Software for indexing

Less than twenty years ago, most indexes were created by writing data on index cards and then editing and sorting the cards. Although cards may still be used occasionally, today most indexes are created using software of one kind or other.

There are three software-based approaches to creating indexes:

1. Embedded indexing, in which the program used to create a document is also used to embed index entries in the document itself. The entries are embedded in the document by delimiting the beginning and ending points of portions of the text and assigning headers to them. The index itself is generated by issuing a command that causes the program to sort the headings and generate the listing. Adobe® FrameMaker® and Microsoft® Word® are the programs used most often for embedded indexing. The advantages of embedded indexing are based on the fact that, when using it, locators are virtual - i.e., they do not exist until the index is printed or displayed. Therefore, if you need to change the size of the page or to print the document in one size for an American audience and another for the rest of the world or to also print a large type version, you do not have to re-index the document because text may move from one page to another when you do those things. Nor do you have to re-index it if you add or delete text. You can also begin indexing earlier in the publication process because you don't have to worry about shifting page numbers. For example: you can begin indexing chapter 7 before chapters 5 and 6 have been written. And, when there is a major revision, These are important advantages in environments where time to market is critical and where documents are changed frequently or need to be printed or displayed in multiple formats. Generally, those are environments that produce lots of technical documentation. The advantages are great, but there are also some disadvantages. The major disadvantage is that embedded indexing is often more time-consuming than other computer-assisted approaches to indexing, especially when editing the index. That is partly due to the fact that the index is not visibly updated as entries are added one after another. The second major disadvantage is that embedded indexing generally requires strict change control because both writers and indexers may have access to the source file. Without it, the indexer may be embedding entries in one copy of the file while the writer is revising another copy. That, as they say in the trade, is a bummer.

2. Using a word processor or desktop publishing system to create an index for another document. One of the advantages of this approach is that it can be used when the indexer cannot access the source file for the document to be indexed, as is frequently the case when freelance indexers are used. Another advantage is familiarity and reduced costs due to the fact that the software used to compile the index is almost certainly used for other purposes as well. The disadvantages, are numerous and may be overwhelming. Some of the potential problems can be solved relatively easily. For example: word processors typically default to automatic hyphenation, which can cause confusing line breaks in indexes. That problem can be easily solved by turning off automatic hyphenation. Other problems, however, are not so easy to solve. Most of them pertain to alphabetization. Alphabetization is a problem with word processors because typically sort text based on ASCII character codes. In the ASCII coding system, all upper case letters are sorted before all lower case letters. That means, for example, that all headings beginning with an upper case A would be positioned before any heading beginning with a lower case a. In indexing, upper case and lower case letters are considered to have equal value during alphabetization, meaning that sorting should not be case sensitive, as it is in ASCII sorting. It may also be difficult to sort multiple levels of entries properly or to switch between letter-by-letter and word-by-word alphabetization. Extensive manual intervention may be the only solution to problems related to alphabetization. Those who seldom compile indexes may find that acceptable; those who frequently compile them won't.

3. Using a standalone indexing program, i.e., a program created specifically for the purpose of compiling indexes. People who seldom compiled indexes are likely to see only disadvantages when looking at such programs: they don't automatically generate indexes, any more than word processors automatically generate term papers; they cost money; they take time to learn; they can't be used for anything else. Why bother with them? To professional indexers, the answer to that question is obvious because they see the advantages that come from using application-specific software that includes facilities needed to complete work done with the least time and effort. Typical standalone indexing programs include facilities for:

• Placing entries in the correct position as you make them
• Toggling between letter-by-letter and word-by-word alphabetization
• Specifying character strings to be included during alphabetization, such as the spelled-out version of a number
• Specifying character strings to be ignored during alphabetization, such as the leading definite or indefinite article in a book title
• Specifying alternative sort schemes for subheadings, such as sorting in page-number order
• Handling various page numbering schemes concurrently, such as Roman and Arabic numbers, and ordering them properly
• Specifying a conflation scheme for page numbers - e.g., specifying whether a page range of 150 to 155 should appear as 150-155 or 150-5
• Specifying the placement and punctuation of cross-references
• Checking cross-references for matching terms and to ensure that there are no blind or circular references
• Viewing subsets of records, such as records created during the current session or records falling within a given page range
• Toggling between indented and run-in styles
• Flipping main headings and subheadings
• Searching for entries matching a pattern you specify
• Merging entries from several indexes or splitting an index into two or more indexes
• Compiling multiple indexes concurrently, such as a name index and a subject index for a book
• Minimizing typing by automatically completing terms once the term's shortest distinguishing sub-substring has been entered

The major standalone indexing programs are:

	CINDEX by Indexing Research 180 Varick Street Suite 1014 New York, New York 10014 Platforms: Windows 95 or higher; Power Macintosh running OS 8.0 or higher Home page:   www.indexres.com
	MACREX by Macrex Indexing Services Beech House Burn Road Blaydon-on-Tyne Tyne and Wear NE21 6JR, England Platforms: Windows 3.1, 95, 98, NT and DOS Home page:   www.macrex.com
	SKY Index Professional by SKY Software 350 Montgomery Circle Stephens City, VA 22655 Platforms: Windows 3.1 through XP Home page:   www.sky-software.com

There are also special-purpose application programs to assist indexers in their work. Some facilitate tasks that may arise when indexing any type of work; some facilitate tasks that are unique to a specific type of indexing, such as legal indexing; most work in conjunction with one or more standalone indexing programs. They include:

•   CaseAbbrev   - ensures that names of legal cases conform to the Blue Book style for abbreviations and removes pinpoint cites in legal indexes
•   CaseRev   - generates reverse entries for case names in legal indexes
•   emDEX   - a FrameMaker plugin that adds a real-time visual means of watching your index grow right before your eyes
•   EM/Index   - splits CINDEX files into separate files for printing, typesetting or electronic usage
•   EM/Index   - splits CINDEX files into separate files for printing, typesetting or electronic usage
•   EntryExpander   - creates multiple entries from a single entry that list multiple co-authors, keywords, etc.
•   HTML Indexer   - a tool for creating and maintaining a back-of-the-book index for Web sites, intranets, and other HTML documents, in classic HTML, HTML Help, and Javahelp formats
•   HTML/Prep   - converts CINDEX files to HTML-coded Web pages
•   Index/Check   - performs style checking on CINDEX files
•   Mapper   - updates CINDEX files by applying lists of old and new locators
•   RepCite/Check   - validates the content and order of reporter cites in case citations in legal indexes
•   SKY Filer   - information management software for Windows 3.1 and higher
•   SumDex   - validates complex cross-references
•   Xelect   - assists in specifying, storing and applying rules for generating derivative indexes from an existing index

These special-purpose programs are used almost exclusively by professional indexers or technical writers.

Guidelines for using indexing software:

• Try the demo version before purchasing the product. All vendors provide free, downloadable demo versions that allow you to explore the functionality and interface of their product by creating small indexes.
• Apply the same evaluation method to each product you try. The functionality of the major standalone indexing programs are similar. None of the demo versions can be used to create indexes large enough to meaningfully evaluate the program's performance. Therefore, the method you use to compare them should be heavily biased toward issues related to ease of use, such as minimization of the number of times your hands leave the keyboard to use the mouse, minimization of the number of keystrokes required to achieve an effect, the quality of the accompanying documentation, your comfort level with the underlying metaphor, etc.
• If you will need to regularly use the software, become an expert user ASAP. Play with it. Try everything. Re-create the most complex indexes you can find. Become efficient before you're under a deadline.
• Participate in the user group or discussion list for the product you select. That is a great way to learn more about it and to get assistance in solving problems at any hour of they day.

To top of page