Re: Abysmal state of GTK build

[Eli Zaretskii]

> From: Po Lu <luangruo@yahoo.com>
> Cc: Gregory Heytings <gregory@heytings.org>,  emacs-devel@gnu.org
> Date: Mon, 22 Aug 2022 21:13:55 +0800
> 
> Lars Ingebrigtsen <larsi@gnus.org> writes:
> 
> > I feel like I'm repeating myself here a lot now.  So to recap:
> >
> > 1) The crash under a Gtk build should be fixed -- in any case, no matter
> > what else we do.  If that means disabling XInput 2 under Gtk, that's fine.
> 
> It's not fine.  Because once that happens, people will stop using the
> XInput 2 build (since it will no longer be the default)

I don't understand: don't the non-GTK builds also support XInput2?  If
so, and if we make one of those non-GTK builds the default, people
will still have an XInput2 build, just not a GTK one.  Or what am I
missing?

> Which is why XInput 2 should stay on.  GTK -- off.

If the default, non-GTK, build supports XInput2, then why is it a
problem to have the GTK build use XInput2 only as an opt-in feature?