Complete, Correct, and Confidential: the inherently unhackable database.
Conference (ADVANCED level)
We think of the software that runs our systems as its beating heart, but the brains of an organisation — the continuity of its identity — resides in its data. The databases must stay complete, correct, and confidential: if any one of the three fails the organisation faces an existential threat.
In a logical world, the most durable and secure part of our systems would be its data, but instead — as a matter of course — we build systems that allow secrets to leak out, allow contamination to leak in, and allow valuable data to be simply forgotten. To make up for those problems, we try to strengthen our perimeter, but that doesn't work, because that's not where the problem lies.
In this talk, Jules explores what’s wrong with our current approach to data persistence, why secure boundaries and the Principle of Least Privilege are not nearly enough, and what we can do instead.
In a logical world, the most durable and secure part of our systems would be its data, but instead — as a matter of course — we build systems that allow secrets to leak out, allow contamination to leak in, and allow valuable data to be simply forgotten. To make up for those problems, we try to strengthen our perimeter, but that doesn't work, because that's not where the problem lies.
In this talk, Jules explores what’s wrong with our current approach to data persistence, why secure boundaries and the Principle of Least Privilege are not nearly enough, and what we can do instead.
Jules May
22 Consulting Ltd
Jules May is a software architect, consultant, and project leader known for designing systems that simply don’t fail. His work spans from safety-critical aviation and automotive software to enterprise cloud platforms, from DSLs and compiler design to algorithmic toolchains.
Starting his career in flight-control systems, Jules learned early that reliability isn’t a feature — it’s a necessity. That ethos has shaped a career focused on clarity, precision, and delivering technology that genuinely works.
He’s the author of Extreme Reliability: Programming Like Your Life Depends on It and the originator of Problem Space Analysis, a practical framework for understanding and solving complex technical problems.
Over the years, Jules has led engineering teams, advised global clients, and spoken at conferences across Europe on reliability, development culture, non-traditional computing and post-quantum security. He’s worked with organisations from start-ups to household names, building solutions that range from embedded systems to high-availability SaaS platforms.
A mathematician by training, a programmer by instinct, and a teacher by inclination, Jules brings a rare combination of theoretical depth and hands-on experience. Whether mentoring developers or shaping strategy as a fractional CTO, he focuses on one thing above all: making software that does what it’s meant to — every time.
Starting his career in flight-control systems, Jules learned early that reliability isn’t a feature — it’s a necessity. That ethos has shaped a career focused on clarity, precision, and delivering technology that genuinely works.
He’s the author of Extreme Reliability: Programming Like Your Life Depends on It and the originator of Problem Space Analysis, a practical framework for understanding and solving complex technical problems.
Over the years, Jules has led engineering teams, advised global clients, and spoken at conferences across Europe on reliability, development culture, non-traditional computing and post-quantum security. He’s worked with organisations from start-ups to household names, building solutions that range from embedded systems to high-availability SaaS platforms.
A mathematician by training, a programmer by instinct, and a teacher by inclination, Jules brings a rare combination of theoretical depth and hands-on experience. Whether mentoring developers or shaping strategy as a fractional CTO, he focuses on one thing above all: making software that does what it’s meant to — every time.