I've been doing a very informal experiment, plotting the size of the blockchain versus time. It appeared in my first run through the block chain, that the size actually retracted from the expected curve for a few weeks in the Feb/2013 timeframe. So, my graph is a nonlinear (curve) - naturally - but it has a little hiccup. Anyone else noticed this? I don't know if I made an error or not, so I'm redoing the entire blockchain test. Maybe I missed something. Kinda weird tho ....
Scanning the data, it almost looks like transactions were future dated in that time frame. I don't mean to start any alarm - I could have transferred something in error. A retest is in order. Unfortunately, my computer's slow and it could take a week to redo... will post when I recheck everything. Maybe someone else could look too ... at late Feb, early Mar 2013 time frame.