Vi has text input modality, unlike essentially, well, anything. It's an entirely different model of operation which it shares with no current software and hasn't for a quarter century.
It's strange to be talking about the UI details of vi/emacs (both, by current standards, about equally super-weird) when the article itself is a kind of perfect vignette of the organizational dysfunction of the group that maintains emacs.
'press i and then basically everything works as you'd expect' is pretty simple; it's got a significantly-smaller barrier to entry than Emacs, and breaks fewer core-assumptions. Is that agreeable to you?
The thread on the mailing list was basically guesses as to why Emacs has lost popularity & methods of regaining it, and the lwn article is primarily focused on just relaying it. The thread itself doesn't point to all that much dysfunction, in my opinion.
'press i and then basically everything works as you'd expect' is pretty simple; it's got a significantly-smaller barrier to entry than Emacs, and breaks fewer core-assumptions. Is that agreeable to you?
It's not. Maybe we're talking about completely different things because I'm not following this at all, just like the other responder downthread. For decades, GUI text entry has been modeless. Larry Tessler one of the GUI's pioneers and prophets (PBUH) had a 'NO MODES' license plate and the idea remains a central theme in UI design. Vi doesn't do that. It does this:
I don't mean to debate the merits of this design but in terms of 'breaking core assumptions, vi 'breaks the core assumption' of typing text into a computer. It's not, per Winnfield, the same ball park, same league or even the same sport.
> 'press i and then basically everything works as you'd expect' is pretty simple
If that means that you can press letters in they keyboard and see them appear in the screen that’s also true in Emacs... and you don’t even have to press ‘i’ for things to start working as expected.
In vi if I press "i" I can't use the arrow keys to navigate the text as normal. In emacs I can. I'd say using the arrow keys for navigation in text fields is a very widely accepted functionality. So isn't it vi breaking the expected behavior here?
It's strange to be talking about the UI details of vi/emacs (both, by current standards, about equally super-weird) when the article itself is a kind of perfect vignette of the organizational dysfunction of the group that maintains emacs.