[11535] in www-talk@info.cern.ch
Re: Enforce reloading of page when using the back-button -Forwarded
daemon@ATHENA.MIT.EDU (danielh@econ.ag.gov)
Fri Mar 17 00:02:02 2000
From: danielh@econ.ag.gov
Message-Id: <200003170456.XAA23230@lycanthrope.crosslink.net>
X-Really-To: <www-talk@w3.org>
Date: Fri, 17 Mar 2000 00:09:36 -0500
To: www-talk@w3.org
In-Reply-To: <s8d168a4.095@ers.usda.gov>
Resent-From: www-talk@w3.org
>This means that is you return a "220 O.K." the browser must treat it like a
>200 O.K. except for the fact that it must not cache the page.
>We also had to add <META HTTP-EQUIV="Pragma" CONTENT="no-cache"> to the
>pages
>( which had previously on it's own had not worked )
>We tested this and it works to break the back button for IE4, IE5.1,
>Netscape 3.04 And 4.04. All of the pages made a new request to the
>server.
>Unfortunately IE 3.01 on NT 3.5 still cached the pages.
I vaguely recollect that you have a reason for forcing reloads on
movements in the history list; but I hope you do realize that, in
general, unasked for re-requesting due to movements in the history
is a major pain in the neck.
It's something that I (as a server author) have had to try and "break"
(by adding expires of a several minutes to otherwise static documents).
Not something I'm proud of, but better then the all too frequent behavior
of most versions of netscape that I've used.
Daniel Hellerstein
danielh@crosslink.net
http://www.srehttp.org
