Of re-counts, and stolen elections

From: David Webber \(XML\) <"David>
Date: Tue May 17 2005 - 22:13:07 CDT

Folks- I've studied this closely and I must say the
objective of the TLV approach is to deliver fair,
unequivical, rapid and clear results. Of course
HAVA says that's what it wants too - but so
far has failed to show any sign of really
intending to.

The Carter / Boxer Commission may be
different.

>From the objective stance - random audits
do not work. If the election is known to be
really, really close from pre-vote polls - then
the incentive to cheat is overwhelming. And
remember candidates are spending $500 million
on elections - so what is another $1M or 5?

$20 per voter in key States campaigning too
such as New Hampshire.

So - we know (see ACM Journal articles) that
adding one vote to one machine per polling
station can be enough to carry an election.
No random audit is going to work here.

Therefore the TLV approach is to have 100%
reconciliation of all three trusted counting
sources - electoral roll vote records, electronic
vote records (voting computer) and verified paper ballot
(printed and cast by voter) and then scanned for
crosschecking.

This allows you to do 100% crosschecking on
election night - to ensure every paper vote has
a corresponding electronic vote, and that the
total of votes cast per station matches the
electoral roll count of voters attending.

By automating this - you can get answers
within hours. This is also crucial - since
legal challenges that take weeks - will fail
to the force majeur effect of the "winner"
having already moved in - picked out
wallpaper and curtains and appointed
their staff.

Anything less than 100% automated audit
checking is a waste of time. This requires
underpinning from the process itself to
carefully separate the sources of the
records that are to be crosschecked.
Combining paper and electronic to
provide a verifiable physical source with
an equivalent machine generated count
and records. This prevents either source
from cheating on the other.

The TLV approach using OASIS EML
explains how this is done.

You can never get 100% secure anything -
but to compromise this approach requires
you to systematically corrupt 4 different
sources of counting, and then across
a geographic region - if an organization
can achieve that - vote rigging is the least
of our worries!!!

The only time I could see doing a small
sample additional hand audit is if you simply
wanted to crosscheck a precinct or so,
just to do a post election sanity check,
and tediously manually check the
computer records and such to make
sure there are no glitches.

DW

_______________________________________________
OVC discuss mailing lists
Send requests to subscribe or unsubscribe to arthur@openvotingconsortium.org
==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Tue May 31 23:17:40 2005

This archive was generated by hypermail 2.1.8 : Tue May 31 2005 - 23:17:52 CDT