Robert N. Charette, an internationally recognized risk analyst and systems expert, has spent the last 20 years studying software failures. Over his 50-year career, it’s been a trend that he’s become deeply troubled by. IT professionals, government leaders, and corporate executives live in a delusional fantasy where software failures are rare and cost little. His extensive body of work includes contributions to IEEE Spectrum, where he has notably highlighted the significant risks posed by software failures in various sectors.
Charette’s insights are particularly relevant in today’s technology-driven landscape. He argues that just like electricity was in the 20th century, software is the key infrastructure of the 21st century. A world with crumbling roads, where we wouldn’t tolerate electricity going out every other day. We’re fine with it when it’s AWS or telcos or the banks that go down,” he said, underscoring our frightening double standard on software reliability.
Charette has written about many such software disasters in recent years — including the Canadian government’s Phoenix paycheck system. This specific example acts as a reminder of the deep financial and emotional pain caused by such breakdowns. As he explains, organizations often fail to do what’s necessary to make prevention of software failure a priority. Such neglect can do serious harm to their economic engines.
“The biggest tragedy is that [software failure] is for the most part predictable and avoidable. Unfortunately, most organizations don’t see preventing failure as an urgent matter, even though that view risks harming the organization and maybe even destroying it.” – Robert N. Charette
The consequence of software failures goes far beyond corporate settings. Charette’s alarm goes even deeper when it comes to medical devices that depend on the integrity of their software. He highlights a troubling statistic: the U.S. Food and Drug Administration recalls approximately 20 medical devices each month due to software issues. These recalls are a chilling reminder of how vital it is to have strong standards when creating software for medical devices.
When it comes to building software for medical devices, there are a lot more standards that you need to satisfy. There’s a heightened fear of the implications of failure,” Charette countered. His expertise sheds light on the pressing need for enhanced regulatory measures and rigorous testing protocols to mitigate risks associated with software in healthcare.
After all, Charette’s seminal article, “Why Software Fails,” first appeared in 2005. It remains a foundation stone in his advocacy for addressing software reliability challenges. He’s incredibly wise about maintaining that ongoing dialogue with readers through platforms like IEEE Spectrum. This coordinated effort continues to educate people about just how prevalent and preventable software-related failures are.
Charette’s happy to report a growing reliance on technology across every sector. She says organizations are still failing to prioritize reliability of software. To do that, he says, would require a change in the whole collective mentality. We must shift from responding after the fact to preventing failures in the first place.