| Thanks, some helpful insights here. (It sounds like the summary is, “set it up and don't fiddle with it too much.”) I agree that the neovim community is great and neovim is of course very hackable, although there's always room to improve: an API for terminal UI would help a lot when writing neovim plugins and make everything feel more consistent, for example. > I don’t know what process you took, but properly learning the tool from the ground up and minimally adding useful plugins as you go should not take much time at all. I built a neovim config from scratch, used built-in docs, community support, and well-known extensions when needed. I wrote small Lua plugins when I couldn't find extensions or workflows that worked for me. And I had fun doing it! I got something good in a weekend, something very decent in a week or two, but still had a lot of gaps between what I had and what I wanted (session save, restore and quick switching; a Magit-like staging and forge management UI for GitHub CLI -driven PR management; invoking and managing long-running terminal processes from the editor; find/replace in large numbers of files across big projects with something more intuitive than a quickfix list; linting with something lighter-weight and easier to configure alongside LSP servers than ale; quicklists for LSP issues; a simple test runner with quicklist for errors; I can go on…). After trying out extensions for some of those things but finding nothing that fit the bill, I played with Emacs and had a pretty similar experience there (weeks of work just to reach parity with VSC+a few extensions). I concluded I was wasting my time — it was pretty clear I was after an IDE more than a text editor at this point. VS Code seemed a nice in-between option. There are things I miss from vim/Emacs (speed, keyboard-driven workflow, getting to enjoy and celebrate editors with a community of likeminded weirdos), which is why I tend to get sucked in by these update announcements and discussions. But for the most part VSC is fine. |