Internet Archive uncloaks open ebook dream machine
Will Google play?
The Internet Archive and various like-minded partners have launched an open architecture for selling and lending digital books online, an effort to consolidate the fledgling market for net texts - and give Google a little food for thought.
Dubbed BookServer, the open platform is meant to provide a standard means for booksellers, publishers, libraries, and individual authors to serve texts onto laptops, netbooks, smartphones, game consoles, and specialized ereaders a la the Amazon Kindle. The Archive has already demonstrated an early incarnation of the architecture with the Kindle and Sony's Reader Digital Book.
"The idea of Bookserver is to integrate business models and organizations that actually have fairly different points of view into a structure that allows the selling, giving away, and loaning of books," Internet Archive founder Brewster Kahle said during a Monday morning mini-conference where the spec was unveiled.
Based on the RSS-like Open Publication Distribution System (OPDS) standard, Bookserver is still in the early stages of development. But a preliminary version of the architecture is now in place at both the Internet Archive and the Feedbooks etext distributor, while tech-book publisher O'Reilly Media - who helped launch the effort - says it has every intention of adopting the architecture to serve up its own titles.
Bookserver arrives as American book giant prepares its own ebook device and the entire industry continues to chatter over Google's epic Book Search project, an effort to digitize and distribute millions of texts inside the world's leading research libraries. In October, Mountain View settled a lawsuit from American authors and publishers over the project, and the Internet Archive - along with Amazon, Microsoft, and other members of the Open Book Alliance - was far from shy about its opposition to the pact.
The pact creates a "Book Rights Registry" where copyright holders can resolve claims in exchange for a cut of Google's revenues. But it also gives Google a unique right to digitize and sell and post ads against "orphan works" - titles whose rights holders have not come forward. Other organizations could negotiate the rights to works in the Registry, but the Registry alone would have the power to set prices.
Following an investigation by the US Department of Justice, Google and the case's other parties are now working to rewrite the settlement, but Mountain View is pushing ahead with its plans to offer up scanned works to the web at large. Last week, at a conference in Frankfurt, Germany, the company said that next year will see the introduction of its first ebook store, dubbed Google Editions.
The company did not respond to our request for comment on the announcement. But according to a report from Bookseller.com, Editions will allow netizens to purchase books directly from Google and read them on any device with a web browser. Google will keep 37 per cent of the payments, and the rest will go to the publisher. The company will also offer versions of the service that allow retailers and publishers to sell books directly. If a book is published through a third-party retailer, the publisher will take 45 percent of the payment, with the remaining 55 per cent split between Google and the retailer. It's unclear how the split would work if the publisher is selling directly through the service.
The service would provide a "cloud library," where buyers could access their texts online after purchasing - provided they can get online. But once they're accessed on a particular device, copies will be cached locally.
Meanwhile, the likes of Amazon, Sony, Barnes & Noble, and others have opened their own online ebook stores, and each has, shall we say, its own ebook philosophy. Whereas Google says its Editions service will serve up titles for reading on any web browser, for instance, Amazon's store only offers titles encoded with the company's DRM-ified .AZW format and the unprotected mobipocket format on which AZW is based. And there are countless other formats in use, from various proprietary creations to the open ePub format to good ol' Adobe PDFs.
'A messy landscape'
"We've created a messy landscape," the Internet Archive's Peter Brantley told the not-for-profit's Make Books Apparent conference in San Francisco, California.
"If I'm looking for a specific title, I don't really know where I should be looking for it, and I have to be concerned about the format as well, because not everything works with everything else. And then I have to actually get the book and put it on a [dedicated book reader] hardware device or other things I'm reading on."
Bookserver seeks to provide a common architecture for use across disparate stores and disparate devices, including not only PCs, laptops, and Kindle-like ereaders but also game consoles and mythical hardware like Steve Jobs' Apple tablet.
"We people really want to be able to find the book they're looking for in the format they know they can use for the devices that they have," Brantley said.
"And book distributors... they want to make books available for people to find, with accurate descriptive data, so people know what they're getting, and they want to this available in any many different places as possible, as many different channels as possible - under the terms they have agreed to for the licensing and acquisitions of their titles."
In pitching this idea, Brantley couldn't help but reference the DoJ's letter to the court over Google's $125m settlement with authors and publishers. Justice proclaimed that book data should be available "in multiple, standard, open formats supported by a wide-variety of different applications, devices, and screens."
"So we know we know we're on the side of good," Brantley said.
The basic idea is to create a web of books much like our web of, well, web documents. "In the same way we're created the massive and inspiring web of information online, where anyone can publish something and make it accessible through search," Brantley continued, "we want to create a web of books, where people can expose information about books online and make it available for people to discover, to harvest, to obtain, to lend, borrow, and buy, and put [texts] on devices of their choice."
Bookserver grew out of work done at Lexcycle, the ebook reader outfit purchased by Amazon earlier this year. Lexcycle began using an ebook catalog system based on Atom, that son-of-RSS standard. With this distribution system, third-party book sellers could serve up their titles to Lexcycle's ebook reader, Stanza, which runs on Apple's iPhone and iPod touch as well as Windows PCs and Macs.
Then, in April, just prior to the Amazon acquisition, Lexcycle teamed up with several others - including Adobe and the Internet Archive - to transform this distribution method into a standard: OPDS.
Bookserver begins with today's OPDS, but the Internet Archive and its partners intend to take the platform much further. Brantley tells The Reg that the aim is to eventually tie the architecture into third-party payment services, so that purchasing books is as easy as browsing them.
Clearly, Amazon is involved with the project - representatives of both Lexcycle and the Kindle were present at today's conference - but it remains to be seen whether Google will join the effort. Traditionally, Google has embraced the sort of lightweight distribution system defined by the OPDS system, and O'Reilly Media publishing technology engineer Keith Fahlgren tells us he's had some sort of discussions with Google over the Bookserver effort. But he too has a wait-and-see attitude.
Like Google, the not-for-profit Internet Archive is busy scanning texts - though on a smaller scale. Brewster Kahle says that the organization now has 20 scanning centers in five different countries and that it now hosts over 1.6 million titles. About half of those are actually Google scans of public domain works that netizens have then uploaded to the IA.
Other partners on the Bookserver project include not only O'Reilly and Feedbooks but Adobe, book distributor Ingram, the University of Toronto, and the One Laptop Per Child Project (OLPC). ®