On Fri, Sep 20, 2013 at 1:11 AM, Daniel Farina <daniel@fdr.io> wrote:
> I think the n-call underestimation propagation may not be quite precise for
> various detailed reasons (having to do with 'sticky' queries) and to make it
> precise is probably more work than it's worth. And, on more reflection, I'm
> also having a hard time imaging people intuiting that value usefully. So,
> here's a version removing that.
I forgot about removal of the relevant SGML, amended here in v6.
