The Box Island Journey
In 2015 I co-founded a video game company Radiant Games. The founding idea of the company is to empower children, by providing the best playful coding experiences possible. We want to introduce kids to the fundamentals of coding because we think it shapes the way they think and interact with the world.
About the game
Box Island is a beautiful mobile coding game that takes kids on an exciting adventure on a charming island. Along the way the kids have to apply the basics of algorithms, pattern recognition, sequences, loops and conditionals to progress through the super fun and brainy game play. The game is truly unique for its beautiful 3D game world and fun story parts that keep the kids engaged and motivated while playing.
Start small
When we first started working on the game we were only the four founders. Three computer scientists and one 2d artist, I took the role as the technical lead which meant that I did all the heavy code lifting. Only three weeks after I joined, we presented a very early prototype at the largest technical conference in Iceland call UTMESSAN.
The early prototype essentially had all of the fundamental ideas already implemented, but not in as pleasing package as the final product. As you can see on the image, the visual scripting language was already in place. You can drag and drop tokens from the toolbox (left hand side) onto the active code block (connected to the OnStart). Where the goal was to get to the yellow tile. (Do you notice that the solution is wrong? 🙊)
Iterate fast
Right from the start we adopted the agile methodology and worked in 2 week sprints, with daily stand ups and retrospectives after every sprint. We also found it to help us focus to have everything printed out and present in the physical world, so our scrum wall was literally a wall. We dedicated every Monday morning to planning if needed. Here below is a GIF depicting a weeks worth of progress on the scrum wall.
Since we used sticky notes for our task we decided to have the color represented different types of tasks. Yellow was for coding, green for art, pink for bug and orange for infrastructure and refactoring. We loved having the literal scrum wall in the office, but as our operation grew with contractors from all over the world it was clear it was not scalable so we ported everything to Asana!
We also made every sprint personal by naming them. The names were drawn from either video games or music, starting from A to Z. Examples are Battlefield, Commandos and Muse for sprints B, C and M repetitively. When development crept closer to a deadline we started having more and more user tests to ensure the quality of our product.
Test, test and test!
Testing was a major factor in our development process. We took every opportunity to conduct a user test, from going to events and schools to conducting formal user test at our office. After every sprint we had a formal user test where we had 4–5 kids come in and play the game while we monitor their reaction and see where they got stuck. This type of testing was especially rewarding since some features which we thought were totally easy to understand were in fact not so easy.🙃
My favorite user test was when we went to a elementary school and the kids did not know that we were the creators of the game and actually expressed their true feelings. It was epic to hear their frustration and the eventual “ah ha!” moment when they figured the puzzle out.
The picture was taken when we were showcasing the game for the Hour of Code initiative in Washington DC.
The Star Problem
Well as I said sometimes when you design something that you think is easy to grasp… think again. Mind you this game is intended to be for 7+ years old and it’s main focus is between 7–12 years old. When you think of a reward system in a mobile game the first thing that comes to mind is stars, like in candy crush or angry birds or many other games. It’s simple, 3 stars means perfect anything less means you are losing out on something.
Fitting the stars to Box Island was trickier than one might think. Our first iteration was to convey the quality of the code, e.g. number of tokens or the path to the goal were not optimal. However, it was almost always the case that if one was not optimal the other wasn’t either. So in our second iteration we decided to only use the number of tokens as the measure. Since the tokens are always in front of the user and can easily be counted.
We thought we were done! It’s totally easy to understand, the more tokens you use the fewer stars you get!… well… nope… displaying it in a easy to digest format was way trickier than we thought. Since there is a reverse relationship between the number of tokens and the number of stars it proved to be hard to convey simply. We tried several iterations of UI designs with varying degrees of polish but to no avail, the kids did not understand know how to get 3 stars when they failed to do so.
Alright we took a step back and asked our self’s what the stars used for? The answer was enforcing good coding practices, basically do not write longer than necessary code. Through the previous iterations we had introduced this same enforcement through limiting the number of available tokens. This is why we decided to totally up end our design two weeks before a deadline and place the stars on the board itself!
At this time the objective of each level was to collect a balloon patch which hovered over a certain tile. The stars replaced the balloon patch and introduced a new mechanic, multiple end points. This decision was one of the best decisions that we made and the game was much more engaging as a result. Not to mention that this change cleaned up a lot of messy UI.
Finally it was obvious how to achieve 3 stars in a level 🎉! Simply collect them!
The glory days
The star problem is just one of many challenging and fun design problems we encountered while making the game. At our peak of operation we were around ten people developing the game from many different disciplines from all over the world.
Since we launched back in 2016 the game has been downloaded more than 800.000 times making one of the most successful Icelandic games ever. Along with more accolades, the game was awarded the Best Fun for Everyone category at Nordic Game 2016.
We received coverage in some of the world’s largest tech-sites, including TechCrunch, TheNextWeb and Mashable and were featured on both the App Store and the Play Store in dozens of countries.
Go Try It!
If you haven’t tried it already, you should! It is available both on iOS (App Store) and Android (Play Store).
Thanks for reading!
