The proportion of comp sci tasks that have practical importance for which a quantum computer can shave off a factor of ~000s seems to be sufficiently small to doubt it will "change the world".