Note that the JIT mechanism mentioned in the paper meant basically JIT from a program code to the bytecode executed in the VM, and directly executing it (very much like PHP), not necessarily directly to machine assembly.

Interesting.... the paper referenced you if you're the correct Anton Ertl :)

That's fascinating. Would love to see what performance such thing can actually perform.

There's a specific issue in implementing / caching the top of the stack with registers. (I'm assuming that you're talking about stuffs related to C's `register` keyword here) CPU registers don't have memory addresses so it would be basically impossible to achieve that, as in the stack machine implementation of this paper, void pointers are stored in stacks

I can only partially agree with this - having a stack exposed to modification would virtually make it not a stack.

GitHub seems fine for me.