Before moving on to text patterns and mixed content, the interleave pattern deserves more attention. As already noted, calling these content models unordered is misleading. Although no order is required by the schema, the nodes will be ordered in instance documents. The order in which they appear in the document can be significant to the applications.

Going back to the example of first and last names: any application managing names will need to know which is a first name and which is a last name. With a little additional effort, they can get the information about whether the first name comes before or after the last name in a XML document. The friendliest of these applications might also want to know whether you prefer to be called by your first or last name first. Do you need to add an additional information item to the schema to carry this information when you could just rely on the order of these elements in the instance document?

In other words, defining content using interleave patterns can be seen as degrading the usefulness of a schema because it looks like the information about the order in which the elements were found will be stripped from the document. That isn’t a real problem; XML processors will still present all the order information to your application. In fact, a content model defined with interleave patterns allows more combinations than a content model that uses group patterns. Thus, with its additional combinations, the interleave patterns can let document creators provide additional information that would otherwise disappear into a fixed structure.

The one downside to using interleave patterns is that the freedom with which they can be used is unfortunately specific to RELAX NG. If you need to insure that it will also be possible to model your vocabulary with a more rigid schema language such as W3C XML Schema, you will often have to restrict the usage of interleave patterns in your RELAX NG schemas.