David Ream and Jan Wright once again took to the stage to elaborate on indexing of digital files. Ream said that there aren’t a lot of usability studies that compare search versus indexing. BNA’s “Using Online Indexes” is one, but it would be interesting to get more universities involved in this kind of research to generate more data.
EPUB 3.0 files will have rich metadata—Dublin Core for publication information, ONIX for supply chain information, and MARC for libraries. The metadata will be key to a digital file’s discoverability—and hence to its sale. Implications for indexers include the following:
- no page or line limitations
- potentially having to index rich media (e.g., time codes)
- potentially having to index interactive ebook features (scripts)
- potentially having to supply semantics of headings and locators (e.g, show only the statutes, show only the people, etc.)
- being able to provide index data in multiple ways
- cumulative indexing—of series, mashups, etc.
Jan Wright then explained the workflow for inserting anchors to EPUBs at the paragraph level. She and Olav Martin Kvern developed scripts that create identifiers for individual paragraphs in InDesign, which can then be used as part of the locator in standalone indexing programs like CINDEX or SKY Index to create a hyperlink from index locator to the paragraph. For now, this is the most realistic way of creating a paragraph-level index that will work for both print and ebook, because InDesign strips out any embedded index entries when it exports to EPUB. The digital trends task force is talking to Adobe separately about this issue, but a fix may be some time away.
Wright and Ream then allowed conference attendees to play around with various devices, from the Kindle, Kobo, and Nook to the iPad, to see the current state of the art of ebook indexing.