by Mark Stone on March 25, 2009 10:00am
On Monday Microsoft and SD Forum held the 3rd annual Open Source ISV Forum. In a day of interesting talks, I was particularly struck by Larry Augustin's talk. As an emphasizing example of the growth of open source projects partnered with a commercial endeavor, Larry mentioned DotNetNuke.
DotNetNuke is, of course, freely available for download and licensed under an open source license. But there is also a professional edition and a range of complimentary commercial services for those who want service, support, or customizations. The business model is classic open source: the free download seeds the market with potential customers, and as some of those run up against the limits of what they are willing to do on their own, they make inquiries about the professional edition. Thus open source creates an inbound channel of qualified sales leads, without the overhead and expense of a sales force working in the field.
This is a great business model, but it's important to think about the actual decision-makers in this adoption process. What's distinctive about DotNetNuke is that it's one of the few mature, open source Content Management Systems (CMSs) that is not written in PHP. It is, as the name implies, ASP.NET based. And this reminds me of a prdocut management position I was in some years ago when (a) the only mature CMS choices were PHP-based, and (b) PHP on Windows was not yet a viable alternative.
The company in question was in a typical position: they were not a technology company, but needed a strong web presence for their business and to connect to the online community of their customers. They had a home-grown CMS solution that wasn't scaling, wasn't secure, and wasn't stable. My product management team put together a good comparison chart of various CMS choices, many of them open source. I sat down with the web development manager to review the choices.
"Some of these are good systems," he said, "but I have a team of .NET developers, and these are all PHP-based. I don't have head count to go out and hire a PHP dev for this project."
Note the decision-making process at work here:
- From an agreed list of candidate software, an engineering team will download something to "test drive";
- One developer will do some testing and make a recommendation;
- From a short list of recommended software a more thorough test will be done with a prototype or pilot project;
- Finally a choice will be made, and money will be spent.
The person with the first vote in the process is not a CIO or any other traditional "IT Decision Maker". It's one dev, talking to his manager. If your software gets vetoed at that level, or -- worse -- never even gets a try-out, then your product isn't in the decision-making process.
A lot has changed in recent years. PHP is now much better supported on Windows, and .NET projects like DotNetNuke are available and much more mature. This is as it should be. For open source to spread to its full potential, it has to be available in the technology adoption decision-making process. And that availability starts with being available to the hidden technology decision-maker: that lone developer who will look at software written in a framework they understand for a platform they work with.