The Quiet 'Retirement'
In October 2019, after more than six years at Dropbox, Guido van Rossum announced he was retiring. This followed his 2018 decision to step down as Python's "Benevolent Dictator for Life" (BDFL), the title he'd held for nearly 30 years. The BDFL resignation
came after a contentious debate over a new language feature, and van Rossum seemed ready for a break from the pressures of leadership. His departure from Dropbox, where he was a Principal Engineer, felt like the closing of a major chapter. The creator of the world’s most popular programming language—a language that powers everything from Instagram to NASA—was hanging it up for good. The tech world gave him a respectful send-off, assuming his direct contributions to the Python ecosystem were largely complete.
An Unexpected Twist
The retirement was short-lived. On November 12, 2020, van Rossum tweeted that he was joining Microsoft's Developer Division, stating simply, "I decided retirement was boring.". The announcement caught many by surprise. Microsoft, once viewed as a fierce competitor to the open-source philosophy that Python embodied, was now hiring its creator. But under CEO Satya Nadella, Microsoft had transformed, embracing open source, acquiring GitHub, and becoming one of its biggest contributors. Van Rossum wasn't just joining any company; he was joining a corporate giant that had become deeply reliant on Python for its strategic ambitions in cloud computing, data science, and artificial intelligence. He was given the title of Distinguished Engineer and, crucially, the freedom to choose his own project.
The Hidden Bet Revealed
So, what was the project? It was the hidden bet behind his return: to solve Python's oldest and most stubborn problem—its speed. For all its famed readability and ease of use, Python has always been slower than compiled languages like C++ or Rust. Van Rossum saw an opportunity to tackle this challenge head-on, with the backing of a corporate heavyweight. His goal was ambitious: make Python significantly faster. This wasn't just about tweaking a few things; it was a fundamental effort to re-architect parts of the CPython interpreter, the default version of the language. The bet was that with Microsoft's resources, he could lead a team to achieve performance gains that had long been considered too difficult or impractical for the volunteer-led core development team to pursue alone.
Microsoft's Side of the Wager
For Microsoft, the bet was just as strategic. The company's growth hinges on its Azure cloud platform and its suite of AI and data science tools, all of which heavily depend on Python. A faster Python makes Azure a more competitive platform for machine learning workloads, improves the performance of tools like Visual Studio Code for Python developers, and strengthens the entire ecosystem. By hiring van Rossum, Microsoft not only gained his unparalleled expertise but also sent a powerful message to the developer community: we are fully invested in Python's future. It was a masterstroke, allowing them to directly fund improvements to a foundational technology that underpins billions of dollars of their business.
Early Returns on the Investment
The project, dubbed "Faster CPython," quickly began to show results. Van Rossum, alongside a small, dedicated team funded by Microsoft, set out to double the speed of CPython. Working with core developer Mark Shannon, the team introduced a series of innovations, including a specializing adaptive interpreter. The fruits of their labor became apparent in Python 3.11, which benchmarks showed was 10-60% faster than its predecessor, Python 3.10, in various tasks. This wasn't a one-off; the plan involves continued, incremental improvements with each new release. The bet is already paying off, delivering a faster, more capable language to millions of developers without sacrificing the backward compatibility that is crucial to the Python ecosystem.
