Site Indexes

In a recent Friday Feast, Shirley notes that the new version of A List Apart could benefit from having a site index. A site index is a book-style alphabetical index, such as that compiled by Adaptive Path for PeopleSoft. All information laden sites like ALA would benefit from an index, however I know from experience that much time, discipline and effort is required to keep an index useful: back in 1996, I created a site index for (the now woefully neglected) Sizzling HTML Jalfrezi, one of the very first HTML & CSS guides.

To maintain a site index for a magazine site like ALA, the author (or preferably a trained indexer; hah!) should scour each new article to extract entries and cross-references would for the index. As explained in Improving Usability with a Website Index, there’s much more to indexing than pulling out a dozen keywords.

Even with a content management system to auto-generate the index from disparate entries, the idea of retro-fitting an index to the existing 300+ articles in ALA is daunting indeed. That said, perhaps the vast readership of ALA could embrace the collective nature of the Web and build a site index collaboratively. Each article could be presented in ‘indexing mode’ where every section (as identified by a heading) could be indexed by readers.

A reader could associate a word or phrase pertaining to the section in question. This index entry could then be stored in a database and displayed with the appropriate links in an index. Obviously while the index was still in its infancy, its usefulness would be small, but I’d love to see how effective the index would be after a few months of contributions.