iBookstore and the Format of eBooks

Increasingly I rely on my Twitter stream for arresting issues in publishing technology. First off, yesterday was a thoughtful blog post at Semantico on the choice facing publishers on whether to go into the iPad platform via individual apps, or via the iBook store. Richard Padley's conclusion:

For straightforward chapter based book content it seems clear there is no longer a compelling case for publishers to deliver e-books as apps. The extra cost of software development, combined with the slowness and lack of scalability in the approval process no longer make sense now that Apple have introduced the iBookstore. (iBooks or Apps? The Publishers Dilemma)

If this conclusion is warranted, it presumably follows that for books that are not 'straightforward chapter based' there remains a compelling case for going down the books as apps route. In point of fact, a very large proportion of books are not straightforward and chapter based. ePub is not a happy format for books with lots of illustrations and tables. So that should keep us busy at Exact Editions. But something else follows from his point: publishers and even worse, readers are going to have to make choices. We are going to expect our audience to read books in one way, as eBooks, on the iPad if they are simple books, but in another way, in another format, say digital editions as apps, if they are not so easy. The experience of reading books will become increasingly fragmented.

Sigh! Life would be a lot simpler if publishers were to consider whether all books might be readable straightforwardly as digital editions (so probably best delivered as individual apps) and then readers would not have to get used to reading books in one way (if they are the kind of book where it does not greatly matter how the page looks and is laid out) and in another way if the layout and design of the book matters. The really odd thing about this, is that the devices are getting better at displaying books as books that we recognise. Because the iPad has a much more generous screen, the need for texts to reflow, or to rescale on the fly, is much reduced. Most book pages on the iPad will be very readable as is.

Furthermore we do not yet know quite how the iBook's reading and display interface is going to work. There are lots of different ways of presenting ePub files. From another tweet, a link to an article in which Hadrien Gardeur notes

(I am disappointed by my).... first glimpse at the iBook's typesetting. "There's not even hyphenation on the page," he said. "If you're designing a reading system I think it's much better to offer optimized typesetting and really create something that's beautiful and easy to read rather than trying to replicate pages in a real book."

Although most readers don't think in terms of kerning and leading, Gardeur's concern was that when they start reading, they'll be able to tell that something's wrong, even if they're not sure why. (from Mediashift -- Dan Brodnitz)

I think its becoming increasingly clear that the ePub format is not going to work equally well across all the many devices that the eBook proponents want it to travel. Part of the reason for this is that the standard file format was designed to solve a problem of how to make 'reasonably straightforward chapter based' books flow and reflow across multiple screen-based systems. The original specification did not allow for the fact that scores of different and somewhat incompatible reading engines would be implemented so that the same text looks so different across different platforms. Here (from some email that Michael Jensen has allowed me to quote) is a heartfelt groan about this inconsistent rendering:

In my own experiments with .epub and other formats, I have yet to find a way of presenting, dependably, any visual model that works the same in Calibre as it does in Stanza/PC as it does in Adobe Editions as it does in Stanza/iPod....

We're still in the late-1990s world of "the same HTML webpages displaying differently in IE than in Netscape," as far as I can tell. Different proprietary software interpreting the same file leads to consumer confusion.

That leads to lowest-common-denominator design: sequential, linear presentations. Sure, leading and line length matter. But c'mon, if I can't have a callout? A wrapped-around image? A dropcap? What a mess, and it's more the rendering software than the format itself.

We are far from making the e-version anything but a pale imitation of the print -- because e-reading software is still in its infancy. For straight prose, it's fine. Anything more? bleh.

My .03 -- we are far away from anything that my ancient typesetter genes would recognize as smart presentation. Perhaps that will change, but I'm betting it'll be 18-24 months before quality and readability becomes dependable, across "e-book browsers."

One can get even more of this pain by reading Mike Cane's accounts of his heroic but frustrating struggles with the ePub format. I think that there is a chance that Steve Jobs is going to decide that ePub is not such a good idea after all (Adobe had a good deal to do with the ePub format and Adobe is Steve's current whipping boy. Can somebody gently point out to Mr Jobs that ePub is a kind of Flash for text -- a flexible but plug-in solution, which complexifies by over-simplifying, where we no longer need it?). Stick with pages! After all, an iPad really is a pad, which holds virtual pages, and Apple has a perfectly good page-oriented word processing program called..... Pages. Admittedly, page breaks in books are arbitrary but they are better than arbitrary implementations of standard files by an unmanageable and ever growing collection of rendering engines.

If all digital books have pages, life will be even simpler if they are all apps. We can then start doing clever stuff with the pages, like linking from pages in one app to pages in another. Will we call that a cross reference or a cross app? Or a page reference from within an app? The idea makes perfect sense. Apps should aspire to them.