So would the use-case mostly be for people who edit lots of files over shells? I guess I was mostly thinking of coding stuff. Or do people use it for coding too?
I mostly use IDEs for day to day coding, and pretty much every IDE supports vim keybindings, which I always have enabled. I also use vim in the terminal for small edits and one-off files, so it's not either/or.
After the initial learning curve and fiddling with settings, it just becomes natural and you can edit code or other text at blazing fast speeds. I also find that it helps with RSI by reducing arm motions reaching for the mouse.
Of course, there are other good options out there, but if vim fits your brain, it can significantly boost your editing speed. For those who say programmers don't spend that much time typing, that's true sometimes, but there are periods after the design/planning phase where we type a lot, and I want that to go as fast as possible while I have an implementation loaded into short term memory.
As someone who used to be a vim skeptic myself, I'd suggest you either give it another look or just accept that it works well for other people and go on with your day.
Over time, you may discover one of the biggest truth about computing - the immense (and sadly disregarded) value of plain text¹. Code is nothing but structured text. The cardinal job of a programmer is text manipulation.
I'm reading this thread in my editor. Why? Because I'm dealing with text - browsers are great for presentation of text, not so nice for manipulating it. What kind of manipulations? Well, I can find, sort, group, narrow, collapse/expand, translate, extract summaries for paragraphs, find all the URLs (including matching a pattern), etc. Good luck doing all that in the browser - even with the extensions.
Most developers don't even get annoyed by seemingly small things. Like how often do you need to bring the url of a given browser tab into your editor? Simplest thing, yet do it dozens of times a day and it gets vexing. For me - it takes milliseconds and a keypress - it even extracts the URL description and converts the link to markdown (if needed).
Or another example - whenever someone's screen sharing, or I'm watching a video and they show a piece of text (let's just for consistency say a url). How would you normally extract it? e.g, for your notes. For me - it takes selecting a region of screen and a keypress - my editor calls a CLI OCRing tool and voila. I don't really care that the source isn't "technically speaking" text - if I can read it, then computer for sure can "read" it too, am I right? That's text manipulation.
There may be dozens (if not hundreds) of such examples (including coding-related) in my workflow that work on top of the idea of manipulating plain text.
Once you grok the cerebral virtue of the idea of having complete and total control over text, you may find that modal editors - nvim/emacs/etc. - are the best instruments to achieve that control.
