https://bugs.kde.org/show_bug.cgi?id=3D346837 --- Comment #19 from Thomas L=C3=BCbking --- (In reply to Ferry from comment #17) > I still don't get the logic of disabling the panel when alt-tabbing to th= e desktop. a) the show desktop thing doesn't belong there in the first place. The tabb= ox would activate the window. I assume that since that would usually (often) implitly raise the window, but the desktop cannot be raised, the desktop showing state is toggled instead (what now actually raises the desktop) > All other windows that are reached in this way have a panel. See above. The panels (usually) reside in the dock layer which is above the normal layer (where most windows hang around). Activating a window will usu= ally raise a window in its layer, but raising the desktop in the desktop layer is NOOP - there's no naturally consistent behavior: the entire feature is an exception by its nature. Thus the main aspect of the RR was to query into which layer the desktop sh= ould be moved, ie. between normal windows and panels, along panels or above pane= ls. Status quo from HIG team is: "The thing is that we have not defined what the dashboard is supposed to be used for. If it is only for glancing at or quickly interacting with desktop widgets, then panels would only be distracting. If it is for interacting wi= th the workspace in general, then of course panels should be accessible. One argument against showing panels would be that clicking on a window in the t= ask switcher would break the mode" Because of that I made the implementation fine grainable: panels can simply opt-in to remain on top by tagging themself transient for= the desktop (what is the consistent flag) (In reply to Ferry from comment #18) > @Thomas Anyways I followed your advise and disabled the aperture effect a= nd > set a rule to enforce docks keep above. >=20 > Still alt-tab behaves odd: tabbing to the desktop it appears but > semi-tranparent, some window is vaguely shown behind it. I think this is due to plasmashell adjusting the opacity during the showdes= ktop mode (to be more dashboard-a-like) - at least KWin does not alter the deskt= op opacity in this case. > Should this be fixed by your patch Yes. The Cover- and FlipSwitch tabboxes are simply broken w/o the patch (I simply forgot them and they use a different code path than the other tabbox= es. Sorry again for that) --=20 You are receiving this mail because: You are watching all bug changes.=