Agreed and I think the reason for this confusion is because string types generally aren't considered a complicated feature of typescript.

Nor is it unique to it. I.e. elexir does the same through pattern matching and it works even better as far as I'm concerned.

It's more about the features that enable extended type inference and modifying these inferred types on-the-fly.

Easy to understand type modifiers like Omit<>, Partial<> etc are fine, but if you lean into that you can create pretty hard to understand typing logic. Especially if you're diving into annotations for code generation