should display hotplug be the only trigger for going in and out of windowed mode

[kevin gunn]

Currently on nexus4 vivid+o + silo0 we have 2 triggers for going in and out of windowed mode use
1) BT mouse
2) hotplug slimport

Should we continue with this? or limit the windowed mode to only hotplug of the hdmi ?
certainly the windowed mode feels odd when on the local phone screen( as in the case of BT mouse only)
using the local screen for windowed mode causes issues like bug 1438465 & bug 1473572 & bug 1488852

UX Solution:
1. If a monitor and a mouse is connected then windowed mode should be enabled.
2. If a pointer device (a mouse) is connected then windowed mode should be enabled.
3. If a monitor is connected with no pointer device connected then windowed mode should not be enabled as a user will not be able to do much without a pointer device.

[Benjamin Keyser (bjkeyser)]

The trigger for windowed mode needs to be the mouse because in larger tablets, it's a conceivable 'real' configuration... especially as screens get more dense.

Having said that, the issues with the bugs mentioned above happen when we don't have fixed minimums on 40 and 50 GU panels... which we are supposed to have.

[kevin gunn (kgunn72)]

@unity8 team, wonder if we could combine the qtinputinfo triggers for mouse with looking at screen size/pix density

[Michael Zanetti (mzanetti)]

Sure, combining information about input and output devices would be possible to determine the current mode.

The other bugs linked in the description are bugs that need to be fixed anyways. Not entirely related to the small screen. Sure, it is easier to trigger them, the smaller the screen is, but you can trigger them just the same on large screen.

[kevin gunn (kgunn72)]

we've almost have the correct implementation, we just need to address #3 from the description

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.