Over the past four days I participated in the Khan Academy 2014 Healthy Hackathon, where my team worked on a gamified version of Khan Academy called KhanQuest.
Aside: The Healthy Hackathon is like any other hackathon in that you get the chance to work on something you wouldn’t normally get to. The big innovation in a healthy hackathon is that you don’t stay up all night or drink even one Red Bull. “Healthy” meals are provided and hackers are kicked out between midnight and 9:30 the next morning.
Khan Academy pays pretty close attention to engagement - there’s a data science team and a (small) growth team. There have been several dashboards charting our usage numbers over the past few years.
Unfortunately no amount of tweaks will make a student genuinely interested in using a website that their teacher is forcing them to use.
Our plan was to take Khan Academy’s content and embed it in a game that’s worth playing. The game is a fantasy rpg with gameplay similar to Final Fantasy (so I’m told). There’s strategy, exploration, levelling up, interesting decisions, and math.
A lot of thought has been poured into the content and structure of Khan Academy and its pedagogy. We wanted to create a tight mapping between the original website and the game so we could reuse as much of that work as possible.
Skills on KA (e.g. finding the area of a polygon) and spells in KhanQuest are equivalent.
Practice tasks and random encounters are equivalent.
Mastery tasks and boss battles are equivalent.
You’re incentivized to review skills because spells grow in power the longer you go without using them - so you’ll want to go back and occasionally use the spells you learned long ago.
… is really difficult!
I expected building the game engine to be difficult, and it was, but the surprising thing for me was the incredible amount of non-coding work there was. Finding and editing art, creating maps, playing through over and over, writing narration, coming up with spells.
We wanted to build a game that would be appealing to both boys and girls. I think it’s fair to say that we failed by making yet another fantasy game. There’s some serious work left to come up with a more universally appealing premise.
We were somewhat limited by the art we could find freely available online. We ended up reusing art from Valyria Tear and Battle for Wesnoth. Both projects are awesome and had the best art we could find. Unfortunately as a result we were more or less forced to make a fantasy game.
We took at least one step to make it not so bad. The default character is a female
mage grocery girl.
The version of the game we demoed is online.
This version of the game is incredibly easy to break, in fact our demo was carefully rehearsed to avoid all of the pitfalls (as a hackathon demo should be).
Some things to look out for:
We only had four days, don’t judge us too harshly!
Even if / when we improve the game I’m going to leave this version the same. I’m proud of what we did on only four days - warts and all.
Building KhanQuest was the most fun I’ve had at work since I can remember. It turns out building a game with a small team is a blast. So is doing work that you care about and you think might make a difference!