urllib3 gets almost a BILLION downloads

PER

MONTH

(a project I authored in 2008 and now co-maintain)

i often think about how inefficient and wasteful our global digital infrastructure is: there's probably a few million monthly primary consumers, plus all the ci instances spamming downloads/installs repeatedly.

urllib3 gets *over* a billion downloads per month.

very demure and mindful number, very cutesy

A doodler and computerer. I like permissive/permissionless open source, smart contracts, p2p systems, room-scale VR, and NixOS.

shazow.net

Currently: WhatsABI

urllib3's secret weapon is all the participating maintainers we've had along the way.

I literally could not have done this on my own (I don't have a multi-decade attention span for a single project, very few people do). I didn't know this at the time, but creating an environment where other people wanted to participate and felt that they had collective ownership in the project is the key to urllib3's longevity.

There was a moment where I had to let go of some ego to level this up (eg move the project from my namespace to its own org, give up some control, let other people make important decisions I may have disagreed with, etc), but in retrospect these were extremely high-value decisions that cost me ~nothing.

Everything else (being early to some ideas, having a high bar of standards like code quality and full test coverage, etc) helped too, but a distant second.

what made you commit to the painstaking task of maintaining it over the years through different python versions?