[E-voting] Disinformation about Ireland - long response re: VVAT question

Catherine Ansbro cansbro at eircom.net
Fri Mar 24 17:35:16 GMT 2006


VVAT in the USA is so far a pure disaster.  There are different rules in 
different states, and different vendors have different solutions, some 
of which are worse than others.  Here's a run-down of some of the issues:

*Getting Access to the VVAT/paper ballots to hand-count them is illegal, 
impossible, or highly restricted*
1) Laws in some states make it illegal (!!!) to hand-count the paper 
ballot or VVAT, even for a recount  (obviously, this makes the VVAT 
completely worthless--or actually worse, because there's a false sense 
of security)
2) Laws in many states make it illegal to hand-count the paper ballot or 
VVAT except in highly unlikely circumstances (e.g., if the margin 
between the winner and loser is extremely small, and if various other 
unlikely criteria is met, such as having to prove ahead of time that a 
recount might change the election results)  (again, this makes the VVAT 
completely worthless--but worse, because there's a false sense of security)
3) Laws in some states require manual audit of some percentage of the 
VVAT (too small a percentage; not always clear how/where the areas for 
recount are to be selected, which means it's easy to manipulate to 
ensure the results are "correct" in terms of matching the machine's poll 
tape)
4) Laws in some states allow manual audit of some percentage of the VVAT 
in certain cases (narrow margin and a losing candidate requests, and you 
have to prove that there's a chance of changing the election results, 
etc.--again, restrictive, and if fraud is only in some areas then one 
can avoid detection)
5) Supreme Court has already judged that ballot boxes with broken seals 
are to be counted, anyways

*Lousy technology - and I mean really lousy - with every manufacturer
*1) Some manufacturers added on a poor quality printer as an 
afterthought, with varying kinds of defects:
--thermal paper is used in some cases (storing it at the wrong temp can 
cause ink to fade)  (After some elections the ENTIRE VVAT was blank!  in 
real elections!)
--There are devices which can cover up the small screen through which 
you're supposed to check the votes (if the cover is flipped over, you 
don't even know there's a place where you can check your vote)
--There's an inadequate magnifying lens, but it only shows you part of 
the ballot
--You can't check all the ballot at once for errors, if you've moved on, 
you can't go back to check if the paper has already scrolled out of view
--Toilet-paper style roll, which can interfere with votor anonymity and 
is horribly impractical for recounting (just imagine)
--Bad paper feed systems that push the paper off-center, causing paper 
jams (really poor design)
--Different results from the printed tally compared to the machine 
tally, but some things not printing out correctly (this was for a 
machine that only prints out the ballots if you're going to do a recount)
--In some cases there are 2 different printers - the built-in one that 
produces a poll tape of the results, and a separate one that presents 
the voter's choices for verification (and of course there's no way to 
know whether or not the two tapes use the data in the same way)
--Clearing a paper jam causes loss of voter secrecy for several voters.
--For some machines, the handicapped can't see the VV screen because of 
its placement
--For some machines, the touchscreen detaches from the printer for 
wheelchair access--which means that voter doesn't get to have a VVAT, 
which throws a monkey wrench into any recount

*When the results differ, vendors and voting officials try to explain 
away the results*
--Computer "glitch" (note:  these are DEFECTS, not glitches!)
--Pollworker error
--Inadequate pollworker training

There have been LOTS of problems with machines breaking down and 
malfunctioning.  In many cases this has left the results unknown even 48 
hours after the polls closed.  In many cases there is no way to know 
what the real voters' intent was.  (E.g. the black audit roll--maybe the 
screen was covering up the verification area and no voters knew the tape 
was blank or that they were supposed to be able to verify?)

The Mean Time Between Error is so low that in some cases, the statistics 
indicate that every machine will malfunction on election day.  (Really.  
No exaggeration)

The problems and irregularities--with VVAT and without--are staggering.  
LOTS of results that defy any sense of logic--but because of the 
electronics or other restrictions, there's no way to check.  The legal 
restrictions that obstruct access to the VVAT for the purposes of 
recount/auditing are devastating.

And guess what--both major parties continue to obstruct access to the 
machines for independent investigation.  This is a non-partisan issue.

It's clear that there has got to be a paper ballot as the official 
voting record.  (No way should an electronic record ever be considered 
as "the vote".)  It's equally clear that all (yes 100%) of the paper 
ballots should be hand-counted at every election.  There are just too 
many indications of hanky-panky, and too many obstructions to 
investigating. 

There is a frightening willingness of vendors (all of them) to lie to 
election officials, to the media, and to the public.  Blatant lies, 
which have been shown to be untruths.  (E.g., claiming a bug was fixed, 
when internal memos revealed that they hadn't been, and they knew it.)  
Lack of access to testing results (what was tested and the results) is a 
problem.  Lack of a physical configuration audit (PCA) - since it's 
"trade secret" (as if it were a trade secret to know what kind of parts 
are in your car).

Election officials explain that they have become highly dependent on the 
vendors.  They take the vendors at their word.  They don't have the IT 
skills to understand the implications of the vulnerabilities.  Vendors 
easily have access to the equipment even while an election is in place.  
Vendors portray anyone asking questions as conspiracy theorists--even as 
completely wrong results roll in, right across the country.  (Wrong as 
in, 100,000 extra votes added in above the number of registered voters 
in a certain county.  Or vote tallies that don't add up correctly at 
all, but election officials didn't even notice till the next day.  We're 
talking gross malfunctions, let alone the subtle ones that no one knows 
about.)

The stench that is coming out of the US elections is unbelievable.  Even 
in primary elections.  (No surprise there, really, if you think about 
election strategy.)

There should be no need to "trust" any aspect of elections.  They've got 
to be open and transparent, and capable of being directly observed and 
understood by average citizens without any special IT knowledge. 

The recent explosive developments are showing that system architecture 
can be used to put in multiple security backdoors into a voting machine 
design, that will not be revealed by checking the software code.  
Checking the software does not ensure security. 

Have a look at blackboxvoting.org or bradblog.com, including the Voters 
Unite daily list of news articles about e-voting incidents. 

Catherine

Michael McMahon wrote:

> Catherine,
>
> Given that VVAT has been mandated in many jurisdictions in the US, do 
> you (or anyone else out there)
> know what the experience has been with it? Has there been any 
> elections yet using
> VVAT?
>
> The Texas system cited by Patrick is not VVAT as far as I can see.
>
> - Michael
>
>
>




More information about the E-voting mailing list