Thanks BB. Great link.
To answer my questions:
1)
Similar wells seem to produce roughly 40,000 barrels/day (+/- 2-5k) at peak production. I haven't really come across a coherent discussion of productive oil flow vs. hypothetical absolutely unchecked oil flow, but it is pretty clear that 70,000 b/d would be an extremely productive well.
2) via xoB, "50 to 100 million." (It's somewhere in the 400+ comments to the post linked above; great reading, kind of hits the full spectrum of the arguments.)
As an aside, one significant complicating factor that is discussed in threads on The Oil Drum attempting to estimate the size of the spill, but that I haven't really come across anywhere else (MSM, here, etc), is that the well is spewing a mixture of both natural gas and crude oil. If you're strictly concerned with an oil slick coming to shore, your volumetric estimates need to account for what that ratio is -- which is unknown, but apparently quite high at this particular well/deposit. Said volumetric estimates are also complicated because a particle velocity analysis needs to somehow distinguish between particle acceleration due to gas expansion and particles just accelerating out of the pipe.
I'm finding my internal, personal guess trending towards something more moderate since I read through those comments. But I'm more convinced than ever that BP is deliberately obscuring all information about the whole damn thing.