Volunteer economics is the notion that even though contributors frequently work without financial cost to a project, there are social, personal, and technical costs that affect their ability to do that work. These costs relate heavily to the question of which projects can sustain themselves on volunteer energy.
We’ll dive into three stories:
The demise of cdn.debian.net: How http.debian.net outcompeted it, through a technical structure that enabled it to grow with fewer people having to collaborate.
Why no one uses “apt-get” to install web applications: Although many server, command line, and desktop software packages are available in Debian, very few of today’s open source web applications are. This section provides a brief overview of Debian’s attempts to make web applications packageable, and an examination of where the efforts have & haven’t succeeded, and considers how volunteer incentives to maintain desktop software don’t apply cleanly to web applications.
The rise of reproducible builds in Debian: What binary-reproducible builds mean, and within 18 months, how a community member was able to create the volunteer energy required to make 80% of all packages truly verifiable.
We’ll conclude by re-visiting some common, cliche questions people ask of free software — “where are all the designers?” “where are all the documentation writers?” — and consider if these stories provide any answers.
The talk assumes no particular background with Debian or maintaining Linux systems. Some of the topics have technical content, but we will provide the necessary background as part of the talk.
(For what it's worth: This is also a talk I gave at Open Source Bridge 2015, compressed somewhat since I expect more familiarity with the topics at Debconf.)