So, it's that time of year again...
Since many of you have asked already, I'll now reveal the proposals that I've submitted for this year's SOC:
1) Animation System Polish
2) Next-Generation Animation Tools
Personally, I'm a bit more interested in #2 than #1 (if it comes down to "sexiness factor"), though #1 is some fairly important work that needs doing too, as recent bug reports would make you believe...
What about the Bullet RigidBody work I was doing last year?
Well, in the immediate term, not much from me. It shouldn't be too hard for someone dedicated enough to easy take the "rigidbody" wrapper module I set up, and hook it up again to some of the changed parts of the Blender sourcecode (particularly the update loop changed since the version the branch was forked from). Key functionality still lacking is support for RigidBody constraints, and/or a few extra templates and editing tools to make it easier to edit rigid body behaviours.
One main reason why I haven't resumed this work thus far has been that I've been busy trying to perfect a workflow+tools to make SVN branch management in future easier so that I don't run into some of the merge-itis that I had with this branch, which meant that I didn't do any updates to this (IIRC, there were also heavy waves of frequent and large-scale changes at the time too, which were not productive to working in branches).
That work is now by and large complete, though I'm a bit more interested in working on the projects I've mentioned above than on physics sims for the near future. This is not to say that I won't return to work on this stuff later (i.e. perhaps if it were to be considered important for a future release in the 2.6 series for example, or when I've wrapped up the key work for these other projects first).