Late last week Matt Asay argued that cloud computing is the natural conclusion of open source, writing that “while open-source entrepreneurs initially intended… services to mean support and consulting, the industry has taken open source to one logical conclusion and has crowned services as the only important software outcome” and that “the cloud takes open source to its logical conclusion, crowning services (the output of software) as king, rather than fettering us to a discussion of software (the input).”
As it happens this was the underlying theme of the presentation I gave at Cloud Expo Europe today. Entitled “How to make millions with open source, or the evolution of open source business strategies” my presentation was essentially a discussion of the evolution of open source business strategies.
The presentation was based on the results of our Open source is not a business model research report although rather than present the findings of that report in detail I decided instead to reflect on some of the lessons learned and best practices we have observed through existing business strategies around open source.
George Greve, president of the Free Software Foundation Europe, recently observed that free and open source business models “are orthogonal, like the three axes of the three-dimensional coordinate system” and that “their respective differentiators are control (software model), collaboration (development model), revenue (business model).”
In OSINABM we also found that there are three dimensions to business strategies around open source that map directly to George’s control, collaboration and revenue definitions, although we referred to them as vendor licensing strategies, development strategies and revenue triggers, respectively.
We also found that there are six different vendor licensing strategies being used by open source vendors today, as well as six different development strategies and nine different revenue triggers. Even ignoring the use of multiple revenue triggers, that equates to over 300 different theoretical combinations, supporting our claim that open source is not, in an of itself a business model.
In order to fully appreciate an individual company’s strategy it is necessary to take into account the development, licensing and revenue aspects, however, it has become clear that the major combined licensing and revenue strategies can be separated into three major categories:
• Pure open source
• Commercial open source
• Commercial – “open source inside”
To which I would add a further category which was outside the scope of our initial research: “business enabler”.
Looking at each of these categories in turn it is possible to identify the evolution of open source business strategies.
- Pure open source
Open source as a distribution model enables the widespread adoption of open source software by a community of users.
In order to generate revenue from that community, early open source vendors such as Cygnus Solutions were formed to provide support for the open source code, as well as make use of the fact that the code is open source to perform custom development, which quickly separated community users from paying customers.
However, Cygnus quickly found that there were problems scaling support and customer development, caused by a small company trying to compete with the needs of widespread open source adoption, and also in terms of a lack of repeatable business, and began to invest in contributing extensions to the core code base.
In order to achieve predictable, repeatable revenue streams, many open source vendors quickly turned to the subscription model, in particular Red Hat, which has been significantly successful through offering 100% open source code, but withholding the binaries for paying subscribers.
The fact that Red Hat generated a profit of $178.1m on $650m in revenue in its last fiscal year disproves the theory that it is impossible to generate revenue with a 100% open source approach.
However, Red Hat has become the exception rather than the rule. When many companies talk about following “the Red Hat model” they invariably mean the subscription model, rather than 100% open source licenses.
And the subscription model itself is not without its challenges. Red Hat’s CEO recently stated that the company’s biggest competitor is not Microsoft or Novell but its own customers deciding they don’t need a support plan.
The model therefore requires vendors to continue to come up with value added services that will encourage customers to renew, and these value added services only being available to paying customers begins to blur the line over what is open and what is closed.
The other 100% open source business strategy is via advertising, although we would see that more as a consumer-oriented strategy that hasn’t been widely adopted for enterprise software suppliers.
- Commercial open source
In order to overcome some of the limitations of the 100% open source model, many vendors have moved to avoid confusion over the licensing of value-added services by introducing commercially-licensed extensions that serve to separate paying customers and community users and provide an incentive for enterprises to use the “Enterprise Edition” product.
Known as the Open Core model, this is a more sophisticated take on the dual license strategy whereby a vendor that develops 100% of the code offers both a community version and a commercially-licensed version of the same code base.
Overall, 100% open source strategies are easy to establish with low start-up costs and can be quickly profitable as vendors take advantage of development and distribution benefits and establish symbiotic relationships with customers.
However support is highly competitive as there are few limits on competition, and expertise required in order to gain a dominant position. There are question marks over whether 100% open source strategies can scale, certainly to the extent required by VCs to deliver revenue on the scale that will fund a lucrative exit (and certainly in terms of a single-product support strategy, as opposed to a multi-product SI approach), and there is the constant need for additional value to encourage renewals.
If done properly the Open Core approach should make it easier to drive revenue via value-add services and provide a better separation of community and paying customers. That’s the theory, although in practice we have seen some vendors that have struggled to identify the differentiating features and have fallen into the trap of constantly having to deliver new services in order to stay ahead of community demand.
The Commercial Open Source approach has been seen as successful as it balances the best of proprietary and open source approaches – as customers are comfortable paying for commercially-licensed versions of open source code and gain the confidence of a traditional licensing arrangement.
On the other hand, this approach stands accused of contradicting the advertised benefits of open source, as many open source adopters are drawn to open source specifically to avoid traditional licensing. It also restricts the benefits of the open source development approach, as only core the core code is actually open source and benefits from community development.
It is essential that vendors are clear about feature differentiation to avoid confusion. I have expressed doubts about whether maintaining this balance is sustainable in the long-term.
Commercial – “open source inside”
It is not necessary to directly monetize open source code in order to generate revenue from it and some traditional software companies generate revenue from open source software by simply selling additional other products and services.
In practice we see more vendors including open source software within larger commercial software packages, however, benefiting from lowering costs and improving quality by being part of open source development projects such as Apache and Eclipse and embedding that code within their products – either commercially licensed software or hardware.
Almost every traditional proprietary software vendor, including Microsoft, Oracle, SAP, and IBM, is generating profits from open source software in this way.
Another emerging strategy, that turns the Open Core model on its head, is to start with commercial software and offer open source extensions on top of the code. Known as the open edge model, it has been used by Microsoft to create a community of partners and users around its customer relationship management software, and I think we may see more of this approach as we see more traditionally-licensed vendors engaging with open source.
We’ve also seen business taking the open source inside approach and utlising it as part of a software as a service strategy, combining the two to minimize both development and distribution costs.
This is where we begin to see open source software as a business enabler, as vendors are not just building business strategies around open source software, but using open source software to enable new business opportunities.
We have also have seen service providers assembling open source software stacks with which to provide hosting services to customers via which they can create their own web sites.
Beyond simple web sites we also see the opportunity for businesses to deploy multiple open source projects, alongside internally-developed extensions in order to create computing platforms that enable new business opportunities, particularly for social networking applications.
The logical extension of that is the use of open source software to create entire computing platforms for web 2.0 applications and social networking, as well as cloud computing.
When you look at the landscape of modern Internet business providers, from Google to Facebook to Twitter to Amazon, almost all of them have relied on open source software to lower their development costs and speed up their time to market with new products and services, as well as exploring the opportunities for opening up those platforms for use by third parties.
Why open source?
We’ve seen how the use of open source as a business enabler provide the opportunity to lower start-up costs and quickly bring new services to market while focusing on providing unique value services.
With this approach there is no need or desire to directly monetize open source and businesses have the ability to modify the code to suit their requirements.
There is no need to contribute back modifications (depending on licensing), so users can benefit from providing value-added features, although over-time we see that vendors are prepared to collaborate on improving the underlying code and getting the full benefits of open source as a development model.
Additionally, as Red Hat has today pointed out: “open source infrastructure technologies for cloud computing offer transparency and view of code and APIs in an effort to ensure strong interoperability from the start – no hidden APIs, no traps”.
Overall we would observe that a pure open source approach provides quick profitability and a sustainable revenue stream, but there are doubts about whether it scales to deliver the sort of revenue investors are looking for.
Commercial open source balances best and worst of proprietary/open source. It is highly successful today but is still relatively immature and there are question marks over its sustainability.
Meanwhile “Open source inside” provides benefits while maintaining the status quo. Established vendors are adopting open source licensing, development and distribution models. Whether that is seen as a victory or a failure of open source depends on your perspective but it reduces the likelihood of open source vendors over-throwing the establishment.
Open source as a business enabler is creating new business opportunities by lowering start-up costs and decreasing time to market.
While the established open source business strategies are not going to disappear there has been a change in recent years and established vendors have adopted open source and cloud computing has change the rules about how business adopt technology.
The long-term opportunities are not in open source products and services, but products and services that are built on open source.