10:39 <kchibisov> dnkl: do you happen to know if anyone proposed escape to ask terminal that it should disable/enable IME input? I'm pretty sure there should be such a thing...

10:41 <dnkl> kchibisov: someone did suggest that to me a long time ago, and we have it in foot. But I haven't tried to raise any awareness about it, and thus I don't think many people know about it.

10:42 <kchibisov> I guess it's some sort of CSI?

10:42 <kchibisov> Or is it an OSC with states?

10:42 <dnkl> it's a private mode flag, 737769

10:43 <dnkl> So, CSI ? 737769 h/l

10:43 <kchibisov> yeah, I understand.

10:44 <kchibisov> Is it used by any real app?

10:45 <kchibisov> Or someone just suggested to add a mode to disable/enable IME?

10:46 <kchibisov> Like if it's completely arbitrary it could be really hard to push it, but pushing anything in terminal ecosystem is nearly impossible.

10:47 <dnkl> it's not used directly by any applications, that I'm aware of

10:48 <dnkl> but some users have configured vim to switch it on/off when changing mode

10:48 <kchibisov> Well yeah, the primary use case is modal editors.

10:49 <dnkl> https://codeberg.org/dnkl/foot/wiki#user-content-enabling-disabling-ime-when-entering-leaving-insert-mode-in-vim

10:52 <kchibisov> I guess you're not interested in pushing escape forward?

10:57 <dnkl> there hasn't been much interest in it, as far as I know. I don't use it myself , and would therefore not be very good at argument for it.

10:58 <dnkl> but I'll be happy to replace what we have in foot, if something else becomes a standard

10:58 <kchibisov> Well, the usual way to write plugins for editors doing dbus to control IME.

10:59 <kchibisov> which is a 'not great', because it could mess things.

10:59 <kchibisov> e.g. https://github.com/vim-scripts/fcitx.vim

11:00 <kchibisov> or https://github.com/h-hg/fcitx.nvim

11:00 <kchibisov> and it's just fcitx, there're plenty of things like that around.

11:01 <kchibisov> And with IMEs without controlling socket obviously none of that will work.

11:07 <dnkl> ah, when I said "something else", I was referring to escape sequences. Not interested in adding anything else... That includes the IME backend as well; I'll support whatever Wayland protocol we finalize on, but will not support dbus or other controlling mechanisms

11:08 <kchibisov> yeah, I've understood, I just gave a 'reason' why pushing escape sequences here is good.

11:09 <kchibisov> because existing solutions to that rely on dbus, which obviously can't work over ssh, etc.

11:09 <kchibisov> but yeah, if you're not interested to push your escape forward, I understand.

11:11 <dnkl> is there even a good place to "push" new things? terminal-wg is useless imho. I could certainly document the sequence and talk about it in social media. But I'm not going to start opening feature requests on a bunch of terminal emulator issue trackers...

11:11 <dnkl> document it *more*

11:13 <kchibisov> Yeah, certain folks made everything to destroy terminal-wg.

11:14 <kchibisov> I remember in the past that OSC 8 was pushed basically like you've described :p

11:20 <kchibisov> Probably documenting it in accessable format and having a dedicated link to it could help.

11:23 <dnkl> I'll see what I can do about increasing visibility

11:26 <kchibisov> It could be nice if it could be indexed by search engines though. Like hyperlink escape is easily discoverable for example due to that.