W3C home > Mailing lists > Public > www-tag@w3.org > March 2010

ISSUE-57 (HttpRedirections) status report

From: Jonathan Rees <jar@creativecommons.org>
Date: Tue, 9 Mar 2010 12:00:28 -0500
Message-ID: <760bcb2a1003090900j6ae5f388qa4a45f5b576b9731@mail.gmail.com>
To: www-tag@w3.org
Noah, I found a note I made to myself during the F2F that says you
want issue status reports by yesterday. Here's my report on ISSUE-57.
This is just a status report, not advice regarding the F2F, so let me
know what else you need. I've omitted references to save time and
because it's pretty easy to find most of them from the issue page in
tracker.  -Jonathan

This issue is a catchall for concerns around the use of 303 as way to
obtain descriptions of things named by URIs.

Here are the sub-issues mentioned in the issue description:

. Cacheability of 303 responses - RESOLVED by HTTPbis

. UI confusion & bookmarkability woes when browser address bar changes

    This concern is shared with 307 and 302 redirects.  We've discussed
    this recently and I will be preparing a blog post.

This question was mentioned in the original email that spawned the issue:

. Overhead (the 303 solution's requirement for two HTTP round trips)

    Suggestions include 203, the entity in the 303 reponse, a new
    HTTP status code. .well-known creates a new possibility, namely
    a sitewide URI rewrite rule - get the rule once, and you can get the
    description resource URI of the resource named by any of the site's URIs.
    (But the latter would fall under ISSUE-62)

The following would fall under "further consideration" of 303s:

. Suitability of 303 for this purpose - is it in spec? - RESOLVED by HTTPbis

No longer tracked under this issue:

. Getting descriptions from the URI's server (see ISSUE-62, Link:, .well-known)

    I've started to write a memo on discovery opportunities presented by Link:
    and .well-known

Not being discussed much:

. "Other ways" of getting descriptions

    Getting descriptions using Link: and .well-known are "other ways". Beyond
    these and 303, nothing is under discussion. (The story is richer
for metadata,
    but that's a different story, see ISSUE-63.)

Tracked under this issue:

. AWWSW TG (because it's investigating the semantics of redirects
  and "information resources")

Topics connected to this issue, might be tracked here if we were to
discuss them:

. "Bringing the httpRange-14 discussion to a conclusion" (HT)

. Preparing a finding that explains the httpRange-14 decision

. Why isn't everyone just using # URIs in the first place?

Note: 303 is now widely deployed and it is possible the community has
learned to live with it.  One way to deal with an issue is to wait until
it gets stale and people stop caring.
Received on Tuesday, 9 March 2010 17:01:16 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:48:20 GMT