You Should Be Giving This Keynote by Mike Shaver, Co-founder, Mozilla Project
In addition to the too-good-to-be-true economics created by the upsurge in software released under increasingly liberal licenses, open development practices and "loosely coupled" projects are demonstrating the power inherent in large and diverse communities. Mike will convince you, through examples, analogies and speaking really, really quickly that the ability to capture even "trivial" contributions from all directions will be more important to the success of modern open source projects than version control, usability guidelines, review processes, licensing, marketing or a really awesome T-shirt design. You will be tested on this material.This was a very good keynote, a great way to start off the conference. Shaver started by emphasizing the importance of project management in any software development endeavor, namechecking Fred Brooks' classic The Mythical Man Month, especially the human skills needed by software developers and their managers. He then went on to discuss the overall theme of his keynote, that small is beautiful. Small in the sense that FOSS projects can survive and thrive on a lot of little contributions from a lot of different people and that small can be a good way to get started. In fact, Shaver says that we should beware people that jump into a project with grand, world-changing ideas as they care often undoable in FOSS projects. Starting small means people with only limited interest or limited skill levels can jump in and make a contribution right away. The danger of small can be that you risk loss of knowledge and continuity when the one advocate/developer of an idea or feature drifts away.
About small, to make it work you have to make it easy for interested people to go from consumers of a product to producers of that same product, find ways to draw people in. Similarly, you also need to make it easy for people to get out too, so that they don't risk burning out. People should be able to make their contribution and withdraw, if that's what they want. Small also needs a culture of testing, to make sure that "small" contributions don't cause the whole system to fail. You also need to make your needs clear on a project, possibly via a public wish list, so people know immediately if they're interested in contributing. Related to this is managing expectations, almost an anti-wishlist so people know what you don't want ie. out-of-scope, clearly for a later version.
Finally, Shaver made a plea for everyone to contribute to FOSS projects, that everyone has some X-Factor, something that they're really good at, that they can contribute. The kinds of competencies he mentioned include: writing skills, math, single-mindedness & focus, artistic ability, reading, listening (ie. tech support, bug finding).
(Update: TOC of my FSOSS posts, FSOSS agenda, video recordings of sessions)