Transplanting a Seattleite

I don't have anywhere near the data (and probably could never get it, either, and if I did, I probably would have had to sign a contract not to let anyone else know any derivative details, thus doing me No Good Whatsoever) to figure this out, but I suspect the largest marginal cost of adding ebooks is that if the ebooks actually find a market, that market is cannibalized from the market that the print books are depending upon, rendering print books more expensive as a result of increased risk/unpredictability. Basically, if you add ebooks, and more people buy them than you expected and fewer people buy print books than you expected, you have to pulp/lose money on the print books. I think that is, in fact, the big cost driver here. Once we're through the transition, it'll be cheaper all around, because they will once again have predictability on the print side.