Today, I am inspired. It could be the sunlight, or just a random chemical reaction in the brain, but for whatever reason, I have above average energy. It could also have something to do with what I have begun to read…a blog (?) on the internet called “The Cathedral and the Bazaar.” Check it out here, but hurry back!
Go ahead, try to tell me you wouldn't hire this guy to build you software.
I am only halfway through it, but needed to share with you some of the authors insights from his experience with software programming. Often, in the business world, I see people and institutions hang on to their aging knowledge and believe that there are some people – i.e. a small amount of “qualified” people – that hold the keys to business success. (Apparently, the good ideas of qualified people are superior to good ideas of unqualified people). If only everyone else could have their knowledge and they could effectively spread that knowledge to the great unwashed of the business world. Boom…utopia achieved.
Reality suggests that upstarts sometimes succeed, and the people who “know” how to run a business fail. Big companies get stale, lose money, and die a slow death replaced by those run by people who don’t know what they are talking about. How can that be?
The man with three first names – Eric Steven Raymond – was one of those careful planning people…until just before he became great. He was an early pioneer of email software (fetchmail for the techies out there – all three of you) that successfully developed the software using the bazaar approach rather than the cathedral approach.
What’s the difference? Cathedral builders carefully plan the design. A small group locks themselves away and carefully engineers all the details. And the bazaar? A system emerges from relationships between a bunch of rowdy, passionate people – some geniuses, some far from. Out of the chaos comes a functioning marketplace. I’ll avoid extending the metaphor any further (although can’t say I wasn’t tempted), but you get the idea.
So why does this work? Causality is something tricky that I refuse to make snap judgements on. I can’t say precisely why, but I know it does. It’s the same principle upon which prediction markets are based. It’s the same reason why ant colonies don’t need a take-charge “leader” to tell them to change course and head toward where the food is. It emerges from a series of behavioral adaptations based on what works.
What say you, Mr. Raymond? “You often don’t really understand the problem until after the first time you implement a solution.” Yep, what else? “Every good work of software starts by scratching a developer’s personal itch.” Yep, tap into people’s passion as a leader and you’ll get great results. Any other advice? “Linus Torvalds’s style of development—release early and often, delegate everything you can, be open to the point of promiscuity—came as a surprise. No quiet, reverent cathedral-building here—rather, the Linux community seemed to resemble a great babbling bazaar of differing agendas and approaches (aptly symbolized by the Linux archive sites, who’d take submissions from anyone) out of which a coherent and stable system could seemingly emerge only by a succession of miracles.”
That is so well-said that I have absolutely nothing to add.
It's more fun than a TPS report... 