The growing diffusion of GPS enabled cell phones coupled with the widespread adoption of social networking tools is opening up new avenues for the study of human behavior. Using a large dataset of geolocated Tweets we present an analysis of both at the temporal and spatial aspects of human communication and interaction. By coupling the analysis of geocoded tweets over a period of two years with tools for automatic language detection we are able to present a large scale study of the geography of language use around the world at scales ranging from country to neighborhood level and how it varies over the course of a year due to seasonal variations of tourism. Finally, we show how the volume of information available in online systems permits the real time gathering of quantitative indicators anticipating the future unfolding of opinion formation events.