With XHTML you have the mixed benefits of being able to parse the structure as an XML document (scanning for heads and subheads for example) while still being able to read it in a Web ...