It’s no secret that I'm a bit of a fan of Sidney Dekker’s work. “Drift into Failure: From Hunting Broken Components to Understanding Complex Systems” was written over a decade ago now, but feels as timely as ever. Unfortunately, it also still feels a little provocative of an exploration of why modern systems fail. This book takes a deep dive - it’s just 200 pages but cognitively dense - into the intricate workings of complex systems, and how, despite our best efforts, they often seem to drift towards failure. Dekker challenges the reader to rethink the very nature of failure and provides a fresh perspective that is both enlightening and essential for anyone dealing with modern systems. Those systems are growing increasingly complex, and our understanding of them has not evolved at the same pace. We are still using outdated, Newtonian cause-and-effect models to analyze failures, doing a disservice to everyone involved. This kind of thinking not only oversimplifies the problem but also impedes our progress in understanding how and why failures occur, as well as dealing with incidents & accidents in a fair manner. In complex systems, the idea of a “root cause” is almost always a mirage; we can only sketch out the contributing factors. Dekker reminds us of a crucial distinction between complicated and complex systems — terms that are often mistakenly used interchangeably. Complicated systems, no matter how intricate, are ultimately predictable. Complex systems, however, are governed by interactions and feedback loops that can lead to unpredictable outcomes. Understanding this distinction is vital, especially as we continue to introduce automation into our systems, thinking it will act as a layer of redundancy. In reality, automation often adds more points of potential failure and, worse, relegates humans to roles they are inherently poor at — the active monitoring of reliable machinery. Introduction of software into seemingly everything is one of the mail culprits in growing complexity. On one hand, it has allowed us to create systems of unprecedented capability; on the other, it has introduced layers of complexity that are often far beyond human comprehension. The result is a system that can fail in ways that are impossible to predict. Drift into Failure challenges conventional wisdom and pushes us to think differently about failure, not as a broken component but as an emergent property of the system itself. Dekker’s work has often been critiqued of lacking immediate real-world applicability; some readers will think that is the case here, too, but I still found the concepts extremely valuable and important. It also does sketch some pathways forward, such as introduction of diversity of all kinds as an antidote to drift into failure. * Rating: 4.5 out of 5 * Dog-ear index: 14 * Who is it for: At least engineers, safety professionals, managers, and people involved in systems thinking or complexity theory.
Hi Sami Mäkeläinen this absolutely feels like a very pertinent read for me and its a topic I am very curious about. But I am wary that it might be a hard read. Do you have an alternative suggestion for someone who hasnt been introduced to Dekker?
Thanks Sami Mäkeläinen for the recommendation 👍🏽
Technologist | Humanist | Pragmatist | ✈️
2wreminder: I highlight important parts of the books I read, and dog-ear the _really_ important pages. The dog-ear index is simply the average number of dog-eared pages per 100 pages] Product link for reference only; please support your local bookstore where possible: https://www.amazon.com.au/Drift-into-Failure-Components-Understanding/dp/1409422216/