Mike Olson of Oracle (previously the CEO for Sleepycat) just posted a blog about his keynote at OSCON. He argues that there is value in collaboration and community, but that it doesn't necessarily require open source code. He mentions VMWare and newer community-based web sites as examples where the power of collaboration is tapped without requiring open source code.
I think if you look at open source as only about collaboration, then yes, you can draw these kinds of conclusions, and you may think, hey, maybe having open source code is not that important.
But as I have mentioned before, I think this misses a crucial point about open source. Open source is not just about collaboration. That may be cool for open source developers, but for users of open source, it's much more than that. It's about freedom. It's about not placing heavy dependencies on a single vendor. Open source and an open community gives you the assurance that the technology you are depending on is not going to be discontinued or put into "maintenance mode," it won't be acquired by someone who you would rather not do business with, and it won't be used as leverage against you to extract money or modify your behavior.
As a representative of Oracle at OSCON, I'm not surprised by Mike taking this perspective. I think we can all agree that open source databases are a threat, to some degree, to Oracle's key source of income. If I were at Oracle, I would be paying close attention to open source databases, and I would be trying to find ways to reduce the panache of open source without coming right out and saying "we hate open source and wish it would go away," which really wouldn't go over well. It's similar to how Sun behaved towards Linux a while back.
That said, MySQL, as the most popular open source database, is I believe less of a threat to Oracle than the other ones. Why? Because it does not have an open community -- it is controlled by a single vendor. This means it's an even playing field. Oracle can compete with MySQL by showing they have better features, or by offering a free version (which they have done), or even, if necessary, by acquiring MySQL or technologies it depends on (which they have done (twice)).
But this tack won't work so well with open source databases that are built in an independent open community. PostgreSQL, by being an open community, provides the additional value of freedom from the tyranny of a single vendor. And you can't "buy" the company that makes PostgreSQL, as there is no company to buy. There's nothing Oracle can do to respond to this, barring opening up their source code to an independent community (shyea, right).
So if PostgreSQL can provide competetive, "good enough" database features and performance (and things are looking good in that arena), and if you can even get support for PostgreSQL, then if I were at Oracle I would start getting a little worried, and just hope that not too many people notice that "other" open source database over there in the corner.
 Note that's what Sun did in response to Linux.