When the clock turns to November 1, the program crashes.

See http://www.talkingpointsmemo.com/archives/241672.php

It’s sad, really. The day before yesterday (I believe) he published graphs that, to my eyes, showed no change going back for I don’t know how long, because the wiggles all were about the same in scale and you could draw a flat line through them, whereas Marshall couldn’t decide whether the graphs showed no change or change twice a week. Indeed, he seemed emotionally pulled towards the latter, despite struggles to go with the former.

As I said, practically no change: http://www.talkingpointsmemo.com/archives/241821.php. Marshall’s use of the term ‘precise’ is deeply misleading, because the tool is too blunt to measure that precisely.

Get yourself a fountain pen: http://www.richardspens.com/?refp=nibs_whats_point