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

Re: Require UAs and intermediary caches to assume Vary: User-Agent

From: Phillip Hallam-Baker <hallam@gmail.com>
Date: Wed, 17 Oct 2012 07:30:48 -0400
Message-ID: <CAMm+Lwjm4rdJ35s-f7FwHZT+a5s=ky_iegGhHdh3zuN9J7GmJQ@mail.gmail.com>
To: Mark Nottingham <mnot@mnot.net>
Cc: Simon Pieters <simonp@opera.com>, ietf-http-wg@w3.org
How much of the Web actually passes through client side proxies these days?
Of those, how many have  caches?

There was a good reason for caching proxies in 1993. I don't see much
justification or utility in fiddling with that part of the spec now. The
content that is relevant for caching these days is huge chunks of video,
audio and images, the part that is generated is text.

But in any case, I can't see why a server that cares can't generate the
vary header for the user agent header...

Bigger problem these days is that there are many sites that just break with
the 'wrong' user agent. All those clueless 'mobile' versions of sites that
don't actually link to any content. Just a page congratulating the
programmer on producing a mobile edition of the site instead of the linked

Net is that even if a site specified the vary header on user agent, it is
probably best ignored just as I switched from safari to chrome on my iPhone
because it gives fewer stupid problems like that.

On Wed, Oct 17, 2012 at 3:32 AM, Mark Nottingham <mnot@mnot.net> wrote:

> Um, no.
> Not only will this retroactively make all intermediary caches
> non-conformant, it'll also make them completely useless, because of the
> large (and unnecessary) amount of variance in User-Agent headers.
> I understand there are security issues here caused by CORS, but that
> doesn't give you the ability to force changes on the rest of the Web.
> Regards,
> On 17/10/2012, at 6:19 PM, "Simon Pieters" <simonp@opera.com> wrote:
> > See http://www.w3.org/mid/507D94AB.3000207@mit.edu
> >
> > It would be useful if the spec required that UAs and intermediary caches
> act as if all responses had Vary: User-Agent specified, since many pages on
> the Web serve different content depending on the User-Agent header but do
> not bother specifying Vary: User-Agent.
> >
> > I would like this to be filed as a new issue in the issue tracker.
> >
> > cheers,
> > --
> > Simon Pieters
> > Opera Software
> >
> --
> Mark Nottingham   http://www.mnot.net/

Website: http://hallambaker.com/
Received on Wednesday, 17 October 2012 11:31:15 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 11:11:07 UTC