The Phoenix Architecture

When I say things like "delete the code, keep the spec," many people seem to interpret that as a rule: write the specification first, and only then are you allowed to program.

Everyone suddenly wants to tell you SaaS is dead.

For decades, architectural knowledge lived inside the implementation, because the implementation was expensive to reproduce.

Imagine joining a software project a few years from now.

Someone asked me recently where the boundary lies between a specification and an implementation.

Every mature system carries lessons no one remembers writing down.

Production used to be the place where software went to fail.

Thirty source files organized into layers.

Every layer of the software toolchain is changing at the same time.

Here’s a scene every working programmer has lived.

The industry is still trying to generate applications.

From Tiny to Deletion Safety In 2014, I gave a talk called Tiny.

The “AI software factory” metaphor is seductive.

Here’s a simple test you can apply to any software system you work on: Imagine deleting the entire implementation.

If you’ve been following this series, you may already be thinking: “This all makes sense for non-UI code.

When code can be thrown away and recreated, the unit of change is no longer lines of code.

Here is a design constraint worth taking seriously: if your system cannot be understood, modified, and regenerated from specification by one competent engineer, it is already too complex.

In late 1999, I found myself inside Extreme Programming movement before most people had heard of it.

We’ve spent decades talking as if “the system” and “the codebase” were the same thing.

As I mentioned in a previous post, at Wunderlist, we had a rule: any new service had to be "this big", a constraint I'd demonstrate by holding my fingers a few inches apart.

In 2013, I wrote about trashing servers and burning code.

If deleting your codebase feels terrifying, your evaluations are insufficient.

If you've been using generative AI regularly for a while, you already know this feeling.

The cheapest system in the AI era is not the one that never changes.

When we talk about software economics in the age of generative AI, it can feel like we’re inventing something entirely new.

Not all software should change at the same speed.

When Machines Learned to Code For most of computing history, programming was bottlenecked by human cognition.

Software is entering a strange new phase.