If you’ve ever dreamed of conquering the world, but lacked the military budget to make it happen, Takeover might be the next best thing. I built this app alongside Mitchell Hart and Pat White to transform the world map into a massive, location-based game where players can compete for territorial dominance in real-time, all while getting some exercise in the process.
Genesis of a Geo-Conquest
Pet projects have always been a good way for me to experiment with new technologies, and explore ideas that might be outside of the day to day work. Takeover emerged from a conversation about the game dynamics of paper.io, and how cool it might be to make paper.io, but IRL. Most fitness apps focus on personal stats and achievements, but we wanted something that would tap into a more competitive nature while encouraging exploration of the world around us.
The concept was straightforward, but ambitious: create a global playground where players could claim “Turf” by physically moving through the world and completing loops. The result would be digital graffiti on top of a global map – you’re essentially tagging real-world space as your own by encircling it.
The mechanics are simple: open the app, start a loop, and physically move to create a closed path. Once completed, that area becomes yours. The larger the loop, the more territory you claim. But there’s a catch – other players can steal your territory by overlapping their loops with yours. This creates a constantly shifting digital landscape of ownership that mirrors real-world movement.
One of the mechanics from paper.io that we initially brought in, but later removed, was “Busts.” In paper.io you can end a player’s game by running into their open path.
It didn’t seem fair that a person could be on an hour long bike ride or run, and have their Take ended by someone who takes a few steps out of their front door. There may be another variation on this mechanic that we may bring back in the future.
Building a World-Sized Game Board
Creating a game that uses the entire planet as its playing field came with unique technical challenges. I needed to:
- Track concurrent user location movements
- Calculate complex polygon intersections in real-time
- Provide immediate feedback across potentially thousands of concurrent users
- Offline support for intermittent internet connection problems
The location tracking alone was a significant hurdle. GPS can be finicky, especially when a user loses signal for a period of time, and we needed accuracy down to a few meters to make the gameplay fair. I explored a variety of different filtering and line simplification algorithms to try and keep DB storage down, without sacrificing too much resolution.
Every time a player completes a loop, we need to calculate how it intersects with existing territories – potentially thousands of them. This had to happen nearly instantaneously to maintain the flow of the game. I eventually built in a spatial indexing system that could efficiently determine which territories might be affected by a new loop without checking every single one. This became necessary fairly quickly as our reddit community grew, and we started getting some superfans that try to take over all of Manhattan.
Real-World Testing (AKA Getting Some Exercise)
There’s something uniquely satisfying about building an app that forces you to physically test it. Unlike most development where you sit at a desk for hours, debugging Takeover sometimes meant going for bike rides, and walks around the neighborhood. My step count hit all-time highs during this project. We very quickly realized that the app sort of forces you to explore parts of your city that you don’t normally travel through because you want to claim as much Turf as possible.
Beta testing was particularly interesting. We’d watch in real-time as players claimed Turf in different cities and countries. In the beginning I had built a debugging tool that would replay various player Takes, so we could get a sense of what was happening in various scenarios. We enjoyed watching them so much that we introduced the “replays” feature that allows you to watch other users’ Takes.
With any game, there are always bound to be some bad actors. We had a few users that would try to cheat the system by spoofing locations, or just driving around in a vehicle (on the highway no less) instead of walking or biking. Along with building various debugging tools for this game, I also had scripts to manually handle this behavior, until I eventually built in some heuristics to account for, and handle these things automatically.
From Code to Community
Building the app was only half the battle. Growing a community of players across the globe presented its own challenges. Since the guys and I are on opposite coasts, we focused on organic growth within the NYC and LA areas, via word of mouth and Reddit. We had friends provide feedback as well.
Anytime I build an app, it’s always a learning experience. Users will always use your app in ways that you don’t anticipate, and it’s fun to have evolve with the community. Being able to export a GPX file of your Take, so that you can add it to Strava or other GPS apps, was a feature we added in response to user feedback. It of course doesn’t work in the reverse direction since Takeover is about real-time movement at the moment, but maybe there can be a different mode to support this in the future.
Future Conquests
One of the bigger asks from the community was to build this for Android. As an Android user myself, I had built most of it, but hosting an app without monetization plans has put a pause on releasing it for now.
In the meantime, if you’d like to join the global competition for territorial dominance, you can download Takeover from the iOS App Store or visit takeover.wtf for more information.