Use structural and content-based elements to enhance your data

It's important to key on structure and content when modeling your XML data. Remember that part of the value of XML is that it separates content from format, enabling you to write your content once, and then apply formatting through stylesheets to reuse it for multiple outputs.

Perform content analysis to determine what category a given piece of information falls into. If a component is content-specific, give it an element name that reflects the content. For example, an instance describing a car might have an element called with an attribute value of "type (standard | automatic)." On the other hand, general information in the same document that doesn't key to a specific component of a car may simply be tagged as a

, which is based on structure, not specific content.

If you formalize your elements into a standard nomenclature, you can then use that vocabulary as a guide when creating new, related data sets. For example, a drug manufacturer may use the phrase "indications and usage" in a document for healthcare professionals, but the term "uses" in information for patients. But both cover the same information. So, if they use the element in the schema for one instance, they should use the same element name in the other as well. If the content type is the same, the element name is the same, even if the intended output may vary. This enhances the ability to search, combine, link across instances, and repurpose information as needed.