Re: HPACK: static header list and Accept header field


note the discussion we have on which was closed as revisit
on change.  Ie if a breaking change is made to the protocol, then we should
optimise the static table.   My testing shows that there is benefit for
trimming the names, juggling the order and providing values.    But I don't
have a very extensive data set to work with.


On 11 January 2015 at 18:07, Julian Reschke <> wrote:

> Hi there,
> I was looking at the static table, and also watching what browsers send in
> their initial request.
> Would it make sense to come up with an entry in the static header field
> that would actually encourage browsers to converge on a common value?
> (Yes, I realize that Chrome also ask for Webp, but that in theory could be
> sent in a 2nd header instance)
> Best regards, Julian

Greg Wilkins <>  @  Webtide - *an Intalio subsidiary* HTTP, SPDY, Websocket server and client that scales  advice and support for jetty and cometd.

Received on Thursday, 15 January 2015 21:05:49 UTC