> Writing a new backend for an existing DE (Gnome3/KDE) is not the same as writing a completely new DE from scratch. Having written substantial parts of the Pl

> Writing a new backend for an existing DE (Gnome3/KDE) is not the same as writing a completely new DE from scratch. Having written substantial parts of the Plasma 5 shell (also nearly a full rewrite, which may just ship stable on Wayland before Unity 8, despite having been started later, and the Wayland conversion even later), I don't agree - "writing a new backend" is a substantial amount of work when the frontend doesn't have a concept of "backend"; a lot of X11 shell code traditionally made a lot of assumptions about being on X11. We had to rewrite our dock[1] as well for example, and that's before you get to talking about turning a X11 window manager into a Wayland display server and its many additional responsibilities. And as mentioned, Wayland handily beat them to the phone, too. I don't see any evidence that Mir hasn't just been an extra expense in resources for Canonical, and that's with freeloading on activity of the Wayland community. How did it in any way accelerate their progress to production to be the odd man out? 1 = https://blogs.kde.org/2016/05/31/new-plasma-task-manager-backend-faster-better-wayland https://blogs.kde.org/2016/05/31/new-plasma-task-manager-bac...