A glitch on your personal computer is irritating. But a glitch on a network that controls critical systems at a major organization can be massively disruptive.
Several companies demonstrated the problem on Wednesday. An "automation issue" at United Airlines led to a mandatory delay for its planes that lasted for almost two hours, affecting 4,900 flights around the world. The Wall Street Journal's homepage went down briefly. And the New York Stock Exchange suspended trading for nearly four hours due to an "internal technical issue."
So far, it doesn't appear that any of these incidents were cyber attacks or anything other than technical hiccups -- and those are basically a fact of life, albeit a potentially damaging one.
The effects of these glitches point to one of the major risks of how modern networks operate: Often systems are centralized, and automated. That can make it easier for various bits and pieces to work together, helping deliver us the world of digital convenience that consumers and businesses rely on.
But that inter-connectivity also means that when something goes wrong with one part of a system, it might be hard to pin down the exact cause of the problem. Or it could even cascade and take down the whole system if there aren't fail-safes in place.
"We receive tremendous benefits from moving older systems to modern technology -- it lowers the cost of services and capabilities, it adds efficiency, and opens up markets to individuals who might otherwise not be able to use them," said Steve Grobman, the chief technology officer at Intel Security. "But with those benefits, we have to tolerate new issues we didn't necessarily have in a less automated environment."
Sometimes the issue will be an inherent flaw with software or some other hard to pin down technical problem. Other times, it will be a small human error that somehow ripples through a system -- like the electrical grid operator who forgot to turn a power monitoring program on after upgrading its software in 2003, contributing to a massive blackout in the Northeast.
"We need to be very diligent to ensure we have redundancies, fail-safes, and good assessments of the cyber-risk profiles these systems have to mitigate them as much as possible," said Grobman. It's unclear what the specific protocols were in place at the companies that had problems Wednesday. But even the best systems with the best oversight still might encounter issues. "What we see often is that assumptions about redundancies will have a corner case where the redundancy won't kick in," said Grobman.
The details remain scant about what appears to be a coincidental rush of technical difficulties. But combined, they're a reminder of the fragility of the technical infrastructures that underpin so much of our lives.