I've been thinking about open source and how it fits into our wider economy recently, given the big push behind it, and also trying to understand the practicality of the fanboys calling for everything to be open-sourced. I'm focusing on the home user market, because that's where my passion lies and where open-source hasn't been that successful.
There are many examples of open-source working in the enterprise environment, with Novell, Red Hat, MySQL and a range of other companies turning a profit. These companies typically make money by offering services to complement the product - e.g. installation and configuration, customisation, integration, training, technical support (especially top-priority level) and value-added services like software maintenance and hosted services like storage. Other companies like Network Box, make money by integrating the open-sourced software into hardware, and selling that. Another strategy is to license the software under an open-source licence, then sell commercial licensing for those who don't like the open-source licence, e.g. Trolltech's Qt. It doesn't always work out though, some saying open-source doesn't work as a business model.
But things are really different when it comes to home users. The whole 'services to complement software' doesn't really work there because that has never existed for home users. And to be honest, I doubt it would work anyway.
If I download some open-sourced software, I don't want to wait for a guy to come round to install it for me (and if that's needed, then there's something seriously wrong with the setup process). There may be some market for customisation, but again, unlikely - if it can't do it out of the 'box', it can't do it - is generally the mindset.
And really, if technical support is needed by a home users, something has gone wrong. Home users have consistently told us they want intuitive products that they can pick up without effort (i.e. without reading a 600 page instruction manual) - they're unlikely to buy into something that requires you to do so, and even worse, shell out money to work it out. And if something does go wrong, most of the time it's the program's fault, not the user, so they shouldn't have to pay to get the program fixed (plus this creates a bad incentive for the developers - the company makes more money when they produce inferior products).
The only services I can think of that would work are hosted ones - e.g. create an open-sourced tool to automatically backup their data on to your servers which the client pays for monthly. Problem with this is, its application is limited and the data gets locked up in a random server, out of the user's control. [sarcasm] I can't wait to have a different account on a hosted service for my mind maps, documents, videos, email, health records, financial records etc. [/sarcasm] That scenario can be reality now with the abundance of web apps, but the beauty of desktop apps is that I remain in complete control of my data and where it is, and it is very easy for me to manage.
Another way an open-sourced software company targeted at home users can make money is through hardware. And there are many that do it well - SlimDevices with the SqueezeBox and Synology with their range of NASes. The new product, BUG, from BUGlabs aims to take this one step further, by letting users program the hardware in an open-sourced environment. But this approach requires the software to be bundled with hardware so therefore its applications is limited as well (can't have a different device for everything).
Content is the final one I can think of. For example, a email newsletter designing app may offer templates designed by a company to purchase. It's one that hasn't been explored much in relation to desktop apps, and only used to a certain extent in web apps. One of the problems with this approach is that the web has broken down the idea of paying for content a lot, with vast amounts of useful legal and illegal content thanks to things like Creative Commons and peer-to-peer networks. It is often not very viable either as a revenue source, as it will probably be very unpredictable.
So where does this leave all the software companies who make software that's partly targeted at home users? Companies like Quicken, MYOB, Cyberlink, Microsoft, EA etc.? Or even companies that have traditionally created fully-featured standalone software but are not entirely home-user focused, like Microsoft Office, Adobe Photoshop and the entire Creative Suite? Unless they're crippled, if these were open-sourced heaps of people could benefit without giving a cent back because they're so powerful as it is.
The reason we have some open-sourced home-user software, things like Firefox and OpenOffice, is because they're all funded by big corporations like Sun and IBM as it complements their products for enterprises. Others are contributed to by developers in their spare time, while they work 9-5 in a related industry, most likely the enterprise software industry. Some exceptions aside, e.g. Firefox which makes millions from bundling Google's toolbar, they are all not for profit, and would be unlikely to make any anyway.
Some have been experimenting with using the web app strategy instead combined with either advertising or a 'freemium' model, i.e. free, but pay for more features. Others are experimenting with hosted software and subscription-based software. But how many of these have succeeded? Are we willing to pay by the month for something we traditionally bought outright, and had complete control over how and when it was used? Or are we willing to relinquish control over our content in return for free usage of a program (e.g. some of the web-based video editors out there, like Jumpcut)? And to be honest, the advertising model sucks. I would hate it if my word processor was showing me ads all day long.
The open-source revolution has given us a lot of very, very cool stuff, and we wouldn't be where we are without it. But it isn't a one-size-fits-all answer. If all software were made open-sourced, who will support the people coding them? Are we all going to be at the mercy of enterprise companies funding these projects? Not all software used by home-users have an equivalent enterprise usage, e.g. media centre software. Programming may have become very cheap in China and India, but even so, people need money, not only to support themselves, but as an incentive to think, work and produce better and better products. The enterprise market may have worked out a way to make ends meet with open-source, but attempts at the home-user market have failed so far, and if it is going to work in a significant way, a radically different approach is needed combined with the right education and marketing, something most open-source projects aren't very good at.
P.S. I find it this whole services revolution quite ironic. On the one hand, we're trumpeting the services model for the web and software in general, giving these companies our money to lock us and our data into their services. Yet on the other, we're shouting abuse at the music and video companies for locking their content with DRM, restricting our rights after we purchase it. WTF?
UPDATE (18/09/2007): Added in a bit about commercial relicensing of enterprise software.