Ebook Formats

One of the more interesting challenges in producing ebooks is the file format question. On the surface, it looks pretty simple: two formats, EPUB and Amazon’s Mobi dominate. Both (loosely speaking) use HTML tags and CSS stylesheets to determine how text and graphics are displayed. So an ebook is a collection of webpage-like pages in a container. Or is it?

When you open an ebook in a reading device, the reader rendering engine displays the pages. The rendering engine is similar to a web browser, and is where the problems begin. If you have been using the Internet for a while, you are probably aware that each web browser works a little differently from all the other web browsers. Internet Explorer does not always render pages the same way Google Chrome does, for example. Some of the rendering differences are subtle; others are substantial; and all are a source of pain for those of us who code web pages.

Web browser rendering problems are “known” in the sense they are well documented and workarounds have become fairly standard. One adds some code to the page header to detect the browser being used, and the code compensates for the deficiencies in that particular browser. The website visitor may not see the page exactly the way you would like (boxes with rounded corners might turn square; shadows bordering images might vanish), but the page is still a comfortable read.

Rendering engines for ebooks are not as well known as web browsers and do not support the kind of code we use on web pages to adjust the presentation. So the first problem one faces is, do you create a different version of the ebook for each ebook reading device (Apple, Kobo, B&N, and so on), and for each model variation in their current line of devices? Or do you code for the lowest common denominator?

Our current approach is what I believe to be the only sensible one: we code for the lowest common denominator. So our EPUB books are coded with the assumption the reader is using an older model, monochrome e-ink reader with limited ability to display fancy typography. This allows for a single EPUB file to be distributed to all the vendors who sell the format. We do the same with files for Amazon: a single file which our testing shows will work on all Amazon reading devices.

Towers of Babel

If the format problem stopped there, everything would be fairly simple. But it does not stop there. A new version of the EPUB standard, EPUB 3, has been ratified and we are expecting all reader hardware makers to begin releasing new reading devices with the ability to render this format around the middle of 2013. EPUB 3 is very attractive. The specification allows for better formatting (read: more complex), better control of typography, and the inclusion of media such as audio and video clips in an ebook. Sounds and looks good.

Naturally, there are problems. The large technology companies who produce and sell the most popular reading devices not likely to release rendering engines that conform 100 percent with the specification. I have little doubt that each will do it a bit differently, just as they have done for years with web browsers. Some will add a few proprietary tweaks; others will leave out some elements. And there will almost certainly be no way of knowing in advance who will do what, or when. Which makes it very difficult to prepare and train staff for the transition.

Will the technology companies release software updates for their existing reading devices, so they can display the new format? Some, such as Apple, probably will, because they have a limited number of devices which they generally replace with newer models once a year. Others will not upgrade older models, because their interest is more focussed on selling newer, shinier devices and persuading users to buy those.

My guess, based on what I have seen in the past, is that we will have a transition from the older devices to the newer ones and it will last a couple of years. Many people who purchased an ebook reading device in 2012 will not want to trash it and buy a new one in 2013, and many will still not want to change in 2014. So I think we may have to produce EPUB format books in both the old and the new versions of the format, starting from late 2013 and continuing into 2015, before we can drop the old.


The problem does not involve only EPUB. Amazon released a new format, KF8, earlier this year for the Kindle Fire. It is a bit more sophisticated than the Mobi format (also known as AZW) which Amazon has used since the original Kindle. KF8, as its name implies, is designed for the Kindle touchscreen, color tablets. Those are fortunately also able to render Mobi files, so we have not felt compelled to work in that format. Yet.

As I indicated in my previous article on how ebook devices are changing, the pure ebook reading devices like the early Kindles, based on monochrome e-ink screens, are not selling as well as they used to. Multifunction, color tablets seem to be the future. So while I expect Amazon will continue to support Mobi on their new devices for a number of years, I also expect the time will come when we will want to transition our ebooks to the KF8 format (or whatever enhanced version of KF8 Amazon produces in the coming couple of years). And that means there will be more staff training for that transition.

Wait, there’s more

So far, I’ve dealt with the two primary formats: EPUB and Mobi. There are others. We can discount, say, Microsoft’s old .Lit format as almost no one uses it. There are other old formats too, but they are either dead or dying. There are also much newer formats, and one or two may emerge as important.

The one I am watching most closely is Apple’s iBooks Author: Apple supplies free software (for Mac OS X only) which allows one to create interactive multimedia ebooks in the .ibooks format. Those books can only be sold through the Apple online store and will only work on the iPad. The format is somewhat similar to EPUB, but with proprietary extensions which make the file unreadable as an EPUB book.

Given the limitations, what makes this format interesting? Apple is aiming, broadly, at the education market. The objective is that books produced with iBooks Author software will be textbooks, for use in the classroom and for self-study. That is a very big market, and it is one where the iPad devices are having an impact. Quite a lot of the books we publish for our authors would fit this market. So I am watching it to see how it evolves, and to try to get a sense of the scale of the market. I have done some very preliminary experimenting with the development software and will continue to do so as time permits. It is something we might find worth pursuing in a year or so from now.

There are other emerging technologies in the ebook world, most of which I suspect will not make an impact. But all need to be watched and understood, on at least a “just in case” basis. Our aim is not to lead the pack in digital technologies, we are too small and, frankly, risk averse for that. We prefer to be somewhere in the middle, where the risk of investing lots of time and energy in new technologies that flop is low, and the rewards are reasonably good.

If nothing else, we are living in interesting times in publishing, as the industry goes through a very substantial transition.


Leave a Reply

Your email address will not be published. Required fields are marked *