On Mon, 1 Aug 2005, Robert Blasius wrote:
[color=blue]
> Thanks. Ok, I will code an browsercheck[/color]
That's a very worrying remark. I'd say 99.9% of attempts to identify
the browser are misconceived, in so many different ways. Whereas, IE
provides its own proprietary syntax for singling itself out, which
(used appropriately, of course) does no harm at all to web-compatible
browsers.
Namely, the so-called "MSIE conditional comments" (these have to be
applied to the HTML, though, not to CSS).
See
http://msdn.microsoft.com/workshop/a...omment_ovw.asp
but note that although most of the syntax described there is harmless,
MS apparently couldn't resist sticking in a few dirty tricks which one
would do well to avoid - specifically, the item which they call
"downlevel-revealed", which isn't acceptable HTML syntax. But when I
needed to use this technique I didn't experience any difficulty
working around that detail.
Their specific examples are of course drivel (no self respecting web
author would talk to their readers like that!!) but the principles
are, I think, clear enough.
What you do in the end is up to you, but don't for heaven's sake try
to detect a browser from the server side. I could give you a hundred
reasons, but one will have to suffice for today:
http://www.mnot.net/cache_docs/ 
