|
Voting
Machine Standards Move Forward
by Terry
Costlow
In contrast to Florida’s problems
in 2000, voting in the 2004 election focused mainly on winners
and losers, not on the process of casting and counting ballots.
Today, engineers around the country are working together to
develop standards that should
help make sure that focus continues to be on vote counts rather
than the way votes are entered and tabulated.
As more and more jurisdictions
adopt electronic voting systems, an ongoing upgrade of standards provides the foundation for these voting machines and their
software. The U.S. Election Assistance Commission
(EAC) is
currently examining Voluntary Voting Systems Guideline proposals
submitted by the National Institute of Standards and
Technology’s (NIST)
Technical Guidelines Development Committee (TGDC).
TGDC supports EAC by providing recommendations on voluntary
standards and guidelines related to voting equipment and
technologies.
The stakes are high
As EAC
Chair Raymundo Martinez recently pointed out, the next big step
required by the Help America Vote Act (HAVA) will be to
distribute $2.3 billion of federal funding to the 50 states and
five voting territories. And those dollars will be augmented by state and local
expenditures for new equipment and training.
Interest in this combination of
money, technology and government actions was
evidenced when the IEEE finished a segment of its
Voting Equipment Standard (P1583). “In the spring of 2003,
we went to ballot and received 1,000 comments,” said Stephen Berger,
chair of IEEE Standards Coordinating Committee 38 (SCC
38). SCC 38 is currently developing two voting standards.
The voting equipment standard (P1583) is charged with developing a standard of
requirements and evaluation methods for election voting
equipment. The standard will provide technical specifications
for electronic, mechanical and human factors that manufacturers of voting machines
can use or by those purchasing such
machines. SCC 38's Electronic Data Interchange Standard (P1622) will develop
electronic data interchange formats to be used by components of
the voting system for exchange of electronic data.
Though NIST and the IEEE have
been working together, the standards are not identical. “At the
end of the day, I think a lot of the IEEE material will be included
in the EAC document and later we’ll publish a separate IEEE
standard that other countries could adopt,” said Berger, who
also serves on the NIST committee.
Need for upgraded standards
Consensus seems to exist over the
need to upgrade existing standards. “Improved standards are very
important. The previous versions are well-known to be
deficient,” said David Dill, a Stanford University professor who
is involved in vote-monitoring
organizations.
“We’re not developing new
standards, we’re augmenting those that are already in place,”
said an EAC spokeswoman.
The people developing the standards note that while the
standards
play an important role in developing and deploying voting machines, standards alone won’t solve all the problems
American voters can create.
“This will answer the question of
‘what is good enough to be a voting machine,’ it’s the starting
gate for vendors. For voting, this minimum requirement needs to
be high,” Berger said.
Standards provide the base on
which the voting systems will be built, but they will have an
enduring impact on both developers and those who must monitor and maintain
voting machines over their long
lifetimes. Up-to-date standards can help keep costs down, and
give buyers some assurance that they are
buying the right equipment.
“The lack of standards has been a
problem in getting equipment deployed. It creates fear,
uncertainty and doubt for the people buying equipment because
they’re afraid the standards might change,” Dill said.
Another significant aspect of
eliminating the
FUD (fear, uncertainty and doubt) factor is to answer critics who contend that electronic
equipment —
and the data they collect and store
— can be
corrupted through tampering. Voting equipment designers, developers
and implementers will be responsible for addressing the lion's
share of these concerns, but the
base for such secure systems is being developed in standards
committees.
“Security is certainly a big aspect of these standards. It’s one
of the premier goals,” the EAC spokeswoman said.
Looking forward
Security will be an even greater
concern in the further evolution of high-tech voting. Balloting
using the Internet has been tested in Michigan caucuses and
other areas, but it’s still a long way from widespread
acceptance.
Security concerns caused the Department of Defense to scuttle plans
to allow overseas military personnel to vote using the Web in the
recent presidential election. However, programs are in
place to see if Internet voting can be used in next year’s
elections.
“EAC tasked NIST to begin
working on standards for Internet voting in FY 2006. The focus
of these efforts is to meet EAC's obligations to assist the
Department of Defense in carrying out another electronic voting
demonstration project, as mandated by the FY 2005 National
Defense Authorization Act,” the EAC spokeswoman said.
For more information
Terry Costlow has been writing
about engineering issues for more than 20 years. He can be
reached at
todaysengineer@ieee.org.
|
E-mail
this page