Who really commits code to open source?

This week I’ve been challenged with the task of producing the annual software industry salary survey for a large US publisher known for its weekly title themed around the word Information.

Although this data is US based, the 4,581 staffers and managers who responded to the questionnaire evidence what I would describe as fairly global trends. In the open source area, I have noted the following:

– after yet another year of glowing press reports detailing the additional flexibility and enhancements that arise from the community approach to development, open source software usage has climbed from 27% to 32% in 2011 — a trend that will no doubt continue over the next 18 months at least.

NB: that’s the first time I have ever quoted myself, shame on me

Linux Foundation.png

OK so serious point, this led me to start thinking about who really commits code that finds its way into open source projects. When I was at the Nokia Qt Developer Days conference in Munich last autumn I got speaking to the guys who work for The Linux Foundation itself and they were a little woolly in some areas.

Let me explain.

Without quoting anyone directly, it does appear that the large percentage of committed code appears to come from customers who take up commercially licenced versions of open sourced project code. This may be because these customers have to work with statically linked (as opposed to dynamically linked) libraries for the code they use, due to security, compliance and mission critical issues — and therefore commit a different “kind” of code contribution when they do.

I do not want to suggest for one second that this is always the case; this is simply what I heard at one event, from one set of (possibly ill informed) spokespeople.

This is an interesting subject to analyse surely? Or should I be committed?