Andreas Röhler <andreas.roeh...@easy-emacs.de> writes: > On 13.10.21 09:34, Kévin Le Gouguec wrote: >> >> "Modern" did not factor in; the goal was to have RET and C-j behave >> consistently in all major modes. > > That does not deliver an argument to change the meaning of RET.
If there is a compelling argument that justifies RET and C-j behaving differently in Org wrt other major modes, I haven't heard it yet. > BTW the costs of such changes are terribly underestimated in Emacs. AFAICT, the costs of user-facing changes are regularly discussed on the Emacs development lists, and different developers have different opinions on how underestimated they are. In the specific case of RET and C-j, I'd argue (and Org maintainers seem to have agreed) that the long-term benefits of Org falling in line with other modes outweigh the short-term costs of annoying long-time users, especially since they are offered ways to bring back the previous behaviour (outlined in ORG-NEWS). And in the specific case of org-adapt-indentation, again, changing the default to nil was the result of extensive discussion on emacs-orgmode, where several users explicitly stated that they did not want text to be indented (neither with RET, C-j, TAB, nor org-indent-line) and never realized that org-adapt-indentation was t because Org ignored electric-indent-mode before 9.4. >> Since electric-indent-mode is enabled globally in Emacs, > > Which IMO was another mistake. > > Preferring a clean editor, which does fancy things only if enabled. There are plenty of things Emacs does by default that I personally find unhelpful; fortunately I can just disable them. And as long as release notes point out changes in default behaviour (and how to revert them), I'm happy with new releases enabling new features. YMMV 🤷
