The content of a character data section is not parsed by the XML parser, so any tags it may include are not recognized as tags. This makes it possible to include special markup delimiter characters directly in the section without using their entity references. <! [CDATA [ content ] ] >