When new developments are made in computer graphics technology or UX research, I want to be able to take advantage of them. I don't want to limit myself to a design based on the state of the art in 2006 just to avoid having to learn anything new.

What's some of the latest and greatest UX research someone might miss out on when they don't use some of the modern desktop systems?

I mean I'm not against new and better ways to interact with a computer, I just like to decide on my own if I want to use the new and shiny thing or not. Fuzzy searching is probably the most significant improvement I opted in in the last couple of years, and it's used basically everywhere on my desktop, while all the "modern" desktop systems like GNOME, Windows 10, macOS, ... still barely make any use of it and their search implementations sometimes even worsened from my point of view (Windows 10).

I can think of two major, useful new features in desktop GUIs in the last, oh, 20 years? More? And both existed before they went mainstream anyway, I think. Both are basically just search: launching programs by search (replaced program-launch menus all but completely for me) and deep search in deeply-nested menus, guiding you to the item at the current level that contains what you need (as in the macOS settings menu—I've actually not seen it done well anywhere else but I want it everywhere that kind of navigation structure exists, having seen how nice it is there)

> launching programs by search

https://en.wikipedia.org/wiki/Command-line_completion#Histor...

The menu-search feature did exist in Unity on Ubuntu at one point but it was not as nice as the macOS one. I think it's a hugely underrated feature in macOS.

Another related feature that I wish some Linux gui framework would adopt is the concept of global keybindings. Every menu in every app in macOS displays by default the shortcut key next to it - if one exists.

These are the kinds of features that are easy to overlook but contribute hugely to the "cohesive" feeling that macOS advocates so often seem to have trouble describing to Linux desktop users.

Yes, that's is such a pain point in most linux setup. Even copy/pasting has not unified keyboard shortcuts (terminal). I know there are valid explanations for that, but from a user point of view this is a disaster.

KDE had global key bindings ab ovo.

Okay.... that's great. Nothing I said precludes that.

Realistically I suspect that at the end of the day the people who are getting things done aren't jumping on every shiny new trend. BUT they also they aren't spending weeks on end tweaking all of their configurations to make them perfect.

I didn't make some grand decision to use emacs (I assume mentioning emacs is what upset you). When I installed linux for the first time the instructions were to use vi or emacs to edit files. I couldn't figure out how to quit vi so after a hard power cycle I use emacs today. It turns out though that if you use a tool for long enough and make small tweaks occasionally it adds up. My environment suits me perfectly. Maybe you would hate it but that's fine too.

Whatever floats your boat. Why be so dogmatic? Why not use a little of the tried and true and mix in a bit of the hip new thing?

I'm not upset and I don't hate emacs (nor love it). I am only addressing the criticism that Windows/macOS/etc change too frequently.

We're using 50 year old UX paradigms. Nothing has significantly improved after Engelbart.

If you really believe that, then why aren't you using a 50 year old desktop environment?

I am, and so are you, I would think. That's how I understand "paradigm" :)

I didn't say, "why arent you using a 50-year old paradigm". I am asking, why are you not literally using a software developed in 1970?

For example, are you using twm? xterm? Admittedly those are only 30 years old but I don't know any older examples that are still supported.