Does this have the same problem as our definition for a function norm where <<f,g>> = 0 doesn't necessarily mean f = g = 0 ?