W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2014

Re: Zero weight for 100 CONTINUE, instead of flow control

From: David Krauss <potswa@gmail.com>
Date: Thu, 3 Apr 2014 14:06:04 +0800
Cc: David Morris <dwm@xpasc.com>
Message-Id: <BF371E36-56F1-45E2-83AB-4B3E37EF54BE@gmail.com>
To: HTTP Working Group <ietf-http-wg@w3.org>

On 2014–04–03, at 5:16 AM, David Morris <dwm@xpasc.com> wrote:

> So I suspect such a feature wouldn't do much in general to improve
> the user experience. Probably better would be a mechanism that allowed
> the recipient to pace the transfer based on a bytes per second target.
> At least some of the major video sites do this now to avoid sending
> content the user will click away from in seconds.

This is an issue with weighting in general, not my particular CONTINUE proposal. (Your response seems to be in relation to Herve’s proposal, though, since mine does not allow a stream to remain at zero weight.)

Any reasonably rich site today is loaded from several CDNs in addition to actual page’s origin. So, I rather doubt that prioritization will have much user-experience effect at all except in the case of opening background tabs from the same site while the foreground tab is still loading something.

The problem with hard, bytes-per-second data flow targets is implementation difficulty. The speed metadata originally inherited from SPDY was dropped. Implementation difficulty might not be a showstopper though, but only an indication that such a feature should be optional as indicated by SETTINGS.


Received on Thursday, 3 April 2014 06:06:42 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:14:29 UTC