Some remarks on Hilbert's (weak) Nullstellensatz