Published: 2022-08-31
The Crowd API
An API providing information about how crowded locations are
We already know that our cellphones and other devices are used to track our location. Companies like Google, Apple and our cell phone service providers are constantly updated with our whereabouts. How about sharing some of this information with developers who can build new services and apps? More privacy intrusive location data from our phones is already commercially available, so I guess there are no legal issues stopping an API like this. Setting a minimum of 10 people to identify a crowd should prevent abusing the API to stalk individuals.
Google already use our aggregated location data to determine when restaurants and other establishments are most crowded. A great feature, but a coordinate based API would open up for others to create cool applications and features as well. It is when the crowd data is combined with other data the real magic can happen. Other developers will probably have far better ideas than the ones I present in this article.
The Crowd API in our everyday life. Visualized on a map of my home town Nyköping, here are some ideas how the API could be used.
1. 250+ people seem to be stuck somewhere along the railroad track where there is no station. Train will probably be delayed. Time standing still, distance to your location and other factors compared to earlier scenarios can provide more details about the delay than you get from the train operator.
2. Compare the number of people before and after the security check on the airport over the last few hours to calculate the average time spent waiting in line. Use this information to calculate when you have to be at the airport.
3. Over two hundred cars standing still on the highway. If you are in your car somewhere close your app will look up incidents and alert you to take another route if needed.
4. 1500+ people at Rosvalla arena. New visitor record for our local team!
Other apps could...
Combine data from several locations and suggest where to go for the quickest lunch. Probability thereof at least.
Suggest reserving a seat on the train if it seems more crowded that usual before it arrives at your station.
Suggest the best time to go to IKEA based on previous crowd data compared to the weather forecast.
Community planning
Authorities, city planners and corporations could make great use of a Crowd API to visualize behavior of the inhabitants as they plan for the both near and distant future.
In a longer perspective crowd data can be used to suggest future infrastructure updates. Overlay people’s actual movements with existing streets and paths to find out new ones should be placed or when to schedule infrastructure maintenance. Crowd patterns can also be used to calculate were new restaurants are likely to succeed or where to place a new office to suit commuters behavior.
In a shorter perspective crowd data can be used to deploy resources to handle of traffic incidents or crowds of hooligans drifting around town. Hospitals can prepare to bring in extra staff. Fewer people at work could mean COVID is on the rise.
Accelerometer data
My friend Gustav had an idea where he wanted to combine real time crowd data with cellphone accelerometer data to get a map over where the best dance parties were held. Including accelerometer data actually a fantastic idea, it would open up to even more possibilities. I had an idea where I would use accelerometer data combined with location data to measure the snow conditions in the alps and locate the best powder. Over time and currently. I also realized that off-piste skiers would never voluntarily share this information. They would find a way to opt out and keep their best runs a secret. On the roads on the other hand, I assume people would gladly share a combination of location and accelerometer data as it can help plan road work and locate pot holes that needs to be filled.