Hi HN, we’ve been building this for the last couple of months, partly to scratch our own itch as avid travellers. We found it increasingly difficult to keep track of changing travel restrictions and policies from dozens of different sources, especially now as many countries have started to implement differing policies for vaccinated travellers. We hope you find it useful. :)
The data is sourced from official sources (mostly government and airline websites) by our data partners, and then we organize and present this information in a way that’s easily searchable and scannable.
For the COVID19 data, We implement the same scoring system as the CDC to grade the risk of each country. We also show leading indicators that could cause restrictions to change, such as new cases, reproduction rate, and vaccination rate.
Soon we’ll be adding the ability to filter on regions that explicitly allow vaccinated travellers to visit. Currently this information is available within the description section for policies and restrictions.
The backend is built in Go using Ent (https://news.ycombinator.com/item?id=26008521) and React as the boring tech for the frontend, which has been a fun combination to try from scratch. We aggregate data from a number of providers who scan official sources daily and then apply it to the graph, making it possible to traverse different regions of the world and find routes.
We’ve built a notification system that can alert you (email only at this stage) of changes as we detect them, which is a few dozen times a day for 247 countries and 138 regions.
What has working with the mapbox map been like? I tried to create a world map website before but I did a custom map. In the end I spent all my time getting overlays on the map to work before I got bored and moved on to something else.
Mapbox certainly has been challenging. I read over a lot of code I found through GitHub search to figure out the overlays, and then went down a big rabbit hole trying to improve the rendering performance without much to show for it — I still think the way we render the overlays is a performance issue, but it could also be that Mapbox loads an entire webgl rendering engine with a lot of features we don't need.
Longer term we'll probably move to OpenLayers + OpenStreetMap because Mapbox data is out of our budget by about 2 orders of magnitude.
5 comments
[ 6.3 ms ] story [ 29.5 ms ] threadThe data is sourced from official sources (mostly government and airline websites) by our data partners, and then we organize and present this information in a way that’s easily searchable and scannable.
For the COVID19 data, We implement the same scoring system as the CDC to grade the risk of each country. We also show leading indicators that could cause restrictions to change, such as new cases, reproduction rate, and vaccination rate.
Soon we’ll be adding the ability to filter on regions that explicitly allow vaccinated travellers to visit. Currently this information is available within the description section for policies and restrictions.
The backend is built in Go using Ent (https://news.ycombinator.com/item?id=26008521) and React as the boring tech for the frontend, which has been a fun combination to try from scratch. We aggregate data from a number of providers who scan official sources daily and then apply it to the graph, making it possible to traverse different regions of the world and find routes. We’ve built a notification system that can alert you (email only at this stage) of changes as we detect them, which is a few dozen times a day for 247 countries and 138 regions.
Longer term we'll probably move to OpenLayers + OpenStreetMap because Mapbox data is out of our budget by about 2 orders of magnitude.