Except Do Not Track basically died in 2018. [1] The W3C abandoned trying to standardize it, and virtually no websites pay attention to it, because the legislative efforts behind it fell apart. It's merely a "feel good" HTTP header that does virtually nothing whatsoever in practice.

I also disagree that "any reasonable person" would expect the header to be used in private browsing. Safari invented private browsing in 2005, and Chrome Incognito mode launched in 2008. Do Not Track didn't even exist as a concept back then.

Incognito mode was never intended to be anti-tracking. It's only ever been intended to hide your browsing history locally, e.g. from family members.

Tracking protection has an entirely different purpose. And if you want protection from tracking, you'd presumably want it in all windows, not just incognito windows, right?

These are the important issues to be kept separate. Anti-tracking is something that should be consistent across all browser windows. It has nothing to do with Incognito. Incognito is about not saving browser history locally. Totally separate.

[1] https://en.wikipedia.org/wiki/Do_Not_Track

> Incognito is about not saving browser history locally. Totally separate.

They're totally separate now, but it's not clear that they should be, and it seems pretty clear that they're not separate in the minds of users.

I'm having a hard time imagining a scenario where a user would want to hide their local history, but are totally cool with people who don't have physical access getting access to their activities.

> I'm having a hard time imagining a scenario

Very easy.

Let's be honest, incognito mode is generally used for watching porn without worrying that it will pop up in the autocomplete box or history later.

But you log into the porn site, in incognito mode, in order to access your saved videos, subscriptions, etc. The porn site knows exactly who you are, tied to your credit card number, etc. Your ISP knows you visit the porn site. Your credit card knows you pay for it.

That's the main use case. The privacy is ONLY regarding local history. That's the only expectation there's ever been.

> That's the only expectation there's ever been.

Given that the judge in this case had different expectations, that's clearly false. That may have been the intention of the feature, but I can easily see why it may not be the expectation of users.

I gave you the scenario. It's reasonable.

The judge in this case is simply seriously misinterpreting the feature. There are always going to be some percentage of users who misunderstand a feature no matter how explicit and clearly it's been described. Even if they're a judge.

Google isn't misleading anyone here. Every time you open an Incognito page it says EXACTLY what it does and doesn't. If some users and even some judges can't read, that's their problem.

Do Not Track died the minute Microsoft killed any possible distinction that header provided. They may have backtracked on that dumb idea with Edge but the damage was done, the momentum of DNT was killed and they effectively "poisoned the well" and ensured that the largest portion of DNT headers seen were not in any way set by the user. The W3C abandoned it because it was already gone.