My Emacs configuration https://git.cjennings.net/dotemacs/atom?h=main 2026-07-10T05:38:21+00:00 2026-07-10T05:38:21+00:00 Craig Jennings c@cjennings.net 2026-07-10T05:38:21+00:00 urn:sha1:02c2e0ee8e00869681595bd7b80a98219563b39f The mechanical half of the "?" convention task. I probed ten modes from their live keymaps rather than from source, because a package's define-key calls don't always agree with what lands in the map. Four modes have both ? and H free, so nov, eat, and the two elfeed modes can adopt the convention without displacing anything. Three bind ? to something small that adopting it would displace. mu4e-headers is the one real conflict: its ? marks mail unread, a state-mutating command in the marking family. magit already binds ? to its own dispatch. The convention isn't new, it's magit's habit generalized. I ran the probe in a throwaway batch Emacs rather than the daemon, since probing calibredb and mu4e in a live session would force-load them. mu4e isn't an ELPA package and needed its site-lisp path.