9 ms·
This is not a user first approach. Ad tech heavily relies on IP address for fingerprinting. Chrome will eventually implement IP Protection for incognito users.
by fumar 2y ago
This is not a user first approach. Ad tech heavily relies on IP address for fingerprinting. Chrome will eventually implement IP Protection for incognito users. I believe Chrome originally planned to roll out IP protection for all users. https://github.com/GoogleChrome/ip-protection?tab=readme-ov-file#availability-of-the-ip-protection-feature https://github.com/GoogleChrome/ip-protection?tab=readme-ov-...
- blackeyeblitzar 2y agoHow does IP address fingerprinting work in IP v6? I am not an expert but as I understand there are just too many addresses to establish patterns.
- nisa 2y agoIPv6 is more like /64 is an IP. Most ISP assign a /56 or /60 to the customer, you can just disregard all the other bits and basically have something like an IPv4 equivalent.
- topranks 2y agoTrue. It’s kind of the reverse of IPv4. In v4 you could have a million people behind one address. In v6 you at maximum have one person behind an address, but that one person may actually be using a million addresses.
- prasadjoglekar 2y agoExactly. IP is but one element of the fingerprint. Browser, OS, language etc. all make it quite possible to reassociate a user to a different IP v4 or 6; just takes a bit of time and processing.
- unethical_ban 2y agoIt's actually easier to track. Less NAT, ISP leases can get more specific,misconfigured privacy options... When I did a geoip lookup on my v4 address last year, it had me about 1.5 miles away from my apartment. When I did it on v6, it was within a block. Using a VPN and NAT will still be useful for privacy with ipv6.
- 1vuio0pswjnm7 2y agoIP protection from Google's ad tech competitors. (As a result of Google's antitrust infractions ad tech is going to become more competitive.) No IP protection from Google and its CDN partners.
- Thorrez 2y ago>Hi, connections through the proxies are encrypted multiple times to prevent Google from being able to access browsing data. In particular, the connection client-website is end-to-end encrypted, and so are the client-proxyA and client-proxyB connections. Because of this, the proxyA (operated by Google) will only be able to see the client IP address but won't be able to know which website is visited. The proxyB (operated by a partner) will be able to see the hostname of the website, but it won't know which client IP is accessing it. Neither proxy can see the URL nor the data due to the end-to-end encryption. With this design, no one - not even Google - can see who visited what website. Regarding log retention for the very limited information that we do have, let me confirm things internally and circle back. https://github.com/GoogleChrome/ip-protection/issues/34 https://github.com/GoogleChrome/ip-protection/issues/34 Disclosure: I work at Google, but not on IP protection.