Running with Scissors: How Tech Evolution Means the Risks Just Change Shape
Back when Docker was first taking the tech world by storm, a CTO summed up the state of software engineering with a phrase that stuck:
“We are running with scissors, and we look around and everyone else is running just as fast with scissors.”
It perfectly captured the chaotic, thrilling, and downright terrifying reality of early cloud-native infrastructure. You were taking raw, unproven technology, slapping it into production, and praying it didn’t blow up — because if you didn’t, your competitors would outpace you before the quarter ended.
Looking at where we are today, it raises an obvious question: Have we finally matured, or does that quote still stand to reason?
The short answer? We’ve definitely grown up. But the scissors have just changed shape.
Where We’ve Matured: Putting Safety Guards on the Infrastructure
In the early Docker era, running containers in production felt like tap dancing in a minefield. You were dealing with raw kernel panics, flaky networking drivers, and a total lack of persistent storage standards. It was pure wild west.
Today, that wild west has been paved over:
- Standardization and Reliability: Containers are no longer experimental; they’re boring infrastructure. Technologies like Kubernetes, managed cloud engines (AWS ECS, Google Cloud Run), and OCI standards have turned containerization into a utility as dependable as electricity.
- Managed Guardrails: Engineering teams rarely build basic plumbing from scratch anymore. Security scanning, automated CI/CD pipelines, and zero-downtime deployments are built directly into modern platforms.
In many ways, the infrastructure we used to sprint with has been replaced with safety scissors.
Why the Quote Still Applies: The New Edge of Innovation
Even though yesterday’s risky tech is today’s standard, the underlying pressure to move fast hasn’t vanished — it has just shifted up the stack.
The AI Rush is the New Docker
Replace “containers in 2013” with “Generative AI and autonomous agents today.” Organizations are rushing non-deterministic models and experimental tools directly into enterprise workflows with minimal security frameworks or long-term safety testing. Why? Because standing still feels like commercial suicide.
Modern Supply Chain Chaos
While our core infrastructure is safer, our software supply chains are vastly more complex. Modern applications rely on thousands of open-source dependencies and third-party APIs. Sprinting at today’s required speed means constantly accepting hidden vulnerabilities just to ship on time.
Accelerated Market Expectations
The baseline speed of business has fundamentally changed. Ship cycles have compressed from quarterly or weekly releases to continuous deployment. Moving cautiously often means getting out-executed by a competitor who is willing to take the risk and fix the bugs live in production.
The Verdict
So, are things different now? Yes and no.
The specific tools that used to keep CTOs up at night during the Docker boom are now safe, structured, and predictable. But technology evolution is a treadmill that only speeds up. The moment one layer of the stack becomes secure, we invent a brand-new, unproven layer on top of it — pick up the sharpest tools available, and start sprinting all over again.