I didn't make it to OSCON this year, so I missed out on more than a few nifty events. One was a panel chaired by Matt Asay of Alfresco, where he cited research to show that companies do switch to open source as a way to save money, but that there are other, much larger goals beyond that.
Matt's discussion of this issue hinges on two things: 1) it costs nothing upfront (well, almost nothing -- let's say a negligible amount) to acquire the software and determine if it's a proper fit for one's business, and 2) the customer can spend that much less all the way down the line to get precisely what they need. And then there are other things that materialize further on down the line -- like how maintenance and support costs break down and where the money goes for that.
What's most interesting about this particular analysis: they based it on real-world research from companies actually doing this stuff, not speculatively-crunched numbers. They went to the people who had actually deployed open source in-house, rather than the vendors themselves. They're the best ones to talk to, since they've actually burned the oil to make things work.
The real goals beyond saving money are things I've seen in other realms, as when I was talking to the FAA earlier this week: avoiding dependence on any one vendor, or having the freedom to roll your own solutions from existing ones. That goes back into something Matt mentioned: "Open source tends to offer best-of-breed solutions that aim to do a limited range of functions well, rather than to be all things to all people." If you know you only want a few things done right, you take those pieces and use them.
This isn't to say that there won't be people who can't thrive on a totally packaged solution. This also isn't to say that someday proprietary software vendors will offer things that are just as flexible in their own way; perhaps some already do. But when you're dealing with business processes, rather than programming-in-the-abstract, the goal isn't "open source". It's lower cost, more flexible ways of getting things done, that whole bag of goodies.
Those should be the real goals of any upscale open source project: to figure out how to offer people ways to do things more flexibly, and to make the whole issue of open source a peripheral one. It won't matter how it's put together, only that it's put together better.
InformationWeek Analytics has published an independent analysis of the next-generation Web applications. Download the report here (registration required).