you know the basic working principle of Linux-Namespaces? And Iirc the JS-intepreter is not spawning a subprocess with a communication pipe for each require? So

you know the basic working principle of Linux-Namespaces? And Iirc the JS-intepreter is not spawning a subprocess with a communication pipe for each require? So good luck with your "easy implementation" - you could also just create a standard library/make PRs removing unnecessary cruft from existing high-level packages. And if you think creating an interpreter in the interpreter is the solution, I'm quite sure the thing will not be secure if it is a natural child of the JS ecosystem.