- From: Anton Vayvod <avayvod@google.com>
- Date: Mon, 10 Aug 2015 23:49:12 +0100
- To: "mark a. foltz" <mfoltz@google.com>
- Cc: Jonas Sicking <jonas@sicking.cc>, "public-secondscreen@w3.org" <public-secondscreen@w3.org>
- Message-ID: <CAOjek6pfazeG=Jk8SmKUCzGLz9snT2GRBhMDKi+7BJSEJZqOMw@mail.gmail.com>
On Mon, Aug 10, 2015 at 10:15 PM, mark a. foltz <mfoltz@google.com> wrote: > > On Thu, Aug 6, 2015 at 5:43 PM, Jonas Sicking <jonas@sicking.cc> wrote: > >> Hi All, >> >> It's not clear to me if getSessions returns only sessions that were >> received by a presenting browsing context? Or if it also contains >> sessions that were created by the same page calling start() on a >> PresentationRequest? >> > > I believe the intent of the current spec is the latter, but I could see > some utility for including sessions created via start/join as well. > Just for clarity: s/latter/former > > >> >> In theory you could imagine an application which opens a page on a TV, >> and then that page on the TV opens another page on a second TV. For >> example in a conferance system where you display slides on one TV, and >> then those slides display a video on a second TV. >> >> Or a video conferance system which connects to a TV to display the >> people you are conferencing with. The TV could then try to connect to >> other TVs in order to distribute the videos of conference members to >> other TVs. >> > > Both totally make sense and would be useful to add to our use cases > document [1]. I'll add this to a pending PR. > It could also be the opposite I suppose: a page starts a presentation on a big screen and then turns itself into a presentation to allow other devices to connect to it. > > >> >> If I read the spec correctly, the "set of presentations" concept >> contains both incoming and outgoing sessions. Is there a reason for >> that? I'd prefer if it only contained incoming sessions. If there's a >> use case for accessing all outgoing sessions (are there?) we could >> always add separate API for accessing those. >> > > The ambiguity stems, in part, from trying to write one spec to define two > different implementations: one for the presenting browsing context, and > another for the controlling/initiating browsing context. > > Francois filed [2] to try to structure the spec in a way to make it easier > to keep these concepts distinct. In the meantime (if we agree with my > interpretation of the spec), we could define two sets of presentations: > > - The set of initiated presentations > - The set of running presentations > > And limit getSessions() to returning the latter. > > m. > > >> / Jonas >> >> > [1] https://github.com/w3c/presentation-api/blob/gh-pages/uc-req.md > [2] https://github.com/w3c/presentation-api/issues/93 >
Received on Monday, 10 August 2015 22:50:00 UTC