Citysense: Lets You Know What Everybody's Doing

citysense everybody

What is everybody in my city doing right now? Sense Networks aims to answer that question for you with their new product Citysense (more info). Citysense shows you how active locations are right now (top screenshot) and which ones are abnormally active (second screenshot). In addition to these sexy visualizations they included an alarm lock that will wake you up earlier if the city is busier than normal before your commute (set it for 7:30 and sometimes it will get you up at 7:28, but on busier days you might be hitting the snooze at 7:21). In exchange for this functionality Citysense will upload your location to its servers. Citysense is still a Beta product and right now it is available only on the BlackBerry (though we can expect an iPhone app shortly) and only has data for San Francisco, CA (there are 6 more cities to come; Chicago will be next — I am jealous on the behalf of all Seattlites).

unusual activity

Sense Networks is the company behind Citysense. They are hedge fund backed and based in New York City. I sat down with co-founders Greg Skibiski (CEO) and Tony Jebara (Chief Scientist and Director of the Columbia University Machine Learning Laboratory) for several hours last week to learn more about the company and their future plans. The company was started after Greg, previously a software architect before the Dotcom Bust, got to know the financial industry. He decided to supply them with data about people’s activities.

Sense Networks uses origin and destination data from cabs to model the city. The Citysense app will feed into the system. The original plan was to use data from mobile usage or other TBD sensors. However, they were able to get taxi data and decided to test it out. They found that the data could be used to get a very clear picture of how a city ebbed and flowed.


Their primary product is Macrosense. This is their money-maker. It’s a dashboard for hedge fund managers. It gives them insights into the comings and goings of consumers (more retail, how far did people come from). In the future Macrosense will also be used for site-selection and retail analytics. Citysense is really a proof-of-concept to see how consumers use this technology and data (it’s also an experiment in gathering their own data as opposed to relying solely on cab companies).

For something that’s just an experiment there’s a lot coming down the pipe for Citysense. Tony, the Chief Scientist, is modeling the flow data and has noticed a pattern in the way people move around a city, where they go. He’s determined that there are 7 different groups (or tribes) of people. Each group is represented by a color and each one has it’s own type of destination (for example Greens are more likely to end up in the hospital after a night out). Very interestingly those destination-types translate across multiple cities — suddenly it becomes a lot easier to figure out where to go when you are in a new city. You can watch two Processing visualizations that shows the tribes moving around SF (one, two ).

The current release of Citysense answers the question “Where’s Everybody?” In the future you can expect Citysense to start to answer the question “Where’s everybody like me?”. If it answers correctly (by your estimation) then you’ll know that Tony’s research has paid off. Tony has agreed to present his findings in a technical session at ETech 2009.

Sense Networks is taking personal data collected management very seriously. One of the co-founders is Sandy Pentland, Director of Human Dynamics Research at MIT. Citysense is a Reality Mining app (something that we experimented with at Where 2.0) and the data collected reveals a lot about a person (beyond just their tribe). As the head of privacy for Sense Networks he set forth the following principles:

  • People should own their own data
  • People have a right to privacy
  • People should receive a meaningful benefit in exchange for sharing data
  • Aggregate anonymous location data should be used for common good

Because of these principles you’re able to delete the last 24 hours of your location data or all of you location data at any time. Sandy, Greg and Tony wisely realized that this type of control would be necessary for any potential users. If you do delete you data you are still able to use the app. Note that if you delete all of your data Citysense will have to relearn Who’s Like You (very similar to the retraining you’d have to for your Tivo if you started with clean slate).

There’s going to be a lot of location-based applications released in the next two weeks (because of the Apple App Store opening). I think that most of them will focus on your friends and social networks. I will be surprised if any of them are able to, right out the gate, are able to provide this type of view into a city to their users, but I, as an iPhone owner am glad that I finally have access to sensor network data from my phone.

tags: , ,