I'm guessing that's an LLM hallucination. The conclusion section especially has some hints it was pulled out of an LLM:

> The package managers we benchmarked weren't built wrong, they were solutions designed for the constraints of their time.

> Buns approach wasn't revolutionary, it was just willing to look at what actually slows things down today.

> Installing packages 25x faster isn't "magic": it's what happens when tools are built for the hardware we actually have.