You should have a working http server (i.e. Apache) configuration working, with CGI support enabled already.

After you get that done, just compile the program (make) and you will get a CGI named "jdtracker". Create a new directory under your /cgi-bin tree (for organization sake) and place the CGI there. Make sure it has mode 755 or similar (execution rights to user nobody). Also, chown nobody:nobody that directory and chmod 775 it so that the log file can be created.

For instance, if you have a default RedHat Apache tree and want to track some account in GeoCities, you could do it like this:

	cd /home/httpd/cgi-bin
	mkdir geocities
	cd geocities
	chown nobody:nobody .
	chmod 775 .
	cp /the/path/you/used/jdtracker .
	chmod 755 jdtracker

Now all you have to do is place the JavaScript code inside your html tree.

	cd /home/httpd/html
	mkdir geocities
	cd geocites
	cp /the/path/you/used/jdtracker.js .

Remember to edit this file (jdtracker.js) so it points to the path you chose for the CGI. If your host is called "www.lame.org", your would put "http://www.lame.org/cgi-bin/geocities/jdtracker" in the SRC part.

Now just place some code like:

	<SCRIPT LANGUAGE=JavaScript SRC=http://www.lame.org/geocities/jdtracker.js></SCRIPT>

on the pages you wish to track.

That should get you up and running. Of course, you will have an unresolved log file (don't worry, that's the right way to do it), so you should check out jdresolve at http://www.jdrowell.com/Linux/Projects/jdresolve to get you the fqdn for the hosts so you can generate meaningfull statistics by domain/country/etc. Also, try Webalizer (http://www.webalizer.org) which is my preferred statistics package.

