On 15 October 2014 15:22, Adrian Cole <adrian.f.cole@gmail.com> wrote: > If we agree that api stuff does in fact use a lot of non-standard > headers, and we believe flipping back doesn't maim normal web, could > that be a strong enough case to revert? > I don't think we should set this up as standard headers vs dynamic headers. Both have very valid use-cases and it is impossible to say which is more deserving of the 1 byte slots. Simply flipping back just changes the bias and causes other problems. If an argument can be made that 2 byte encodings are still too large for dynamic headers, then instead of flipping back let's investigate how the 1 byte slots can be shared between static and dynamic. Reducing the size of the static table does not appear to have drastic effects on compression, so if the slots thus freed are a benefit for dynamic headers then we have a good compromise. I'm also OK with the status quo, as I think we are talking +/- 1% when we are already saving >60% But I'm happy to run the data if people think it is more than 1%. cheers -- Greg Wilkins <gregw@intalio.com> @ Webtide - *an Intalio subsidiary* http://eclipse.org/jetty HTTP, SPDY, Websocket server and client that scales http://www.webtide.com advice and support for jetty and cometd.
Received on Wednesday, 15 October 2014 04:59:57 UTC