Assuming that this is intended to work along the lines of Andy's option A from the first page, then this works quite well. I wonder though if it reflects the desired effect of larger domains being substantially more expensive to maintain to the degree that Bjorn apparently intended.
Perhaps the progression after 10 levels could increase by .02 rather than .01 (i.e. 1, 1.2, 1.4, 1.6, etc.) and have the break points every 20 levels, rather than every 40.
That would give a slightly less linear progression and also get the cost of the really big domains up closer to the levels Bjorn had intended.
