I\'ve got an XML document containing news stories, and the body element of a news story contains p tags amongst the plain text. When I use XSL to retrieve the body, e.g.
If you have control over the input document, CDATA is the right way to go.