While reading Haacked today, a question I’ve had for a long time finally boiled over to the point of writing this post. 
Phil commented on this post which puts up a good argument for why Microsoft should financially support Open Source projects. Like Dave mentioned in the original post, I too have used DNN for implementing customer sites. Without
it, development costs would have been too high for at least one project and I most likely would have lost the gig. I’m also one of the many users of NUnit, NDoc, Nant, log4net and Subtext. 
Ok now for my question: What drives people to write open source software? What drives a person to write something as good as NUnit or DNN and then give it away? What drives someone to create something like NDoc that’s used by
jillions of developers, give it away, and then close shop because no donations were given? Why not charge from the outset?
Is it because by charging people, you’ve taken on more responsibilty when it comes to support? Is it because you don’t think people would pay for your product? Someone please enlighten me because I truly am curious.
I can’t speak for everyone, but many projects try to take advantage of the power in numbers.
I think Subversion is an example of this. It was developed as a better source control. The company needed this tool, but realized they could develop it faster by open sourcing it.
As for me personally, I find it fulfilling to hack on Subtext and to lead a community of like minded developers.
Why do people form running groups, or biking groups, or any kind of group to share an interest? I don’t see participating on Subtext as much different. It’s a group of people doing something they enjoy.
Geeks are pretty unique in that they often do for a job the same thing they do for leisure. Writing code is fun, especially when you can choose exactly what code you want to write. Turning that product you made for fun into a commercial venture is not that much fun. It’s long hours, low pay and high risk with the (small) chance of a big payoff at the end.
Craig,
I agree that writing code can be fun, especially when it’s something you want to write. I guess I still have a tough time with the idea that turning a product into a commercial venture == long hours, low pay, etc.
I guess I don’t see how turning a product into a commercial venture is any tougher than producing an open source product and asking for donations.
Why do so many open source projects ask for donations? Wouldn’t it be better to take the risk of turning it into a commercial venture from the outset?
Phil,
You said:
“Why do people form running groups, or biking groups, or any kind of group to share an interest? I don’t see participating on Subtext as much different. It’s a group of people doing something they enjoy.”
Seems like apples and oranges to me. If I’m running in a group, I’m not asking for donations for new shoes, right?
I feel a follow-up post coming soon.
I guess that’s the problem with analogies, you look hard enough for differences and they all end up being apples to oranges.
Let me refine it. I used to do a big-brother like mentoring program. Found it fun and fulfilling. But we also had costs, so we asked for donations.
Another (albeit odd) analogy is if I started a museum with friends and we wanted the most number of people to enjoy it. We have costs so for those who can afford to, we’d request donations. It’s a public service and we have costs.
As for turning it into a commercial venture, not every open source project is necessarily commercially viable. Since many are community projects, they serve a small niche audience.
So founders often want to reach the most users they can to attract a larger development team than they would otherwise be able to afford as a company.
This doesn’t mean I’m not mulling options on ways to build a commercial venture around open source software (my consulting firm implements DNN for example). Just that it’s not so simple all the time.
Phil,
You said “Let me refine it. I used to do a big-brother like mentoring program. Found it fun and fulfilling. But we also had costs, so we asked for donations.”
That is a much better analogy.
I do see your point about some projects not being commercially viable due to the relatively small size of their user base, but what about all the open source development tools out there (that aren’t based on existing open source projects)? I’d say the audience for those is pretty large.
As I pointed out with Subversion, some of these projects are founded by people who need these tools for their primary business (the tools themselves not being the primary business).
They figure rather than building out an expensive team to develop the tool, they can take advantage of free labor while also providing something to the community by simply open sourcing it. That’s one motivation I’ve heard.
As for others, it’d be interesting to get their perspective as I’m not an open source tool developer.
However I do occasionally contribute to tools such as log4net because I may run into a problem in using it for my needs. I love being able to open up the code, seeing the exact issue, and submitting a patch. Try that with a commercial tool. In part, I think that explains one of the many motivations for a *contributor*.
But for the founders of log4net (and log4j its parent in effect), I’d love to hear what their thoughts are.
Phil,
Thanks for taking the time to reply again. I appreciate your feedback. You’ve given me some things to think about in regards to open source software.
I’m working on a follow-up post that I hope to have up today or tomorrow.