Producing an eBook – How hard can it be?

At the beginning of this year, Maureen MacGlashan, editor of The Indexer, came up with the idea of publishing collections of Indexer articles based around particular themes. We had already made every single issue, right back to 1958, available online but being told ‘there’s 54 years of articles – what you want is in there somewhere’ is understandably a bit daunting. Having the best and most relevant articles selected for you would be useful.

At first, we thought of publishing the collections as paper books through the print-on-demand company, Lulu, which I have used for a long time and through which we recently started publishing single issues/back issues of The Indexer. Looking at the costs, it became clear that we could achieve the same income for the The Indexer but with much lower prices for purchasers by producing an ebook. Essentially, we eliminated the cost of printing, packing and shipping heavy paper. I had already made a Society publication, OP5 Indexing Children’s Books, available on Kindle, with a linked index, in 2011. We would need this new book to be available on all platforms, including for those without eReaders, but how hard could it be?

The Kindle file formats are based on HTML, so they are essentially long web-pages and the Kindle behaves very much like a browser. I have been indexing websites for a long time and wrote a book on it in 2006 so I was on familiar ground there. To publish on all platforms I decided to produce an ePub file which could then be converted to a Kindle file. The ePub file would work on iPad, Nook and most other readers, and could be read on a desktop machines, both PC and Mac, by using Adobe Digital Editions (ADE).

To allow Maureen to do the indexing I devised a system using numbered paragraphs which would then allow me to write software which would link the index automatically. This appeared to work, until I checked the index out on the iPad. The iPad and Kindle work in quite different ways. Instead of operating like a browser, the iPad formats the text into a particular pagination and that pagination stays fixed unless something major changes, in which case a new pagination is constructed. So, if the font size is changed, or the iPad is rotated from portrait to landscape, the entire book is repaginated and the new pagination becomes the fixed pagination. With a Kindle, the formatting is done a page at a time, so if you jump to a particular point, by using a bookmark or an index entry, then that point will be positioned at the top of the screen. With the iPad, the target point could be anywhere on the page, even the last line. My paragraph numbering scheme had pointed to the start of paragraphs, which worked wonderfully on websites and the Kindle, as the relevant paragraph appeared at the top of the screen, but was disastrous on iPad. The first index entry I looked up, I couldn’t find anywhere. It turned out to be an 8 line paragraph, only the first line of which was on the page I was directed to, and which had the index topic in its last line – part way down the next page. Consequently I had to go through every one of the 905 locators, consider where it should be positioned and move it to the precise position.

Even tasks like proofreading proved problematic. When I first checked I saw in the first article:

"I began my professional career as a bibliogra-
pher selecting books in certain subjects for the
University of Toronto Library."

All good, until I rotated the iPad and saw:

"I began my professional career as a bibliogra-pher selecting
books in certain subjects for the University of Toronto Library."

Most of the material had come from previously typeset material rather than original electronic text, and end-of-line hyphenations in the original were wrong (both ways around – sometimes words like cross-reference had lost their hyphens when not on a line boundary). Furthermore, each device handles a separate set of formatting commands, so that formatting which works on one device might not work on the others. Italics, for example, might work by using an italic font or by skewing a Roman font, depending on what fonts are available on the device. There is no universal set of formatting guaranteed to work on all. Handling of images is particularly problematic – apart from anything else, the iPad handles colour and has a very high-resolution screen, whereas Kindle e-ink models are monochrome. So proofreading had to be done on the iPad twice, in portrait and in landscape, and again on the Kindle and again on ADE.

One author actually wanted to proofread their own article – actually, not at all an unreasonable request as their article involved explanations of XML coding and the formatting was crucial, but I suddenly had to work out how to have her see the text and how she could communicate any changes back to me. I extracted the chapter and created a new, stand-alone ebook, hoping that I included all the formatting so that it appeared the same as the complete book, but that didn’t provide any page numbers or fixed reference points. I created instructions, explaining what to report and what not to report, and produced a PDF printout of the text. The PDF wasn’t in the same format as the ebook, so couldn’t be used for proofing, but she could use it to mark her changes on.

Finally the ePub and Amazon files were ready, and I had just to upload them, but still I wasn’t finished. I uploaded the ePub file, only to have it rejected with a “failed 1.1.0 validation” error – which meant as much to me as it does to you now. After a lot of research, I tracked down the validation being used, downloaded version 1.1.0 of the appropriate software from an archive site, ran it (from the command line) and (using Unix piping) managed to get out the list of reasons why it was failing. The two “errors” would have been perfectly valid in a normal HTML document and simple to fix, and so I uploaded again. This was successful, but two days later, I found that the book had been rejected from the iBookstore because the subtitle did not have the correct capitalization. Apparently it has to be title-case! That was fixed and the ePub book had arrived.

Uploading the Amazon book had no problems with the uncapitalized subtitle, but, bizarrely, they inverted the colours on the cover! I still don’t know why. The colours went from orange to blue when I uploaded it, but I thought that was because the image was the “currently selected object”, so inverted, but when it arrived on the Amazon store, it was blue.

At last, the book is published – Newcomers: A Selection of Articles for Those New to Indexing is available. The price varies according to format, but the book covers everything those starting out need to know, including indexing efficiently, choosing software, producing a quote, negotiating with clients, and more. We will start work soon on the next book. This time it will be much easier. After all, second time around, how hard can it be?

Originally published online November 6, 2012, and in SIDelights (Society of Indexers Newsletter)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s