PostgreSQL Is Not a Democracy

I found it very odd that in [ Josh Berkus’s post today on OSS project leadership], he attempted to assert that the [ PostgreSQL] project was a democracy. Now, I know of some OSS projects that are democracies, [ Debian] being one of the biggest examples, and I know Josh has been involved with democratic organizations in the past, such as [ SPI], but PostgreSQL does not follow this mold at all. The folks “in charge” of the project are known as the “PostgreSQL Core Team”, of which Josh is a member. But these people are not elected in any way, shape, or form; they are invited to join the core team strictly by choice of the existing core team members. Core team members serve an indefinite period of time, and cannot be removed by public demand; the only two core team members to lose this designation both resigned after becoming univolved with the project in the daily lives. Does that sound like a democracy to you? At best what we have in PostgreSQL is something like a benevolent oligopoly. Not that this is neccesarily bad. Sure, I’ll admit I’ve discussed what it would be like for the project to have things like core team elections and limited terms over a pint with other PostgreSQL community members, but on the whole I have no strong evidence to show that the project would be better off given rotating leadership (after all, there’s no way to ensure that people would vote the way I would want them too [[image /xzilla/templates/default/img/emoticons/wink.png alt=”;-)” style=”display: inline; vertical-align: bottom;” class=”emoticon” /]]) But, I think it is important that we recognize the type of leadership structure that we have, and be consistent when explaining it, especially from those who are in those leadership positions.