- From: Zhong Yu <zhong.j.yu@gmail.com>
- Date: Sat, 21 Jun 2014 13:45:28 -0500
- To: K.Morgan@iaea.org
- Cc: Roland Zink <roland@zinks.de>, HTTP Working Group <ietf-http-wg@w3.org>, C.Brunhuber@iaea.org
On Sat, Jun 21, 2014 at 1:02 PM, Zhong Yu <zhong.j.yu@gmail.com> wrote: > On Sat, Jun 21, 2014 at 11:34 AM, <K.Morgan@iaea.org> wrote: >> We have rapidly growing append-only files. Currently we use gzip Transfer-Encoding to compress an identity range request. But since T-E went away (again) in h2, it's no longer possible. > > damn it... I just implemented the feature on my server. another fool's errand. Sorry if that's misleading. I meant that I implemented it on a 1.1 server. > I realize it's a controversial issue, but if I understand correctly, h2 is supposed to replace *everything* you could do in 1.1. I'm not keeping up with h2, but I assume h2 will have its own mechanism to compress bytes on transfer? And h2 intermediaries can interact with h1 clients/servers using TE/T-E headers? Then it's not a problem. Zhong Yu http://bayou.io
Received on Saturday, 21 June 2014 18:45:55 UTC