Earlier quoted context omitted.
The contention might be on the time frame. Desktop OS and apps won’t go away for the next 10 or 20 years by sheer inertia of the corporate market, and the halo effect on home users. And I guess the past month/years issues don’t affect business users (corporate managed desks will have any problematic feature disabled from the start) “Everyone” migrating to OSX/linux won’t happen for many many years, even if they keep…
PC gamers won't be switching to an Xbox, and so we require a desktop. It won't just be halo effect on home users, but ironically also Halo effect. The remastered collection being ported over as we speak, starting with Halo: Reach iirc.
Some of that software is Windows-only. There are quite a few good products that run on Mac. Relatively little runs on open platforms like Linux. And for now, macOS has its own issues for long-term professional use, starting with Apple's reluctance to give any sort of concrete guarantees about support longevity. So that can be quite a barrier to migrating away from Windows for businesses.
It would be nice to think that in time Linux or other free platforms would fill the gap, but it's a chicken and egg problem. As an interesting contrast, within a very short time after Adobe went cloud/subscription with the dominant package of graphics/publication software, we had several smaller and more agile competitors entering the market at much more attractive (and non-subscription) price points: Sketch, the Affinity products, etc. We don't know to what extent any of these might already have been in development before and how much they were motivated directly by the shift to Creative Cloud, but we do know that the closest OSS equivalents have been around for years and made negligible impact, which doesn't bode well for the viability of Linux desktops as an alternative platform for "normal" business use (as opposed to, for example, specialist work like software development).