Types of eBook Indexes

Missing index: no index exists. This is often the case when the publisher converts a book from print to e-format and assumes the reader can just “search” for what they want. If the user doesn’t know exactly what they are looking for or how to spell it, they’re out of luck.

“Dead” index: the index that existed in the print book is simply reproduced. It looks exactly like the print index, but does not contain hyperlinks to locations within the book. The only advantage of a “dead” index over a missing index is that it provides an outlined overview of the material in the book and can assist the user in coming up with search terms.

Hyperlinked index: entries or page numbers (or section heads) are hyperlinked to locations within the text. Now we’re getting somewhere. There are several varieties of hyperlinked indexes.

• Index headings and subheadings are linked to locations within the text. This works as long as each topic only comes up once within the book (which may be the case with certain reference books). Otherwise, it is insufficient, because there is no way to map to multiple locations.
• Index page numbers (from the print index) are linked to locations (usually the top of what corresponded to the numbered “print” page) within the text. This is often the result of “converting” the print index and adding hyperlinks automatically (which many of the larger conversion houses offer as a service). This works OK for pdf (fixed-page) format where there are fixed page numbers, but falls short with reflowable formats (Kindle, ePub) where there are no fixed page numbers. The problem is as follows: suppose you click on index page number “62.” This brings you to a location that corresponds to the top of page 62 in the printed book. If you are reading this book on your iPhone, you may have to scroll through dozens of screens to find what you’re looking for. Moreover, you have no way of knowing when you’re reached the end of “page 62.” This can be very frustrating for the user.
• Index page numbers are hyperlinked to an exact location in the text. This is obviously preferable to the previously discussed method, but it does require placement of anchors (tags) at specific locations within the text. This job is usually done by the indexer, since it requires some knowledge about how indexes work. If the tagger prepared the print index, they are already familiar with the index terms.
• Hyperlinked section heads (following index entries) are used in place of page numbers. The links can be either to the beginning of a particular section or a location within a section. This approach is often seen with HTML or other web-like layouts. The advantage of this approach is that the index does not rely on “meaningless” page numbers.
• Hyperlinked indexes can be either static or dynamic. This is a characteristic of how the index is produced, rather than the final result. A static hyperlinked index is produced by simply adding target anchors in the text. For example, an indexer can hyperlink index page numbers and insert target anchors within the text. This is basically what you get when you have one of the larger conversion houses “convert” the print index. A dynamic hyperlinked index, on the other hand, is based on index tags (rather than generic anchors) placed within the text. These tags can be used to dynamically generate the index. The advantages of this are several: 1) the index can be customized for different formats (e.g., hyperlinked index page numbers for pdf, hyperlinked section heads for HTML); 2) sections within the text can be moved around and the index will regenerate automatically; 3) sections (individual chapters) of the book can be removed, and the index entries for that section/chapter will be intact. To get the full benefit of dynamic indexes, the tags should be inserted in a “common-denominator” format like XML. XML transforms (like XSLT) are used to generate the index (as well as the entire book or ebook).