Joyent's David Young has written an excellent treatise on why "clouds" (as in cloud computing) should be open and not proprietary. He details nine attributes of an open, "platform as a service" cloud.
My favorite? Young's contention that while an open cloud could lead to everyone "rolling" their own, the rationale behind doing so is, well, not so rational:
If you're writing an application, and you want to be able to achieve tremendous scale, the answer shouldn't be to move off the cloud onto your own "private" cloud of dedicated servers. Of course, if the cloud computer is open, as we've described, you can build your own cloud. It's also true (that) you can generate your own electricity from coal, if you want to bother. But why bother?
This is a fundamental tenet of open-source businesses. There's much that you could do to fork an open-source project and create your own splinter project, but generally, it's not worth the bother.
The Slashdot community piled on to comment on Young's post, with some insightful questions as to the viability of uploading a company's "crown jewels" to the cloud, "for all the world to see." Others suggest that "'cloud computing' is just the latest marketing promotion designed to move us to renting software."
In the wake of all this chatter, many have missed the funding of open-source cloud-computing vendor 10gen. OStatic has a good write-up on the company. Commenting on its revenue model ("Despite the fact that 10gen's software stack is free and open source, hosting and customer service are fee-based"), OStatic's Sam Dean suggests that this may well prove to be the business model of choice for cloud services: make it easy (and free) to develop, but charge for services rendered.
Regardless, I concur with Young that too much is at stake in cloud computing to leave it to proprietary clouds. Why build oneself into a corner? For cloud computing to succeed, it needs to be transparent in its ways and means, so that people won't have to worry about what will happen with their data.