Footstreams: Clickstreams for the Physical World

People vote with their feet. An individual person visits places that are in some way important to that person; he or she does not visit the vast majority of places at all. This simple observation makes clear that there is substantial information contained in human activity in the real world. To date, that information has been, for all intents and purposes, inaccessible. Happily, though, for the first time in human history, location-based services provide us with the ability to capture, in digital form, the places people go. And “places” does not mean just the lat/longs, the cities or zip codes or neighborhoods. For dramatically improved contextual value, we can capture which businesses or other points of interest individual people visit. This data set is the real-world analog of a clickstream in the Web domain; in fact, we might call it a “footstream.”

The promise of a world in which footstreams are commonplace is vast. We will be able to view human activity in the physical world as digital information, much the way we can view human activity on the Web. In its simplest form, a person’s footstream can be thought of as a story of one’s life. Family and friends can live vicariously through the actions and voice of another person. When two people have an experience together, the story can be told collaboratively, (e.g. there can be two versions of what happened). These stories are captured while the experiences unfold, they are indelible, and they can be shared as narrowly or as broadly as desired.

Stepping up a notch in complexity, we can analyze patterns of visits across many people and intersect the data with the history of an individual, to create highly relevant recommendations such as places for that individual to visit (e.g. “people who visit the places you visit also visit this place, and you have not been here yet” or “your friends like this place, and you have yet to visit it” or “people who visit this place often go to these places afterward.”) We can aggregate footstream data in real time to display a heat map of activity occurring right now, and we can show past activity as well as forecasted future activity. These are examples of how footstreams serve to enable an application that to date has been impossible to build: serendipitous discovery for the real world.

Close examination of footstreams and the value they unlock have already begun to create a new class of application at the intersection of augmented reality and reality mining. In order for these new and innovative applications to be successful, however, one must commit to three basic conclusions:

1. People’s voting with their feet will start to be digitally tracked (which implies people will become comfortable with this). 2. In aggregate, the “footstream” data can be used to drive serendipitous discovery in the physical world. 3. Annotating your personal footstream provides your context to the world, which is an important value for users.

And, all of this leads to a vision of the web of the physical world.

The presentation will share explicit data taken from the Whrrld community showing how aggregating footstream data in the right way leads to serendipitous discoveries based on user data from our product (e.g. how many places we recommend for users in their home town, what the follow-through rate on the recommendations are, how recs change when a new friend is added or the user adds a new place to his/her footstream, etc.). It will also provide compelling examples of annotated footstreams creating value. Historical footstream data has incredible value if utilized in the right way — the specific examples and insights discussed today will help to illuminate a new model of where the physical world web can go.