Re: Not your ordinary barcode

From: William B. Cushman, Ph.D. <cushman_at_cox_dot_net>
Date: Mon Apr 19 2004 - 12:28:32 CDT

Alan Dechert wrote:

>William,
>
>
>
>>Which is all well and good, except for two things. In the real world I
>>can absolutely guarantee you that the barcode will NOT be used simply to
>>assist blind voters, it will be used for the actual count . . .
>>
>>
>>
>How do you know that? I don't think you've look at the OVC architecture.
>
>The precinct level tabulation is actually done using electronic ballot
>images (EBIs) from the voting stations that have been verified to have
>matching paper ballots.
>
>The verified EBIs get transmitted to the county HQ.
>
>There are numerous checks and cross checks. I will be virtually impossible
>for someone to make the barcode say one thing while the text says something
>else--without being detected.
>
>Alan,
>
    Actually, I was assuming a best case scenario. Transmitting EBIs is
even worse, from my perspective. There is way too much room for
mischief. You seem to be depending on obscurity and inaccessibility for
security and this will never work. This is the approach Microsoft
takes, and look what always seems to happen to them!

    Consider again, please, the Canadian style of election. Voters mark
their ballots with a pencil, drop them in a box, and they are later
counted in public by humans. I can't imagine any way to hack into this
kind of system, simply because it is so open. I think that your system
could be *equally as open*, and by analogy just as secure! The key to
this goal is to use the computer exactly like that Canadian pencil . . .
to simply produce a ballot. A simple, human readable barcode is a nice
extra . . . and will certainly serve to help "sell" the system . . .but
is really more dangerous than useful. The only reason for the barcode
at all is to facilitate counting (hopefully by many parties) and
possibly to "read" the ballot for a visually impaired person. Of course
the OCR approach being worked on could do this nicely. And, the OCR
could be used for counting as well . . . and will be if it is available
to do so . . . which gets us back to that same bag-of-worms concerning
the visibility of the process in its entirety. A pencil, paper, and
public count is visible and thus secure. If you introduce a computer
anywhere in that chain except where it is most useful and directly
auditable by the voter, i.e. as a means of producing the ballot in the
first place, then that visibility goes out the window along with your
security. Have you /ever/ seen a system that you could not hack, given
sufficient time, resources, and will, that was based on secrecy or
obscurity? NSA makes a business of doing just that, and if you think
that the people that would be interested in such mischief don't have
similar resources . . . well, I think that you are wrong, but why take
the chance if it is not necessary?
In this case it is /not /necessary!

    I know that you have built in security checks /ad nausium/ and
layered them upon each other to the point where it will make my head
spin to try and sort them out. Which is simply another way of saying
that you have provided lots of opportunities for mischief as well as
providing the illusion of security. I freely grant that you believe
that your security cannot be breached, and you could no doubt easily
prevent me from breaching it as well. But you have to be twenty times
smarter than the people who will try to breach that security to make
sure that there are no holes to exploit at all. The people trying to
breach your defenses only have to find that one little hole that you
overlooked.

    You have probably guessed that I have become somewhat cynical and
un-trusting in my old age, and you would be right about that. I am on
my local county Democratic Executive Committee and have seen too many
"anomalies" to really be comfortable with the behavior of my fellow
citizens when it comes to counting votes. Now, for many reasons, things
are getting very serious for our country. Our (Florida) Secretary of
State tells us to "trust her" . . . but I simply don't. She wants to
eliminate recounts entirely for DREs, and nearly did so a few weeks
ago. And there are so many "technical" reasons why these nice expensive
DREs can't print anything as well. I admit it, I'm paranoid as all
hell. I think we are being set up . . . again.

    I won't pester you any further, I think that I have said all I can
on this subject. Good luck with the project in any case. And just to
give you a warm, fuzzy feeling I have copied our Secretary of State's
last press release concerning DREs below. Keep in mind that:

"right now, without any further innovation, voters can have every confidence that their votes will count."

                                                                         
Bill Cushman

*Every vote will count when Floridians go to the polls*

By Glenda Hood

MY VIEW

Re: "Recount '04: Florida is headed for crisis" (column, April 4)

Reasonable people can disagree and often do. And there's certainly plenty of
room within the facts for reasonable people to disagree about the best way
to build an elections system for Florida.

But, unfortunately, too much of the debate this year has been based on
knee-jerk reactions to secondhand information. As someone who assumed
responsibility for an elections system that I did not choose or design, I
offer my views because I believe they represent the perspective of an
outsider who has quickly gained an insider's education.

Much has been made about the touch-screen technology in use in 15 Florida
counties, from conspiracy theories to predictions of crises and election
doom, as in Mike Pope's column.

Personally, I have a high degree of confidence that touch-screen machines
will record accurate election results this fall. Here's why: They use
computer technology, which makes them fast and accurate, but they are not
computers as we think of them in several important ways. They were built for
the sole purpose of being voting machines.

Each touch-screen machine has three independent hard drives recording votes.
While each of us can easily imagine a computer failure, three simultaneous
failures are much less plausible. Second, each machine is a stand-alone unit
 It is not connected to the Internet or to any network. So it cannot be
hacked" into.

Most importantly, each individual machine is subjected to rigorous logic and
accuracy testing by local, nonpartisan elections employees before every
election. The public is invited to attend and witness this testing to assure
public confidence.

The demand by some to add "printers" to touch-screen machines conjures up
images of something we could buy off the shelf at any computer store. But
that's as much of a misnomer as it is to call touch-screen voting machines
computers." It's not that simple. We are talking about developing a
specially designed, tamper-proof machine that doesn't exist right now - one
that prints a report that voters can see but not handle and that won't be
counted if voters say it's incorrect.

In the headlong rush to try to solve a problem that hasn't been demonstrated
to exist, we may create a few more problems along the way if we aren't
careful. First, the minor problems we've experienced in elections since 2002
have been the result of human error, not equipment failure.

It will be no small task to retrain tens of thousands of poll workers in how
to use these voter verified printers when they do become available. And as
with any new technology, especially one developed quickly, there may be
problems and unforeseen consequences, such as printer jams that may
significantly complicate Election Day.

Many Florida counties went to touch-screen voting because it's fast,
accurate, less prone to human error and less subject to ambiguous voter
intent, not to mention that it allows visually impaired and disabled voters
the opportunity to vote privately and independently for the first time ever.
These are benefits that would be largely lost by reverting to a paper system
 While our lawmakers ultimately may be willing to forego those benefits to
have a paper receipt, we should not rush to that conclusion without careful
consideration of all the facts.

For now, the most important facts are these: Before a voting machine with
printer capabilities can fully be developed and certified, standards must be
set on the national level by the Elections Commission and the National
Institute of Standards and Technology, and at the state level by the
Legislature, on the use and handling of the printed receipts.

In addition, the manufacturers of the touch-screen equipment used in Florida
tell us they cannot develop, test, complete the certification process, mass
produce and install voter verified printers on Florida's 35,000 touch screen
machines in the next five to seven months so that they can be used in the
fall elections.

We can and will have accurate, fair and decisive elections in 2004 without
printers. As technology continues to advance, Florida will continue to be on
the leading edge of it. But, right now, without any further innovation,
voters can have every confidence that their votes will count.

Glenda Hood is Florida's secretary of state. Contact her at
secretaryofstate@dos.state.fl.us.


==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Fri Apr 30 23:17:12 2004

This archive was generated by hypermail 2.1.8 : Fri Apr 30 2004 - 23:17:29 CDT