> (how do you write software that can tell the difference between a 'real' link and a spam link?)
I was more thinking along the lines of Google's perhaps defining a class of Web sites as "spam magnets" and mandating that they use "no follow", with non-compliance resulting in no more spider visits.
There are, after all, any number of Web sites, FTM included, that are not dedicated to product recommendation and whose purposes do not include product recommendation other than on a casual basis, and it seems like ignoring links posted on those sites would not skew Google's rankings but might put a significant dent into link spam.
Identifying the sites that belong in that class wouldn't be easy, but the folks at Google are a pretty creative bunch.
By the way, is "no follow" strictly a Google thing, or do other engines' spiders pick up on it?
