After Tim posted his findings regarding the indexing of flash websites, I suggested to him that he should carry out similar research on PDFs, because you can’t nofollow a link in a PDF.
As it happened he was already running a test, and he has just published the results to his PDF SEO research
Companies already pay for hosted pages, why not pay for hosted white papers or product catalogues?
On the negative side, PDFs when distributed virally might end up pointing to old resources, and it is hard to track them if you use direct links to resources. You should always at the very least use a redirect link with a 301, but preferably something with built in tracking.
Best practice? Always have lots of links in every PDF you create