In this note we will consider the following problem: how many coupons we have to purchase (on average) to complete a collection. This problem, which takes everybody back to his childhood when this was really "a problem", has been considered by the probabilists since the eighteenth century and nowadays it is still possible to derive some new results, probably original or at least never published. We will present some classic results, some new formulas, some alternative approaches to obtain known results and a couple of amazing expressions.