John Horgan, in a blog on the Scientific American site, warns us against getting too far ahead of ourselves on the Big Data hype. Although he agrees that correlation can sometimes yield usable results, he has misgivings about some of the other claims.