Actually, that's not true - I've produced plenty of open source software over the years. However, in a sense, it is true: only the very best actually get paid to work on open source software full time, and I'm not one of them. People like Linus Torvalds. People like Guido van Rossum, although even he supposedly divides his time, and does not work on Python full-time.
Think about that. Python is a hugely popular programming language used by many companies and individuals who get a lot of value out of it. But the creator doesn't even work on it full-time. Now, that's just one example - perhaps Guido enjoys the things that Google sends his way to work on outside of Python in any event - but I think it's representative of open source in general.
Back to me: I've produced some small bits of open source code that other people find useful. Several people have even built products on Hecl that make money. But I'm not good enough to work on open source full time - I'm not one of those famous, brilliant coders who is so good that someone will find a way to pay them to work on stuff that gets given away for free. I am, however, a good enough programmer to work on people's proprietary code, and have never had too much trouble finding someone with a project they're happy to pay me for. Why is that? Because it's so much easier to funnel money back into a proprietary project. If people like the product and buy it, the company gets money, which can be used to pay the developers. With open source, millions of people might use it and get a lot of value from it, but the developer has no right to receive any of that back as cash, which he or she can use to pay for things like food and rent.
So, I can code tolerably well, and I could conceivably contribute more open source code, but instead I spend my time working on proprietary code because it pays the bills. Clearly, when I can, I use open source software in these endeavors, and contribute back whenever I can, but the "secret sauce" remains closed. That's a pair of hands lost to the creation of more open source.
I know I'm not alone in this, either - tons of people work on mostly proprietary projects the world over, but relatively few people get paid to work on open source code full time.
So when I read about people debating the utility of copyright bring up the existence of open source as some sort of counterexample, it irritates me a bit. The right level of protection and enforcement of copyright is a complex debate that I'm not going to get into here. What I want to point out is "that which is not seen". Sure, open source exists. But how much more of it would exist it there were more money to fund work on it? How much open source software remains an idea in the developers head that does not get realized for lack of time? People often criticize the "Linux desktop" despite its extraordinary strides in recent years. Well, how much farther along would it be if there were more people paid to work on the 'boring stuff', like usability testing? Ubuntu and Redhat pay a few people to do that kind of stuff, but how many more people do Microsoft and Apple have for that kind of work?
That's not to say that open source "doesn't work" or some such nonsense. It obviously works quite well, but it really shines where the currency is code, not money. Developers can and do give back lots, in terms of code, bug reports, suggestions, documentation, and so on to open source projects, which make them better for all involved. Where open source doesn't seem to work quite as well are in small, fast-moving, consumer-oriented products. My guess is that 99% of iPhone users could care less about the source code for their apps, but on the other hand, a large portion of the Emacs user base more than likely has written at least a few lines of Elisp.
In any event, the point isn't to beat up on open source software, but to counter this idea that "intellectual property" is in no way shape or form necessary because the existence of open source software somehow "proves" that "things will get made just the same". Yes, maybe they will, but in lower quantities than consumers might find desirable. After all, most of us aren't good enough to work on open source software.