I can assert that a large majority of people at CERN leave GNU/Linux for servers and use Windows/macOS on their desktops.

It was like that 20 years ago, and it looks quite the same when I visit it for Alumni events.

Same, it's either linux or macos in compute intensive research for what I've seen in Europe. I've seen groups working with Windows based data analysis tools but those groups and the ones that need python rarely intersect.

WSL2 is a godsend for people forced on Windows by blind IT policies, but fortunately IT doesn't have that kind of control in academia.

What you've "seen" has nothing to do with what's actually popular. Why does the concept of observer bias have to be explained on this site?

My response was based on my own experience. Hence the remark.

What I've observed was also dictated by my own choices.