For many non technical consumer users it no longer really matters whether they're running Windows 10, MacOS or something weird like a linux+xorg/KDE desktop environment.
We don't click on links that go to web sites we don't trust. We don't open email attachments, we're paranoid... because our operating systems weren't designed for the age of the internet.
linux based ideas like qubes are are a step in that direction.
To deploy using a 30GB partition it only takes about 5 minutes to write zeros to the 30GB, another 5 minutes to quickformat and deploy the WIM file to the fresh partition (in naturally well-defragmented condition), then a judicious BCDBOOT command to make it boot from your present BOOT or EFI folder.
Moreover, the desktop is a very difficult market to enter. Writing a new desktop environment that is competitive with even Windows 7 and Mac OS X Snow Leopard is going to take a lot of work, even if they piggyback on existing operating systems such as Linux in order to avoid the full work of writing a new operating system. Moreover, where is there a viable business model for selling desktop operating systems? Be, Inc. tried with BeOS in the 1990s; the company had a hard time getting PC companies to agree to shipping their PCs with BeOS preinstalled due to agreements these companies had with Microsoft regarding preinstalling Windows. There's also the problem with software incompatibility, though, interestingly enough, this may be less of an issue today than it was in the 1990s due to the dominance of the Web and due to the "Electronization" of desktop applications.
My opinion is such innovation on the desktop will come from a hobbyist open-source project or from a small business that is willing to cater to a niche of users who want compelling alternatives to Microsoft, Apple, and various Linux desktops.