Unfortunately the way the PC market is makes it basically impossible for a new desktop OS to show up at this point. The hurdle in drivers alone would be insane, and the easy solution to that is to only target a small set of hardware which effectively makes you a hardware company (kinda like Apple), but that probably won't work either since people are a lot less likely to try out your OS if they need to buy a new computer to do so.
> ...the easy solution to that is to only target a small set of hardware which effectively makes you a hardware company (kinda like Apple)
Perhaps with this idea the only companies able to do this are Apple, Microsoft and Google. Every other OS aiming to be a new desktop OS has essentially failed. Including the Linux desktop community and everyone else.
The desktop OS market only has room for 2 or luckily 3 and always requires at least a billion-dollar (in profit) tech company with such resources to even plan for years to do such a thing. Hence the fierce competition in this space, some may ask themselves why bother.
> but that probably won't work either since people are a lot less likely to try out your OS if they need to buy a new computer to do so.
The exception to that is Google with ChromeOS (Laptops) and Android (Mobile) which they're probably both replacing with Fuchsia. I'd expect Fuchsia running pixelbooks and phones to arrive in this decade. Which will make the desktop market as Windows, macOS and Fuchsia OS.
Perhaps one should build their new OS around a compatibility layer to support Windows drivers. I hardly am too much of a system programmer and don't know how much sense would that make but AFAIK this is possible. E.g. I remember using Windows WiFi NIC drivers on Kubuntu a decade ago.
