Jarrett Billingsley wrote: > Whether it'll be going away, I'm not sure. I guess Walter is the only > one who knows that. I see no point to breaking invariant() again.