However bad people think X11 is, the engineering decision (or more likely lack of any intentional design decisions beyond the initial extremely narrow use cases?) to "replace" a huge portion of its functionality with windowing toolkits is a disaster for open source. I have written about this before, but the amount of waste and rework that it induces is completely unsustainable. If you want to see another generation wasted in the bazaar then stick with wayland. Wayland is very good at the use case it was developed to serve, which is fast booting systems for automobile backup cameras. Everything beyond that is at best an afterthought. Forcing dbus on everyone to replace xlib is like chopping off your legs because you heard that you can run faster with the new carbon fiber prosthetics -- it will work fine on a track, but the second you try to go hiking off the trail with boulders you will fall and break your skull.
Gnome developers are the ones advocating for “dbus everything”, wlroots prefers wayland protocol extensions.
And since no one company is behind the linux ecosystem, everyone works on what they want to work on, so every change is fundamentally starts with chaos in linux. But it will stabilize on a good enough solution (eg. I don’t see much ill with the current 3-4 implementation of the protocol with one of them being a minimal “lib”, that can be used for small niche window managers to build on)
