Comments only, relating to Roman's new built-in rule for seq