Hello Matthias, I agree the two reply from Jonas, and Olivier.
I just add to Olivier comment, that even we tried to remove MWM hints (i.e. they do not belong any up-to-date standards), they are still used by up-to-date software and implemented by popular WMs. This use case show that (1) those hints are useful and (2) they do not have noticeable draw back. My primary complain is that draw states should be merged with the previously defined window/surface states, because by definition a draw state is a state for a window, just like the state activated for example. Best regards. On 07/06/2016 10:00, Olivier Fourdan wrote: > Hi Jonas, > > ----- Original Message ----- >> On Mon, Jun 06, 2016 at 09:50:19AM -0400, Matthias Clasen wrote: >>> To me, 'draw states' sounds a lot like 'mwm hints' - which would be a >>> 180 degree reversal from semantic states back to presentation states. >>> Do we really want to go down that road ? >> >> I'm not too familiar with MWM hints, but this does not affect the >> semantic states, nor will it replace any of them. They'll all be >> optional, and they are mostly meant to be used by DE:s that wants to >> opt-in on non-default behaviours. > > Yeah, but still not unlike EMWH vs. MWM hints, in X11 you can have a EWMH > semantics and apply additional presentation tweaks using MWM, that's how gtk+ > achieves CSD in X11 for example, it uses MWM hints to tell the WM not to > decorate the window. > > Cheers, > Olivier > _______________________________________________ > wayland-devel mailing list > [email protected] > https://lists.freedesktop.org/mailman/listinfo/wayland-devel > _______________________________________________ wayland-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/wayland-devel
