I posted this a few weeks ago to the ubuntu-motu mailing list. Perhaps planet would like to see it too.

I'm writing this on the flight back from UDS1. Traditionally this week is when I contemplate my usefulness as a MOTU.

We had a couple of sessions this week about ways we can involve new developers (I'm using developer here to mean what you might incorrectly call a packager. Think ~ubuntu-dev) and find interesting things for them to do.

One idea I had is holding "projects". Say I'm running a Haskell transition which involves rebuilding a couple of hundred packages in the correct order, checking their dependencies are correct and then uploading to the archive. I could create a project out of this by describing what needs to be done and what level of experience someone will need to help me out. Others will do the same for their projects (whatever they may be), and contributors can then scan this list for something to work on in partnership with an existing developer. The developer gets some useful help and the contributor gets some useful experience. Get in contact with me when you next have a project (good candidates may be transitions or porting work) and we'll work out how to get it listed.

Mr. Motivator

Anyway, that wasn't the point of this post.

What I now realise that we completely missed is helping with the motivation of existing developers. Taking myself as an example, while my motivation in the specific areas I find interesting remains high, for the goals of MOTU more generally it is as low as ever. I'm a terrible sponsor. I almost never look at this or this or this. Do you?

Dear MOTU, how do you feel about your work on Universe and the team in general? How can we make continued contributions to Universe more rewarding? It is my impression that MOTU is often seen as a gateway to the more sexy work available in the main component. Is this accurate? If, as I suspect, it is then why is that?

I think a big part of the problem is that a lot of the work is difficult, repetitive and ultimately dull2. Who wants to grind through a huge list of failed upgrades and tease out what went wrong? Certainly not me.

The task is daunting, the team small and the rewards scant. When I first got involved I had little to no perception of the scale involved.

These are my feelings, but I'm really much more interested in hearing from my colleagues in MOTU. Please let's have a discussion about this. I care a lot for Universe and feel that it is often neglected in discussions around the distribution as a whole. We're here to take care of it, and I feel that, on the whole, we could be doing a better job.

I apologise for the negativity of this post. Actually I do feel that, despite constantly treading water, we do manage to deliver a decent Universe component every cycle. If only we could keep the old hands as enthusiastic as they were when they started. Yes, a big part of our job is to help new contributors be effective, but we should also be helping ourselves to lead by example.

Epilogue: Since I wrote this post, I've gotten involved with some aspects of release management, namely running the transition tracker and joining the release team. Having some kind of responsibility has helped motivation somewhat.

  1. Since you asked, it was a really fun and productive week with some interesting ideas. And Budapest is a really lovely place. And the Hungarian LoCo team has some great tour guides. ↩

  2. (advance warning: profanity ahead) In fact, in a similar discussion at UDS, we described the job of MOTU as "shovelling shit". ↩