Text in a document that should not be parsed is put within a CDATA section. This allows for the inclusion of characters that would otherwise have special functions, such as &, <, >, and so on. CDATA sections can be used in the content of an element or in attributes.