Should the designation of the components of a system reflect its known hierarchy or rather the history of their discovery? With the recent progress in, say, radial velocity techniques, the old famous order in which components were used to be discovered (inner to outer components for spectroscopic systems) is somehow altered. In the past, capital letters were used for visual companions and lower case letters for spectroscopic components and there was almost no overlap between the two groups. The situation has changed from both ends of the orbital period interval. In some rare cases, we think letters should be re-distributed and re-assigned in order to reflect the structure of the system. With an adequate choice of the data structure, such a change of the companion designation is rather straightforward to implement in modern databases (such as SB9). The only foreseen drawback is related to the cross-reference with some old papers: the letter B would not designate the same component in a 1970 paper and in a 2003 one. For instance, the former secondary of an SB2 system might now refer to the unseen companion and an astrometric triple.