Ask HN: Can Big Data Cure Disease?

I had an idea tonight. What if 100,000 people tracked everything they ate for four years -- and a team of researchers mined that data for correlations with certain health outcomes. Would this yield significantly useful information? Because I was thinking tonight about ways the data could be collected. College students might gladly opt-in to a special low-priced meal service provided in exchange for their participation. Maybe large-scale volunteer organizations (like the Peace Corps) might find people willing to participate in the study, just as a way to help identify health risks. Maybe there could be a new movement -- instead of MOOCs, we have could massive online volunteer opportunities... There's even ways the government could get involved (with voluntary participation from servicemen or prisoners). Even before you get to the possibility that someone might actually try to privately fund the massive study with an eye to eventually benefiting from its data...