Neither - in order to be "half full" i.e. "half empty" (for indeed verily they are the same thing) it would be necessary to know the exact number of molecules of liquid in the glass, and the exact number of molecules which the glass could potentially hold as a maximum amount, and for the former to be exactly half the latter. These two figures would not be constant, because they would both be changing constantly due to evaporation, friction, expansion, etc. And the definition of what constitutes a "full" glass would need to be defined in terms of the level of the surface, the meniscus, the rim of the glass, the lowest point on the rim, the angle at which the glass is resting on a surface, etc. The extent to which the glass is defined as "full" would also depend on the shape of the surface of liquid being affected by the viscosity of the liquid, and the distortions caused by the local gravitational field. It would be impossible to know all of these measurements exactly, due to Heisenberg's uncertainty principle.