We have investigated the practical implementation of a database model that can represent DOM node trees, and hence is able to store the content of any well-formed XML document. Since we are working at a higher level of abstraction than a schema-specific model, query building can prove more complex, but it does have the advantage of genericity. The merging of DOM with the Nested Sets Model allows us to navigate the document without recourse to walking the entire node tree. This feature allows for fast serialization and querying, but at the expense of slower updates (since insertions and deletions may require updates to the (x, y) coordinates of many nodes).
The major benefit of this exercise is that it provides the relational database programmer with the opportunity to experiment with XML data storage without recourse to implementing much (if any) new technology, and to do so in such a way that the XML can sit alongside traditional relational data without the need to build additional interfaces. The generic nature of the repository means that we are not constraining ourselves to any specific schema?a real advantage to those programmers that are not yet sure exactly what XML can do for them, or how it will fit with their future applications.