IPS-English Q&A: ”There Is Such a Thing as Technology Overkill”
 
Date: Tue, 29 Jan 2008 14:14:14 -0800

 

Interview with Dr. Avi Rubin, technical director of the Johns Hopkins
University Information Security Institute

TAMPA, Florida, Jan 29  (IPS)  - Millions of votes will be counted
by computers this U.S. election season, and it seems that the chasm
between security gurus who argue that the existing systems are fatally
flawed and the manufacturers who claim that e-voting is both efficient
and airtight has never been wider.

Since 2002, Congress has disbursed 3.9 billion dollars to the states
to ensure that the confusion and bitter partisan wrangling surrounding
the 2001 presidential election never recurred. One result was a massive
investment in Direct Recording Electronic (DRE) technology, which
tallies votes with the swipe of a card or touch of a screen.

While many systems have appeared to function well in recent election
cycles, others have crashed, assigned votes to the wrong candidates,
and in at least one notorious instance, counted backwards. 

Also worrisome is a major new five-year study of five current e-voting
systems and one prototype by several U.S. universities, which found
that 3 percent of people voting electronically selected a candidate
they did not intend to choose, implying that problems can arise not
just from software glitches, but from confusion caused by poor user
interfaces.

Twenty-three states still do not require a paper record of all votes.

Avi Rubin, a widely respected authority on electronic voting and
the author of ”Brave New Ballot” (2006), believes that such a verifiable
paper trail could go a long way toward dispelling the anxieties induced
by e-voting.

He is also somewhat of a rarity among academics who study the topic
because unlike most, he has done tangible and practical work in the
field. In 2006, he acted as an election judge in Maryland's primary,
and has testified twice before the U.S. Election Assistance Commission.

IPS correspondent Mark Weisenmiller spoke to Rubin about the pros
and cons of these systems.

IPS: Did your work as an election judge teach you anything that seems
applicable in the 2008 primaries and presidential election?

AR: I learned that there is such a thing as technology overkill.
In the 2006 primary we had all kinds of problems in my state. All
of them could have been easily avoided, and in hindsight it was clear
how to do so, but at the time, the sheer complexity of the electronics
-- poll books, the smart-cards, and the electronic voting terminals,
coupled with an elderly workforce at the polls and buggy software
in the e-poll books -- led to many problems. I recently attended the
2008 election judge training class, and it appears that at the very
least, the e-poll book bugs have been fixed. However, some of the
poll workers in the training still seemed rather confused by all of
the technology.

IPS: Can you tell us a little about the organisation you helped found?

AR: The National Science Foundation's ACCURATE centre is a research
centre, funding professors and graduate students to help develop technologies
that could lead to more auditable, accurate, secure, transparent and
accessible elections. NSF typically funds basic and forward-looking
research, so while it is unlikely that we will develop a new voting
machine, we will develop the fundamental technology on which such
future machines can be built. There is also an important educational
and outreach component to our centre. We work with election officials
at all levels and provide education to our communities and the national
media about what the issues are with electronic voting.

IPS: Is there a particular region of the United States that seems
to experience the most problems with electronic voting?

AR: This is a tricky question because not all problems exhibit visible
signs. For example, if an election is run with electronic voting machines,
and a software glitch causes a percentage of the votes to accidentally
be tallied for the wrong candidate, it is possible that we would never
know that this happened. So would that election be viewed as problematic?
Probably not. What is required for unproblematic elections is the
ability to audit the election and to perform recounts. In that regard,
I don't think that it makes sense to classify by region, but rather
by which technology is used in each area.

IPS: Much of what we hear about the various types of computerised
voting machines is negative in nature. Which state in your opinion
does a superior job and how have they managed it?

AR: I believe the best voting system today is a ballot marking device,
used to make selections on a touch screen that prints out paper ballots,
which are then verified by the voters and optically scanned and counted
in the precinct. These systems provide a tangible audit record that
can be recounted, namely the paper ballot. The optical scanner in
the precinct allows for the elimination of residual votes, which are
errors by voters, such as a stray mark on the ballot or picking too
many candidates for a race. Voters have a chance to correct their
ballots if there is a problem. This is the most transparent and reliable
system I know of.

IPS: Do you foresee any problems in this year's state primary elections
or, more importantly, in the upcoming Nov. 4 presidential election?

AR: This is impossible to predict. The potential is there for problems.
The more we rely on software, the more things can go wrong in unpredictable
ways. Clearly the states that use paper ballots with optical scan
have the largest safety net if something goes wrong. Other states,
such as my home state of Maryland, will be in trouble. There is no
way to perform a meaningful recount in Maryland.

IPS: Will your company, Independent Security Evaluators, be involved
in the primaries or presidential elections in 2008?

AR: No, my company has a policy that we do not do any work related
to elections or electronic voting. I like to keep my two lives completely
separate and avoid any potential conflicts of interest between my
research and my business.



*****
+ POLITICS-US: Attempts to De-Bug Voting Systems Before 2008 Elections
ű Aug 2007 (http://ipsnews.net/news.asp?idnews=38921)
+ POLITICS-U.S.: Citizens Revolt Against Paperless Voting ű Jun 2004
(http://ipsnews.net/interna.asp?idnews=24044)
+ More IPS Coverage of the 2008 U.S. Elections (http://www.ipsnews.net/new_focus/us_elections2008/index.asp)


(END/IPS/NA/IP/IC/EL/MW/KS/08)


 
= 01292338 ORP015
NNNN