Lucie Haskins—Jumping on the embedded indexing bandwagon—or should I? (ISC conference 2015)

Embedded indexing is still evolving as the relatively new ebook industry finds its legs. Ebook indexing is so new that it’s a bit of a Wild West, with different software, standards, and processes competing for space. Clients may hear the buzzwords and turn to you for answers. Should you make the jump to embedded indexing? Lucie Haskins looked at some of the issues you should consider when deciding.

Unlike back-of-the-book (BoB) indexing, in which you receive designed files, either in hard copy or PDF, from the client and write an index in RTF or DOCX format, which the client then typesets, embedded indexing is done in the native file, whether it’s in Framemaker, Word, InDesign, XML, or HTML. You tag the text with index terms and send the file back to the client. In Haskins’s words, “You receive their baby, you manipulate their baby, and you send it back to them. It’s a huge responsibility.”

Some limitations of native indexing modules

Creating terms

  • No index preview
  • No autocomplete of index entries
  • Tiny marker boxes
  • Poor control of special strings, such as page range, italic or bold formatting, and cross-references

Editing terms

  • No change propagation of index entries
  • No index preview
  • No viewing indexing entries in the document
  • No temporary grouping of index entries

As a result, Haskins said that you can expect to spend 50 to 100 percent more time on embedded indexing compared with BoB indexing.

Some benefits of native indexing modules

Creating terms

  • Autogenerated entries

Work process

  • Indexer can start before final pages
  • Indexing concurrent with proofreading
  • Potential reuse in future editions, other formats

Issues specific to embedded indexing

  • access control and time constraints
  • software versions
  • version control on files and downloading/uploading

You and your client will have to discuss what software (and what version of that software) to use. For example, if you and your client are using different versions of InDesign, one of you will have to convert the file to IDML. If you don’t have the client’s fonts, your system will substitute a font that will affect flow and pagination, which means that the final index would have to be regenerated by the client. At that point, the client would have to be responsible for formatting text to italic, because InDesign doesn’t allow italicized text in index entries. Each entry has to be formatted manually. and the formatting disappears whenever the index is regenerated.

Should you bother with embedded indexing? Haskins says you shouldn’t feel you have to, unless existing or prospective clients have approached you directly about it and you have an interest in it. Haskins doesn’t recommend jumping on the bandwagon otherwise, because the field may evolve into something else entirely in a few years. For example, there are hints that BoB indexing using anchors at the paragraph level may be where the field ends up. It would use techniques familiar and intuitive to indexers and would obviate the need for specialized software. Buying all of the software and upgrading your equipment would be a significant investment of money; educating yourself and your client on the software and the process would be an investment of time.

If you do want to learn embedded indexing, however, Haskins suggests