Open samuelweiler opened 7 years ago
Discussion of "Panopticlick for Private Browsing Mode" and possible IETF hackathon project in July 2019 happening on the PING list. https://lists.w3.org/Archives/Public/public-privacy/2019AprJun/0077.html https://lists.w3.org/Archives/Public/public-privacy/2019AprJun/0081.html
In their 14 June 2019 doc, the TAG set a goal that private browsing modes should not be detectable.
Also, PING now has a draft privacy mode doc: https://github.com/w3cping/privacy-mode/blob/master/private-browsing.md
September 2019 workshop had a consensus among browser vendors that these modes should not be detectable. Mechanisms for that include providing all APIs in these modes, with partitioned (and wiped) storage. Browsers are making incremental progress. Unclear if standards work is helpful.
There is consensus that private mode should not be detectable (see below), and the work to make that happen is in individual specs. The open question is what clean-up work needs to happen in existing APIs to achieve that goal. Leaving this item open for now.
As mentioned above, the W3C TAG Observations on Private Browsing Modes says:
Spec authors and browser vendors should work towards achieving private browsing mode work in a way indistinguishable for i sites from the normal mode, to respect the user’s users’ privacy in choosing it.
In addition, the TAG's Web Platform Design Principles says:
Make sure that your feature doesn’t give authors a way to detect private browsing mode.
As sites begin to turn away browsers in incognito or private mode, will browsers be forced to run in modes that provide less privacy? Can we make incognito or private mode less identifiable?
Example of the day: https://arstechnica.com/information-technology/2017/05/boston-globe-website-no-longer-lets-you-read-articles-in-private-mode/
Some sample detection code: https://gist.github.com/cou929/7973956