[House Hearing, 108 Congress]
[From the U.S. Government Publishing Office]
TESTING AND CERTIFICATION FOR
VOTING EQUIPMENT: HOW CAN
THE PROCESS BE IMPROVED?
=======================================================================
HEARING
BEFORE THE
SUBCOMMITTEE ON ENVIRONMENT, TECHNOLOGY,
AND STANDARDS
COMMITTEE ON SCIENCE
HOUSE OF REPRESENTATIVES
ONE HUNDRED EIGHTH CONGRESS
SECOND SESSION
__________
JUNE 24, 2004
__________
Serial No. 108-65
__________
Printed for the use of the Committee on Science
Available via the World Wide Web: http://www.house.gov/science
U.S. GOVERNMENT PRINTING OFFICE
94-316 WASHINGTON : 2004
____________________________________________________________________________
For Sale by the Superintendent of Documents, U.S. Government Printing Office
Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800; (202) 512�091800
Fax: (202) 512�092250 Mail: Stop SSOP, Washington, DC 20402�090001
______
COMMITTEE ON SCIENCE
HON. SHERWOOD L. BOEHLERT, New York, Chairman
RALPH M. HALL, Texas BART GORDON, Tennessee
LAMAR S. SMITH, Texas JERRY F. COSTELLO, Illinois
CURT WELDON, Pennsylvania EDDIE BERNICE JOHNSON, Texas
DANA ROHRABACHER, California LYNN C. WOOLSEY, California
KEN CALVERT, California NICK LAMPSON, Texas
NICK SMITH, Michigan JOHN B. LARSON, Connecticut
ROSCOE G. BARTLETT, Maryland MARK UDALL, Colorado
VERNON J. EHLERS, Michigan DAVID WU, Oregon
GIL GUTKNECHT, Minnesota MICHAEL M. HONDA, California
GEORGE R. NETHERCUTT, JR., BRAD MILLER, North Carolina
Washington LINCOLN DAVIS, Tennessee
FRANK D. LUCAS, Oklahoma SHEILA JACKSON LEE, Texas
JUDY BIGGERT, Illinois ZOE LOFGREN, California
WAYNE T. GILCHREST, Maryland BRAD SHERMAN, California
W. TODD AKIN, Missouri BRIAN BAIRD, Washington
TIMOTHY V. JOHNSON, Illinois DENNIS MOORE, Kansas
MELISSA A. HART, Pennsylvania ANTHONY D. WEINER, New York
J. RANDY FORBES, Virginia JIM MATHESON, Utah
PHIL GINGREY, Georgia DENNIS A. CARDOZA, California
ROB BISHOP, Utah VACANCY
MICHAEL C. BURGESS, Texas VACANCY
JO BONNER, Alabama VACANCY
TOM FEENEY, Florida
RANDY NEUGEBAUER, Texas
VACANCY
------
Subcommittee on Environment, Technology, and Standards
VERNON J. EHLERS, Michigan, Chairman
NICK SMITH, Michigan MARK UDALL, Colorado
GIL GUTKNECHT, Minnesota BRAD MILLER, North Carolina
JUDY BIGGERT, Illinois LINCOLN DAVIS, Tennessee
WAYNE T. GILCHREST, Maryland BRIAN BAIRD, Washington
TIMOTHY V. JOHNSON, Illinois JIM MATHESON, Utah
MICHAEL C. BURGESS, Texas ZOE LOFGREN, California
VACANCY BART GORDON, Tennessee
SHERWOOD L. BOEHLERT, New York
ERIC WEBSTER Subcommittee Staff Director
MIKE QUEAR Democratic Professional Staff Member
JEAN FRUCI Democratic Professional Staff Member
OLWEN HUXLEY Professional Staff Member
MARTY SPITZER Professional Staff Member
SUSANNAH FOSTER Professional Staff Member
AMY CARROLL Professional Staff Member/Chairman's Designee
ADAM SHAMPAINE Majority Staff Assistant
MARTY RALSTON Democratic Staff Assistant
C O N T E N T S
June 24, 2004
Page
Witness List..................................................... 2
Hearing Charter.................................................. 3
Opening Statements
Statement by Representative Vernon J. Ehlers, Chairman,
Subcommittee on Environment, Technology, and Standards,
Committee on Science, U.S. House of Representatives............ 10
Written Statement............................................ 11
Statement by Representative Mark Udall, Ranking Minority Member,
Subcommittee on Environment, Technology, and Standards,
Committee on Science, U.S. House of Representatives............ 12
Written Statement............................................ 13
Panel I:
The Hon. Rush Holt, a Representative in Congress from the State
of New Jersey
Oral Statement............................................... 14
Written Statement............................................ 15
Panel II:
Mr. Thomas R. Wilkey, Chair, Independent Testing Authority (ITA)
Committee, National Association of State Election Directors
Oral Statement............................................... 38
Written Statement............................................ 40
Ms. Carolyn E. Coggins, Director, ITA Services at SysTest Labs
Oral Statement............................................... 42
Written Statement............................................ 44
Biography.................................................... 52
Financial Disclosure......................................... 53
Dr. Michael I. Shamos, Professor of Computer Science, Carnegie
Mellon University
Oral Statement............................................... 54
Written Statement............................................ 56
Biography.................................................... 59
Dr. Hratch G. Semerjian, Acting Director, National Institute of
Standards and Technology (NIST)
Oral Statement............................................... 60
Written Statement............................................ 62
Biography.................................................... 65
Discussion.......................................................
Election Management Best Practices and Acceptance Testing of
Voting Equipment............................................. 65
Should All Computer-based Voting Equipment Be Required to Have
a Paper Trail?............................................... 68
Technologies for Reducing Voter Fraud.......................... 71
Role and Ability of NIST to Address Voter Equipment Testing and
Evaluation Issues............................................ 72
What Does NIST Need to Fulfill This Role?...................... 74
What Do States and Other Entities Need to Do to Improve the
Technological Aspects of Elections?.......................... 76
Appendix: Answers to Post-Hearing Questions
.................................................................
Ms. Carolyn E. Coggins, Director, ITA Services at SysTest Labs 82
Dr. Hratch G. Semerjian, Acting Director, National Institute of
Standards and Technology (NIST)................................ 85
TESTING AND CERTIFICATION FOR VOTING EQUIPMENT: HOW CAN THE PROCESS BE
IMPROVED?
----------
THURSDAY, JUNE 24, 2004
House of Representatives,
Subcommittee on Environment, Technology, and
Standards,
Committee on Science,
Washington, DC.
The Subcommittee met, pursuant to other business, at 2:20
p.m., in Room 2318 of the Rayburn House Office Building, Hon.
Vernon J. Ehlers [Chairman of the Subcommittee] presiding.
hearing charter
SUBCOMMITTEE ON ENVIRONMENT, TECHNOLOGY, AND STANDARDS
COMMITTEE ON SCIENCE
U.S. HOUSE OF REPRESENTATIVES
Testing and Certification for
Voting Equipment: How Can
the Process Be Improved?
thursday, june 24, 2004
2:00 p.m.-4:00 p.m.
2318 rayburn house office building
Purpose:
On Thursday, June 24, 2004, the House Science Subcommittee on
Environment, Technology, and Standards will hold a hearing to examine
how voting equipment is tested against voting system standards and how
the independent laboratories that test voting equipment are selected.
Each election season, a small number of newly deployed voting
machines fail to perform properly in the field, causing confusion in
the polling places and concerns over the potential loss of votes.
Because these machines have already been tested and certified against
standards, these incidents have raised questions about the reliability
of the testing process, the credibility of standards against which the
machines are tested, and the laboratories that carry out the tests.
While most of the national attention on voting systems has been focused
on the subjects of computer hacking and voter-verifiable paper ballots,
press reports (see Appendix A) have also highlighted the problems of
voting machine testing.
A focus of the hearing will be how the implementation of the Help
America Vote Act (HAVA) is intended to improve the way voting machines
are tested, the role of the National Institute of Standards and
Technology (NIST), and what changes can be implemented in time for the
2004 election and beyond.
Witnesses:
Dr. Hratch Semerjian--Acting Director, National Institute of Standards
and Technology (NIST).
Mr. Tom Wilkey--Chair of the National Association of State Elections
Directors (NASED) Independent Testing Authority (ITA) Committee. He is
the former Executive Director of the New York State Board of Elections.
Ms. Carolyn Coggins--Director of Independent Testing Authority Services
for SysTest Laboratories, a Denver laboratory that tests software used
in voting machines.
Dr. Michael Shamos--Professor of Computer Science at Carnegie Mellon
University. He has served as an Examiner of Electronic Voting Systems
for Pennsylvania.
Overarching Questions:
The Subcommittee plans to explore the following questions:
How are the accreditation of testing laboratories and
the testing and certification of voting equipment conducted?
How should voting equipment standards and laboratory
testing be changed to improve the quality of voting equipment
and ensure greater trust and confidence in voting systems?
What can be done to improve these processes before
the 2004 election, and what needs to be done to finish these
improvements by 2006?
Background:
Introduction
In October 2002, Congress passed the Help America Vote Act (HAVA)
to help correct the problems with voting machines that were brought to
the public's attention during the 2000 federal election. Under HAVA,
the States are receiving $2.3 billion in fiscal 2004 to purchase new
voting equipment. To try to encourage and enable states to buy
effective voting equipment, HAVA reformed the way standards for voting
machines are developed and the way voting machines are tested against
those standards. However, HAVA does not require any state or
manufacturer to abide by the standards.
Before the passage of the Help America Vote Act (HAVA), the Federal
Election Commission (FEC) established voting system standards. A non-
governmental group of State elections directors (the National
Association of State Elections Directors, or NASED) accredited the
laboratories, also known as Independent Testing Authorities (ITAs),
which then tested whether voting systems met the standards. With the
passage of HAVA, the responsibility for issuing voting system standards
and for accrediting the ITAs was transferred to the Election Assistance
Commission (EAC). Under HAVA, the EAC is to select ITAs based on the
recommendations of the National Institute of Standards and Technology
(NIST). For more information on HAVA, see Appendix B.
The transition to the new standards regime, however, has been slow.
Members of the EAC were appointed at the end of 2003. Congress provided
little funding this year to the EAC and none at all to NIST to begin to
carry out its duties under HAVA. (At the Science Committee's
instigation, the Administration was able to find $350,000 for NIST to
carry out some of the most urgently needed work.) As a result, the
current testing regime is essentially identical to that which existed
before Congress passed HAVA.
The FEC Testing Regime
The standards used today were first issued by the FEC in 1990 and
last updated in 2002. Those standards, known as the Voting System
Standard (VSS), deal with performance, security, and other aspects of
voting systems have existed since 1990. The FEC developed the standards
on a limited budget with input from NASED, voting experts,
manufacturers, and interest groups, such as the disabled and the League
of Women Voters, many of whom participated on a volunteer basis.
Although no federal mandate requires that the standards be used, some
States have adopted them as mandatory requirements.
To qualify voting machines under the FEC standards, manufacturers
must send their equipment to a NASED-approved laboratory (ITA) for
testing and inspection. There are three ITAs: Wyle Laboratories, which
tests hardware; and CIBER and SysTest laboratories, which test
software.
Prior to HAVA, the Federal Government had no official role in
approving ITAs. The FEC did cooperate informally with NASED to identify
laboratories that could become ITAs. However, few laboratories were
willing to participate because they viewed voting machine certification
as a risky venture that was unlikely to generate much revenue.
Once a voting machine or its software has passed the current
testing process, it is added to the NASED list of ``Qualified'' voting
systems, which means they have met the FEC standards. The only publicly
available information is whether a particular machine has passed
testing; the complete tests results are not made public because they
are considered proprietary information.
Voting technology experts have raised a number of concerns about
the standards and testing under the FEC system. They include:
Some of the FEC Voting System Standards are
descriptive rather than quantitative, making it more difficult
to measure compliance.
Many of the FEC Voting System Standards are described
very generally, for example those for security. Although this
avoids dictating specific technologies to the manufacturers,
the standards may require more specificity to be meaningful and
effective.
The ITAs do not necessarily test the same things in
the same way so a test for a specific aspect of computer
security in one lab may not be the same test used in another.
Hardware and software laboratories do not necessarily
know each other's testing procedures, and although
communication takes place between them, they are not required
to integrate or coordinate their tests.
The ITAs, once chosen, are not regularly reviewed for
performance. Reaccreditation would help ensure that quality and
expertise did not decline or otherwise change over time, and
that any new testing protocols were being carried out
appropriately.
Few States effectively test voting machines once they
are delivered even though ITA testing--like most product
testing--tests samples rather than every unit of a product.
When Georgia, in association with Kennesaw State University,
conducted their own independent test of their new machines, the
State sent five percent of them back to the manufacturer for
various defects.
Companies offer, and States install, last-minute
software ``patches'' that have not been subjected to any
testing. California recently decertified new voting machines
because they included untested software patches.
The small number of ITAs limits the amount of
competition on the basis of either price or quality.
As is the case in most product testing,
manufacturers, rather than disinterested third parties, pay for
the testing.
The Pending NIST Testing Regime
To fully implement HAVA, NIST will have to develop, and the EAC
will have to approve standards that the voting equipment must meet (to
replace the FEC Voting Systems Standards); tests to determine whether
voting equipment complies with those standards; and tests to determine
whether laboratories are qualified to become ITAs. NIST has begun
preliminary work on some of these tasks, but has been constrained by
scarce funds.
Under HAVA, NIST is also to conduct an evaluation of any laboratory
that wishes to become an ITA (including ITAs that were already
accredited under the NASED system). Accreditation would then be granted
by the EAC based on NIST's recommendations. HAVA also requires NIST to
monitor the performance of the ITAs, including, if necessary,
recommending that the EAC revoke an ITA's accreditation. (These
provisions of HAVA originated in the House Science Committee.)
NIST has not yet begun to implement this aspect of HAVA, but NIST
recently announced that it will soon convene a meeting for those
laboratories that are interested in becoming ITAs to discuss what
qualifications they must meet.
Since NIST has just begun developing lab accreditation standards,
as an interim measure, NIST will probably accredit laboratories as ITAs
using a generic, international standard for laboratories, known as ISO
17025. NIST uses that standard already as part of its existing program
for certifying laboratories for other purposes, known as the National
Voluntary Laboratory Accreditation Program (NVLAP).
Obviously, none of this will be done in time to affect the purchase
of equipment for the 2004 elections, and many States are making large
purchases of voting equipment now with the money available under HAVA.
However, a number of large States have not yet purchased equipment
partly because of uncertainty about what the new standards will be.
Limitations of Laboratory Testing in Reducing Errors in Voting
Equipment
An improved federal certification process is a necessary, but not
sufficient condition for improving the performance of voting equipment.
According to experts, among the issues that remain are:
No one is required to abide by the new system,
although presumably States will want to buy equipment that
meets the EAC standards and has been tested in federally
certified ITAs.
Laboratories cannot test every situation that may
arise in the actual use of voting machines. Election experts
say States should do their own testing, including simulated
elections. Some States, for example Georgia, California, and
Florida, are implementing tests of their own.
Pollworker training and voter education are critical
to reducing human error and resulting problems with voting
equipment. Technology that works perfectly can still be
confusing to the users.
WITNESS QUESTIONS
In their letters of invitation, the witnesses were asked to respond
to the following questions:
Questions for Dr. Semerjian:
1. How should the accreditation of testing laboratories and
the testing and certification of voting equipment be changed to
improve the quality of voting equipment and ensure greater
trust and confidence in voting systems?
2. What can be done to improve these processes before the 2004
election, and what needs to be done to finish these
improvements by 2006? Do enough Independent Testing Authorities
exist to carry out the needed tests? If not, what can be done
to increase the number of laboratories?
3. What progress has NIST made in carrying out the
requirements of the Help America Vote Act?
Questions for Mr. Wilkey:
1. How should the accreditation of testing laboratories and
the testing and certification of voting equipment be changed to
improve the quality of voting equipment and ensure greater
trust and confidence in voting systems?
2. What can be done to improve these processes before the 2004
election, and what needs to be done to finish these
improvements by 2006?
3. Do enough Independent Testing Authorities exist to carry
out the needed tests? If not, what can be done to increase the
number of laboratories?
Questions for Ms. Coggins:
1. How should the accreditation of testing laboratories and
the testing and certification of voting equipment be changed to
improve the quality of voting equipment and ensure greater
trust and confidence in voting systems?
2. What can be done to improve these processes before the 2004
election, and what needs to be done to finish these
improvements by 2006?
3. How do standards affect the way you test voting equipment?
Questions for Dr. Shamos:
1. How should the accreditation of testing laboratories and
the testing and certification of voting equipment be changed to
improve the quality of voting equipment and ensure greater
trust and confidence in voting systems?
2. What can be done to improve these processes before the 2004
election, and what needs to be done to finish these
improvements by 2006?
3. How important is NIST's role in improving the way voting
equipment is tested? What activities should States be
undertaking to ensure voting equipment works properly?
APPENDIX A
Who Tests Voting Machines?
New York Times Editorial
May 30, 2004
Whenever questions are raised about the reliability of electronic
voting machines, election officials have a ready response: independent
testing. There is nothing to worry about, they insist, because the
software has been painstakingly reviewed by independent testing
authorities to make sure it is accurate and honest, and then certified
by State election officials. But this process is riddled with problems,
including conflicts of interest and a disturbing lack of transparency.
Voters should demand reform, and they should also keep demanding, as a
growing number of Americans are, a voter-verified paper record of their
vote.
Experts have been warning that electronic voting in its current
form cannot be trusted. There is a real danger that elections could be
stolen by nefarious computer code, or that accidental errors could
change an election's outcome. But State officials invariably say that
the machines are tested by federally selected laboratories. The League
of Women Voters, in a paper dismissing calls for voter-verified paper
trails, puts its faith in ``the certification and standards process.''
But there is, to begin with, a stunning lack of transparency
surrounding this process. Voters have a right to know how voting
machine testing is done. Testing companies disagree, routinely denying
government officials and the public basic information. Kevin Shelley,
the California Secretary of State, could not get two companies testing
his State's machines to answer even basic questions. One of them, Wyle
Laboratories, refused to tell us anything about how it tests, or about
its testers' credentials. ``We don't discuss our voting machine work,''
said Dan Reeder, a Wyle spokesman.
Although they are called independent, these labs are selected and
paid by the voting machine companies, not by the government. They can
come under enormous pressure to do reviews quickly, and not to find
problems, which slow things down and create additional costs. Brian
Phillips, president of SysTest Labs, one of three companies that review
voting machines, conceded, ``There's going to be the risk of a conflict
of interest when you are being paid by the vendor that you are
qualifying product for.''
It is difficult to determine what, precisely, the labs do. To
ensure there are no flaws in the software, every line should be
scrutinized, but it is hard to believe this is being done for voting
software, which can contain more than a million lines. Dr. David Dill,
a professor of computer science at Stanford University, calls it
``basically an impossible task,'' and doubts it is occurring. In any
case, he says, ``there is no technology that can find all of the bugs
and malicious things in software.''
The testing authorities are currently working off 2002 standards
that computer experts say are inadequate. One glaring flaw, notes
Rebecca Mercuri, a Harvard-affiliated computer scientist, is that the
standards do not require examination of any commercial, off-the-shelf
software used in voting machines, even though it can contain flaws that
put the integrity of the whole system in doubt. A study of Maryland's
voting machines earlier this year found that they used Microsoft
software that lacked critical security updates, including one to stop
remote attackers from taking over the machine.
If so-called independent testing were as effective as its
supporters claim, the certified software should work flawlessly. But
there have been disturbing malfunctions. Software that will be used in
Miami-Dade County, Fla., this year was found to have a troubling error:
when it performed an audit of all of the votes cast, it failed to
correctly match voting machines to their corresponding vote totals.
If independent testing were taken seriously, there would be an
absolute bar on using untested and uncertified software. But when it is
expedient, manufacturers and election officials toss aside the rules
without telling the voters. In California, a State audit found that
voters in 17 counties cast votes last fall on machines with uncertified
software. When Georgia's new voting machines were not working weeks
before the 2002 election, uncertified software that was not approved by
any laboratory was added to every machine in the state.
The system requires a complete overhaul. The Election Assistance
Commission, a newly created federal body, has begun a review, but it
has been slow to start, and it is hamstrung by inadequate finances. The
commission should move rapidly to require a system that includes:
Truly independent laboratories. Government, not the voting machine
companies, must pay for the testing and oversee it.
Transparency. Voters should be told how testing is being done, and
the testers' qualifications.
Rigorous standards. These should spell out in detail how software
and hardware are to be tested, and fix deficiencies computer experts
have found.
Tough penalties for violations. Voting machine companies and
election officials who try to pass off uncertified software and
hardware as certified should face civil and criminal penalties.
Mandatory backups. Since it is extremely difficult to know that
electronic voting machines will be certified and functional on Election
Day, election officials should be required to have a non-electronic
system available for use.
None of these are substitutes for the best protection of all: a
voter-verified paper record, either a printed receipt that voters can
see (but not take with them) for touch-screen machines, or the ballot
itself for optical scan machines. These create a hard record of
people's votes that can be compared to the machine totals to make sure
the counts are honest. It is unlikely testing and certification will
ever be a complete answer to concerns about electronic voting, but they
certainly are not now.
APPENDIX B
The Help America Vote Act (HAVA)
In 2002, the President signed the Help America Vote Act (HAVA) into
law, which included a number of measures intended to improve the U.S.
election system. Among other things, HAVA banned the use of punch card
and lever voting machines and provided funds to the States to replace
them. It established an Election Assistance Commission (EAC) to assist
in the administration of federal elections and the administration of
certain federal election laws and programs, and otherwise oversee the
reforms recommended under HAVA. HAVA also established a number of basic
requirements that voting machines and systems should meet, and a
process by which new voluntary technical standards could be developed
to ensure the reliability and accuracy of new voting equipment.
The Science Committee included provisions in HAVA that designated
the Director of the National Institute of Standards and Technology
(NIST) to chair the Technical Guidelines Development Committee (TGDC),
a 14-member panel charged with the development of voluntary voting
system guidelines, or standards. HAVA also created a 110-member
Standards Board consisting of State and local election officials, and a
37-member Board of Advisors consisting of representatives from various
associations, who together would review the standards recommended by
the TGDC. The EAC was given the final word on whether these standards
would be officially adopted. Once adopted, it would still be up to the
States to determine whether the equipment they bought needed to meet
the standards, since they are meant to be voluntary, not coercive.
Chairman Ehlers. It is my pleasure to call this hearing to
order. It is a hearing on Testing and Certification for Voting
Equipment: How Can the Process be Improved? And we--I apologize
for the delay in starting. That is the bad news. The good news
is we are now unlikely to be interrupted by votes for the
remainder of the hearing, so we should be able to proceed
directly through it.
I am pleased to welcome you today to today's hearing on
improving the testing and certification of voting equipment.
Most of the national attention on voting systems has focused on
the subjects of computer hacking and voter verifiable paper
ballots. However, recently, the New York Times and other
organizations have brought more public attention to the subject
of voting machine testing, the laboratories that test the
machines, and the development of standards used to conduct the
tests.
All new models of voting machines sold in the U.S. today
are certified by the National Association of State Elections
Directors after having passed a series of tests administered by
Independent Testing Authorities, known as ITAs, which are
private laboratories. These tests are conducted to ensure that
the machines meet certain standards for environmental
tolerances, logic, and accuracy, computer security, and other
metrics that make them fit for use in elections. Voting
machines must also be certified by individual states before
they can be purchased by State or local election officials.
However, each election season, a small number of newly
deployed voting machines fail to perform properly in the field,
causing confusion in the polling places, and concerns over the
potential loss of votes. Because these machines have already
been tested and certified against Federal Election Commission
standards, these incidents have raised questions about the
reliability of the testing process, the credibility of
standards against which the machines are tested, and the
laboratories that carry out the tests. We must resolve this
issue soon, because states are already receiving billions of
federal dollars under the Help America Vote Act, or HAVA, to
modernize their voting systems. It is crucial that voting
systems be easy to use, accurate, verifiable, secure, and
reliable, and all of those criteria must be met.
The Science Committee, through HAVA, gave the National
Institute of Standards and Technology, known as NIST, the role
of improving the accreditation process of the laboratories
carrying out the tests, and the standards against which
machines must be tested and certified. Ultimately, NIST's
activities under HAVA will improve the overall quality and
performance of voting machines.
Unfortunately, NIST did not receive any funding for these
activities for this fiscal year, and the Administration did not
request any for 2005. I am working with my colleagues to
rectify this situation and provide NIST the money it needs. I
am also encouraged that the Election Assistance Commission,
which was created in HAVA to oversee overall voting reform, is
requesting specific funding in 2005 for these important NIST
activities.
I look forward to hearing from our distinguished panel on
how best to improve the testing and certification process for
voting equipment. And I would like to add that this has been a
project dear to my heart ever since the Florida election of a
few years ago. I do have to say that what happened there was
absolutely no surprise to me whatsoever. Anyone who has been
through the electoral process before knows how easy it is for
mistakes to occur, typically using poll workers who do it only
a few times a year, and in fact, in my very first election,
there was a problem because my opponent's listing and mine were
switched in one polling place. I still won, but there was that
problem, and it could have swung the election.
The--it is very important for us to ensure the integrity of
the voting process, and I must add I am particularly concerned
about the possibilities of fraud, even though those of you
testifying here today obviously are not the sort of persons who
would commit voter fraud, but there is, I believe, an
increasing trend of voter fraud across the country. We managed
to get rid of Tammany Hall and all the other political machines
of the past, where the fraud was quite obvious and deliberate,
but I, in my work on the committees dealing with elections in
the House, I have discovered that there are increasing problems
with fraud in various parts of the country, and so we have to
make sure that all our machines are fraud-proof to the greatest
extent possible.
Having said that, I would like to turn to the Ranking
Member for his opening statement.
[The prepared statement of Chairman Ehlers follows:]
Prepared Statement of Chairman Vernon J. Ehlers
Welcome to today's hearing on how to improve the testing and
certification of voting equipment.
Most of the national attention on voting systems has focused on the
subjects of computer hacking and voter-verifiable paper ballots.
However, recently the New York Times and other organizations have
brought more public attention to the subject of voting machine testing,
the laboratories that test the machines, and the development of
standards used to conduct the tests.
All new models of voting machines sold in the U.S. today are
certified by the National Association of State Elections Directors,
after having passed a series of tests administered by Independent
Testing Authorities, which are private laboratories. These tests are
conducted to ensure that the machines meet certain standards for
environmental tolerances, logic and accuracy, computer security, and
other metrics that make them fit for use in elections. Voting machines
must also be certified by individual States before they can be
purchased by State or local election officials.
However, each election season, a small number of newly-deployed
voting machines fail to perform properly in the field, causing
confusion in the polling places and concerns over the potential loss of
votes. Because these machines have already been tested and certified
against Federal Election Commission standards, these incidents have
raised questions about the reliability of the testing process, the
credibility of standards against which the machines are tested, and the
laboratories that carry out the tests. We must resolve this issue soon
because States are already receiving billions of federal dollars under
the Help America Vote Act (HAVA) to modernize their voting systems. It
is crucial that voting systems be easy to use, accurate, verifiable,
secure, and reliable.
The Science Committee, through HAVA, gave the National Institute of
Standards and Technology (NIST) the role of improving the accreditation
process of the laboratories carrying out the tests, and the standards
against which machines must be tested and certified. Ultimately, NIST's
activities under HAVA will improve the overall quality and performance
of voting machines.
Unfortunately, NIST did not receive any funding for these
activities for this fiscal year and the Administration did not request
any for 2005. I am working with my colleagues to rectify this situation
and provide NIST the money it needs. I am also encouraged that the
Election Assistance Commission, which was created in HAVA to oversee
overall voting reform, is requesting specific funding in 2005 for these
important NIST activities.
I look forward to hearing from our distinguished panel on how best
to improve the testing and certification process for voting equipment.
Mr. Udall. Thank you, Mr. Chairman. Along with the
Chairman, I want to welcome all of you to this hearing today.
As the Chairman mentioned, we are going to address a very
important topic, which is the testing and certification of
voting equipment and systems. And although this sounds like a
set of dry topics, as the Chairman has mentioned, it is
something that we rely upon every day. And I want to provide
you the example I rely on, as I think everybody here does, on
certification from Underwriters Laboratories, or UL, to tell me
that my electric appliances are safe. I may not understand the
standard and the test performed by UL, but I do understand that
the result is a safe and reliable electric appliance. And that
is exactly what we are here to examine today, how to ensure
that voters can depend on the voting equipment that they use to
be safe and reliable.
This isn't an easy task. As the 2000 election pointed out,
this is--was a wakeup call for our country, in that it exposed
many problems with our voting equipment. And I should note that
I think all of us, or most all of us have forgotten that back
in 1988, some 16 years ago, NIST identified problems with punch
card ballots, and recommended that they be retired from
service. Unfortunately, that advice, that prescient advice, was
ignored by the FEC and by State election officials.
Four years after the events in Florida, in the last
Presidential election, very little has been done to assure the
public of the accuracy and integrity of our voting systems. In
fact, with the press coverage of problems with the new
generation of voting equipment, I wouldn't be surprised to find
the public even more skeptical than they were four years ago.
We have mentioned earlier HAVA, H-A-V-A, which passed with
great fanfare, on the critical issue of testing and
certification. The Administration has never requested the funds
for NIST to do its job. And Congress, including this committee,
have been lax, I believe, in its responsibilities, by not
conducting appropriate oversight of the implementation of HAVA.
My biggest concern at this point is that now we are faced
with the sole option of too little, too late. I don't doubt
that with time and money, NIST, the head of the Technical
Guidelines Development Committee, could develop a rigorous set
of standards, testing criteria, and an independent lab testing
system. But we are less than four months from the November
elections. We can't afford to be complacent and hope that the
next election will run smoothly. And I think if there are
problems, we may spend years rebuilding the public's confidence
in our voting system. We need to squarely face the fact that
there have been serious problems with voting equipment deployed
across the country in the past two years.
Let me end by reassuring the witnesses that I am not here
to find blame. I think the blame, if there is blame to be
apportioned, rests squarely with this Administration and this
Congress. What I hope to learn today is that we can do some
things to assure the public that the voting systems that they
use are accurate, reliable, and secure.
So I look forward to the testimony, and I would also add
that we are--Mr. Chairman, we have been joined by Carolyn
Coggins, who will sit on the second panel, who is a resident of
Colorado, and whose business operations are in the 2nd
Congressional District in part. So I want to welcome her in
particular. With that, I would yield back if I have any time
left.
[The prepared statement of Mr. Udall follows:]
Prepared Statement of Representative Mark Udall
Good afternoon. I'd like to welcome everyone to today's hearing.
Today we are going to address a very important topic--the testing
and certification of voting equipment and systems. Although testing and
certification sounds like a dry topic, it is something that we rely
upon everyday.
For instance, I rely on certification from Underwriters
Laboratories, or UL, to tell me that my electric appliance is safe to
use. I may not understand the standard and test performed by UL, but I
do understand that the result is a safe and reliable electric
appliance. That's exactly what we're here to examine today--how to
ensure that voters can depend on the voting equipment they use to be
safe and reliable.
This is no easy task. The 2000 election was a wake-up call for this
country in that it exposed problems with our voting equipment. I should
note that many people have forgotten that back in 1988, NIST identified
problems with punch-card ballots and recommended that they be retired
from service. Unfortunately, NIST's advice was ignored by the FEC and
by state election officials.
So in March 2001, Democratic Members of the Science Committee and I
introduced the first bill that called upon NIST to lead a Commission to
develop standards and testing procedures for election equipment and
systems. This base concept was eventually incorporated into the Help
America Vote Act (HAVA), which brings us to today's hearing.
Four years after the last presidential election, very little has
been done to assure the public of the accuracy and integrity of our
voting systems. In fact, with press coverage of problems with the new
generation of voting equipment, I would not be surprised to find the
public more skeptical than they were four years ago.
Although HAVA was passed with great fanfare, on the critical issue
of testing and certification the Administration has never requested the
funds for NIST to begin to do its job. And Congress--including the
Science Committee--has been lax in its responsibilities by not
conducting appropriate oversight of the implementation of HAVA.
My biggest concern is that we are now faced with the sole option of
``too little, too late.'' I don't doubt that with time and money,
NIST--as the head of the Technical Guidelines Development Committee
(TGDC)--could develop a rigorous set of standards, testing criteria,
and an independent lab testing system.
But we are less than four months from the November elections. We
can't afford to be complacent and hope that the next election will run
smoothly. If there are any problems, we will spend years rebuilding the
public's confidence in our voting systems. We need to squarely face the
fact that there have been serious problems with voting equipment
deployed across the country in the past two years.
I want to reassure the witnesses that I'm not here to find blame--
the blame rests squarely with this Administration and the Congress.
What I hope to learn today is what can be done to assure the public
that the voting systems they use are accurate, reliable and secure.
I look forward to your testimony.
Panel I
Chairman Ehlers. I thank the gentleman for yielding back.
We will begin with the first panel, consisting of one person,
and at this time, I am pleased to introduce my colleague from
New Jersey, my fellow physicist, Representative Rush Holt, who
will provide his comments on this important topic.
As both Rush and I know, physicists are both omniscient and
omni-competent, and so I am looking forward to hearing his
testimony.
Mr. Holt.
STATEMENT OF HON. RUSH HOLT, A REPRESENTATIVE IN CONGRESS FROM
THE STATE OF NEW JERSEY
Mr. Holt. Thank you, Mr. Ehlers, Mr. Udall, Mr. Burgess,
Mr. Gutknecht, Mr. Baird, Mr. Matheson. Thank you for having me
here today. I have some prepared testimony that I would like to
leave with you, but let me give a few summary remarks, if I
may.
We should begin by noting that it was the advent in the use
of computers in voting that precipitated the development of
national standards in the voting systems. The 2001 Caltech MIT
Voting Technology Project reported that the first national
effort to develop standards, a joint project of the then Bureau
of Standards and the General Accounting Office Office of
Federal Elections, focused on the accuracy and security of
computerized voting systems. That was more than 25 years ago.
Now, in the wake of 2002 elections, despite the enactment
of the Help America Vote Act, what are we experiencing? Well,
one after another, incidents or irregularities reported, on
various computer voting systems. 100,000 votes disappearing
into cyberspace, or maybe 100. Xes jumping from one candidates
name to another. You know, 100,000 votes being recorded in a
district where only 19,000 are registered to vote, and only
5,000 turned up that day at the polls. In one jurisdiction
after another election officials are being given pause.
Now, like you, Mr. Ehlers, I am not surprised about this.
As a physicist, I have programmed computers. I understand the
kinds of things that could go wrong, and I am sure you and I,
or any of us, could swap election stories of apparent
irregularities, or close calls, or recounts, or whatever. What
it comes down to today, a fundamental fact, that with the
computer voting devices today, there is a gap between the
casting of the vote and the recording of the vote that makes
the process quite a bit different than what we have been used
to before.
When voting machines were simple, mechanical devices, no
one much cared if the manufacturers helped local officials
select and maintain their equipment, but with more
sophisticated, computerized machines, and the sudden
availability of hundreds of millions of dollars in federal
subsidies, it has raised questions in the minds of members of
the public and election officials.
You know, in November 2003, allegations surfaced to the
effect that uncertified software had been used in electronic
voting systems in at least two California counties. In response
to these allegations, the Secretary of State of California
ordered an independent review be conducted of all voting
systems in the state, and he has subsequently imposed a number
of requirements on future voting in the state, particularly
with regard to electronic or computerized voting machines.
The Caltech MIT Voting Technology Project, to which I
referred earlier, said that, quote, existing standards--the
existing standards process is a step in the right direction,
but it does not cover many of the problems that we have
detected, the project has detected. Important things are not
reviewed currently, including ballot and user interface
designs, auditability, and accessibility. Well, HAVA went a
long way in improving accessibility, and despite a certain
amount of, well, some mention of auditability, I think it
failed to really deal with that question, and it is on that
that I wanted to spend a couple of minutes, because I think it
has important implications for the certification process.
With the computers, the mechanism is not transparent. Any
of us who has programmed computers or has tried to debug
someone else's program knows how easy it is for an error to lie
undetected. A bug can enter the system in various ways,
inadvertently or by malicious hacking. With the difficulty of
monitoring all machines at all times, and the ease with which
changes could be made, the ease with which changes could be
concealed, or as I say, escape detection, it means that there
is a much higher burden, and it is not good enough to just
certify a certain machine, or even a certain class of machines.
What is possible is that these problems could go undetected,
and what concerns me even more than all of these reported
irregularities that we have read about in the papers are the
ones that have gone undetected, that we will never know about,
that will not be subject to a recount because the margin maybe
wasn't so small. There could be errors that we would never know
about, and therefore, the certification process, I think, has
to be designed to get at that, and the only way, I believe,
that we can get at that problem is through auditability. In
other words, a verifiability that is built into the system, and
that is part of the audit process.
I commend the Committee for holding these hearings, and I
think it is important that we ensure that the testing and
certification procedures used to scrutinize and safeguard the
equipment have the highest possible caliber, but it is
different from auditing other machines. It is different from
auditing ATM or bank machines, because it is a secret ballot,
and each ballot is secret, and therefore, it is impossible for
the manufacturer and the vendor, or any election official, to
reconstruct the intention of the voter in that secret booth.
Only the voter knows his or her intention, and only the voter
is in a position to verify whether the vote is recorded the way
that she or he intended. That is why it is important that a
process be built in to the system for verification, and I would
argue that verification must belong to the voter, and I think
the implications for certification are what should be explored
in that context.
[The prepared statement of Mr. Holt follows:]
Prepared Statement of Representative Rush Holt
Distinguished Members of the Committee, thank you for inviting me
to come before you today to address the matter of the testing and
certification of voting systems used in the United States, as well as
the accreditation of independent testing authorities (ITAs). As the
Committee knows, the integrity of the electoral system in the United
States is a matter of great concern to me. Any and all current
shortcomings in existing testing, certification and accreditation
procedures must certainly be addressed, but in addition, the inherent
limits in the protection that may be provided by even the best such
procedures must also be acknowledged.
It should be noted that it was the advent of the use of computers
in voting that precipitated the development of national standards for
voting systems. Prior to the use of computers in the electoral system,
there were no national standards for voting systems, nor, I expect, did
anyone particularly see the need for them. When voting systems were
strictly paper-based, or strictly mechanical, the average citizen--or
election official--could readily understand all there was to know about
the system, and implement it without extensive study or training. With
the advent of computer voting systems, the average citizen--and the
average election official--has become almost completely reliant on the
representations of the system vendors, and the technologists who test
and certify them, that the systems will function properly.
The 2001 Caltech MIT Voting Technology Project reported that the
first national effort to develop standards, a joint project of the
National Bureau of Standards and the General Accounting Office's Office
of Federal Elections, ``focused on the accuracy and security of
computerized voting systems.'' Published in 1975, more than 25 years
ago, the report, entitled ``Effective Use of Computing Technology in
Vote Tallying'' stated that ``one of the basic problems with this
technology was the lack of evaluative standards and testing procedures
for election systems.'' That 1975 report led to Congressional action,
which resulted in the development of voluntary voting system standards
by the Federal Election Commission (FEC) and the National Institute of
Standards and Technology (NIST) in 1984, which were used by the FEC to
promulgate national standards and testing procedures in 1990. Those
1990 voluntary standards covered punch card, optical scan, and direct
recording electronic (DRE) voting systems, and have been adopted by
more than half of the states for use in certifying the voting systems
used in those states.
The Caltech MIT Voting Technology Project continued, however, by
saying that ``[t] existing standards process is a step in the right
direction, but it does not cover many of the problems that we have
detected. . .important things are not reviewed currently, including
ballot and user interface designs, auditability, and accessibility.''
Auditability is, and obviously must be, among the very most critical
aspects of any testing and certification process. The Caltech MIT study
further stated, under the heading ``Create a New Standard for Redundant
Recordings,'' ``[a]ll voting systems should implement multiple
technological means of recording votes. For example, DRE/touchscreen
systems should also produce optical scan ballots. This redundancy
insures that independent audit trails exist post-election, and it helps
insure that if fraud or errors are detected in one technology there
exists an independent way to count the vote without running another
election.''
The Caltech MIT study reported the results of a 12-year study
covering elections between 1988 and 2000. It was the joint effort of
computer scientists, human factors engineers, mechanical engineers and
social scientists; the project organizers met with leading election
officials, researchers and industry representatives. In their joint
statement releasing the report, the Presidents of the California
Institute of Technology and the Massachusetts Institute of Technology
said that in the aftermath of the 2000 election ``America learned that
at the heart of their democratic process, their `can-do' spirit has
`make-do' technology as its central element. For many years, we have
`made do' with this deeply flawed system, but we now know how poorly
these systems function. Until every effort has been made to insure that
each vote will be counted, we will have legitimate concerns about
embarking on another presidential election.''
In the wake of the 2000 election, hundreds, if not thousands, of
the best minds in our country were working on the problem of our flawed
election system. The 2001 Caltech MIT study was released well before
the Help America Vote Act (HAVA) was passed in October 2002. And yet,
HAVA did not mandate what this critical study recommended--standards,
if not actual laws--requiring an independent audit mechanism. Not a
privatized audit mechanism, not a vendor-verified audit mechanism, but
a meaningful, independent audit mechanism.
In the wake of the 2002 election, and despite the enactment of
HAVA, what are we experiencing? One after another incident of
irregularities reported on computer voting systems. 100,000 votes
disappearing into cyberspace, or even just 100. ``X''s jumping from one
candidate's name to another. More than 100,000 votes being recorded in
a district where only 19,000 were registered to vote, and only 5,000
voted. In one jurisdiction after another, election officials are being
given pause.
Despite the fact that national standards have been developed and
implemented and improved upon over the past three decades, and despite
the fact the standards in use today do cover and have been used to
certify DRE and other electronic voting systems, electronic voting
system irregularities have not been prevented. Let's consider the
example of California.
In November 2003, allegations surfaced to the effect that
uncertified software had been used in electronic voting systems in at
least two California counties. In response to those allegations,
Secretary of State Kevin Shelley ordered that an independent audit be
conducted of all voting systems used in the state. In his press release
announcing the audit he said ``[T]o ensure that the integrity of
California's elections process has not been compromised, I will make
certain that all California systems are in compliance with State
security standards.'' The result of the audit--it was discovered that
Diebold Election Systems had used uncertified software in all 17
California counties in which it's electronic voting equipment was used.
Fourteen of those counties had used software that had been federally
qualified, but not certified by State authorities. The other three used
software that had not been certified at the State nor qualified at the
federal level. In April 2004, Secretary of State Shelley banned the use
of touch screen systems in four counties and decertified all touch
screen systems in California for use unless and until those systems
were brought into compliance with additional security measures. Kevin
Shelley's Decertification Order, and his recently release standards for
Accessible Voter Verified Paper Audit Trail Systems, are attached as
Appendix A.
California is in a sense an extreme example, but perhaps only
because Secretary of State Shelley acted upon the first indication of a
problem, and discovered and confronted those problems. But again,
reports of irregularities on electronic voting systems abound, and have
occurred in states from one shore of this country to the other. In how
many other states might similar deficiencies in testing or
certification be found? As we all know, the voting systems Secretary of
State Shelley decertified in 2004 had just been used in the recall
election in California in 2003. And those touch screen systems were not
independently unauditable. Three decades of work developing and fine
tuning national standards did not protect voters in the State of
California, and have not necessarily protected voters elsewhere. Were
those three decades of effort all for naught? Of course not. Were the
standards developed worthless? Of course not. But we can plainly see by
this one example that perfecting testing and certification procedures
is not, nor will it ever be, the end of the inquiry.
Johns Hopkins Computer Scientist Aviel Rubin, co-author of the
analysis released in the summer of 2003 that described ``stunning,
stunning'' flaws in the software used in Maryland's touch screen voting
systems, has issued a challenge, entitled ``Can a Voting Machine that
is Rigged for a Particular Candidate Pass Certification?'' In it he
says ``[p]roponents of DREs argue that the ITA [Independent Testing
Authorities] process would catch any attempts to manipulate the
results. They argue that Trojan horse programs would have to have
magical properties and that they would be detected. They further argue
that techniques such as parallel testing, where machines are selected
at random and elections are run on them on election day where they are
checked for accuracy, ensure that no such rigging is possible. Security
experts do not buy these arguments.''
In short, Professor Rubin proposes that a team of computer security
experts be given access to one of the major vendors, full authority to
produce a rigged machine, and that that machine then be presented to an
ITAs that is unaware of the challenge, along with all the other
machines, to determine whether the ITA could discover the rigging. If
not, that would demonstrate that voting system vendor's employee could
rig an election. Would any of the ITAs accept this challenge? Would any
vendor? I think it would be a worthwhile endeavor, although, as
Professor Rubin points out, the testing and certification process is
analogous to airline security procedures--``just like successfully
catching an agent with a concealed weapon at the airport does not mean
the next guy won't get through,'' even if the ITA in question discovers
the rigged machine in question, that doesn't mean the next rigged
machine won't get through.
Even in the absence of such a challenge, the Committee should leave
no stone unturned in determining exactly how the Diebold systems used
in California, Maryland other jurisdictions have passed muster with the
ITA's in question. In every instance in which an irregularity has been
reported in connection with the use of any electronic voting system,
the same inquiry should be made. In every instance, the Committee
should ask, are testing and certification procedures capable of being
implemented with perfection? Will they find every flawed or rigged
machine? In the wake of September 11, despite the obviously heightened
security at our airports, has every single weapon sought to be smuggled
onto an aircraft, has every mechanical malfunction, been found before
take-off?
It is also of critical importance to note that the ``revolving
door'' for employees between vendors, testers and certifiers perhaps
ought to be closed, permanently. Going back to California, take for
example the recent report in the San Francisco area online periodical
the Contra Costa Times:
``Critics say. . .close, often invisible, bonds link election
officials to the equipment companies they are supposed to
regulate. When voting machines were simple mechanical devices,
no one much cared if manufacturers helped local officials
select and maintain their equipment. But a switch to
sophisticated computerized machines, and the sudden
availability of hundreds of millions of dollars in federal
subsidies, has raised questions about counties' dependence on
private firms. While a revolving door between government
service and private-sector jobs is common, some observers argue
that such cozy familiarity has led public officials to overlook
flaws in controversial electronic voting systems, putting
elections at risk.''
Attached as Appendix B to my statement is a copy of an editorial
published in the New York Times on June 13, 2004, entitled ``Gambling
on Voting,'' which makes the point that slot machines are subject to
more rigorous testing and certification procedures than voting systems.
I would like to commend the Committee for holding this hearing, and
for taking action to ensure that the testing and certification
procedures used to scrutinize and safeguard the equipment used in our
elections are of the highest possible caliber. But I would at the same
time urge the Committee to recommend, as was recommended by the Caltech
MIT study, that DRE/touch screen systems produce optical scan or other
paper ballots, so that an independent audit trail will exist in each
election, and help insure that if fraud or errors are detected there
will be an independent way to count the vote short of running another
election. We most definitely ``can-do'' this, and ``making-do'' without
it does nothing short of placing this very democracy at risk.
Appendix B
Gambling on Voting
Published in the New York Times, June 13, 2004
If election officials want to convince voters that electronic
voting can be trusted, they should be willing to make it at least as
secure as slot machines. To appreciate how poor the oversight on voting
systems is, it's useful to look at the way Nevada systematically
ensures that electronic gambling machines in Las Vegas operate honestly
and accurately. Electronic voting, by comparison, is rife with lax
procedures, security risks and conflicts of interest.
On a trip last week to the Nevada Gaming Control Board laboratory,
in a State office building off the Las Vegas Strip, we found testing
and enforcement mechanisms that go far beyond what is required for
electronic voting. Among the ways gamblers are more protected than
voters:
1. The State has access to all gambling software. The Gaming
Control Board has copies on file of every piece of gambling
device software currently being used, and an archive going back
years. It is illegal for casinos to use software not on file.
Electronic voting machine makers, by contrast, say their
software is a trade secret, and have resisted sharing it with
the states that buy their machines.
2. The software on gambling machines is constantly being spot-
checked. Board inspectors show up unannounced at casinos with
devices that let them compare the computer chip in a slot
machine to the one on file. If there is a discrepancy, the
machine is shut down, and investigated. This sort of spot-
checking is not required for electronic voting. A surreptitious
software change on a voting machine would be far less likely to
be detected.
3. There are meticulous, constantly d standards for gambling
machines. When we arrived at the Gaming Control Board lab, a
man was firing a stun gun at a slot machine. The machine must
work when subjected to a 20,000-volt shock, one of an array of
rules intended to cover anything that can possibly go wrong.
Nevada adopted new standards in May 2003, but to keep pace with
fast-changing technology, it is adding new ones this month.
Voting machine standards are out of date and inadequate.
Machines are still tested with standards from 2002 that have
gaping security holes. Nevertheless, election officials have
rushed to spend hundreds of millions of dollars to buy them.
4. Manufacturers are intensively scrutinized before they are
licensed to sell gambling software or hardware. A company that
wants to make slot machines must submit to a background check
of six months or more, similar to the kind done on casino
operators. It must register its employees with the Gaming
Control Board, which investigates their backgrounds and
criminal records.
When it comes to voting machine manufacturers, all a
company needs to do to enter the field is persuade an election
official to buy its equipment. There is no way for voters to
know that the software on their machines was not written by
programmers with fraud convictions, or close ties to political
parties or candidates.
5. The lab that certifies gambling equipment has an arms-
length relationship with the manufacturers it polices, and is
open to inquiries from the public. The Nevada Gaming Control
Board lab is a State agency, whose employees are paid by the
taxpayers. The fees the lab takes in go to the State's general
fund. It invites members of the public who have questions about
its work to call or e-mail.
The federal labs that certify voting equipment are
profit-making companies. They are chosen and paid by voting
machine companies, a glaring conflict of interest. The voters
and their elected representatives have no way of knowing how
the testing is done, or that the manufacturers are not applying
undue pressure to have flawed equipment approved. Wyle
Laboratories, one of the largest testers of voting machines,
does not answer questions about its voting machine work.
6, When there is a dispute about a machine, a gambler has a
right to an immediate investigation. When a gambler believes a
slot machine has cheated him, the casino is required to contact
the Gaming Control Board, which has investigators on call
around the clock. Investigators can open up machines to inspect
their internal workings, and their records of recent gambling
outcomes. If voters believe a voting machine has manipulated
their votes, in most cases their only recourse is to call a
board of elections number, which may well be busy, to lodge a
complaint that may or may not be investigated.
Election officials say their electronic voting systems are the very
best. But the truth is, gamblers are getting the best technology, and
voters are being given systems that are cheap and untrustworthy by
comparison. There are many questions yet to be resolved about
electronic voting, but one thing is clear: a vote for president should
be at least as secure as a 25-cent bet in Las Vegas.
Chairman Ehlers. Thank you, Mr. Holt. As you well know,
normally Members are not questioned by their colleagues,
because we have ample opportunities to discuss it with you.
I would just add one quick comment to illustrate the
difficulty of what you are referring to, and that is that I
have also programmed computers many times--it is even possible
to program the computer to present to the voter a verifiable
notification of some sort, and yet record a different result in
the memory, and so that--even that verification has
difficulties. So, we have a lot of problems to deal with, but
thank you very much for your testimony. I appreciate--you
certainly----
Mr. Boehlert. Mr. Chairman, is this the witness and the
Chair 100 percent of the House physicists caucus?
Mr. Holt. This you see before you the bipartisan physics
caucus of the 108th Congress.
Chairman Ehlers. And as soon as we can find a phone booth
for the straw court, we will have our office.
Mr. Boehlert. Well, thank you, Dr. Ehlers, and thank you,
Dr. Holt.
Chairman Ehlers. Thank you. Thank you very much. Thank you
for coming, Mr. Holt.
Mr. Baird. Mr. Chairman, if I may. Mr. Chairman. I would
just like to express my profound respect and appreciation for
the gentleman's work.
Chairman Ehlers. Yes.
Mr. Baird. I can tell you, I receive letters and phone
calls from constituents who are profoundly concerned about
this, and there are no PACs, there are no political
contributions that go with is. This is a Member of the Congress
fighting for a fundamental principle of one person, one vote,
and that votes be fairly counted, and I have a tremendous
admiration and gratitude for the gentleman, and we all owe him,
as Americans, a debt of appreciation.
Mr. Holt. Thank you. I thank Mr. Baird. And I would say,
Mr. Udall said that this may seem to be a dry topic. Let me
tell you that this is a topic that has excited hundreds of
thousands, if not millions of Americans. Since four years ago,
I think we have had an education here in the United States
about voting, and it has excited many people, and I am
certainly pleased to see that so many Americans believe their
vote is sacred, and they are taking steps to see that their
votes are protected.
Chairman Ehlers. I thank you for your comments, your
testimony, and let me assure you this subcommittee shares that.
That is why we wrote the legislation two years ago, and wish it
had been even stronger in the final version, and fully funded.
Thank you for being here.
If there is no objection, all additional opening statements
submitted by the Subcommittee Members will be added to the
record. Without objection, so ordered.
We will now ask the second panel to take their places at
the table. At this time, I would like to introduce our second
panel of witnesses. Mr. Tom Wilkey is the Chair of the National
Association of State Election Directors, also known as NASED,
and he is Chair of the Independent Testing Authority Committee,
I believe. Ms. Carolyn Coggins is the Director of ITA Services
at SysTest Labs, an Independent Testing Authority for software,
based in Boulder, Colorado. Dr. Michael Shamos is a Professor
of Computer Science at Carnegie Mellon University. And a
familiar face, Dr. Hratch Semerjian, is the Acting Director of
the National Institute of Standards and Technology.
As our witnesses presumably have already been told, you
will each have five minutes to offer your spoken testimony. If
your written testimony is longer than that, we ask you to
summarize it within the five minute time periods. And after you
complete your five minutes, then we will each question you, and
each of us will have five minutes to do so. The timer, in case
you haven't been told, will display green during the first four
minutes of your talk, yellow during the last minute, and red,
all sorts of exciting things happen. So try to wrap up before
it turns red.
At this point, we will open our first round. Mr. Wilkey,
you may proceed. Would you please turn on your microphone?
Panel II
STATEMENT OF MR. THOMAS R. WILKEY, CHAIR, INDEPENDENT TESTING
AUTHORITY (ITA) COMMITTEE, NATIONAL ASSOCIATION OF STATE
ELECTION DIRECTORS
Mr. Wilkey. Thank you, Mr. Chairman, and I am Thomas
Wilkey. I am the former Executive Director of the New York
State Board of Elections, having retired from that position
last August. However, I continue to chair the NASED Voting
Systems Board, and I am pleased to appear before you today to
discuss the work that has been done by the National Association
of State Election Directors, NASED, with regards to the
selection of Independent Test Authorities, and its program to
encourage states to adopt the federal voting system standards,
and to utilize test reports which have been issued by these
ITAs.
My involvement in the development of the Federal Voting
System Standards began several years before NASED became an
official organization. Several of my colleagues worked with me
on an advisory panel in assisting the FEC in the development of
the first set of voluntary standards in 1990. These standards
were developed over a 5-year period, between 1985 and 1990, and
the initial drafts were contracted to the late Robert Naegele
of Granite Creek Technology, who had for many years worked in
the area of voting system testing for the State of California.
Following the adoption of the standards in 1990, it became
evidence that states were not adopting these standards. Because
the Federal Government was not interested in the selection of
qualified Independent Testing Authorities, the standards were
destined to lie on a shelf collecting dust, and the hard work
of developing them would have been in vain. At that time, NASED
was formed, and at one of their earlier meetings, discussions
took place to try to develop a program that would encourage
member States to adopt the standards, select and qualify
testing laboratories that would not only test equipment and
software, but provide reports to states which needed them as a
component of their own certification process.
Identifying laboratories qualified to do this testing, and
by having member States participate in this program, vendors
would need only go to one or two laboratories to have
comprehensive testing completed, thus saving time and money by
avoiding duplicate testing in each state.
Needless to say, our plans did move quickly in those early
years, as it was difficult to find laboratories that were
willing to do the work, given the economic realities of the
times, and a somewhat less than perfect fit into their overall
business plans.
At the outset, a handbook was developed by Bob Naegele,
which was utilized as a checklist for prospective laboratories,
outlining the necessary personnel and equipment to do the work.
This handbook was revised several years ago, and a copy has
been provided to the Committee.
NASED was very pleased that Wylie Laboratories in
Huntsville, Alabama stepped up to the plate to become our first
ITA. Their expertise in the testing of hardware and software
for NASA and other U.S. government agencies is internationally
recognized, and they have continued to this day to work with us
toward the qualification of numerous voting systems in use
throughout the country.
Over the years, Wylie has been joined by Ciber, Inc. of
Huntsville, Alabama, and SysTest Laboratories of Denver,
Colorado, who have been qualified as software laboratories.
SysTest has recently been qualified to test hardware as well,
and joins us today in our presentation to the Committee.
Over the years, while we have encouraged other laboratories
to join this project, the consideration of the sheer volume of
business and the negative publicity of late caused most others
to decline this opportunity. We continue to encourage others to
look at this program as we transition this program to the
Election Assistance Commission and to NIST in the next several
months.
NASED's involvement in the development of the 2002
standards was twofold. In the late 1990's, NASED requested the
FEC provide funding for revisions that NASED thought were
needed, based on the testing and evaluation that had been done
over the past several years, and the fact that standards were
now nearly 10 years old. New technology and issues not
considered in the original standards needed to be addressed.
The FEC acted on our request and authorized a contract with
Mantec, Incorporated, to conduct a needs assessment and
evaluation to determine if the project indeed needed to be
done, and if so, the scope of the work to be done.
As a result of the needs assessment, the FEC awarded a
contract to AMS Consulting to draft the revised standards and
prepare them for a series of public comment periods required by
federal law. NASED's contribution to the project included the
involvement of NASED's Voting Systems Standards Board, as
members of an ad hoc advisory group.
It is important for the Committee to understand several
important facts as they relate to NASED's role in the selection
of ITAs.
First, there is a misconception that NASED certifies voting
equipment, or voting systems. NASED's role is solely limited to
review and qualify perspective ITAs, and provide for the review
of reports by its technical subcommittee before they are sent
to the vendors, and to, ultimately, State ITAs and others
designated by states to receive and review them.
NASED, through its Secretariat, who for many years, had
been the Election Center, had placed on its web sites
information regarding systems which had been qualified under
the standards, so that States and local jurisdictions,
particularly those who had no formal certification process,
could know that a system had met the voluntary federal voting
system requirements. This secretarial role was turned over to
the Election Assistance Commission in November of 2003.
Member of NASED's Voting System Board served on a voluntary
basis, receiving no salary or compensation, and in many cases,
traveling at their own expense to intense sessions held on
weekdays or on weekends in Huntsville, or in other areas across
the country. The Election Center received no compensation
whatsoever, except for reimbursement of room expenses. The sum
and substance of this was that this program operated on a
purely voluntary basis without any funding from the Federal
Government, nor, with the exception of the travel expenses for
some members, without any State or local funding.
NASED has worked closely since January of 2003 with NIST on
the transition of this program to the Technical Guidelines
Development Committee, under the Election Assistance
Commission. Regular meetings will hopefully provide for a
smooth transition and eventual reevaluation of ITAs by the EAC
and NIST, and the consideration of other issues which we have
dealt with as part of our program.
NASED is proud of what we have tried to accomplish. We know
that there have been weaknesses in the program, but that it is
finally the day to get the day to day full-time attention that
is needed under the EAC and NIST.
Voting System Board members, election directors, and
dedicated experts in the field of technology have given
thousands of hours of their personal time and talent to this
program, because they wanted to make a difference.
Together, colleagues rose to meet a tremendous challenge,
with a single goal in mind, to help ensure the integrity of
America's voting systems and processes. Absent these bold
motives almost 15 years ago, recent scenarios would have been
significantly worse.
Many people have said to me over the past several months
that given the current media attention on voting systems, it
would have been understandable had we thrown in the towel on
this critical issue. But looking back, I can say with
confidence that we can be proud of what we accomplished, as we
tried to do something rather than nothing at all.
Thank you for the opportunity to testify today and for your
interest in this matter.
[The prepared statement of Mr. Wilkey follows:]
Prepared Statement of Thomas R. Wilkey
Mr. Chairman and Members of the Committee;
I am pleased to have the opportunity to appear before you today to
discuss the work that has been done by the National Association of
State Election Directors (NASED) with regards to the selection of
Independent Test Authorities (ITA) and it's program to encourage states
to adopt Federal Voting System Standards and utilize test reports which
have been issued by these ITAs.
My involvement in the development of the Federal Voting System
Standards began several years before NASED became an official
organization. Several of my colleagues worked with me on an Advisory
panel in assisting the FEC in the development of the first set of
voluntary Standards in 1990.
These standards were developed over a five-year period (1985-1990)
and the initial drafts were contracted to the late Robert Naegele of
Granite Creek Technology who had for many years, worked in the area of
voting system testing for the State of California.
Following the adoption of the standards in 1990, it became evident
that States were not adopting the standards. Because the Federal
Government was not interested in the selection of qualified Independent
Testing Authorities, the standards were destined to lie on a shelf,
collecting dust and the hard work of developing them would have been in
vain.
At that time NASED was formed, and at one of their earlier
meetings, discussions took place to try to develop a program that would
encourage member States to adopt the standards, select and qualify
testing laboratories that would not only test equipment and software,
but provide reports to states which needed them as a component of their
own certification process.
By identifying laboratories qualified to do this testing and by
having member States participate in the program, vendors would only
need to go to one or two laboratories to have comprehensive testing
completed, thus saving time and money by avoiding duplicative testing
in each state.
Needless to say our plans did not move quickly in these early
years, as it was difficult to find laboratories that were willing to do
the work, given the economic realities of the times, and a somewhat
less then perfect fit into their overall business plans.
At the outset, a handbook was developed by Bob Naegele which was
utilized as a check list for prospective laboratories, outlining the
necessary personnel and equipment to do the work. The handbook was
revised several years ago and a copy has been provided to the
committee. Mr. Steve Freeman, who joins me on the panel today is here
to briefly outline the steps taken to qualify a test laboratory as he
has been involved in this task for NASED and has received training
under the National Institute of Standards and Technology (NIST) to do
so in future evaluations.
NASED was very pleased that Wylie Laboratories in Huntsville,
Alabama stepped up to the plate to become our first ITA. Their
expertise in the testing of hardware and software for NASA and other
U.S. Government agencies is internationally recognized and they have
continued to this day to work with us toward the qualification of
numerous voting systems in use throughout the country.
Over the years, Wylie has been joined by Ciber Inc. of Huntsville,
Alabama and SysTest Laboratories of Denver, CO. who have been qualified
as software laboratories. SysTest has recently been qualified to test
hardware as well and joins us today in our presentation to the
Committee.
Over the years, while we have encouraged other laboratories to join
this project, their consideration of the sheer volume of business and
the negative publicity of late, caused most others to decline this
opportunity. We continue to encourage others to look at this program
and as we transition this program to the Election Assistance Commission
and to NIST in the next several months we know that they will be
reaching out to all interested parties as well.
NASED's involvement in the development of the 2002 standards was
two-fold:
In the late 1990's NASED requested that the FEC provide funding for
the revisions that NASED thought were needed, based on the testing and
evaluation that had been done over the past several years and the fact
that the standards were now nearly ten years old. New technology and
issues, not considered in the original standards needed to be
addressed.
The FEC acted on our request and authorized a contract with Mantec
Inc. to conduct a needs assessment and evaluation to determine if the
project indeed needed to be done and if so, the scope of the work to be
done.
As a result of the needs assessment, the FEC awarded a contract to
AMS Consulting to draft the revised standards and prepare them for a
series of public comment periods required by federal law. NASED's
contribution to the project included the involvement of NASED's Voting
System Standards Board as members of an ad hoc advisory group to review
the document and make suggestions for improvement. The 2002 Standards
were released in the fall of that year.
It is important for this Committee to understand several important
facts as they relate to NASED's role in the selection of ITAs, the
development of standards, and our overall program.
First, there is a misconception that NASED ``certifies'' voting
systems. NASED's role is solely to review and qualify prospective ITAs
and provide for the review of reports by it's technical subcommittee,
before they are sent to the vendors and ultimately to State ITAs and
others designated by the states to receive and review same.
NASED, through it's secretariat, who for many years has been the
Election Center, has placed on its web sites, information regarding
systems which had been qualified under the standards, so that States
and local jurisdictions, particularly those who had no formal
certification process, can know that a system has met the voluntary
federal voting system requirements. This secretariat role was turned
over to the Election Assistance Commission in November 2003.
Members of NASED's voting system board served on a voluntary basis,
receiving no salary or compensation and in many cases traveled at their
own expense to attend sessions held on weekdays as well as weekends in
Huntsville and the Election Center served as our Secretariat did so
without any compensation, except for the reimbursement of meeting room
expenses. The sum and substance of this was that this program operated
on a purely voluntary basis without any funding from the Federal
Government, nor with the exception of travel expenses for some members,
without any State or local funding.
NASED has worked closely since January of 2003 with NIST on the
transition of this program to the Technical Guidelines Development
Committee under the Election Assistance Commission. Regular meetings
will hopefully provided for a smooth transition, and the eventual re-
evaluation of ITAs by the EAC and NIST, and the consideration of other
issues which we have dealt with as part of our program.
NASED is proud of what we have tried to accomplish. We know there
have been weaknesses in the program, but that it will finally get the
day-to-day full-time attention that it needed but never realized under
the voluntary nature of our program.
Voting System Board members, election directors and dedicated
experts in the field of technology have given thousands of hours of
their personal time and talent to this program because they wanted to
make a difference. Together, colleagues rose to meet a tremendous
challenge, with a single goal in mind--to help ensure the integrity of
America's voting systems and processes. Absent those bold motives
almost 15 years ago, recent scenarios would have been significantly
worse.
Many people have said to me over the past several months, that
given the current media attention on voting systems, it would have been
understandable had we thrown in the towel on this critical issue. But
looking back, I can say with confidence that we can be proud of what we
accomplished, as we try to do something rather than nothing at all.
Thank you for the opportunity to testify today and for your
interest in this important matter.
Chairman Ehlers. Thank you for your comments. Ms. Coggins.
STATEMENT OF MS. CAROLYN E. COGGINS, DIRECTOR, ITA SERVICES AT
SYSTEST LABS
Ms. Coggins. Mr. Chairman and Members of the Committee, I
am Carolyn Coggins from SysTest Labs. We are the only combined
hardware and software NASED Independent Test Authority. Thank
you for inviting us here today to speak about qualification
testing.
NASED qualification testing is, to the 2002 FEC Voting
System Standards. All testing conforms to two VSS processes.
This first is the physical configuration audit. It addresses
the source code, software, hardware configuration, and
documentation.
The functional configuration audit addresses all testing.
SysTest has created a test methodology incorporating physical
and functional configuration audit-specific reviews and tests.
Standard templates are customized for each unique voting
system, but the overall process is always the same for every
voting system.
To have confidence in the voting system, one needs to have
confidence in the testing. NASED qualification testing is the
second level of four levels of testing identified by the Voting
System Standards. The first level of testing is vendor testing.
The vendor tests to the design requirements. The second level
is qualification testing. The ITAs examine the vendor's testing
for adequacy and completeness. We run a standard set of end-to-
end functional tests customized for the specific voting system
to ensure that it meets the VSS. We also test for any
additional functionality that is non-VSS required.
Qualification testing means that the hardware, software,
and all documentation of the voting system have been defined,
reviewed, and tested for conformance with the requirements of
the Voting System Standards. It means the voting system
contains a method to create elections, provide a ballot, record
votes, report tallies, and produce an audit trail. It means
voting is secret, accurate, and reliable. It means all code
used in testing has been reviewed by an ITA, and it means that
the documentation required to help jurisdictions run elections
is accurate and sufficient.
The qualification testing does not mean that testing has
been sufficient to confirm that voting systems meet the
specific laws of all states, or for that matter, any State.
This responsibility falls to the third level of testing, State
certification. Qualification testing also does not mean that
the voting system the vendor delivers is exactly the system
that was qualified or certified. This aspect falls to the
fourth level of testing, local acceptance testing.
All four levels are essential to the voting process. We
suggest that the 1990 Voting System Standard implementation
plan be used as a baseline guide. While never fully
implemented, it contains an excellent structure for issues
associated with all levels of voting testing. Additionally, we
recommend that the new EAC standards define specific reporting
methodologies and poll worker usability, to assist the States
and local jurisdiction to understand and use ITA qualification
reports and voting systems themselves.
To ensure confidence in testing, you have to have
confidence in the test labs. Currently, environmental testing
and all functional software and hardware testing of the polling
place equipment is assigned to the hardware ITA. The functional
testing of ballot preparation and the central count
functionality, and then the integration of end-to-end testing
is assigned to the software ITA.
As technology has evolved, we feel this scope should be
reexamined, because polling place software cannot be fully
tested without integrating ballot preparation and counting
software. Integration testing repeats much of the polling place
functional testing. New voting systems today tend not to have
separate applications that neatly divide these functions.
Vendors must artificially divide code in order to conform to
current lab assignments. Lastly, polling place issues that are
found in end-to-end testing by a software ITA must go back to
the hardware ITA for code review and functional testing. Then,
the hardware ITA must send the code back to the software ITA to
rerun their tests.
The Subcommittee has asked us to provide suggestions for
future accreditation of labs. We would suggest that the
accrediting of primary labs responsible for all hardware and
software testing. We would also suggest that primary labs may
have qualified subcontractors to perform environmental testing,
but they must demonstrate their ability to monitor all
subcontractor work.
Lastly, to ensure confidence in voting systems testing and
labs, one must have confidence in the standards. Criticism of
the 2002 standards generally is focused on security in terms of
active attack code such as backdoors. When you look at security
from a broader view, the requirements of the VSS are more
comprehensive. Testing for accuracy and reliability helps
secure the vote. Testing the functional requirements dealing
with election creation, voting, counting, and auditing helps
secure the vote. Documenting the processes to ensure physical
security and detect intrusion help secure the vote.
In terms of active attack code, the VSS supplies some
detail, and there are some sections that provide very wide
latitude to the labs. These sections give the individual labs a
great deal of discretion, but it does not provide the detail
consistency across all ITAs. The role of the ITA is to hold the
vendor's feet to the fire, but it is not to build the fire.
HAVA tasks the EAC in this to address this issue in the future.
The Subcommittee has asked us to provide suggestions for
changes to improve the process before the 2006 Election. The
2002 VSS implementation plan has a process for issuing
clarification bulletins. We would suggest a NASED, EAC, and
NIST transition clarification bulletin addressing any
significant issues.
Thank you for the opportunity to speak here, and we thank
you.
[The prepared statement of Ms. Coggins follows:]
Prepared Statement of Carolyn E. Coggins
SysTest Labs is pleased to provide the Environment, Technology, and
Standards Subcommittee with information about ITA (Independent Testing
Authority) Qualification Testing of Voting Systems for the National
Association of State Election Directors (NASED) to the Federal Election
Commission (FEC) Voting System Standards (VSS).
Three labs currently provide NASED Qualification Testing. All of
the labs test to the VSS, but each has their own methods. Our comments
here reflect the methods used by SysTest Labs.
My discussion shall identify:
SysTest Labs' qualifications and accreditation as an
ITA;
The standards, in addition to the VSS, that govern
qualification testing;
How the Voting System Qualification Test process is
defined in the VSS;
How SysTest Labs implements the VSS Voting System
Qualification Test process;
How SysTest Labs maintains quality and manage process
improvement; and
Observations and recommendations regarding lab
accreditation, the VSS and qualification testing.
Accreditation as a NASED Qualification ITA
SysTest Labs is full service laboratory specializing in all areas
of software testing. Our work ranges from Independent Verification and
Validation for software development efforts of State unemployment
insurance systems to large and complex software laboratory testing for
major telecommunication companies to web site performance testing for
major retailers to software test staff augmentation. SysTest Labs has
successfully completed over 500 software testing or quality assurance
projects for over 250 clients worldwide. Regardless of the test effort,
all aspects of our quality program, test methodology and test engineer
training are guided by Institute of Electrical and Electronic Engineers
(IEEE) standards and the SysTest Labs quality procedures.
In order to become a software and hardware ITA, SysTest Labs had to
apply to NASED and then be audited by the NASED Technical Committee. To
my knowledge, we are the only lab that has sought and been awarded both
software and hardware accreditation, to become a full service ITA. We
initially applied and qualified as a software ITA in 2001. We recently
granted acceptance as a hardware ITA. Our hardware ITA status is
provisional, i.e., our audit was successfully completed, NASED has
recommended accreditation and our initial hardware qualification test
effort will be monitored by a NASED auditor.
Quality Program, Test Standards and Test Methods
The NASED audit process requires that we provide documentation and
demonstrate our quality program. In addition, we have had to provide
documentation and demonstrate our test methodology and processes for
NASED Qualification Testing of voting systems. While the requirements
we test to are governed by the standards, we must define the method of
testing and processes to ensure the consistency, adequacy, accuracy,
and overall quality of our NASED Qualification Testing.
While the 2002 Federal Election Commission Voting System Standard
is the primary standard, there are a number of other standards used in
our voting system testing. The VSS itself incorporates a number of
other standards, which are included in NASED Qualification Testing (see
Volume 1 Applicable Documents). The primary standards we use in NASED
ITA Qualification Testing are:
Federal Election Commission
Federal Election Commission Voting System Standards,
Volume I Performance Standards and Volume II Test Standards,
April 2002.
National Association of State Election Directors
NASED Accreditation of Independent Testing
Authorities for Voting System Qualification Testing, NASED
Program Handbook NHDBK 9201, a National Association of State
Election Directors (NASED), May 1st, 1992.
NASED Voting System Standards Board Technical Guide
#1, FEC VSS Volume I, Section 2.2.7.2, Color and Contrast
Adjustment
NASED Voting System Standards Board Technical Guide
#2, Clarification of Requirements and Test Criteria for Multi-
language Ballot Displays and Accessibility.
Institute of Electrical and Electronics Engineers
IEEE Standard for Software Quality Assurance Plans
IEEE STD 730-1998
IEEE Standard for Software Configuration Management
Plans IEEE STD 828-1998
IEEE Standard for Software Test Documentation IEEE
STD 829-1998
IEEE Recommended Practice for Software Requirements
Specifications IEEE STD 830-1998
IEEE Standard for Software Unit Testing IEEE STD
1008-1987
IEEE Standard for Software Verification and
Validation IEEE STD 1012-1998.
Federal Regulations
Code of Federal Regulations, Title 20, Part 1910,
Occupational Safety and Health Act
Code of Federal Regulations, Title 36, Part 1194,
Architectural and Transportation Barriers Compliance Board,
Electronic and Information Technology Standards--Final Rule
Code of Federal Regulations, Title 47, Parts 15 and
18, Rules and Regulations of the Federal Communications
Commission
Code of Federal Regulations, Title 47, Part 15,
``Radio Frequency Devices,'' Subpart J, ``Computing Devices,''
Rules and Regulations of the Federal Communications Commission.
American National Standards Institute
ANSI C63.4 Methods of Measurement of Radio-Noise
Emissions from Low-Voltage Electrical and Electronic Equipment
in the Range of 9Khz to 40 GHz
ANSI C63.19 American National Standard for Methods
of Measurement of Compatibility between Wireless Communication
Devices and Hearing Aids.
International Electro-technical Commission.
Electromagnetic Compatibility (EMC) Part 4: Testing and
Measurement Techniques
IEC 61000-4-2 (1995-01) Section 2
Electrostatic Discharge Immunity Test (Basic EMC
publication)
IEC 61000-4-3 (1996) Section 3 Radiated
Radio-Frequency Electromagnetic Field Immunity Test
IEC 61000-4-4 (1995-01) Section 4 Electrical
Fast Transient/Burst Immunity Test
IEC 61000-4-5 (1995-02) Section 5 Surge
Immunity Test
IEC 61000-4-6 (1996-04) Section 6 Immunity to
Conducted Disturbances Induced by Radio-Frequency
Fields
IEC 61000-4-8 (1993-06) Section 8 Power-
Frequency Magnetic Field Immunity Test. (Basic EMC
publication)
IEC 61000-4-11 (1994-06) Section 11. Voltage
Dips, Short Interruptions and Voltage Variations
Immunity Tests.
Electromagnetic compatibility (EMC) Part 5-7: Installation and
mitigation guidelines
IEC 61000-5-7 Ed. 1.0 b: 2001 Degrees of
protection provided by enclosures against
electromagnetic disturbances.
Military Standards
MIL-STD-810D (2) Environmental Test Methods and
Engineering Guidelines.
NASED Qualification Testing of Voting Systems ITA Process
SysTest Labs performs qualification testing in conformance with the
two processes required in the 2002 VSS. The results from Qualification
reviews and testing are documented throughout the process (ITA
documentation of testing in red):
Physical Configuration Audit (PCA in blue) addresses
the physical aspects of the voting system, including:
Review of the Technical Data Package (TDP)
documentation
Verification of the configuration of the hardware
and software
Identification of the code to review
Source Code review
Observing the building of the executable from the
reviewed source code.
Functional Configuration Audit (FCA in green)
addresses the functional aspects of the voting system,
including:
Review of all testing performed by the vendor
Test planning
Test Case preparation and/or customization of
Standard Test Cases
Test execution.
While the VSS outlines the overall PCA and FCA process, SysTest
Labs has defined specific processes for each area of testing or review
to ensure a consistent, repeatable test methodology. These processes
include specific review and test templates that have been prepared in
conformance with the VSS, IEEE standards, NASED accreditation policies
and SysTest Labs quality procedures. Each voting system is unique.
While qualification testing must be customized for the unique
requirements of each specific voting system, the overall process is
exactly the same for every voting system.
The VSS does not designate software and hardware ITA
responsibilities. These responsibilities are assigned by NASED
accreditation policies. The processes documented here note processes or
test approaches that can be applied to either the software or hardware
ITA.
PCA Technical Data Package (TDP) Review: The TDP is
reviewed to confirm required documentation is present, conforms
in content/format and is sufficient to install, validate,
operate, maintain the voting system and establish the system
hardware baseline associated with the software baseline.
Results of the review are provided to the vendor in a Pre-
qualification Report.
PCA Source Code Review: The source code is reviewed
for:
Maintainability--including the naming, coding and
comment conventions, adherence to coding standards and
clear commenting.
Control Constructs--to determine the logic flow
utilizes standard constructions of the development
language, its used consistently, the logic structure
isn't overly complex and there's an acceptable use of
error handlers. Where possible automated tools are
used.
Modularity--confirming each module has a testable
single function, unique name, single entry/exit,
contains error handling and an acceptable module size.
Security and Integrity of the Code--including
controls to prevent deliberate or accidental attempts
to replace code such as unbounded arrays or strings,
including buffers to more data, pointer variables and
dynamic memory allocation and management; and other
security risks, such as hard coded passwords.
PCA Test Environment: The Hardware and Software ITAs
document the setup of the voting system configuration to assure
a consistent test environment. The ITAs observe building of the
executable from reviewed source code. The Hardware and Software
ITAs work together to confirm that all testing is performed
only on ITA reviewed code built under ITA observation.
FCA Test Documentation Review: The ITA reviews and
assesses prior testing performed by the vendor. Based upon the
assessment of vendor testing the ITA identifies scope; designs
testing; and creates the Qualification Test Plan.
FCA Testing: Each ITA tests to their identified
scope, using their own internal processes.
Polling Place System Testing: The Hardware ITA
initiates environmental operating and non-operating
tests; functional testing of polling place hardware/
software, and user manuals for all VSS-required and
optional vendor supported functionality; testing the
capability of the voting system to assist voters with
disabilities or language; and accuracy and reliability
testing.
Election Management System Testing: The Software ITA
initiates functional testing of the Ballot Preparation
and Central Count hardware/software, and user manuals
for all VSS-required and optional vendor supported
functionality.
System Level Testing: The Software ITA initiates
end-to-end testing of the integrated EMS and Polling
Place System, including testing of the system
capabilities and safeguards, claimed by the vendor in
its TDP.
Creating the Test Methodology and Maintaining Quality
In structuring our review and test methodology we are guided by a
continual quest to improve the process and quality. From the foundation
of our first ITA project we have continually examined our methods.
Through ten completed or active projects we have honed and revised our
processes. Some changes have been based upon internal `lessons learned'
and others have come from the external changes in the ITA process, such
as the update to the 2002 VSS.
The process we followed in creating and maintaining the NASED
Qualification Testing was to define and document a review and test
process for both management and test activities. This process needed to
be standardized, repeatable and integrated into the overall structure
for all SysTest Labs testing projects. Within this standard structure
we tailored the individual methods to the unique requirements of
software ITA qualification testing based upon the 1990 VSS. Processes
addressed in this phase included VSS requirements management, test
elements (plans, test cases, reviews and reports), test management,
defect tracking, basic training, quality assurance, configuration
management (vendor materials and our testing) and project management.
Our next step was to work with and observe and improve the process
through successive test efforts. In this phase we broadened our view to
training needs, organizational coordination of the individual test
tasks and peer reviews. With each effort we reworked some processes and
identified other areas for potential process improvement.
At the point the 2002 VSS was implemented, we had a solid structure
and the perfect opportunity to implement several identified process
improvements, in conjunction with a conversion to the new standards.
While we continue to observe our processes, we are also moving into
an optimization phase. In our expanded role as a hardware ITA we will
be initiating some new processes that will follow our historic model,
but will also look at some of our old processes and optimize them for
an increased workload.
Observations and Recommendations for Lab Accreditation
The majority of VSS requirements for qualification testing involve
software. There are unique environmental tests that address hardware
specifically, but the VSS requires that a portion of software testing
for accuracy and reliability be performed in environmental chambers. In
doing so there is an overlap. The most effective way to handle this
overlap is to create a structure that permits joint testing of the
hardware and software. NASED structured the scope of testing so that
the hardware ITA was responsible for functional software and hardware
testing on the polling place equipment and environmental testing of the
hardware. The software ITA has been responsible for the ballot
preparation and central count functionality along with integration
testing of the entire system (end-to-end elections processes). While
the software ITA does not review all the code, they must receive all of
the code in order to perform end-to-end testing on the integrated
system.
We feel this scope should be changed due to the following issues:
Polling place software cannot be fully tested without
integrating the entire voting system. Today's new voting system
vendors do not develop separate applications. In the majority
of systems we see, a vendor is forced to artificially divide
their code in order to give the polling place software to the
hardware ITA and the balance to the software ITA.
The ITA labs try to keep duplication of effort down
to a minimum, however integration testing must repeat much of
the polling place functional testing.
Vendors are required to return to the hardware ITA
for regression testing if issues are uncovered during
integration testing. If the software ITA uncovers an issue in
the polling place during integration testing, they must notify
the hardware ITA. While the software ITA must rerun their tests
with the new version of the code, the hardware ITA is
responsible for reviewing the code changes to fix the issue and
functionally testing to confirm the fix. In addition, there
have been times when ITA labs have an inconsistent
interpretation of the standards and a vendor's solution will
overlap between the hardware and software ITA.
While environmental hardware testing requires
specialized equipment and testing, the environmental test
methodology is not unique to voting systems and generally does
not require specialized knowledge of voting. Furthermore,
effective software testing does require specialized knowledge
of voting practices.
We recommend that accreditation of labs include the following:
Primary labs that bear responsibility for all
testing, review and reporting. Primary labs may have qualified
subcontractors to perform specialized testing, e.g., hardware
environmental testing. The primary lab must demonstrate their
ability to monitor the work of the subcontractors and verify
that all subcontractor work reflects quality processes equal to
or greater than those of the primary lab;
Validation of an understanding of the unique
functional requirements of voting systems and voting system
standards;
Validation of manual and automated software testing
experience, methodology and software quality engineering
practices meet a minimum of CMMI Level 3; and
Validation of test equipment and chambers sufficient
to perform all VSS defined environmental testing, as well as
environmental testing experience, methodology and quality
engineering practices.
Observations and Recommendations for Voting System Standards
One hears much discussion on the adequacy of the 2002 FEC Voting
System Standards with extensive criticism against the adequacy of
security standards, but perhaps these critics are not taking a broad
view of how the VSS addresses security. Basic functionality
requirements, such as printing the name of an election and date on all
reports, are an aspect of security. Voting system, accuracy and
reliability are aspects of securing the vote. Any functional
requirement of the VSS that deals with election creation, voting,
counting or auditing is an aspect of securing the vote. The VSS
requirement for a vendor to identify the weight of paper deals with the
security of the vote. Additionally, the VSS requirements call for
documentation of the process to ensure physical security of a voting
system and the ability to detect intrusion. When looked at from this
broad view, the requirements of the VSS are quite comprehensive.
Criticism is generally is focused on the narrower view of security
in terms of active attack code such as viruses, worms, Trojan horses,
logic bombs, backdoors, exploitable vulnerabilities, and programming
flaws. The VSS provides some detail here. There are also sections in
the VSS that provide the labs with some wider latitude. In Volume 2
Section 1.5 the VSS states ``Additionally, new threats may be
identified that are not directly addressed by the Standards or the
system. As new threats to a voting system are discovered, either during
the system's operation or during the operation of other computer-based
systems that use technologies comparable to those of another voting
system, ITAs shall expand the tests used for system security to address
the threats that are applicable to a particular design of voting
system.'' A statement like this allows the individual lab a great deal
of discretion in testing. What it does not do is provide the detail for
consistency across all ITA testing.
Is providing more detail being addressed? HAVA specifically
identifies a review of the security and accessibility requirements of
the VSS and creation of new voting standards by the EAC, with the
support of NIST.
Is there anything that can be done to enhance the VSS without
waiting for the writing of new standards? Yes. The 2002 FEC Voting
System Standards Implementation Plan identified a process for issuing
clarification bulletins. This year NASED Voting System Standards Board
Technical Guides 1 and 2 were issued with clarifications of two VSS
requirements dealing with accessibility. Although NASED has a mechanism
to issue clarifications, we are not aware if they have the physical or
financial resources to meet this responsibility.
In terms of the HAVA mandated review of the VSS to be performed by
the EAC and NIST, we offer the following suggestions for greater
guidance in the standards:
Coding flaws--These may have security implications,
such as vulnerable constructs. Some languages and their
supporting libraries provide security vulnerabilities within
their functions. This can allow for a buffer overflow (which is
addressed in the VSS Volume 2 Section 5.4.2.d, ``For those
languages with unbound arrays, provides controls to prevent
writing beyond the array, string, or buffer boundaries'') or a
stack overflow attack. Additional, and potentially more
harmful, is the vulnerability to access the wrong program or
data file. This makes the program susceptible to the
introduction of external malicious code. We suggest providing
language specific prohibitions of vulnerable constructs.
Currently these vulnerable constructs can be used in programs
without malicious intent but it is difficult in a static review
to detect the security implication with their use.
Race conditions--Synchronization issues, such as race
conditions, present security vulnerabilities. Automated code
checking tools can detect the potential for this situation but
typically detect a number of ``false positives.'' We suggest
guidance on the acceptability of race conditions within the
code.
Global Variables--These variables are recognized
throughout the program and in some cases are used to store
critical status information that a number of programs need and
therefore provide a valuable service; however, their potential
for error and abuse should discourage their use. We suggest
guidance on when they can and cannot be used.
We would also suggest that the standards include the following:
Code Review Requirements for the vendors to provide
documentation identifying the known security weaknesses of the
programming language(s) they used, and their process for
mitigating those weaknesses.
Requirements for the vendors to provide documentation
of their security practices. The standards need to also provide
the ITAs with guidance for the review of this documentation to
assure that security is incorporated into the vendor's
development process.
Observations and Recommendations for NASED ITA Qualification Testing
The greatest challenge for NASED ITA Qualification Testing is the
lack of understanding of what it is, what it is supposed to do, what it
does not do and the role it should play in the entire election process.
What is NASED ITA Qualification Testing? It is the second of four
levels of testing identified in the VSS.
Level 1 Vendor Testing: The vendor tests to ensure
that their system meets their design specifications, the
requirements of the VSS, and any specifically supported State
requirements.
Level 2 NASED ITA Qualification Testing: The vendor's
testing is reviewed for adequacy and additional testing is
performed by software and hardware ITAs to ensure that the
voting system meets the requirements of the VSS, and any
additional functionality supported by the voting system as
defined in the vendor's design specifications performs as
specified.
Level 3 State Certification Testing: State personnel
or contractors perform testing under the direction of the State
to ensure that the voting system meets all of the State's
requirements.
Level 4 Acceptance Testing: Individual jurisdictions
perform testing prior to each primary or general election to
ensure that the voting system operates as required.
What is the objective of NASED ITA Qualification Testing? The
intent of qualification testing is to ensure that only voting systems
that pass independent testing to the minimum requirements of the 2002
FEC Voting System Standards are issued a NASED Qualification Numbers.
This means:
The elements of the voting system (hardware,
software, any required materials, and all documentation) have
been defined, reviewed and tested for conformance with the
requirements of the VSS;
The voting system contains a method to successfully
create elections, provide a ballot, record votes, provide
report tallies, and produce an audit trail;
Using the vendor's documented procedures and
mandatory security processes, ensuring that voting is performed
in a secret, accurate, reliable and secure manner;
The source code has been reviewed and meets the
requirements for modularity, maintainability, consistency,
security, integrity, and the use of error handling;
The code is sufficiently well commented so if the
vendor cease to support the code it can be reasonably
maintained by another entity;
The code installed on the voting system for testing
was built from the source code reviewed by an ITA and witnessed
by an ITA;
The Vendor's documents required by the VSS the
requirements for content and format;
The Vendor documentation required to assist the
states and jurisdiction to configure, use and maintain the
voting system (hardware, software, other required materials and
documents) is accurate and sufficient to perform all supported
functions;
Security has been achieved through the demonstration
of technical capabilities in conjunction with the documented
mandatory administrative procedures for effective system
security;
Vendors have an established set of quality procedures
and have supplied evidence of their implementation through
development, internal testing, and ITA testing;
The elements of the voting system configuration have
been identified, tested and tracked by the ITA;
Upon completion of testing a report has been issued
to the NASED Technical Committee for peer review;
The report has been accepted and retained by the
NASED Technical Committee/EAC, the vendor and the ITA.
NASED issued a qualification number.
What NASED ITA Qualification Testing does not mean:
It does not mean that testing has been sufficient to
confirm a voting system meets the specific laws of all the
states or for that matter any state. There is much election
functionality in the VSS that is optional. The VSS only
requires that this work in terms of the vendor's own
requirements for a function. Taking an example to the extreme,
the VSS does not require a vendor to support primary or general
elections; these are both optional functions. A vendor must
support some sort of election, but the VSS allows the vendor to
specify exactly what they choose to support.
It does not mean that the code the vendor delivers
installed on the voting system is exactly the code that was
qualified. It does not mean that the hardware that was
delivered by the vendor matches the qualified hardware
specification. While a version number may be the same, without
a verification methodology at the State and local level, it is
possible for unqualified versions to be used in an election.
While security risks are significantly reduced, it
does not mean that the voting system does not require an
external audit process by the local jurisdiction for detection
and prevention of irregularities. The same stringent audit
processes jurisdictions apply should include the voting system.
What role should NASED ITA Qualification Testing play in the
election process?
If one goes back to the implementation program for the 1990 Voting
System Standards, one will see the direction that was originally
intended. Qualification testing was just the first step. Additional
phases were planned for State certification and local acceptance
testing. There was a structure outlined for the accreditation of labs
by NVLAP/NIST. The FEC was supposed to be a clearinghouse to make the
reports available to State and local officials. Additionally, the
States and local jurisdictions were encouraged to report their
certification and acceptance testing to the clearinghouse. Escrow
agents were envisioned to hold qualified versions of the code and
assist the States and local jurisdictions in validation of qualified
versions of code.
For unknown reasons, the later phases were not implemented. NASED
assumed the role for accreditation. No official clearinghouse or escrow
was established. States and local jurisdictions moved forward
independently. NASED informally provided a meeting place to exchange
information. The job of holding the report and source code fell to the
NASED ITAs. As the vendors and the ITAs had non-disclosure agreements,
delivery of the report beyond the NASED Technical Committee was at the
request of the vendor.
While the vendor controls delivery of the report, it does not mean
State and local officials do not have the right to see the report. The
report is only confidential if the State certification or a local
purchaser allows it to be a confidential. We receive instructions from
the vendors to send their reports to State agencies.
We would suggest that in going forward:
The 1990 Implementation Plan shall be used as
guidance in completing the future structure of the
qualification, certification and acceptance testing of voting
systems. Whatever structure is implemented, it must minimally
address the functions outlined in this baseline plan;
A risk and needs assessment be performed against the
roles outlined in the 1990 Implementation Plan to identify the
capabilities of the players to understand and perform their
roles;
The needs of the State certification and local
jurisdictions for using, understanding and interpreting the
qualification report should be incorporated into the new
standards from the EAC. The standards should define any
specific reporting methodology to assist the States and local
jurisdiction in understanding the reports;
An annually updated, centralized database of all
State specific voting requirements shall be made available to
the ITAs, vendors, and election officials.
Biography for Carolyn E. Coggins
Director of ITA Voting Services and Senior Project Manager
Carolyn Coggins has a BA in Economics from University of
California, Berkeley. She heads up all voting projects at SysTest Labs
and has signature authority for Independent Test Authority (ITA) Voting
System recommendations to NASED (National Association of State Election
Directors) for voting system approval. She serves as a ex-officio
member of the NASED Technical Committee. In this capacity she provides
technical assistance for NASED to the Election Assistance Commission
(EAC), State election officials, and voting system vendors. Carolyn is
the Chair of the Technical Data Package Special Task Group of the IEEE
Project 1583 Voting Equipment Standards.
On the voting system test efforts, her responsibilities include
development and maintenance of the quality processes that defines all
policies, procedures and templates required to perform ITA
certification testing for voting systems, ITA certification test
planning development, execution, and reporting, management of ITA
testing resources, and interfacing with other ITA. She communicates and
enforces the policies and procedures of SysTest Labs and NASED
including Test Engineering best practices for the testing of voting
systems. She oversees ITA daily testing and approves the reports
generated in ITA test projects. Recently she managed the efforts
associated with the expansion of SysTest Labs NASED accreditation to
Full ITA status including both hardware and software. In addition,
Carolyn had led several highly complex testing projects for
telecommunications efforts, e-commerce efforts, large migration and
conversion projects. She has been with SysTest Labs since 1998.
Chairman Ehlers. Thank you. Dr. Shamos.
STATEMENT OF DR. MICHAEL I. SHAMOS, PROFESSOR OF COMPUTER
SCIENCE, CARNEGIE MELLON UNIVERSITY
Dr. Shamos. Mr. Chairman and Members of the Subcommittee,
my undergraduate degree is in physics, and my first graduate
degree is in physics, so whatever claim to omniscience that may
entitle me to in this room, I gladly accept.
I have been a faculty member in the School of Computer
Science at Carnegie Mellon University since 1975. I am also an
attorney admitted to practice in Pennsylvania and before the
U.S. Patent and Trademark Office. From 1980 until 2000, I was
statutory examiner of electronic voting systems for the
Commonwealth of Pennsylvania. During those 20 years, I
participated in every voting system examination conducted in
that state. From 1987 until 2000, I was statutory examiner of
computerized voting systems for the State of Texas, and during
those 13 years, I participated in every voting system
examination conducted in that state. All in all, I have
personally examined over 100 different electronic voting
systems.
In my opinion, the system that we now have for testifying
and certifying voting equipment in this country is not only
broken, but is virtually nonexistent and must be recreated from
scratch, or we are never going to restore public confidence in
elections. The process of designing, implementing,
manufacturing, certifying, selling, acquiring, storing, using,
testing, and even discarding voting machines must be
transparent from cradle to grave, and must adhere to strict
performance and security guidelines that should be uniform for
federal elections throughout the United States.
The step of qualification is testing to determine whether a
particular model of voting system meets appropriate national
standards. Unfortunately, no adequate standards currently
exist. The Federal Voting System Standards, FVSS, formerly
known as the FEC standards, are not only incomplete and out of
date, but there exists no effective procedure for even
repairing them.
Even if suitable standards existed, the current process of
qualification testing by Independent Testing Authorities
certified by NASED is not effective. As proof, I need only cite
the fact that the voting systems about which security concerns
have recently been raised in the popular press, such as Diebold
Accuvote, were all ITA-qualified. Some of these systems
contained security holes so glaring that one wonders what the
ITA was doing when they were doing the testing.
Well, one may wonder, but one cannot find out. The reason
for that is that the ITA procedures are entirely opaque to the
public. The NASED web site contains the following peremptory
statement: ``The ITAs do not and will not respond to outside
inquiries about the testing process for voting systems, nor
will they answer questions related to a specific manufacturer
or a specific voting system. They have neither the staff nor
the time to explain the process to the public, the news media,
or jurisdictions.'' By the way, the emphasis in that quotation
was theirs, not mine. I emphasize the capitalized words from
the NASED web site.
The next step, after qualification, which is certification,
the process that I participated in, certification to individual
State requirements, is also flawed. Many states that formerly
had statutory certification procedures have abdicated them in
favor of requiring no more from a vendor than an ITA
qualification letter, in some cases, even less. Alabama, for
example, requires no certification at all, but relies on a
written guarantee by the vendor that its system satisfies that
State's statutory requirements. Mind you, these are
requirements over which experts may differ as to their meaning.
My own State, Pennsylvania, I am embarrassed to say, abandoned
certification in the year 2002, because it believed the ITA
process was sufficient. We are, therefore, less safe in 2004
than we were 20 years ago, and possibly less safe than we even
were in the year 2000.
Even certified machines may not operate properly when
delivered to a jurisdiction, and must undergo acceptance
testing, but I am not aware of any State that makes such
testing a statutory requirement. It may be recommended in the
standards, and the ITAs may recommend it, but there is no body
that actually forces the states to go through acceptance
testing.
So far, we have ignored the matter of where the software
used in the machine actually comes from. It may have worked
when delivered by the vendor, but may have been modified or
substituted, either deliberately or innocently, by persons
known or unknown. We need a central repository for election
software, to which candidates and the public has continuing
access, so that it may be known and verified exactly what
software was used to present the ballot to the voter, and to
tabulate a specific election.
I was provided in advance with three questions to which I
understand the Subcommittee desires answers. One related to the
accreditation of testing laboratories, and whether that should
be changed to ensure greater public confidence. I believe that
there certainly is room for testing laboratories. I am not
against the ITA process. I just think it needs to be revamped.
Testing laboratories should be certified and rigorously
monitored by the EAC, or such other national body as Congress
may create. The cost of testing should be shouldered by the
states on a pro rata basis, possibly out of HAVA funds. I don't
believe that the laboratories should be paid by the vendors,
which is the current method.
In testing laboratories, we have faced the following
paradoxical situation. It is bad to have just one, because
there is no competition, but it is also bad to have more than
one, and the reason that is bad is that if there are multiple
laboratories, undoubtedly one of them will have the reputation
of being the most lax, and that is the one that every vendor
would like to have examining its equipment. So, I can't decide
whether there ought to be one laboratory or multiple
laboratories, except that if there are multiple laboratories,
and the vendor has no participation in the decision as to which
laboratory will be used to test his equipment, then we would
have no conflict of interest.
What can be done to improve these processes before the 2004
election, and what needs to be done by 2006? Well, the answer
to the first question is simple. I don't think there's anything
one can meaningfully do in the next 130 days that remain before
the 2004 election. Even if it were possible to enact
legislation, the states would be powerless to comply in so
short a time. The saving grace, though, is that the mere
presence of security vulnerabilities in voting systems does not
mean that actual security intrusions will occur. We have had a
successful record of using DRE machines in the United States
since the late '70's. We have had a nearly perfect record of
using them in Pennsylvania since 1984. There has never been a
single verified incident of actual manipulation of DRE voting
results in this country. We may thank our lucky stars for that.
It may be happenstance that that occurred, but nonetheless,
there has been a tremendous hullabaloo raised over incidents
that have never actually occurred.
And how important is NIST's role in improving the way
voting equipment is tested? I believe that NIST has an
important role, but we are not just talking about simple
electrical or mechanical specifications for equipment. We are
talking about standards from beginning to end of the entire
voting process, from where the machines come from, how they are
deployed, how people are trained to use them, et cetera. And so
I think NIST is part of the process, but the EAC, which has
great election expertise, needs to be the primary force behind
such processes.
Thank you very much.
[The prepared statement of Dr. Shamos follows:]
Prepared Statement of Michael I. Shamos
Mr. Chairman: My name is Michael Shamos. I have been a faculty
member in the School of Computer Science at Carnegie Mellon University
in Pittsburgh since 1975. I am also an attorney admitted to practice in
Pennsylvania and before the United States Patent and Trademark Office.
From 1980-2000 I was statutory examiner of electronic voting systems
for the Secretary of the Commonwealth and participated in every
electronic voting system examination held in Pennsylvania during those
20 years. From 1987-2000 I was statutory examiner of electronic voting
systems for the Attorney General of Texas and participated in every
electronic voting system examination held in Texas during those 13
years. In all, I have personally examined over 100 different electronic
voting systems. The systems for which I have participated in
certification were used to count more than 11 percent of the popular
vote in the United States in the year 2000.
I have not received any federal funding for my voting work.
I am here today to offer my opinion that the system we have for
testing and certifying voting equipment in this country is not only
broken, but is virtually nonexistent. It must be re-created from
scratch or we will never restore public confidence in elections. I
believe that the process of designing, implementing, manufacturing,
certifying, selling, acquiring, storing, using, testing and even
discarding voting machines must be transparent from cradle to grave,
and must adhere to strict performance and security guidelines that
should be uniform for federal elections throughout the United States.
There are a number of steps in the process of approving and using
voting systems that must be distinguished. The process of
``qualification'' is testing to determine whether a particular model of
voting system meets appropriate national standards. Unfortunately, no
such standards currently even exist. The Federal Voting System
Standards (FVSS), formerly known as the FEC Standards, are incomplete
and out of date.
For example, one of the principal election security worries is the
possibility of a computer virus infecting a voting system. Yet the FVSS
place virus responsibility on the voting system vendor and do not
provide for any testing by the Independent Testing Authority (ITA).
Furthermore, the standards do not even require that a voting system
contain any virus detection or virus removal software at all: ``Voting
systems shall deploy protection against the many forms of threats to
which they may be exposed such as file and macro viruses, worms, Trojan
horses, and logic bombs. Vendors shall develop and document the
procedures to be followed to ensure that such protection is maintained
in a current status.'' It is hardly reassuring to have the fox
guarantee the safety of the chickens.
Even if there were suitable standards, it is a significant question
how to assure the public that a particular machine meets them. The
current process of qualification testing by Independent Testing
Authorities certified by the National Association of State Election
Directors (NASED) is dysfunctional. As proof I need only cite the fact
that the voting systems about which security concerns have recently
been raised, such as Diebold Accuvote, were all ITA-qualified. Some of
these systems contain security holes so severe that one wonders what
the ITA was looking for during its testing.
One may wonder, but one cannot find out. The ITA procedures are
entirely opaque. The NASED web site contains this peremptory statement:
``The ITAs DO NOT and WILL NOT respond to outside inquiries about the
testing process for voting systems, nor will they answer questions
related to a specific manufacturer or a specific voting system. They
have neither the staff nor the time to explain the process to the
public, the news media or jurisdictions.'' I don't believe that either
Congress of the public should allow ITAs to behave this way. Did I say
``ITAs''? Allow me to correct that. For hardware testing, there is only
a single NASED-certified ITA: Wyle laboratories of Huntsville, Alabama.
I find it grotesque that an organization charged with such a heavy
responsibility feels no obligation to explain to anyone what it is
doing.
It should be understood that qualification to standards addresses
only one part of the problem. A qualified machine may not meet State
statutory requirements even if it functions perfectly. A further
examination, called certification, is needed to learn whether the
machine can actually be used in a given state. Even a certified machine
may fail to function when purchased unless it is tested thoroughly on
delivery, a form of evaluation known as acceptance testing. I am not
aware of any state that makes such testing a statutory requirement.
Assuming that the machines operate properly when delivered, there
is no assurance that they will be stored, maintained, transported or
set up properly so they work on Election Day. While many states provide
for pre-election testing of machines, in the event of a large-scale
failure they can find themselves without enough working machines to
conduct an election.
The machines may work according to specification but if they have
not been loaded with the appropriate set of ballot styles to be used in
a polling place they will be completely ineffective. The process of
verifying ballot styles is left to representatives of the political
parties, who may have little interest in the correctness of non-
partisan races and issues.
In this whole discussion we have ignored the matter of where the
software used in the machine comes from. It may have worked when
delivered by the vendor but may have been modified or substituted,
either deliberately or innocently, by persons known or unknown. We need
a central repository for election software to which candidates and the
public has continuous access, so it may be known and verified exactly
what software was used to present the ballot and tabulate the results.
I was provided in advance with three question to which I understand
the Subcommittee desires answers.
1. How should the accreditation of testing laboratories and the
testing and certification of voting equipment be changed to improve the
quality of voting equipment and ensure greater trust and confidence in
voting systems?
Testing laboratories should be certified and rigorously monitored
by the EAC, or such other national body as Congress may create. The
cost of testing should be shouldered by the states on a pro-rata basis,
possibly out of HAVA funds. The laboratories should certainly not be
paid by the vendors, which is the current method.
In testing laboratories we face the paradoxical situation that it
is bad to have just one, but it is also bad to have more than one. A
single laboratory has scant incentive to do a good job, but every
incentive to please its customers, namely the vendors. If there are
multiple laboratories, however, then some will acquire the reputation
of being more lax than others, and the vendors will seek to have their
system tested by the most ``friendly'' laboratory. This problem can be
alleviated by monitoring the performance of the laboratories and
according the vendors no role in their selection.
The existence of federal standards and ITAs has actually had a
counterproductive effect. Many states that formerly had statutory
certification procedures have abdicated them in favor of requiring no
more from a vendor than an ITA qualification letter, and in some cases
even less. Alabama, for example, requires no certification at all but
relies on a written guarantee by the vendor that its system satisfies
the State's requirements. My own State, Pennsylvania, abandoned
certification in 2002 because it believed the ITA process was
sufficient. We are less safe in 2004 than we were 20 years ago.
2. What can be done to improve these processes before the 2004
election, and what needs to be done to finish these improvements by
2006?
I do not believe that Congress can act meaningfully in the 130 days
that remain before the 2004 election. Even if it could, the states
would be powerless to comply in so short a time. A saving race is that
the mere presence of security vulnerabilities does not mean that
tampering will or is likely to occur. We have been holding successful
DRE elections in the U.S. for over 20 years. The problem this year is
that many states, wishing to avoid the negative experience of Florida
in 2000, have rushed to acquire new voting systems with which they are
unfamiliar. This will undoubtedly lead to machine failures long lines,
and dissatisfaction at the polls in November. It is not likely to lead
to security intrusions. I should mention that since DREs were
introduced in the late 1970s, there has not been a single verified
incident of tampering with votes in such a system. There have been
numerous allegations, all of which vanish into thin air when
investigated. The most important factor right now in running a
satisfactory election is training of the people who must operate the
voting machines.
For 2006 there are many actions that can be taken:
The process of conducting elections in the U.S. is
highly fragmented. Election administration is left up to 3170
individual counties, except in a few states, such as Georgia,
which have statewide voting systems. This means that there is a
huge variance in elections budgets and level of expertise
across the country. The states should be encouraged through the
mechanism of HAVA to adopt systems and procedures that are as
uniform as possible within each state. The more different
voting systems a State operate, the more difficult it becomes
to keep track of the software and firmware that is used to run
them.
No jurisdiction should be forced to deploy a new
voting mechanism before it is ready. The availability of large
amounts of HAVA funding has not been helpful in this regard.
The rush to rid the Nation of punched-card systems, while
generally laudable, has propelled counties having no experience
with DRE elections into errors whose consequences will take
years to overcome. A partial solution is gradual deployment and
transition to the newer systems rather than overnight
replacement.
The need for voter and poll worker training cannot be
over-emphasized. The best and most secure voting machine will
not function properly if poll workers do not know how to
operate it and voters don't know how to use it.
A comprehensive regime of qualification,
certification, acceptance and operational testing is needed.
We need a coherent, up-to-date, rolling set of voting
system standards combined with a transparent, easily-understood
process for testing to them that is viewable by the public. We
don't have that or anything resembling that right now, and the
proposal I have heard are not calculated to install them.
The means by which voting machines are modified,
updated and provided with ballot styles and software should be
tightly controlled, with meaningful criminal penalties for
violations. Right now, a vendor who distributes uncertified
software risks little more than adverse newspaper coverage.
3. How important is NIST's role in improving the way voting equipment
is tested? What activities should States be undertaking to ensure
voting equipment works properly?
I believe that NIST has an useful role to play in developing
standards for voting system qualification, but it should not be a
dominant one.
NIST claims to have expertise in the voting process, and cites the
fact that it has produced two published reports on the subject. The
first of these, which appeared in 1975, was a ringing endorsement of
punched-card voting, now recognized to be the worst method of voting
ever devised by man. The second report, 13 years later, corrected that
error. Both, however, were written by a single individual who is not
longer with NIST. The NIST voting web site, vote.nist.gov, contains a
table of 16 ``cyber security guidelines'' that NIST asserts are
responsive to the risks of e-voting. These guidelines occupy more than
2000 printed pages, yet the word ``voting'' appears nowhere within
them.
While it is true that stringent voting machines standards are
required, the task of developing them should not be assigned to NIST
merely because the word ``Standards'' is part of its name. For voting
standards are unlike any other in that they must be capable of being
understood and accepted by the entire public. An airline passenger may
place his trust in the pilot to verify that the plane both are about to
fly in has been properly maintained. The hospital patient relies on the
doctor for assurance that equipment in the operating room will not kill
him. The voter has no one to turn to if her vote is not counted and
therefore must develop a personal opinion whether the system is to be
trusted. Suspicion about the manner of making and testing voting
machines harms everyone. Arcane technical standards make the problem
worse.
Having a successful, error-free and tamper-free election is not
simply a matter of using a voting machine that obeys certain published
criteria. Everything about the process, including the input of ballot
styles, handling of vote retention devices, testing and subsequent
audit must follow controlled protocols. If voting were done in a
laboratory, it could be instrumented and observed carefully by
engineers following precise procedures. However, voting is conducted
using over one million volunteer poll workers, many of whom are senior
citizens with scant computer experience. In fact, almost 1.5 percent of
the U.S. voting population consists of poll workers themselves. The
reality that elections are not run by engineers is an important
consideration in the development and implementation of standards.
In short, expertise in the process of voting and the human factors
and fears that attend that process have not historically been within
NIST's expertise. I do not doubt that NIST could acquire the necessary
experience given sufficient time, money and mandate. But the Nation
does not have that kind of time. A repeat of the Florida 2000
experience will have a paralytic effect on U.S. elections.
Instead, I propose that standards for the process of voting be
developed on a completely open and public participatory basis to be
supervised by the EAC, with input from NIST in the areas of its
demonstrated expertise, such as cryptography and computer access
control. Members of the public should be free to contribute ideas and
criticism at any time and be assured that the standards body will
evaluate and respond to them. When a problem arises that appears to
require attention, the standards should be upgraded at the earliest
opportunity consistent with sound practice. If this means that voting
machines in the field need to be modified or re-tested, so be it. But
the glacial pace of prior development of voting standards is no longer
acceptable to the public.
I may have painted a depressing picture of the state of voting
assurance in the United States. That was my intention. However, I have
a number of suggestions by which the process can be made to satisfy
most of my concerns. In addition to the proposals presented above, I
add the following:
1. There are too many organizations that appear to have
authoritative roles in the voting process, including the FEC,
NASED, the Election Center, NIST and the EAC. Most assert that
compliance with their recommendations is voluntary, and legally
it may be. But election officials abhor a vacuum, and the mere
existence of published standards, good or bad, is enough to
cause states to adopt them. A coherent scheme needs to be
devised, at least one that will assure that voting machines
work and are secure. I do not propose to sacrifice State
sovereignty over voting methods and procedures so long as they
are safe.
2. There is a Constitutional reluctance in the United States
to having the Federal Government control elections, even those
over which it may have authority to do so. I have long believed
that states must be left to determine the form of voting.
However, there is no contradiction in requiring that they obey
minimum standards necessary to ensure that all citizens have
their votes counted and moreover are confident that their votes
have been counted.
3. The reality is that states cannot assume the expense of
conducting multiple elections on the same day using different
equipment and procedures, so if standards are mandated for
elections involving federal offices they will almost certainly
be used for all elections.
4. The current pall that has been cast over computerized
voting in the U.S. can only be lifted through greater public
involvement in the entire process.
I thank you for the opportunity to present testimony here today.
Biography for Michael I. Shamos
Michael I. Shamos is Distinguished Career Professor in the School
of Computer Science at Carnegie Mellon University, where he serves as
Co-Director of the Institute for eCommerce, teaching courses in
eCommerce technology, electronic payment systems and eCommerce law and
regulation.
Dr. Shamos holds seven university degrees in such fields as
physics, computer science, technology of management and law. He has
been associated with Carnegie Mellon since 1975.
From 1980-2000 he was statutory examiner of computerized voting
systems for the Secretary of the Commonwealth of Pennsylvania. From
1987-2000 he was the Designee of the Attorney General of Texas for
electronic voting certification. During that time he participated in
every electronic voting examination conducted in those two states,
involving over 100 different voting systems accounting for more than 11
percent of the popular vote of the United States in the 2000 election.
Dr. Shamos has been an expert witness in two recent lawsuits
involving electronic voting: Wexler v. Lepore in Florida and Benavidez
v. Shelley in California. He was the author in 1993 of ``Electronic
Voting--Evaluating the Threat'' and in 2004 of ``Paper v. Electronic
Voting Records--An Assessment,'' both of which were presented at the
ACM Conference on Computers, Freedom & Privacy.
Dr. Shamos has been an intellectual property attorney since 1981
and has been an expert witness in Internet cases involving the Motion
Picture Association of America and the Digital Millennium Copyright
Act. He is Editor-in-Chief of the Journal of Privacy Technology, an
all-digital publication of the Center for Privacy Technology at
Carnegie Mellon.
Further information is available at http://euro.ecom.cmu.edu/
shamos.html.
Chairman Ehlers. Thank you very much, and Dr. Semerjian.
STATEMENT OF DR. HRATCH G. SEMERJIAN, ACTING DIRECTOR, NATIONAL
INSTITUTE OF STANDARDS AND TECHNOLOGY (NIST)
Dr. Semerjian. Thank you, Mr. Chairman, and Members of the
Committee. Thank you for the opportunity to testify today on
NIST responsibilities under the Help America Vote Act,
specifically on testing and certification of voting equipment.
Clearly, major changes are taking place in the way we
conduct elections. We are running into more and more optical
scanners or touch screen systems, and as a result of these
changes, Congress enacted the Help America Vote Act, commonly
known as HAVA, and mandated specific roles for NIST.
Many of the issues we are examining today are directly
related to standards and guidelines. Congress understood the
importance of standards in voting technologies, and
specifically gave the Director of NIST the responsibility of
chairing the Technical Guidelines Development Committee,
otherwise known, TGDC, a Committee reporting to the Election
Assistance Commission under HAVA. The TGDC is charged with
making recommendations to the Election Assistance Commission
with regard to voluntary standards and guidelines for election-
related technologies that have an impact on many of the issues
we are discussing.
While we have considerable experience in standards
development, NIST understands that as a non-regulatory agency,
our role is limited, and we need to understand the needs of the
community. To this end, NIST staff have started to meet with
members of the election community. Also, at the request of
Congress and the National Association of State Election
Directors, NIST organized and hosted a symposium on building
trust and confidence in the voting systems last December. Over
300 attendees from the election community were at the seminar
to begin discussion, collaboration, and consensus building on
voting reform issues.
Mr. Chairman, at this time, I would like to enter a copy of
the CDs that contain the video transcripts of the symposium
into the record. Thank you.
Chairman Ehlers. Without objection, so ordered.
Dr. Semerjian. As required under HAVA, NIST recently
delivered to the EAC a report which assesses the areas of human
factors research and human-machine interaction, which feasibly
could be applied to voting products and systems design to
ensure the usability and accuracy of voting products and
systems. The EAC delivered the report to Congress on April 30
of this year. Again, the specific recommendations of the report
are included in my written testimony.
NIST views as a top priority accomplishing its
responsibilities mandated in the HAVA legislation, in
partnership with the EAC. These mandates include the
recommendation of voluntary voting system standards to the EAC
through its Technical Guidelines Development Committee. The
first set of voluntary standards is due nine months after the
appointment of the 14 members by the EAC. Last week, the EAC
announced the membership of the TGDC, and their first meeting
has been scheduled for July 9.
Under HAVA, NIST is directed to offer formal accreditation
to laboratories that test voting system hardware and software
for conformance to the current voting system standards.
Yesterday, NIST announced in the Federal Register the
establishment of a laboratory accreditation program for voting
systems. NIST will carry out the accreditation of these
laboratories through the National Voluntary Laboratory
Accreditation Program, otherwise known as NVLAP, which is
administered by NIST.
NVLAP is a long-established laboratory accreditation
program that is recognized both nationally and internationally.
NVLAP will also conduct a public workshop with interested
laboratories in the near future to review its accreditation
criteria, as well as receive comments and feedback from the
participating laboratories and other interested parties. After
the workshop, NVLAP will finalize specific technical criteria
for testing laboratories and make the necessary logistical
arrangements to begin the actual assessment of the
laboratories. It is our intention that laboratories will be
able to formally apply to NVLAP and initiate the assessment
process in early 2005, if not sooner.
Laboratories seeking accreditation to test voting system
hardware and software will be required to meet the NVLAP
criteria for accreditation, which include the ISO/IEC 17025
standard, the 2002 Voting System Standards, and any other
criteria deemed necessary by the Election Assistance
Commission. To ensure continued compliance, all NVLAP
accredited laboratories will undergo an onsite assessment
before initial accreditation, during the first renewal year,
and every two years thereafter to evaluate their ongoing
compliance with specific accreditation criteria.
Only after a laboratory has met all NVLAP criteria for
accreditation will it be presented to the EAC for its approval
to test voting systems. The EAC may impose requirements on the
laboratories in addition to the NVLAP accreditation.
Finally, NIST has compiled best security practices relevant
to election security from current Federal Information
Processing Standards, FIPS. These standards are available on
both the NIST web site and the EAC web site. This compilation
is intended to help State and local election officials with
their efforts to better secure voting equipment before the
November 2004 election.
NIST realizes how important it is for voters to have trust
and confidence in voting systems even as new technologies are
introduced. Increasingly, computer technology touches all
aspects of the voting process, voter registration, vote
recording, and vote tallying. NIST believes that rigorous
standards, guidelines, and testing procedures will enable U.S.
industry to produce products that are high quality, reliable,
interoperable, and secure, thus enabling the trust and
confidence that citizens require, and at the same time,
preserving room for innovation and change.
Mr. Chairman, thank you for the opportunity to testify, and
I will be happy to answer any questions.
[The prepared statement of Dr. Semerjian follows:]
Prepared Statement of Hratch G. Semerjian
Mr. Chairman and Members of the Committee, thank you for the
opportunity to testify today on NIST's responsibilities under the Help
America Vote Act, specifically testing and certification of voting
equipment. Major changes are taking place in the way we conduct
elections. Our trusty old ballot boxes often are being replaced by a
host of new technologies. Citizens are now much more likely to
encounter optical scanners or touch screen systems at the polling place
than a wooden box with a sturdy lock. As a result of these changes,
Congress enacted the Help America Vote Act, commonly known as HAVA, and
mandated specific research and development roles for the National
Institute of Standards and Technology (NIST).
Many of the issues we are examining today are all directly related
to standards and guidelines. As we like to say at NIST, if you have a
good standard, you can have a good specification, and with proper
testing you will be assured that the equipment performs as required.
Congress understood the importance of standards in voting technologies
and specifically gave the Director of NIST the responsibility of
chairing the Technical Guidelines Development Committee (TGDC), a
committee reporting to the EAC under HAVA. This committee is charged
with making recommendations to the Election Assistance Commission (EAC)
with regard to voluntary standards and guidelines for election-related
technologies that have an impact on many of the issues we are
discussing.
While we have considerable experience in ``standards development,''
NIST understands that as a non-regulatory agency our role is limited
and has started to meet with members of the ``elections community,''--
ranging from disability advocacy groups, voting advocacy groups,
researchers, State and local election officials, and vendors--to learn
about their concerns. Ultimately, in coordination with the EAC and the
broader ``elections community'' we want to apply our ``standards
development'' experience to election-related technologies so that, when
voting is complete, the vote tally will be accurate and done in a
timely manner.
NIST is by no means a newcomer to the issues related to electronic
voting. Previous to the HAVA, NIST's involvement in studying voting
machine technology resulted in the publication of two technical papers
in 1975 and 1988. NIST's recent activities related to voting system
technology have been preparatory to the implementation of HAVA and
fulfilling the initial mandates of the law.
At the request of Congress and the National Association of State
Election Directors, NIST organized and hosted a Symposium on Building
Trust and Confidence in Voting Systems in December of 2003 at its
Gaithersburg headquarters. Over three hundred attendees from the
election community attended the seminar to begin discussion,
collaboration and consensus on voting reform issues. Symposium
participants included State and local election officials; vendors of
voting equipment and systems, academic researchers; representatives of
the cyber security and privacy community; representatives from the
disability community, standards organizations and independent testing
authorities, as well as newly appointed U.S. Election Assistance
Commissioners. Representative stakeholders participated with NIST
scientists in panels addressing:
Testability, Accreditation and Qualification in
Voting Systems;
Security and Openness in Voting Systems; and
Usability and Accessibility in Voting Systems.
Attendees agreed that they all shared the goals of:
Practical, secure elections, with every vote being
important;
The importance of looking at the voting system end-
to-end;
The need for good procedures & best practices in
physical & cyber security;
The need to improve current testing & certification
procedures;
The need to separately address both short-term and
long-term challenges; and
The benefits of the election community working as a
team.
As required under HAVA, NIST recently delivered to the EAC a report
``which assesses the areas of human factors research and human-machine
interaction, which feasibly could be applied to voting products and
systems design to ensure the usability of and accuracy of voting
products and systems, including methods to improve access for
individuals with disabilities (including blindness) and individuals
with limited proficiency in the English Language and to reduce voter
error and the number of spoiled ballots in elections.'' The EAC
delivered the report to Congress on April 30, 2004.
The report titled ``Improving the Usability and Accessibility of
Voting Systems and Products,'' assesses human factors issues related to
the process of a voter casting a ballot as he or she intends. The
report's most important recommendation is for the development of a set
of usability standards for voting systems that are performance-based.
Performance-based standards address results rather than equipment
design. Such standards would leave voting machine vendors free to
develop a variety of innovative products if their systems work well
from a usability and accessibility standpoint. Additionally, the report
emphasizes developing the standards in a way that would allow
independent testing laboratories to test systems to see if they conform
to the usability standards. The labs would employ objective tests to
decide if a particular product met the standards.
In total the report makes 10 recommendations to help make voting
systems and products simpler to use, more accurate and easily available
to all individuals--including those with disabilities, language issues
and other impediments to participating in an election. The
recommendations highlight the need to:
1) Develop voting system standards for usability that are
performance-based, relatively independent of the voting
technology, and specific (i.e., precise).
2) Specify the complete set of user-related functional
requirements for voting products in the voting system
standards.
3) Avoid low-level design specifications and very general
specifications for usability.
4) Build a foundation of applied research for voting systems
and products to support the development of usability and
accessibility standards.
5) To address the removal of barriers to accessibility, the
requirements developed by the Access Board, the current VSS
(Voting System Standards), and the draft IEEE (Institute of
Electrical and Electronics Engineers) standards should be
reviewed, tested, and tailored to voting systems and then
considered for adoption as updated VSS standards. The
feasibility of addressing both self-contained, closed products
and open architecture products should also be considered.
6) Develop ballot design guidelines based on the most recent
research and experience of the visual design communities,
specifically for use by election officials and in ballot design
software.
7) Develop a set of guidelines for facility and equipment
layout; develop a set of design and usability testing
guidelines for vendor- and State-supplied documentation and
training materials.
8) Encourage vendors to incorporate a user-centered design
approach into their product design and development cycles
including formative (diagnostic) usability testing as part of
product development.
9) Develop a uniform set of procedures for testing the
conformance of voting products against the applicable
accessibility requirements.
10) Develop a valid, reliable, repeatable, and reproducible
process for usability conformance testing of voting products
against the standards described in recommendation 1) with
agreed upon usability pass/fail requirements.
NIST views as a top priority accomplishing its impending
responsibilities mandated in the HAVA in partnership with the EAC.
These mandates include the recommendation of voluntary voting system
standards to the EAC through its Technical Guidelines Development
Committee. The first set of voluntary standards is due nine months
after the appointment of the fourteen members by the EAC. Last week the
EAC announced the membership of the TGDC. The first meeting of the TGDC
has been scheduled for July 9, 2004.
Under HAVA, NIST is directed to offer formal accreditation to
laboratories that test voting system hardware and software for
conformance to the current Voting System Standards. This week, NIST is
announcing in the Federal Register the establishment of a Laboratory
Accreditation Program for Voting Systems. NIST will carry out the
accreditation of these laboratories through the National Voluntary
Laboratory Accreditation Program (NVLAP), which is administered by
NIST. NVLAP is a long-established laboratory accreditation program that
is recognized both nationally and internationally. NVLAP accreditation
criteria are codified in the Code of Federal Regulations (CFR, Title
15, Part 285).
NVLAP will conduct a public workshop with interested laboratories
in the near future to review its accreditation criteria, as well as
receive comments and feedback from the participating laboratories and
other interested parties. After the workshop, NVLAP will finalize
specific technical criteria for testing laboratories and make the
necessary logistical arrangements to begin the actual assessment of the
laboratories. NVLAP must identify, contract, and train technical expert
assessors; laboratories must complete the NVLAP application process;
rigorous on-site assessments must be conducted; and laboratories
undergoing assessment must resolve any identified non-conformities
before accreditation can be granted. It is our intention that
laboratories will be able to formally apply to NVLAP and initiate the
assessment process in early 2005 if not sooner.
Simply stated, laboratory accreditation is formal recognition that
a laboratory is competent to carry out specific tests. Expert technical
assessors conduct a thorough evaluation of all aspects of laboratory
operation that affect the production of test data, using recognized
criteria and procedures. General criteria are based on the
international standard ISO/IEC 17025, General requirements for the
competence of testing and calibration laboratories, which is used for
evaluating laboratories throughout the world. Laboratory accreditation
bodies use this standard specifically to assess factors relevant to a
laboratory's ability to produce precise, accurate test data, including
the technical competency of staff, validity and appropriateness of test
methods, testing and quality assurance of test and calibration data.
Laboratory accreditation programs usually also specify field-specific
technical criteria that laboratories must meet, in addition to
demonstrating general technical competence.
Laboratory accreditation thus provides a means of evaluating the
competence of laboratories to perform specific types of testing,
measurement and calibration. It also allows a laboratory to determine
whether it is performing its work correctly and to appropriate
standards.
Laboratories seeking accreditation to test voting system hardware
and software will be required to meet the NVLAP criteria for
accreditation which include: ISO/IEC 17025, the 2002 Voting System
Standards, and any other criteria deemed necessary by the Election
Assistance Commission (EAC). To ensure continued compliance, all NVLAP-
accredited laboratories undergo an on-site assessment before initial
accreditation, during the first renewal year, and every two years
thereafter to evaluate their ongoing compliance with specific
accreditation criteria.
Only after a laboratory has met all NVLAP criteria for
accreditation will it be presented to the Election Assistance
Commission for its approval to test voting systems. The EAC may impose
requirements on the laboratories in addition to NVLAP accreditation.
Finally, NIST has compiled best security practices relevant to
election security from current Federal Information Processing standards
(FIPS). These standards are available on the NIST web site (http://
vote.nist.gov/securityrisk.pdf) and will be available on EAC's web site
(http://www.fec.gov/pages/vssfinal/vss.html). This compilation is
intended to help State and local election officials with their efforts
to better secure voting equipment before the November 2004 election.
NIST realizes how important it is for voters to have trust and
confidence in voting systems even as new technologies are introduced.
Increasingly, computer technology touches all aspects of the voting
process--voter registration, vote recording, and vote tallying. NIST
believes that rigorous standards, guidelines, and testing procedures
will enable U.S. industry to produce products that are high quality,
reliable, inter-operable, and secure thus enabling the trust and
confidence that citizens require and at the same time preserving room
for innovation and change.
Thank you for the opportunity to testify. I would be happy to
answer any questions the Committee might have.
Biography for Hratch G. Semerjian
Hratch G. Semerjian is serving as Acting Director of NIST while
Arden Bement serves in a temporary capacity as the Acting Director of
the National Science Foundation. Dr. Semerjian has served as the Deputy
Director of NIST since July 2003. In this position, Dr. Semerjian is
responsible for overall operation of the Institute, effectiveness of
NIST's technical programs, and for interactions with international
organizations. NIST has a total budget of about $771 million, and a
permanent staff of about 3,000, as well as about 1,600 guest
researchers from industry, academia, and other national metrology
institutes from more than 40 countries. Most of the NIST researchers
are located in two major campuses in Gaithersburg, Md., and Boulder,
Colo. NIST also has two joint research institutes; the oldest of these
is JILA, a collaborative research program with the University of
Colorado at Boulder, and the other is CARB (Center for Advanced
Research in Biotechnology), a partnership with the University of
Maryland Biotechnology Institute.
Dr. Semerjian received his M.Sc. (1968) and Ph.D. (1972) degrees in
engineering from Brown University. He served as a lecturer and post
doctoral research fellow in the Chemistry Department at the University
of Toronto. He then joined the research staff of Pratt & Whitney
Aircraft Division of United Technologies Corp. in East Hartford, Conn.
In 1977, Dr. Semerjian joined the National Bureau of Standards (now
NIST), where he served as Director of the Chemical Science and
Technology Laboratory (CSTL) from April 1992 through July 2003. Awards
he has received include the Fulbright Fellowship, C.B. Keen Fellowship
at Brown, the U.S. Department of Commerce Meritorious Federal Service
(Silver Medal) Award in 1984, and the U.S. Department of Commerce
Distinguished Achievement in Federal Service (Gold Medal) Award in
1995. In 1996, he was elected a Fellow of the American Society of
Mechanical Engineers. In 1997, he received the Brown Engineering Alumni
Medal. Dr. Semerjian was elected to the National Academy of Engineering
in 2000.
Discussion
Chairman Ehlers. Thank you very much, and thank all of you
for your testimony. We will now begin with the questioning, and
I yield myself five minutes for that purpose.
Election Management Best Practices and Acceptance Testing of
Voting Equipment
A couple of things, first of all. We are concerned about
the initial testing of the equipment and the software. We want
to make sure that it meets the design criteria, specifications,
that it works as it is intended to work. The second aspect is
to preserve that as time goes on, and ensure that it continues
to operate properly. Let me just, for my own information, ask a
question about that. Perhaps Mr. Wilkey would be the one to
answer. Others may want to comment.
On the newer electronic machines, do the manufacturers
provide some type of self-test routine that you run the
computers through before each election? In other words, you
insert this in, it runs through, checks the software, and makes
sure it is doing what it is supposed to do, that no one has
tinkered with it? Is that standard or is that just not done at
all?
Mr. Wilkey. Mr. Chairman, thank you for asking that
question, because it gives me an opportunity to talk about
something that I have been on my soapbox for over 15 years, and
now, as a private citizen, it may be the last time I have a
chance to talk about it publicly.
Certainly, what we have tried to do in the area of
standards development and testing of an initial product is only
50 percent of the battle.
Chairman Ehlers. Yeah.
Mr. Wilkey. The next 50 percent, and perhaps even the most
important part of what we are talking about, is what needs to
happen once the product is delivered to the jurisdiction. And
that is where we have consistently talked about doing
acceptance testing of a quality that is developed by the
jurisdiction and not the vendor, is done by the jurisdiction
and not the vendor, and similarly, all of the maintenance
activities and the pre-election testing that must occur,
ongoing, throughout the process.
One of our biggest problems in election administration in
this country is that there are over 13,000 election
jurisdictions. Many of them, as you know, Mr. Chairman, in your
own State, and in mine, are very small. They are mom and pop
operations with the county clerk that may have a number of
responsibilities, or a town clerk, if you are talking about the
New England states. They don't have the expertise always
available to them to do this, so on many occasions, they are
relying on the vendor to do this.
This is a practice that we are trying to stop, and what we
are hopeful, with the new Election Assistance Commission, that
they will get the necessary funding to be able to do what I
have talked about for the last 15 years, and that is the
management operational standards, that--it is a huge project.
But it needs to be done, because jurisdictions need to be able
to go some place to say I have bought this system. This is how
I do an adequate test. This is how I develop the test. This is
how I do ongoing maintenance. This is the kind of maintenance
logs I have to keep, and on and on and on. Because it is only
that 50 percent of the battle that we are seeing in the news
media today.
And another part of our problem, which I think the EAC
hopefully will address, and which the Chairman has addressed
already in his remarks a couple of weeks ago, is that we keep
hearing there are problems out there across America with these
systems. One of the things that we are not able to determine is
how many of these units are out there, and how many of these
units have problems, and what are these problems?
Hopefully, and the Chairman of the Commission, Chairman
Soaries, has called on every election jurisdiction in the
country to report to the EAC the problems that they are having
with their equipment, so that we can begin to see what is going
on, and we can see a pattern, and that the TGDC can begin to
take a look at the problems, then, and try to prevent them from
happening in the future.
So, thank you, Mr. Chairman. I am glad you asked that
question.
Chairman Ehlers. Well----
Mr. Wilkey. Because it is very important.
Chairman Ehlers. And my point simply was, it seems to me
because there are a lot of mom and pop operations, and I am
very familiar with that, that we should expect the
manufacturers to provide the testing software and materials to
test--at least test the software that is on the machine. The
county clerk or the township clerk can do--can set up 10
machines and run a quick fake election with employees, and make
sure that it works, to really make sure it hasn't been tinkered
with.
Mr. Wilkey. Yes, I agree, Mr. Chairman, and one of the
things we encourage everybody to do, one of the projects that
is going on right now at the EAC, and which I have been
involved in, is to do a series of best practices that will be
out and available to election jurisdictions over the next
several weeks.
And through this project, jurisdictions will be able to go
to the EAC web site and take a look at some examples of tests
that should be done on various equipment. I think it is good
for the vendor to provide this information to the jurisdiction,
but I think the jurisdiction has to go beyond that.
Chairman Ehlers. Yeah.
Mr. Wilkey. And so if a vendor says you have got to test
this, this, and this, the jurisdiction should be taking and say
yes, we are going to test this, this, and this, but we are
going to do it four times.
Chairman Ehlers. Dr. Shamos, I thought I saw you indicating
a twitch when I asked the question.
Dr. Shamos. Yes, you did, Mr. Chairman.
The processes that we are talking about here are much more
out of control than anyone is willing to admit. There are
essentially no controls on how software enters a voting machine
these days.
We know how it gets there when the machine is sold.
However, it is often necessary for the vendor to fix bugs, add
new features, make customizations that are specifically
requested by the jurisdiction. There may be statutes that
require them to submit that software for recertification, but
there is nothing that physically prevents them from putting a
chip into a FedEx envelope and sending it directly to the
county clerk, with instructions to install this chip, whose
contents they have no knowledge of, into a voting machine.
And the problem, of course, is exacerbated by the fact that
we have over 3,100 counties in the country, so essentially,
3,100 different elections, and that is another place where the
degree of sophistication or lack of it comes into play. They
are simply not equipped to know what to do to test this new
thing. Now, the idea that the vendor would be able to supply
testing software whose specific purpose is to reveal flaws in
the activities of the vendor doesn't seem to be a stable
situation to me.
There are certain kinds of tests that one naturally
performs, is the processor operating? Are the lights on the
panel operating, et cetera. But if the allegation that has been
made by security specialists is rational, that a vendor could
himself introduce malicious code, or code designed to influence
the outcome of an election, then we certainly can't rely on the
vendor's testing protocols to reveal that.
And so, I believe there have to be nationwide standards
that apply, otherwise we are going to run into an equal
protection issue, that a voter in one state will not be
accorded the same degree of--literal protection against having
his vote tampered with than a voter in another state.
Chairman Ehlers. Ms. Coggins.
Ms. Coggins. I would concur.
Currently, the voting system standards do actually have an
operational test that must be performed, and the labs have to
test for that. But at this point, there is no standard that
tells a jurisdiction how do you go and do this validation? How
can you check to see that the code you have matches the code
that was qualified by the lab, or is certified by your state?
I would suggest that actually, as Dr. Shamos has added that
into the standards. The whole process is jurisdictions are not
exempt from audit, and that audit is not--just because the
voting system has been tested, it doesn't mean that you still
don't have to run the same kind of manual audits that you ran
against your registration system.
It is not yes, you have a computer system, but you know,
test, but verify. I mean, that is--trust, but verify, sorry.
Chairman Ehlers. And test, as well.
Ms. Coggins. Yeah. That is right. And also, just in terms
of the clearinghouse role, you know, that was part of the
original intent of the 1990 implementation plan, that there was
a clearinghouse where all of this information could be reported
back on, on this anecdotal information. If the EAC could
somehow have a reporting mechanism, where you can go online and
you can, as a local jurisdiction, you can type into a database,
and the form is set up in a way that it is--a software
reporting, defect reporting, something along those lines, where
it is structured, where you can really guide people, okay, here
is the information we need you to get. I would also suggest
that, in terms of the overall end-to-end process of education
for elections, you look at putting something out there that can
help local jurisdictions report back to this clearinghouse.
Chairman Ehlers. Thank you all, and I--my time has expired.
I will now yield to the gentleman from Colorado, Mr. Udall.
Mr. Udall. Thank you, Mr. Chairman. I want to thank the
panel as I begin. It was very helpful. As you all know, I think
you raised more questions than you answered, but that is the
purpose of having a hearing.
Should All Computer-based Voting Equipment Be Required to Have
a Paper Trail?
If I could direct a question to Dr. Shamos, I think this
maybe gets at one of the questions we all ask ourselves,
particularly given what Congressman Holt had to say. There are
a number of computer experts that strongly recommend that all
computer-based equipment have a paper ballot trail. You alluded
to this. Congressman Holt alluded to it.
What are your views on this recommendation?
Dr. Shamos. Congressman, there are already requirements in
place for DRE machines in certain states to have paper audit
trails. These are not the so-called voter verifiable audit
trails, but they are a continuous roll of paper that records a
complete ballot image of every vote cast, and in fact, I
haven't recommended certification of any DRE system that didn't
possess such a capability.
We are talking about the voter verified paper trail, the
one that produces a piece of paper that the voter may see, so
that he can verify that his vote has--corresponds to the
buttons that were pressed, or whatever actions had to be taken.
And the idea is that that voter verified ballot is not taken
away from the voting booth with the voter, but is deposited in
a secure receptacle within the machine, so it is available for
some process later on, whether that is an audit, or a recount,
or some other activity associated with an election contest.
I don't have anything against paper in general. The problem
that I have with those proposals, and particularly, that single
sentence in Representative Holt's bill, is the sentence that
says that the paper record shall be the official one, and that
it shall take precedence over the electronic record. The reason
I take issue with it is that this country has a very long and
sorry history of vote tampering, and that vote tampering has
almost exclusively been conducted through the use of physical
ballots, whether they were ordinary paper ballots, punched
cards, mark-sense, or otherwise.
The New York Times, which has recently been so fond of
supporting the concept of a paper trail, has published over
4,700 articles during its history on vote tampering around the
United States with paper ballots. And those 4,700 articles date
back to 1852, and if you do the division, it is that the New
York Times has published such an article on an average once
every 12 days since it began publishing in 1851, it has decried
the use of paper ballots as a way of recording votes. Yet in
2004, when nothing had changed, the New York Times decided
suddenly that paper was the right mechanism.
What has not occurred here, and what the computer
specialists who recommend paper trails have not done, is to do
a security comparison between the security vulnerabilities of
DRE systems and the security vulnerabilities of paper. If, on
balance, paper is safer, then that is the system we should be
using. But it is the reason we don't use paper. The Kolth, or
lever machines, beginning in the 1890's, which led in 1925 to
New York adopting lever machines, was specifically to combat
chicanery with paper ballots.
So once the paper ballot becomes the official one, anybody
who has any mechanical capability at all is able to fiddle with
paper ballots, but they can't fiddle with properly secured
cryptographically encoded electronic records. That is why I am
not in favor of them becoming official.
Mr. Udall. You may be very popular around here, because
there are certainly a lot of people who look for instances in
which the New York Times contradicts itself.
Chairman Ehlers. They are not that hard to find, actually.
Mr. Udall. So in effect, you are saying there are those
that hear all of the arguments about DREs and the problems who
might say why don't we just say to ourselves, look, technology
isn't the answer to everything. Let us just go back to paper
ballots, because they are verifiable. They are in your hand.
There is no hidden software, but you point out that that,
although on the surface, may seem like a viable option, it has
its own problems, and fraught with its own history.
Dr. Shamos. I have asked those experts personally. I said
tell me, make a list of problems that you believe that paper
trails are intended to solve, and then demonstrate to me the
way in which the paper trail solves the problem, and they are
unable to do it with a single exception, and I will give them
this, that when the voter goes into the voting booth, she wants
to be sure that her choices have been properly understood by
the machine. She needs some feedback that says that. The paper,
the piece of paper does, indeed, provide that feedback. There
are numerous other ways of providing it that are completely
electronic, but the paper does it. The fallacy is in believing
that once that piece of paper is cut off and drops into a
receptacle, that it will be around for a recount, that it will
not have been modified. It will not have been deleted. It will
not have been augmented. There is no, absolutely no assurance
that those things will not happen. So, they solve, of the top
10 problems with DREs, it is possible that paper trails solve
one.
Mr. Udall. I see my time has expired. I do fall back to
some extent on an ATM analogy. I know, at least it is my habit.
I deposit some money, or I remove some money, and I get a
little paper receipt, and I stick it in my wallet, and carry it
along with me, and sometimes I check, and sometimes I don't, to
see if that it is, in fact, what has been recorded in my
savings or checking account.
Dr. Shamos. Well, I am glad you raised that analogy,
because if you read Reg E of the Federal Reserve Board, which
requires such paper receipts from ATMs, you will find that the
paper receipt is not the official record of the transaction.
All it is is a piece of evidence, and if there is a discrepancy
between the electronic record and the piece of paper, that is
the starting point for the bank's investigation. It is not the
endpoint, and I believe it should be exactly the same with
voting systems. If there is a discrepancy between the paper
audit trail and the electronic record, that is where we start
looking, and we do a forensic examination to see who did the
tampering. We don't simply take the paper record and say, that
is it. We don't have to look at the electronics any more,
because all that means is we are simply returning to hand-
counted paper ballots.
Mr. Udall. Thank you.
Chairman Ehlers. If I may just interject here, I assume you
would agree with my statement to Mr. Holt that it would not be
too much trouble to program the computer to store one record
that is different from the one that is printed out.
Dr. Shamos. Oh, one can certainly program a computer to do
that.
Chairman Ehlers. Yes.
Dr. Shamos. However, I don't agree that it would be
possible to do that in such a way that it would not be detected
during testing, qualification----
Chairman Ehlers. Yes. Yes. Right. I agree. Next, I am
pleased to yield to Mr. Gutknecht, my friend from Minnesota.
Mr. Gutknecht. Well, thank you, Mr. Chairman, and I want to
thank the distinguished panel today. I appreciate the
testimony.
I am still sort of torn on this whole issue, because I
guess there are sins of omission, there are sins of commission,
and I am not sure how many problems we have with various voting
machines, but I do believe we in the United States and,
frankly, even in my own State, on occasion, have problems with
people who would try to alter the outcome.
In fact, in my own district, we had a very disputed State
senate election last time. We--and it was paper ballots, and
you could say we had an audit trail, and in one of the most
disputed precincts, one of the election judges inexplicably
took a bunch of ballots home and burned them in her fireplace.
Dr. Shamos. It must have been cold.
Technologies for Reducing Voter Fraud
Mr. Gutknecht. It was cold. It was Minnesota, and it was
November, December, by the time they got to this. But I guess
what I really, and maybe this is a question for the folks from
NIST. It seems to me if we are going to get serious about
really cleaning up the elections, we have to do something to
make certain that the people who are actually voting are the
people that say that they are. In other words, most of the
examples, I think, where we have had what I would describe as
voter fraud is where people who were not eligible to vote
voted, and where people may have voted in more than one
precinct, and unfortunately, I think that has been happening
more than most people would like to admit.
And so far, we have talked an awful lot today about, you
know, voting machines and making certain that they tabulate
correctly, and that the voters' wishes are expressed, but I
guess the question I would have is how do you ultimately, as
Mark Twain once observed, you know, we as politician's are
America's only native criminal class, and so there is always
this temptation to figure out ways to tweak the system to
somebody's advantage, and I really have been less concerned
about the tabulation by the machine than I have what some of
the political machines might do to try and change the outcome
illegally.
And have you worked at all on trying to tie those ends
together?
Dr. Semerjian. We have not, so far, but I think that will
be probably one of the major agenda items for the TGDC, in
terms of how do you assure that the person who presents himself
or herself there is the person, and then, how do you--I mean,
we have a lot of different technologies.
Mr. Gutknecht. Right.
Dr. Semerjian. Some of them are being used today, with you
know, some of the magnetic cards that they give you, based on
your presentation of an ID, so I think the technologies are
there. The issue is how are they implemented locally, and a lot
of the uncertainties probably come from local implementation of
these issues. So, frankly, TGDC and the EAC can provide
guidelines, standards, for all those issues, but these are,
after all, voluntary standards. They will be voluntary
standards, so it will be up to the local jurisdictions to
decide how far they go.
Mr. Gutknecht. Well, I thought for a long time, there ought
to be a way that when someone votes, that they leave a
fingerprint, and the technology is relatively simple on
biometrics. I mean, I say that relatively, but--and more
importantly, it is not that expensive nowadays to really
confirm that, you know, that person is who they say they are,
but more importantly, that they haven't voted anywhere else
that day. And I really think that NIST could be helpful in
perhaps bringing some of that technology together, and at least
demonstrating to local election officials that this is
available now, and yes, we could do all we can to make certain
that the technology that we are using is accurate, but at the
end of the day, you know, the other side of that equation is we
have got to make certain that the people who are voting are
eligible to vote, and that they haven't voted more than once.
Dr. Semerjian. Well, I wouldn't have agreed with you three
years ago, but today, certainly, the technology is there,
because of the visa, you know, entry, and that technology is
certainly available. But there are, of course, philosophical
issues. Not everybody is--we don't have everybody's
fingerprint, and how would that be accepted in the community as
a whole? And whether the costs of implementing a system like
that in all of the jurisdictions would be acceptable.
I don't think that is a technology issue.
Mr. Gutknecht. Correct.
Dr. Semerjian. I think it is an implementation issue, cost
issue, and some philosophical issues, whether we will require
the whole country to have, basically, a fingerprint of every
eligible voter.
Mr. Gutknecht. Well, I think if we wait until we have a
complete consensus, we will never move on any kind of a
universal system, so that we do have that kind of technological
guarantee. And that is where I think NIST can play an important
role, as we begin to say to communities and States, look, this
stuff exists, and it can be integrated. Now, it may not happen
overnight, but if you don't start today, you will never get
there. And I really think that is a very important part of this
story, that you know, I am not as worried about the machines
that we use in Minnesota not counting correctly, as I am about
large numbers of people in some precincts that maybe half a
dozen people or a dozen that could change the outcome of a
school board election, or a State legislative election, or even
a Congressional election.
And so, I do hope that as you go forward, you will at least
keep open to that, and try to at least let folks know that this
technology is out there. It is not all that expensive. I think
the concern I have with, you know, with your immediately going
to the philosophical question. You may well be right. But I
think generally speaking, the public has always resisted new
technologies. I mean, there were people who thought putting
electrification inside of houses was ludicrous because people
would die. And of course, they were right. I mean, people have
died from being electrocuted. But, you know, we figured out
that it is a risk we are willing to take, and we take it every
day. And I think that is going to be true with this technology.
I think at first, there will be resistance, but more and more
people realize it is for their protection as well.
I yield back the balance of my time. Thank you.
Role and Ability of NIST to Address Voter Equipment Testing and
Evaluation Issues
Chairman Ehlers. The gentleman's time has expired. I will
ask some additional questions. And let me just interject here,
in the midst of all this gloom and doom about fraud, error, and
so forth, that I am pleased that we live in a country that, by
and large, values the integrity of their elections, and the
majority of the people, in fact, are honest and want honest
elections.
So, it is not all bad news, but the point is, we want to
protect it and make sure that people can be assured, first of
all, that their vote counts, and secondly, that there are no
fraudulent votes counted, and that all votes are counted
accurately.
This is a question for anyone on the panel. How important
is NIST's role at this point in solving the problems we have
discussed here today? What specific assistance do you need from
NIST, or do you think NIST should provide, both what they
already are doing, and what they might potentially do? And then
I would like to ask NIST to respond whether or not they can
meet these needs, and how much funding would be required.
Mr. Wilkey, we will start with you.
Mr. Wilkey. Thank you, Mr. Chairman.
One of the issues that the Chairman of the EAC has come out
with in the last couple of weeks--I mentioned them earlier--he
has called on--and I know that he has personally called every
one of the vendors, and asked them to voluntarily place their
software and source code into the NIST software library--and
this is something that we had been talking about in all of our
discussions, going back a year ago, that one of the great
benefits that NIST brings to this whole program is to be able
to have a single repository for software source code, all of
the versions, because there are so many versions out there. It
is one of the most difficult things that we have to deal with,
or that the ITAs have to deal with, is version control. And to
bring them into this library, similar to the one that they host
now for the law enforcement agencies all over the country,
would be a great benefit to this program.
Let me just interject also, and I may have mentioned this
before, but I--we came away from our initial meetings with NIST
so gratified that the little baby that we tried to raise is now
kind of grown up, and we can turn it over to them, and feel
confident that they are going to give it the day to day
attention that it really needs.
We were particularly gratified because NIST, and we didn't
know this before we began meeting with them, is that NIST has
the ability, being who they are, to bring the very best in
technology to the table to look at these issues, and to study
these issues, and to make the very best recommendations that
they can. And so, we are very pleased from our end, and as non-
technical people. I am not a technician, never claimed to be. I
am just a school teacher who ended up going into the Election
Office 35 years ago, and here I am today.
But I think all of us in NASED who have been working on
this have particularly been very much pleased with what we have
seen at NIST, and we know that they will do a great job in this
area.
Chairman Ehlers. Well, let me just thank you for that
statement, because you have no idea how many objections I
received from members of your organization when I first
proposed NIST.
Mr. Wilkey. Chairman, we were a little skeptical, but we
were quick learners. Let us put it that way.
Chairman Ehlers. More than a little skeptical. Ms. Coggins,
do you have any comment on the question of how important NIST
is, and what the appropriate role is?
Ms. Coggins. I think I would just say that a reexamination
of the voting system standards is appropriate, and we
definitely support, you know, any help that can be provided by
NIST. I think, you know, it is good to have an organization
such as them helping with that process.
Chairman Ehlers. Okay. Dr. Shamos, any comment?
Dr. Shamos. Yes, Mr. Chairman.
I think the nature of voting system standards, they differ
from other kinds of standards. The Chair mentioned we have
Underwriters Laboratories testing of various electrical
devices, so we believe we are safe from shock. But if half the
people who use toasters got electrocuted, we would look very
carefully at what Underwriters Laboratories was doing.
So most people in their daily life do not need to
understand the testing procedures or even the standards that
are being applied to toasters, because our experience is that
they are safe. However, so much hue and cry has been raised
about security problems and reliability problems with voting
systems that I do not believe that the public will be satisfied
with standards that the public cannot observe and understand.
And therefore, I think that the proper role of NIST is to
coordinate the development of standards with massive input from
the public, and massive transparency and visibility, similar to
the way Internet standards are developed, by having Requests
for Comment, engineers all over the world look at the
protocols, make comments, and what happens is that the cream
rises. And if someone has an idea that is bad, there are 100
people who explain why it is bad.
Instead of looking to a super-organization who,
essentially, takes on the mantle of we are the experts, trust
us. The word trust is rapidly disappearing from the process of
voting and counting votes. We can just never get the public to
buy the concept that some distinguished gentleman ought to be
trusted simply because they have been around a long time. And
we need much more public involvement.
What Does NIST Need to Fulfill This Role?
Chairman Ehlers. Thank you. And to wrap this up, Dr.
Semerjian, two questions. Can NIST meet these needs? How much
funding will it require? And HAVA gave you nine months to
develop a standard. Can you meet that deadline?
Dr. Semerjian. First of all, we are very pleased to be
involved in this. Our mode of operation has always been to be
open and transparent in anything. We don't have many smoke-
filled backrooms where things get decided. Indeed, the
standards setting process, everything that we do is open,
through the normal procedure of publishing notices in the
Federal Register, giving sufficient time to people to comment,
or almost invariably, having workshops to not only welcome, but
indeed solicit comments from the public, and the technical
community.
So, I certainly have no reservations in terms of meeting
the kinds of requirements that Dr. Shamos has in mind. I mean,
indeed, this is an area where public trust and confidence, just
the perception, is a very important issue. The fact that
scientists or engineers can sit and convince each other that
this works or this is right is not sufficient. The process has
to be open enough, transparent enough, so that everybody
understands, as he pointed out.
So, it is very important, and indeed, our process of doing
any of these kind of activities, have been along these lines.
And we don't normally just sit and decide on one particular
standard. As you know, in the encryption standards, for
example, we opened the field to the whole world, basically
asked scientists, engineers, to come with proposals for the
kinds of standards that we should have. So, I expect a similar
process. I think our only problem will be we are running on
such a short time scale that----
Chairman Ehlers. And that was a question.
Dr. Semerjian. Yeah.
Chairman Ehlers. Can you meet the time?
Dr. Semerjian. I think so. But it will--I mean, we already
have the--as I pointed out, we have put--the Federal Register
notice came out yesterday. We expect the workshop within a
month or so, and we will certainly give our best try to meet
the nine-month deadline to come up with a draft standard.
Chairman Ehlers. And the funding?
Dr. Semerjian. Well, I guess that is really hard to say,
but we will--I know you are working very hard to come up with
resources for NIST, and we will try to get that done within----
Chairman Ehlers. Yeah. And as you know, I did try to take
some of the funding that was for the new voting machines, and
just divert a very small fraction of that to you, but received
objections from NASED, for which I will never forgive them, and
so that wasn't accomplished. But perhaps that can still be
done.
Thank you. My time has expired. We are rejoined by the
Ranking Member, Mr. Udall, if you have further questions.
Mr. Udall. Thank you, Mr. Chairman.
I am glad to hear that NIST believes that you can get the
job done. But I do think it is incumbent on us to provide you
with the resources, and I hope you will continue to make that
case, as will members of the panel, to the Congress. The
squeaky wheel gets the oil is certainly a principle that works
in the Congress.
I, Mr. Chairman, want to just for the record note that I
talked to our Secretary of State, who I think may be familiar
to some of the panel, Donetta Davidson, last week, and asked
her some questions about what was unfolding in Colorado, and
she is, Mr. Chairman, for the record, she is a moderate,
thoughtful Republican.
Chairman Ehlers. All Republicans are moderate and
thoughtful.
Mr. Udall. Thoughtful at all times, I know. And a well-
respected public servant, and her point was get NIST the
resources, and get NIST on the case, and we can move to where
we need to be, that the 2006 deadline is bearing down upon us,
and that was her focus, not the 2004 election. I do fear that
we may have the potential in 2004 for a repeat of 2000 at the
Presidential election level, but be that as it may, we
certainly have that 2006 deadline to meet.
Dr. Semerjian. Mr. Udall.
Mr. Udall. Yes.
Dr. Semerjian. I don't know if you are aware of it, but Ms.
Davidson is on the TGDC.
Mr. Udall. Yes.
Dr. Semerjian. She is a member of the TGDC, and we are very
pleased to have that expertise on the Committee.
Mr. Udall. She brings, of course, a county perspective,
because she served in that role as the county clerk in Arapahoe
county, which is a very populated county south of Denver, and
now, she is the Secretary of State. And as I mentioned, highly
respected by Members of all parties in Colorado.
What Do States and Other Entities Need to Do to Improve the
Technological Aspects of Elections?
I thought I might try and stir things up with my last
question, because I think the answer to this will--I want to
give everybody a chance, but I want to return to Dr. Shamos,
and he raised a number of questions about the current standards
and testing procedures, as well as some recommendations on how
they could be improved, and I thought I would love to hear from
each of the panel members, your views on Dr. Shamos' testimony
in that regard. Let me start with Mr. Wilkey and move across.
Mr. Wilkey. Thank you for your comments about the Secretary
of State, who is a very good personal friend of mine, and I
have spent so much time in your State lately over the last
year, I nearly meet the qualifications to be able to vote for
you. So, if I defect, or----
Mr. Udall. Mr. Gutknecht may want to get your fingerprints
before you----
Mr. Wilkey. Okay.
Mr. Udall [continuing]. You are allowed to vote.
Mr. Wilkey. While I appreciate Dr. Shamos' statements, I
want to reiterate, as I did in my testimony, that we certainly
have done the very best job we could do on a voluntary basis,
not having any staff, not having any funding, but trying to
keep it going for the benefit of our member States and our
jurisdictions. Certainly, there are some areas that need to be
addressed, and we are hopeful that the Technical Guidelines
Development Committee, which will be having its first meeting
in the next couple of weeks, will be able to address those.
Certainly, and I want to re-emphasize again the role the
states and jurisdictions need to play in this process. You
know, you can take a toaster, for example, as was already
mentioned here, and you can put it in the lab, and you can test
it, and you can, you know, similar to what we do with voting
systems. You know, we put them in a chamber, and run them for
163 hours, and shake them and bake them. And you know, can come
out with a clean bill of health, but if you don't do what is
necessary at the local level, you have lost all of that,
essentially, because you are only testing one unit. And so it
is absolutely necessary that--and something that we have talked
about in NASED for a long time, that our states have to take
the bull by the horn in doing that, similar to what they are
doing in your State, Congressman, in your State, Mr. Chairman,
in some of the states that have put funding to adequately do
this job, places like my own State, where we have a dedicated
staff that does that. The State of Florida, State of
California, Georgia, and others, that have seen the need to
have their own people on staff to be able to continue to make
that whole process work.
And so, I think that this is the most important thing that
we need to understand in this whole process.
Mr. Udall. Ms. Coggins.
Ms. Coggins. Well, I think one of the points is that--and I
don't take it that Dr. Shamos is saying that there is--the labs
have an integrity problem at all. I am not interpreting it--it
is a transparency problem, is part of his view. And we agree
that there can be greater transparency in this process. You
know, I think I keep going back to this 1990 implementation
program.
One of the original issues in that was that the reports
were supposed to be provided--the FEC was going to have this
clearinghouse where they could distribute the reports to the
states, and somehow, that didn't happen, and you wound up
making the person who distributes the report someone who has a
nondisclosure agreement. And so, at this point, whether or not
a state gets a report or not, it depends upon the vendor to
request the lab to send it. You know, I would say in terms of
that, if the State and the local person don't request the
report, the report remains confidential because they allow it
to remain confidential.
But we agree that there can be greater transparency in the
process. We have also tried to be, by coming here today, and
other things that we do to support NASED, we have gone before
the California taskforce. We went to the NIST conference. We
try and get our processes out. Quite frankly, I start talking
and eyes glaze over in one minute, when you start talking about
test process. So, I know that there is an interest in greater
participation, and we definitely feel that, you know,
transparency, in terms of reports, in terms of the
accreditation, we don't have an issue with that.
Mr. Udall. Dr. Semerjian, I think maybe it is--I don't know
if it is inappropriate for you to answer, but I know this is
what--the area in which you are going to do some work. If you
feel comfortable responding, I would welcome your input.
Dr. Semerjian. No--I see no reason why reports should not
be available, whether it is the accreditation report, or the
test reports. The other comment, I thought Dr. Shamos was
making, that you know, if people send you a chip, and you know,
somebody can just plug it in, and that is perfectly okay, that
is not an acceptable procedure under ISO 17025 standard. You
can't just plug things in and take things out, make changes,
without proper notification and proper documentation of those
changes. So, I think just by implementing more rigorous test
procedures and standards, I think we should be able to get over
some of those difficulties.
I think his concern is well-placed, in the sense that we
need to be worrying about not just a box, a piece of apparatus
here. We need to--or just the chip inside. We need to worry
about the integrity of the whole system, the whole system being
first, the machine itself, and second, not just the voting
machine, but how does that data get transmitted to a central
location where the vote is tallied, etc.
So, clearly, our concerns have to be not just limited to
one particular box, one particular element of the system, but
the entirety of the system. I think clearly, we have to look at
the totality of the systems that are being used.
Mr. Udall. Spoken like the head of NIST. Thank you, and
again, I want to thank the panel. I think we are going to
conclude here. But what I heard, Dr. Shamos, you saying in the
end, that this is more about human error than it is about
fraud, although we always have to be worried about fraud. But
that, in the end, that is more where you would place your
concerns, given the multiplicity of systems around the country,
and the difficulty in arranging some sort of a fraudulent
conspiracy, if you will.
Dr. Shamos. Well, we must be thoroughly vigilant to make
sure that the systems are not vulnerable to fraud. We shouldn't
engage in the fantasy that electronic frauds are going on all
the time, or that that is the major problem that we face in
elections. Most of the incidents that I have read about involve
machines that simply won't boot up properly on election day.
That has nothing to do with fraud. What it has to do with is
either people not knowing how to turn them on, or machines that
have not been adequately tested, stored, or set up properly.
That is an education problem.
But I am certainly not suggesting that security is not a
vital concern.
Mr. Udall. And thanks to the panel. This is very
informative.
Chairman Ehlers. The gentleman's time has expired. Mr.
Wilkey, did you have something you wanted to say? It looked
like you wanted to make a comment.
Mr. Wilkey. And I would like to do just a quick followup on
the question that Congressman Udall asked.
You know, we have often been accused of--because this was a
voluntary effort of having a rinky-dink process here, this is
the handbook that NASED uses to qualify our ITAs. When we move
this process over to NIST, they will use a process called ISO
17025. It is a very familiar accreditation standards for
independent test authorities. It is almost a carbon copy of the
handbook that we have been using for a number of years, because
it was developed when the first draft of 17025 was being
drafted by the late Bob Naegele, who did all of this work, and
who worked closely with NVLAP and NIST at that time.
Further, we have had some of our own questions regarding
these reports. We have consistently told our member States that
have been involved in this program, and believe me, it took a
long time to get 41 states to adopt the voluntary federal
standards, a lot of talking, and a lot of arm-twisting. But we
finally did it, and one of the things that we have consistently
told these states is that they must get copies of these reports
turned over to State ITAs, if they have a State ITA, or if they
are a large jurisdiction buying a voting system, that it needs
to be part of their contract with the vendor, that you don't
sell a product here unless we see a copy of this report, or
have it reviewed by somebody that is willing to do a
confidentiality agreement.
I agree that it has not been, it has been the most
disconcerting of everything we have done, because of the fact
that there has been so little funding available for us to be
able to go out and do this on our own, it was necessary for the
vendor to pay for this process, and it has been a very
expensive process, at least if you listen to them screaming and
hollering. And so, that product becomes their property, but
that in no way means that a State or a jurisdiction cannot get
their hands on this report if they go through the right process
to do so, and we have encouraged them to do that.
Chairman Ehlers. Thank you very much. Just a few other
comments. First of all, I have, over the years as a county
commissioner, State house member, State senator, and now here,
worked with many county clerks, city clerks, township clerks,
and poll workers, and I have to say that by and large, they are
the salt of the Earth. They are very dedicated, they really
want to do it right, and we have to recognize that. And so, our
purpose here is not to condemn them, or to denigrate them, but
simply say we want to try to help you to do it right.
We also have to recognize the federal role in elections is
Constitutionally limited. We, of course, can worry about
federal elections, but there are a lot of other elections,
city, township, State, and so forth, that we do not have
jurisdiction over, unless there is evidence of fraud that our
Justice Department would have to investigate.
So, and it has been a very difficult road to get where we
are. I am pleased with where we are, except we should have been
here two years ago. But we will get this done, and we will have
a safe and secure system to the extent possible.
I, also, with Mr. Gutknecht's comment about fingerprints, I
was reminded of an election story--this is true--some years
ago, in an unnamed jurisdiction, where the county political
boss was in the habit of registering his dog to vote as well as
himself, and this became general knowledge, and the people just
sort of lived with it. However, he overreached when he
registered the dog in three different precincts, and the dog
voted in all three precincts. That was the end of the political
boss. So, fraud is not exactly new, and not even imaginative.
But it is pleasure to thank you for your participation
here. It has been a good hearing, and we are very pleased with
the progress. As I say, it is later than we would like, but we
are looking for good results, and we hope the next election,
Presidential or otherwise, will be far better than it was four
years ago.
I thank the panelists for coming here. You have all
contributed substantially to the hearing, and I appreciate it.
If there is no objection, the record will remain open for
additional statements by the Members, and for answers to any
follow-up questions that the Subcommittee may ask of the
panelists by writing, and we would appreciate it if you would
respond to those if we send them to you.
Without objection, so ordered, and the hearing is now
adjourned.
[Whereupon, at 4:01 p.m., the Subcommittee was adjourned.]
Appendix:
----------
Answers to Post-Hearing Questions
Responses by Carolyn E. Coggins, Director, ITA Services at SysTest Labs
Q1. How do the standards and testing methodologies for voting
equipment differ from standards and testing methods for other kinds of
equipment that your company tests? Are tests for voting equipment
generally more or less specific or comprehensive?
A1. An ITA performs two distinct types of testing on electronic Voting
Systems. These are as follows:
1. Software and integrated system testing where the focus is
on testing to ensure the functionality, security, accuracy,
etc., of either software or firmware.
2. Hardware environmental testing where the focus is on
ensuring that custom developed hardware meets all applicable
environmental standards.
Hardware Environmental Testing: The standards for this are fairly
specific and straightforward, having been derived from the
manufacturing industry for hardware components and hardware devices.
The methods used for hardware environmental testing are very similar to
methods used for testing other kinds of equipment. The requirements
within the 2002 Federal Election Commission Voting System Standards,
VSS, and methods for hardware environmental testing directly resemble
the international standards and many of the standards within the VSS
either call out or reference both national and international hardware
environmental testing standards, e.g., FCC, OSHA, ISO and Mil
standards.
Software and Integrated System Testing: The methods for testing
software and integrated systems can be as varied as there are different
software applications and industries. In addition, although standards
from the FEC (the 2002 VSS), IEEE, the SEI, FDA, DOD, ISO and others
exist for software, there is no uniformly adopted testing approach for
the software development world. SysTest Labs has a testing methodology
that governs our testing processes and procedures. This methodology,
Advanced Test Operations ManagementTM (ATOMTM) ensures that SysTest
Labs follows the same basic techniques, regardless of the type of
system. ATOMTM was audited by the NASED Auditors and approved for use
in testing of electronic Voting Systems. Having ATOMTM in place at
SysTest Labs ensures that we take a robust and repeatable approach to
each and every test effort, from banking systems to electronic Voting
Systems. The only difference between our testing of electronic Voting
Systems and other systems is in the depth of testing.
The depth of testing for other systems is defined by many factors.
SysTest Labs has separated systems into three basic categories related
to the criticality or the magnitude of impact/risk of the system. These
are:
Low Criticality or Magnitude of Impact/Risk: General Commercial
Testing is performed to customer requirements.
Customer assesses the risk and determines if testing
is sufficient.
Testing is often viewed as a cost center item as
opposed to a profit center item. Customers or Vendors may try
to minimize the time and money spent on testing.
There are no uniformly adopted standards for these
types of systems and the methods for testing can vary from ad
hoc (no planning) to extremely systematic and robust.
Acceptance criteria: Sufficient can be fluid,
responding to influences like the benefit of ``first to
market'' and budgets.
Medium Criticality or Magnitude of Impact/Risk (e.g., Electronic VOTING
SYSTEMS, Gaming, Telecommunications, Banking, and others)
Testing can be required to meet regulatory standards
with either government or fiduciary oversight.
Testing is still viewed as a cost center item as
opposed to a profit center item. This translates to customers
or Vendors trying to minimize the time and money spent on
testing.
Level of testing is determined by financial risk,
penalty, or governed by published guidelines and/or standards.
Acceptance criteria: Customer may set the acceptance
criteria or the acceptance criteria may be defined by
regulatory standards. The customer may define which
requirements the system will meet, i.e., the regulations or
standards do not force the customer to meet all system
requirements but a minimum set of requirements.
High Criticality or Magnitude of Impact/Risk: (e.g., DOD, NASA, FDA):
Life critical systems.
The systems must meet very stringent standards and
requirements.
The methods used for testing are required to meet
very stringent standards and requirements.
Oversight and enforcement by DOD, NASA or the FDA.
Comprehensive level of testing determined by class;
class defines severity of risk, i.e., life and/or injury.
Acceptance Criteria: Meets all requirements and
standards, must be free of defects. Per the 2002 VSS and NASED
guidelines, an ITA is required to ensure that a voting system
being tested meets the ``minimum requirements of the voting
system standards.'' The VSS specifies what minimum set of
requirements a voting system must meet in order to be
recommended for qualification. The VSS does not specify an
exhaustive set of requirements for software and these
requirements tend to be at a very high level leaving
significant room for interpretation by both the Vendor and ITA.
It is not to say that all voting systems tendered for ITA
Qualification only meet the minimum requirements of the VSS.
However, it is important to recognize that the intent of the
standards is to define a minimum set of requirements that all
voting systems must meet in order to be recommended for
qualification at a federal level. The individual functionality
required by each state is not addressed in the standards other
than to task the ITA to test additional functionality to the
``Vendor's requirements.'' This assumes the Vendor designed to
the correct requirement.
ITA software and integrated system testing for voting equipment is
very specific. All voting systems submitted for testing must pass a
standard set of tests based upon the minimum requirements of the VSS,
customized to the individual voting system design. However it is
generally less comprehensive than testing for other systems. This is,
in part, because the VSS requirements stipulate that the Vendor has an
internal quality assurance and testing program. ITAs may accept Vendor
testing if a review of their pre-submission testing is found to be
comprehensive. Unlike other testing we perform, we cannot make
recommendations regarding the design of a system. In testing a system
we must remain impartial. We can make observations about a design or
function that is less than optimal but if it meets the VSS, we cannot
withhold a recommendation. Although testing has shown that many Vendors
exceed the VSS, when an issue is encountered and there is a dispute
between the Vendor and the ITA, the Vendor will assert that the ITA's
charter is to hold them to ``the minimum requirements of the
standards.''
Q2. To your knowledge, do the tests used by SysTest to evaluate the
performance of voting machines differ from the tests used by the other
Independent Testing Authorities? Does NIST need to develop uniform
testing procedures that would require every lab to use exactly the same
test?
A2. SysTest Labs believes that the hardware environmental tests
performed between Wyle Labs and SysTest Labs are virtually the same.
Again, these types of tests have been required for hardware components
and devices for some time and are standard throughout the industry.
SysTest Labs does not have access to the tests used by Ciber as a
Software ITA. Therefore, SysTest Labs cannot provide an objective
determination of whether or not our tests differ. SysTest Labs can
state that within the last three years, our methods and tests have been
reviewed by NASED Auditors (at least four times) and that software
testing for our first ITA test effort was closely monitored and
observed by the Auditors during late 2001 and early 2002.
Having NIST develop a uniform set of software testing procedures
would be very difficult. Each electronic Voting System will have a
different approach and solution to meeting the requirements of the VSS.
For example, touch screen devices can take the form of small screens,
full-face ballots, systems that produce paper ballots from the touch
screen, etc. In addition, the solution for election management systems
can take many different forms depending on the database, reporting
mechanisms, etc. This is the challenge that Software Testing faces when
designing tests to ensure that an electronic Voting System meets the
requirements of the VSS. The overall objectives will generally be the
same, but the specific steps required to test out functionality will
vary greatly from system to system. In addition, since there are
requirements within the VSS that are not mandatory, some systems will
require tests that others may not (depending on whether or not the
Vendor states that they support the optional requirements).
An alternative would be for NIST to work with the ITAs and
together, design and develop the following items:
1. Testable scenarios and objectives for ballots, contests,
voting, tabulating, etc., or identify specific types of tests,
configurations, ballots, contests, etc. but allow the ITA lab
to control their actual test procedures.
2. Provide State-by-State requirements for handling of voting
variations. (Help identify conflicting requirements.)
3. Define and standardize the format, data, and acceptance
criteria upon which the ITA must report.
Q3. Besides the recommendations you provided in your testimony on what
specific kinds of computing problems need to be addressed by NIST
during standards development, are there other activities that NIST
could carry out to help the ITAs improve the testing process?
A3. SysTest Labs suggests the following items that NIST could carry out
to help the ITAs improve the ITA Qualification Testing process:
1. Issue technical bulletins and clarifications as needed for
ITA Qualification Testing.
2. Develop a process for reporting disagreements between the
ITA and the Vendors regarding interpretation of the VSS
requirements or when an ITA requires a ruling on an issue with
a Vendor's system.
3. Standardize the reporting elements. Provide a Qualification
Report format and structure that allows ``apples to apples''
comparisons of reports.
4. Provide state-by-state requirements for handling of voting
variations. (Help identify conflicting requirements.) This is
not only beneficial to the ITA but providing this information
to Vendors will help ensure that they build better voting
systems.
5. Recognize and understand that testing of an electronic
Voting System is not just the responsibility of the ITA
Define what should be considered public information
and what should remain proprietary.
Provide a basic set of guidelines for testing at
state certification and local acceptance testing
levels.
Provide guidelines and methods to local
jurisdictions on the use of on-going Vendor services
for programming and acknowledge that local
jurisdictions have responsibilities for performing
independent testing or oversight of Vendor ballot
programming.
A representative from NIST must be required to read
and evaluate qualification and certification reports.
Include report criteria in the standards so that there
is a common output with a focus on providing
information that can be used and understood by state
and local election officials.
Help the EAC to develop a common definition for all
functional elements of a voting system including
software, hardware, and documents.
Help the EAC to define a clear process and timeline
for submitted Qualification Report review and
acceptance/rejection by the EAC and NIST. (Method of
submission, timeframe to review, method of acceptance/
rejection, veto, appeals, etc.)
Help the EAC to develop a document and library
structure as the clearinghouse for Qualified Voting
System software and hardware systems.
Help the EAC to define the clearinghouse role and
identify responsibilities: report retention, source
code and executable retention, voting system
documentation retention, policy for access to reports,
policy for obtaining/tracking results of state
certification, and national database to track voting
system problem reports.
Answers to Post-Hearing Questions
Responses by Hratch G. Semerjian, Acting Director, National Institute
of Standards and Technology (NIST)
Q1. To your knowledge, do the test protocols used by testing
laboratories to evaluate similar or identical pieces of equipment (not
necessarily voting equipment) vary widely among different testing labs,
or do they use identical tests? If there is a significant variation,
does NIST need to develop uniform testing procedures for voting
equipment as part of its responsibilities under the Help America Vote
Act.
A1. NIST has no information about the test protocols used in the past
by the NASED testing laboratories (ITAs). However, a well-written test
protocol is always preferable to a less well-written test protocol.
NIST could contribute considerably to the development of test protocols
that are within NIST's scope of expertise. The improved test protocols
would most likely result in better agreement among EAC accredited
laboratories.
In general, when detailed test protocols are used, e.g., the IEC
61000 series (see http://www.iec.ch/about/mission-e.htm), different
laboratories would be expected to report equivalent test results and
when test protocols are not detailed, it is not possible to determine,
in advance, if equivalent test results will be reported. When a test
method involves sampling, the results will depend on the sample.
Voting equipment and systems are usually tested four times: during
product development, qualification testing, certification testing, and
acceptance testing. At each stage, there is the possibility of
different test methods being used. In some cases, a different test
method must be used, e.g., determination of inter-operability of system
components versus conformance of a component to a specification or
determination that the system incorporates the laws of a particular
locality.
Q2. Mr. Shamos says in his testimony that the performance of a
particular machine against national standards is considered
proprietary. Should that information be revealed to the public?
A2. Within recognized national and international accreditation
programs, accredited laboratories are not permitted to reveal
proprietary or confidential information belonging to their clients. A
vendor may share a test report that it owns with anyone that it wishes.
A laboratory may provide information only if specifically requested to,
in writing, by the owner of the information.
Intellectual property rights must be respected. A requirement to
reveal information may violate those rights. Unless the specifications,
standards, test methods, test results, interpretations, and
requirements are all provided, a statement of ``performance'' would be
meaningless and potentially damaging to some or all of the parties
involved in the contract.
As the rule-making body under HAVA, the EAC could choose to require
the public disclosure of certain information about voting systems as
part of an accreditation process. States and localities could do the
same. There would have to be publicly available requirements and
conditions defining the requirement. The EAC, the States, or localities
could require disclosure of information in the contract between vendor
and purchaser. That information could, by contract, be declared
publicly available or proprietary, again by the EAC and not NIST.
Q3. What laboratories have indicated their interest to NIST in
becoming testing laboratories under HAVA and how long do you anticipate
the accreditation of these labs to take?
A3. As a matter of procedure, the National Voluntary Laboratory
Accreditation Program, NVLAP, does not reveal the names of laboratories
that express an interest in NVLAP programs or accreditation (http://
ts.nist.gov/ts/htdocs/210/214/214.htm).
In August, NIST/NVLAP held an initial workshop to gauge interest
within the laboratory community (see: http://ts.nist.gov/is/htdocs/210/
214/whatsnew.htm). An archived webcast of the workshop is available for
viewing at: http://www.eastbaymedia.com/NVLAPworkshop.
Approximately 10 laboratories attended the initial workshop. They
were not all voting systems laboratories. They may or may not be
interested in becoming accredited. A formal call for interested
laboratories will be made shortly. Another workshop will likely follow
in the December time frame.
The length of time it takes to accredit laboratories depends on the
laboratories and how ready they are to meet ISO 17025 standards for
laboratory accreditation. The laboratories must meet the requirements
of NIST Handbook 150 (http://ts.nist.gov/ts/htdocs/210/214/docs/final-
hb150-2001.pdf) and any program specific requirements (yet to be
developed). Given the complexity of this program, it could well take
one year for the laboratories to meet the requirements, be assessed,
resolve findings, and receive accreditation. In addition to the writing
of program specific requirements, it is necessary to identify and train
appropriate assessors. Assessor teams of one or more experts will be
assigned for each laboratory. The size and make-up of the assessor team
will depend on the scope of accreditation of the laboratory. Because of
the uncertainty involved in the accreditation process, the EAC could
decide to ``grandfather'' the current ITAs (laboratories), for a period
of time to maintain continuity.