I head an open source project that provides bilingual semantic analysis of Chinese (http://www.adsotrans.com http://www.adsotrans.com). The commercial opportuni

I head an open source project that provides bilingual semantic analysis of Chinese (http://www.adsotrans.com http://www.adsotrans.com). The commercial opportunities we get come from having the best and most flexible software in the market. (If anyone wants to launch a semantic search or advertising network in China, I'm your man). Being open source is irrelevant. There have been three advantages to being open source in my experience: (1) it can help potential clients find you and simplifies the sales process since everyone is already crystal clear what you provide, (2) it can attract users to contribute things that aren't software, but that the software helps you generate (content, data), or (3) you want other people to contribute to the actual code base. Neither #1 nor #3 are good enough reasons to go open source in my opinion if your goal is making money. The only advantage you'll have over a closed source company comes with #2 - in which case you aren't a software company at heart. All of these rules go out the window if you're working in a really, really niche area, but then your business is really consulting. My advice would be that if you REALLY want to go open source for some reason, use a license that permits others to use and develop your software, but doesn't permit them to redistribute it. This will prevent others from forking the code commercially, but won't deter hobbyists and/or developers from using it to build systems themselves. Restrictions on redistribution also give you the freedom to change the license in the future when you figure out what you should be doing while protecting early adopters from having the rug pulled out from under them. At that point they can either continue developing the software themselves, or start paying for it.