5 Basic Components of Your Content Model
Once you have completed the content audit (see our last post), you can build your content model. Your content model should include:
1) Which topic types to use in which situations (such as task, reference, or concept)
2) The specializations you have decided to build
As we noted in our last post, you should consider your use of specializations carefully – they may lead to complexity in the future. Use specializations when they will provide significant benefits in expressing the meaning of your information, ease the authoring of your documentation, and facilitate publishing.
3) A strategy for organizing maps (which represent your publications), sub-maps and content references (conrefs), with the ultimate goal of re-purposing content.
This should also include:
a. A convention for standardized usage of conditional attributes and attribute values
b. A convention for standardized usage of linking, eliminating uncontrolled cross-linking between files, using methods such as map hierarchies and relationship tables
4) A metadata model or subject scheme model which defines your taxonomy and how you will classify your content. For example, taxonomy and classification will allow your content to be filtered for different products, information types, and user roles in knowledge base and help systems.
Metadata is very important not just for ensuring the accessibility and searchability of your content, but also for your style sheet development – defining how your content will appear, whether in a PDF, online help, or in a mobile ePub. For example, there may be many types of publication information that you would like to include in your PDF or other publishing formats. This information needs to be populated in your DITA files using metadata. Your metadata model often impacts how your CMS and other tools should be configured.
5) Templates for new topics and publications so that your authors can get up and running and create new content as quickly and easily as possible. Using the right tagging and the right content model (from the last step) will help you not only make the process easier for new authors, but also make the maintenance of your style sheets a lot more efficient. When you use a very consistent content model, it is relatively easy to enforce the consistent usage of tags by using templates, schematron and RELAX NG rules. This allows you to drive down maintenance costs and reduce headaches when you go online with your style sheets.
Reality Check: Your information architecture and content model will continue to evolve. More often than not, down the road there will be different outputs that you want to provide, whether a new type of content like release notes or a different format like HTML5. Your content needs will change as time goes on. Your content model is a process, and your information architecture will evolve as you migrate more content. You will need to stay in touch with your information architect to ensure your system is evolving in the way you want to go.
Are we finished? Of course not! Tune in next time for “Conversion to XML: What to Do First.”