One Parent, Two Strategies
by Doug Williams
Information Mapping ™ and DITA both derive from cognitive science research about how people read. But one child of this research applied predominantly to the world of print, and the other to hypertext.
Information Mapping was an influential content information strategy from the 1990s that many of us were taught. While the company went bankrupt in the “oughts” (its assets were bought and are being repurposed for DITA, with greater or lesser success), it lives on in the tendency of many writers to create navigation tables at the top of their content. In Information Mapping, these types of tables were the information strategy of preference to use. They were called “overview map” topics, which provided a point of prospect for readers in the form of content lists. Here are the original guidelines (1996) for their use, which even then noted that there was an online issue:
When to use
For paper, an overview Map should always introduce a series of Maps in a volume, part, chapter, section, or an entire document.
Online, you do not generally use overview maps. The design of the online document an the way the user accesses information eliminates the need. Instead, rely on the organization of the document and the Block labels to advance organize your audience.
The Information Mapping guidelines were derived from cognitive science research, and broadly hold true today. In fact, block labels are essentially the same idea as a DITA container topic (also known as an orientation topic in some parts of Oracle). But that nuance of usage didn’t prevail for many legacy publications, or for writers taught the overview map/navigation table way of thinking. In the 1990s, Oracle customers could order printed books, and these physical books were the primary access method for documentation. Designing an overview map as a point of prospect made sense. However, in our online world, even content optimized for printed use, such as PDF, is at least as commonly used online for its capability of keyword searches as it is used printed out. And online, the block labels are always available to the reader, assuming you have correctly organized your content under user-goal focused labels. You can find the block labels/container topics/orientation topics on the left-hand navigation tree of the PDF or HTML documentation. Users can select the sections of content that they require, and read or print it out selectively.
The function of a summary overview is perfectly valid: It is an expression of a key principle of usability design, the point of prospect. If there is additional content overview required to provide users with a summary of a task, or API list, or set of concepts, then you can use a table as a checklist to provide an additional point of prospect for readers. Just don’t include the “more information” column providing links to your documentation contents, or to external references within our release collection content. That function is better served by the container/orientation topic, with its auto-gen link list, and its always-current headings and short descriptions picked up from the target topics.
Old style block label topics, in the form of chapter topics or section topics with manual link lists, are also obsolete. Maintaining manual link lists increases your workload to maintain the xrefs, makes the topics that contain these lists unreusable in other contexts, and is unnecessary when you use the DITA Open Toolkit as it is designed to be used.
Leveraging the modern capabilities of the DITA block label that we call container or orientation topics, and the DITA-OT Auto-Gen link list, is something that both makes your reader’s life easier, and builds on solid reading research to facilitate content chunking, F-shaped reading patterns, and a variety of other things I’ll leave for another day.