Full-text indexing of resources on the World Wide Web is limited to simple content types, such as HTML and plain text. More complex content types, such as Postscript, PDF and proprietary word-processing formats are excluded, despite the fact that such documents are usually rich in content.