Dear Flavio,
I wanted to compare the mathematical structure of classical and quantum evolution equations, to see what kind of interpretations they allow. That is, I wanted to "ground" the discussion in the formalism.
From that perspective, my last paragraph was ill-chosen and obscure. I agree with what you write in response.
I think we need to highlight interference as a distinguishing feature of quantum evolution equations, in addition to the uncertainty principle and the quantum of action. Interference is the reason why we cannot "decompose" the evolution of an ensemble in phase space in quantum mechanics to the evolution of its subsets, or to the deterministic evolution of its points. And that's the reason, as I see it, why quantum mechancis cannot be interpreted deterministically - in a certain sense.
Trying to be more precise what I mean by this, I'm talking about the standard postulates of quantum mechanics (thus excluding the additional stuff in Bohmian mechanics), and the evolution of observables, or points in phase space. (In contrast, the determinism in MWI resides in the wave function, that is, in the deterministic evolution of points in Hilbert space.)
The standard example is, of course, the double-slit experiment. You cannot "decompose" the evolution of the setup to the alternatives R1 and R2 where the particle goes through the left and right slit, respectively. Then you lose the interference pattern, which exludes som final particle destinations that are allowed in the "decomposed" evolution.