Which is why "The Year of Desktop Linux" will keep being a meme (30 years to 4%), unless we are talking about running Linux Desktops from within WSL and Virtualisation Framework, or for the oldies, VMWare Workstation and Virtual Box.
"Linus Torvalds Explains The Reason Linux Lost The Desktop"
M$FT revenue is currently around $300bn a year, with a 40% profit margin. Canonical's revenue is less than $300m.
Even if Canonical wanted to create a perfect version of Ubuntu with driver support for almost everything and as many desktop bugs removed as possible, they still can't afford to do it.
A credible Linux-on-the-desktop needs a global or multinational consumer-oriented foundation that can herd all the cats and turn desktop Linux into a professionally-managed project with proper beta testing and some interest in user experience.
That's a 180 from Linux culture, so it's unlikely to ever happen.
The Linux Foundation is not that Linux foundation, because it's not remotely desktop or consumer-oriented. freedesktop.org is a nice side project but doesn't have anything like the budget or the profile or resources required.
That is literally the only way to make a desktop version that can compete with Windows and OS X.
There's an obvious opportunity for someone to do this for governments now - a standard gov-optimised distro with enhanced security and access control and a collection of boring productivity software.