On unifying “toy languages” and “real languages”