[IP] Re: Escape From the E-Voting Quagmire]

[David Farber <dave@farber.net>]

________________________________________
From: Gordon Peterson [gep2@terabites.com]
Sent: Monday, January 28, 2008 3:32 AM
To: DallasDemocrats@yahoogroups.com; bushlied@yahoogroups.com; DigitalDemocrats@yahoogroups.com; Clean-ESS-Elections@yahoogroups.com; DiehardDems@yahoogroups.com; David Farber; unger@cs.columbia.edu
Subject: [Fwd: [IP] Escape From the E-Voting Quagmire]

While I have a few quibbles about the article referenced, it's overall a
pretty good discussion about the issue, and with some good links.

In the 2004 Presidential election, one of the key thefts that put Shrub
back into power happened in Ohio, and didn't happen in the polling
places and the electronic voting machines... it was apparently done at
the state level in the totalling and tabulating software.  Around
12:30am, as more precincts reported in, Kerry's vote (count) totals
abruptly WENT DOWN at the same time that Shrub suddenly jumped ahead to
just over 50% of the vote.  It's worth noting that this type of change
in the totals is EXACTLY the behavior produced by the Florida
demonstration vote-changing program delivered by a NASA contractor there
to (then) Floria rep. Feeney.

While it's only part of the solution, one key to preventing such fraud
would be (and the example I'm using is based on the ESS optical scanners
we use here in Dallas) very simple:

    1.  At the end of the election day, after closing the polls, the
judges would print the results tapes as we presently do.
    2.  The election judge would then print two to four EXTRA sets of
result tapes from the optical scanner(s).  These additional tapes, like
the existing tapes, would be signed at the polling place by the election
judge and the clerk(s).
    3.  When the election judge drops off their results at the central
vote counting station, he or she would hand one copy of each of the
additional tapes to the representatives there at the dropoff site from
each major political party.
    4.  This would allow the major parties to INDEPENDENTLY TOTAL THE
VOTES using their own spreadsheets.  Obviously, the partys' vote totals
should agree with the count reported by the county at the end of the
night... and in cases where it doesn't, obviously that needs to be
closely scrutinized.
    5   County parties could forward their spreadsheet totals to the
state parties, which (again) can use their own spreadsheets to verify
the totals reported by the state.  If there is a discrepancy, again, it
will stand out like a sore thumb.

   This approach doesn't do much for mail-in absentee ballots, or early
voting, but other comparable approaches should be developed for those
scenarios too.

   There are a variety of other approaches which can be used to help
make the voting more robust at the polling places.

   One further point that should be emphasized.... WE MUST NOT ALLOW
INTERNET VOTING!  There is NO way to ensure that the voter is not voting
under duress, with someone watching over their shoulder to make sure
they vote the "right" way... and perhaps even selling their vote.  THERE
IS *NO* TECHNICAL SOLUTION FOR THAT WHATSOEVER!  Internet voting, which
if adopted would probably quickly become 60% or more of total votes,
would be a HUGE and incredibly attractive target for such manipulation.
  It would also (of course) inevitably make it "easier" for people with
computers and internet connectivity to vote, and disadvantage those
without such connectivity.

-------- Original Message --------
Subject: [IP] Escape From the E-Voting Quagmire
Date: Sun, 27 Jan 2008 23:11:16 -0800
From: David Farber <dave@farber.net>
Reply-To: dave@farber.net
To: ip <ip@v2.listbox.com>


________________________________________
From: Stephen Unger [unger@cs.columbia.edu]
Sent: Sunday, January 27, 2008 3:23 PM
To: David Farber
Subject: Escape From the E-Voting Quagmire

Dave,

Here is another item you might consider for ip.

Steve
.............
The folly of the rush to e-voting has become increasingly clear over
the past year. An  opinion piece that I wrote summarizing the
situation and suggesting a plausible course change was printed today
in the Journal News. It can be viewed at
http://www.lohud.com/apps/pbcs.dll/article?AID=/20080127/OPINION/801270311/1076/OPINION03

Steve
.........

Stephen H. Unger
Professor, Computer Science (currently on leave-of-absence)
Columbia University

-------------------------------------------



--

Gordon Peterson II
http://personal.terabites.com
1977-2007:  Thirty year anniversary of local area networking

-------------------------------------------