The original plan was to integrate both the Skyscanner API with the Esri API to come up with a hack that would allow people to identify peak travel periods to different popular destinations around the world. This would be useful to those who are avoiding the crowd, and who are looking for the cheaper destinations.
What it does
It generates a heatmap on the WebGL API, which identifies the destinations with lots of flights, using a self-written Gaussian function. It draws data from the Skyscanner API to do so.
How we built it
The original idea was to use the Esri API; however, there were technical issues with parsing the data into a geolocation database object that can be rendered by the layer, so we migrated to the WebGL API, feeding it with the JSON data drawn from the Skyscanner API.
Challenges we ran into
The Esri API did not offer some of the functionalities that we needed, i.e. we could not process the data to be passed into the Layer, without having access to their server (which is a paid service). There were issues with making calls to the Skyscanner API, which was only resolved in the night. We had to manually write the gaussian blur for the heat map, before mapping it to the globe. After mapping it to the globe, we had to re-plot the points on the globe, without the data stacking.
Accomplishments that we're proud of
Despite the setbacks, we finally managed to complete the project, even though we had to start from scratch during the night.
What we learned
How difficult it is to hack into APIs without proper documentation and few features, as well as the difficulty in deploying to AWS.
What's next for jetblazinit
Mapping the data to history would allow people to plan their holidays in advance, so that they can make the best choices. We would also like to be able to write an algorithm that would identify the next best (nearest and cheapest) destination that people can look for as an alternative.