Rafael Harth comments on Matt Botvinick on the spontaneous emergence of learning algorithms