|
The 04 "inexplicably OFF-register" was discovered.
It wasn't inexplicable. The ballot printer (Eagle) subcontracted to
another print house that had lousy machines. Optical stretch, bad
paper, and zero testing. About 13000 ballots failed to be recognized
properly by the scanners. In some cases these ballots weren't able to
be scanned at all, because the registration marks were out of
tolerance. About 1000 ballot were visually hozed, the rest looked fine,
but the Hart scanners still couldn't read them properly. Hart proposed that their print shop in Texas was the better choice (nice marketing), and on the next round Hart was the printer. The ERC proposed that ballots be tested by running them blank (unfilled) through the scanners before sending them out. It is my understanding that Hillary ordered a fractional test. Not all the ballots were tested, but some where. There are two issues with a fractional test: bad ballots may show up in a test (or not) because they are in batch order, but if the batch for testing is pulled from a good stack, then overall errors would be missed. This leads us to the second issue. The ballots are not returned in batch order. As costly as it may be, the best test of the ballots would be to test all of them before they leave the clerk's office. However, once the ballots are distributed and out of the lab and into the real world, we have no control. No clue as to what conditions they will be in once outside of the elections office. Mail ballots are subjected to all sorts of environmental and human handling conditions. Less so if they were voted in precincts. Even better - don't scan them, hand count them. paul tiger Lou Puls wrote:
|