In a cluster core, hot gas can radiate energy and cool. Why is cooling not always observed to proceed unchecked at the rate a simple radiative-cooling estimate might suggest?