Well, sometimes the user expects the focus to shift, like the example mentioned when you click a link in one application and your browser in the background should come into focus when it opens a new tab.
Some applications just decide to steal focus when the user does not expect it, and that is the problem.
somat|6 months ago
Now there may be a legitimate case to steal focus, but I am unable to think of one at the moments and your click link example fails to convince me.
I also sort of hate modal dialogs/windows. I think modals are in general an indicator of lazy/bad design. That being said, there are legitimate cases for modals. but "stop the world and handle me" should be a last resort not the first.
TheAceOfHearts|6 months ago
In purely software spaces there's usually nothing that urgent happening. If something is really urgent it's usually going to be because it carries significant real-world consequences. Emergency alerts on smartphones are a good example of taking immediate priority over all other activities, especially if it's something like a warning of incoming missiles.
Dylan16807|6 months ago
zamadatix|6 months ago
BobaFloutist|6 months ago
watusername|6 months ago
nottorp|6 months ago
Can't even say "browsers are allowed to grab focus" because they'll grab it for a stupid window telling you to update the browser or what new features no one cares about they introduced.
I'd prefer to have to switch focus to the browser manually than have the stupid ubuntu update manager steal focus when i'm typing in the terminal...
elehack|6 months ago
It isn't perfect, because there's no way to know that the browser isn't using the token to request focus for something else, but maintaining and validating chain of custody for focus across applications is exactly the problem it looks like they are working on solving.
Ukv|6 months ago
To my understanding, the approach described in the article is that the currently active program requests a token and then passes that along to the program that it wants to take focus. Compositor can also check what triggered the request (mouse click? global keybind?) to decide if the request is legitimate.
That seems reasonable to me, opposed to requiring the user to switch over to a new window every time they `right click -> show in file browser` a file in their IDE, or after they press a hotkey to open a screenshot tool, or so on.