The short (somewhat obvious) answer, aside from errors introduced by OCR: because publishers cut corners by laying off proofreaders and copy editors, then rush the manuscripts out too quickly for their skeleton crews to catch gaffes. At least that’s what one editor confessed to Virginia Heffernan at the New York Times. Another editor, however, says in the era of word processors authors have gotten lazier and stupider: “It is amazing how little review seems to have occurred before the text is sent to the editor. Seriously, you have no idea how sloppy some of these things are.”
Still, if you’ve ever felt that a publisher wasn’t taking the ebook edition seriously, perhaps it will please you to know that typos are increasing in print editions as well. Heffernan noted two stinkers, “Buckminster Füller” and “habberdasher”, in recent first edition hardcovers.
Read “What Typos Mean to Book Publishing” at the NYT’s Opinionator blog.
A lot of the ebook typos I’m seeing have nothing to do with the authors or even spelling. They are strictly formatting issues. The last couple of China Mieville novels I read had the first five or six words of every new section in small caps. I’m reading my second Charles Stross novel and both have the first letter of each new section in lower-case. It’s like reading my kids’ text messages.
I still don’t understand why, in an era of word processors, these books need to be OCRed at all. Why not just convert to the doc format to epub/mobi etc?