Even if you don't live in the United States, you've likely heard of Netflix, the Internet-based DVD rental service responsible for ushering in the final days of Blockbusters and Hollywood Videos on every suburban street corner. For those who haven't used it, it's a very simple service: Add movies to your queue and within one business day, whatever is at the top of your queue magically ends up in your mailbox.

Simple service it may be, there's a lot of science (and magic) involved. Aside from the logistical aspect of managing, tracking, and dispersing millions of discs nationwide every single day, Netflix has the daunting task of being responsible for recommending movies based on your taste. However, the corporation's robominds are, to their own admission, fairly linear in their logic... Liked Dawn of the Dead 2004? You'll probably like 28 Days Later... So in an attempt to get a leg up on any future competitors, Netflix created the Netflix Prize in 2006.

The rules were simple: The first person or team to create a new collaborative filter algorithm that improved the recommendation accuracy of Netflix' own algorithm by at least 10% would take home $1,000,000. It may be a testament to how good Netflix' in-house system already was, but it took nearly 3 years for a team to make enough progress to claim the million dollar prize. Even then, deciding the winner came down to the wire; the second place team submitted their bid 20 minutes later than the winning team, which must be maddening in the scope of a 3-year long competition.