Re: Universal Stylus

On Fri, Aug 26, 2016 at 2:24 PM, Fleck, Dave <Dave.Fleck@wacom.com> wrote:

> > On 26/08/2016 17:35, Rick Byers wrote:
> >> 3. The property can't be used to uniquely identify a user - for web
> >>    privacy reasons (i.e. serial numbers can't be exposed to the web as
> >>    a result).
> >
> > Unfortunately, I think the properties listed (Transducer Serial Number,
> Preferred Color,  Preferred Line Width, Preferred Line Style) may fall foul
> of this (particularly the TSN, the others could be used to at least
> infer/narrow down users). I think one thing that *would* be nice would be
> to ensure that the same transducer always gets the same pointerId across
> interactions (not the same as the TSN, but unique for the duration of the
> browser/tab session)
> >
>
> Perhaps stating the obvious - but - the idea is that I can upload into my
> stylus my preferred color (I like purple because it’s a corporate color)
> and width (I tend to write small so I like a fine tip pen) and where-ever
> (any application) whenever (even tomorrow) I will get those settings by
> default.
>

I'm personally much more concerned about unique IDs (like TSN) than just
properties that can be used to narrow down a user.  The latter is
effectively unsolvable and comes up when almost any new API is added.
Chromium has a blanket policy (I can dig up the link) of not worrying about
"this new API provides a bit more details which could be used to
fingerprint a user" - as such arguments can basically be used to stop all
progress.  Eg. just the fact that the user is using a stylus at all is very
powerful...