You'll need to either "round-trip" back to the server after detecting
javascript and do a response.redirect (ala BrowserHawk) or redirect through
the navigator object on the client. If you're using IIS 5 (i.e. Windows
2000) you can take advantage of server.transfer or server.execute instead of
using SSI, but you'll still need to round-trip after capabilities detection.

Hope This Helps,

Jim

"Boris Feldman" <boris@feldmangroup.com> wrote in message
news:37fc183e@NEWS.DEVX.COM...
> E.H.,
>
> You can't do what you want. Server Side Includes happen on the server,
> JavaScript runs inside the browser. By the time the browser gets the HTML
> and JavaScript for the page the SSI's have already happened.
>
> Boris.
> ----------------
> Boris Feldman
> feldmangroup
> boris@feldmangroup.com
>
>
> E.H. Scholl <scholl@ipass.net> wrote in message
> news:37fb8e24@NEWS.DEVX.COM...
> >
> > I want/need to be able to detect a browser version so I can use one of

two
> > navigation schemes - one
> > with JavaScript and one that's text-only.
> >
> > I've been using JavaScript to do the Browser Detect, and then trying to

> use
> > SSI's to include one or the
> > other navigation file (so if something changes, I don't have to change

it
> > in hundreds of files,
> > just the two...) but I've been having a heck of a time.
> >
> > What I have is this (or variations thereof...):
> >
> > <SCRIPT LANGUAGE="javascript">
> > <!--
> > if (is_nav4up)
> > {
> > document.write (<!--#include file="nav.html"-->)
> > }
> > else
> > {
> > document.write(<!--#include file="navtxt.html"-->)
> > }
> > file://-->
> > </SCRIPT>
> >
> > I get the "main meat" of the page, but not the navigation.
> >
> > I've tried many variations on the theme but I can't seem to get anything
> > to work.
> >
> > I notice it also tends to get part way through the script in the

included
> > file and then starts
> > mis-interpreting code, so instead of a </SCRIPT> in one spot, I get:
> >
> > </XMP><PRE><B>&lt;<FONT SIZE=+0 COLOR="#551A8B">/SCRIPT</FONT>&gt;</B>
> >
> > This is consistent no matter what it is I'm during my debugging.
> >
> > Any clues?
> >
> > Please, if you can, also email me a response at scholl@ipass.net
> >
> > Thanks a LOT!

>
>