The index is one of the most valuable—and most popular—features of software documentation because it can lead users to exactly the right information. This chapter covers index methodology: how to decide what to index, levels of detail, phrasing, building, and proofreading.
Plan Your Indexing Strategy
There are two methods for creating an index: manual, in which you make notes as you read through a document, or electronic, in which a software indexing feature identifies terms.
Manual Indexes
To create a manual index, the document must be essentially complete so that page numbers are correct. The steps are as follows:
¨ Review the user analysis to remind yourself of how the software will be used
¨ Read or scan the page for index entries (main topics, concepts, and tasks; tables, figures, captions, examples, definitions, acronyms, and abbreviations; or menu names, tool buttons, and keyboard shortcuts)
¨ Record all index entry locations in a word processing file
¨ Alphabetize, edit, and format the index for consistency and usefulness
Electronic Indexes
An embedded indexing function is part of most word processing software programs.
These allow you to highlight terms and assign them to categories. It’s possible to index as you create the document, and then compile the index when it’s finished. Advantages to this method include automatic alphabetizing, automatic formatting, and easy revisions. The process is similar in some ways to manual indexing:
¨ Review the user analysis, keeping this in mind as you create the document
¨ Mark index entries as you write or edit
¨ Build the index: select the location and use the software to create the index
¨ Edit the index to eliminate inconsistencies, double entries, missing entries, etc. (Note: corrections must be made in the text, not the index.) Format, if necessary.
Decide What to Index
Try to make your index “support the activities and actions the user will undertake in applying the program to workplace needs.” Have developers, users, and other writers review the index specifications. Consider commands and functions (menu terms), concepts related to subject matter, user synonyms for program terms and proper names, user questions, glossary terms, and tasks and procedures.
Identify the Level of Detail
A very simple one-level index organizes the table of contents alphabetically. A standard two-level index includes main headings (categories), subheadings, and locator numbers. Three or more levels are very complex. An index may also be described by the number of “indexable” items per page of the document: a light index has 2-3 items, a medium one has 5-7, and a heavy index has 8-9 items per page.
Decide on Phrasing and Format
Writers need to plan the phrasing and formatting of index entries. Cue the primary locator numbers by bolding, capitalize terms consistently, make entries sound like sentences, and cue special terms.
Edit and Proofread
Carefully edit and proofread the completed index for mistakes such as indentation errors or inaccurate page numbers. Ask a user to review and test the index, if possible. Consider what content to emphasize and how to group information.
Discussion
Because indexing is often left until the end of a project, indexers may feel pressured to finish quickly because of publication deadlines. However, an index accesses the document from the user’s point of view and directs the user to the information they need. It is useful to beginners and advanced users alike.
A print index, located at the back of a manual, contains terms printed alphabetically in a two- or three-column format. An online index lists terms alphabetically and allows users to either scroll through the list or click on a letter from an alphabet display. A keyword search allows users to select a word from a list or enter one manually; a search engine then matches it to related topics. The writer is responsible for creating the electronic links. This type of search is often more limited than other types of indexes.
Microsoft Word, Adobe Framemaker, and other software programs have indexing capabilities. Some programs also work with HTML files. Problems occur when search engines retrieve too much irrelevant information to be useful, because they examine the entire text or web page. “Meta” tags can be used to locate certain responses identified by the web designer. However, information retrieval is complicated by partial words or misspellings.
Indexers rely on reference resources such as dictionaries, thesauri, encyclopedias, fact books, phone directories, organizational listings, geographical maps, and other collections of information (for example, science, medicine, business, law, agriculture, or computer terminology). As Barker notes, “These kinds of resources can help the indexer make the crucial matches between the vocabularies of a software program (represented by the words that make up the interface) and the terminology used by and familiar to the user.”