Andrew Plato wrote: >Theoretically, you could bribe polling working just as easy as you could >bribe a postal carrier. > No you couldn't. To bribe a mail carrier, you only have to bribe one person with no witnesses. The polling workers, OTOH, work under the observation of both Democrat and Republican observers. To effectively bribe poll workers, you would have to bribe observers from both parties, or be the observer from one party and bribe your opponent's observer to look the other way, which is improbable. >Electronic systems merely add a new mechanism for abuse. The only >difference with electronic systems is that a very few people really >understand how they work. Hence, you have a group of people with the >capability to cause problems and do so virtually undetected. This is why >such systems need audit trails that can be followed later. > That's not true either. A highly skilled expert in computer voting systems, standing outside the voting booth and looking at it, cannot tell whether the software inside has been corrupted. This is qualitatively different from a paper ballot system, where there is a pile of papers that can be re-counted if there is dispute. >One of the complaints about the Deibold system was that it was >essentially a black box. There was no way to audit the function of the >system. Hence, some bright hacker could corrupt the system at some point >and for the rest of the day all votes for George Bush could get >attributed to Al Sharpton. > Exactly. >Oregon will NOT be using Deibold's system. Deibold was under >consideration, but the elections group selected a local firm - Saber >Consulting - to design a new system. > Does this new system have VVAR (Voter Verifiable Audit Record)? This is a critical feature to make electronic voting safe. Essentially, the voting machine should generate a piece of paper clearly indicating the cast vote in human readable form and show it to the voter. The voter then affirms that the vote was cast as intended, and the piece of paper goes into the recount box of paper slips. Without VVAR, digital voting systems are all corruptable, no matter how much source code inspection you apply. >It would seem to me that election fraud isn't nearly as serious of a >problem as voter turn out. When only 40% of the eligible people >participate in an election, then essentially only 40% of the population >is making decisions for the remaining 60%. Anything that encourages >voter turn out without causing significant opportunity for fraud seems >like a good idea to me. > True, vote by mail meets your criteria. Digital voting without VVAR does *not*, because it really does open the door to completely corruptable elections, totally owned by whoever is the powerful elite in charge. Crispin
This archive was generated by hypermail 2b30 : Thu Sep 11 2003 - 17:28:57 PDT