The followup to the wildly successful Netflix Prize has been canceled due to privacy concerns. From a blog post by Neil Hunt, Chief Product Officer for Netflix:
In the past few months, the Federal Trade Commission (FTC) asked us how a Netflix Prize sequel might affect Netflix members’ privacy, and a lawsuit was filed by KamberLaw LLC pertaining to the sequel. With both the FTC and the plaintiffs’ lawyers, we’ve had very productive discussions centered on our commitment to protecting our members’ privacy.
We have reached an understanding with the FTC and have settled the lawsuit with plaintiffs. The resolution to both matters involves certain parameters for how we use Netflix data in any future research programs.
In light of all this, we have decided to not pursue the Netflix Prize sequel that we announced on August 6, 2009.
This is very sad news. The first Netflix Prize did a lot of good in showing how combinations of different approaches can work better than any individual approach, and also in showing how difficult it is to find economically significant models for some types of recommendation systems. I, along with many, had looked forward to the new insights (and possible visibility for operations research) another Netflix Challenge would bring.
For now, we are left with predicting murders in Philadelphia. Unless someone finds privacy concerns there.
There is another good other prize competition:
The international nurse rostering competition 2010 (the follow up on the international timetabling competition 2007) just started:
http://www.kuleuven-kortrijk.be/nrpcompetition
Though, it could use some more attention it seems.
I think this is great news: when I rent a movie from netflix, I have an expectation of privacy, and in the first competition, they shared my movie rental records with the whole world (not in raw form, but in a form that enable easy deciphering). The “privacy concerns” were not idle worries, but based on real attacks that were demonstrated. In my opinion, the argument that I have nothing to hide is as weak as saying that it’s ok to put everyone under house arrest on snow days, since I wasn’t going to go out anyway. I am glad that my rental records will not be made public again.
While we will all be delighted to get insights from data, I think it should be up to me to decide whether to share my rental records to help research or to help improve recommendations.
Kunal – I’m curious what real attacks were demonstrated?