Recently I got to talking open source with a friend of mine who has worked in the proprietary database market, especially at Sybase, for several years. He made a few interesting observations during the conversation. For example, we talked about open source competition in databases, and he said a few good things about MySQL, but also said that the proprietary databases often outdo MySQL based on just a few small features, and also commitment to legacy systems that many companies have. His take on what Oracle might do with MySQL is that it might treat it as an open source on-ramp to its own proprietary databases, but that Oracle will have to keep MySQL like WordPad --Microsoft's text editing application, which falls far short of the features found in Microsoft Word.
When we got down to discussing business models for successful commercial open source companies, my friend especially liked the Red Hat model of providing support and services for free software, which has provided the company with substantial financial success, and also consistency. There are quite a few companies following the Red Hat model, and there are also signs appearing that that model will end up immeasurably improving important open source software applications and platforms--not just business bottom lines.