[caffeine]

You have to do your own thing ... usually that's bad advice but for people with the temperament you describe I think it's reasonable.

Also a lot depends on expectations and finding good partners. In a previous job I worked with another guy as a team. I would start projects and get an MVP out and delivering value - then move on to the next thing. He would go through and essentially rewrite them to be high quality, solidly engineered products, well integrated with the rest of the stack.

We both got to do what we enjoyed and were good at: I am very fast at breaking new ground and delivering new value, and find engineering a bit boring. He was very good at improving existing systems, but too slow and plodding to try out new ideas effectively.

[nyanpasu64]

i have the unmet rambling dreams of the first person, and the anxiety-driven perfectionism and revulsion to poorly written code of the second person... so i'm unhappy thinking about unfulfilled ideas, unhappy writing new programs, and unhappy fixing old programs.

"i can only do so much and of course it's never enough"

i want the perfect programming language, the perfect gui library api and theme, the perfect program, the perfect ide... and i can't accomplish a single one

and of course C++ is an all-devouring Cthulhic monster that does everything but poorly, Rust has immature gui libraries and doesn't fully align with my values and desired features (I want linear typing, strong typedefs/subclasses of integer types, polymorphic variants/anonymous unions, prioritizing iterator generators over async), qt is basically legacy code and Qt Widgets is mostly unmaintained but difficult to fork (to build, create Windows installers, and convince Linux distributions to accept behavior-changing bug fixes), and my personal dream project (https://gitlab.com/exotracker/exotracker-cpp) is vaporware no matter how much i burn myself out making it (trying to both explore new ideas, and engineer them well, at the same time).