- Software development
- DevOps/system administration
- Industrial automation
- CAD
- (Serious) image/video editing
There are also many "legacy" business processes that are technically doable on a phone but are so cumbersome/limited there that the phone is only really used for things like quick review, small touch-ups, etc. "Microsoft Excel runs the world" is not a trope for nothing.
There are four main reasons for these limitations IMO, in rough order of importance:
1. The phone O/S vendors, who are happy to make most of their platform money through rents on physical-service (often extractive) companies that only pretend to be software companies (like ridesharing and food delivery), gambling, and "free-to-play" video games
2. The software ecosystem, which has largely ignored the use cases I listed above, and so remains fragmented and incomplete (at best) where they are concerned
3. The form factor, which is much more limiting than a monitor+keyboard+mouse/trackpad setup, and can't easily be adapted to compete with the latter except through a dock, but those remain clunky and a bit gimmicky, and no phone even has a Thunderbolt port or other high-bandwidth display-capable connection to properly dock with anyway
4. The hardware, which still shares a lot in common with that of PCs, but has been optimized for things like extremely low power consumption, limited external peripherals, and default-distrust of the end user
And the above is entirely focused on things that make money, this doesn't touch on hobbies, games, and other personal-interest niches.
The rest of the listed stuff seems super niche and goes back to the rounding error initial comment.
Like take a step back and think from the average person in the world, not some system administrator, what can the PC do that the phone can't?
The home-consumer PC market is already contracting and will likely continue to contract. That is not the claim I took issue with. You keep conflating market size with importance. These "rounding errors" run the modern world and will not stop doing so anytime soon.
If it "worked best" and actually mattered, the market would not shrink.
Sure. It doesn't (yet), so that won't happen (entirely).
However, it does work "good enough" in enough other areas to help explain the contraction. Maybe it will expand into being "good enough" in more areas, and maybe younger people will cause a generational shift that this "unc" can't foresee. But I laid out a bunch of reasons why that's unlikely to happen. I think both PCs and phones are market-saturated, but the replacement treadmill for phones is much faster than for PCs, so the current price shock is hitting PC sales harder.
I think we need an out here to say that PCs will remain for the foreseeable future for businesses and hobbyists, but this won't be enough to sustain the PC market as we know it.