Stuart_Armstrong comments on Partial preferences and models