Even though you can rent Million Dollar Baby from Netflix (NASDAQ:NFLX), now is your chance to earn it. Yesterday, the leading provider of DVD rentals by mail announced a contest for data-crunchers who are up to the challenge of improving on the company's algorithms for member recommendations.

With more than five million subscribers, the $1 million in prize money would be chicken feed. What company wouldn't be willing to pay less than $0.20 per member to get to know its user base a little better, providing a more valuable service along the way?

Is this a publicity stunt? A genuine effort for improvement? Is Netflix nuts for putting its users' anonymous ratings out there for anyone to benefit from? Is Netflix even more nuts for offering to share its findings by publishing a "detailed description of the winning approach" that may enrich its rivals?

Leave it to Netflix to turn burning questions into Molotov cocktails. True to cinematic form, these are also the ideal seeds for a cliffhanger.

For the love of publicity
By releasing the data behind 100 million movie recommendations, Netflix is clearly flexing its muscles. It's a message for less seasoned rivals like Blockbuster (NYSE:BBI). In fact, Netflix wants the world to know that these 100 million ratings represent less than 10% of its current subscriber base (and cover less than 18,000 of the 65,000 titles that Netflix stocks). That big number is really just a tiny sample of the data that the company's own Cinematch software has been digesting over the years.

The contest will run for at least five years, or until some statistics-loving nerd is able to win the big prize by proving a 10% improvement over the company's own system. If five years for a deadline seems a little out there, consider that Netflix is basing its contest on the British government's Longitude Prize, which was initiated in 1714 to improve transoceanic travel. The winner there didn't emerge for 47 years.

In the end, of course, the contest is genuine. Who wouldn't want a better algorithm? You really couldn't think of a better recruiting tool, either. It's like the Google (NASDAQ:GOOG) billboards that posed a brain-busting challenge. The solution was a web address that provided a back door into the company's job application process.

If Netflix is able to milk a little publicity out of the process, that's really just a welcome bonus. The problem is that, like so many things in corporate life, things can go wrong.

Spitting fire can backfire
The timing of the contest could have been better. Too many people still recall the blunder made by Time Warner's (NYSE:TWX) America Online this summer, when it leaked the search request data of 650,000 AOL subscribers. Even though the AOL data was anonymous, a little journalistic sleuthing made it easy to put a name to some of those anonymous users.

Naturally, you won't be able to name names this time. Good luck trying to find that one person who loved Mona Lisa Smile but tagged Walk the Line with just a single star. It can't be done. You and I realize that much, but leave it to a Nervous Ned or Nellie out there to start some online petition, erroneously chastising Netflix for divulging private data.

This isn't the only way that this contest can backfire. What if Cinematch is exposed as a lightweight? What if several aspiring statisticians can improve on the algorithms, and the distant competition is injected with life at the possibility of marketing itself as a smarter, more effective alternative to Netflix?

Amazon.com (NASDAQ:AMZN) had such great respect for the Netflix model that it chose to launch a similar service in the United Kingdom instead of charging ahead stateside. Wal-Mart (NYSE:WMT), a retailer that barely backs out of a fight, bowed out of the DVD-rental-by-mail business and handed its tiny base of users to Netflix.

Terms of the contest state that the winner must divulge to the world how it was able to improve the process, and hand over the secret sauce to Netflix in a non-exclusive event. In other words, the winner is free to negotiate with the competition in offering the process that outsmarted Cinematch.

The power of community, revisited
Netflix wants to be a good Samaritan. It also wants to be a brainier one. Leveling the playing field accomplishes the former, but not the latter. There is value in the data that a successful company mines over the years. This can be readily seen at Amazon, where the site's interface now knows what to recommend to you based on past shopping patterns.

Our new Motley Fool CAPS stock rating beta service also cashes in on what former Fool writer Jim Surowiecki calls "the wisdom of crowds." That's been part of the art of the Netflix model. I have been a subscriber since 2002, and all the time I've since spent rating flicks has helped to ensure that I've never been lured away by the cheaper Blockbuster service. The end result of my click-happy labor has been the discovery of some pretty amazing movies, titles I would have never unearthed on my own.

If this contest makes Netflix better, awesome. However, just as the discussion board that Netflix opened is bringing up some pretty rival-propping queries (like someone wondering whether they can combine the data with Amazon's IMDB.com feeds in search of a better solution) let's hope that Netflix knows what it's doing here. Once you let the genie out of the bottle, it's pretty hard to trick the genie back in.

Netflix , Time Warner, and Amazon have all been recommended to Motley Fool Stock Advisor newsletter subscribers. Wal-Mart is an Inside Value stock pick. Whatever your investing style, try the Foolish newsletter that's right for you free for 30 days.

Longtime Fool contributor Rick Munarriz has been a Netflix subscriber -- and shareholder -- since 2002. He is also part of the Rule Breakers newsletter research team, seeking out tomorrow's ultimate growth stocks a day early. T he Fool has a disclosure policy.