The fingerprint is a hash of the visitors IP, User-Agent and a salt.
So the fingerprint is not unique for vistors from a corporate network behind a NAT router where multiple individuals share the same public IP and use the same corporate mandated browser. In these cases, it's back to cookies --- unless you don't really care if your analytics are faulty or not.
Or maybe just stick with cookies since it works in almost every case.
Tracking will never by accurate. Pirsch reads the Forwarded header which is set by most proxies. Apart from that, the User-Agent will be different in a lot of cases for people on the same network behind NAT.
Cookies do not work accurately either, uBlock for example shields me from Google Analytics. More and more browser start to block these cookies by default. So that's not really an argument.
You simple cannot track accurately anymore without heavily invading the visitors privacy or technical effort (like analyzing the HTTP/2 pattern the client uses/accepts).
jqpabc123|5 years ago
So the fingerprint is not unique for vistors from a corporate network behind a NAT router where multiple individuals share the same public IP and use the same corporate mandated browser. In these cases, it's back to cookies --- unless you don't really care if your analytics are faulty or not.
Or maybe just stick with cookies since it works in almost every case.
marvinblum|5 years ago
Cookies do not work accurately either, uBlock for example shields me from Google Analytics. More and more browser start to block these cookies by default. So that's not really an argument.
You simple cannot track accurately anymore without heavily invading the visitors privacy or technical effort (like analyzing the HTTP/2 pattern the client uses/accepts).
marvinblum|5 years ago
https://marvinblum.de/blog/server-side-tracking-without-cook...