/tech/ - Soyence and Technology

randomization only if you can do it without fucking up

randomization is probably the better way to do it in existing browsers
why not just make a browser that literally cannot be fingerprinted, doe (because no differences in the environment it runs in should affect what the browser does to start with)

Standardization.

I work on browser fingerprinting as part of my job and randomization can be trivially detected by cross referencing reported values against other signals and even turned into part of the fingerprint. Standardized setups for it like Brave aren't too bad, easier to detect than custom setups but doesn't make you too unique, but rolling your own setups with a hodgepodge of extensions and user.js tweaks can easily single you out as one in a million since nobody else is spoofing the exact same set of signals as you.

The Tor Project is aware of this and that's why the Tor browser uses standardization instead of randomization. Their implementation pretty effective too, you can't extract much entropy at the browser level from Tor users other than their operating system and CPU arch.

>>35281
that will only work if everyone uses a standardized browser which wont happen, otherwise you wont be able to use many websites due to disabled features or they will just see that its you because nobody else is using librewolf in your ip address area.
you just need to pretend to be a different browser each session to blend in and spoof the entropy. and perhaps make extensions undetectable somehow by using css or dom namespaces.
so basically randomization of standards.

>>35219 (OP)
standarization on apple devices, random on everything else.

bump

>>35243
is that possible though?