In this commentary I point out that Palmer mislocates the source of the inverted spectrum, misrepresents the nature of colour science, and offers no reason for prefering one colour machine over another. I conclude nonetheless that talk about 'colour machines" is a step in the right direction.