- From: Anton Vayvod <avayvod@google.com>
- Date: Tue, 16 Sep 2014 14:16:14 +0100
- To: "Kostiainen, Anssi" <anssi.kostiainen@intel.com>
- Cc: "public-webscreens@w3.org" <public-webscreens@w3.org>
- Message-ID: <CAOjek6ou8rMzs0wH9KAA=kg2UY=sttuXVZ0JqMkZxT5HmwzJSA@mail.gmail.com>
On Tue, Sep 16, 2014 at 1:23 PM, Kostiainen, Anssi < anssi.kostiainen@intel.com> wrote: > Hi Anton, > > On 15 Sep 2014, at 20:49, Anton Vayvod <avayvod@google.com> wrote: > > > We don't need to distinguish between when the user explicitly dismissed > the dialog or simply ignores it, > > Right. From the web content perspective the end result is the same (no > access granted) whether the user actively dismisses the dialog or just > ignores it. No need to let the web content know the reason. > > This is aligned with the principle of data minimization [1]. > > > but probably need to distinguish between when there was nothing selected > and something was selected but we couldn't connect. > > I’d say nothing selected is analogous to dismiss or ignore. If you choose > no screen, you are practically dismissing the dialog, right? > Right, I used "nothing selected" as another way to phrase "dismissed/ignored". > If we can figure out we cannot connect to a particular screen before > starting the session, I’d say it is a similar scenario, since technically > there’s no screen if one cannot connect to it. > So in all cases the promise would become rejected without specifying the reason, right? > > OTOH, if the error happens after the session has been established we > should provide means to detect that and inform the controlling page. > > Btw. at W3C there’s a Privacy Interest Group (PING) [1] that consists of > privacy folks who are able to conduct privacy reviews on our work when we > move on to the standards track in the WG. > > Thanks, > > -Anssi > > [1] http://www.w3.org/2001/tag/doc/APIMinimization.html > [2] http://www.w3.org/Privacy/
Received on Tuesday, 16 September 2014 13:17:02 UTC