TITLE: Single-Source Indexing

AUTHORS(s) & AFFILIATION(s): Jan C. Wright, Wright Information Indexing Services, Inc., Sandia Park, New Mexico, and Caroline Parks, Indexellence, Inc., Sandia Park, New Mexico

KEYWORD(s): Indexing, Single-Sourcing

PRESENTER / CONTACT PERSON: Jan C. Wright/Caroline Parks

CONTACT EMAIL: Jancw@wrightinformation.com; cparks@indexellence.com 

ABSTRACT:

As more and more content is being produced and distributed in multiple formats, the issue of providing indexes for various formats becomes important. Indexes can be considered as residing in interfaces, rather than just more pages accompanying a printed piece, or a tab stuck in a online help file. Designing indexes that work equally well in any of the various interfaces a document may be displayed in, whether print, help, PDF, or on the Web, presents a real challenge.

Understanding the structure and relationships available in each destination format allows the indexer to design the index to work well in each instance. Ignoring an output format or assuming that the index is a simple construction leads to poorly designed online indexes, in which one format has been sacrificed for another s output needs. In addition, print indexes do not translate well to online without consideration of screen design and user behavior.

In this paper we discuss the interface indexing design issues for print, online HTML Help, PDF, and plain HTML. We outline the problems and benefits for each interface, and discuss what tradeoffs and compromises must be made to make an index work well in each environment. We discuss the structure and components of indexes that become problematic in multiple formats, introducing such terms as one-to-one interfaces, one-to-many interfaces, active and passive entries, and stubheads. Real-world illustrations of index content help demonstrate how much the different index interfaces can affect the index.

Last modified October 28, 2001 by Scott Tilley.