By Kurt Ament, infotektur.com
Information modeling is extreme document planning. Like traditional document plans, which outline strategies for developing usable documents, information models outline strategies for developing re-usable document components. If you restructure information for re-use, you make it usable in multiple documents and formats. With this in mind, you can leverage information models as “usability multipliers.”
From usability to re-usability
When developing a document in the traditional document development process, you follow a document plan, or blueprint. This blueprint describes how you will structure your document (for example, the table of contents), how you will deliver it (for example, as a printed manual or online help system), and what usability standards you will follow (for example, your corporate style guide).
What happens to your blueprint when you decide to re-use information? The moment you decide to re-use information, your strategic goal changes. Before choosing re-use, your goal was to develop usable documents. Now your goal is to develop information that can be re-used in different documents and formats. Of course, you want the resulting documents to be at least as usable as their predecessors. That is, you want your information to be usable in any context.
Unfortunately, traditional document blueprints focus on one document and one format, from beginning to end. These document-based blueprints are too specific to plan re-use in many documents and formats. What you need are generic blueprints that define information elements that can be re-used anywhere. These element-based blueprints are known as information models.
From blueprints to models
Collectively, information models define your re-use strategy. Before building information models, you make an inventory of the your existing documents. You look for chunks of information, or elements, you can re-use. You then classify the elements by type (for example, procedures). Some elements serve as containers for other elements (for example, procedures contain titles, introductions, and steps). The size of the elements you re-use determines the “granularity” of your strategy. The smaller the elements you re-use, the more you have. The more elements you have, the easier they are to re-use, but the harder they are to manage.
When structuring these elements, you associate meaningful categories, or “semantic elements,” with generic publishing tags, or “base elements” (for example, headings, paragraphs, and lists), as well as metadata (for example, cataloging information) and architectural information (for example, re-use guidelines and maps). This process often involves complex SGML- or XML-based tools (for example, Adobe FrameMaker+SGML) and templates (for example, DocBook) that separate content from format. You use these tools and templates to assemble elements into different documents in different formats, as needed.
Separating content from format is not just a technology issue. It is a technical communication issue. Stripping away document-based format reveals the element-based structure of your content. For example, your procedures may contain titles, introductions, and steps. If you standardize each element in this structure, you are one step closer to re-using any procedure in any document. Although your procedures may have many authors, you can structure them to speak with one voice. Consistently structured content fits in anywhere.
From elements to standards
To ensure that shared content can be integrated seamlessly into any document or format, you can integrate writing standards into your information models. For example, you might specify that procedure and process titles begin with verbs, and that conceptual and reference titles begin with nouns. Likewise, you might make active voice, present tense, and second person mandatory for all text-based elements. And you might replace document-based terms (for example, “above” and “below”) with electronic cross-references.
Conversely, you can integrate information models into your style guide. For example, you might set up guidelines for writing consistently structured procedure elements. You might specify that each procedure title must begin with an infinitive, that each procedure introduction must point explicitly to steps, that each step must describe one action only, and that all steps must point in the same direction. Such parallel construction ensures that elements written by many authors appear to have been written by one author when re-used side-by-side in the same (assembled) document.
When developing writing guidelines for information models, it is wise to focus on your worst-case scenario: online help. People have been exposed to online media for only a few decades. As a result, online helps systems are not nearly as fault-tolerant as printed manuals. In effect, users are forced to view content in flashcard-sized chunks. This limitation has a silver lining: information that is usable online is even more usable on paper. By structuring your information for online viewing, you increase its usability in print.
From theory to reality
To succeed, information models require the expert advice of technical writers like you. Only you really understand how and where your information can be re-used effectively.
If your organization is starting to model information for re-use, you owe it to yourself to get involved in the process. At the very least, you can vet draft information models early, providing the information modeling team with much-needed reality checks that correct unworkable models before they are hard-coded into your information infrastructure. If you are more ambitious, you can help build your own usability standards into the information models themselves. In the process, you may even develop a new career niche for yourself.
Kurt Ament is a senior member of LASTC, and the author of two books: Single Sourcing: Building Modular Documentation and Indexing: A Nuts-and-Bolts Guide for Technical Writers.