

The “unless you know what you’re doing” part tells me it’s totally worth it in some highly exceptional situations. You just need to be able to justify spending a few hours to figure out exactly how to do it safely.
Best thing about Linux is that you can do literally anything you want. If it works, it’s awesome. If you break your system, you get to keep the pieces and learn something new along the way.
I’m utilizing this liberty by being a lazy admin who updates things like eventually™ or soon™. Haven’t learned any hard lessons yet, so I guess it’s ok. Or maybe I just know what I’m doing…

Back in the X11 days, I actively avoided GNOME, because Cinnamon, KDE and XFCE were so much better. I had so many issues with the design philosophy, that using GNOME felt impossible.
However, when Wayland started having some support in GNOME, I got very curious and gave it a try. Then, I also bought my first touch screen laptop, and simply had to try GNOME with it. Turns out, GNOME wasn’t that bad, as long as you’re not trying to tweak every little thing about it. If you’re a tweaker, KDE is definitely the way to go. If not, GNOME might be tolerable or even good.
I’ve done so much tweaking already, that I don’t really have that itch any more. Sure, some things like custom keyboard shortcuts have to be just right, but that’s why you have GNOME Tweaks and the dconf Editor.