People who write open source do it for analogous reasons, except the "strategic incentives" are often personal. Those reasons can change quickly when shiny new things appear. There's a wealth of abandoned OSS projects that illustrate this point.

Anyone working on Mesos these days?