Peter Hutterer <peter.hutterer@...> writes:
> On Wed, Feb 08, 2012 at 08:30:58AM -0800, Jason Gerecke wrote:
> > Any feedback on the changes or ideas on exposing the properties?
> 
> sorry for the delay. short answer - I don't think we should merge this
> support as properties. properties are not well-suited for this, they are too
> generic and too much knowledge must reside in the client setting them.

Exactly. But the client should be the one handling the mode switching, and the
button actions in the first place. So I don't see what's wrong with exposing
it as a property. It would make it available now, and make it possible for me 
to make the mode switching work in GNOME 3.4 (I really don't want to have mode
switching enabled if I can't show the user that the mode has indeed switched).

So I'm in favour of one property per LED group, and a bitmask as to which one
is lit up (if even just an index, if we want to force 1 LED lit at all times).

> Instead, we should extend the protocol for XI 2.3 to add a LED class and the
> required requests to modify the LEDs support. This gives us
> more flexibility handling LEDs and less driver-dependent behaviour.

That's nice. It doesn't require using properties. but it does still require
the client having special knowledge of the device. It's not like one would set
the LEDs the same way on a Wacom tablet and a PS3 joypad for example.

Cheers


------------------------------------------------------------------------------
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing 
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
_______________________________________________
Linuxwacom-devel mailing list
Linuxwacom-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linuxwacom-devel

Reply via email to