[AnemicOak]

Mare here are some posts from a couple of other threads by Charles Stross (cstross) that might give some of the info you want.

Originally Posted by cstross

My understanding is that Hachette spent something like EUR 16 million on their ebook virtual warehouse infrastructure.

Because they treat vendors like Fictionwise as virtual wholesale customers, and have to handle accounting for ebook sales exactly the same way they would handle dead tree sales, and in turn allocate royalty payments accordingly. And stick DRM checksums on each outgoing ebook copy; sell 5000 copies, ship 5000 files (which each have to be generated separately on an encryption server).

They've done their best to port their existing business model to the electronic world, because nobody has the authority to say "this business model is obsolete -- here's a completely new one" (not merely because it would screw their career if they made a wrong call, but because there's internal corporate resistance to it: lots of people will lose their jobs if the old business model is junked).

Originally Posted by cstross

Well yes: indeed, I've got all the facilities and resources I need to do it myself, if it made sense to do so. (Got a colo server: check. Got a web platform that has been fire-tested through regular slashdotting/boingboing-ing: check. Married to a typesetter with an Adobe DTP workstation: check. Can hack perl: check. Got a Paypal merchant account: check.)

I think, however, that you underestimate the people problems.

And you overestimate the profitability of ebooks, too. Bluntly: yes, Baen have managed to make money at it. But it's nothing spectacular -- their ebook sales volumes are comparable to their hardcover sales volumes, and an order of magnitude below their mass market paperback sales. This is excellent going compared to the rest of the publishers (who in many cases sell so few DRMd/overpriced books that they don't even recoup the cost of paying someone at their typesetting bureau to hit the "Publish" button in Mobipocket Creator), but it's not setting the publishing world on fire. And, more importantly, it's not profitable enough to goad the execs in the big publishing houses to risk their jobs by going contra the received wisdom.

This will change over time (indeed, I believe Tor is making progress in this direction, but not yet publicly). But it's going to take years, not months. Because? As I said earlier: the practices of the publishing industry have evolved, and they're the vector sum of those practices that didn't make some other publisher go bust at some point in the past two centuries. And what you're asking for goes against some of those survival-honed best practices.

Originally Posted by cstross

It's all about internal publishing industry politics.

1. All the major publishing imprints are divisions of large multinational media conglomerates these days. For example: if Ace or Roc go bankrupt, don't expect to read about it in the Financial Times -- they're units of Penguin Group, which in turn is part of Penguin-Putnam, who happen to also publish the FT.

2. The execs in charge of the publishing units are, like execs in any other multinational, expected to contribute to the corporate bottom line -- or else.

3. The structure of these publishing units is geared around a production pipeline that takes in manuscripts at one end and ships provisions to a warehouse for distribution via retail outlets at the other, typically processing several books per week. This process is reflected in the rights contracts that their legal departments negotiate with the authors. It is very hard to change this structure because it's How The Industry Works. (Which is inviolable, because the publishing industry works the way it does because its practices are the vector sum of all practices that have not caused some other publisher to go bankrupt at some previous point in the past two hundred years. In other words, it doesn't make sense to look at what it does, you need to understand it by looking at what it doesn't do.)

4. Consequently, ebooks are generated by the same production process as dead tree books, and are "shipped" to an imaginary "warehouse" from which copies are distributed to imaginary "wholesalers" like, say, Fictionwise.

4a. It is worth noting at this point that the cost of paper-and-ink contributes maybe 10% of the cover price of a novel. Another 10% goes to the author (plus or minus), about 10% goes to the publisher in profits, and 70% is gobbled up by the wholesale and retail supply chain.

5. This means that there's little or no flexibility in the process, and the publishers can't easily cut their overheads by redesigning the publishing pipeline to work sensibly.

6. They recognize that they could boost ebook sales by cutting the prices. But if they did that, they'd risk cannibalizing the sales of their dead-tree editions, and reducing their profitability. Their hands are tied so they can't streamline their processes to make a bigger profit on the ebooks; if they cut the ebook prices by even 10% below the dead-tree prices, they're eating into their profits. And that brings us round to point #2 above.

7. Moreover, ebook sales of DRM'd high-cover items are so piss-poor -- figures in double digits are not uncommon, compared to the four to five digit sales expected of a hardcover -- that there's no pay off in sight to act as an incentive for executive risk-taking (like, oh, selling ebooks for what everyone understands is a realistic price).

Finally,

8. You try selling a novel to a major publisher without giving them the ebook rights! I've tried it, and my agent just gets knocked back -- withholding ebook rights is a deal-breaker, because despite all of the above nonsense, everyone knows that ebook rights are going to be worth something sooner or later. So they insist on buying them, even though they can't use them effectively. And at the end of the day, we writers have to earn a living so we can eat and excrete more of the product, so we knuckle under.