As a designer/developer for many years, here's what I've noticed: if a project isn't originated and run in a well-defined, well-managed environment with common goals/incentives it's very likely to flounder. The best incentive is, of course, money.

A good example for me is the gaming mapmaking/modding community. Most of the custom environment map models I've seen for games are crap (specifically the custom edition of Halo for the PC). Too often you get several immature hacks with possibly great creative skills but poor management and teaming skills "collaborating" on a disaster. I see easily 25 poor to mediocre game maps for every good one. As an experiment, I tried a true open source approach to a game map, even posting aspects of my project that most others kept secret, but discovered very quickly that the approach was doomed; one common cause (in lieu of money) that keeps a map team on target and producing a good map is secrecy and the ability to bestow testing rights on a chosen few, creating a cult of sorts. My true open source projecy was supposed to be based more on altruistic goals and apparently that takes the type of team members I was unable to assemble.

That's one example of open source hell, but it applies to every discipline.

That is not to say, however, that all OS projects fail. As I said, the successes are just few and far between in my own experience. Specific to the tablet developer community, there are some earnest hack jobs and some true gems along with various other flavors, and I remain optimistic that Nokia's guidance and firm development foundation will *eventually* bear more fruit. IMO we just need more solid developer teams and fewer "cowboy" efforts... no offense to the cowboys.