Why Our Governance Doesn’t Work

5:12 pm OpenSolaris

Every day I come across examples of why the OpenSolaris governance is really proving to be an obstacle with zero value return.

Consider this, James Carlson’s proposal for networking documentation sent late February. Since James is looking for a source code repository to do his work, it needs to be a Project within the scope of opensolaris.org (since that is what the website infrastructure can handle). According to the guidelines laid out in the project instantiation policy, Community Group’s are supposed to sponsor project requests, with their own discretion on how projects are discussed and approved. Once the Community Group approves the project, it gets sent to project-setup@opensolaris.org so the website team can create the project page. James is in this case using the Consensus voting system, requiring three +1′s and no -1′s from the Core Contributors of that Community Group.

17 days later, James is still looking for the required votes to start what, in my opinion, seems to be a great project. You can’t help but think how many people would have given up by now.

There’s no substitute to JFDI – or improving our governance model so that it doesn’t get in the way of suitably motivated people who want to contribute.

5 Responses

  1. Stephen Says:

    So, since the title suggests ignoring the specifics of the discussion so far, what do you propose to change about the policy or the constitution?

  2. Peter Tribble Says:

    Amen to that. I have a project proposal from Jan 20 that still hasn’t managed to qualify.

    Interestingly, as the constitution doesn’t really define projects I think we’re overinterpreting the constitutional requirements and could just change the Project Instantiation Policy to make it much much easier.

  3. gman Says:

    Stephen: Aww, you mean I have to follow up a rant with a proposal? I know it’s very easy to jump into criticizing the constitution, but I can’t get away from the feeling that this should be easier. It’s entirely possible that Jim shouldn’t be using the Consensus voting method, despite the Constitution currently suggesting it. Projects should be zero cost, in my opinion – even the worst run, least successful project is going to allow those leading it to make enough mistakes to learn from their experiences. While there’s obvious cost from a corporate point of view in terms of man months, it’s not clear what costs there are within an open source point of view.

  4. mc Says:

    > Every day I come across examples of why the OpenSolaris governance is really proving to be an obstacle with zero value return.

    I’m pretty sure this is what Palpatine said before he overthrew the Galactic Senate and took over the galaxy.

    But I agree though, the impotent and useless OpenSolaris community as it is should be discarded and reformed under the total and explicit governance of Sun Microsystems.

  5. Stephen Hahn Says:

    @gman: Yes, I would probably still pursue project approval via Consensus, but would be a super-nagger among my core contributors. But, if Consensus is too hard, Assumed could be used instead. We could revise the policy to make this more clear.

    @Peter: Which proposal is stuck? (I’ll reiterate my point on nagging, but if it’s in a group I vote in, I’ll help you nag.)