I believe that was the implication, too. One's level of expertise will determine what is "too complex for $SHELL". When that point is reached, it's time to swi

I believe that was the implication, too. One's level of expertise will determine what is "too complex for $SHELL". When that point is reached, it's time to switch to something else, not time to add more features to $SHELL. (If, when something becomes too complex for $SHELL, you switch to assembly, you might be too much of a $SHELL expert!)