W3C home > Mailing lists > Public > ietf-http-wg@w3.org > July to September 2012

Re: Connection limits, was: Straw-man for our next charter

From: Julian Reschke <julian.reschke@gmx.de>
Date: Wed, 01 Aug 2012 00:17:54 +0200
Message-ID: <50185992.70101@gmx.de>
To: ietf-http-wg@w3.org
On 31.07.2012 22:19, Mike Belshe wrote:
>
>
> On Tue, Jul 31, 2012 at 12:08 AM, Julian Reschke <julian.reschke@gmx.de
> <mailto:julian.reschke@gmx.de>> wrote:
>
>     On 31.07.2012 06 <tel:31.07.2012%2006>:04, Mike Belshe wrote:
>
>
>
>         On Sun, Jul 29, 2012 at 10:45 AM, Larry Masinter
>         <masinter@adobe.com <mailto:masinter@adobe.com>
>         <mailto:masinter@adobe.com <mailto:masinter@adobe.com>>> wrote:
>
>              Your post is consistent with the assertion that there isn't
>              agreement yet about what "faster than HTTP/1.1" means, or
>         how to
>              compare proposals for improvement. And neither measured
>         worst case
>              latency or real network traffic with buffer bloat, or
>         situations
>              that would detect the impact of HOL blocking.
>
>
>         While SPDY leaves a tiny HOL issue, it fixes the massive one from
>         HTTP/1.1, which can only load a couple of resources in parallel per
>         domain (2 by spec, 6 by implementation best practices).  The
>         tradeoff
>         turns out to be a boon in terms of reduced latency while also using
>         fewer network resources.
>         ...
>
>
>     "By spec" in RFC 2616, but not in HTTPbis (this has been fixed a
>     LONG time ago!).
>
>
> Alright :-)  spec fine, but still a practical issue in all major browser
> implementations today.

It's not a spec problem anymore, it's an implementation problem. Adding 
a new spec doesn't solve *that* problem.

Best regards, Julian
Received on Tuesday, 31 July 2012 22:18:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 31 July 2012 22:18:55 GMT