> how are we to objectively evaluate who the great programmers are, when it has often been argued (including by a certain essayist) that this can't be measured?

This isn't that hard. Let companies bid for H1-B slots. The more they are willing to pay, the better the programmer.