This is just a general comment, but it does involve MMS. This is also a personal statement, not something necessarily company-endorsed.

Shipping a working mobile OS is already a lot of work. The more features you add, the more work is involved - but the amount of people you have to split that work in our case is quite constant (and the lesson enshrined by the "mythical man month" is that adding a bunch more people won't necessarily solve that problem any more quickly - it may make the situation worse, even).

So you have three essential parameters you can tweak: features (+/-), quality (+/-), schedule (+/-). Pick any two.

We chose to ship something that we hope is of an acceptable quality on our target schedule, and look at adding more features through software updates. Hopefully when we look back at this in 6 months, that'll seem like a good choice.