Open Core or Open Snore?
by John Mark Walker - Oct. 01, 2009Comments (4)- Related Blog Posts
- The Great Software Freedom Debate, Part 368
- Sun Exec Proposes Software Freedom Definition and Vendor Scorecard
- More on Open Core
- EnterpriseDB: A New Stake from IBM, and its Novel Approach
If you had asked me in 1999 if we would still be having discussions on the viability of open source business models in 2009, I would have looked at you incredulously. One positive thing I can say is that the discussion is no longer about *whether* open source will survive, but how and in what form. Be that as it may, it seems like we're taking an awfully long time to learn the lessons of what works and what doesn't. Take, for example, the recent discussions around Open Core (see Andrew Lampitt's original post), Open Edge and other business models around open source. When it comes to Open Core, I have to agree with Tarus Balog when he says it's fauxpen source - although I think Tarus' argument is a bit simplistic, and I've told him as much.
What we should be asking ourselves is not what the VC's want nor what makes us ideologically pure, but rather 'what's best for our respective communities?' Too often this arugment breaks down along pre-existing faultlines in the Free Software vs. Open Source camps, which is a shame - that discussion never gives the decision-makers the information nor proper context they need. If communities are indeed the lifeblood of open source projects, doesn't it make sound business sense to maximize community success? Do IT managers and ISV executives actually know what a well-run community looks like?
Quick, name an Open Core technology which took over its target market - and no, I'm not including MySQL, because when it was first GPL'd, it was the entire code base. For a few years, there was no "enterprise version" of MySQL. I can't name any, either, although I can think of multiple success stories, both non-commercial and commercial, from projects and companies that employed different models - Linux, Apache, Eclipse, Subversion, and Fedora, off the top of my head. Therein lies my beef with Open Core: despite many attempts, it never actually seems to work. By now, one would thing that the "smart people with investment money", ie. the VC's, would recognize this. Yet, most venture-backed open source startups still prefer this model, unless they're angling for SaaS or Cloud-derived revenue.Â
The more I think about it, the more I realize how ingenius it was for Red Hat to stop making Red Hat Linux, start the Fedora project, and sell Red Hat Enterprise Linux. One of the great disappointments for me personally is witnessing a failure of execs at open source companies to learn from past successes and failures. One would think that Red Hat's success would make it an example for others to learn from - and yet, inexplicably, it's not. In that respect, the open source execs have performed no better than their proprietary peers. Remember the days when just being an open source company required that your leadership consist of true visionaries? These days, too many open source execs peddle in the derivative, unadventurous and unimaginative, and far too few think very seriously about what it actually takes to achieve maximum community success. If they were a soccer team, they'd be Chelsea.
At this point, I think we can write down a few markers of success:
1. Your community has a life of its own and is not your company's "mini me." You have entrusted your community with managing its own brand, a la Fedora.
2 . Your project is not a wholly-contained feature set of whatever product you're selling. Again, see Fedora. See JBoss, back in the day. Think of the business case - if you do this, you're designing your project to fail, because it will be forever consigned to runtware status and will be viewed, both by your community and your staff, as something only useful for upgrading to the enterprise product. It also incites unproductive conversations about freeloaders and how "your biggest competitor is your open source product."Â
I'm not suggesting that a company can never find some success with an Open Core model; I'm just saying that you're putting a cap on the possible success and losing the innovation opportunity that comes with a wildly successful community.Â
For more on this, I created a presentation on some of these topics and more at SlideShare . I am, of course, available to give this presentation to your organization, community, or conference :)
Share Your Comments
