The collection of diaries, weblogs, and status updates found online represents a vast, unsolicited source of information on the thoughts and feelings of millions of people. The nature of these writings lends itself quite naturally to automated sentiment analysis, for the purpose of understanding how people of different countries, climates, sexes, and age groups are feeling. Do emotions spread across the network of blogs with a dynamic similar to viral media? How does happiness vary with age? We are using data from wefeelfine.org (created by the artists Jonathan Harris and Sep Kamvar) and twitter.com to pose these questions and others.  Among other observations, we have found that happiness rises and then falls with increasing age as well as distance from the equator.  It peaks on holidays and weekends, and appears to be growing worldwide over the last several years. We have also found that music lyrics have become more negative over the last 50 years, primarily due to the influence of punk, metal, and industrial music.  Our results have been published in the Journal of Happiness Studies and covered in the press by Science magazine and the New York Times, among others. We have also set up the website One Happy Bird to track happiness trends in twitter posts. The ultimate goal of the project is to develop a hedonometer, a real-time remote sensor of population level happiness.  We also hope to develop novel methods for measuring how humans influence each other’s behavior and understand the collective dynamics governing the decision making process. This is collaborative work with Peter Dodds.