The incidental anonimization may be a nice feature, but there is existing free servies that provide this function without serving as a Man-in-the-Middle.

Finally, the idea that this information could ever be anonymized seems silly – I believe Google claims this concerning the location data they are gathering – Seems any forensics guy worth his salt could attach this type of file to an individual with minimal effort if they have some basic work and home location info to begin with.

One answer is use two factor authentication (using something you have, such as a user name, and something you know secretly such as a PIN) or multi-factor authentication which would be two factor authentication plus adding “something one is” (such as biometrics).

None of these methods of authentication will be useful in the long run if they are not stored securely, which is likely to happen with ubiquity of use. This requires different types of authentication for access to different services. This solution, however, creates complexity with having to remember multiple user names and passwords. Lots of work going on a finding and marketing a better solution.

I agree with you. It seems as though the FTC is seeking a middle ground between having “no tracking” as the default (which would be very disruptive), consumer opt out through a “Do Not Track” list, and what you suggest — dealing with the harm of inapparopriate data usage. The “Do Not Track” list does, however, provide a means to address the concerns of those who feel harmed by the online data collection itself. The FTC acknowledges in its report that it had been too narrow in the past in looking at the privacy harms and is now considering being creeped out (my words) and reputation damage into its calculus. I’m wondering how today’s brouhaha over “history sniffing” would fit with “Do No Track.”