English specs can now be "compiled" to runnable code by LLMs.

When we make programs, we compile the source code down to an executable binary.

But then we make sure to keep the source code, so when we tweak it in the future we can compile a new binary.

Or, when we get better at compiling, or want to target a different environment, we can recompile.

The source code is far more precious than the binary.

But with "LLM compiled" specs to code, we currently keep the "binary" (the source code) and throw out the spec.

That seems backward!

This implies that LLM-native software development will place much more emphasis on the spec / PRD / design doc than the code.