Message-ID: <87u0rxe0qb.fsf@sanosuke.troilus.org>
From: mdpoole at troilus.org (Michael Poole)
Subject: Re: Evidence Mounts that the Vote Was Hacked

Jei writes:

> On Tue, 9 Nov 2004, Jay D. Dyson wrote:
> 
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA1
> >
> > On Mon, 8 Nov 2004, Atom 'Smasher' wrote:
> >
> >> Evidence Mounts that the Vote Was Hacked
> >
> > 	Read the whole thing and didn't see any evidence.  Just wild
> > speculation and baseless conjecture.  Hell, there were countless counties
> > across the nation in which more people were registered to vote than were
> > eligible residents, but -- for some reason -- that ain't news.
> 
> It would be _major_ news, were it not America where it happened.
> Even India managed to hold a secure digital election recently,
> without any such major exit poll or other discrepancies happening.

Can you then explain the results of India's election this year given
exit poll numbers like those shown at
http://www.indian-elections.com/opinionpoll/ ?  While the range for
NDA is only about 21% (230 * 1.208 = 278), the range for the Congress
party (plus its allies) is about 123% (92-205), and Congress+ won 220
in the end (220 - 205 = 15, approximately 3% of the 542 seats being
voted on -- and 205 is the absolute upper range of exit poll
projections).

I'll admit I am not familiar with any of those five media or with the
Indian electoral system, but that looks like a rather more significant
discrepancy than seen in the US's exit poll numbers.  I suspect this
news is being inflated _because_ it happened in the USA.

> Also note that Americans aren't the only people in the world with
> capable intelligence agencies. Teenage kid hackers aren't the only
> people who might influence US elections' outcomes, given a viable
> chance. You need to consider all the factors.

One of Kerry's main campaign themes was "more respected abroad."  It
seems frivolous to invoke intelligence agencies as possible meddlers
in Bush's favor when most of them would have apparently preferred a
Kerry victory -- and when so many of the paperless precincts voted
strongly in Kerry's favor.

> Digital voting needs to be as secure and reliable as bank accounts
> are from an independent (democratic) nation's national security point
> of view. A digital vote discrepancy == national bank account
> discrepancy, in it's importance, in this regard.
> 
> Arguing that vote discrepancies don't really matter, is like a system
> admin arguing that system binary checksum discrepancies do not matter.

Assuming vote discrepancies in the absense of good evidence is
foolish.  People with much stronger analytic backgrounds than Thom
Hartmann have debunked the idea that the early exit poll numbers were
representative of all voters[1] as well as the idea that there were
relevant -- or even consistent -- discrepancies between exit polls and
final returns[2].  Others have explained why so many Florida counties
with high Democratic registration vote for Republican presidents[3].

Michael Poole

[1]- They did not adequately represent those who voted on Nov 2 and
entirely omitted absentee and early vote results; in some states,
those totalled 20 or 30% of registered voters.
[2]- http://vote.caltech.edu/Reports/VotingMachines3.pdf
[3]- In a teacup: Historic Democratic affiliation; very few local
Republican candidates, so many elections are settled in the (closed)
Democratic primaries; and a historic trend to vote for Republican and
southern presidential candidates.  These counties largely voted as
they did in 2000, and about half voted for Dole in 1996 in the mist of
what was largely a Clinton landslide victory.

Powered by blists - more mailing lists