-jM_Y72=YHCVqO0g@mail.gmail.com>
> , James M Snell writes:
> 
> >As discussed within draft-tarreau-httpbis-network-friendly-00, and as has
> >been mentioned several times in discussion on list, handling of headers
> >within the current SPDY framing, and in particular the layering of HTTP/1.1
> >messages into SPDY frames is less than optimal. [...]
> 
> While I appreciate your willingness to keep trying to modify SPDY,
> I think it is premature to do so, certainly until we know what
> HTTP/2.0 should actually do and how it should do it.
> 
> For instance if we decide to do the envelope/substance split,
> (probably envelope/metadata/substance actually), that has pretty big
> implications for the requirements to the actual protocol.
> 
> Such fundamental questions should be resolved before we waste time
> trying to fine-tune representations.

I think such contributions here on the WG are useful. As you probably
have noticed when we spent countless hours rotating bits in every
possible combinations, starting from a blank page is extremely difficult,
and most of the progress comes from contributors who are able to comment
on what they like or don't like in the design.

Whatever will be retained as a basis for HTTP/2.0, this exercise is
useful and may incite other users to provide very valuable feedback.

Best regards,
Willy

Received on Friday, 13 July 2012 22:51:29 UTC