I’ve just returned from an exciting and vibrant show at LinuxWorld Expo UK to be plunged back into the 80 hour weeks with OpenNMS that make up my normal life. It was great to see so many people at the OpenNMS booth and so many people starting to see the advantages of open source software.
However, it does seem that some people are giving up on the open source model. Nessus announced that they were closing its source due to competitors using it. I don’t have that problem – heck I view people using OpenNMS as an endorsement of its worthiness.
I blogged about Cittio’s WatchTower a couple of weeks ago. It was pointed out to me that some features of WatchTower resembled OpenNMS, and so I started to investigate it. I set up a call with Jamie Lerner, the CEO/President of Cittio, and I just got off the phone with him.
Let me say from the start that Jamie comes across as a nice guy. He was very upfront, didn’t try to dodge any of my questions, and seemed very sincere about insuring that Cittio maintained a good relationship with the communities of the open-source projects they used.
OpenNMS is a part of WatchTower, but they are using a version from mid 2003 (something like 1.1.0). They haven’t modified any of the code. They use various daemons in OpenNMS to generate both information in the database and RRD files, and their other processes access that data outside of OpenNMS.
But, you might ask, their Orbital Architecturefeature where they can make changes to the configuration files without restarting implies they modified the code. From what Jamie tells me, they ”do” restart the OpenNMS daemons they use when a change is made, just not the WatchTower application itself.
They are starting to run into some of the scalability issues that plagued us back then, and they would like to get involved in addressing the data collection issues we run into on very large networks. I first advised them to check out a later version of OpenNMS (those of you who have been around for a couple of years know that we’ve improved by orders of magnitude) and then decide in what areas they would like to contribute, and we’ll set them up with a branch on Sourceforge.
The last topic of discussion was credit. The OpenNMS code is a volunteer effort, and just as I always try to acknowledge contributions in the CHANGELOG, it would be nice to see some sort of “we use OpenNMS” blurb with respect to WatchTower. I brought up with Jamie a similar issue with The Gimp on OS X by the fink project. People who work on open-source code like to have that work acknowledged. Jamie promised me that they would address that as well.
All in all it was a good call. There is room in this world for commercial and open source to play together. My goal is for OpenNMS to be the default management platform for everyone, and that means that the applications that use that platform will span the gamut of licensing models (Ciscoworks anyone? Bueller? Bueller?).
OpenNMS itself, however, will always be free as in freedom. Everyday I see more and more evidence that open source works, which is why I am disheartened by the decisions of folks like Nessus. They state that they don’t get many contributions from the community. That’s a shame – and something I hope to avoid. By inviting companies like Cittio to get involved, we can all benefit.