But this way you are storing utf-8 encoded strings in the database without the database knowing about it. This could potentially lead to errors, e.g. when some code depends on the string length or is using postgresql's substring function.
Propably not very likely that it bites you, but possible.
