> it offers almost no protection against the websites you may browse to (how could it?) > All you have to do is combine the "anonymized" data with data from oth

> it offers almost no protection against the websites you may browse to (how could it?) > All you have to do is combine the "anonymized" data with data from other sources I'm struggling to understand why. Can you give a very specific example of how I, using Tor browser to browse web, can still be tracked as an individual? What are "other sources"? Each session of Tor browser has a randomized IP and user agent, and moreover if JavaScript requests for instance the "screen width" of the device, it gives some random numbers. Letterboxing, anti-fingerprinting in action: https://support.torproject.org/tbb/maximized-torbrowser-window/ https://support.torproject.org/tbb/maximized-torbrowser-wind... > True anonymization is possible: it requires the collector to just keep general aggregate statistics and to immediately delete Most businesses store all that "raw" data (pageview and other events) to process it later. It's reasonable – hard to build and unify it otherwise, but at the end of the day it gets aggregated. To add, they have their own data retention policy. So the best and minimally required one can do IMO is to use aka "incognito mode" when browsing, if it matters.