This is exciting, but I feel the need to mention the Planning Fallacy. I'll be very happy if it gets done in 20 or 25 years.
After reading the nextbigfuture article and hackernews thread, I still don't understand what this project plans to do.
They appear to be aiming for whole brain emulation, trying to scale up previous efforts that simulated a rat neocortical column.
http://nextbigfuture.com/2011/05/henry-markram-and-human-brain-project.html
Markram's TED talk:
http://www.ted.com/talks/henry_markram_supercomputing_the_brain_s_secrets.html