Sebastian Seung gave a talk here at FHI a few weeks ago. Here is a recording: Sebastian Seung on connectomes, slicing neurones, retinas, and gamers - YouTube. (the Q&A-session)
Some very interesting ideas here about how to bootstrap the necessary computer vision methods to handle the imminent flood of connectomics data. As a computational neuroscientist I'm itching to take these reconstructed neurons and turn them into neural networks just to see what happens.