04:41 <leon-p> have fun!

13:41 <tiosgz> whoops, i hope if has notifications off. didn't really expect today would be another of my think-aloud days...

13:56 <leon-p> tiosgz: regarding your comment: focus-follows-cursor will be WM policy. Both pointer_enter and pointer_motion could trigger it and some WMs may also want to check if a focusable thing is under the cursor after each set_position

13:56 <leon-p> or am I missing your point?

13:58 <leon-p> hmm... I am not too happy with the pointer_motion event, because it makes the update event, which is already where all interesting logic sits, into a hot path, especially for high poll rate pointing devices. But I also see no alternative for the "always" policy of focus-follows-cursor we offer right now

14:09 <tiosgz> leon-p: my point is that the pointer enters the window and is moved, but the former example makes no distinction between the window being moved under the pointer (which doesn't trigger ffc in normal mode)...

14:09 <tiosgz> ...and the pointer entering it just after the window was moved (which should trigger ffc)

14:10 <vyivel> something something pointer warp https://gitlab.freedesktop.org/wayland/wayland/-/merge_requests/340

14:35 <tiosgz> aahh, now i'm back to thinking about actually simultaneous events. expect me getting totally confused and later coming back with the conclusion that the protocol is actually indeed okay

15:41 <tiosgz> so either i've finally wrapped my head around it and the proto is ok (in which case i've edited the comment in the PR), or i'm irreversibly confused and need to find less difficult things to think about

15:49 <leon-p> now I for my part have accepted that both the thing I am trying to understand is OK /and/ that I am confused and probably should just go out with my friends instead

15:52 <leon-p> what was keeping me up the last few days is thinking about which things really need double buffering on the WM side. I decided to just double buffer everything for now and later simplify the things that don't need it

19:47 <Guest71> By any chance: is anyone able to get screensharing to work on FreeBSD? Everything else works great for me.

19:47 <Guest71> I have this line in my init file and it works fine on Linux:

19:47 <Guest71> riverctl spawn 'dbus-run-session pipewire'