Honestly, I think the optimal path is somewhere in between.

I doubt you're ever going to see a huge developer presence on Maemo. Sure, the Linux community might (_might_) grab on to it once Nokia _really_ gets behind it. But it's Linux. The mainstream consumer apps developer community isn't likely to make that leap.

Maemo needs 3 things that Android has:

1) a mainstream programming environment -- a Java based makes for an easy learning curve for the huge base of Java programmer who consider becoming an Android developer. It doesn't have to be Java, but it has to be a common language with a common (more common than Linux, Gnome, and probably Qt) API.
2) an open and optional App Market (as opposed to no central distribution channel for 3rd party devs that older Maemos have had, or the closed and mandatory distribution channel that Apple has)
3) the HUGE marketing push and effort that Google, HTC, T-Mobile, etc. have put into Android (or that Apple has put into the iPhone).

And, it wouldn't hurt to have a huge base of devices that it runs on as well. Apple hasn't needed that, but it's helping Android quite a bit, IMO.

#3 is probably the biggest key. But #1 and #2 would be a huge benefit to Maemo, IMO.

I'm not sure that Android _needs_ (for its intended goal/audience) anything from Maemo ... but I'd like to see it have a sort of "pro" version that gives you a more typical Linux experience at the command line, apt-get/apt-cache, perl, and not need to be "rooted" for some of the more interesting capabilities, and things like that. All things that you get with Maemo.

That's why my first choice is a hybrid of Maemo at the base, and Android on top.