Jan Wright, a leader in the field of ebook indexing, gave the keynote address at the Indexing Society of Canada’s annual conference this morning. We are witnessing a watershed moment, she says, where we are trying to define what the markup for our content should look like, no matter where it is—whether it ends up on paper or on a device like e-reader or smart phone. This development is in its infancy right now, with conflicting formats on different platforms, and Wright is part of a working group of indexers actively involved in shaping the EPUB 3.0 standard to include indexing concerns.
Current ebook indexing is either nonexistent or ineffectual. Ebook indexes may be missing or static, and there are almost no ebook indexes that index at a paragraph level. They are not an integrated navigational tool, they are difficult to get to, and they are hard to browse, especially if they’re typeset in two columns.
Existing platforms try to mimic certain features of indexing, but they don’t provide all of the functionality of a traditional index. For example, iBooks Author conflates an index with a glossary and limits the function of indexes as navigational tools. Amazon’s X-Ray, currently available only on the Kindle Touch, shows all occurrences of a particular term by page, chapter, or book, but it is merely recall—without the precision of an index—and offers terms in chronological order. In other words, it’s a brute force attempt at indexing.
When considering ebook indexes, we have to take into account a reader’s mental patterns and search behaviours. Some readers have never read the book and need to know if it adequately covers a given topic; some have read the book and know that their search topic is in there, but they have to find it. We must also keep in mind that reading styles differ whether you’re reading for education or for pleasure, fiction or nonfiction. Using physical cues, such as the position in a book or location on a page, to locate content, as well as behaviours like skimming, are disrupted in ebooks. Some platforms attempt to mimic a paper metaphor, but really, paper is just another interface. The key is to figure out what each interface does best and playing to those strengths, because the paper metaphor doesn’t carry over well onto a small screen. The danger with today’s ineffective ebook indexes is that they are training the reader to believe they are unpredictable and thus to question why they should bother using them at all.
The ideal ebook index has features that have been implemented in other contexts before and so should be completely feasible. Wright gave us a demo of what an effective ebook index should do. It should be accessible from every page; the “Find” feature should reflect the best hits, as identified by index; it should show the search results with snippets of text to offer context; it should allow cross-references to help you refine search phrasing; and it should remember that you’ve been there before and let you go back to it. Ebooks would also allow for additional functionality, like bringing up all indexed terms in a highlighted swath of text in a kind of “mind map” that offers additional information showing how concepts are connected.
So what can we do now? First, Wright says, is to get ready for the eventual use of scripts and anchors in EPUB 3.0. A goal is to develop a way to add anchors or tags to content at the paragraph level, which would allow for hyperlinking directly to the relevant content. Once prototypes of the interactive ebook index have been developed, we must assess their usability to ascertain what’s best for readers.
A big takeaway from this keynote speech is that advocacy and outreach are essential. With the standards at a nascent, malleable stage, this is the time for indexers to have their concerns addressed as the technology develops so that indexers’ workflow can be taken into account. (But more on this in a later post.)