[gleenn]

Sounds like if you have a record of a lot of location/timestamp data for people, you look at the distance difference divided by the time difference. Now you have average speed for any pair of points. Now filter where the average speed is as fast as a Boeing jet. That filters out most of the data except for people who are almost certainly on a plane. Et voila, you now look at those data points geolocation and you have people who traveled from one city to another because you already have the location. Compare City1 -> City2 with any public flights in those cities around those times and you know who flew on what flight from where to where and at what time.

[timeon]

I'm more interested in this part:

> you have a record of a lot of location/timestamp data for people

What is the source of that data?

[jacobr1]

from the parent post: `social media and/or ad data`

So if you have ad impression data you have IP geolocation, or maybe better, along with the timestamp. Similarly for socials sometimes you get location metadata, and with image uploads you can can get location metadata (though today these are often stripped, historically they weren't).

[Macha]

Do people post on social media at high enough resolution to do this?

Especially since the claim was in 2012, is airplane wifi and roaming data reliable enough for people to view enough ads to do this?

Also where are you getting all a users ad impressions across different providers to have this kind of timing information?

There's a lot of creepy data available linked to ads and lots of companies doing enrichment with different fields if you have already have a real identifier for a user (e.g. phone number, email, cc, isp account number), but this sounds way more like "we created a load of hypothetical simulated data and pretended we had access to all of it"