Skip to main content

Guido van Rossum Held the Title of BDFL Until He Stepped Down in 2018

Python spent its first twenty seven years under the rule of one person. Guido van Rossum, who started the language as a hobby project in December 1989 and shipped the first public release in early 1991, carried the title Benevolent Dictator For Life, usually shortened to BDFL. The title sounded like a joke, and Guido treated it lightly, but the power behind it was real. When two camps disagreed about how Python should work, Guido decided, and his decision was final.

The mechanism around that power was the Python Enhancement Proposal, or PEP. Anyone in the community could write one, assign it a number, and argue for it in public. The PEP process is one of the reasons Python's history is so well documented: you can read PEP 8 for the style guide, PEP 20 for the Zen of Python, or PEP 257 for docstring conventions, and see exactly which arguments won and which lost. What made the model unusual was that the final accept or reject decision sat with a single person rather than a committee or a vote.

That concentration of authority made Python fast to steer. It also made Guido the target of every dispute, and in 2018 one dispute was enough to end the arrangement. The topic was PEP 572, which introduced assignment expressions and the walrus operator, the ":=" syntax that lets you assign a value inside a condition or comprehension. The debate ran for months across mailing lists and issue trackers, with strong feelings on both sides. Guido accepted the proposal, then announced on July 12, 2018 that he was stepping down from the role, writing that he no longer wanted to fight so hard for a PEP only to find that so many people despised his decisions.

What happened next is the part that matters for anyone who depends on Python professionally. The community did not appoint a replacement dictator. It wrote down the options as PEPs, debated them, and voted, choosing a governance model in PEP 8016: a five member steering council with annual elections. The PEP process survived. Instead of one decider at the end of the line, there is a council that owns the overall direction plus delegates for narrow areas, so that experts in typing or packaging can rule on their own turf. Guido remained a respected voice, later returning to work on interpreter performance, but the veto is gone.

For a senior developer, language governance is not an abstract topic. It decides how quickly breaking changes can land, how long deprecations last, and how predictable an upgrade path will be. A single benevolent dictator can move quickly, but a language used by millions of production services needs a process that outlives any individual, since the cost of an unpredictable change lands on every team that has to schedule the migration. Watching how Python handled the handover tells you something practical: the project kept its written, archived decision trail and only changed who signs off at the end.

The walrus operator itself shipped later in Python 3.8, released in October 2019, and it is now ordinary syntax in modern codebases. The lesson is not about ":=". It is that the health of a language you build on for a decade depends on the health of its decision making, and a project that writes everything down and documents the transfer of power is a safer bet than one that relies on a founder staying forever famous, available, and willing to absorb the argument.

We use cookies for ads and analytics.