Linux

For the record, in the anecdote in question the professor was teaching marketing in a non-tech degree, so I'm not sure about that one. The argument, IIRC was about them arguing that the Win95 launch campaign had been one of the, if not THE most successful marketing campaign ever, which all the millenials in the room were not having. Prof argued "nobody even knows what the second most successful PC OS would be" and the Linux incident happened. It was very funny.

Anyway, on the underlying point I agree that you could change the usage numbers in many ways, but the argument here is not that the low usage info proves the bad usability, necessarily. I'm saying the bad usability and compatibility issues are a major problem that makes the OS hard to embrace for most users. That's the hypothesis. The info that after decades of public, free availability Linux remains a marginal choice is a piece of info that reinforces that hypothesis. It doesn't prove it by itself, but it's certainly very consistent with it.

I'd argue that the fact that Linux is free and it's not preinstalled more often also reinforces that point. In fact some PC builders would offer it as a fallback if you didn't want to pay for Windows, especially back in the 00s when the functionality gap was actually narrower than it is now, and that didn't seem to help much, with most people still paying the fee to get a OEM Windows install.

But all of that is still indications we see in the market of the ripple effects of Linux's reputation, which would be ripple effects of its UX and compatibility issues. It's not the entire picture, but it sure fits in the picture, if you see what I'm saying.