Inspiration
I wanted to know how close I'd been to meeting my friend before we actually met. We both lived pretty close to each other for our entire lives, so it was pretty likely we'd crossed paths many times.
What it does
Twined finds the occasions in you and your friend's past where you intersected. It can tell you how you were just 10 metres away in a Sainsbury's in 2015, or how you both visited the same cinema less than a day apart.
How we built it
We collate your Google Maps location history with your friend's, efficiently finding all intersections, then display these in a pretty, readable format using React.
Challenges we ran into
Finding a way to quickly deal with potentially around 2 million data points wasn't so easy. The raw data itself can easily top 2GB. In the end we came up with an algorithm and data structure that can handle this large task in as little as 5 seconds.
However, actually finding a way to display all this data presented an even larger challenge. We had to convey these intersections across both time and space without confusing the user.
Accomplishments that we're proud of
We were most proud of the idea. We regularly got distracted during the hack by some of the fascinating and unexpected insights our half done code was providing.
What we learned
UI/UX is hard, and is just as important as functionality. There's no point in having a cool program if nobody can figure out what it does.
What's next for Twined
We have many possible expansions planned, from a history you can scrub through, to more in-depth maps, to complex automatically generated infographics; there's no shortage of ideas that build on Twined.
Log in or sign up for Devpost to join the conversation.