Well, the article is really about Smashwords as a publisher. If you look at it in that context, the graph makes much more sense: as a publisher, their cut of the pie is minimal. But it also sounds like the services they offer are minimal (they vet books, don't edit them; they only offer electronic editions, not electronic and print). As such, the risk to them is much lower than with traditional publishers.

Will the Smashwords model work out? I don't know. They are really offloading the risk onto the author and the reader. In the end, either one or both may revolt. But it is bound to be good in one sense: traditional publishers are going to have to do a better job if they are going to survive. They're going to have to push harder to generate more revenues for authors and improve the quality to hold onto readers. Otherwise they just won't be able to justify the large cut that they are asking for.