> If this exponential doesn’t turn out to be a sigmoid in disguise I'd put a lot of money on this in fact being the case, but I take your point that it matters

> If this exponential doesn’t turn out to be a sigmoid in disguise I'd put a lot of money on this in fact being the case, but I take your point that it matters how long there (roughly linear) part of the curve lasts.