It should never have happened in the first place. It was always going to cause problems
Some major changes in the language were inevitable after the introduction of new-style classes. Python 3 was more or less inevitable but the release was very badly handled. Thankfully, the core developers realised this at some point and did something to fix it. It's no longer an issue for the vast majority of developers.
Smirk. I've seen plenty of people get into deep, deep trouble with Python chasing idiotic bugs long after deployment
And I could point to a long list of projects that were saved by migrating to Python…
Also there's the mad, lunatic idea of having a package manager as part of the language installation
This is unavoidable because there is no installer that works for all OSes. I maintain a popular library and much as I moan about Python's packaging I'm so glad I don't have to package it for all the different Linux distros, MacOS, Windows.
It makes deploying Python application highly painful.
Strangely I rarely hear this, in fact usually the opposite. Deployment is such a huge problem that there are entire conferences devoted to the infrastructure.
Python on Linux does some things subtly differently to Python on Windows.
Python itself doesn't do things differently but invariably the different OSes have different APIs, file handling springs to mind. It's the same for any language that interfaces directly with the OS.
I'll be sticking to my C, C++, C#.
I'm happy for you. Why can't you be happy for people who, despite your predictions of doom, are successful with Python?