Having said that, I found this worrying:
We hope to use our findings to develop a software application that could be installed on home computers and mobile devices. It would monitor your Internet usage and alert you when your usage patterns might signal symptoms of depression. This would not replace the function of mental health professionals, but it could be a cost-effective way to prompt people to seek medical help early. It might also be a tool for parents to monitor the mood-related Internet usage patterns of their children.
One of the things I've noticed over the period of creating a lot of applications and sites is that any interaction with a computer can be interpreted in the opposite way from which it was intended. (I nominate this as Markham's Law of Technology Reciprocity) Upvote an article on HN? You're not just tagging articles for quality, the system is also tagging you to see what kinds of articles you upvote (whether or not that information is used or not is not important here). Have a lot of tabs on your browser and switch between them? You're not just engaging sites with the browser; the browser is also engaging you on how you spend your time. Show Facebook who you would like to invite to be a friend? Facebook is watching who you pick as folks you want to associate with socially. You are not just interacting with the net; the net is closely watching you. It's a two-way street.
This can lead to some scary places. Having a computer monitor my actions and "help" me identify my emotional state and mental health could be awesome -- or it could be a tool that lets outsiders climb inside my head in ways no human has ever dreamed of before. If you think keeping your DNA information secret is important, you'll be having heart palpitations when you start thinking this stuff through. There's this world of data mining out there that mostly has to do with your internal thoughts, beliefs, and emotions. I'll repeat: I think this is a good thing. But hell if I'm happy about the trend in general.