The flavour problem and family symmetry

We show how two different family symmetries can be used to address the flavour problem in SO(10)-like models. The first is based on a gauged $U(1)_F$, whose problems dissappear in the context of a type I string model embedding. The second is based on $SU(3)_F$, the maximal family group consistent with SO(10); in this case the family symmetry is more constaining, so we merely look at it in the context of a supersymmetric field theory