Tracking protection defaults on trusted and untrusted sites

13 September 2017

(I work for Mozilla. None of this is secret. None of this is official Mozilla policy. Not speaking for Mozilla here.)

Setting tracking protection defaults for a browser is hard. Some activities that the browser might detect as third-party tracking are actually third-party services such as single sign-on—so when the browser sets too high of a level of protection it can break something that the user expects to work.

Meanwhile, new research from Pagefair shows that The very large majority (81%) of respondents said they would not consent to having their behaviour tracked by companies other than the website they are visiting. A tracking protection policy that leans too far in the other direction will also fail to meet the user's expectations.

So you have to balance two kinds of complaints.

"your dumbass browser broke a site that was working before"
"your dumbass browser let that stupid site do stupid shit"

Maybe, though, if the browser can figure out which sites the user trusts, you can keep the user happy by taking a moderate tracking protection approach on the trusted sites, and a more cautious approach on less trusted sites.

Apple Intelligent Tracking Prevention allows third-party tracking by domains that the user interacts with.

If the user has not interacted with example.com in the last 30 days, example.com website data and cookies are immediately purged and continue to be purged if new data is added. However, if the user interacts with example.com as the top domain, often referred to as a first-party domain, Intelligent Tracking Prevention considers it a signal that the user is interested in the website and temporarily adjusts its behavior (More...)

But it looks like this could give large companies an advantage—if the same domain has both a service that users will visit and third-party tracking, then the company that owns it can track users even on sites that the users don't trust. Russell Brandom: Apple's new anti-tracking system will make Google and Facebook even more powerful.

It might makes more sense to set the trust level, and the browser's tracking protection defaults, based on which site the user is on. Will users want a working "Tweet® this story" button on a news site they like, and a "Log in with Google" feature on a SaaS site they use, but prefer to have third-party stuff blocked on random sites that they happen to click through to?

How should the browser calculate user trust level? Sites with bookmarks would look trusted, or sites where the user submits forms (especially something that looks like an email address). More testing is needed, and setting protection policies is still a hard problem.

Bonus link: Proposed Principles for Content Blocking.