W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2008

Re: Page Relocate

From: Christophe Strobbe <christophe.strobbe@esat.kuleuven.be>
Date: Fri, 05 Sep 2008 11:29:16 +0200
Message-Id: <>
To: w3c-wai-ig@w3.org


At 09:48 29/08/2008, David Woolley wrote:

>Ryan Jean wrote:
>><META HTTP-EQUIV=Refresh CONTENT="0; URL=test.htm">
>This is explicitly a SHOULD NOT in the HTML specification.  I 
>imagine they would have given it a MUST NOT except that the HTML 
>specification isn't normative for the "refresh" value in meta.
>This breaks the back button.

In my tests, http-equive="Refresh" did not break the back button in 
Mozilla browsers (Mozilla, Firefox, SeaMonkey), Internet Explorer 6 
and Opera (all only tested on Windows) if the timeout is 0 seconds, 
but it broke the back button in IE 6 on Windows 2000 (but not on 
Windows XP) if there was a timeout.
See my collection of tests and test results at 
<http://purl.org/NET/error404/xp/scripting/redirect/>. (If anyone 
sends me test results for other browsers and platforms, I'll add them.)

>Incidentally, this is a mix of HTML and vendor extensions to 
>HTTP.  It will actually work with a real HTTP header, and on non-HTML content.
>This breaks the back button, so is an accessibility no-no.

In my tests, this broke the back button in Internet Explorer 6 and 
Opera (because the redirect fires again) but not in Mozilla browsers.

>Whilst I believe there are scripting techniques that don't break the 
>back button, not everyone has JavaScript, e.g. Lynx and Amaya have 
>no support, and many who do disable it for security reasons.  One 
>common piece of advise in security advisories is to disable browser 
>scripting, and, in secure environments, the standing risk from such 
>a level of pragrammability is generally considered too high.

Even though I prefer server-side redirects, there are situations 
where an author doesn't have sufficient control over the server to 
implement them.

The Techniques for WCAG 2.0 contain the following relevant techniques 
and failures:
* SVR1: Implementing automatic redirects on the server side instead 
of on the client side
   (with example code in JSP, ASP and PHP, and an Apache configuration example)
* G110: Using an instant client-side redirect
* H76: Using meta refresh to create an instant client-side redirect
* F41: Failure of Success Criterion 2.2.1, 2.2.4, and 3.2.5 due to 
using meta refresh with a time-out

Christophe Strobbe

Christophe Strobbe
K.U.Leuven - Dept. of Electrical Engineering - SCD
Research Group on Document Architectures
Kasteelpark Arenberg 10 bus 2442
B-3001 Leuven-Heverlee
tel: +32 16 32 85 51
Please don't invite me to LinkedIn, Facebook, Quechup or other 
"social networks". You may have agreed to their "privacy policy", but 
I haven't.

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm
Received on Friday, 5 September 2008 09:30:37 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:38 UTC