Resumen
The main concept of rough set theory (RST) is an indiscernibility relation normally associated with a set of attributes. The existence of functional dependencies stressing that they cannot be found by just examining the data, and hence must be specified by the database designer. The mechanics of RST applied to data is indeed deterministic. However, it is doubtful this makes the entire method objective since the preprocessed data is heavily contaminated with subjective judgments during the initial roughing stage. Roughing data involves so much initial subjective processing that there is no scientific basis for believing any kind of miraculous enhancement by further objective or rather deterministic processing would change them into a more precise or objective image of reality. RST is, potentially, an important tool in the analysis of data with important applications in data mining and knowledge discovery. However, claims of its superiority over other approaches must be substantiated by scientific evidence.