Today started out with a Marshmallow Challenge exercise organized by Alex Finger.

He saw this originally presented as a TED talk. The idea is that you divide into teams, and each team is given 20 pieces of dry spaghetti, a yard of tape, a yard of string and a single jumbo marshmallow. Each team has 18 minutes to elevate their marshmallow as high as possible by building a free standing structure using just those materials.

It was fun, although my team was robbed as Alex ignored the fact that Seth and Antonio’s structure fell over after time had expired and they had to set it back up, but I guess this is to be expected since Alex can be considered a French judge.

Sometimes life is not fair.

Anyway, the rest of the day was pretty cool. With so many people here from Europe we discussed opening a European office of the OpenNMS Group, and I think this is pretty close to happening. From the frequency of commits, lots of code is getting written, and folks seem to be having a great time.

Any Dev-Jam post wouldn’t be complete without talking about dinner. Chris Rodman at Papa John’s Pizza FedEx’d us some coupons for free pizza and wings, which went well with the evening’s Jason Straham film festival (Transporter and Transporter 2 ).

While very unstructured, we do try for some organization at Dev-Jam and it starts (and pretty much ends) with a morning standing scrum.

We had to switch to a standing scrum after the first day since people were so excited to be working together again, scrum was taking forever to complete. By making everyone stand, it helped shorten things considerably.

I spent the day updating the OpenNMS Wikipedia page. This was my first foray into editing Wikipedia, and while none of my edits have caused problems, we did end up having the page for The OpenNMS Group deleted rather quickly. Wikipedia has a policy of not including anything but notable companies (for which we don’t qualify) but I wanted to have a page there, not to promote The OpenNMS Group, but to better separate the .com side of the project from the .org. Oh well, I understand their reasoning, so all we need is about 30-40 independent articles on the OpenNMS Group to try again. (grin)

Toward the end of the day the room filled with the wonderful smell of smoked BBQ, courtesy of Mike Huot.

Once again we had an amazing meal, and I really appreciate the work he put into it.

Dev-Jam feels less to me like a conference than a tradition. Most of the people here have been here before. There is no need to really organize or direct anything – it just happens.

One of the traditions is that everyone starts out with grand plans. By Day Two they tend to get scaled back quite a bit. I started out with the idea of organizing the wiki. Now I’ll be happy if I can get a decent Wikipedia article on OpenNMS written.

Yesterday I teased DJ Gregor about missing his first Dev-Jam. He responded by sending us 14 flavors of Jeni’s Ice Cream.

All is forgiven.

I am, however, worried about the geeks now playing with the dry ice.

Tonight we also continued a tradition we started last year by having Brasa cater dinner. If you are ever in Minneapolis-St. Paul, be sure to check out this restaurant, and if you can’t get around to it, at least plan to come to Dev-Jam next year.

It may become a tradition.

Back in 2005, I invited a number of OpenNMS contributors to come to Pittsboro, stay at my house and spend the week working on the project. That was the first annual OpenNMS “Dev-Jam”. We had five people show up, and it was such a success that we decided to make it a yearly event.

This year we have 19 guys hanging out at the University of Minnesota for this year’s Dev-Jam. They come from eight countries (Canada, France, Germany, Italy, New Zealand, the UK, the USA and Venezuela) and three of them have made every Dev-Jam since that first one, oh so long ago (DJ Gregor, who, prior to this year had made them all, can’t make it due to some wussy work obligation).

It is one of my favorite times of the year, and I can’t wait to see what new amazing OpenNMS goodness arises out of this week. We’re having a cookout on Wednesday, so if you are in town, drop me a note and feel free to stop by.

The last day of Dev-Jam is always about commitment – well, commits to git anyway. After a solid week doing all things OpenNMS, parting is always bittersweet.

We took our group picture late, so we’re missing Bill and Matt R. The camera is Alex’s but the photo credit goes to Jen, one of the dorm advisor’s that Alex convinced to take our picture. I was stylin’ in my Doktor Kaboom t-shirt.

After we cleaned up the Club Room, 19 of us headed over to Town Hall Brewery for our last meal together this year. I think most people felt this was the best Dev-Jam ever: great facilities, awesome bandwidth and wonderful weather.

On Saturday the shuttle picks half of us up at 9am, while the rest will leave over the course of the day (well, except for Ronny who has decided to stay another week in the US). It will be nice to be home, but I’m not looking forward to a return to the hot and humid weather I’ve managed to avoid for the last two weeks.

We should be publishing a Dev-Jam wrap up in the next week or so, and I’m already looking forward to next year.

Things are starting to blur together now, so I can’t really remember all I worked on Thursday. I know I played around more with RT, and in the early afternoon, Ethan came over and we did a podcast with John Willis and Michael Coté.

I’ll post a link when Coté puts it up, but I think it was one of my favorite podcasts of all time. Willis got to gloat when I said that I liked “the cloud” and I got to talk about some of the scalability features of OpenNMS, such as the ability to discover and manage devices with 32,000 interfaces each (virtual, of course) and a test we ran for the Department of Energy where OpenNMS was handling 125,000 syslog message a minute – more than the Netcool/Omnibus syslog probe could handle.

Oh, and OpenNMS did it for 8 straight hours before we stopped the test.

Ethan got to talk about Nagios XI and we had a friendly debate on the open source services model and the commercial software model. If you are in to that sort of thing, it will be worth a listen.

For dinner that night we ate leftovers, and then went to Big 10 for the weekly pub quiz. We started to play but got distracted by a game that Ben introduced to us called Mafia.

There is a moderator, who removes all of the aces and face cards from a standard deck, except for two aces, a king and a jack. They then add enough “plain” cards so that everyone playing gets one, and they are dealt out.

The people with the two aces are “mafia”. The person with the king is the “inspector” and the person with the jack is the “doctor”. All of the rest are villagers. The moderator then launches into a story about night falling on a village and every goes to sleep. Everyone playing shuts their eyes.

He then instructs the “mafia” to open their eyes, and then silently decide which person in the game they wish to kill. Once a decision is made, the moderator has them shut their eyes.

He then asks the “inspector” to open their eyes. The inspector can then indicate a person at the table and ask the moderator if that person is mafia. The moderator will then indicate “yes” or “no” and the inspector closes their eyes again.

Finally, the “doctor” opens his eyes and the moderator asks them to indicate if there is a person at the table they want to save, and then they close their eyes.

The moderator starts their story again, stating that dawn has come to the village, and tragically someone has died. The game is then opened up for discussion and the villagers must decide on someone to lynch. That player “dies” and the game repeats until either all of the mafia are dead or all of the villagers are dead.

If the mafia targets the person the doctor chooses to save, no one dies in the night.

It’s actually a pretty fun game. Even if the inspector knows who a mafia member is, it is doubtful that they would flatly state they were the inspector since the remaining mafia member would obviously target them next. It is also doubtful that the doctor would save anyone but themselves in the beginning (although if the inspector identified himself the doctor might protect them in the next round).

In our game the mafia targeted me in the second round, but the doctor saved me so I didn’t die (you don’t learn this during the game but I was told afterward). The villagers were victorious but it had nothing to do with me, since the final mafia member was Antonio and I kept arguing that it was stereotypical to blame the only Italian at the table.

I was wrong. (grin)

This was an incredibly long and busy day.

In a normal week I think about OpenNMS a lot, but during Dev-Jam there is so much energy that I think about it more (if that is possible) and it makes it hard to sleep. I stayed up late the night before working on the new opennms.org website and I woke up around 5am and couldn’t get back to sleep for all of the new ideas swimming around my head.

So I got up, did a bunch of .com work (including payroll – it’s that time of the month once again) and wrote what may be my last post on open core (probably not, but we can always hope). I then went downstairs to join the rest of the team.

It was distracting. Mike Huot brought in his smoker and was cooking what would become our dinner and it smelled amazing. They had to start on it the night before and, to jump ahead, the results were delicious.

There was a lot of work going on, and on a whim Bill decided to discover the network we were using. Here is a screenshot of the unfiltered nodes:

I worked with Alex on RT for most of the day, and managed to take a short nap just before Ethan Galstad and Mary Starr showed up.

Ethan has been a visitor to Dev-Jam before (he lives nearby) and Mary is his business partner at Nagios Enterprises. Nagios Enterprises is Ethan’s commercial software company that builds on the open source Nagios platform to deliver an extended and supported commercial version. That differs greatly from the OpenNMS business model, so we had a lively debate about it.

First, even though Nagios is open source, Nagios XI is presented as commercial software. Just like my Hyperic example yesterday, go to nagios.com and search on “open source”. No matches. Ethan is 100% transparent about the commercial nature of his product. Nagios XI is not open core.

Second, I have often said that I see software taking two paths: either becoming a commodity or becoming open source. Ethan has structured his business around commoditizing the Nagios platform and it is priced accordingly.

Finally, people have been building proprietary software add-ons on top of Nagios for nearly a decade, and Ethan quite simply wants a part of it. As the main person responsible for the project, he has built a brand of considerable value. Just now a Google search on “Nagios” returns “About 9,470,000 results” (OpenNMS is only around 123K hits). That’s an impressive number.

It does illustrate a difference between the communities around Nagios and OpenNMS. From the moment I took over the administration of the project, I have relied heavily on the community to keep it going and make up for my considerable shortcomings. In contrast, Ethan has been the primary author of most of the Nagios core code.

He asked me point blank why we didn’t produce a commercial version of OpenNMS. I pointed out that our market was squarely aimed at “open source network management” and that we didn’t have any expertise in selling commercial software, but the truth of the matter is that I don’t feel the same ownership over OpenNMS that Ethan has toward Nagios. While The OpenNMS Group does hold 100% of the copyright, it would just seem wrong to me to build on that work and not give it back to the community, in any fashion.

If you have never met Ethan, please understand that he is one of the nicest guys I’ve met in this business, and his business partner Mary seems very competent and brings a strong business background to the company. Note that what I have written here our my thoughts on our conversation and Ethan may have a different take on some or all of them.

So we stuffed ourselves with barbecue and talked business for several hours, then Ethan and Mary left and many of the rest of the gang went off to an outdoor showing of Sherlock Holmes.

It had been a long day. I went to bed.

I am happy to announce that I was finally able to get the new www.opennms.org website to a point where it could go live. We started this process nearly a year ago and managed to get the www.adventuresinoss.com site finished early in the year, but for a variety of reasons we just couldn’t finish the other one.

Outside of cosmetic changes, not much has changed. At the heart of the site is still a wiki, but we wrapped a few information pages in front of it to help introduce people to the project.

There is also a tighter coupling between the .org and the .com sites, but we will strive to keep the commercial content on .com and the project content on .org. I do hope that this will end the occasional question we get on the mailing lists about whether or not one can get commercial support for OpenNMS, however.

I’ve been pretty heads down on the site, so I’m not sure what everyone else has been working on, but there is a scratchpad page that is tracking some of it.

Dinner was catered by Brasa and most agree that it was in the top five Dev-Jam meals of all time.

I’m not sure how much blogging I’ll be able to do this week, since it is way more fun to hang out with the gang, but I thought I needed to post at least one picture.

Thanks to Chris Rodman and Papa Johns sponsoring dinner, we were inundated with pizza and wings yesterday and it was nice to be able to eat without interrupting work.