The 'Do Not Track' proposal is a travesty, because it depends on the participation of the violator. Furthermore, it completely invalidates itself, because the only method of compliance is through tracking!
Privacy control needs to be in the hands of the user, before a tracking site is even resolved via DNS. A combination of extensions like RequestPolicy and BetterPrivacy go a long way towards achieving this (but new tracking techniques are always being developed). Browsers need to have some of this functionality built-in, instead of trying to protect the sheep by supporting a protocol developed by wolves.
From what I understand, the do not track header is added to ever single browser request when the option is enabled. With virtually nobody actually obeying the do not track header, all it is for the time being is a waste of bandwidth. Certainly not a lot, but it probably adds up.
Yes, in theory a decentralized system for indicating privacy settings is a great idea, but the large scale offenders, the only people whose networks span a great enough breadth in order to collect useful information on people, happen to be a pretty small number. A quasi centralized system with opt out cookies has less of an overhead, less ambiguity and most importantly of all, already exists and works.
Plus, tracking on the web is not a binary thing. Maybe for a simple blog or ad, it might be less ambiguous. But are you still allowed to be able to know how many visitors you have on a given day? If your on a web app, does do not track mean you can't save any documents because they constitute a digital trail of what you've done?
A system to fight tracking needs to be nuanced. Something like that disconnect extension, where it's not simply on or off.
If you think about it, google doesn't really have any real reason to hinder the adoption of DoNotTrack to preserve it's data mining business. A do not track header standard would hinder competitors equally, whereas the opt out cookies would single out google. If i were in that position, I would choose the option that hinders my competitors as well.
uninformed rant: Do-not-track headers feel like a sham. When you access content and services funded by advertising, you opt in to advertising. I think a service would be fine to ignore the header, because the user has presumably already implied permission to be tracked by accessing the service.
Consumers want 'cheap' far more than they want privacy. They sign up for retail credit cards to save $20 on a single order. They scan their grocery keytag every time they shop. They give their mailing address for a free t-shirt. Maybe you guys don't do this, but I still find myself irrationally hemming and hawing over a small purchase. To fix this, the way we think about money as a culture would have to shift.
Privacy awareness is a good thing coming out of this, and consumers understanding privacy can lead to better choices and more competition, but in the short term I don't see much practical impact.
i think its interesting what might happen if this movement gained enough traction to hurt google's adwords revenue. How much is a tracking cookie worth to them? would they require opting-in to tracking to use google services? is no-track enough of a competitive distinction that mainstream consumers will pay money for no-track services? is it enough money for a no-track business to compete, and will they find unexplored ways to make money?
Privacy control needs to be in the hands of the user, before a tracking site is even resolved via DNS. A combination of extensions like RequestPolicy and BetterPrivacy go a long way towards achieving this (but new tracking techniques are always being developed). Browsers need to have some of this functionality built-in, instead of trying to protect the sheep by supporting a protocol developed by wolves.