What about CSS extensions like .webkit? Same thing really. AFAIK the standard way to do browser-detection is to use the user-agent string, which your browser se

What about CSS extensions like .webkit? Same thing really. AFAIK the standard way to do browser-detection is to use the user-agent string, which your browser sends with every request it makes (but can be overridden) anyway, so them programming an if(browser == firefox) statement does nothing whatsoever for their knowledge of you. They already get told what browser you are running. And I am no webdev, but a quick google seems to show that it is possible to do browser detection completely client-side anyway with Javascript, so even if user-agent strings did not exist they could detect your browser without any kind of fingering. Granted, if you wanted to do so perhaps you could use this to circumnavigate people hiding their user-agent by having the generated link be a hosted redirect link, that way you know that if someone clicked redirect-firefox.theirdomain.org they were using firefox even if they hid their user-agent string. But it's not really fair to assume that in this case. (and i have to wonder what the value of an ip-adress/browser pair is).