Frankly speaking ...

A Reader Pushes Back Why Lean? argued that formal methods are the industry's answer to a world where testing can no longer keep up with the cost of failure. A reader, commenting on that post , pointed out something the post had left out entirely: sometimes bugs survive not because testing failed to catch them, but because nobody was ever going to fix them. That comment, and the conversation it started, turned into more than a reply. It was the second half of the argument. The Bugs We Choose Not to Fix The reader's point was blunt and specific: a company they had just left refused to fix known bugs because the market and investors wanted to see year-on-year growth, and growth comes from shipping new features, not from fixing what already exists. Bugs get "fixed" only when a client trips over one in production. In effect, the user base becomes an informal, unpaid QA layer — crowdsourced testing whether they like it or not. This is a different failure mode from ...

The Real Issue Is Trust Modern society runs on software that is increasingly essential and increasingly fragile. The problem is not simply that code has bugs. The deeper issue is that we have built a digital civilisation on systems whose correctness is assumed rather than demonstrated. The examples are familiar, and they are getting more costly. In 2025, a routine firewall upgrade caused the Optus emergency calling outage : ordinary calls were automatically rerouted to another network without incident, but Triple Zero calls were silently left undelivered for hours. The fault was narrow, yet it landed precisely on the one function that mattered most, raising serious questions about how deeply society depends on systems we assume are safe and reliable. In 2026, the Telstra outage brought widespread disruption across a continent, showing how a single infrastructure failure can cascade through communications, commerce, and daily life. In a world where a single logic error can disabl...

1. The Problem: AI Amnesia We’ve all been there. You’re working with an AI agent, and for a moment, it feels like magic. Then, five minutes later, it forgets the architectural decision you just made and defaults to some generic, shallow solution. It’s disconcerting. The truth is, AI-assisted development often lacks discipline. We’ve moved away from rigorous systems and into "vibe coding"—just hoping the right prompt will magically give us a maintainable codebase. It rarely does. Matt Pocock nailed the core frustration: You have access to a fleet of middling to good engineers that you can deploy at any time. But these engineers have a critical flaw: they have no memory. They don't remember things they've done before. I’ve learned that the only way to keep these agents on track is to stop treating them like magic chatboxes and start treating them like disciplined (if forgetful) engineers. The fix? Agent Skills. These are modular, encoded processes that force...