The XML Information Set

The XML Information set, also termed the infoset, is an abstract data model for most of an XML document and is intended to be used by other XML-related specifications from the W3C.

An information item is similar to a node as used by DOM and XPath. However, the detail of how an XML document is modeled differs from both DOM and XPath.

The XML information set is the newest of the data models specified by the W3C but is viewed as the foundation for many future XML-related specifications produced by the W3C. For example, the infoset is used as a basis for the W3C XML Schema language, together with some augmentation of that infoset to express notions specific to the activity of schema validation.

Which Data Model?

The fact that the W3C has released three data models that model the same thing—an XML document—can seem confusing. In practice, the choice of data model to use depends on what you want to do.

The DOM is used when you want to create, manipulate, or delete parts of an XML document using programming languages such as Java. XPath is used primarily to navigate the in-memory representation of an XML document, to make selected parts of that document available to other XML-based application languages such as XSLT, the XML Pointer Language, and XForms.

The information set is used in W3C XML Schema and is proposed as the basis for the XML Query Language (XQuery) and version 2.0 of XSLT and XPath (currently under active development at the W3C).