Dev-Jam 2008 – Day 5

Dev-Jam 2008 is pretty much over but the crying. We sent a few people people home today and the rest of us head out tomorrow (well, except for Alejandro who is making a side trip to California on Sunday). Things are packed and ready to load into the van, and my guess is that everyone is ready to be home.

We also hit a milestone. At around 7:30 pm Atlanta time Alejandro Galue submitted the 10,000th code commit to the OpenNMS repository (a unit test for a RESTful interface to retrieve node data).

w00t!

This time last year we were at 6985, so that’s 3015 commits in a year. Not bad since this code goes back to 2000, and thus 30% of all commits to the OpenNMS code base were done in the last 12 months. The project is really ramping up.

Plus, our favorite analyst, Micheal Coté, put his interviews up today. Skip mine and get to hear from some of the other folks involved in the project.

Measuring Open Source Success: Become Superfluous

This has probably been the most successful developer’s conference we’ve ever had. The main reason is that the core community is becoming mature with respect to the code.

Think about it. Most of the highly successful open source projects you have heard of are code-centric, i.e. they tend to be used by programmers or people with a strong programming background.

OpenNMS is different. The main users of the product are network and systems administrators. Sure, they’re used to writing “glue” code and scripts, but they usually don’t have any formal training in programming.

For the last several years we’ve been holding these conferences to help convert those decades of management knowledge into high quality code. Matt (who does have a tremendous amount of programming knowledge) has been a a great mentor and as this conference shows we’ve come along way.

When I took over maintaining OpenNMS in 2002, it was just me. I was pretty much responsible for everything, and so I pretty much knew all there was to know about OpenNMS. If a bug got fixed, I fixed it. If a question got answered, I answered it. I did as much as I could, and at times it was scary and it was always overwhelming.



Back Row: Alex, me, Alan, Craig G., Craig M., Jeff, Matt, Mike, Alejandro
Front Row: Johan, DJ, Walt, Jonathan, Rob, Dave, Bill, Ben
(photo by Alex)

Not anymore. The project is growing so fast it is amazing. Today Mike Huot pointed out that even with the majority of the people involved with the project here in Atlanta, the mailing list is pretty much maintaining itself. People are coming up with features, writing code, and fixing things without any interaction with me.

I have become superfluous.

So many open source projects revolve around a single person or a very small group. For whatever reason they never seem to let go. Others only exist because some corporation controls them and the majority of the work is done by the employees of that company. I’ve always found that such organizations quickly become isolated from the needs of the end users and their efforts start to focus more on what generates press or short term revenue than on solving real-world problems. OpenNMS is special in that, for some strange reason, we’ve created something with a life of its own; a whole that is bigger than the sum of its parts.

This frees me up to tell stories and write silly blog posts and to play with an OpenNMS that gets better every day. I get to go out and solve management problems using a powerful tool purpose-built for the job. We must be doing something right, because when a long time customer called me today to schedule our annual week of training, I looked at my calendar and replied “December”.

So let me say “thanks” to everyone involved with OpenNMS. Let’s go do great things.