> intercepting/sending keystrokes in particular had to be redesigned to make to make it more secure. without that there would not have been a point to wayland a

> intercepting/sending keystrokes in particular had to be redesigned to make to make it more secure. without that there would not have been a point to wayland at all. Sure, and I really would like to enjoy that security improvement. But would an evolutionary approach have been possible? For example, would it have been possible to start by running Wayland with all programs in a privileged status, then migrate to using something like OpenBSD’s pledge(2) to spawn programs preferentially in an unprivileged status, then migrate to spawning some programs (e.g. legacy X11 display managers) in a privileged status? It sure seems like Wayland requires a lot more work from software using it than X11 did. Wayland has been out for 16 years and still isn’t almost universal; 16 years after X11’s 1984 release it was well-nigh universal. Again, I get that no-one owes me anything. But it seems to me that the Wayland developers committed the fundamental error of trying to replace a working system with vapourware. It’s still vapourware, in that in 2025 Wayland is still not a drop-in replacement for X11. > your actual apps that just display stuff on the screen and take your input all still work though. My display manager is an actual app that is a fundamental part of my workflow.