The great thing about vi(m) is that it's always there. SSH'd into some random server and need to edit some random config? vim.
And Sublime is vim with a user-friendly-ish GUI wrapped around it.
And Sublime is vim with a user-friendly-ish GUI wrapped around it.
I hate that, I ssh in, then maybe I want to edit cron and I am hit with vim(someone changed the simple nano with it), in the past you had no choice then google how to quit vim but now it will tell you how to do it(I ssh and change config files 1 or 2 times a year). I never learned how to quit, just o use EDITOR=nano command to get a simple editor, I don't need a powerful editor for config files.
Would it be better if Nginx would have never been written and we'd all use Apache instead?
What about Docker never having been written and us having to ship VM images with no tool chain around them?
What about JetBrains never creating new IDEs and everyone being stuck with Eclipse or NetBeans?
What about no systemd and having to use sysvinit or OpenRC? Okay, perhaps that last one is debatable, due to some valid concerns.
If Vim is so powerful, then this power should be embraced and transformed into a software package that's easier to use, be it with nano style hints or simpler shortcuts - Vim isn't good because of its accidental complexity after all!
Of course, that might take a few generations, because no one who knows Vim wants to relearn it, whereas the newer people will oftentimes simply resist using it.
But my point is that the weaknesses of each software package should be addressed and eventually fixed, either in itself or another to replace it.
Could vim be easier? Probably. But it’s still useful and is an archetype. Remember I’m not saying new code should be written in Awk. It’s just useful to know how to use the tools that everyone knows, regardless of whether they should be replaced or improved.
The handsaw isn’t perfect or that beginner friendly, but if you refuse to learn to use one because there are other cutting instruments, you’ll probably have a bad time compared to if you didn’t.
This is a perfectly reasonable argument to make and probably reflects the reality of working in the industry pretty well.
That said, I'm still stoked to see what people come up with in the next 20 years.
Maybe gradual improvements like Neovim tried to do. Or perhaps Vim-like text interaction in VSC or JetBrains IDEs, with all of their plugins and ecosystem. Or maybe something even more grand!
Vim's power comes from it being 1) text-based, 2) very fast to load and manipulate files, and 3) very fast to use if you are familiar with the syntax. Like many power tools, it's the last part here that's hard, but it's also very hard to fix without affecting one of the other two qualities. For example, Emacs nails points 1 & 3, but is so much slower to load that few people use it as their default $EDITOR, and has a sizable learning curve as well.
It makes no sense to learn vim because I am hit with it once a year, I better learn tools like "tail" and "grep" this ones are useful for what I do.
But sure if you are a sysadmin or ssh daily and edit big complex config files the sure learn vim if you really want that.
Or you can keep resetting the default editor to nano. You do you, friend.
nano /etc/congig.cfg
my only issue was with crontab, since it was using nano but someone changed it to vim so I was surprised by it, I really had to google how to exit it. Since I edit cronjobs once a year and for config files I can use nano it would be a waste to learn vim, I would just forget all the shortcuts - not sure how other people memory works by me I forget stuff I do not use , like regexp I use them 2 times a year and each time I need to read again what symbol does what.
I am not sure who to blame for making vim the default (in EDITOR), I could swear it was nano and someone in our team had to switch it vim to fuck with the rest,
Ended up settling on MacVim.app which has nice window support.
I’ve had good experiences with the visual studio vim plug-in that’s written in F#