> It's fine to think of entropy as messiness; that's the Boltzmann picture of statistical mechanics. But it should be added: If one leaves the field of kinetic

> It's fine to think of entropy as messiness; that's the Boltzmann picture of statistical mechanics. But it should be added: If one leaves the field of kinetic gas theory and focuses on scenarios where cohesion becomes relevant, such as in crystallisation, an increase in entropy means an increase in order.