Re: Electronic Voting Systems "Audited"

From: Ron Crane <voting_at_lastland_dot_net>
Date: Sat Aug 20 2005 - 10:47:45 CDT

Charlie Strauss wrote:

> Below is an intriguing article. An organization tried the following
> experiment which they mistakenly call parallel testing.
> Voters voted on an Diebold Accuvote, and then they were asked to vote
> again in a different manner.
> the result showed a 4% difference.
> The article wants you to believe this shows that the accuvote is
> inaccurate.
> Such a conclusion is of course unjustified till we reject more
> plausible explanations.

Really? "Of course?" For what reason does the "Accuvote" deserve this
deference? Has it been demonstrated to be built and operated
consistently in a trustworthy manner? Has its vendor been completely
open about every aspect of its software and hardware? Has its vendor
done its best to steer clear of partisan politics? Has its vendor
promoted vigorous public inquiry into its machines and into the
procedures for using them? Does its vendor prescribe hand recounts in a
randomly-selected statistically-significant set of precincts, to check
the system's performance?

> If we assume that in fact the accuvote were perfectly accurate then I
> can think of three possible reasons for the discrepancy.
> 1) the experiment was misconducted and is a pile of steaming poo
> 2) The voters are mischevious and lie.
> 3) Voters intrisicly make huge errors
> While 1 and 2 are highly likely contributing factors it's worth
> considering 3) further....

Factors (2) and (3) should produce randomly-distributed errors that, on
average, should not bias the results. Note, further, that both races
were biased in one partisan direction.

Factor (1) could explain the results (if, for example, those who ran the
"parallel election" cheated), but we don't have any specific evidence
that they did.

The results suggest cheating in the official tallies. That doesn't
necessarily implicate the "Accuvote", but it might very well. The
results call for a total *hand* recount and a thorough investigation,
and suggest, once again, that computer-assisted voting equipment
generally cannot be trusted and should not be used.


OVC discuss mailing lists
Send requests to subscribe or unsubscribe to
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Wed Aug 31 23:17:29 2005

This archive was generated by hypermail 2.1.8 : Thu Sep 15 2005 - 11:44:12 CDT