image © Utah Count Votes

Utah Count Votes! Information to maintain Utah's excellent reputation for honest accurate elections. Please help ensure that "Utah Counts All Votes" Paper-Under-Glass Voter Verified Paper Audit Trail puts election integrity at risk. Accuracy Depends on Being Able to Detect and Correct Errors. Thermal paper roll "voter verifiable paper audit trails (vvpat)" are not practically countable by a state's own election officials! In Nevada the Nov. 04 election resulted in large discrepancies between the exit polls and the presidential election results. The Green party tried to obtain a recount of the DRE paper roll audit trail only to find out that outside experts recommended by the voting machine vendor had to be hired at huge expense to count them. No recount was performed. Later, the NV Secretary of State claimed that all counties audited the paper rolls, but this story was found to be false by investigators who phoned several counties and found out that some counties had merely tested the print programs by re-printing the electronic counts and had sealed and never opened or examined the paper rolls.

• Counting the paper rolls by hand could be a difficult process and requires a special piece of equipment to set the paper roll into in order to advance it. Counting the paper rolls electronically requires paper roll advancers, scanners, laptops, software, and technicians. The same errors made in the original electronic ballots could be made on the barcodes or in the software to count the barcodes.
• The voter must take an extra step to verify that the small paper roll writing matches the computer screen. Studies have shown that most voters do not bother to verify their votes. Of voters who do attempt to verify, often mistakes or omissions on the paper rolls are missed.
• A computer hacker could easily make the paper record be wrong and match a wrong electronic record for all ballots where the voter does not check the paper roll. This hack could steal roughly 2/3rds of the target votes in a way that a manual audit could not detect.
• The voter has no way to know if the votes are recorded properly inside the computer or on the barcode on the paper roll.
• Voter anonymity is not protected from poll workers because votes are stored in the same order as voters enter the polling booth. Poll workers who have to fix a printer, etc may find out how we voted.
• Flimsy paper rolls are impractical and do tear while in use, are not durable to store and use for recounts. If as in some systems, the paper is cut, how would we to stack, store, and count curled up, flimsy pieces of paper? The only way to catch possible errors in electronic vote tallies is to count the paper ballots, which may be very inconvenient with paper-under-glass rolls.
• Storing the ballots of all previous voters in the voting booth may have interesting advantages for hackers who might bring pliers, box cutters, or other tools into the voting booths to steal prior voters' ballot records. Will we need metal detectors at every polling place and frisk persons in wheelchairs to prevent mischievious theft of ballot records?

In sum, while paper-under-glass VVPAT might introduce new fun for election hackers and poll workers wanting to know how we voted, it is doubtful that they would increase election integrity. Computer scientists in each state are better qualified to set requirements for voter verified paper ballots to ensure honest accurate elections, than are commercial vendors.

Here is a much better easily re-countable ballot that can be printed, verified, and then placed in a ballot box by the voter. Try out the online demo and print the ballot it generates to see for yourself. Better voting systems designed by computer scientists can be developed in time to meet HAVA 2006 deadlines.