I’ve talked in the past about a number of “open source” business models. There’s the MySQL “dual license” model, where all the software is available under an open license but a proprietary license can be purchased. There’s what I call the “shareware” model where some software is offered under an open license and the rest is proprietary, and then there’s the “parasite” model where companies try to build a business by exploiting the work of other open source projects. I am sure there are many more, and I’d like to add one called the “OpenView” model. This is where a company wants to use the term “open source” to create a community like the OVForum, but keep a portion of the software closed so they can reap licensing revenue.
The current darlings of open source seem to be those who have managed to sell investors on an open source business model based on license revenue. They scoff at services businesses. One recently stated that in order to be successful, an open source company couldn’t have more than (approximately) 25% of its revenue from services.
So, what’s wrong with the statement that an open source company cannot be successful unless 25% (or less) of their business is services? Well, if I’m doing the math right, that means 75% (or more) of their revenue must come from something other than services, and that can only be software licenses.
The power of open source ultimately comes from the community. The community will work together to build something wonderful, and almost by definition for that to be successful the community must own the final product. If the product is actually owned by a specific entity, for the purposes of revenue generation, the potential of that community is lessened, and so is the product.
Is there anything wrong with that? It depends. The OpenView community was one such community and it thrived for a time, but ultimately it faded. Certain “open source” management companies seem to have the goal of replicating OpenView (hence my label of the “OpenView model”) by creating software that is easy to expand via plugins (although the core code or enterprise-level code is closed), but since people are fleeing OpenView in droves I don’t believe that a similar model is sustainable for a long time.
If you write plugins for these companies, please don’t take what I am saying as an insult. As an open source guy I’m a firm believer in that what works, works. Almost all of these firms provide a lot of code under the GPL and that makes them head and shoulders above OpenView, but I have to wonder about a company that bases the majority of their revenue of software licensing and also calling themselves “open source”. What would happen if someone in the community contributed a patch that threatened that revenue stream? How can one encourage code contribution when the more code that gets contributed results in less money for the company? They can try to funnel the effort into “plugins” and tell everyone else to ignore the code behind the curtain, but that really defeats the purpose of open source.
These firms may be able to pull off creating a company worth a lot of money just by providing a useful, albeit non-free, product, but as free products become better and better, their software subscription model becomes weaker. I’ve heard it called “owning the bottom” and after originally disagreeing with the concept, I’ve become a believer. I’m betting that the next OpenView will come not from another commercial software company, but from an open source community, and I’m also betting that community is OpenNMS.
In the long run it will be the company that does services best that is worth the most. Even now the trend is moving toward services or software as a service. Which brings me to the future: what makes a services company successful?