Re: Cost analysis: (was: Getting to Consensus: CONTINUATION-related issues)

On 21 July 2014 01:06, Mark Nottingham <mnot@mnot.net> wrote:

> If you look at how I formulated the question, it was mostly about a/b (c
> still looks like it doesn’t have broad support), with a bit of wiggle room
> on x/y/z (and I do think we can answer a/b/c and then tackle x/y/z).


Hmm I saw it a little differently.  Most of the discussion was really
related to the impact of a header limit and very little about the transport
of large headers and if they need to be fragmented etc.

However, it does look like there is good support for y, which is really
your b (status quo with regard to large header transport and max header
expressed in uncompressed bytes).     So that's a worthwhile step anyway...
just not sure it is sufficient to quell discontent about continuations.

regards








-- 
Greg Wilkins <gregw@intalio.com>
http://eclipse.org/jetty HTTP, SPDY, Websocket server and client that scales
http://www.webtide.com  advice and support for jetty and cometd.