Thanks for taking the time to respond, Siem.

I’ll say it again: I benefit from lots and lots of OSS projects, and the model makes lots of sense in lots of contexts. It’s definitely the right choice for tools that help machines talk to each other, like OIPA. I think it’s the wrong choice when you want to make software that “civilians” — dentists, homebuilders, foreign aid field workers — will be using in their day-to-day work.

The point of my article was not that we’d be better off without open source software. It was that the current ICT4D orthodoxy — this knee-jerk assumption that OSS is inherently more virtuous and more appropriate for this field — is incorrect and counterproductive. Why do we think it’s more virtuous to sell consulting hours around software, or add-ons to software, or advertising powered by software, than just to sell software outright?

These other business models are fine, but they’re complicated and they obfuscate how the software is actually being paid for. My deal has the advantage of being super simple and transparent: I make software, and you can pay to use it if you want. If people want to use my software, I can pay my mortgage. If not, I have to find a different line of work. That’s a powerful incentive for me to care about what users want and need (as opposed to what a donor might be persuaded to fund).