W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2007

Re: Accept- & Content-Resolution headers proposal

From: Mike Dierken <dierken@gmail.com>
Date: Thu, 7 Jun 2007 09:17:17 -0700
Message-ID: <7cd8e0930706070917o726d11edqa6e1a39f28c9e0f9@mail.gmail.com>
To: "Nicholas Shanks" <contact@nickshanks.com>
Cc: "Julian Reschke" <julian.reschke@gmx.de>, ietf-http-wg@w3.org

> Images, and anything that refers to images would need to support it.
> You may have existing content that is named foo-hires.jpeg and foo-
> lowres.jpeg and want to switch between them without doing the
> negotiation on the images themselves (so that old references don't
> need to be fixed)

Ahh, well there's yer problem right there...
The "... so that old references don't need to be fixed..." is the
issue. It turns out that on the Web it's much better to have two
references and work on 'fixing up' the sources that point to one or
the other. One concrete example is a caching proxy that has been
taught to only look at the URI - without updating the implementation
of all those existing caching servers, the wrong bits will be returned
to clients that ask for "foo.jpg" because the cache hasn't been taught
to look for this new header. In this case perhaps using the Vary
header in a response may help, but I believe there are other areas
where having different representations for the same URI may create
problems.
Received on Thursday, 7 June 2007 16:17:28 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 06:50:10 GMT