> Anyone out there have a systematic method for identifying Robots hits on WWW > pages? > > So far, all I go on is the User Agent string in the log files.
I wrote something many, many moons ago called BotWatch. This used a variety of approaches, including the presence of robots.txt fetches, originating URLs and User-Agent strings to attempt to identify robots. You can find it at http://www.tardis.ed.ac.uk/home/sxw/robots/botwatch.html The configuration file hasn't been updated in a long time. Your mileage may vary. Cheers, Simon.