You keep saying things like that, but it's really hard to develop for a device which may or may not have a physical keyboard, may or may not have a GPS, may or may not have a phone, has a screen or unknown dimensions, has an unknown amount of memory, has an unknown arrangement of hardware buttons, has some as-yet-undeciphered second type of keyboard input on one variant, may or may not have its apps distributable through Nokia's app store, may or may not be able to drive an external screen, might but probably won't have a capacitive screen, might but probably won't support multi-touch, may or may not support haptics beyond the simple "vibrate" that we know about, may or may not have a transflective screen, has unknown battery life, may or may not have a hardware stylus, may or may not fit in a jeans pocket, etc.

Success is not just a matter of developing against an SDK and hoping that the device will be well-suited to the app.

If Nokia actually wants to reap the benefit of the open source model, they need to be open themselves.

Someone once said that "there's not much community around S60, except for the community that Nokia pays to be there". That's also true of Maemo to some extent. Nokia is enlightened when it comes to subsidising conference attendances, funding a community debmaster, etc etc. But that's an expensive way to tap into open source when Nokia could reap huge benefits simply by being ... open!

Regards,
Roger