I have sometimes been curious to see a map of all the flights I have taken in my life, and learn about their total distance. At first as a way to quantify personal climate impact, but later also to just visualize my personal “globalization”.
References
In August 2024, I saw a visualization on the Mastodon profile header of John O’Nolan that looked close to what I was looking for myself (his current profile as presented by Mastodon, even shows a 3D globe screenshot). I think John also explained the meaning of the colors once somewhere, but I can’t find that back.
Another blog post by Jason Becker around the same time described more aggregated insights on his flight history gained through the apps Flighty, TripIt and meticulous tracking.
What I would try
I haven’t done any active tracking of my flight data in any app.
Perhaps various governments would have some of it, but I can imagine that would be difficult to acquire.
What I do have is an Gmail email account that likely received details about 90+% of the flights I have taken: tickets, receipts, check-in reminders. Some of those were also automatically added to Google Calendar.
A good starting point would be to try to query or scrape my historical emails for my flight details. Perhaps using a Google email search API as a first pass, or alternatively, starting by using a local archive (Google Takeout) of all my emails.
Since this data will largely not be structured, some form of AI would likely be helpful as a part of the pipeline to filter and extract relevant information.
Another question is: which information would be relevant? And what can be generally gleaned from these emails?
I would let a script scan for:
- departure and arrival time(s)
- departure, intermediate and arrival airport codes
- flight number(s)
- people who were assigned to the flight, in order to exclude tickets I purchased for others
Other data, such as the plane type and exact plane trajectory could perhaps be retrieved from (likely paid) third party APIs with historical flight data and then cross-referenced with my flight data. FlightRadar24 has one, at least. This more precise data could be interesting to calculate CO2_e emissions, or aggregate flights by plane types, but the price of far-away history seems to go wildly beyond my willingness to pay for a tiny side project.