Contributers first, users as a result

I've come to a growing realisation that given the amount of time I have to contribute to CruiseControl is unlikely to grow and yet it seems a shame to let the project stagnate and fade away as the prominance of Hudson and to a lesser extent, the commercial CI tools grow.

It seems appropriate at least that we should stick around if only to keep Hudson honest.

How can the situation be resolved?  Limited time, legacy code base, but still a large number of existing users and community.

I've encountered a phrase in the Lean community: "Employees first, customers second, shareholders as a result"

I think this is what needs to happen with us.  Instead of employees though, we should think contributers.  And not just the committers, who are all scarce on free time, but mainly leveraging the efforts of community contributers.  Make it easy for the percentage of users who are eager to contribute.  So I as a committer, now focus first on making life easy for contributers, trusting that they will focus on making life easy for users.

That's my new strategy.

What makes things easier for contributers?  A few thoughts...

1. Responsiveness to contribution - patch triage lead time, answering questions on the dev list, etc.
2. Less cruft - confusing code, setup, and boring tasks that should have been automated
   
3. Clarity on how to contribute - obvious on what is expected in terms of coding style, tests, documentation, etc.

In order to keep me honest, I plan to use this blog to record the progress of this strategy and see how it goes.

Next step... triaging reported issues and patches immediately...