- From: Martin Thomson <martin.thomson@gmail.com>
- Date: Sun, 4 Jan 2015 20:10:32 -0800
- To: Ryan Hamilton <rch@google.com>
- Cc: Jyrki Alakuijala <jyrki@google.com>, Roberto Peon <grmocg@gmail.com>, Mark Nottingham <mnot@mnot.net>, Dave Garrett <davemgarrett@gmail.com>, HTTP Working Group <ietf-http-wg@w3.org>
On 4 January 2015 at 19:46, Ryan Hamilton <rch@google.com> wrote: > This seems like an interesting intellectual discussion. Since HTTP/2 is in > Last Call, it seems like there would need to be a major problem with HPACK > to reopen the header compression algorithm. I'm not seeing anything like > that in this thread. I think that is a fair summary. There are, in any process like this, alternatives that are equally valid in the long term, and it could be that what Jryki is arguing for here is one such alternative. In this space, it's definitely not a lack of alternatives that was ever the problem. But this particular alternative hasn't been thoroughly analyzed, and it definitely isn't clear that it is superior either (Roberto's criteria being those we would judge by, compression efficiency is definitely low in the priority order). As Ryan observes, even something superior would be insufficient; HTTP/2 is done, the only thing that would change that fact is if there were a demonstrable flaw. I'd add that a change would only occur if that flaw didn't have a good workaround.
Received on Monday, 5 January 2015 04:10:59 UTC