My understanding is that the mean-free-path of conduction electrons in even OFHC copper is quite short -- scattering on impurities is the dominant loss mechanism, not the boundaries of the conductor. Is the theory advanced above that the morphology of the wire matters, or is there another mechanism at work?