There was this strange sound. Clack, clack, clack! Was it coming from the tires? Clack, clack, clack! Just a few minutes after we hit the Autobahn to drive back home all the way across Germany. Clack, clack, clack! Maybe I’ll better have a look. I pulled over and stopped at the filling station. I was lucky: There was a mechanic from the ADAC, the German automobile association, just filling up his car. I walked over and asked him if he could have a look at the wheels I had just changed a few hours before. “Well, these are the wrong bolts.” He looked worried. “This could have killed you.” It turned out that I had used the longer bolts from the steel rims of the winter tires to tighten up the summer tires. But because they didn’t fit, they had started to come loose. A few more kilometers and a white Golf I with two students on the way back to their parents would have lost a tire at 80 miles an hour. And all of this because of one small human error.
When we design systems, we want them to be fail-safe. We want them to be as resilient as possible. To make a system resilient, we often focus on the design of technical aspects like durability, reliability, and security. We try to build it from individual parts that are as simple as possible to reduce complexity and, thus, the possibility of technical failure. We try to add in redundancies so that when one component of the system fails, another one can take over so that the whole system keeps running. This is also the reason why we often use a decentralized approach to structure our system. All of this because we know that a resilient system has to be able to respond to change and therefore has to be as adaptable and agile as possible.
But what we often overlook are ourselves. Humans. In our love for technological solutions, we forget that humans will have to work with a system. Humans who architect, use, maintain, change, and enhance the system. Humans with different skillsets and levels of knowledge. Humans who make mistakes.
Humans are significant components of every system. Sometimes, they are the predominant elements in the system. Sometimes, for example in a team, humans even constitute the entire system itself. Be it a spacecraft, a hospital, or a pattern library for a website – if the system is truly resilient, it doesn’t allow for an individual human to become the single point of failure.
This is the 18th post of my 100 days of writing series. You can find a list of all posts here.