By Olivia Snaije
Books are a wonderful form of addiction. Last year my addiction increased when I became an e-book reader. Now about half of what I read is in e-book form. Unlike print books, e-books can instantly be purchased and delivered. In my case, most of my shopping is done in bed late at night, which can be extremely dangerous. A click here and there and five new books can be sitting in one’s Kindle with an alluring tab on the side marked “new.”
But there is a downside to e-book lust: As I began reading e-books I noticed something very irritating. There were often spelling mistakes, formatting problems and line breaks or odd symbols appearing instead of letters, which interrupted the pure pleasure of reading seamlessly. The first time I noticed the errors was when I read Lisa Alther’s 1984 novel Other Women in a digital edition. I was so annoyed that I asked Amazon to refund me, and to their credit, they did. (Amazon will also offer customers new versions of e-books when corrections appear.)
A quick look at the Nook, Kindle and Amazon online forums revealed I was hardly alone in my irritation.
The worst affected seemed to be even older books, such as Ayn Rand’s Atlas Shrugged, where a reader complained that the number “1” and the letter “I” were often switched. (Considering Rand’s Objectivist philosophy focuses on the “virtue of selfishness,” perhaps this was merely a fateful, karmic mistake.) But more recent titles have fallen victim, too. One reader complained about problems with Dean Koontz’s Lightning (a book from 1987), another was exasperated that Barbara Kingsolver’s 1998 novel The Poisonwood Bible was “riddled with errors,” and yet another person noted spelling mistakes every five pages in a recent Harlan Coben thriller.
Why — after we’re well into the digital age — are publishers not able to exert rudimentary quality control at a level equal to print? Is it too much to ask?
Perhaps it is. E-books exist in a variety of different formats, which themselves change frequently. Earlier versions of e-book software were inflexible and often difficult to work with and many publishers outsourced their conversion activities to specialized companies. Adding additional steps to workflow and each new time the file is opened invites the introduction of errors and mistakes — and finding out who is responsible for fixing them can be even more problematic.
What’s more, when an older book is scanned to convert it to an e-book, the images are run through optical character recognition software, and then go through a series of steps before they are converted into a basic e-book format. In the case of these pre-digital titles — and Alther’s Other Women would be considered pre-digital — the original copy may have aged, the ink could be faded and blurred, or there could be dust on the page when it was scanned, resulting in text that is difficult for the computer to read.
Recent titles are “born digital” — with the original files in digital formats — making this less of a problem. Or so one would assume.
When I asked Open Road Integrated Media, the digital publisher and multimedia content company responsible for re-issuing Other Women, why so many mistakes persist in e-books, the company — which responded via e-mail through a spokesman — was circumspect:
“Open Road recognizes that this is a challenge and therefore has a stringent process in place and puts in a lot of resources to produce quality e-books. The books go through a thorough proofread and at least two subsequent levels of quality assurance before being finalized. While the technology is certainly new and there will always be some mistakes when you publish hundreds to thousands of e-books a year, we are proud of our track record and will continue to invest in and improve our already thorough process.”
At Aptara, a prominent digital publisher and conversion house, Sriram Panchanathan, senior vice president for digital solutions, explained that mistakes are often introduced in the conversion process because the software used to convert to e-book formats “is very good for handling structured information, but not very good at identifying and processing information that varies and is inconsistent.”
Panchanathan added: “Software will get you 90 percent of the way, but the rest needs to be controlled manually. To ensure top quality you need quality control editors to look at every line and every page…the challenge is when you are converting a large number of books and you’re not willing to spend the hours of quality control.”
Christèle Blay of Jouve, a French content management company that makes e-book conversions primarily for US and UK publishers, concurred, noting that the number of errors often depends on the condition, state and accuracy of the original material (true, it must be noted that many print books also contain errors). Blay underscored the fact that many backlist, out-of-print and out-of-copyright titles are simply being converted with the objective of making the books available once again, albeit in a digital format.
Indeed, in the rush to get these older books back into print and capitalize on this otherwise moribund intellectual capital, it appears that expediency has often been prioritized over accuracy. When a book may only sell a modest number of digital copies and “good enough” will do, proofing likely seems like an unnecessary step.
In the end, Panchanathan summed it up best when he said that in order to avoid errors, there always needs to be a human involved in finalizing the text, which inevitably entails additional expenses.
“You can design software to drive a car but you can’t have a car without a driver…the day cars drive themselves is the day you will find a perfect e-book without the manual touch.”
This, I somehow found reassuring.
Editor’s Note: Yes, we know, you may well likely find an error or two in this piece as well, and any errors will be corrected with alacrity.