privacy-model

> It should not be technically possible to track a user across sites. Chrome agrees with you on that: https://github.com/michaelkleber/privacy-model The replacements they are talking about attempt to maintain the economic benefits without cross-site tracking.

> There are a handful of companies that can track you across multiple websites, so any real solution has to start by enumerating and addressing those companies. I think it's more like hundreds than handfuls. But they're all connecting your behavior across sites using the same few techniques: * Explicit methods: cookies, link decoration, and other browser-supported ways of adding entropy. Browsers are working on removing these, but if they move too aggressively here then adtech just moves to: * Fingerprinting: using existing browser entropy. Generally worse than explicit methods because the user doesn't have control (ex: shared fingerprint between successive private browsing sessions). Browsers are also working on reducing this, see the article, but it's very hard because the number of techniques is large and they generally use features us

I think one thing overlooked is that Google's supposed rationale for providing this service is effectively because they suppose that their plans to "more aggressively block fingerprinting" will leave advertisers with no other choice. Personally I find this highly unlikely. The very nature of fingerprinting makes it difficult to prevent without some pretty severe compromises, and it wouldn't surprise me if advertisers enter an arms race with Google instead of caving in. Of course the economic strategy for Google to avoid that is to reduce the price/limits of this service to so low that this isn't worth doing, but if the effort Google puts into preventing fingerprinting is insufficent then this would lead to a "worst of both worlds" situation - Google gets to pseudonymously track the internet with their "privacy sandbox", and nefarious trac

There's one more repo > https://github.com/michaelkleber/privacy-model I thought it was odd and annoying that the 5 proposals are in different place, and it took me so many clicks to get to them from the blog post.

Interesting to me is that Google's Privacy Model suggestions of "Identity is partitioned by First Party Site" and "Third Parties can be allowed access to a first-party identity"[1] come off as more coherent and effective than Apple's ML based Tracking Prevention policies[2]. Issue is Google's third suggestion, "A per-first-party identity can only be associated with small amounts of cross-site information", is clearly advertiser-focused, not privacy-focused. I hope all can rally around the partitioning concepts and leave the others behind. [1]https://github.com/michaelkleber/privacy-model [2]https://webkit.org/blog/8828/intelligent-tracking-prevention...