[Voterescue] [EDA Data Analysis] 1-20-08 Holt's emergency bill to assure secure elections is a FRAUD.

[Jerry Lobdill]

Oh yeah! And how would we know that the compiled code was a faithful 
version of the disclosed open source even if there was such?

Jerry

At 12:33 PM 1/20/2008, Brent Turner wrote:
>Not to mention no open source / disclosure language, as Microsoft made sure-
>BT
>
>-----Original Message-----
>From: analysis-bounces at electiondefensealliance.org
>[mailto:analysis-bounces at electiondefensealliance.org] On Behalf Of Jerry
>Lobdill
>Sent: Sunday, January 20, 2008 6:48 AM
>To: voterescue at voterescue.org; EI Legislation List; 'Analysis EDA'
>Cc: Art Brender
>Subject: [EDA Data Analysis] 1-20-08 Holt's emergency bill to assure secure
>elections is a FRAUD.
>
>Holt's emergency bill,
>
>H.R.5036
>
>Emergency Assistance for Secure Elections Act of 2008 (Introduced in House)
>is totally inadequate to assure the integrity of an election. This bill is
>WORTHLESS. Here is the relevant language in the bill:
>
>
>
>
>http://thomas.loc.gov/cgi-bin/query/F?c110:1:./temp/~c1103UiRUC:e2303
><http://thomas.loc.gov/cgi-bin/query/F?c110:1:./temp/~c1103UiRUC:e2303>  :
>
>                 (c) Number of Ballots Counted Under Audit-
>
>
>
>                         (1) IN GENERAL- The hand counts of the paper ballots
>administered by the Election Auditor of a State, county, or equivalent
>location under this section with respect to an election shall occur in at
>least 3 percent of all precincts or equivalent locations (or alternative
>audit units used in accordance with the method provided for under paragraph
>(2)) in which ballots were cast in the election.
>
>
>                         (2) USE OF ALTERNATIVE SAMPLING MECHANISM-
>Notwithstanding paragraph (1) (and subject to the completeness requirement
>set forth in subsection (b)(2)), a State, county, or equivalent location may
>adopt and apply an alternative sampling mechanism to determine the number of
>paper ballots which will be subject to hand counts in accordance with this
>section with respect to an election, so long as the alternative sampling
>mechanism uses the paper ballots verified by voters or paper ballot
>printouts verifiable by voters at the time the vote is cast to conduct the
>audit and the Director of the National Institute of Standards and Technology
>or a panel of 3 independent statisticians appointed by the Election Auditor
>of the State, county, or equivalent jurisdiction determines that the
>alternative sampling mechanism will be at least as statistically effective
>in ensuring the accuracy of the election results as the sample size
>specified under paragraph (1).
>
>
>
>Dr. Holt, a physicist, knows precisely what the flaw in this language is.
>The question is, Why is he introducing a bill that he knows will not provide
>a definite and controllable probability of detection of vote fraud? The
>following does not even consider the damning facts that Holt's rescue bill
>permits the use of cash register thermal print "paper trail" records that do
>not bear any resemblance to the display screens a voter sees, and electronic
>ballot scanner/counters, which have demonstrated to be as vulnerable to
>fraud as the paperless voting machines.
>
>Here's an example of what will occur under the provisions of this bill.
>
>Suppose there are 1000 audit units (precincts, election districts,
>whatever). If you choose 3% of these (30) at random for a full hand recount,
>what does that buy you in terms of probability of choosing at least one
>corrupted audit unit?  The answer is, "It depends on the announced winning
>margin of votes, the distribution of audit unit size, how many units were
>actually corrupted in producing a fraudulent win, and the sizes of the units
>selected at random for audit."
>
>Just for example, suppose the precincts are all the same size, so that
>completely random selection is a proper strategy for audit. Suppose the
>margin of the winner is stated to be 5% and the corruption is evenly spread
>through 15 of the largest audit units (easily done by reference to county
>election records).
>
>What is the probability that your audit will find one of the corrupted
>precincts?  Why, it's about 0.37!
>
>What if the corruption is spread through 25 of the audit units? Then the
>probability of finding one corrupt unit in the audit is 0.66.
>
>So you see, the language of Holt's emergency bill is flawed and, considering
>his background and the work that was done on his previous bills before the
>politicos took over and emasculated them, you'd be a fool to think he is
>unaware of this flaw. Therefore, he is a fraudster with very high
>probability.
>
>You can read more about the details of why this is fraud in the following
>papers (which were available to Dr. Holt in 2006).
>
>Considering Vote Count Distribution in Designing Election Audits Rev.
>2, 11-26-06 , by Jerry Lobdill:
>
>http://vote.nist.gov/Considering-Vote-Count-Distribution-in-Designing-Electi
>on-Audits-Rev-2-11-26-06.pdf
><http://vote.nist.gov/Considering-Vote-Count-Distribution-in-Designing-Elect
>ion-Audits-Rev-2-11-26-06.pdf>
>
>Election Audit Sampling Design It's Not Just About Sampling Without
>Replacement 10-09-06, by Jerry Lobdill`:
>
>http://vote.nist.gov/Election-Audit-Sampling-Plan-Design-Its-Not-Just-About-
>Sampling-Without-Replacement-10-09-06.pdf
><http://vote.nist.gov/Election-Audit-Sampling-Plan-Design-Its-Not-Just-About
>-Sampling-Without-Replacement-10-09-06.pdf>
>
>Jerry Lobdill