A remark on discretization of the uniform norm