Just after midnight Eastern Time on July 19, 2024, the enterprise cybersecurity company CrowdStrike YOLOed a software update to millions of Windows machines. Or as they put it: On July 19, 2024 at 04:09 UTC, as part of ongoing operations, CrowdStrike released a sensor configuration update to Windows systems. That sensor configuration update caused the largest IT outage in history. Overnight, about 8.5 million computers blue screened, affecting hospitals, banks, 911 systems–as the New York Times put it, “It is more apt to ask what was not affected.” The answer is Linux, Macs, and phones. The outage highlighted a different kind of digital divide. On one side, gmail, Facebook, and Twitter kept running, letting us post photos of blue screens located on the other side: the Windows machines responsible for actually doing things in the world like making appointments, opening accounts, and dispatching police. They also run airlines. Here’s a visualization of the chaos that CrowdStrike caused for airlines from the New York Times: Airline cancellations is a good metric, but I want to look directly at air traffic: How many planes were in the air? How many planes should have been in the air? At about noon UTC, 8 hours after the CrowdStrike update hit, someone posted a video to Twitter that they made with FlightRadar24 showing air traffic over the United States. It was described as a 12-hour timelapse of American Airlines, Delta, and United plane traffic that showed the nationwide ground stop of the three airlines due to CrowdStrike. Here’s the video: It’s not a good visualization of the impact because there’s no basis for comparison. It clearly shows fewer planes flying at night, but that happens every day. Was that night different from any other night? There’s no way to tell. In Bellingcat’s “OSHIT: Seven Deadly Sins of Bad Open Source Research”, sin #4 is “Lacking Context for Occurrences, Common or Otherwise”. In this post I’ll show the effects CrowdStrike had on air...