Even an imperfect estimate of the volume would be useful: for example, perhaps we only find some of the edges and conclude the volume is some fraction of its true value. I have the distinct sense of talking past the point you were trying to make, though.
No, that sounds more or less right.