The Future

Occasionally OpenNMS gets mentioned alongside well funded “open source” management companies. Since we haven’t announced any multi-million dollar investment, the press often ignores us. But I much prefer it when we get mentioned favorably in the company of HP and IBM. That’s where I want to see the community take OpenNMS.

So, how do we get there?

First off, it takes time. You can try, but you can’t “buy” a community. We’ve worked very hard to build the OpenNMS community over the last seven plus years. We’ve done it by focusing on two things: OpenNMS will always be free, and OpenNMS will never suck. By trying to keep the .org and .com sides separate, we hope that those people active in the community will never fear that we would take their work and commercialize it. I like saying that you can’t stick nine pregnant women in a room for a month and get a baby. Some things take as long as they take, and no amount of money can speed that up.

It has taken OpenNMS over seven years to get where it is today, and I believe that some investment at this time would go a long way toward making the product stronger. Not $24 million, $10 million or even $4 million, but something more along the lines of $2 million.

Why “just” $2 million? Two main reasons: I don’t think the market is ready to support a large open source management company, and OpenNMS is not ready to spend more than that.

Since I like to explain things, let me ‘splain.

When a company takes in venture capital, the VCs don’t want it to sit around in the bank. They have banks, nice banks, that could take better care of their money than most start-ups. They want it spent, and they want a return of 5 to 10 times their investment in about 5 years. Thus a company that raises $20 million has only a couple of years to become worth well over $100 million dollars. That’s a mighty big challenge, and my guess is the temptation to raise that money on the backs of their customers will become really strong. How would they do that if it is open source? If it was truly open source, they couldn’t, but by using a software subscription licensing model they can lock the customer in, just like HP tries to do with OpenView.

I really don’t think any of the open source options out there, including OpenNMS, can both be true to open source and become worth that much in a couple of years. It is a totally new concept for many enterprises to not be forced to pay money for software. Current commercial OpenNMS clients are obviously ahead of the curve (they are quite the smart bunch, and good-looking, too). Many companies don’t believe that they can get something equivalent to much of what OpenView and Tivoli provide for considerably less money, or that software developed by a consortium can work better than one developed in a monolithic organization. It will take time, I think a couple of years, before it will be possible to really bring something like OpenNMS to the mainstream. Everyone is looking for the “catch”, the “gotcha” and the prevalence of “open source” as a marketing term isn’t helping.

Even if OpenNMS was able to raise many millions of dollars, I don’t think what we’ve started can be rushed, and thus I don’t think i could spend it wisely. I am so blessed to be able to work with the folks I work with in the OpenNMS community that I won’t do anything to jeopardize it, even if it means short term financial benefit for myself.

So what would I do with $2 million? I call it my $2M/2 year plan. Over the next two years I would use that money to focus on the real intellectual property of the OpenNMS Group: our knowledge of network management. We make a little money each year, and thus we have a good idea of what it takes to run a services business. OpenNMS could definitely use some polish, in terms of documentation and ease of use/installation, that an infusion of cash would enable. Plus the biggest comment I get when talking about OpenNMS is that few people have heard of us. I’d like to be able to market ourselves a little better. It’s a great testament to the product that we’ve been able to survive solely on word of mouth advertising.

Now, the biggest criticism of a services company is that you have to find and hire an army of people to provide those services. Aye, there’s the rub, eh? At OpenNMS we will get around that in a number of ways.

The last two people I’ve hired have been taken directly from the Order of the Green Polo (OGP). The OGP is the core team behind OpenNMS. Imagine hiring someone and having them be productive days before they even start their job. The people in the OGP are in it because they love OpenNMS, and you can’t find better. I’d like to hire them all.

So what happens when they run out? The “top secret” success of the OpenNMS Group is that we’ve learned how to use a core group of dedicated people to inspire others, and we have formalized a process where you can take a motivated person and quickly turn them into someone who can help others with the product and maintain a “laser-like” focus on the customer. They might not know everything in the beginning, but they will be helpful and, better still, they will know where to go for the “everything”.

Identifying and training people like that is hard, but we’ve learned both through experience and through services clients of ours like Rackspace that it is possible.

One final criticism of services companies is that the people are always on the road. The answer to that is simple: we won’t be doing the traveling. We are working though the community to find partners around the country and around the world to deliver on services related to OpenNMS. The OpenNMS Group will focus on support (which can scale), custom development (which drives new features) and training, while our partners, who already have great relationships with their own clients, can have the services revenue associated with deploying and customizing OpenNMS. It’s worked well in Italy, and we have plans underway for partners in the UK, the rest of Europe, Turkey and Saudi Arabia. Since we don’t want to “own it all” we can make this happen, and use certification to insure a high level of quality.

I think getting all of the kinks out of this will take 2 years. Even if I had more money, I don’t think it can be done any faster. But I am certain by the end of those 2 years OpenNMS will be poised to become a real contender for replacing OpenView, and at that time we will have a story for the VCs that will make sense.

Now the only trick is to find an angel with $2 million to invest willing to wait 2 years. Answers on a postcard please …

It’s now about 3 hours since I started writing this, although with about a hundred edits, so I just want to sum up and go to bed. The future of software is in services, be it direct “I pay a person to do stuff for me” or software “as a service”, like Salesforce.com. Services are key to the success of open source projects. If it weren’t for services we’d have never discovered the market need for such things as an integration between CentricCRM and OpenNMS, much less gotten it done in less than a month.

Hybrid business models such as the “shareware” model or the “OpenView” model are short lived at best. Being totally free, products like OpenNMS will own the bottom, constantly driving software licensing revenue down and without a strong service model there is no way to make up for that loss. Finally, those companies that focus on service as the product, and can successfully scale it while maintaining a high quality of customer experience, will ultimately prevail.

I’ve bet the farm on it.