Model independent constraints on a heavy neutral vector boson from present and future LEP and SLC data