Gotta work together if you’re gonna catch ‘em all

A cuddly Psyduck from the 1998 launch of Pokemon Red/Blue. Picture by Mrs Gemstone, CC-BY-SA.

The launch of Pokémon Go has seen the rebirth of a 90s craze in the smartphone era. It’s a rebirth that leads to a tale about different worlds, data infrastructure, sensors and working together to improve our real world.

Pokémon Go is an AR (augmented reality) game. It is not the first AR game, it is not even the first where people roam their town catching monsters, but it is certainly the most popular so far. On one day last month 25 million people played it in the USA. The craze might fade but we will see a wave of AR games and products over the next few years.

Pokémon Go takes place across three worlds

We can think of Pokémon Go as taking place across three worlds: the real world that the player is walking around; the Pokémon world where the characters, gyms and pokestops are; and a world of data, or data infrastructure, that connects together the real and Pokémon worlds. Many other AR games and services will show a similar pattern.

A sample Pokemon map, with a pokestop in the pub just across the road from my flat.

Players wander the real world whilst their smartphone displays a map from the world of data with local parks and streets. The map also shows the location of characters in the Pokémon world that people can catch. When people try to catch a Pokémon the phone will overlay an image of the character on a live image of the real world captured through their phone camera. The map also shows places where people can collect items, called pokéstops, and gyms where people can battle each other using Pokémon characters.

The Pokemon world’s gyms and pokéstops are overlaid on real places: pubs, churches, restaurants that exist in the real world and whose name, type and location is also stored in the data infrastructure.

Pokémon Go shows us some of the gaps in our data infrastructure

Digital maps and lists of places pinpoint where we live, work and visit; connect us to local communities and services we rely on; and help us find our way around the world. They are important parts of the data infrastructure that underpins so many parts of a modern society and economy. I think about that data infrastructure a lot in my current job.

Unfortunately there are problems with our data infrastructure. If my address wasn’t in that data infrastructure then I might struggle to register to vote, get insurance or even order a pizza. Other parts of our data infrastructure are inaccessible because its locked away or priced too high for people to use. Other parts are simply missing or broken, no one’s got around to building or fixing them yet.

Despite this data infrastructure underpins Pokémon Go and will underpin other AR games and services in the future. We can see some of it’s weaknesses in this game.

A vicious looking Pidgey that got through security in a UK Parliament building. Characters in the Pokémon world don’t respect barriers in the real world because the barriers aren’t in the data infrastructure.

Shortly after the launch of the game someone who lived in a converted church reported that their house had been marked as a pokéstop and that players were gathering outside. The information about the place was out of date. People had to be asked to stop playing the game in Auschwitz museum. The information about this area of the real world was incomplete. Some areas and towns have few pokéstops or gyms because these areas are rural or contained few players of previous AR games so there is a lack of information. Pokémon Go is only playable in a few parts of South Korea due to national security concerns. Meanwhile any Pokémon player in an area with recent construction is likely to see bits of the map that are incorrect with missing roads or a park that has since been built over.

Collaboratively maintained data infrastructure will create better AR

Pokémon Go has ways for people to report problems with the data, but these are not easy to use and only pass data back to the people who control the Pokémon world rather than the people who maintain the underlying data infrastructure. As more AR services launch will the people who live in the converted church or people who live in poor and rural areas have to report the same problems to each of the new service providers? Will each service provider have to clean up the same data? This seems rather time-consuming and expensive when people can work together and solve the problem once.

Weak data infrastructure will be a a common problem for AR services, stronger, collaboratively maintained data infrastructure that anyone can use will provide a common solution.

A road found by a Pokémon Go player will be available to a car driver using Google Maps to find their way around a strange city. Players in one game can mark an area as out of bounds and players in other games will receive the same warning. Data about a new restaurant will be available to every service whether it be a game or an AR service that tries to entice you into that restaurant.

If the data is handled carefully, and privacy and openness are brought together to build trust, then perhaps some of the camera images could also be incorporated. Perhaps Pokémon Go could automatically spot and report a pothole caught on your camera whilst you were playing the game. People are more likely to trust uses of their data that respect their privacy and benefit society. This trust can lead to increased use of the AR service benefiting the service provider.

Collaboratively maintained and open data infrastructure will help build a better future

The benefits go beyond improving the AR services. Information about poorly mapped areas of cities can be gathered and used by AR services, public services or delivery firms. On election day a government could publish information about candidates and polling stations to the data infrastructure and every AR service would have access to the information and be able to incorporate it into their virtual worlds to encourage more people to vote.

The coming explosion in data. Image from the Open Data Institute, CC-BY-SA.

We are about to see an explosion in the number of sensors capturing data about our world. In Pokémon Go these sensors are smartphones and their cameras, but we are also seeing the rise of the Internet of Things; automated cars that will transport people whilst capturing data about the streets they travel; and delivery drones capturing images from the sky.

All of these sensors can capture data, respect privacy and publish useful information back to the underlying data infrastructure to help collaboratively maintain and improve it. By working together to maintain this data infrastructure and publishing data as openly as possible so that anyone can use it then it can help build a future which maximises the value we get from the data. To put it simply it can help make the real world a better place. One that can handle a growing population and our expectation for ever cheaper and better services. It will also improve AR games and might make Pokémon in their Pokémon world a bit easier to catch too.

It might seem strange to go from the 90s Pokémon craze, to an augmented reality game using three different worlds, to building a collaborative and open data infrastructure that can make the world a better place to live.

Perhaps it looks like a crazy dream that’s come from throwing too many balls at too many cartoon characters but it is an achievable future and one that comes from working together. Even if we start by throwing a small ball at some strange cartoon characters, we can still dream big.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.