I'm not sure that Nokia's phones have been the raging success with either the average joe or otherwise, the last time I checked. Can you check the numbers? I thought that Nokia lost marketshare in the period since the internet tablets were around--so it's probably not the just tablets' fault.
Near as I can tell, this whole new market that Nokia just about created and led (the Internet Tablet) is something for which they could have continued to pioneer--and it appears to be the trend you're seeing with several competitors that have popped up recently (Pandora, Archos 5 Internet Tablet, ODROID, etc.).
Instead, they've relegated the N900 to another iPhone wannabe. That'll be a raging success in the "mass market" alright. :P Treat it like the openly expandable, portable general computing device that it should be and it'll do better than the iPhone wannabe that it seems poised to be.
I also don't buy the soldered battery argument. NOTHING excuses a soldered-in battery.. not size, not weight, not anything. Cell-phone batteries are thin enough and last well enough not to use that sorry excuse to charge people money to swap out a battery and make sure there's no third party market or competition.
Seems to me a company with Nokia's size and experience should have the resources and the intelligence to be able to make a small module that could be used across many devices to support a carrier.
I remember hanging around people in Silicon Valley, back when I lived in Santa Clara in the late 90's and early 2000's, that used to build their own cell phones. I'm not sure if these are useful for your interests: http://www.opencircuits.com/Open_Mob...ts#GSM_modules
You might even want to take a look around the whole wiki for interesting project resources and information.