Business of open source: my take on “open core”

April 1st, 2010 by Eric Barroca Leave a reply »

The “open core” debate has been around since the inception of the model itself, and gained significant adoption among so-called “commercial open source companies”. The debate is heating a bit these day, and I got inspired by 2 recent blog posts from Brian Prentice and Matt Aslett. I advise to read both posts to understand the reactions bellow. :-)

I think a lot about business models leveraging open source as it’s what makes our business run and I’m deeply convinced that “open core” is fundamentally flawed. This conviction has been formed by real world observations: running a real business for several years now. And I’m happy to see high-profile analysts reach the same conclusion with a sharper analysis and better process! :-)

About Sales & Marketing Savings

James Dixon from Pentaho argues — not suprisingly — that open core is the right model because it reduces the sales and marketing costs to distribute the software. And that might be true. But it is not linked to open core neither to open source, intrinsically. It’s linked to the internet distribution model and the rise of the web as a way for new and small companies to become known.

I would also add that I think the open core market is less powerful as a distribution model than a freemium (or free trial)-one based leveraging proprietary software. The reduction in sales and marketing spend is more related to the online distribution of the software than the open source aspect of “a software core”. So, of course, when you distribute a software online, you have to make it easy to install (hence improve packaging compared to traditional proprietary offerings). But it is really linked to internet-enabled distribution, not to the open access to the source code of the software.

Atlassian (or Skype) might be the flagship of success for this model… without being open source at all (but while contributing to it)!

About the “open core” model

With an open core model, you have to exclude/remove features (hence value) from the open source software to in order to preserve your business: you could be forced to even limit innovation in the open source branch because it could damage your business. It does not leverage any benefits that should be derived from the open nature of the code — which is the very core aspect of open source. It only leverage free distribution not the open aspect of the source code.

I would add that a subscription-based model (where subscription is for maintenance and support services) is superior on the long-term because it drives the company to align and organize itself around the value created for its customer. The more the customer deploy and benefit from the software, the more revenue goes to the vendor. So I think it’s a great way to align vendors and users on the same value stream.

Also, using an open source license does not require free download (advocated by open core users): RHAT is, I think, considered as the most successful company in the open source world. Yet their software is not freely downloadable, while 100% open source.


I read a lot about “hybrid models“, based some blend of proprietary and open source software, to justify/support an “open core” model. I don’t believe in this at all. Of course, proprietary software is not going away anytime soon. It’s a valuable and proven business model that works and can deliver value to customers when properly applied. Same for open source: it’s here to stay for good. But the “open core” model offer no long-term benefit for customers: it just blurs the real value of open source for them, only leveraging the distribution channel it implies for the vendor benefit.

I believe in clear business models. Successful companies use clear business models because that’s what enable trust from customers. Open core is not one of them: there is no clear line between open source / proprietary, neither serious justification for the customer. Open core is just the good old proprietary model, nothing new here. All vendors uses open source software today!

From a higher perspective, I believe that the whole IT market is moving toward service-based approaches — SaaS paved the way — because it aligns customer value with vendor revenue. That’s why we — at Nuxeo — won’t use the open core model even if it could increase short term revenue. We’re here to stay and we believe that basing our revenue stream on the value we create for our customers is the best way to create sustainable growth.



  • James Dixon

    Hi Eric, Maybe I have this wrong, but doesn’t Nuxeo Connect include ‘Premium Tools’. As far as I am concerned you are the CEO of an open core company. Yet you are bashing your own business model. James Dixon

  • Eric Barroca

    James, Happy to see you there. Nuxeo Connect include SaaS. Not tools / code / additional features to apply on the open source software. Pretty much like RedHat for its OS or for JBoss. So, no, we’re not an open core company. And again: I don’t have any problem with the open core and the proprietary approach. I do believe there is value delivered through these models. I’m just saying, it’s important to not blur the positioning and undermine the value of the ones that take a different route. Open core software is not open source software (not more than proprietary software embedding open source components is open software software), just because open core is a business model (most known as freemium) while open source is not (it’s just a licensing / development model). While I agree open core is a great and fast way to monetize open source, cause it leverages the open source brand value a lot more than proprietary software could do. I just don’t believe open core is the best way to deliver value from OSS to customers. But that’s just my opinion. ;-) Cheers, EB.

  • Rahul Sundaram

    So let’s keep this simple: Is any part of Nuxeo that you sell proprietary code is it all 100% free and open source? Otherwise, the comparison to Red Hat model is invalid.

  • Eric Barroca

    @Rahul: 100% of the code we sell / you install and run is LGPL. Period. So I think the comparison is very valid. :-)

  • James Dixon

    So can we imply that the bits we don’t install – the Premium Tools in you SaaS offering – are proprietary? If so, then you are wrapping proprietary bits around open source software, and selling that (in a hosted service) to your customers. If I had to say whether this was an open core model or a pure-play (professional services / support only), I’d say you’re running an open core model. James

  • Eric Barroca

    Your answer does not bring anything to this argument. I am not saying proprietary models are evil neither trying to play the game “I’m more open source than you”. I am arguing against the open core model because I think it is flawed in a long-term perspective and suppresses the main benefit (in my view) of open source software for the customers. I believe a model based on (recurring) services (be that cloud, packaging, update, etc.) is better because it allows to combine the benefit of open source for the customer (enable collaboration & innovation around the source code) and predictable revenue for the vendor. If you want to mix proprietary and open source code in the software you sell, I believe the “patron model” or the model of “proprietary software using open source components” (aka all proprietary software nowadays) are perfectly fine and viable. They are clearer and easier to understand for the customers + still enabled a lot of innovation around the open source components (living outside the vendor’s home). That said I have nothing personal against your company neither its model and wish it succeed. I just believe open core is not a good business model on the long run. Merry Easter, EB.