I was privileged to attend the NumFOCUS Summit this year. The summit itself took place Nov 2 and 3, and I stayed an extra day to attend the sprints that followed. It was a great opportunity for informal discussions with the NumFOCUS-sponsored project teams as well as representatives from some of the organizations that sponsor NumFOCUS.
It may be because I am now more involved in the internals of the open source world, but it seems as though the idea of paid work on open source projects is taking hold across the spectrum of projects working in data science. Both from the perspective of contributors wanting to turn their passion into a full-time gig, and the perspective of prospective funders. Working on open source is no longer viewed as something that must be done only by volunteers. NumFOCUS presented the various paths they provide to help sponsored projects obtain funding: direct corporate or private foundation sponsorship, targeted donor campaigns, partnering with research institutes, and more. The total funds available via these channels is growing every year, leading to the next challenge: managing a modern open source project.
Not all the sessions were about funding. There were interesting talks about understanding the bigger picture of open source in the context of the society around it. Open source software is one example of a "public good", which is a well-defined economic concept. There are economic and social forces shaping the way open source projects develop: what happens when a large commercial entity forks a project and begins to develop without the original team? How can we create incentives so that companies who contribute time and money to open source are not put at a disadvantage versus companies who only take without giving back? These are open questions that were very much in the foreground of discussions at the summit.
There was plenty of time for socializing at the meals and in the evenings. It was fun to bounce ideas off other attendees, and learn a little more about people who I only "talk" with on issue trackers and mailing lists. It turns out many people have lives outside of coding!
I especially enjoyed the "unconference" sessions where anyone could put up a note, grab a room, and meet and discuss topics near and dear to their heart. I attended sessions on the Outreachy program and diversity, discussions on designing websites and user surveys, and how to interact with your employer around topics of contributing to the wider community.
The last day I was sprinting with people from other NumFOCUS-sponsored projects. We shared information on a number of technical topics around integration, delivery, and continuous integration.
In short, it was a great opportunity to grow, share, and meet with like-minded people. Thanks to the NumPy community and NumFOCUS for making it possible.