Thinking recently about Open Source, I realized that I really didn't understand the drivers that made it such a popular 'thing' in recent years. So I did a little digging and read about Open Source licensing (Andrew M. St. Laurent book and Lawrence Rosen book - content available free [as in "free speech", and as in "free beer" I guess, too ;) ]) and economics to get a better idea. What I found is that Open Source as competition to proprietary software creation seems to be an inappropriately applied concept, as the two approaches target different audiences for different reasons and represent different models for Intellectual Property investment.

So, the first thing that i did not fully understand is that that Open Source is simply a model for licensing Intellectual Property. The definition of Open Source specifies that a piece of software can be freely re-distributed without royalties or other fees, source code is made available, original software may be modified or new works derived from the original, and a few other tenets (see the complete definition). Open source does not automatically equal "free software".

Open Source's primary economic driver is to reduce cost of commodity software development/consumption. Commodity is being defined in general as anything that doesn't bring market differentiation to the business, is available in equal manner to all businesses, but is required for effective business operations. Email, for example, or office productivity suites could be viewed as "commodity"; while a proprietary search ranking algorithm (a la Google's PageRank) commodity considered would not be (Yoda-speak).

Theoretically, the concept of "commodity software" is crucial to Open Source. A company has a lot of interest in sharing cost of developing commodity software alongside its peers or competitors. If no one gains competitive edge from the use of the commodity software, everyone involved in developing it benefits: each company spends much less than they otherwise would have had they developed it on their own; and on top of it all there's an active developer community that is focused on creating features that are specifically asked for, rather than trying to develop a mass-marketed piece of software with 80% of the features that may not be used by more than 10% of the users.

While the definition is easy to understand, I think commodity is easy to define but may be difficult to find in the real world. There aren't many pieces of software that aren't used to drive some efficiency in the business. Efficiency typically produces competitive advantage, so in theory any software (when properly implemented and scoped) would add to a company's competitive advantage. If it doesn't - why would you bother having it in the first place?

It appears that Open Source software is primarily important to hardware manufacturers who want to share the costs of making their hardware useful. Look at Sun, for example, or IBM: by making the Operating System that powers their hardware "Open Source", they can use the global development community to make their boxes more valuable! Makes perfect sense to me...

Unfortunately, Operating Systems are not, despite Sun's and IBM's great efforts to convince everyone of the contrary, commodity software. A great deal of Intellectual Property that makes an OS useful goes into its creation. This IP comes from years of research and investment. Taking this IP and making it, essenially, free, can only happen if you're able to subsidize it somehow (say, from your successful consulting practice, or hardware sales). I think support of Open Source is the greatest "play" by IBM & Sun could have made to further their hardware goals and, at the same time (which is especially relevant in Sun's case) undermine other Operating System manufacturers while furthering their own hardware/software/application ecosystem.

So, without getting into too much depth here, my current understanding of Open Source is that it makes sense as a development license in some narrow cases where Intellectual Property you're investing will bring you no business advantage. In non-business situations such as scentific inquiry, academic research, hobbying, etc. - it's a great way of lending your brain-share to a project. Economic prospects of Open Source are still up in the air, but in my not-quite-fully-educated opinion, in the long term, Open Source has questionable value to any business that uses technology as an efficiency-driving tool. The "free" aspect is great for cash-strapped college students, but everyone should equally realize that intellectual property requires investment, and any worthwhile investment should have a cost.

I can still change my mind - do you have any info that will help me do so?