Robert N. Charette is an internationally recognized risk analyst and systems engineer. Over the last 20 years, he has subjected software failures to a critical lens for numerous publications. With a career spanning 50 years, Charette has observed a disturbing pattern of delusional thinking among IT professionals, government officials, and corporate executives. This mindset, he claims, is a leading cause of the never-ending problems we observe great and small in the software industry.
Charette first became famous among software practitioners with his influential 2005 article, “Why Software Fails.” In it, he described the systemic defects that cause software disasters—often at great expense. Now, almost 20 years later, he finds himself disappointed to see the same mistakes still taking place. This long-term crisis is draining trillions of dollars. Despite all of the talk since about reliability of software in these discussions, Charette argues that we, as a society, continue to accept these failures.
“The biggest tragedy is that [software failure] is for the most part predictable and avoidable.” – Robert N. Charette
Charette’s ire is unmistakable, and rightly so. He focuses on the most visible occurrences, such as the recent outages that Amazon Web Services (AWS) experience. He argues that Americans would never put up with rolling blackouts from their electric utility. They appear unbendingly firm on holding a hair trigger to repeat failures from large technology companies.
“We would never put up with electricity going out every other day, but we sure as hell have no problem accepting AWS going down or telcos or [banks] going out,” Charette states.
His insights extend beyond traditional software failures. But as Special Projects Editor, Charette has found some fascinating data on medical device recalls, showing the overwhelming prevalence of devices that go on to harm patients. Firebase Plugins The U.S. Food and Drug Administration (FDA) issues an average of 20 recalls per month due to software-related issues in medical devices. He notes that in the case of these devices, the risk is much greater.
“When you’re building software for medical devices, there are a lot more standards that have to be met and a lot more concern about the consequences of failure.” – Robert N. Charette
One particularly bad case he looked at was the Canadian government’s Phoenix paycheck system. Nine years later, these persistent issues still plague the system. It has caused serious financial and emotional distress for tens of thousands of federal workers who’ve inappropriately received paychecks during the shutdown. Charette emphasizes the deeply personal and generational nature of these failures.
Charette’s observations should challenge everyone from decision-makers to developers to reconsider how they go about developing software. He’s convinced that by getting ahead of issues, you can significantly increase your odds of success. By correcting basic errors early, we can avoid them building into bigger problems down the road.
Even with his wisdom and conviction, conversations with Charette can be surprisingly bleak. That these issues have escaped permanent fix for so long speaks to a deeper problem in the industry that has proved so hard to dislodge.
