[Seaside] URL link and caching

[Avi Bryant]

On Tue, 19 Mar 2002, Kamil Kukura wrote:

> I think it shouldn't hurt the system to add response header
>
>     Cache-Control: private
>
> which in HTTP 1.1 means that response should not be stored in
> proxy-chache [shared-cache] memory.

Ah, ok.  How about you file an issue for that (bugs.beta4.com)?  Although
I don't think it's a problem without it, since the session key is part of
the url, and only someone else with that same session key would get the
cached page back.  But we might as well save SQUID from doing unnecessary
caching.

> Btw, this "save browsing" feature is deathblow to my Internet Explorer
> 6. Sometimes it complains page couldn't be loaded, sometimes it goes to
> neverending wait for response or spits out the message that page has
> expired. Netscape 6 is doing just fine.

Hmm, that ain't good.  Anyone else seeing the same thing?
In case it helps, let me make clear what "Safe Browsing" does.  Each page
view is split into two parts - a request ending in /action, which contains
the POST or GET data, and performs the action methods, which then
redirects to a request ending in /view, which simply renders the next
page.  This nice thing here is that reloading the /view page never
resubmits any data or causes any nasty side effects.  The cost is an extra
redirect and, apparently, issues with IE... I don't develop on Windows, so
I don't have any advice in that dept.  Julian, comments?

Avi