Search engines targeting content other than hypertext documents require a crawler that discovers resources identifying files of certain media types. Naive crawling approaches do not guarantee a sufficient supply of new ...