Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?

To: combjelly who wrote (128516)	11/14/2000 12:23:18 PM
From: Jim McMannis	Read Replies (2) \| Respond to of 1570498

The problem is withe the handling of ballots and the chads (punch outs) The more you handle them to more fall out. They have even names them. You have the corner chad (one corner broken) You have the swing door chad (two corners off) The hanging chad or flapping in the wind chad (three corners off) The pregnant chad (no corners off but bulging..) Now you see the problem... Best to take the machine count and fix the problem next time...after all, the ballots were made for machine counting and the machine is neither Dem or repub. Jim

To: combjelly who wrote (128516)	11/14/2000 1:56:57 PM
From: pgerassi	Read Replies (1) \| Respond to of 1570498

Dear Combjelly: Our computer systems that we are using today have a soft error rate (non repeatable errors) of less than 1 in a trillion reading from memory. The hard error rate is much less than that. Optically scanned ballots have a machine error rate (usually undervoted ballots) of less than 1 in a hundred. After a video scan, the error rate can be less than 1 in a 10,000 easy. Any rejected ballots would number only 580 for all of Florida. These can be decided by people who could agree on all but less than 1% of those leading to an error rate of less than 1 in a million over a quick tally that could be accomplished in less than 48 hours possibly even 24 hours with full audit trail, if needed. These numbers are upper bound of the actual error rates since even Oregon law states that a recount is necessary with a error of 0.2% on the initial run (better than 3 sigma probably). Thus they have confidence of their system of optically scanned ballots (their holdup is in the registration phase of the ballots not the counting phase) of having a standard deviation of less than 1 in 1500 for the first count including undervotes (typical in optically scanned ballots (mark is a little too far off for the fast machine but not for a video system using say template matching)). Thus, a better scan would only need to cover on average 1000 ballots for the whole state (1% would pop out 15,000 abnormal ballots there). When even the video scan machine cannot determine the vote, the ballot could be kicked out for humans to make the final determination. In all phases, a randomly selected sample could be used for Q/A and statistical certification that everything is running accurately. They could run all ballots through three times, and if they match exactly, the phase is considered to be correct. If slightly off, an agreed to test plan could troubleshoot the process at that phase. The last phase will result in a probability that only a few votes would be disputed enough in a really tight election but any arguments will cover less than a few votes (say under a dozen) in the whole election of a state. That is an error rate of a few vpm (votes per million). In that case, the challenger should get the benefit of the doubt due to the inherent advantages of the incumbent. If both candidates are not, the party not having the office at present (used even in college bowl selections as a tie breaker in places where ties are more frequent). This should satisfy all the parties and keep very close elections from snafuing an election. This is actually how it worked in the last two close elections, Carter (democrat from a republican Ford) and Kennedy (democrat from a republican Eisenhower)(I do not recall the election in the late 19th century but the republicans did not even exist then and the democrats took positions that would be considered republican today). This would clear up these close elections and get rid of the complaints of "squeaker" losers (very few winners complain). Pete