OldBoy:
Sorry, dudes, but somethings are just too complicated. We can make great strides in weather prediction, but making the leap from 3-4 days in the future to accurate predicitions a week into the future are quite difficult.
As a similar question, where will you be and what will you be doing at 19:38 next Thursday? Can you make an accurate (100% guaranteed, no prospect of a deviation from the plan, including your own death) prediction and then justify it (assuming you're not alread dead and reading this, in which case, it's probably pretty certain where you'll be and what you'll be doing)? Ok, now set 10000 PCs to evaluate your life and see if it can accurately predict what you'll be doing.
And yes, I know that this whole subject is a cultural icon of being a Northern European, but either way, the difficulty in the problem doesn't change.
H.
My Gallery.
flickr gallery