i told my browser to disable caching of html documents alltogether, but it doesn't help. it must be my proxy's fault. i hardly had any trouble at home (where i don't use a proxy), just here at the office. when i reload a page, the http request is slightly different and the proxy fetches a new version of the page, but i can't tell my browser to send that kind of request upon klicking a link (or can i? how?).

well, in my case, the pages expire rather instantaneously, but even putting this line in and setting the value to 1 didn't help at all. i'll go talk to the admin of the proxy.

one server-side measure that would surely work would be to append some kind of session ID to every URL, as seen on orf.at. the proxy could/would differentiate between different versions of the same document and act accordingly. but don't do these IDs make the URLs look ugly?

you did get me wrong. as I can see in your code, you entered this:

meta http-equiv="pragma" content="no-cache" / meta http-equiv="expires" content="1" /

and this will not work; if you give it a try with "3600" instead of "1", it should work and the page will refresh automatically one hour after your last visit...

you can set a date, too, if you prefer this (see Self-HTML or here)

And: some of these things will not work for Netscape, others not for IE !

What I did is: I wrote an AppleScript, that cleans ALL my browser-caches within two seconds - I run it everytime I log out of the internet, so there is nothing left to reload and the browsers have to load the latest version of the page from wherever it is. This is for Macintosh only; don't ask me how to do this for Windows...

i got you wrong? maybe you think so because you absolutely failed to comprehend anything you read today. i'll try to explain:

1. expires is not for automatic "refreshing", it states for how long a cached document will be valid and may be pulled from a cache. this is also what the reference you gave is saying.
2. i did mean one second, because a dynamic page that has to be generated from scratch each time it's loaded isn't valid in any cache (so i actually should've put a 0 in there, just wasn't sure if this is a valid value -- thanks for the reference, which carifies that). i already explained that above.
3. consequently, setting a date would be completely out of the question.
4. by browser, which is neither IE nor Netscape, but implements the standards quite nicely, doesn't cache anything. i already stated this above.

the issue is: the caching proxy should also pay attention to those tags, but mine doesn't. but as i can reload pages okay, there could be a way to load them correctly the first time, by tricking the faulty proxy somehow.

What about being a bit more kind to people who are trying to help you? And while we're at it, also to people you are, um, trying to help?

@ nex: "... you absolutely failed to comprehend anything you read today ..." ? Well, you are so right, my friend. So sorry, I bothered you. Fact is: I was trying to help, like other people helped me. According to my humble English/German-capacities "you did get me wrong" means "Du hast mich mißverstanden" and is nothing more than eine Vermutung. I could not know, how smart you are; so, forgive me.

@ hns: thank you for your remark

hns, i'm sorry about the entry there (really!), i updated it.

docbuelle: your help is appreciated (you had pretty good ideas more than once), but your attempt to do it again above looked to me as if you're treating me like an idiot ("let me have a look at your code... oh! you did it all wrong! try again!", where i had actually corrected it so it made sense), and as it didn't help me, i was frustrated. maybe it's just me and my inability to interpret emoticons and self-righteous sarcasm :-)

ascertainment, conclusion, determination, statement: "it is."

assumption, conjecture, guess, guesswork, presumtion, speculation, supposition, surmise: "i think/guess/presume that it is."

mkay?