They are. The fact that HN doesn't follow it just speaks to the anti-Microsoft bias here. They've had the OneCore project back in 2015-2016 to unify the kernel and core OS pieces, which was a success and is now powering every Windows device out there from Xbox to the PC. They have the UWP, their unified app framework.

They are also working on CShell, which is a unified UI/windowing/compositing system that will work across PC, Xbox, Mobile, etc.

CShell + OneCore + UWP are essentially becoming part of a "brand new" OS they're calling Core OS, which will strip the Win32 backward compatibility requirements and just work with UWP apps, primarily targeted at devices like phones and low power tablets.

You can search Project Polaris for more information

They're taking a different approach. They're positioning .NET to be the Platform regardless of the underlying Hardware/OS.

They tried to rewrite the OS in a C# derivative and could never meet the perf needed while still fighting backwards compatibility issues[0]. At this point mobile OS's are a lost cause, but the PC market is stable. Windows server has never been as big as Linux, but all of Azure runs the Windows hypervisor. I'm not sure they're losing at all actually. [0]: https://en.wikipedia.org/wiki/Midori_(operating_system)

Not really, they are winning the desktop and tablets (2-1) war.

Desktops aren't going away, and Macs are pretty much an US phenomenon, alongside a couple of other countries.

Likewise, Android tablets aren't that good, with most of the apps being phone apps. Google now is playing with ChromeOS for tablets, which also remains to be seen where it goes.

So, most consumer shops around here are now mostly selling iPads and Windows 10 tablets.

They've already tried. There's no interest currently for a 3rd also-ran OS, especially from a vendor that would be changing up the toolkit yet again.

I would have agreed with you a year or so ago. I've been on MBP for about 4 years, my next laptop is going to be a PC again.

If Fushia/Flutter gets adopted, it would be a better outcome than MS botched effort with universal desktop/mobile apps. The fragmented tooling and support was like it was made by sales/marketing types rather than engineers.