Hello,
please try to note that Wayland/Weston have already refined a concept of
seat, which includes a group of input devices (keyboard, mice, ...)
associate with et defined set of one or multiple displays.
That definition will be used in a good number of cases as well.
Dominig ar Foll
Senior Software Architect
Open Source Technology Centre
Intel SSG
Le 18/09/2014 16:25, Andersson, Gunnar a écrit :
Jussi,
On Sep 18, 2014, at 12:44, Jussi Laako <[email protected]> wrote:
On 18.9.2014 13:12, Andersson, Gunnar wrote: Yes I would say display is
still better than system. You are here dealing with the scenario of one
single system (more precisely one ECU) driving multiple displays / input
peripherals. Often people will say system when they mean ECU. And if
that's not what they mean, they mean several interconnected ECU make up a
system, like the car electrical system.
The term "Seat” is used in GENIVI and if you like it you could adopt it.
We have seats already, but "display" is now the one that doesn't have a
physical seat, but spans multiple seats (holds shared resources). It's kind
of the "everybody inside the car" -seat.
Not sure I understood. Do you want to name a concept that only is used
for a "shared" display position in the car? Examples: Main screen between
driver and passenger in the front (a.k.a. the "center stack”)
Or an all-encompassing term that covers all possible display positions
(including those that would be clearly in front of a single specific physical
car seat?)
In current GENIVI proposal a user login is associated with a Seat
This is also how our setup works...
Hmm, it works the same way when using GENIVI terminology, or with a
different meaning of seat?
But I would guess there is still need for some way of knowing where a logged
in user would prefer their apps to appear. Similarly, you need to associate
input peripherals with the person operating them. It won't be like a
desktop system where you point your mouse to a window and the input goes
there.
Actually our setup works just like that, with Weston desktop shell, you can
point and click (or tap the screen) to start applications and move their
windows around within the display of the seat. Each seat has it's own mouse
cursor and they can be typing on the two keyboards simultaneously input going
to what ever window they have activated…
Yup, I have run the Tizen image. I was asking more about how a real product
would work, where I'm not envisioning a desktop look and feel. Tizen provides
flexibility clearly; I'm just looking for clarity on how that can then be used
effectively.
With IVI shell, things are also
constrained to the particular seat, both input methods and display output.
OK so that's handled then most likely. But need to talk some more about this
in contradiction to Tizen's ability of moving user apps from one display to the
other. But I'll wait for clarity on the above first so we understand each
other.
It's kind of a setup of single desktop computer that feels like being two
desktop computers (it could be more, with quad-output GPU).
In addition you can have apps as the "display" user that simultaneously span
multiple displays and input methods controlling those.
Got it. These are like "system applications" as opposed to "user applications"
right?
- Gunnar
_______________________________________________
IVI mailing list
[email protected]
https://lists.tizen.org/listinfo/ivi
