A-Z Index Examples

Here’s a collection of A-Z index examples on UX Refresh.

Generally, I’m a fan of A-Z indexes. But at the same time I realize they are really difficult to create and maintain, particular in dynamic online settings. So the real value of them remains elusive to me. I don’t think I’d really try too hard to convince someone they need an A-Z index to organize information in a digital space.

That said, I did make a point in my presentation at the Euro IA Summit 2008 in Barcelona that things like indexes and taxonomies make sense within bounded domains–more so than in open domain contexts. (See also a summary in the ASIST Bulletin: “Navigating the Long Tail.”) Even Clay Shirky agrees with that. Here’s my point:

As we collectively move down the long tail, bounded domains–or niche markets, as Chris Anderson calls them–will increase and solidify, and so we will also see an increase in the need for indexes, taxonomies, and ontologies to help organize these domains.

So maybe there’s hope for A-Z indexes after all. In fact, I recently came across an excellent implementation of an A-Z index not included in the collection summarized on UX Refresh: EMBASE, a bibliographic database from Elsevier. From the Embase website:

EMBASE.com is a biomedical and pharmacological bibliographic database, which provides access to the most up-to-date citations and abstracts from biomedical and drug literature via EMBASE and Medline. It contains over 19 million indexed records from 7,000+ peer reviewed journals, covering 1947 to date, with more than 600,000 additions annually.

EMBASE is indexed using the Elsevier life science thesaurus, EMTREE and Medline records are mapped to EMBASE before adding to EMBASE.com.

The interesting part about it is that the index is integrated into the auto-complete suggested terms feature from the main search field–with “use:” references and all:

Auto-complete suggestions are most often alphabetical anyway, so this makes a lot of sense. And since biomedicists become familiar with the standard terms in their bounded domain, their understanding of the index should be fairly high. I’d even predict that people would expect to have access to standard index terms in this context.