Let me suggest you a Jedi mind trick which just may work out for you: Imagine a situation where you could fork() the curent universe and run two different copi

Let me suggest you a Jedi mind trick which just may work out for you: Imagine a situation where you could fork() the curent universe and run two different copies of the world in parallel on two different CPUs. Now you can make a certain choice in one copy, and an opposite choice in another copy, and then let the events run their course and see which one was the better choice. Wouldn't that be nice? You could then see the unpredictable consequences of your choices, and there are always some of those. Sadly, we only have one CPU to run the world, and there is no place for context-switching. But! Turns out there is one situation where you can do just that - if the choice is suicide (yes/no), you can pretend that you have taken that choice and killed yourself; so now you know the outcome of that choice, and the CPU is free to run the other fork() of the world. Since you already know what's down the first path, might as well go look through the second one.