Home

Personal Blog of Shaun Walker

 
     

DNN's Embedded Open Source Opportunity

By Shaun Walker on 2/19/2015
Despite the fact that I created the largest open source project native to the Microsoft platform and also founded a successful company based on its widespread adoption, I still feel like a student when it comes to open source business models. Although open source software has been around for 30 years it has only been in the past decade when the marriage of open source and commercialization became more mainstream. And since these two concepts are such strange bedfellows, it stands to reason that companies have chosen many different approaches to try and achieve success. To date no single model has ever emerged as a clear winner in terms of open source business strategy. This is because each open source community has such distinct philosophy and culture, there can not be a "one size fits all" approach. Ultimately, open source communities exercise a significant amount of influence over commercialization efforts, as the freedoms of open source software licenses regulate the amount of control that any single entity in an open source ecosystem can maintain over a project.

I recently came across a blog that was written in 2009 by Matt Aslett of The 451 Group. Matt is currently a research director for data management and analytics within 451 Research's Information Management practice. However he is also highly regarded as an expert in open source software, as he has covered the enterprise open source market since 2006. The blog was written in 2009 and is related to open source business models. It contains a variety of observations and predictions that I found particularly interesting to read in hindsight, especially based on my own real world experiences with DNN. Coincidentally, DNN Corp began its own commercial open source journey at the same time that this blog was published, launching DotNetNuke Professional Edition in early 2009.

The blog was focused on the Open Core business model, which is essentially the distribution of a fully functional open source product and a separate commercial edition with proprietary features and professional support ( this was the business model which DNN Corp chose to pursue in 2009 ). And despite the popularity of this model at the time, Matt questioned whether the Open Core model was sustainable in the long-term, and he predicted a shift away from vendor dominated open source projects ( ie. MySQL, etc.. ) to vendor dominated open source communities ( ie. Eclipse, etc.. ). He also offered some advice to companies that were utilizing the Open Core model; suggestions that were exceptionally prophetic when I consider them within the context of the DNN ecosystem. 

In looking at the specific points that he raised, I thought it would be interesting to consider them in hindsight from the perspective of DNN's own commercial evolution and offer my own personal opinion on how effective we were at achieving each one:

Truth in Advertising

Matt's advice was that it is important to be absolutely transparent about licensing. If the open core commercial product is being offered under a license that is not a standard OSI-approved license, then the commercial product should not be described as open source. This was indeed the case with our DNN Professional product, as it was distributed under a proprietary software license which had different rights and restrictions than the standard MIT license used for the DNN core. My assessment is that DNN Corp was transparent about the commercial license - it was front and center in every sales engagement and customers would usually review it extensively as part of their diligence process. The only instances where challenges occurred was when the commercial license terms were modified in a way which affected customers rights retroactively.

Don't Promise What You Can't Deliver

Matt's guidance was to avoid overselling the benefits of open source in the context of the commercial product. I believe DNN Corp did an effective job of focusing its marketing efforts on the value-added services and the benefits of the commercial product in terms of how it solved business problems for customers. The fact that the product was built on top of a mature and widely adopted open source product was merely an attribute - but not the primary sales pitch delivered to prospective customers.

Separation of Church and State

This particular point was our largest challenge with the Open Core business model. Matt eloquently described the problem as "the further removed the proprietary functionality is from the needs of community users, the less likely it is that the community will develop their own alternatives, and the easier it will be to manage the needs of both commercial and community users". The reality is that once it was decided that feature differentiation was going to be a primary sales lever for the DNN commercial editions, it created numerous challenges. This was primarily due to the fact that there was never a concrete set of guidelines that could be used to judge if a particular feature belonged in the open source offering or the commercial offering. This led to artificial restrictions to the user interface and API of the core product for capabilities that would have been beneficial for all members of the ecosystem. It also led to endless debates with the community about whether DNN Corp was acting in good faith as a proper steward of the open source project.

Care in the Community

DNN Corp was fortunate to be able to hire many of the most experienced and respected community members from the ecosystem, cementing a strong relationship between the company and the community. It also created a Forge and numerous other community-focused services and channels to support the needs of the ecosystem. Unfortunately over time the resources devoted to these types of activities were scaled back, impacting their benefit to the community and causing some discouragement. In addition, commercial partners were actively discouraged from participating in community channels which created fragmentation within the ecosystem and a loss of potential opportunities.

Keep an Eye on the Bigger Picture

There are always trade-offs in business but the best strategies do not sacrifice long term potential for short term results. Unfortunately the pressure of showing high growth fincancial results quarter-over-quarter, sometimes results in business decisions that place an emphasis on hitting current numbers at the expense of achieving future objectives. In the open core model it is easy to get so consumed with the customer conversion process that you lose sight of the care and feeding of your customer acquisition channel. Open source communities require continuous investment to ensure that they are healthy. This includes investment in both the open source product to ensure it remains relevant, as well as in community services which provide mechanisms for users and companies to engage with one another.

Matt's comments about a shift toward "vendor dominated communities" as opposed to "vendor dominated projects" was interesting. "Vendor dominated communities" do not necessarily mean open source communities that are 100% dominated by vendors - it simply means that it is important to have a diverse set of commercial stakeholders involved in the evolution of an open source product. This is also possible in a "vendor dominated project" but more often than not, the vendor will have very specific goals they wish to achieve which often restrict or limit the product or community in various dimensions. Vendor dominated communities require more collaboration but also tend to cultivate more innovation and creativity.

Matt indicated that he saw "embedded open source" as a more scalable long term commercial strategy than open core. His definition of "embedded" was open source code that is embedded within a larger proprietary product. This is especially relevant for open source platforms or libraries, which are often embedded within other software applications. The benefit of the embedded model over open core is that a vendor is not focussed on the challenge of trying to upsell open source users into a slightly more professional product edition. The features and capabilities of the products are clearly differentiated, with the open source platform and proprietary application targeted at totally different audiences. A good example would be 37 Signals and how they offer Ruby on Rails as a widely popular open source platform for developers, but drive their business using BaseCamp, a project management application for business users ( where Ruby on Rails is embedded ).

An interesting thing happened as I read Matt's blog. I had an epiphany. With the rebranding of DNN in 2014, the company no longer needs to pursue the traditional open core business model. There is now a clear branding distinction between the open source DNN Platform and Evoq commercial solutions. Essentially the DNN Platform is now simply an embedded part of Evoq. The company no longer needs to sell the benefits of DNN as an open source platform; instead it can focus on selling commercially licensed solutions under the Evoq brand.

This creates an opportunity for other commercial "distributions" as well - where DNN Platform is embedded. This is actually not a new concept, as companies have been embedding the DNN Platform into other commercial products for years ( it was originally conceived as a web application framework afterall ). The difference is that DNN Corp no longer needs to be the sole project steward for the DNN Platform, the responsibility can be shared with other vendors who are also embedding DNN within their applications. 

This actually opens the door for the DNN ecosystem to evolve into what Matt referred to as a "vendor dominated community". Similar to Eclipse or Linux or Apache, multiple vendors within the ecosystem could step up with resources to invest in the DNN Platform. The responsibility would be shared and DNN Corp could participate like any other vendor. The benefit would be access to a larger pool of resources, greater and more diverse investments in product capabilities and innovation, greater confidence in the long term viability of the technology, and more collaborative involvement from the ecosystem.

Obviously for this concept to evolve further, it will require additional effort - for example many aspects of the current infrastructure would need to be moved to a shared location. The good news is that some of this has already happened and the pendulum seems to be swinging further in this general direction.

What are your thoughts?

3 comment(s) so far...

The DNN journey has been a great learning experience for many of us. And for that alone I'm eternally grateful. Sustainable open source is incredibly hard as you point out. And what we began with the community with "DNN Connect" is beginning to take shape as we all search for this "vendor dominated community"-future that you allude to. It is thrilling to see the community re-engage and to witness this emerge.

My first impression is that you are indeed pointing to the right direction. I myself was very much into-DNN by 2007/2008 and then, a few years later... when I saw that the community was being sort-of disregarded, I drifted away, focusing on other things. It was mainly because of the commercial overtake that -as I saw it- was taking place, and the consequences of that, the urge to deliver releases that were not as strong tested as they should have been, the disappearance of several long-lived and high participating members... It was discouraging. It is with great pleasure that I read your words in this post. There have been members who were very critic along the path, and now I would like them to come and leave an opinion. I applaud your move and wish you godspeed.

I've been yelling this stuff at everyone for years... the DNN open source opportunity is everyone's opportunity to build what DNN Corp built; a company supporting DNN targeting a certain need (I laud the guys at Arrow for creating a proprietary ecommerce solution (HotCakes) and supporting it). Every time I hear DNN described as a CMS I wince... because it is a web application framework (I was using DNN for years as such before I moved away from program development (intranets) to focus on the marketing potential inherent within public facing DNN). DNN Corp IS just another DNN vendor, but it had a massive marketing advantage (which created angst within the hard-working community), and head-hunted within the community... but the fact they are the stewards of the open source project is still blurring the lines (don't get me wrong, I think the community needs leadership) in that there cannot be one organization with two visions... that's called double-vision. ack! My hope is that the quality of the open source application framework will continue to improve allowing hundreds/thousands of people the opportunity to see DNN as a vehicle by which they can improve the quality of their lives. (Among those people will be leaders who will hopefully be given the opportunity to improve the core platform).

Shaun Walker has 25+ years professional experience in architecting and implementing enterprise software solutions for private and public organizations. Shaun is the original creator of Oqtane and DotNetNuke, web application frameworks which have cultivated the largest and most successful Open Source community projects native to the Microsoft platform. He was one of the original founders of DNN Corp, a commercial software company providing products, services, and technical support for DotNetNuke, which raised 3 rounds of venture capital from top tier Silicon Valley investors. Based on his significant community contributions he has been recognized as a Microsoft Most Valuable Professional (MVP) as well as an ASPInsider for over 10 consecutive years. He was recognized by Business In Vancouver as a leading entrepreneur in their Forty Under 40 business awards, was a founding member of the Board of Directors of the Outercurve Foundation, and is currently the Chair of the Project Committee for Microsoft's .NET Foundation. Shaun is currently a Technical Director and Enterprise Guildmaster at Cognizant Softvision.

1baiser.comporno