Okay, now...
Do you see anything in this methodology that would seem to artificially reduce the number of sectarian attacks on civilians?
In particular, since it's obvious that the same methodology would have been used for the pre-surge numbers, so the methodology wouldn't be weighting the numbers anyway, is there an actual point to this?
It's the 22nd of September, and the count so far this month of Iraqi civilians killed is 494, compared to almost 1,500 last month.