I think many companies are basically stuck with Microsoft (Excel, Word, Teams, Sharepoint, Onedrive etc).
Tbh, office and collaboration tools are the least of our worries; there are plenty of good alternatives which, with some financial support, can be adapted to suit companies' needs in very little time. What should worry us more are the tons of critical applications tailored to a very specific area of administration and business. The software that runs the power plants and hospitals. The software that manages logistics and industrialised production. The software that runs our accounting and HR. That's the Windows lock-in that we're not going to shake off over night.
Oh, that and the hardware vendors + hyperscalers...
Not true. We've seen the U.S. leverage big tech against its perceived enemies by shutting them down remotely. Unless Europe bends over backwards to serve the U.S. fascists' every whim, this form of blackmail-by-proxy is going to continue, and it's only going to get worse.
So the calculation is not: "cost of switching" vs. "cost of continuing the existing vendor lock-in".
It is: "cost of switching" vs. "cost of shutting down operations for good".
It's really easy maths if you break it down to its core. It's maths that even the most bone-headed of our leaders and business "experts" can do.
You know much more about the world around you than the orange turd that claims to be the leader of the free world. Don't sell yourself short, mate! It'll take people like you to rebuild once this is over.
I assume device drivers, but probably it is much more complicated than that
Yes, device drivers are an issue. Reverse engineering them is a removed and slows you down, particularly if you want to support a wide range of models and those models keep getting hardware updates.
But that's not all, software ecosystem is another big one. Android and iOS have seen two decades of people developing software for them. In order for them to want to port their software over to your cleanSlateOS, it would have to have a significant user base. And in order for cleanSlateOS to draw that significant user base, it would have to have an attractive suite of apps to run on it. It's a catch-22.
You could, in theory, try to develop emulators or compatibility layers so that Android apps will also run on cleanSlateOS. But that, again, is time-consuming, will never be free of friction, and require you to make compromises with regard to security and privacy (many apps simply don't run properly without Google's main piece of spyware, the Play Services). It will also kind of tie you to Google again - and that was the thing you were trying to get away from in the first place...
The problem is: at least where I live, banks use mobile apps to generate second factor authentication. (No TOTP, all proprietary / homegrown mechanics.) No second factor - no login.
So, basically Microsoft's "we know what's best for you" style? No. Fuck that, no matter the purpose.