Re: Notes on two recent fingerprinting papers

> On Jul 5, 2016, at 3:20 PM, Greg Norcie <gnorcie@cdt.org> wrote:
> 
> So do you think it's fair to say we've made some forward progress on reducing fingerpintability?

I don't think we have enough data to know whether fingerprintability on the whole has increased or decreased for all users and use cases over time. I do think we have evidence that some features that were being used for fingerprinting have become less effective for many users (Flash/fonts, plugins, window sizes on mobile devices), which I would consider forward progress. However, new features have increased fingerprintability (canvas, in particular, but also apparently User-Agent in some mobile devices).

> I often see fingerprinting presented as a threat that will always be at it's current effectiveness (or become more effective as browser features are added), so anything that challenges that assumption is very interesting.

I think there's definitely evidence here that not all features used for fingerprinting will remain at their current effectiveness, which might be part of that broader assumption. (Indeed, we should have noticed that years ago with the rise of popular mobile devices with a common screen size, auto-updating operating system and limited plugin support.)

—Nick