Free software and good user interfaces should be required reading in all Computer Science 101 courses. In fact, probably in Engineering 101. I'm not going to pick that article to pieces, but I am going to elaborate on one key idea that it, and all discussions of free software projects, seem to completely miss.
The biggest undocumented advantage is that Someone Can Say No. What I mean by this is the following: in commercial development, generally, marketing or a specific customer drives the feature list. Engineering is generally behind the 8-ball if it wants to try to state "this can't be done", and even more so if it wants to state "this shouldn't be done". At some point, the engineers will be told that they simply will do it that way if they would like to continue being paid ...
The "shouldn't be done" argument is especially difficult to make because it is so hard to quantify. Intuition about what makes "good design" is the most under-appreciated of all engineering skills, and trying to communicate this to someone who simply wants Feature X is very difficult. (The above citation is the clearest explanation of this I've ever seen in print).
But what the author misses is that in the non-paid, free software world, there is no such backstop. If (to use the most famous example) Linus Torvalds thinks your latest frob to the Linux kernel is junk, he'll simply reject it. This irritates many people who don't understand that not only is this behavior not a bug, it's a feature. In fact, I'll go out on a limb and say it's the main feature behind Linus' sucess, and even that of most other sucessful free software projects. Why? Because there is someone in the position to say "this is outside the scope of what this piece of software ought to do" or "this implementation is terrible" or "this idea is a Wrong Thing in the first place". A lot of suggested cruft winds up in /dev/null this way, and that's exactly as it should be.
So the next time you're reading one of these development mailing lists, and you start to have a knee-jerk reaction to someone saying "no, this is a brain-damaged idea, don't do it", think twice about whether it's worth listening to that objection, and keep this context in mind.