There is no "say" about it, that is a fundamental rule of oxidation state theory, which as I have said is only an approximation (as there is no discete ionic states in the molecules).
Yes it is! The F atom is the most electronegative in the molecule so it will pull electron density towards it. You have massively oversimplified the model of the molecule by assuming that it is a discrete pair of ions, this is incorrect. Quantum modelling calculations (which are also inaccurate, but a damn site better than oxidation state theory) show that the flourine has significant electron density around it in the ground state (indicating that it probably carries a delta - partial charge). The outer electrons are most likely to be dissociated over the molecule (i.e. full charges are unlikely to reside on one individual atom. A bond between oxygen and fluorine will be polarised with the flourine having a delta -ve charge (and, neccesarily, the oxygen having a delta +ve charge) because fluorine is more electronegative (i.e. it's nucleus has a higher charge denisty and very low inner shell shielding effects and so exerts a larger attractive force than the oxygen nucleus on the bonding electrons).