Why Python’s ‘Worse is Better’ Philosophy Still Shapes Modern Development

Python’s Origin and Rise

Guido van Rossum started developing Python during the 1989 Christmas holidays, creating a prototype in three months. By May 2025 Python topped the Tiobe index with a score of 25.35% and led the PyPL index with a 30.41% search share, surpassing Java.

“Worse is Better” Philosophy

At the 2025 Python summit, Van Rossum asked whether the “Worse is better” principle still applies. He recalled that early Python development was driven by the need for a quick, simple language, borrowing ideas from the ABC language and Unix.

“We wanted Amoeba to be self‑hosting, but it required many user‑level tools that Unix could not provide, so we had to rewrite them in C, which was slow.”

He described how the lack of long integers, reliance on C stdio, and the absence of classes were intentional compromises that allowed Python to run within three months.

Two Philosophical Paths

One path, exemplified by MIT/Stanford languages such as Common Lisp and Scheme, pursues “the right thing” with formal elegance. The other, “Worse is better,” favours simple, portable implementations, as seen in early Unix and C.

Build a cross‑platform, low‑resource core.

Release the “good enough” version quickly.

Iteratively improve functionality and performance.

Is the Philosophy Still Valid?

Van Rossum argued that Python’s early compromises proved successful, but modern development often requires years of effort and large teams. He questioned whether today’s contributors must deliver perfect prototypes or can return to the “first build, then improve” model.

Python and Rust Collaboration

He highlighted the PyO3 project, which lets developers write Python modules in Rust, following the same “worse is better” approach: start with a runnable core and refine it over time.

Conclusion

Van Rossum suggested embracing community involvement and pragmatic development to keep Python evolving, even if it means revisiting the “Worse is better” mindset.