[House Hearing, 111 Congress]
[From the U.S. Government Publishing Office]
CYBERSECURITY ACTIVITIES AT NIST'S
INFORMATION TECHNOLOGY LABORATORY
=======================================================================
HEARING
BEFORE THE
SUBCOMMITTEE ON TECHNOLOGY AND INNOVATION
COMMITTEE ON SCIENCE AND TECHNOLOGY
HOUSE OF REPRESENTATIVES
ONE HUNDRED ELEVENTH CONGRESS
FIRST SESSION
__________
OCTOBER 22, 2009
__________
Serial No. 111-59
__________
Printed for the use of the Committee on Science and Technology
Available via the World Wide Web: http://www.science.house.gov
______
U.S. GOVERNMENT PRINTING OFFICE
52-857 WASHINGTON : 2009
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing
Office Internet: bookstore.gpo.gov Phone: toll free (866) 512-1800; DC
area (202) 512-1800 Fax: (202) 512-2104 Mail: Stop IDCC, Washington, DC
20402-0001
COMMITTEE ON SCIENCE AND TECHNOLOGY
HON. BART GORDON, Tennessee, Chair
JERRY F. COSTELLO, Illinois RALPH M. HALL, Texas
EDDIE BERNICE JOHNSON, Texas F. JAMES SENSENBRENNER JR.,
LYNN C. WOOLSEY, California Wisconsin
DAVID WU, Oregon LAMAR S. SMITH, Texas
BRIAN BAIRD, Washington DANA ROHRABACHER, California
BRAD MILLER, North Carolina ROSCOE G. BARTLETT, Maryland
DANIEL LIPINSKI, Illinois VERNON J. EHLERS, Michigan
GABRIELLE GIFFORDS, Arizona FRANK D. LUCAS, Oklahoma
DONNA F. EDWARDS, Maryland JUDY BIGGERT, Illinois
MARCIA L. FUDGE, Ohio W. TODD AKIN, Missouri
BEN R. LUJAN, New Mexico RANDY NEUGEBAUER, Texas
PAUL D. TONKO, New York BOB INGLIS, South Carolina
PARKER GRIFFITH, Alabama MICHAEL T. MCCAUL, Texas
STEVEN R. ROTHMAN, New Jersey MARIO DIAZ-BALART, Florida
JIM MATHESON, Utah BRIAN P. BILBRAY, California
LINCOLN DAVIS, Tennessee ADRIAN SMITH, Nebraska
BEN CHANDLER, Kentucky PAUL C. BROUN, Georgia
RUSS CARNAHAN, Missouri PETE OLSON, Texas
BARON P. HILL, Indiana
HARRY E. MITCHELL, Arizona
CHARLES A. WILSON, Ohio
KATHLEEN DAHLKEMPER, Pennsylvania
ALAN GRAYSON, Florida
SUZANNE M. KOSMAS, Florida
GARY C. PETERS, Michigan
VACANCY
------
Subcommittee on Technology and Innovation
HON. DAVID WU, Oregon, Chair
DONNA F. EDWARDS, Maryland ADRIAN SMITH, Nebraska
BEN R. LUJAN, New Mexico JUDY BIGGERT, Illinois
PAUL D. TONKO, New York W. TODD AKIN, Missouri
DANIEL LIPINSKI, Illinois PAUL C. BROUN, Georgia
HARRY E. MITCHELL, Arizona
GARY C. PETERS, Michigan
BART GORDON, Tennessee RALPH M. HALL, Texas
MIKE QUEAR Subcommittee Staff Director
MEGHAN HOUSEWRIGHT Democratic Professional Staff Member
TRAVIS HITE Democratic Professional Staff Member
HOLLY LOGUE Democratic Professional Staff Member
DAN BYERS Republican Professional Staff Member
VICTORIA JOHNSTON Research Assistant
C O N T E N T S
October 22, 2009
Page
Witness List..................................................... 2
Hearing Charter.................................................. 3
Opening Statements
Statement by Representative David Wu, Chairman, Subcommittee on
Technology and Innovation, Committee on Science and Technology,
U.S. House of Representatives.................................. 6
Written Statement............................................ 6
Statement by Representative Adrian Smith, Ranking Minority
Member, Subcommittee on Technology and Innovation, Committee on
Science and Technology, U.S. House of Representatives.......... 7
Written Statement............................................ 7
Prepared Statement by Representative Harry E. Mitchell, Member,
Subcommittee on Technology and Innovation, Committee on Science
and Technology, U.S. House of Representatives.................. 8
Witnesses:
Ms. Cita M. Furlani, Director, Information Technology Laboratory,
National Institute of Standards and Technology
Oral Statement............................................... 9
Written Statement............................................ 10
Biography.................................................... 15
Dr. Susan Landau, Distinguished Engineer, Sun Microsystems,
Burlington, MA
Oral Statement............................................... 16
Written Statement............................................ 17
Biography.................................................... 20
Dr. Phyllis Schneck, Vice President, Threat Intelligence, McAfee
Corporation
Oral Statement............................................... 21
Written Statement............................................ 23
Biography.................................................... 26
Mr. William Wyatt Starnes, Founder, CEO, and President,
SignaCert, Inc.; Founder, Tripwire, Inc.
Oral Statement............................................... 27
Written Statement............................................ 28
Biography.................................................... 36
Dr. Fred B. Schneider, Samuel B. Eckert Professor of Computer
Science, Cornell University
Oral Statement............................................... 37
Written Statement............................................ 38
Biography.................................................... 41
Mr. Mark Bohannon, General Counsel and Senior Vice President for
Public Policy, Software & Information Industry Association
(SIIA)
Oral Statement............................................... 42
Written Statement............................................ 44
Biography.................................................... 47
Discussion....................................................... 47
Appendix: Answers to Post-Hearing Questions
Ms. Cita M. Furlani, Director, Information Technology Laboratory,
National Institute of Standards and Technology................. 58
Dr. Susan Landau, Distinguished Engineer, Sun Microsystems,
Burlington, MA................................................. 59
Dr. Phyllis Schneck, Vice President, Threat Intelligence, McAfee
Corporation.................................................... 60
Mr. William Wyatt Starnes, Founder, CEO, and President,
SignaCert, Inc.; Founder, Tripwire, Inc........................ 61
Dr. Fred B. Schneider, Samuel B. Eckert Professor of Computer
Science, Cornell University.................................... 63
Mr. Mark Bohannon, General Counsel and Senior Vice President for
Public Policy, Software & Information Industry Association
(SIIA)......................................................... 65
CYBERSECURITY ACTIVITIES AT NIST'S INFORMATION TECHNOLOGY LABORATORY
----------
THURSDAY, OCTOBER 22, 2009
House of Representatives,
Subcommittee on Technology and Innovation,
Committee on Science and Technology,
Washington, DC.
The Subcommittee met, pursuant to call, at 2:07 p.m., in
Room 2318 of the Rayburn House Office Building, Hon. David Wu
[Chairman of the Subcommittee] presiding.
hearing charter
SUBCOMMITTEE ON TECHNOLOGY AND INNOVATION
COMMITTEE ON SCIENCE AND TECHNOLOGY
U.S. HOUSE OF REPRESENTATIVES
Cybersecurity Activities at NIST's
Information Technology Laboratory
thursday, october 22, 2009
2:00 p.m.-4:00 p.m.
2318 rayburn house office building
1. Purpose
On Thursday, October 22, 2009 the Subcommittee on Technology and
Innovation of the Committee on Science and Technology will hold a
hearing to review the recommendations made in the Cyberspace Policy
Review that may be appropriate for the National Institute of Standards
and Technology (NIST) and the proposed reorganization of the NIST
Information Technology Laboratory.
2. Witnesses
Ms. Cita Furlani is the Director of the Information Technology
Laboratory at NIST.
Dr. Susan Landau is a Distinguished Engineer at Sun Microsystems. She
is a former member of the Commission on Cyber Security for the 44th
Presidency and the NIST Information Security and Privacy Advisory
Board.
Dr. Fred Schneider is the Samuel B. Eckert Professor of Computer
Science at Cornell University and a current NIST Information Security
and Privacy Advisory Board member.
Dr. Phyllis Schneck is the Vice President of Threat Intelligence at
McAfee. She served as a commissioner for the Commission on Cyber
Security for the 44th Presidency and on the National Board of Directors
for the Federal Bureau of Investigation's InfraGuard.
Mr. William Wyatt Starnes is the Founder and CEO of SignaCert, Inc. He
is formerly a member of the NIST Visiting Committee on Advanced
Technology.
Mr. Mark Bohannon is the General Counsel and Senior Vice President,
Public Policy at Software & Information Industry Association (SIIA).
Prior to working at SIIA, Mr. Bohannon was the Chief Counsel for
Technology at the U.S. Department of Commerce where he helped oversee
NIST cybersecurity activities.
3. Brief Overview
On May 29, 2009, the Administration released its 60-day review of
federal cybersecurity activities entitled, ``Cyberspace Policy
Review.'' The review team acknowledged the difficult task of addressing
cybersecurity concerns in a comprehensive fashion due to the large
number of federal departments and agencies with cybersecurity
responsibilities and overlapping authorities. The document detailed a
number of near-term and mid-term action plans and stated that it would
not only take increased organization and coordination within the
Federal Government, but extensive public-private partnerships and
international collaboration to achieve these recommendations.
The witnesses were asked to address any recommendations from the
Cyberspace Policy Review, focusing on three specific recommendations:
the need for a single locus for Federal Government involvement in
international standards, an increased public awareness and education
campaign, and a larger focus on identity management.
4. NIST Background
The NIST Information Technology Laboratory (ITL) is currently
organized into six divisions that perform research and development in
the areas of network technology, computer security, information access,
mathematics, statistics, software and systems. ITL has a budget request
of $72 million for FY 2010.
Computer Security Division (CSD)
CSD is tasked with protecting the federal non-classified
information technology network by developing and promulgating cyber
security standards for federal civilian network systems. CSD developed
minimum security requirements for these systems in Federal Information
Processing Standard (FIPS) 200. CSD also does work in cryptology,
electronic identity management, methodology for assessing effectiveness
of security requirements, and developing tests to validate security in
information systems. Cybersecurity tasks were appointed to NIST in the
Computer Security Act of 1987 (P.L. 100-235), the Cyber Security
Research and Development Act of 2002 (P.L. 107-305), and the Federal
Information Security Management Act of 2002 (P.L. 107-347).
Advanced Network Technologies Division (ANTD)
ANTD works to improve the quality of networking specifications and
is currently focusing on advanced areas of cryptography, domain name
system security, and evaluation of wireless networks for first
responder communication.
Information Access Division (IAD)
IAD provides measurements and standards in areas such as speech
recognition, biometrics, and inter-operability of interactive
technologies.
Mathematical and Computational Sciences Division (MCSD)
MCSD performs research and development in areas of mathematical
modeling, mathematical software, and their scientific applications.
Software and Systems Division (SSD)
SSD develops software testing tools and methods to improve the
quality of software and testing in areas such as health care
information technology, computer forensics, and voting systems.
Statistical Engineering Division (SED)
SED provides statistical consulting to the NIST laboratories and
performs statistical research to improve statistical modeling and data
analysis.
5. Issues and Concerns
Recommendations from the Cyberspace Policy Review
The Technology and Innovation Subcommittee has asked the witnesses
to discuss recommendations from the Cyberspace Policy Review that may
be appropriate for NIST and to specifically address three of the
recommendations:
The need for a single locus for Federal Government
involvement in international cybersecurity technical
standards--Currently, the United States is represented by an
array of standards setting organizations, both federal and
private industry. The Cyberspace Policy Review calls for a
single entity to coordinate federal representation for
cybersecurity technical standards and develop an engagement
plan for use with international standards bodies.
The need for an increased public awareness and
education campaign--the CSD currently conducts limited
cybersecurity outreach and education through its Small Business
Corner. Also, NIST has a well-established program called the
Manufacturing Extension Partnership (MEP) that provides
services and information to businesses from regional MEP
Centers. NIST can expand upon these resources to increase
cybersecurity education and public awareness amongst private
citizens and small business, as well State, local, and Federal
governments.
The need for a larger focus on identity management--
The Cyberspace Policy Review states that cybersecurity cannot
be improved without improving identity management. It goes on
to say that identity management is not only about
authenticating people, but that online transactions involve
trustworthy data, hardware, and software for networks and
devices. As noted above, NIST has extensive expertise in
identity management areas throughout its six divisions.
The report states that future cybersecurity technical standards
plans must address the convergence of information technologies and
infrastructures. NIST represents an opportunity to address these
recommendations because of its broad array of expertise in
cybersecurity technology standards and established relationships with
private industry and international standards organizations.
Reorganization of ITL
The ITL Director, Ms. Furlani, has proposed a reorganization that
would, as part of its actions, split the CSD and combine its programs
with others to form two new divisions. Cybersecurity experts are
concerned that the split of CSD will take focus away from cybersecurity
and are not clear on how the reorganization will improve the function
and future capabilities of ITL. Witnesses were asked to assess the
reorganization and discuss how it may improve the outcomes of ITL
activities.
Chairman Wu. This hearing will now come to order. I would
like to welcome everybody this afternoon to this hearing on
cybersecurity, and we increasingly put all sorts of
information, including personal information, online. Our
nation's entire infrastructure, from traffic systems and air
traffic control to manufacturing to power distribution, depends
on internet networked systems. I can think of few topics as
important for this subcommittee to address than cybersecurity.
And I want to welcome all witnesses here this afternoon for
this very, very important hearing.
As anyone who has seen movies recently, including movies
like Ocean's Eleven, thieves have become increasingly
sophisticated in their method of heists, and it should be no
surprise that cybercriminals in real life are becoming also
more sophisticated in their crimes.
Congress realized the dangers of networked systems as far
back as the 1980s, and in 1987, this committee wrote the
Computer Security Act, which charged NIST (National Institute
of Standards and Technology) with developing the technical
standards to protect non-classified information on federal
computer systems. Congress has remained concerned about cyber-
threats, and since 1987, Congress has passed 13 laws related to
cybersecurity.
Today OMB (Office of Management and Budget) reports that
federal agencies spend approximately $6 billion per year on
cybersecurity to protect a $72 billion IT (Information
Technology) infrastructure. In addition, the Federal Government
funds $356 million in cybersecurity research each year. I don't
believe that simply spending more money or creating more
programs is the means to improve cybersecurity. We also need to
use our existing resources more efficiently and with specific
achievable goals in mind. This is also the main conclusion of
the Administration's recent cybersecurity review.
The focus of today's hearing is not to review what NIST has
done but to address what should be its focus going forward.
Since NIST is the only federal agency tasked with protecting
non-classified federal computer systems, the testimony we hear
today will have a vital and long-lasting affect on our nation's
economic and national security.
We have a distinguished panel of witnesses who have a long
history of working with NIST and detailed knowledge of NIST's
IT activities. I want to assure them that this subcommittee is
prepared to act on their recommendations.
And now I would like to recognize Ranking Member
Representative Smith for his opening statement.
[The prepared statement of Chairman Wu follows:]
Prepared Statement of Chairman David Wu
I want to welcome everyone to this hearing on cybersecurity. More
and more of our personal information is making its way online, and our
nation's entire infrastructure--from traffic systems and air traffic
control to manufacturing--depends on Internet networked systems. I can
think of no topic more important for this subcommittee to address than
cybersecurity.
As anyone who has seen Ocean's Eleven can tell you, thieves have
become increasingly sophisticated in their heists. It should be no
surprise that cybercriminals are also becoming progressively
sophisticated in their crimes.
Congress realized the inherent dangers in networked systems as far
back as 1987, when this committee wrote the Computer Security Act,
which charged NIST with developing the technical standards to protect
non-classified information on federal computer systems. Congress has
remained alert to cyber-threats. Since 1987, Congress has passed 13
major laws related to cybersecurity.
Today OMB reports that Federal agencies spend $6 billion on
cybersecurity to protect a $72 billion IT infrastructure. In addition,
the Federal Government funds $356 million in cybersecurity research
each year. I don't believe simply spending more money or creating more
programs is the means to improve cybersecurity. We need to use our
existing resources more efficiently and with specific achievable goals
in mind. This is also the main conclusion of the Administration's
recent cybersecurity review.
The focus of today's hearing is not to review what NIST has done,
but to address what should be their focus going forward. Since NIST is
still the only federal agency tasked with protecting non-classified
federal computer systems, the testimony we hear today will have a
vital, long-lasting effect on our country's security.
We have a distinguished panel of witnesses who have a long history
working with NIST and detailed knowledge of NIST's IT activities. I
want to assure them that this subcommittee is prepared to act on their
recommendations.
Mr. Smith. Mr. Chairman, thank you for calling this hearing
today on cybersecurity, the fourth in a series of hearings held
by the Committee this year. Thank you to the witnesses as well.
While our earlier hearings reviewed cybersecurity through a
relatively broad lens, today we are here to examine the
specific role NIST plays or should play in supporting computer
and network security.
Our starting point for this review is the White House's 60-
Day Cyberspace Policy Review which was released in May and
which provided a broad outline of the actions the
Administration intends to emphasize moving forward. A number of
these actionaries appear well-suited to NIST's capabilities and
expertise. With respect to security practices and standards,
NIST is a proven and trusted entity within the Federal
Government, the private sector, and even around the world.
It is also well-known for its excellence in advancing
research and the fundamental science of computer security. For
these reasons, it is important for the Committee to consider
more closely the specific additional or expanded activities
which makes sense for NIST to undertake and what, if any,
associated legislative authority or direction is necessary to
enable this.
In doing so, I think it is also important that we work to
prioritize these activities and identify those which provide
the greatest security returns, recognizing the universe of
computer security activities we would like NIST to do is
significantly larger than any realistic budget expectations.
Additionally, and as I emphasized in our prior hearings, I
think we should also be careful to delineate which activities
NIST shouldn't undertake, particularly with respect to anything
which could take on a regulatory nature, either directly or
indirectly.
I thank the Chairman and the panel today. Thank you for
dedicating your time and donating your time to this productive
discussion. Thank you, Mr. Chairman.
[The prepared statement of Mr. Smith follows:]
Prepared Statement of Representative Adrian Smith
Mr. Chairman, thank you for calling this hearing today on
cybersecurity--the fourth in a series of hearings held by the Committee
this year.
While our earlier hearings reviewed cybersecurity through a
relatively broad lens, today we are here to examine the specific role
NIST plays--or should play--in supporting computer and network
security.
Our starting point for this review is the White House's 60-day
Cyberspace Policy Review which was released in May and which provided a
broad outline of the actions the Administration intends to emphasize
going forward.
A number of these action areas appear well-suited to NIST's
capabilities and expertise. With respect to security practices and
standards, NIST is a proven and trusted entity within the Federal
Government, the private sector, and around the world. It is also well-
known for its excellence in advancing research and the fundamental
science of computer security.
For these reasons, it is appropriate for the Committee to consider
more closely the specific additional or expanded activities which make
sense for NIST to undertake, and what if any associated legislative
authority or direction is necessary to enable this. In doing so, I
think it is also important we work to prioritize these activities and
identify those which provide the greatest security returns, recognizing
the universe of computer security activities we would like NIST to do
is significantly larger than any realistic budget expectations.
Additionally, and as I emphasized in our prior hearings, I think we
should also be careful to delineate what activities NIST shouldn't
undertake--particularly with respect to anything which could take on a
regulatory nature, either directly or indirectly.
I thank the Chairman for assembling an excellent panel today, and I
look forward to a productive discussion.
Chairman Wu. Thank you very much, Mr. Smith, and if there
are any Members who wish to submit their opening statements,
the statements will be added to the record at this point.
[The prepared statement of Mr. Mitchell follows:]
Prepared Statement of Representative Harry E. Mitchell
Thank you, Mr. Chairman.
As the world becomes increasingly connected through the Internet,
it is critical to ensure that we have a secure and reliable cyberspace
policy.
Today we will discuss the findings and recommendations of the Obama
Administration's 60-day Cyberspace Policy Review.
Specifically, we will review that recommendations made in the
Cyperspace Policy Review that may be appropriate for the National
Institute of Standards and Technology (NIST) and the proposed
reorganization of the NIST Information Technology Laboratory.
I look forward to hearing more from our witnesses.
I yield back.
Chairman Wu. And now it is my pleasure to welcome our
witnesses. Ms. Cita Furlani is the Director of the Information
Technology Laboratory (ITL) at the National Institute of
Standards and Technology. Dr. Susan Landau is a Distinguished
Engineer at Sun Microsystems, and a former member of the
Commission on Cybersecurity for the 44th Presidency. I thought
that was a mistake at first, but that is the title of the
group, and the NIST Information, Security and Privacy Advisory
Board. Let us see, we have a different order here. Dr. Phyllis
Schneck is the Vice President of Threat Intelligence at McAfee.
She serves as a commissioner also on the Commission on
Cybersecurity for the 44th Presidency and is on the National
Board of Directors for the FBI's InfraGard. Mr. William Wyatt
Starnes is the Founder and CEO and a great Oregonian, I might
add, of SignaCert. He is formerly a member of the NIST Visiting
Committee on Advanced Technology. Professor Fred Schneider is
the Samuel B. Eckert Professor of Computer Science at Cornell
University and is a current NIST Information Security and
Privacy and Advisory Board Member. And finally, our last
witness is Mark Bohannon who is the General Counsel and Senior
Vice President, Public Policy at Software & Information
Industry Association. Prior to working at SIIA, Mr. Bohannon
was the Chief Counsel of Technology at the U.S. Department of
Commerce where he helped oversee NIST cybersecurity activities.
The witnesses will each have five minutes for your spoken
testimony, and your written testimony will be included in its
entirety in the record for the hearing. When you complete your
testimony, we will begin with questions, and each Member will
have five minutes to ask questions of the panel.
Ms. Furlani, please proceed.
STATEMENT OF MS. CITA M. FURLANI, DIRECTOR, INFORMATION
TECHNOLOGY LABORATORY, NATIONAL INSTITUTE OF STANDARDS AND
TECHNOLOGY
Ms. Furlani. Chairman Wu, Ranking Member Smith, and any
other Members of the Subcommittee. I am Cita Furlani, the
Director of the Information Technology Laboratory at the
Department of Commerce's National Institute of Standards and
Technology. Thank you for the opportunity to appear before you
today.
Cybersecurity is a vital, central mission of our
laboratory. The impacts of NIST's cybersecurity activities
extend beyond providing the means to protect federal IT
systems. They provide the cybersecurity foundations for the
public trust that is essential to our realizing the national
and global productivity and innovation potential of electronic
business and its attendant economic benefits.
Consistent with our mission and the recommendations of the
Administration's Cyberspace Policy Review, NIST is actively
engaged with many others in coordination and prioritization of
cybersecurity research, standards development, standards
conformance demonstration, and cybersecurity education and
outreach activities.
The Review observed that it is our total national
information infrastructure which is under attack. The President
has developed a coordinated national response approach that
places leadership for cybersecurity-related policies amongst
the team within the White House. This team provides an
effective means for coordination and collaboration across the
Federal Government and with the private sector.
The intelligence community, the other elements of the
national security community, and NIST are actively coordinating
their standards and processes for cybersecurity. This effort is
producing a single set of requirements. For the first time,
NIST has included security controls in its catalog for both
national security and non-national security systems. The
updated security control catalog incorporates best practices
and information security from the defense, intelligence and
civil agencies, an historic achievement.
The Review recommended building a cybersecurity-based
identity management, vision and strategy. In response, NIST is
working with OSTP (Office of Science and Technology Policy),
OMB and the NSC (National Security Council) through a new sub-
interagency policy committee focusing on on-line identity
management. Working with OMB and other agencies, NIST is
helping to develop a security and privacy profile that will
provide guidance to enterprise architects on integrating
information security and privacy requirements into the Federal
Enterprise Architecture.
NIST hosts the Information Security Automation Program
which is an effort to enable the automation and standardization
of technical security operations including automated
vulnerability management and policy compliance evaluations. The
NIST National Vulnerability Database is one such tool. It makes
available information on vulnerabilities, impact measurements,
detection techniques and remediation assistance. It provides
reference data that enable the information security automation
program's security automation capabilities. This database also
is key to the payment card industry in their efforts to
mitigate vulnerabilities in credit card systems. The Review
recommended a national public awareness and education campaign
to promote cybersecurity. NIST, working with the SBA (Small
Business Administration) and the FBI, has put an instructional
video on YouTube and published a guide to help small businesses
and organizations. In addition, the Review recommended
strengthening federal leadership and accountability for
cybersecurity. In response, NIST was asked by OMB to contribute
to the Security Metrics Task Force to develop new metrics for
information security performance for federal agencies.
The Review recognizes the role of international standards
in protecting our information infrastructure. We are actively
working with others in fostering international standards and
protocols that are conducive to a free and safe information
processing and interchange environment. NIST also actively
contributes to the NITRD (Networking and Information Technology
Research and Development) program and its five-year strategic
plan.
Consistent with the Review's recommendation, NIST works
with other members of the Cybersecurity and Information
Assurance Interagency Working Group in establishing research
and development priorities to address actions that compromise
or threaten to compromise computer and network-based systems.
NIST has undertaken an internal assessment of its
operational structure and allocation of resources to ensure
that our programs fully reflect the complex interdisciplinary
nature of today's threats. Based on the feedback we continue to
receive, I have decided to put the proposed reorganization of
ITL on hold. We have received expressions of both support and
concern from various stakeholders. We are seriously considering
this input and plan to reevaluate how to ensure that our
structure is as flexible and efficient as possible in meeting
the many challenges and opportunities ahead. Regardless of
whatever recommendations emerge from this internal assessment,
the technical program of work currently performed by the
Computer Security Division (CSD) would not change. ITL welcomes
and appreciates all input and looks forward to continued
conversations on this matter.
Thank you for the opportunity to testify. I would be happy
to answer any questions you may have.
[The prepared statement of Ms. Furlani follows:]
Prepared Statement of Cita M. Furlani
Chairman Wu, Ranking Member Smith, and Members of the Subcommittee,
I am Cita Furlani, the Director of the Information Technology
Laboratory (ITL) at the Department of Commerce's National Institute of
Standards and Technology (NIST). Thank you for the opportunity to
appear before you today to discuss our role in cybersecurity and our
perspective on the Administration's Cyberspace Policy Review
Recommendations.
As one of the major research components within NIST, the
Information Technology Laboratory accelerates the development and
deployment of information and communication systems that are reliable,
usable, inter-operable, and secure; advance measurement science through
innovations in mathematics, statistics, and computer science; and
develop the measurements and standards infrastructure for emerging
information technologies and applications. In addition to research into
cybersecurity technologies, NIST is responsible for development of,
publishing, and providing explanatory support for federal cybersecurity
standards, guidelines, and best practices. Just as the standards
function extends beyond writing federal standards to playing an active
role in the development of national and international consensus
standards, the support function is extended to State and local
governments and private sector elements that voluntarily adopt NIST-
developed cybersecurity standards.
NIST doesn't rely solely on Federal resources and insights. We
employ collaborative partnerships with our customers and stakeholders
in industry, government, academia, and consortia to take advantages of
their technical and operational insights and to leverage the resources
of a global community. We are actively seeking to expand the scope of
these collaborative efforts in general, and of our private sector
collaborations in particular.
The impacts of NIST's cybersecurity activities extend beyond
providing the means to protect federal IT systems. They provide the
cybersecurity foundations for the public trust that is essential to our
realizing the national and global productivity and innovation potential
of electronic business and its attendant economic benefits.
The cybersecurity standards and support capabilities of NIST's
Information Technology Laboratory rest on the foundation of the
laboratory's cybersecurity research and development activities. Based
on input from our customers and stakeholders, we have focused our R&D
agenda on eight broad program areas: complex systems; cyber and network
security; enabling scientific discovery; identity management systems;
information discovery, use and sharing; pervasive information
technologies; trustworthy information systems; and virtual measurement
systems.
Many of our vital programs impact national security in ways that
extend beyond what are generally recognized as the boundaries of
cybersecurity. Examples of these impacts include improving the accuracy
and inter-operability of biometrics recognition systems and
facilitating communications among first responders. The combination of
our mission and legislative mandates such as the Federal Information
Security Management Act (FISMA), the Cyber Security Research and
Development Act, the USA PATRIOT Act, the Enhanced Border Security Act,
and the Help America Vote Act lead to rich programmatic diversity.
Cybersecurity is a vital, central mission of our laboratory. NIST's
mission in cybersecurity is to work with federal agencies, industry,
and academia to research, develop, and deploy information security
standards and technology to protect information systems against threats
to the confidentiality, integrity, and availability of information and
services. Consistent with this mission and with the recommendations of
the Cyberspace Policy Review, NIST is actively engaged with private
industry, academia, non-national security federal departments and
agencies, the intelligence community, and other elements of the law
enforcement and national security communities, in coordination and
prioritization of cybersecurity research, standards development,
standards conformance demonstration, and cybersecurity education and
outreach activities.
The Cyberspace Policy Review observes that it is our total national
information infrastructure, not just the federal information
infrastructure, which is under attack, recognizing a national response
is necessary to prevent catastrophic consequences for society,
including those critical infrastructures which integrate information
systems into their operations. To provide for such a national response,
the President has developed a coordinated approach that places
leadership for cybersecurity-related policies within the White House.
This includes the appointment of a Chief Technology Officer, located in
the Office of Science and Technology Policy, a Chief Information
Officer in the Office of Management and Budget, and the pending
appointment of a Cyber Advisor in the White House. This team provides
an effective means for coordination and collaboration across the
Federal Government and with the private sector. This includes
integrating the responses of national security organizations and those
of federal organizations that do not have a primarily national security
mission. In fact, we observe that the intelligence community, the other
elements of the national security community, and NIST are, in response
to the Federal Information Security Management Act of 2002, actively
coordinating their standards and processes for cybersecurity. This
effort is producing a single set of requirements, rather than the
past's three independent sets of requirements for consumers and
providers of information processing and interchanges resources.
A key output of this initiative to develop a unified information
security framework for the Federal Government and its contractors
occurred on August 1, 2009, when NIST announced the release of Special
Publication 800-53, Revision 3, Recommended Security Controls for
Federal Information Systems and Organizations. NIST Special Publication
800-53, Revision 3, is historic in nature. For the first time, NIST has
included security controls in its catalog for both national security
and nonnational security systems. The updated security control catalog
incorporates best practices in information security from the United
States Department of Defense, Intelligence Community, and civil
agencies, to produce the most broad-based and comprehensive set of
safeguards and countermeasures ever developed for information systems.
This unified framework provides a standardized method for expressing
security at all levels, from operational implementation to compliance
reporting. This allows for an environment of information sharing and
interconnections among these communities and significantly reduces
costs, time, and resources needed for finite sets of systems and
administrators to report on cybersecurity to multiple authorities.
The NIST Identity Management Systems Program is pursuing the
development of critical standards and metrics to support the effective
management of digital identities for large-scale enterprises throughout
their life cycle. These efforts will improve the strength, usability,
and inter-operability of identity management systems; protect users'
personal data; and assure that U.S. interests on this issue are
represented in the international arena. We have been heavily involved
in Federal Government identity management efforts, including developing
the standard for the personal identity verification (PIV) card in
response to HSPD-12 and co-chairing the National Science and Technology
Council (NSTC) Identity Management Task Force.
The Cyberspace Policy Review included in its top ten action items,
``Build a cybersecurity-based identity management vision and strategy
that addresses privacy and civil liberties interests, leveraging
privacy-enhancing technologies for the Nation.'' To this end, NIST is
working with the Office of Science and Technology Policy, the Office of
Management and Budget (OMB), and the National Security Council staff to
determine how to address this action item, through a new Sub-
Interagency Policy Committee which will focus on online identity
management.
NIST is taking other proactive steps to increase the long-term
security of federal information systems. Working with the Office of
Management and Budget and several federal agencies, NIST is helping to
develop a Security and Privacy Profile that will provide guidance to
enterprise architects on integrating information security and privacy
requirements into the Federal Enterprise Architecture. This initiative
will ensure that information security and privacy requirements are
built into federal information systems early in the system development
life cycle rather than attempting to add these requirements after
systems are deployed into operational environments. NIST will also be
working with its partners within the Federal Government to publish
guidance on best practices in systems and security engineering to
address the effective integration of commercial information technology
products into federal information systems. This guidance will build on
the excellent work published by the National Security Agency as part of
the Information Assurance Technical Framework over a decade ago and
make the information widely available to both public and private sector
entities.
NIST hosts the Information Security Automation Program (ISAP),
which formalizes and advances efforts to enable the automation and
standardization of technical security operations, including automated
vulnerability management and policy compliance evaluations.
The NIST National Vulnerability Database (NVD), which is funded by
the National Cybersecurity Division of the Department of Homeland
Security, is the United States Government repository of standards-based
vulnerability management reference data. The NVD makes available
information on vulnerabilities, impact measurements, detection
techniques, and remediation assistance. It provides reference data that
enable the ISAP's security automation capabilities. NIST's security
automation program is based on the NIST Security Checklist program and
the Security Content Automation Protocol (SCAP) activity. The SCAP
Validation Program performs conformance testing to ensure that products
correctly implement SCAP. NVD also plays a pivotal role in the Payment
Card Industry (PCI) in their efforts to mitigate vulnerabilities in
credit card systems. The PCI has mandated that NVD's vulnerability
severity scores be used for measuring the risk to payment card servers
worldwide and for determining which vulnerabilities must be fixed.
In addition to the initiatives described above, NIST has
implemented an aggressive outreach program to work with State, local,
and tribal governments as well as private sector entities to raise the
awareness of government officials and corporate executives with regard
to the ongoing and increasingly sophisticated nature of cyber threats.
The outreach program will help organizations external to the Federal
Government have a better understanding of NIST's suite of security
standards and guidelines and provide an opportunity for voluntary
adoption of the standards and guidelines by those organizations to
facilitate an increased level of information security for the Nation's
critical information infrastructure.
On a broader scale, in response to the Cyberspace Policy Review's
recommendation to initiate a national public awareness and education
campaign to promote cybersecurity and as a contribution to October's
Cyber Security Awareness Month, NIST, working with the Small Business
Administration and the Federal Bureau of Investigation, has published a
guide to help small businesses and organizations understand how to
provide basic security for their information, systems, and networks.
The 20-page guide, Small Business Information Security: The
Fundamentals, uses simple and clear language to walk small business
owners through the important steps necessary to secure their computer
systems and data. The guide provides ten ``absolutely necessary steps''
to secure information, which includes such basics as installing
firewalls, patching operating systems and applications, and backing up
business data, as well as controlling physical access to network
components and training employees in basic security principles. NIST
also created a video that explores the reasons small businesses need to
secure their data.
We are encouraged to observe that the Cyberspace Policy Review
recognizes that cybersecurity strategies and solutions must be
structured in a manner that accommodates commerce, economic growth,
scientific collaboration, and individual liberties. The report reflects
the notion that we are not looking for ``lockdown solutions'' that
achieve security at the expense of essential services or civil
liberties. Recognizing the economic impact of cyberspace, NIST is
working to provide measurement techniques to facilitate offsetting the
cost of both public sector and private sector security solutions by
decreases in losses or cost of insurance or increases in business due
to increases in trust. In order to meet the cyber threat to our total
national infrastructure, we must demonstrate that implementing measures
that increase security is good business sense. We'd note that not all
of these measures need to be technical or regulatory in nature. Some
simple procedural steps can have a materially positive effect on
security. One example is the financial sector's having introduced a
delay into the conversion of electronically transferred funds into
tangible assets, a delay sufficient to permit invocation of fraud
detection processes.
As acknowledged in the Cyberspace Policy Review, measurement of
information security performance can benefit organizations in many
ways, by increasing accountability, improving the effectiveness of
safeguards, demonstrating legislative and policy compliance, and
providing quantifiable inputs for risk-based resource allocation
decisions. The Cyberspace Policy Review recommended strengthening
federal leadership and accountability for cybersecurity, including
identifying cybersecurity as a management priority and assessing the
progress of federal agencies against cybersecurity goals, ultimately
leading to increased accountability, compliance with cybersecurity
policies, and effective implementation of cybersecurity safeguards.
Because of its strengths in measurement science and cybersecurity, NIST
was asked by OMB to contribute to the Security Metrics Taskforce. This
taskforce was established to develop new outcome-focused, rather than
compliance-focused, metrics for information security performance for
federal agencies, resulting in more effective provisioning of security
controls and resources, and improved protection in support of critical
mission and business processes.
We were particularly encouraged by the report's recognition of the
role of international standards in protecting our information
infrastructure. Our infrastructure is inextricably integrated into a
complex of global networks. NIST's role in documentary standards has
long been established in law and executive direction. We are actively
working with our sister agencies, including the Department of State, on
improving our common understanding of how we can collectively
participate, in cooperation with the private sector, in fostering
international standards and protocols that are conducive to a free and
safe information processing and interchange environment.
Recognizing the importance of security-related standards beyond the
Federal Government, NIST leads national and international consensus
standards activities in cryptography, biometrics, electronic
credentialing, secure network protocols, software and systems
reliability, and security conformance testing.
Under the provisions of the National Technology Transfer and
Advancement Act (P.L. 104-113) and OMB Circular A-119, NIST is tasked
with the key role of encouraging and coordinating federal agency use of
voluntary consensus standards and participation in the development of
relevant standards, as well as promoting coordination between the
public and private sectors in the development of standards and in
conformity assessment activities. NIST works with other agencies such
as the State Department to coordinate standards issues and priorities
with the private sector through consensus standards organizations such
as the American National Standards Institute (ANSI), the International
Organization for Standardization (ISO), the Institute of Electrical and
Electronic Engineers (IEEE), the Internet Engineering Task Force
(IETF), and the International Telecommunication Union (ITU).
Key contributions NIST has made include:
Development of the current federal cryptographic and
cybersecurity assurance standards that have been adopted by
many State governments, national governments, and much of
industry;
Development of the identity credentialing and
management standard for federal employees and contractors (also
becoming the de facto national standard);
Development of the standard and conformance test
capability for inter-operable multi-vendor fingerprint minutia
capture and verification;
Development and demonstration of quantum key
distribution;
Establishment of a national cyber vulnerability
database; and
Establishment and oversight of an international
cryptographic algorithm and module validation program. (This
Cryptographic Module Validation Program [CMVP] achieved a
significant milestone on August 15, 2008, by issuing the
program's 1,000th certificate.)
Understanding the value of interagency coordination of research as
well as of standards development, NIST actively contributes to the
Networking and Information Technology Research and Development (NITRD)
program and the development of the NITRD five-year strategic plan.
Within the past year, the NITRD Program has assumed expanded
responsibilities for coordination of federal cyber research and
development, and NIST is well represented in, and leverages, these
activities.
The Cyberspace Policy Review challenged the federal networks and
Information Technology (IT) research community to develop a framework
for research and development strategies that focus on game-changing
technologies. Over the past year, through the National Cyber Leap Year
and a wide range of other activities, the government research
community, including NIST, sought to elicit the best game-changing
ideas from the broader research and technology community.
NIST works with other members of the Cyber Security and Information
Assurance Interagency Working Group in establishing priorities for
research and development to prevent, resist, detect, respond to, and/or
recover from actions that compromise or threaten to compromise the
availability, integrity, or confidentiality of computer- and network-
based systems. These systems provide both the basic infrastructure and
advanced communications in every sector of the economy, including
critical infrastructures such as power grids, emergency communications
systems, financial systems, and air-traffic-control networks. These
systems also support national defense, national and homeland security,
and other vital federal missions, and themselves constitute critical
elements of the IT infrastructure. Broad areas of concern which NIST
research addresses include Internet and network security;
confidentiality, availability, and integrity of information and
computer-based systems; new approaches to achieving hardware and
software security; testing and assessment of computer-based systems
security; and reconstitution and recovery of computer-based systems and
data.
There are others ways in which NIST's expertise can help to drive
improvements in the cybersecurity arena. NIST has integral roles in a
number of Administration initiatives, including Health Information
Technology, Smart Grid, Broadband, and Web 2.0. NIST can continue to
work on more effective metrics (security controls effectiveness
determination), expand education and other outreach, improve product
assurance processes, expand national and international cybersecurity
standards participation, and automate security controls. This is in
addition to our cryptography, technical guidelines, and best practices
work.
To address the interdisciplinary nature of security in cyberspace,
ITL also has programs in the usability of systems such as voting
machines, health information technology and software interfaces;
research in mathematical foundations to determine the security of
information systems; the National Software Reference Library, computer
forensics tool testing, software assurance metrics, tools, and
evaluation; approaches to balancing safety, security, reliability, and
performance in supervisory control and data acquisition and other
industrial control systems used in manufacturing and other critical
infrastructure industries; technologies for detection of anomalous
behavior, quarantines; standards, modeling, and measurement to achieve
end-to-end security over heterogeneous, multi-domain networks; and
biometrics evaluation, usability, and standards (fingerprint, face,
iris, voice/speaker, multi-modal biometrics.) Research activities in
ITL range from innovations in identity management and verification, to
metrics for complex systems, to development of practical and secure
cryptography in a quantum computing environment, to automation of
discovery and maintenance of system security configurations and status,
to techniques for specification and automation of access authorization
in line with many different kinds of access policies.
We, at NIST and the Department of Commerce, recognize that we have
an essential role to play in realizing the vision set forth in the
Cyberspace Policy Review. NIST will continue to conduct the research
necessary to enable and to provide cybersecurity specifications,
standards, assurance processes, training, and technical expertise
needed for securing the U.S. Government and critical infrastructure
information systems to mitigate the growing threat. NIST will continue
to closely coordinate with domestic and international private sector
cybersecurity programs and national security organizations. Finally,
consistent with the NIST Three-Year Planning Report, NIST plans to
broaden its focus on cybersecurity challenges associated with health
IT, the Smart Grid, automation of federal systems security conformance
and status determination, and cybersecurity leap-ahead research.
Cybersecurity is a vital, central mission of our laboratory. Given
the increasing importance and complexity of cybersecurity, NIST has
undertaken an internal assessment of its operational structure and
allocation of resources to ensure that ITL programs fully reflect the
complex interdisciplinary nature of today's threats. For example, NIST
is considering whether it needs to strengthen the authority and purview
of the NIST Chief Cybersecurity Advisor. Regardless of whatever
recommendations emerge from this internal assessment, the technical
program of work currently performed by the Computer Security Division
would not change. NIST welcomes, through our Advisory Committee, key
external stakeholders, and this subcommittee, input on NIST operations
and structure and looks forward to continued conversations on this
matter.
Thank you for the opportunity to testify today on NIST's work in
the cybersecurity arena. I would be happy to answer any questions you
may have.
Biography for Cita M. Furlani
Cita M. Furlani is Director of the Information Technology
Laboratory (ITL). ITL is one of nine research Laboratories within the
National Institute of Standards and Technology (NIST) with an annual
budget of $85 million, 335 employees, and about 150 guest researchers
from industry, universities, and foreign laboratories.
Furlani oversees a research program designed to promote U.S.
innovation and industrial competitiveness by advancing measurement
science, standards, and technology through research and development in
information technology, mathematics, and statistics. Through its
efforts, ITL seeks to enhance productivity and public safety,
facilitate trade, and improve the quality of life.
Furlani has several leadership responsibilities in addition to
those at NIST. Currently, she is Co-Chair of the Interagency Working
Group on Digital Data, Co-Chair of the Subcommittee on QuInformation
Science, and Co-Chair for Strategic Planning for the Subcommittee on
Networking and Information Technology Research and Development, all
under the auspices of the National Science and Technology Council. She
also serves as Co-Chair of the Technology Infrastructure Subcommittee
of the Interagency CIO Council.
Furlani has served as the Chief Information Officer (CIO) for NIST.
As CIO, Furlani was the principal adviser to the NIST Director on the
planning, execution, evaluation, and delivery of information technology
services and support.
Furlani also served as Director of the National Coordination Office
for Networking and Information Technology Research and Development.
This office, reporting to the White House through the Office of Science
and Technology Policy and the National Science and Technology Council,
coordinates the planning, budget, and assessment activities for the 12-
agency Networking and Information Technology R&D Program.
Previously, Furlani was Director of the Information Technology and
Electronics Office within the Advanced Technology Program (ATP) at
NIST. Before joining ATP, Furlani served as Chief of the Office of
Enterprise Integration, ITL, NIST, coordinating Department of Commerce
activities in the area of enterprise integration. Furlani also served
as special assistant to the NIST Director in the Director's role as
Chair of the Committee on Applications and Technology of the
Administration's Information Infrastructure Task Force. Previously,
Furlani was on detail as technical staff to the Director of NIST in the
position of Senior Program Analyst. Prior to August 1992, she managed
research and development programs within the NIST Manufacturing
Engineering Laboratory, applying information technology to
manufacturing since 1981.
She earned a Master of Science degree in electronics and computer
engineering from George Mason University and a Bachelor of Arts degree
in physics and mathematics from Texas Christian University. She was
awarded two Department of Commerce Bronze Medal Awards in 1985 and 1993
and the Department of Commerce Silver Medal Award, in 1995.
Chairman Wu. Thank you very much, Ms. Furlani. Dr. Landau,
please proceed.
STATEMENT OF DR. SUSAN LANDAU, DISTINGUISHED ENGINEER, SUN
MICROSYSTEMS, BURLINGTON, MA
Dr. Landau. Thank you very much, Mr. Chairman, and Members
of the Committee. I am a distinguished engineer at Sun where I
concentrate on security and public policy issues. I have done
this for ten and a half years. I served on ISPAB, Information
Security and Privacy Advisory Board, that advises NIST and got
a chance to see firsthand what a terrific job the people at the
Computer Security Division do, but I had seen that earlier in
my work in cryptography. They have a very difficult job and a
very complicated situation. The designing and security
standards and guidance for federal agencies, those are their
customers, but the work that they do actually gets used by
businesses, private sector as well as being used
internationally. That is when they do things right, and they do
things right most of the time. I am very impressed.
But the reason it is a complicated job is because they--in
order for them to do their work, providing standards for the
Federal Government for federal civilian agencies, they need not
only to do just basic research but mostly applied research and
security guidance, and they are doing that within an agency,
NIST, that focuses on scientific research. So doing the applied
work is often a complicated dance for NIST, for the Computer
Security Division. And I think they do it extremely well. They
do it extremely well because they listen to their customers and
they work well with the industry. They are seen as an honest
broker.
The 60-Day Review was very clear on the need to work
internationally. In order to work internationally, it is
extremely useful to have a scientific agency at your side
providing guidance. We show up in this country with NIST to do
that. Sun was part of a group of industry that had concerns
over the Chinese government trying to impose mandatory security
requirements on 13 different products. We showed up at the
table with NIST, not NSA (National Security Agency), not DHS
(Department of Homeland Security). Having NIST at the table was
extremely important because the Chinese government saw that as
an agency that was not interested in snooping, not interested
in finding out about things from China that it shouldn't, but
as a scientific agency. And it really helped the decisions that
happened, and we as industry are quite happy with the results,
and we really relied upon NIST to do that.
That was part of what the 60-Day Review said, the
importance of international agreements, and that calls for an
elevated role for the Computer Security Division. There are
other things that the Computer Security Division should be
doing, and I am delighted to hear, by the way, that the
reorganization is off the table because I thought that that was
problematic. But there are other things that the Computer
Security Division should be doing.
We need to address privacy standards. In recent months,
there had been technical work that describes how easy it is to
take data that looks as if it is anonymized and re-identify it
with other data outside that particular data set, taking
information from Netflix that has been anonymized and comparing
it with data outside the Netflix database and being able to
figure out who the people are. We need scientific standards,
technical standards, to talk about how data should be handled
to protect privacy. In the past, NIST has worked almost
entirely on computer security standards and not on privacy
standards, and I think that this role is very important,
especially as we move forward with health care. We need NIST,
we need the Computer Security Division to be active in the
international arena, we also need greater independence for the
Computer Security Division. It is impossible to separate policy
from security. I am not asking here for NIST to be setting
government policy on security. What I am asking is for NIST to
be providing advice when a computer security issue comes very
close to a policy issue, whether it is about identity
verification, identity management, or any one of another
technical issues. NIST has that expertise and should be using
it more in government.
It is also important to keep the branding of the Computer
Security Division which is well-known both within the
government now as a result of FISMA (Federal Information
Security Management Act) and outside the government because of
all the excellent work that CSD does.
For all these reasons, I think it is time to elevate the
Computer Security Division to the level of a laboratory. I
think that that would help a great deal in international work,
I think it would be appropriate in terms of the policy effort
that I think a computer security group should be doing, I think
it is important for privacy standards.
Thank you very much, and I would be happy to answer
questions.
[The prepared statement of Dr. Landau follows:]
Prepared Statement of Susan Landau
Mr. Chairman and Members of the Committee:
Thank you for the opportunity to testify today on the Computer
Security Division and its role in developing computer security
standards and guidance for the Federal Government and the wider
community. I am a distinguished engineer at Sun Microsystems, where I
concentrate on security, cryptography, and public policy. I have been
involved in Sun efforts on cryptography and export control, security
and privacy of federated identity management systems, developing our
policy stance in digital rights management, and in analyzing security
risks of surveillance in communications infrastructures. I am a member
of the Commission on Cyber Security for the 44th Presidency,
established by the Center for Strategic and International Studies, and
I serve on the advisory committee for the National Science Foundation's
Directorate for Computer and Information Science and Engineering. I am
also a former member of NIST's Information Security and Privacy
Advisory Board, where I served six years. I have been a strong
supporter of the Computer Security Division for many years.
Fulfilling the Cyberspace Policy Review Recommendations
Over the last decade there have been many discussions and reports
regarding the ways and means to achieve cybersecurity. The problem is
partially technical and a great deal policy. The most recent Cyberspace
Policy Review\1\ raises several new points.
---------------------------------------------------------------------------
\1\ Cyberspace Policy Review: Assuring a Trusted and Resilient
Information and Communications Infrastructure, 2009.
---------------------------------------------------------------------------
One of these is the need to work internationally in order to
achieve security in cyberspace. With the somewhat boundaryless nature
of the Internet, this point is abundantly clear, but this direction has
not been a focus of recent U.S. policy. It should be.
Working with other nations on securing cyberspace requires policy
efforts--treaties and international agreements of various sorts--but it
also requires technical work--standards, for example. NIST is the
appropriate agency for the latter. I would expect the Computer Security
Division (CSD) at NIST to work hand-in-hand with the Department of
State in forging international agreements to secure cyberspace. CSD has
a proven history of working well with multiple partners inside and
outside the Federal Government. It has played an excellent role in
developing standards accepted by the international community. This
combination of collaboration and insistence on technical and scientific
integrity means that CSD will be a respected partner in discussions
with other nations and scientific societies. It is the only U.S.
Government agency able to play this role on the civilian side. In fact,
it has already been doing so.
Two years ago, for example, the Chinese government notified the
World Trade Organization that it planned to impose new mandatory
information security certification rules for thirteen product areas.
The proposed rules might have barred several types of U.S. products
from China's marketplace. Industry, working with the Department of
State, the U.S. Trade Representative, and NIST held a series of policy-
level and technical level discussions with the Chinese government and
impacted the rules finally promulgated this year. CSD's help in this
was invaluable.
The Cyberspace Policy Review points out the need for defined
performance and security objectives. The organization with experience
to develop these is CSD.
Indeed, while this was undoubtedly not the intent of the review,
the document is a ringing call for the skills, activities, and
interventions of CSD. The report certainly makes the case for an
expanded role for the division. The review underscores the fact that
cybersecurity is a problem that will need international cooperation,
emphasizes the importance of working with private industry, and
stresses the need for protecting privacy and civil liberties rights
while securing cyberspace. The U.S. Government agency with a history
and a reputation for scientific integrity and with an ability to work
well with civilian groups outside the Federal Government is NIST's
Computer Security Division.
In light of such additional responsibilities, it is appropriate to
ask how should the CSD be structured to achieve these goals. In one
sense, no change is needed: the organization works. In another, some
change will be needed because of the additional responsibilities.
NIST's Information Technology Laboratory is proposing a restructuring
of the division within ITL. I believe such a change is a mistake and
will actually hinder CSD's new roles rather than enhance them. I
believe that instead that the Computer Security Division should become
its own laboratory, the Computer Security Laboratory. CSL more properly
suits the U.S.'s cybersecurity needs for the twenty-first century.
What the Computer Security Division Contributes
I look at the proposal to reorganize the Computer Security Division
from the perspective of the cryptographic standards DES and AES, and
the superb job that CSD did in organizing the competition for the
Advanced Encryption Standard. Not only did the division run the
competition in an open way that encouraged submissions from around the
world, the division even asked for comments on the proposed
requirements and changed those requirements in order to fit public
needs. This openness resulted in a standard that was accepted
immediately almost everywhere. This acceptance of AES is a tremendous
win for security. I note that the situation is in sharp contrast to
that for 1970's algorithm, DES, about which doubts about secret back
doors and weak keys persisted for many years; these impeded the
algorithm's acceptance.
The fact is that CSD knows how to work with industry and in a
public environment. That means better security not just for the
civilian Federal Government, whose computer security standards and
guidance the division develops, but also for the U.S. private sector
and the world.
What Needs to be Sustained and What Needs to be Changed
Developing security standards for federal civilian agencies has
various components. In addition to basic research, it requires applied
work and guidance documents. Successful security means knowing what
customers--in CSD's case, that is the federal civilian agencies--need.
It also means knowing how to work with industry to develop the
standards and guidance documents that enable computer security to be
implemented. This means computer security not just for federal
agencies, but for much broader constituencies.
Having CSD within NIST is complicated, because CSD's efforts,
including the guidance documents, are out of synch with NIST's research
mission. But nonetheless it is NIST, and not DHS or NSA, that is the
right home for CSD. In order to be effective CSD must work with
industry, developing standards that function at both a technical level
and a policy one. A standard that is too complex to implement, or that
contradicts customer needs, is a standard that will not be widely
deployed. For this reason, the correct home for CSD is the Department
of Commerce, the U.S. department that works with industry and that has
responsibility for U.S. competitiveness and e-commerce.
CSD is viewed as vendor neutral and an honest broker. The honesty
with which CSD does its work and the openness in which it develops its
standards and guidance, contribute to the work's broad acceptance and
usage. Over the last dozen years, CSD has done a superb job in
developing standards and guidance that works, from AES, to SCAP, to the
new work on hash standards (Because SHA-1 is increasingly vulnerable to
attack, NIST's decision to pursue a SHA-3 algorithm seems to have been
prescient). NIST's work on cloud computing has provided reference
definitions upon which the Cloud Security Alliance relies; NIST has
definitely provided thought leadership in this important and emerging
area.
CSD guidance and standards are ones that make sense in a civilian
context. The health care industry, for example, which keeps 95 percent
of U.S. health care records does not want to adopt computer security
standards developed by the military; it wants standards developed for a
civilian context. Many CSD standards are used by private industry and
in countries around the world. Both U.S. industry and computer security
benefit from this.
At the same time, there are things that are missing within CSD.
Although the division is not a policy setting organization, CSD needs
to be more willing to be involved in policy decisions that verge on
technical ones. This includes the Personal Identity Verification (PIV)
standards, where CSD should have pushed back on OMB, and said that
these standards cannot be implemented effectively within the time
frame; there will be security costs, there will be privacy costs that a
slower timetable would alleviate. Other discussions in which CSD should
be involved on the policy level includes the current Identity,
Credential, and Access Management (ICAM) effort on identifiers for
Level of Assurance 1.
CSD also needs to work more on usability and security, and on
usability and privacy. Security controls that are too complex to use
and privacy standards that are unclear help neither security or
privacy. I understand that CSD has begun active work in this direction.
Finally--and this is a long-term challenge--CSD could do a better
job of making its work public. From the state of its web page, in which
it is challenging to find information (this is a subject about which
the Information Security and Privacy Advisory Board, and probably
others, have raised concerns), to its lack of sufficient workshops on
implementing its standards, CSD does not do sufficient outreach. It is,
for example, CSD which should be running workshops for small businesses
on security (and not the FBI). CSD produces high quality, vendor-
neutral security guidance, and this high quality information should be
much more broadly publicized--and therefore used--than it is.
If CSD is to develop privacy standards and to do effective
outreach, CSD will need an increased budget. These are new
responsibilities and CSD's people are already stretched thin. These are
difficult budget times and funding is tight, but given the criticality
of our nation's cybersecurity needs, such increased appropriations are
both appropriate and necessary. The money spent now will prevent higher
costs to society as a result of weak cyber protections; it would be
money well spent.
The Proposed Reorganization
For reasons that are not entirely clear, the Information Technology
Laboratory is attempting a reorganization. Some aspects of this seem
excellent--moving the head of CSD to the secretary's office to work on
policy-related aspects of computer security is a smart plan--but others
raise great concern. The argument is being made that there would be
increased synergy by moving aspects of security, such as identity
management, into other parts of the organization. I disagree.
Synergy is best achieved by keeping members of the Computer
Security Division together. Researchers find commonalities in security
issues, whether it is protecting VoIP or securing virtual worlds, when
they work closely together. While spreading security across an IT
support organization might be useful, the same is not true for an
organization doing research. The rationale for one split, moving
identity management to the testing division and separating that group
from most of computer security, is that identity management is
intimately tied up with testing. This is correct, but in fact identity
management is also intimately tied to computer security, and separating
the two areas weakens the whole. Dividing different groups supporting
CSD's mission will be detrimental to the work CSD does. Ultimately the
effect will be to weaken CSD's impact on federal civilian security.
In addition, having multiple sources for federal civilian computer
security standards and guidance will cause CSD to lose its identity as
the ``go-to'' organization for federal civilian security, and the
division will lose the branding recognition that has already occurred.
The proposed reorganization, if it should happen, will make it more
difficult for people to locate the NIST computer security information
they need (a problem that is already too difficult). This is the wrong
step at the wrong time.
I believe that instead we should be looking to create a separate
Computer Security Laboratory within NIST. There are many arguments for
such a change.
The first is that there are new responsibilities the division
should take on. In the world of massive databases and such privacy-
threatening technologies as social networks, the CSD mission should
create privacy standards. This includes, for example, how to handle
data to prevent loss of privacy due to data aggregation, what suitable
anonymization techniques are, etc. This is a new and important job for
CSD.
A second issue is that increasingly we will need to bring to the
bilateral and multilateral bargaining table a government partner on
technical cybersecurity issues. This partner must be one that is
trusted by all sides and this means the division will be part of a U.S.
team negotiating internationally on issues of cybersecurity. In such
negotiations, NIST's technical people must be perceived as having the
right stature. The elevation of the division to a laboratory would be
very useful to U.S. interests and fits in with the actions proposed by
the Cyberspace Policy Review.
A third important reason is that a NIST laboratory-level computer
security organization would provide the correct level of independence
for such an organization. The director would be in a better position to
provide the policy guidance needed in discussions related to computer
security and privacy. Note that I am not talking about setting
government policy, but advising on the policy implications of what
appear to be purely technical decisions, whether in the adoption of a
PIV card that allows the biometric authenticator to be read without a
guard present, or in the use of OpenID as a Level of Assurance 1
identifier.
In elevating CSD to a laboratory within NIST, CSD's branding is
retained. This is important to the effective filling of the CSD
mission.
As we all know, cybersecurity will only increase in importance with
time. A separate Computer Security Laboratory will enhance CSD's
visibility, and ensure that CSD's work is not diluted by other,
excellent work in ITL (but work that is unrelated to the computer
security effort). In order to function effectively, CSD needs to be a
single unit, but with more independence, with strong support from its
parent agency of NIST, and with the ability to speak with an honest,
scientific voice. A separate laboratory within NIST is the right way
for CSD to go at this time.
Thank you very much for the opportunity to address the Committee. I
eagerly await any questions you might have.
Biography for Susan Landau
Susan Landau is a Distinguished Engineer at Sun Microsystems
Laboratories, where she works on security, cryptography, and policy,
including surveillance and digital-rights management issues. Landau had
previously been a faculty member at the University of Massachusetts and
Wesleyan University, where she worked in algebraic algorithms, and she
held visiting positions at Yale, Cornell, and the Mathematical Sciences
Research Institute at Berkeley.
Landau is co-author, with Whitfield Diffie, of ``Privacy on the
Line: the Politics of Wiretapping and Encryption'' (MIT Press, original
edition: 1998; updated and expanded edition: 2007), which won 1998
Donald McGannon Communication Policy Research Award, and the 1999 IEEE-
USA Award for Distinguished Literary Contributions Furthering Public
Understanding of the Profession.
Landau participated in the 2006 ITAA study on the security risks of
applying the Communications Assistance for Law Enforcement Act to Voice
over IP, and is also primary author of the 1994 Association for
Computing Machinery report ``Codes, Keys, and Conflicts: Issues in US
Crypto Policy.'' Prior to her work in policy, Landau did research in
symbolic computation and algebraic algorithms, discovering several
polynomial-time algorithms for problems that previously only had
exponential-time solutions.
Landau is a member of the Commission on Cyber Security for the 44th
Presidency, established by the Center for Strategic and International
Studies, and serves on the advisory committee for the National Science
Foundation's Directorate for Computer and Information Science and
Engineering. She is also an Associate Editor for IEEE Security and
Privacy and a section board member of Communications of the ACM. Landau
serves on the Executive Council for Association for Computing Machinery
Committee on Women in Computing, as well as on the Computing Research
Association Committee on the Status of Women in Computing Research.
Landau served for six years on the National Institute of Standards and
Technology's Information Security and Privacy Advisory Board. She has
been a member of ACM's Advisory Committee on Privacy and Security and
ACM's Committee on Law and Computing Technology as well as an Associate
Editor of the Notices of American Mathematical Society.
Landau is the recipient of the 2008 Women of Vision Social Impact
Award, a Fellow of the American Association for the Advancement of
Science, and a Distinguished Engineer of the Association for Computing
Machinery. More information on her publications and awards can be found
at http://research.sun.com/people/slandau
Landau received her Ph.D. from MIT (1983), her MS from Cornell
(1979), and her BA from Princeton (1976).
Chairman Wu. Thank you very much, Dr. Landau. Dr. Schneck,
please proceed.
STATEMENT OF DR. PHYLLIS SCHNECK, VICE PRESIDENT, THREAT
INTELLIGENCE, MCAFEE CORPORATION
Dr. Schneck. Good afternoon, Chairman Wu, Ranking Member
Smith, Members of the Subcommittee. My name is Phyllis Schneck.
I am the Vice President of Threat Intelligence at McAfee. We
are headquartered in Santa Clara, California. A core of our
cyberlabs and our cyber research is in Beaverton, Oregon.
I testify today on behalf of the BSA, the Business Software
Alliance. Thank you for the opportunity to testify on
cybersecurity and the role of the ITL. I commend the
Subcommittee for focusing on these important issues.
McAfee and BSA believe that innovation and standards are
among the most important tools we have to improve our
cybersecurity. Therefore, our primary recommendation regarding
the role of the ITL in implementing the recommendations of the
60-Day Review is to contribute to an integrated, U.S.
Government strategy to influence the development of
international standards on cybersecurity.
Please allow me to explain the important links between
innovation, cybersecurity and international standards. First,
we believe innovation is key to cybersecurity. Those persons
intent on doing harm, whether cybercriminals, spies, hostile
nations, even terrorist groups, find new ways to attack. They
adopt those new technologies all the time, and we must stay
ahead of them, and to do that innovation is key.
Second, we believe that global industry-led voluntary
standards are critical to innovation. This is because first,
they facilitate interoperability between systems built by
different vendors. Second, they facilitate competition between
those vendors, leading to greater choice, lower cost. Finally,
they spur the development and the use of innovative and secure
technologies because they are regularly updated.
Cybersecurity depends on innovation which in turn depends
on global industry-led standards. This is why we urge the
United States to support and uphold these standards by
developing a comprehensive, international cybersecurity
standards strategy.
Currently the U.S. Government's involvement in standards
development is ad hoc, incomplete and uncoordinated. The 60-Day
Review recognized this lack of coordination and called for a
comprehensive strategy that defines what cybersecurity
standards we need, where they are being developed and what
agencies will represent the United States for each.
NIST has expertise in standards and in cybersecurity and is
internationally respected, so it should play an important role
in the creation and implementation of such a strategy.
Conversely there are missteps the government should avoid.
Most importantly, we should not impose country-specific,
government-created technology standards for cybersecurity. This
would set a dangerous precedent that other nations would follow
to create their own divergent standards. This would be at odds
with the global nature of the Internet, it would Balkanize the
global marketplace, and it would inhibit inter-operability. We
believe our position is fully consistent with President Obama's
statement when he released the Cyberspace Policy Review on May
29. President Obama said, ``My Administration will not dictate
security standards for private companies. On the contrary, we
will collaborate with industry to find technology solutions
that ensure our security and promote prosperity.''
I will now address the proposed reorganization of the ITL
and CSD. We believe the success of CSD depends first on budget
and manpower. CSD is already under-resourced and understaffed.
As we give them new missions in the context of tighter federal
budgets, they will need sufficient resources. We will also need
to ensure that NIST funds intended for Congress for
cybersecurity are not spent on other projects.
Second, CSD works with a wide range of industry and
academic partners. The process under way needs to be open and
transparent so that it can be informed by the views of the
stakeholders.
And third, whatever we do, we should avoid diminishing the
visibility, priority and resources accorded to cybersecurity
within NIST.
Finally, I would like to close my testimony with a few
other recommendations about further activities of CSD. As
Congress considers how to elevate cybersecurity as a government
priority, including how to reform FISMA, the Federal
Information Security Management Act, CSD should produce the
following. First, government-wide standards and guidelines for
real-time monitoring, audit and analysis of data about the
security of federal networks. And second, government-wide
standards and guidelines developed jointly with industry for
sharing threat and vulnerability information among federal
agencies and with the private sector.
NIST must also continue to invest in cybersecurity research
and development. BSA has called for the creation of a national
cybersecurity R&D plan, and we believe that NIST would play an
important role under such a plan, given its own R&D work and
its private-sector relationships.
Thank you, and I look forward to answering any questions.
[The prepared statement of Dr. Schneck follows:]
Prepared Statement of Phyllis Schneck
Chairman Wu, Ranking Member Smith, Members of the Committee, thank
you for the opportunity to testify today on the important issue of
cybersecurity, and the role of the National Institute of Standards and
Technology (NIST)'s Information Technology Laboratory (ITL).
My name is Phyllis Schneck, and I am the Vice President of Threat
Intelligence at McAfee. McAfee is the world's largest dedicated
security technology company. McAfee is committed to relentlessly
tackling the world's toughest security challenges. The company delivers
proactive and proven solutions, services and global threat intelligence
that help secure systems and networks around the world, allowing users
to safely connect to the Internet, browse and shop the web more
securely.
As Vice President of Threat Intelligence, I am responsible for the
design and application of McAfee's Internet reputation intelligence,
strategic thought leadership around technology and policy in
cybersecurity, and leading McAfee initiatives in critical
infrastructure protection and cross-sector cybersecurity.
I testify today on behalf of the Business Software Alliance (BSA),
of which McAfee is a member. BSA is the foremost organization dedicated
to promoting a safe and legal digital world. BSA is the voice of the
world's commercial software industry and its hardware partners before
governments and in the international marketplace.\1\
---------------------------------------------------------------------------
\1\ BSA members include Adobe, Apple, Autodesk, Bentley Systems,
CA, Cadence Design Systems, Cisco Systems, Corel, CyberLink, Dassault
Systemes SolidWorks Corporation, Dell, Embarcadero, HP, IBM, Intel,
Intuit, McAfee, Microsoft, Minitab, Quark, Quest Software, Rosetta
stone, SAP, Siemens, Sybase, Symantec, Synopsys, and The MathWorks.
---------------------------------------------------------------------------
My testimony will address three questions:
1. What could NIST do to address some of the recommendations
of the Cyberspace Policy Review?
2. What is our assessment of the proposed reorganization of
NIST's ITL, and how will it improve the outcomes of ITL
activities?
3. Given the current emphasis on information assurance and
cybersecurity, what recommendations do we have on how ITL might
improve its effectiveness or expand the scope of its activities
and their impact?
1. What could NIST do to address some of the recommendations of the
Cyberspace Policy Review?
McAfee and BSA welcomed the 60-day review ordered by the President.
We believe that cybersecurity needs to be elevated as a priority of
this country. We also welcomed the openness of the review process,
which allowed a wide range of stakeholders, and in particular owners
and operators of critical cyber infrastructure, to provide their views
and recommendations. In the end, while the final report contains many
recommendations and so will require that industry remain engaged
throughout their implementation, McAfee and BSA were broadly supportive
of the Cyberspace Policy Review's conclusions.
I would like to touch on a few of the recommendations of the
Cyberspace Policy Review that we believe are of particular importance
and relevance to NIST.
Firstly, we strongly support the Cyberspace Policy Review's call for an
integrated U.S. Government strategy to influence the development of
international standards on cybersecurity.
Such a strategy would recognize the important links between
innovation, cybersecurity and international standards.
We believe innovation is key to greater cybersecurity. Those
persons intent on doing harm, whether profit-motivated cyber criminals,
cyber spies, hostile nations or terrorist groups, find new ways to
attack and adopt new technologies all the time. We must stay a step
ahead of them. To do this, innovation is key.
A necessary element of ensuring continued innovation is sound
standards policy. Global, industry-led, voluntary standards and best
practices create the environment where multiple innovative solutions
can flourish by:
Facilitating inter-operability between systems built
by different vendors.
Facilitating competition between vendors, leading to
greater choice and lower cost.
Spurring the development and use of innovative and
secure technologies, because industry-led standards are
regularly updated.
This is why we urge the U.S. Government to support and uphold
global, industry-led standards and best practices on cybersecurity, by
doing the following:
First, the U.S. Government needs to develop a
comprehensive international cybersecurity standards strategy.
What we have currently is a collection of ad hoc, incomplete
and uncoordinated efforts. The White House Cyberspace Policy
Review recognized this lack of coordination. NIST should play
an important role in the creation and implementation of such a
strategy. The strategy needs to answer the following questions:
1. What cybersecurity standard development efforts is
the U.S. currently involved in?
2. What cybersecurity standards do we need?
3. Where are they being developed?
4. What agencies will represent the U.S. for each of
them?
Second, the government should identify the relevant
international industry-led cybersecurity best practices, and
recognize and promote their use in federal systems. Government,
industry and academia should collaborate to identify
international industry-led best practices, and McAfee and BSA
would eagerly contribute to such a process.
But there are also missteps the government should avoid. Most
importantly, the government should not impose country-specific
technology standards for cybersecurity, in particular standards
developed by government agencies, except in narrowly tailored national
security situations. This would set a precedent that other nations
would follow to create their own, divergent standards. The end result
would be at odds with the global nature of the Internet, would
contribute to breaking up the global marketplace into national markets,
and would inhibit rather than promote inter-operability.
Finally, I would add that if NIST were tasked with creating and
mandating such domestic standards, it would lessen the high regard it
enjoys not just in the United States, but also internationally, as an
arbiter of a process grounded in science.
Therefore, cybersecurity policy-makers should support the global
nature of the IT marketplace, rather than contribute to breaking it up
into national markets.
We believe our position is fully consistent with President Obama's
statement, when he released the Cyberspace Policy Review on May 29:
``My administration will not dictate security standards for private
companies. On the contrary, we will collaborate with industry to find
technology solutions that ensure our security and promote prosperity.''
Secondly, I would like to say a few words about the Cyberspace Policy
Review's recommendation to launch a public education and awareness
campaign.
Educating the public about threats and about common sense measures
it can adopt to protect itself, is important. That is why the CEOs of
BSA raised this issue when they met with Secretary of Homeland security
Napolitano this year. Many BSA members, including McAfee, have made
important investments in educating the public about cybersecurity, for
example by actively supporting and sponsoring the National Cyber
Security Alliance (NCSA), the preeminent public-private partnership
between industry, the U.S. Department of Homeland Security (DHS) and
non-profit institutions, to promote cybersecurity awareness for home
users, small and medium size businesses, and primary and secondary
education.
McAfee and BSA believe a major education and awareness campaign on
the scale envisaged by the Cyberspace Policy Review should build upon
the foundation of the NCSA. If NIST were to take a role in education
and awareness, we recommend that it do so through the national campaign
that NCSA should coordinate. NCSA should be the focal point, using and
expanding the relationships and brand it has already built with a
multitude of local stakeholders--schools and universities, community-
based organizations, local governments, local chambers of commerce,
home-owners associations, etc.
Thirdly, NIST has a valuable role to play in carrying out the
Cyberspace Policy Review's call for building a cybersecurity-based
identity management vision and strategy.
Identity and authentication are foundational building blocks of a
modern and fundamentally secure cyberspace. The Administration is
already working to implement this recommendation of the Cyberspace
Policy Review, and we expect them to issue a draft document in the
coming months to the public for comment.
NIST should play a critical role in crafting and implementing this
government strategy, on the basis of the important contributions it has
made to previous federal identity and authentication initiatives, such
as the implementation of Homeland Security Presidential Directive 12
(HSPD-12). As identity and authentication can apply not only for
individuals, but also for devices, NIST's ability to advise and
influence this strategy will be critical to ensuring its technical
feasibility and operational success.
As the Cyberspace Policy Review notes, it is important that the
government not mandate the use of specific identity management systems,
but rather ensure that they are available as opt-ins. We also agree
with the Review that a variety of inter-operable systems should be
offered, rather than the government picking a single provider or
technology, which would stifle innovation.
2. What is our assessment of the proposed reorganization of NIST's
ITL, and how will it improve the outcomes of ITL activities?
BSA has not had the opportunity to reach a common position among
its members on the reorganization of the ITL. However, I would like to
make the following comments about what is at stake.
First, we believe two important factors in the future success of
the Computer Security Division (CSD) of the ITL are budget and
manpower. CSD is already under-resourced and under-staffed. As we give
them new missions in a context of tighter federal budgets, sufficiency
of resources will be a key concern. We will also need to ensure that
NIST funds intended by Congress for cybersecurity are not spent on
other projects, and this can be achieved by requiring that ITL
regularly report to this committee on how it spends funds designated
for cybersecurity.
Second, the process that will determine the future course of the
ITL needs to be open, transparent and based on the input of the wide
range of stakeholders, in particular from the IT industry and academia,
who work with CSD.
And third, the guiding principle should be to avoid diminishing the
visibility, priority, and resources accorded to cybersecurity within
NIST.
3. Given the current emphasis on information assurance and
cybersecurity, what recommendations do you have on how ITL might
improve its effectiveness or expand the scope of its activities and
their impact?
First, McAfee and BSA want to restate their deep appreciation for
the outstanding work done by the ITL and CSD over the years.
I would like to highlight two reasons in particular that have
contributed to establishing ITL as a widely-respected leader:
1. ITL works collaboratively with stakeholders. Its work
products are well regarded because they draw upon the best
contributions of leading experts in their fields, from industry
but also from academia. One of the most salient examples is the
AES encryption standard, whose underlying cryptographic
algorithm had been developed by Belgian academics and selected
through a rigorous competition. The openness of the selection
process has greatly contributed to inspiring confidence in AES
and thus in its wide adoption outside the Federal Government.
2. For the security of federal systems, and with very few
exceptions, ITL does not in fact enact mandatory technology
standards. Rather, it offers guidance--through its Special
Publications 800 (SP 800) series--that are flexible enough to
allow each agency to adopt the security posture most
appropriate to its risk profile. We need to ensure that federal
agencies more consistently implement this guidance.
As Congress considers how to reform FISMA to place greater emphasis
on actual security of federal networks and systems, federal agencies
will need in particular that CSD expand its scope of activities,
building on its legacy of public-private collaboration and non-
mandatory guidance, to produce the following:
Government-wide standards and guidelines for real-
time monitoring, auditing and analysis of data about the
security, performance and health of federal networks and
systems across the entire Federal Government. This would
contribute to providing holistic, end-to-end security of
federal networks, rather than focusing on the security of
single points of failure.
Government-wide standards and guidelines for sharing
threat and vulnerability information among federal agencies and
with the private sector. While we think, as I said before, that
NIST should always work collaboratively with stakeholders,
given the private sector impact of information sharing, any
NIST effort in this area should be undertaken jointly with the
private sector, in coordination with DHS.
Global, industry-led standards must continue to underpin the global
IT ecosystem. Therefore, these two categories of NIST standards and
guidelines should draw from global, industry-led standards to the
greatest extent possible.
Importantly, in producing such standards and guidelines, NIST
should spur innovation by always striving to, per the terms of the
National Institute of Standards and Technology Act, ``ensure that such
standards and guidelines do not require specific technological
solutions or products, including any specific hardware or software
security solutions; ensure that such standards and guidelines provide
for sufficient flexibility to permit alternative solutions to provide
equivalent levels of protection for identified information security
risks; and use flexible, performance-based standards and guidelines
that, to the greatest extent possible, permit the use of off-the-shelf
commercially developed information security products.'' \2\
---------------------------------------------------------------------------
\2\ Section 20 of the National Institute of Standards and
Technology Act (15 U.S.C. 278g-3), subsection (c)(5-7).
---------------------------------------------------------------------------
Finally, NIST must continue to push at the edges of cybersecurity
research and development. BSA has expressed in the past to this
committee the importance that we attach to research and development
(R&D) to improve our nation's cybersecurity, and we have called for a
national cybersecurity R&D plan. We believe that NIST would play an
important role under such a plan, given its own R&D work and its
ability to reach out to the R&D arms of many companies.
In conclusion, I want to reiterate the importance that we attach
to:
Innovation as a major tool to improve our
cybersecurity;
The role that R&D and international, industry-led
standards play in spurring innovation and in improving
cybersecurity; and
The development by the U.S. Government of an
international cybersecurity standards strategy.
Biography for Phyllis Schneck
For more than a decade, Dr. Phyllis Schneck has held a
distinguished presence in the security and infrastructure protection
community. Currently serving as Vice President of Threat Intelligence
at McAfee, she is responsible for the design and application of
McAfee's Internet reputation intelligence, strategic thought leadership
around technology and policy in cybersecurity, and leading McAfee
initiatives in critical infrastructure protection and cross-sector
cybersecurity.
Schneck recently served as a commissioner and a working group co-
chair on the public-private partnership for the CSIS Commission to
Advise the 44th President on Cyber Security. Schneck also served for
eight years as Chairman of the National Board of Directors of the FBI's
InfraGard program and as Founding President of InfraGard Atlanta,
growing the InfraGard program from 2,000 to over 26,000 members
nationwide. Named one of Information Security Magazine's Top 25 Women
Leaders in Information Security, Schneck holds three patents in high-
performance and adaptive information security, and has six research
publications in the areas of information security, real-time systems,
telecom and software engineering.
Before joining McAfee, she served as Vice President of Research
Integration at Secure Computing. Schneck holds a Ph.D. in Computer
Science from Georgia Tech where she pioneered the field of information
security and security-based high-performance computing.
Chairman Wu. Thank you very much, Dr. Schneck. Mr. Starnes,
please proceed.
STATEMENT OF MR. WILLIAM WYATT STARNES, FOUNDER, CEO, AND
PRESIDENT, SIGNACERT, INC.; FOUNDER, TRIPWIRE, INC.
Mr. Starnes. Good afternoon, Mr. Chairman, and respected
Members of the Committee. I appreciate the opportunity to
present today before the Committee.
As you know, my name is Wyatt Starnes. I am the founder of
a company called Tripwire, Incorporated. Tripwire has been
heavily used in both government and commercial security
practice, and I currently serve as the CEO and President of
SignaCert, also involved in information assurance issues.
We have been working with both companies very closely with
commercial and government sectors in the areas of information
assurance and cybersecurity for better than a decade.
For purposes of my testimony and for reasons better
described in my written testimony, we prefer the term ``cyber
assurance,'' and the reason we tend to think this way is we
deal both with non-malicious and malicious activity and have
found empirically that non-malicious activity, unauthorized
changes and uncontrolled changes can cause up to 90 percent of
the failures in complex information technology systems. We
really believe that that view needs to be broader than just
cybersecurity.
Relative to NIST and the 60-Day Review, my personal
experience tells me that NIST is already ahead of the curve in
most of the key areas discussed in the report. What I would
observe in general about the report is it lacks substantive,
out-of-the-box thinking. There are bigger and more important
things we can be doing than pure black list-based
cybersecurity, which is the goal of keeping the bad guys out of
the systems. We must more broadly assure that the systems are
intact as designed.
But NIST's contributions relative to all of these issues,
cyberassurance and cybersecurity, have in fact been formidable.
So I am going to talk about three of those.
One and perhaps most importantly is the 800-series body of
work which is literally volumes of work, and this work has
contributed significantly to the state-of-the-art for both
federal and commercial IT software and systems management.
Secondly, I would like to focus on some extension of that
work on a practical sense, and that is a multilateral and both
private and public partnership and teaming that has been in
place to effect the security, content, automation protocol, or
SCAP methodology. Ms. Furlani referred to that in generalized
security cataloging. We as an industry participant see this as
an extremely important method and protocol, leveraging heavily
the work of NIST with the 800-series documents as well as
bringing in the best of some of the intelligence community and
DOD (Department of Defense) work.
In my opinion, the SCAP method and the increased emphasis
on continuous monitoring as opposed to pure accreditation and
auditing methods represent far and away the most important
advance federal IT systems management that I have seen.
I think I can be even briefer on the subject of the
reorganization of ITL. My personal belief there, having worked
inside and outside of NIST, that the management team is very
capable of making decisions like this. I would expect that the
goal of these changes are to align the expertise with a
changing mission requirements and budgetary requirements and
would also believe that this movement to a broader view of
cyberassurance as opposed to pure computer security is a
motivation. IT best practices are increasingly a horizontal
cross-agency issue, and therefore it is logical to consider
this reorganization.
Relative to contributions on the 60-Day Review, the main
missing element that we saw is again the focus on the defensive
architecture. We actually see moving to a more offensive
position. The SCAP framework leads us a long way down that
path. So it is more than just keeping the bad guys out. It is
making sure that the systems are good and deployed as we
intended them. So there is a software supply chain issue. There
is a change management detection issue. A lot of that is being
encompassed in the work at SCAP, and generally industry refers
to these methods as whitelisting methods, in complement to the
black listing methods. Make sure the bad code is kept out, make
sure the good code is good. The combination of those methods is
very powerful.
So in conclusion, I would like to urge NIST to continue
their great work multilaterally with their peers in government
and industry to distill the best of the best ideas into the
NIST standards and methods on a timeline that fully recognizes
that we are behind and heavily exposed.
Thank you, and I welcome any questions.
[The prepared statement of Mr. Starnes follows:]
Prepared Statement of William Wyatt Starnes
Good afternoon Mr. Chairman and respected Members of the Committee.
I appreciate the opportunity to present before this committee today.
My name is Wyatt Starnes, a Founder of SignaCert, Inc. and
Tripwire, Inc., and currently the CEO and President of SignaCert.
Please see my narrative biography for more details on my background and
experience.
I should note for the record that I did serve as member of the
National Institute of Standards and Technology (NIST) Visiting
Committee on Advanced Technology (VCAT), and while I have some recent
experience with NIST and the Information Technology Labs (ITL), I am no
longer serving as a VCAT member.
As you are aware Mr. Chairman, I have been working closely with
both the commercial and government sectors in the areas of information
assurance and cyber security for many years. For the purposes of this
testimony I will generally reference the Information Assurance and
Cybersecurity issues as ``Cyber Assurance'' for the following reasons:
In my opinion labeling our challenge as ``Cybersecurity'' is
limiting. Our full goal must be to address ALL issues that
relate to improving the security, availability, stability and
reliability of the computing devices used to create and deliver
complex IT business processes.
We must address the risks that are hostile in source and
nature (malicious), as well as hardware and software design,
delivery, and maintenance weaknesses (non-malicious) that are
also known to induce risk.
It is well established that undetected non-malicious changes
do increase malicious risk, and also cause IT business service
delivery instability and failure.
It is my belief that we are at a very critical time in our nation's
history with regards to our Cyber Assurance practices. We must act now,
and bring increased creativity, technology and innovation to these
challenges.
I would like to commend this subcommittee, led by Congressman Wu
and his staff, for continuing to direct focus to our cyber assurance
challenges, and the important contributions that NIST has made, and
continues to make, in support of these critical national cyber
assurance priorities.
Specific questions posed by the Subcommittee
The Committee posed three questions for me to address during this
hearing:
1. What could NIST do to address the recommendations in the
60-day review?
2. What are my thoughts and comments on the Reorganization of
ITL?
3. Given the current emphasis on Information Assurance and
Cybersecurity, what are my recommendations on how ITL might
improve its effectiveness or expand its scope/activities and
impact?
NIST and the 60-Day Review
Relative to question one, regarding NIST and the Cyberspace Policy
Review: Assuring a Trusted and Resilient Information and Communications
Infrastructure (the 60-day review), my personal experience tells me
that NIST is already ahead of the curve with its contributions to the
key issues and priorities presented in the 60-day review document that
was delivered to the President.
Before I address these specifically, I would like to briefly
comment on the role of NIST and its legislated mission and budgeted
charter.
As the Committee knows, NIST is a non-regulatory agency founded on
March 3, 1901, as the National Bureau of Standards and was the Federal
Government's first physical science research laboratory.
While it may surprise many citizens, it is no accident that NIST
was created as an agency within the Department of Commerce where its
primary mission is to promote U.S. innovation and industrial
competitiveness by advancing measurement science, standards, and
technology in ways that enhance economic security and improve our
quality of life.
An even simpler way to state this mission is to reduce the friction
of commerce by advancing measurement science, standards and technology.
NIST's role against the 60-day review is clearly in relation to
creating and administering IT measurement standards, technology and
methods to enable better, and more standardized methods for optimizing
the efficacy of cyber assurance methods.
For the purposes of this, my written statement, I would like to
elaborate on some of the specific work accomplished by NIST. While
there is much more Information Technology Labs (ITL) work that deserves
acknowledgment, I will focus these comments on the following areas:
The 800-series Information Technology Support for
Federal Information Security Management Act (FISMA).
The National Software Reference Library (NSRL) work,
and it relationship to the Help America Vote Act (HAVA), and
its potential contributions to FISMA and the Security, Content
and Automation Protocol (SCAP).
The multilateral (public and private) effort to
establish and enhance the SCAP method.
FISMA and the ``800-Series'' body of work:
From the NIST special publication 800-53 Revision 2 (The bold text
was added by this author for emphasis):
The Information Technology Laboratory (ITL) at the National
Institute of Standards and Technology (NIST) promotes the U.S.
economy and public welfare by providing technical leadership
for the Nation's measurement and standards infrastructure. ITL
develops tests, test methods, reference data, proof of concept
implementations, and technical analyses to advance the
development and productive use of information technology. ITL's
responsibilities include the development of management,
administrative, technical, and physical standards and
guidelines for the cost-effective security and privacy of other
than national security-related information in federal
information systems. The Special Publication 800-series reports
on ITL's research, guidelines, and outreach efforts in
information system security, and its collaborative activities
with industry, government, and academic organizations.
With the charter and intent of the work described here (from the
same publication):
This document has been developed by the National Institute of
Standards and Technology (NIST) to further its statutory
responsibilities under the Federal Information Security
Management Act (FISMA) of 2002, P.L. 107-347. NIST is
responsible for developing standards and guidelines, including
minimum requirements, for providing adequate information
security for all agency operations and assets, but such
standards and guidelines shall not apply to national security
systems. This guideline is consistent with the requirements of
the Office of Management and Budget (OMB) Circular A-130,
Section 8b(3), Securing Agency Information Systems, as analyzed
in A-130, Appendix IV: Analysis of Key Sections. Supplemental
information is provided in A-130, Appendix III.
This guideline has been prepared for use by federal agencies.
It may also be used by non-governmental organizations on a
voluntary basis and is not subject to copyright. (Attribution
would be appreciated by NIST.) Nothing in this document should
be taken to contradict standards and guidelines made mandatory
and binding on federal agencies by the Secretary of Commerce
under statutory authority. Nor should these guidelines be
interpreted as altering or superseding the existing authorities
of the Secretary of Commerce, Director of the OMB, or any other
federal official.
Mr. Starnes Observations on the 800-series work:
While the creators and authors of the 800-series publications have
been consistently humble relative to their contributions in bringing
this important work forward, the impact to both government and industry
has been enormous.
I congratulate the dedicated teams across NIST for their work and
I'd like to specifically commend the Director of ITL, Cita Furlani, for
her steadfast vision and support of the implementation of this work by
NIST ITL in order to serve these critical national needs.
Additionally I would like to recognize Ron Ross, Stu Katzke, Arnold
Johnson, Marianne Swanson, Gary Stoneburner and George Rogers and many
others for their contributions to this foundational body of work.
Areas for NIST improvement:
In general, the areas I outline below are already well underway by
NIST, and I raise them to encourage continued focus only:
Make the 800-series documents and recommendations
easier to read and use by the targeted constituencies. Bigger,
in terms of content volume, is not necessarily better. I
support the effort to streamline the 800-series documents
making them more concise and easier to utilize.
Continue to drive emphasis with all federal IT
practices, including FISMA and the supporting standards and
methods, from ``Certification and Accreditation'' (C&A) and
periodic compliance to ``Continuous Monitoring.''
Help America Vote Act (HAVA) and the National Software Reference
Library (NSRL):
From the NIST web site:
The Help America Vote Act:
The Help America Vote Act (HAVA) of 2002 (Public Law 107-252)
was passed by Congress ``to establish a program to provide
funds to States to replace punch card voting systems, to
establish the U.S. Election Assistance Commission (EAC) to
assist in the administration of federal elections and to
otherwise provide assistance with the administration of certain
federal election laws and programs, to establish minimum
election administration standards for states and units of local
government with responsibility for the administration of
federal elections, and for other purposes.''
NIST's roles under HAVA:
HAVA established the Technical Guidelines Development
Committee (TGDC) to assist the EAC with the development of
voluntary voting system guidelines. HAVA directs the Director
of the National Institute of Standards and Technology (NIST) to
chair the TGDC and to provide technical support to the TGDC in
the development of these voluntary guidelines.
In addition HAVA directs NIST to conduct an
evaluation of independent non-federal laboratories to carry out
the testing of voting systems and to submit recommendations of
qualified laboratories to the EAC for accreditation. HAVA also
charges NIST with monitoring and reviewing laboratories
accredited by the EAC.
National Software Reference Library:
From the NIST web site:
This project is supported by the U.S. Department of Justice's
National Institute of Justice (NIJ), federal, State, and local
law enforcement, and the National Institute of Standards and
Technology (NIST) to promote efficient and effective use of
computer technology in the investigation of crimes involving
computers. Numerous other sponsoring organizations from law
enforcement, government, and industry are providing resources
to accomplish these goals, in particular the FBI who provided
the major impetus for creating the NSRL out of their ACES
program.
The National Software Reference Library (NSRL) is designed to
collect software from various sources and incorporate file
profiles computed from this software into a Reference Data Set
(RDS) of information. The RDS can be used by law enforcement,
government, and industry organizations to review files on a
computer by matching file profiles in the RDS. This will help
alleviate much of the effort involved in determining which
files are important as evidence on computers or file systems
that have been seized as part of criminal investigations.
The RDS is a collection of digital signatures of known,
traceable software applications. There are application hash
values in the hash set which may be considered malicious, i.e.,
steganography tools and hacking scripts. There are no hash
values of illicit data, i.e., child abuse images.
Mr. Starnes' Observations on HAVA and NSRL:
In my opinion, HAVA comprises some of the most important technical
work underway by USG to automate and enforce technical and social trust
that helps enable our democratic process. HAVA can and should serve as
a lighthouse for other countries to follow for enabling a seamless,
automated and trusted voting and vote aggregation system.
I note HAVA in my testimony because the methods and technologies
specified under the guidance, and the software measurement methods
developed under the NSRL programs, have tremendous importance and
utility over and above the HAVA use cases.
Essentially HAVA and NSRL represent a practical instantiation of a
``trust-based'' compute model. I believe that trust-based computing
methods are crucial to achieve better and more transparent, holistic
Cyber Assurance for both the government and commercial sectors.
A major tenet of the HAVA/NSRL method is the ``positive system
attestation'' methods required by the HAVA language. Under HAVA,
Software used to operate electronic voting apparatus must be
cryptographically measured and validated to a trusted reference. NSRL
data is used to create the ``trust reference'' for software
attestation.
Generally referred to as software ``Whitelisting'' by industry,
these capabilities promise to ``close the blind spot'' in our view of
IT by establishing the capability to ensure the ``as-deployed''
software state (and ONLY the as-deployed software state) is currently
in place on the IT device or system.
This ``positive trust-based method'' has broad ramifications for
government and industry. By fully utilizing whitelisting techniques we
can:
Reduce the exposure of malicious and hostile software
that is ``hiding in plain sight.''
Establish and prove supply chain validity
(provenance) of the software that is deployed on our mission
critical IT devices ranging from Servers to Blackberry's. This
is increasingly important in the ``outsourced'' and ``open
source'' world that we now rely on.
Increase the transparency and automation of complex
IT system management by creating a systematic ``closed-loop''
measure/validate method. This addresses both malicious and non-
malicious change quickly and efficiently.
Enabling continuous monitoring of the positive state
of the software stack has been shown to dramatically increase
IT uptime and stability, while reducing the labor and manpower
required for the delivery of that capacity.
Mr. Starnes' Recommendations to NIST on Whitelisting:
NIST should explore its role with industry (companies
and standards groups) relating to whitelist content exchange
standards (XML schemas, etc.) in order to ensure that industry
and government content and methods are ``interchangeable.''
This not only serves government customers with improved
frameworks such as SCAP (discussed below), but it also enables
industry to better serve broader government initiatives, such
as HAVA and other extended NSRL-like use cases, such as
improved cyber forensics.
NIST should encourage industry (especially platform
and software vendors) to support supply chain validation
methods, such as whitelisting methods and content, as a
standard practice for IT systems management and security.
Broader adoption and support of Common Platform Enumeration, or
CPE, should also be stressed as a part of the software
measurement for operational monitoring and supply chain
assurance purposes.
The Security, Content and Automation Protocol effort:
The SCAP method is described below.
From the NIST web site:
``The Security Content Automation Protocol (SCAP) is a
synthesis of inter-operable specifications derived from
community ideas. Community participation is a great strength
for SCAP, because the security automation community ensures the
broadest possible range of use cases is reflected in SCAP
functionality. This web site is provided to support continued
community involvement. From this site, you will find
information about both existing SCAP specifications and
emerging specifications relevant to NIST's security automation
agenda. You are invited to participate, whether monitoring
community dialogue or leading more substantive activities like
specification authorship.
NIST's security automation agenda is broader than the
vulnerability management application of modern day SCAP. Many
different security activities and disciplines can benefit from
standardized expression and reporting. We envision further
expansion in compliance, remediation, and network monitoring,
and encourage your contribution relative to these and
additional disciplines. NIST is also working on this expansion
plan, so please communicate with the SCAP Team early and often
to ensure proper coordination of efforts.''
Mr. Starnes' Observations on SCAP:
A major goal with SCAP was to create a normalized ``content'' view,
specifically around IT vulnerability and configuration intelligence.
Using several databases, vulnerabilities and configurations can be
mapped to government IT platforms. This helps serve prescriptive IT
device provisioning and deployment, operational compliance, continuous
monitoring and remediation.
SCAP provides a powerful and extensible set of methods, content and
embedded IT best practices, enhancing system visibility while improving
the validation periodicity for complex IT environments.
SCAP is the culmination of many years of public-private cooperation
and, within government, one of the best examples of multilateral
government -to-government cooperation this witness has seen.
I applaud the efforts of NIST, NSA, DHS, DISA, MITRE and many
others for bringing this ground-breaking best practices and content
method to fruition.
Industry is already working to extend SCAP methods in several ways
including known-provenance image management, as shown within the blue
circle below.
Mr. Starnes' Recommendations on SCAP:
Government IT professionals, including NIST staff and management,
are demonstrating pervasive IT leadership with the SCAP methods. It is
my belief that these methods will become ``de facto standard'' not only
for Civilian Agencies and DOD, but potentially within the commercial
sector.
Vendor support and momentum of the Federal SCAP initiative is
growing rapidly and is already impacting commercial companies on both
the supplier and end-user side. Most of the major information security
companies have, or are readying, SCAP-compliant products for use by
their customers.
Additionally, ISVs are adding SCAP protocol to their software
measurement content, such as the Common Platform Enumeration (CPE)
fields utilized by SCAP.
My personal opinion is that SCAP represents the most significant
and impactful IT standard, content delivery and best practice framework
ever conceived and delivered by the government IT community.
Again I applaud the NIST team, and broader Federal IT community,
for their strong leadership role to conceive and deliver SCAP.
General Observations for the Committee:
We must begin to better focus our IT legislation targeting the
specific results that we want the constituencies to deliver. I favor
emphasis on the use more carrots versus bigger sticks. It is important
to recognize the leadership that led to the creation of important
methods such as SCAP. We must also reward the political-will of the
departments and agencies that are voluntarily stepping up to implement
these important new methods ahead of any regulatory requirement to
adopt.
Rethinking our budgeting and regulatory processes to drive faster real
results:
With FISMA, government has traditionally focused on Certification
and Accreditation (C&A) and periodic compliance checks for agency IT
systems and infrastructure. This has resulted in a ``check list''
mentality where getting a ``better grade'' becomes the focus. This does
not necessarily yield a more secure and robust IT environment.
Additionally, literally millions of dollars and thousands of man
hours are spent by government every year to fill three-ring binders
that are immediately out of date and irrelevant when the C&A process
has been completed. This is driving a false sense of security and is
wastes tremendous capital and consumes precious manpower without
significantly improving our real cyber risk.
We MUST move to systematic and continuous monitoring solutions that
address and adapt to the current realities and dynamic demands of
today's cyber world.
Our risk profile now mandates that we move to a more complete
``sensor'' view (whitelist plus blacklist), along with the active and
systematic vulnerability and configuration checking enabled by the SCAP
framework. We must change our C&A and compliance mindset, to one of
``We are always exposed, so we must continually monitor report and act.
This is just common sense.
I urge our legislators in both the House and the Senate to observe
and support the tremendous technical work being done by government in
partnership with the commercial sector with the SCAP framework.
We (industry and government) are already working side-by-side on
live deployments where broad near real-time continuous monitoring is
the goal. We believe that these goals are immediately feasible and
expect they will quickly prove dramatic improvements in our IT
operational readiness.
There is significant and immediate leverage to be gained by
shifting dollars allocated for FISMA-based C&A and compliance projects
to full-scope continuous monitoring using the SCAP framework. I
strongly recommend to this committee, and other committees involved in
oversight and legislation for targeting improved cyber assurance and
regulation, to consider these suggestions.
If we do this (with the close cooperation of the legislative
branches, EOP/OMB and DOD), significant national cyber assurance
progress can be realized without significant incremental budget impact.
Realigning IT budgeting and spending to our current challenges, and
moving from pure C&A to SCAP-enabled Continuous Monitoring, is likely
budget neutral to positive. Further, it is expected that the immediate
automation advantage will lower the demand for qualified IT personnel
and reduce long-term IT operational expense.
Reorganization of ITL
On this point I can be quite brief. It is curious to me that an
internal reorganization, conducted by the capable and professional
management staff of NIST, should draw as much attention as it has.
While I am not privy to the precise catalysts of, and motivations for,
the contemplated and/or actual organizational event, it seems like the
benefit of any doubt should be yielded to the Acting Director and staff
at NIST.
That being said, like most organizations--government or otherwise--
I would expect that the intent of the reorganization was to realign the
human resources with the changing mission requirements. In this case I
would further expect that NIST has realized that CYBER ASSURANCE
methods and best practices are increasingly a horizontal-cross agency
issue, and its core-competencies should not remain in a silo within
NIST.
If this is the case, I applaud NIST for adjusting to changing
needs, and my only advice perhaps would be a bit more advance marketing
and communication to affected NIST constituencies.
Recommendations on how ITL might improve its effectiveness or expand
its scope/activities and impact in Information
Assurance and Cyber Security
Having worked with NIST from several perspectives for nearly a
decade, I have only the deepest appreciation for the dedicated
scientists and staff at NIST. I often use the story with family and
friends to explain the reach and impact of NIST in the physical world
by using the following statement:
In any room, in nearly any country, in any sector of our commercial
endeavor--look around that room and I can almost assure you that at
least SOMETHING in the environment has been touched, driven or impacted
by work done at NIST.
Now when I look from my day-job perspective--and take that same
view from a cyber assurance point of view and ask ``What impact has
NIST had on the security, reliability, stability, and utility of the
operational computing infrastructure?'' . . . We still have work to do.
I encourage NIST, perhaps with even a greater sense of urgency, to
continue with its core mission of standards and best practices as they
relate to the broader cyber assurance goals and objective.
I further encourage NIST and its government partners in these areas
including NSA, DISA, DHS and others, to embrace more ``out of the box''
thinking around the cyber assurance challenges that the Nation is
facing.
TIME IS OF THE ESSENCE:
Mr. Richard Marshall, senior information assurance representative
for the Office of Legislative Affairs at the National Security Agency
(NSA) said at a public event recently, ``We're polishing stones instead
of creating stones,'' he said. ``If we don't do something in the near-
term, there won't be a long-term. We are running out of time.'' I
agree.
I encourage NIST to consider the following actions:
Continue to create and advance measurement standards
and methods for Cyberspace.
We must do this by continuing to improve our
NEGATIVE AND DEFENSIVE posture:
This is the Risk and Vulnerability
perspective--are we effectively identifying the
``Bad things and risky things'' in our computer
environment--and improving the common language
to express and communicate these risks.
NIST has done some great work in
these areas including the Common Vulnerability
Scoring System (CVSS) and National
Vulnerability Database (NVD).
We need to continue to emphasize
these as OPERATIONAL METHODS as opposed to
(only) Certification and Accreditation (C&A)
and compliance methods.
We need to supplement these negative detection and
enforcement methods with an improved POSITIVE POSTURE:
This is where the prescribed ``good state'' perspective
is captured and enforced. We need operational methods
and standards that measure ``the known and good state''
to assure that our deployed computer environments are
intact. We can also address important supply chain
provenance issues with these same techniques.
NIST has already worked in these
areas but they appear ``less connected'' with
the some of the methods described above. Much
of this work is apparent in the National
Software Reference Library (NSRL) and the Help
America Vote Act (HAVA).
Many of the same ``positive
attestation'' and trust attestation controls
required by HAVA can and should be applied to
SCAP-enabled IT operational best practices.
In my view there are MANY parallels between the ways NIST has
contributed to this in the physical world for the last 108 years.
Software, software assemblies and indeed entire software ``stacks''
used to enable and enhance our way of life, can and should be measured
and operationally attested.
I urge NIST to continue to work multilaterally with their peers in
government and industry on all the methods I mentioned above, and to
distill these ``best of the best'' ideas into NIST standards and
methods on an even a faster cycle than normal.
Summary
We are a crucial time in our history on multiple fronts. While I
fully acknowledge that we are a vendor of methods used to improved
cyber assurance, my primary motivation to ``join the team'' around SCAP
and other important developments has been citizen-centered.
We are in a race of dramatic proportions and potential risk, and we
are behind. Our National and Economic Security are at risk and if we
can improve this as a team, then we must take action now.
We must advance the state-of-the-art in Cyber Assurance in order to
get to the next level of visibility, control and efficiencies. Extended
SCAP methods, along with Continuous Monitoring, are our best chance of
getting ahead of our adversaries, and scaling that advantage quickly
and efficiently across the federal enterprise.
I respectfully submit that our technical teams have given us the
tools to significantly raise our odds of closing the large cyber
assurance gap we now face. It is imperative that our legislative and
executive branches show the political-will, and the program and
financial resources to enable us to succeed.
Thank you and I welcome any questions from the Committee.
Biography for William Wyatt Starnes
William ``Wyatt'' Starnes was born in Atlanta, Georgia in October
1954. Mr. Starnes had a deep and immediate interest at an early age in
everything mechanical, electrical and the emerging electronics
industry. He built his first photocell sensor electronic project for a
science fair while in still in elementary school. He went on to
graduate from Ygnacio Valley high school in Concord, California in 1972
knowing that computers and electronics would become his life's work.
After graduating, Mr. Starnes was restless and ready to go to
Silicon Valley to begin his career. He took a highly focused path
graduating from Control Data Institute of Technology with an Associate
Art's degree in computer science, and began his professional career
with Data General (DG) Corporation in Sunnyvale, California in 1973.
Mr. Starnes' insatiable curiosity about ``how things work''
continued in Silicon Valley involving himself in ``everything
semiconductor'' for the first several years. This work included
everything from detailed courses in semiconductor physics to software
design and engineering with many of the early programming languages.
His early career was centered on semiconductor automated testing and
measurement. Mr. Starnes not only helped design the first semiconductor
memory and microprocessor devices for DG, he wrote or co-wrote all of
the test programs used to verify the functionality of these complex
chips.
Data General was the first of many successive entrepreneurial
experiences for Mr. Starnes. He went on from DG to Monolithic Memories
and helped to build the first MOS and CMOS processes and devices,
including the 1k and 4k MOS dynamic RAM's. While still focused on
programming of Automatic Test Equipment (ATE), he went on to Maruman
Integrated Circuits, creating one of the first ``Fab-less
Semiconductor'' resources in Silicon Valley. Maruman produced (and Mr.
Starnes wrote the test programs for) much of the Atari game devices in
late 1970's.
After having made significant technical contributions in the ATE
industry, in 1978 Mr. Starnes took an early stage management position
with MegaTest Corporation. Megatest revolutionized ATE by inventing and
delivering the most cost-effective test and measurement equipment ever
delivered. This contribution was viewed as critical to Intel
Corporation, AMD, National Semiconductor and many others, for testing
complex integrated circuits at a fraction of the cost of previous
solutions. Interestingly this breakthrough had to do with ``reference
testing,'' which would become a model for additional breakthroughs in
software assurance methodologies.
Mr. Starnes moved to Tokyo Japan in 1981 for two years to found
MegaTest Japan. This provided much needed international market
perspective to Mr. Starnes resume, and he continues to be very active
in the Asian market.
After a 20-year career in semiconductor manufacturing and testing,
Mr. Starnes made the shift to software in 1993. Having moved from
Silicon Valley to Portland, Oregon in 1989--Mr. Starnes began a new
chapter of his career with Infinite Pictures (now iMove). This company
did pioneering work in 3-dimensional visualization software and
hardware. iMove is now one of the leading producers of fixed and mobile
surveillance devices for industry and government.
While Mr. Starnes has always maintained his deep technical roots,
he has continued to expand his management, sales and marketing
expertise. He has been deeply involved at the senior management level
of every company he has worked in since 1973. This has allowed him to
remain both technically adept at the ``street level,'' while
maintaining senior executive relationships across many enterprise and
government sectors.
While at Infinite Pictures Starnes met Gene Kim, with whom he went
on to found Tripwire, Inc. with in 1997. The Tripwire software was
developed by Gene Kim under the close guidance of Purdue University
professor Eugene Spafford (aka Spaf) beginning in 1991.
While CEO of Tripwire Mr. Starnes grew the company rapidly and was
awarded Inc. Magazine's 20th Fastest growing company in America award
in 2002. More importantly, the Tripwire technology and products began
to alter the state-of-the-art in information security and assurance by
bringing the notion of integrity management to the market.
Due to a medical issue (early stage cancer) in the summer of 2003,
Mr. Starnes left Tripwire to seek a cure with his family and doctors
support. Quickly recovering after successful treatment, he returned to
the software assurance and cyber security industry in spring 2004 with
the formation of SignaCert, Inc.
It was in this timeframe that Mr. Starnes was invited by the Acting
Director of NIST to serve on the Visiting Committee on Advanced
Technology, or VCAT, which he served in that capacity from 2005 to
2008. Mr. Starnes also presided as the first Chairman of the IT
Subcommittee under the NIST-VCAT Oversight Committee in 2007 and 2008.
While at SignaCert, Mr. Starnes and his team have continued to
drive the ``think differently'' vision in dealing with complex
information security, compliance and information assurance. The
fundamental breakthrough, covered now by two U.S. Patents, is that
software can and should be ``measured.'' This led to the long-term
development of Global Software Trust Services based on the measurement
of software that is built by the Independent Software Vendors, or ISVs.
In a way similar to the ATE methods now commonly used by companies
such as Intel to test and verify semiconductor devices (as developed by
MegaTest), SignaCert builds ``reference views'' of software, using
software measurements, or ``whitelists'' to assure that IT devices
(servers, workstations, routers, mobile devices, etc.) are in alignment
to the prescribed reference measurement set, or ``gold image.''
This information assurance method, is complimentary and additive to
traditional perimeter-centric, reactive and defensive IT methods (such
as firewalls, intrusion detection, and anti-virus) by ensuring the
established, known and presumed trusted, IT state is maintained over
the deployment and usage life cycle of that IT device.
Knowing that the ``as-deployed state'' is accurate to a control
reference has been shown to contribute immediate benefits for all
market sectors and customers that depend on complex IT to deliver
critical business and mission services.
In addition to improving cyber security against both inside and
outside risk and adversaries, the method has been shown to dramatically
increase mean-time-between-failure (MTBF) and reduce mean-time-to-
repair (MTTR), which serve to increase IT business process stability
and availability, while reducing the requirement for trained people to
manage complex and broadly scaled IT infrastructure.
Mr. Starnes continues to passionately pursue his primary career
mission of improving cyber assurance by providing greater efficacy and
more transparency. Critical to this mission is lowering both costs and
resource requirements through enabling automation across all critical
enterprise sectors and geographies.
Chairman Wu. Professor Schneider, please proceed.
STATEMENT OF DR. FRED B. SCHNEIDER, SAMUEL B. ECKERT PROFESSOR
OF COMPUTER SCIENCE, CORNELL UNIVERSITY
Dr. Schneider. Thank you, Mr. Chairman. NIST's Computer
Security Division serves today as a trusted source of expert
information about secure computing. The recent proposal to
reorganize the division in my opinion threatened its
effectiveness and thus could have undermined a key national
resource for civilian cybersecurity. Therefore, my remarks here
will focus on CSD organization, but I will be prepared to
answer other questions later.
What had been proposed involved two elements. The first
element had the head of CSD reporting higher up in NIST's
management chain. This would have been good. Higher levels of
NIST's management increasingly will want to understand and
champion computer security activities so they can secure needed
resources and can provide guidance throughout the Federal
Government.
The other element of the reorganization involved redefining
which projects are part of CSD. CSD would no longer be the home
for all cybersecurity activities within the information
technology lab. I have not heard a compelling rationale for
this, and I am not sure one exists.
First, I fear that having computer security activities
outside of CSD would erode the CSD brand. This brand is a
valuable asset. It keeps CSD visible to its customers so they
know where to come for help, and it enables CSD to attract
talent because CSD employees are seen to have an impact on
computer security, both domestically and internationally.
Second, I am concerned about loss of budget accountability for
computer security activities. Put all the activities in a
single division and it will be easy to ascertain that funds
appropriated to NIST for cybersecurity are used as intended.
Disburse cybersecurity activities over multiple divisions and
the funds will be intermixed with funding for other activities.
Finally and perhaps most important, I see no intellectual
basis for deciding what computer security activities to place
outside of CSD and what other activities to place inside of
CSD. However, I do see difficulties when people who are working
on closely connected initiatives are not under the same
management. It removes imperatives for cooperation, for
rational budgeting, and makes comparisons of people and
projects difficult. So the proposed reorganization seemed to
offer few benefits.
But a slightly different reorganization actually could have
been a very wise move. Looking ahead, CSD will have to assume a
larger role because trustworthy computing is so central to the
future of our nation's critical infrastructures, private sector
systems, and the Administration's new initiatives in healthcare
and SmartGrid. Growth will be necessary to meet these needs.
Although the recent reorganization proposal makes no allowance
for such growth, there is a plan that does. Elevate CSD to
become a laboratory in NIST so that it is parallel to the
information technology lab currently housing CSD. With this
alternative proposal, the director of the new lab would report
higher up the NIST management chain, the CSD brand would be
protected and perhaps even strengthened. Budget control and
accountability are facilitated by having all and only computer
security activities under one director, and there would be no
need to separate various efforts that intellectually are
closely related.
In sum, I find that entertaining a reorganization of
today's CSD is sensible, but the recently proposed
reorganization lacks a rationale and seems to create problems
without offsetting benefits. An alternative reorganization that
elevates CSD to form a new computer security lab at NIST has
much to recommend it. Thank you.
[The prepared statement of Dr. Schneider follows:]
Prepared Statement of Fred B. Schneider
Mr. Chairman and Members of the Committee, I appreciate this
opportunity to comment on the role, activities, and proposed
organizational changes within the Computer Security Division at the
Information Technology Laboratory of NIST. I am Fred B. Schneider, a
Computer Science professor at Cornell University and Chief Scientist of
the NSF-funded TRUST\1\ Science and Technology Center, a collaboration
involving researchers at U.C.-Berkeley, Carnegie-Mellon University,
Cornell University, Stanford University, and Vanderbilt University.
---------------------------------------------------------------------------
\1\ Team for Research in Ubiquitous Secure Technology.
---------------------------------------------------------------------------
I have been a Computer Science faculty member since 1978, actively
involved in research, education, and in various advisory capacities for
both the private and public sectors. Besides my work at Cornell, I
today serve as member of the Computing Research Association's Board of
Directors and as a council member of the Computing Community
Consortium. I also co-chair Microsoft's TCAAB external advisory board
on trustworthy computing. And perhaps most relevant to today's hearing,
I have served since Sept. 2006 on the Information Security and Privacy
Advisory Board (ISPAB), a Congressionally mandated FACA board that
advises NIST, the Congress, and OMB about cybersecurity in Federal and
civilian computer systems. The comments that follow are my own
opinions, however.
Our nation's needs for secure systems will surely grow over the
next decade. The networked computing systems employed today to operate
critical infrastructures (e.g., energy distribution, banking, finance,
transportation, and communication) are vulnerable to attack. Systems
running our civilian government offices and private sector business are
also vulnerable. And we, as a nation, are now discussing a ``smart
grid'' for energy distribution and a new health care system that will
depend critically on computing systems that must be trustworthy.
Activities performed by Computer Security Division (CSD) are critical
to the success of all.
CSD plays a special and important role for the Federal Government
and the private sector, by serving as a respected source of objective
information about ways to build and operate secure computing systems.
This role is possible only because
CSD is able to attract top talent,
CSD is situated within an institution-NIST-where
research is valued and is being conducted (even though only
some CSD activities are, in fact, research), and
CSD can be trusted as an advocate of security, by
virtue of not being part of a law enforcement or national
security organization, since there is then no basis for concern
about CSD developing standards with a hidden purpose of
collecting information.
Question: The Cyber Space Policy Review makes a number of
recommendations to improve federal efforts for cybersecurity. Examples
of these recommendations include the establishment of a single federal
entity to act as a locus for U.S. involvement in international
standards, increased public education and awareness, and a larger focus
on identity management. What could NIST do to address these and other
recommendations from the Cyber Space Policy Review?
NIST--and within NIST, CSD--indeed serves as a locus for U.S.
involvement in international standards, increased public education and
awareness related to cybersecurity, and a larger focus on identity
management. Despite a modest budget, CSD has succeeded admirably in
these tasks; I urge that it be supported to continue and expand these
activities.
There is also much other work to be done in support of civilian
system cybersecurity, especially with the crying need to revise FISMA
and with the Administration's initiatives to create the expertise and
standards for smart grid and health care. NIST is the right place to do
this work and should aggressively embrace these challenges by
increasing the size and funding for CSD.
Moreover, as noted above, CSD is ideally situated to provide
cybersecurity information that its customers can trust. Other federal
agencies (e.g., DHS, NSA, FBS, CIA, DOD) also have important roles to
play in the cybersecurity landscape, but each has a mission that can
only engender suspicion by a private sector wary of government
surveillance. So these other federal agencies could neither replace nor
host CSD activities.
Question: NIST is proposing a reorganization of ITL. What is your
assessment of this reorganization and how will it improve the outcomes
of ITL activities?
Plans for the reorganization of NIST's Information Technology
Laboratory (ITL) and CSD first came to my attention about four months
ago, in July. All of the details have still not been made public, but
there was a public discussion of some aspects of a proposed CSD
reorganization about two weeks ago (at the Oct. 7, 2009 ISPAB meeting).
The key parts of the reorganization described to me have two
elements:
The Office of the Associate Director for
Cybersecurity Research and Development reports higher-up in the
ITL management structure.
The set of projects under CSD is changed slightly,
with a few projects whose names suggest they concern
cybersecurity being moved outside of CSD while other projects
whose names suggest they have a significant content that does
not concern cybersecurity being moved into a new CSD with a new
name.
Note, the two elements are largely independent.
The first element, having CSD report-in higher-up the management
chain, seems wise and even prescient, given the growing need for
services that CSD now provides or will need to be providing in the near
future. Higher-levels of NIST's management will have to understand and
champion the activities of CSD, to ensure sufficient resources are
available to support cybersecurity efforts and to provide guidance to
other federal and civilian decision-makers in a world where
cybersecurity matters are growing pervasive. Notice, also, that this
first element of the proposed reorganization directly impacts a small
number of people but offers enormous leverage.
The second element of the proposed reorganization affects a much
larger number of people--all those involved in CSD projects plus some
others within ITL. Any reorganization that potentially affects many
people tends to be disruptive (and this one already seems to have had a
significant impact on the esprit de corps within CSD), so such change
is best contemplated and undertaken only when there are significant
gains to be had. In evaluating any proposed reorganization of CSD, I
think that we should want to know:
To what extent does the proposed reorganization
leverage investments and personnel? For example, what is the
overhead for management and for communication within the
proposed reorganization, as compared with the current
organization?
To what extent does the proposed reorganization
facilitate or impede inefficiencies, collaborations, synergies,
and informed trade-offs by virtue of shared management. For
example, how would changing which projects share mangers
benefit or harm each effort as it competes for budget, other
resources, ratings, promotions, etc.
Does the proposed reorganization change the
visibility of CSD activities to NIST management (which must
make budget trade-offs and advocate for CSD outside of NIST) or
to CSD customers (Federal Government civilian agencies and the
private sector).
Does the proposed reorganization facilitate better
accountability for budget appropriations intended to enhance
activities in computer security?
Does the proposed reorganization better position NIST
to support expected future needs (such as changes to FISMA to
require continuous monitoring of systems and improved security
metrics, the Administration's new smart grid and health care
initiatives, and our nation's ever-increasing dependence on
networked systems both within the government and private
sectors)?
Yet I am aware of no analysis that answers the above questions. I
myself am not familiar enough with the details of ITL and CSD to
attempt such an analysis. But I can offer some general guidelines for
designing a good CSD organizational structure.
The CSD brand is a valuable asset. It serves as a clear and obvious
point of engagement for customers. That both (i) increases the
efficiency of interactions between CSD and customers and (ii) increases
the chances that those in need will know to seek CSD expertise and to
embrace CSD standards and other guidance.
The CSD brand also means that
(1) CSD accomplishments,
(2) the unique role and impact CSD has on the computer
security landscape internationally (through encryption
standards) as well as domestically (through other standards and
guidance, too), and
(3) the problems CSD addresses
together make CSD an exciting place to work. This, in turn, has enabled
CSD to recruit an outstanding staff, despite the scarcity of computer
security experts and despite competition for their services (with
considerably better compensation) from the private sector. A CSD
reorganization that erodes the CSD brand by eliminating the name or by
diffusing the organization's efforts into a larger pool of computer
science activities should therefore not be undertaken lightly.
In addition, mixing computer security activities and other computer
science efforts complicates accountability of computer security budget
appropriations. Creating decreased management visibility into how
budget is divided seems unwise, as we enter an era where Congress will
doubtless be providing increased budgets to NIST in order to serve the
ever growing computer security needs of our nation.
Finally, I see no benefits from dividing cybersecurity activities,
locating some in an organization that is mostly populated by
cybersecurity experts but others in an organization that is not.
I can see no intellectual basis that could be used to
decide today on such a partitioning of cybersecurity projects,
much less to decide on a partitioning that is likely to remain
sensible for a future where our understanding of cybersecurity
will almost certainly have evolved. To give an extreme case,
there once was a time when it made sense for those studying
privacy and other policy matters to be organizationally
separated from technologists. That separation is no longer
sensible, however--technologies are typically useless when
developed by people ignorant of policy, and policy developed by
people who don't understand technology is often damaging to
innovation and growth. So CSD ought to include both, yet the
proposed new reorganization seems to be considerably narrower
and includes only a subset of the technologists.
There is also a matter of styles. Some members of CSD
engage in research, and some engage in activities that have a
very different character-writing standards, compiling best
practices, etc. The rest of ITL is primarily concerned with
research. If all computer security activities were located in
CSD, then this difference would be accommodated by the
organizational structure. In contrast, diffusing the one kind
of activity within the other will likely lead to an
organization that is difficult to manage and has various
different classes of citizens.
From my analysis and the guidelines I proposed above, I conclude
that NIST management would be wiser to be contemplating a new
laboratory--CSL (instead of CSD)--in parallel to ITL, instead of making
changes to the organization of ITL. Choosing which specific projects to
place in CSD, as advocated by the second element of the proposed
reorganization, simply offers no leverage but has the potential to
create problems. A new CSL structure, however, would satisfy all of the
requirements I noted above: (i) the director would report higher-up in
the NIST management chain, (ii) CSD function would be even more visible
and have a stronger identity, (iii) budget control and accountability
is facilitated, and (iv) there is no need to separate projects that are
closely related.
Question: Given the current emphasis on information assurance and
cybersecurity, what recommendations do you have on how ITL might
improve its effectiveness or expand the scope of its activities and
their impact?
Looking to the future, the functions performed today within CSD
will play a bigger and bigger role in how the Federal Government and
the private sector protect their computer systems. Smart grid and
computerized support for health care, for example, raise new computer
security questions. The current discussion about ``accountability of
action'' for enforcing security on our networks raises numerous issues
involving both technology (e.g., how to attribute packets in transit)
and policy (e.g., how to manage trade-offs with privacy)--topics that
fall squarely in the expertise of CSD. And no matter what happens with
a U.S. universal identity card, questions about federated identity
still need to be sorted out as various public sector and private sector
organizations create identity management systems on the Internet.
In short, the need is there today for a CSD that is much larger
than its current size; and the needed work cannot be done in the
private sector, because of inherent conflicts of interest and
commitment. I conclude that CSD will have to grow in size significantly
over the next five to ten years.
But CSD growth raises another issue about the recently proposed
efforts to reorganize ITL and CSD. The proposed reorganization does not
group all cybersecurity efforts together in a single CSD presumably
because that division would be too large. So yet another reorganization
would be required to accommodate significant growth in CSD activities.
If, instead, a CSL is created today, then we would be putting in place
an organization that not only satisfies its requirements for today but
would continue to meet its requirements for a long time to come. And
that strikes me as by far the more sensible course.
Biography for Fred B. Schneider
Fred B. Schneider is Samuel B. Eckert Professor of Computer Science
at Cornell University. He joined the Cornell faculty in Fall 1978,
having completing a Ph.D. at Stony Brook University, preceded by a B.S.
in Engineering from Cornell in 1975. Schneider currently also serves as
the Chief Scientist for the NSF-funded TRUST Science and Technology
Center, which brings together researchers at U.C.-Berkeley, Carnegie-
Mellon University, Cornell University, Stanford University, and
Vanderbilt University.
Schneider's research has focused on various aspects of trustworthy
systems--systems that perform as expected, despite failures and
attacks. His early work concerned formal methods to aid in the design
and implementation of concurrent and distributed systems that satisfy
their specifications; he is author of two texts on that subject: On
Concurrent Programming and A Logical Approach to Discrete Mathematics
(co-authored with D. Gries). He is also known for his research in
theory and algorithms for building fault-tolerant distributed systems.
For example, his paper on the ``state machine approach'' for managing
replication received an SOSP ``Hall of Fame'' award for seminal
research. More recently, his interests have turned to system security.
His work characterizing what policies can be enforced with various
classes of defenses is widely cited, and it is seen as advancing the
nascent science base for security. He is also engaged in research
concerning legal and economic measures for improving system
trustworthiness.
Schneider was elected Fellow of the American Association for the
Advancement of Science in 1992, the Association of Computing Machinery
in 1995, and the Institute of Electrical and Electronics Engineers in
2008. He was named Professor-at-Large at the University of Tromso
(Norway) in 1996, and was awarded a Doctor of Science honoris causa by
the University of NewCastle-upon-Tyne in 2003 for his work in computer
dependability and security.
Schneider has served since Sept. 2006 as a member of the
Information Security and Privacy Advisory Board (ISPAB), which advises
NIST, the Secretary of Commerce, and the Director of OMB on information
security and privacy issues pertaining to Federal Government
Information Systems. He chaired the National Academies CSTB study on
information systems trustworthiness that produced the 1999 volume Trust
in Cyberspace. He also served as a member of CSTB from 2002-2008 and
served from 2004-2007 on the CSTB study committee for improving
cybersecurity research. Schneider was a member of the NSF CISE advisory
committee 2002-2006. And in Fall 2001, he chaired the United Kingdom's
pentennial external review of research funding for academic Computer
Science.
In 2007, Schneider was elected to the Board of Directors of the
Computing Research Association (CRA) and appointed to the steering
committee of CRA's Computing Community Consortium. CRA is an
association of more than 200 North American academic departments of
computer science, computer engineering, and related fields; part of
it's mission is to strength research and advanced education in the
computing fields and to improve public and policy-maker understanding
of the importance of computing and computing research in our society.
Schneider is a frequent consultant to industry, believing this to
be an efficient means of implementing technology transfer as well as
learning about the real problems. He is Co-Chair of Microsoft's
Trustworthy Computing Academic Advisory Board, which comprises outside
technology and policy experts who meet periodically to advise Microsoft
about products and strategy. He also provides technical expertise in
fault-tolerance and computer security to a variety of firms, including:
BAE Systems, Fortify Software, Lockheed Martin, and Microsoft.
Chairman Wu. Thank you very much, Dr. Schneider. And Mr.
Bohannon, please proceed.
STATEMENT OF MR. MARK BOHANNON, GENERAL COUNSEL AND SENIOR VICE
PRESIDENT FOR PUBLIC POLICY, SOFTWARE & INFORMATION INDUSTRY
ASSOCIATION (SIIA)
Mr. Bohannon. Thank you, Mr. Chairman, Ranking Member
Smith, Congresswoman Edwards. It is a pleasure to be here today
on behalf of the more than 500 members of SIIA, the principal
association of software companies, to discuss with you NIST's
cybersecurity activities in the context of the 60-Day Review.
As has already been indicated by the panel, that review was
quite comprehensive in its outlook. Even by its own statement,
it touched virtually everyone and everything we do in our
society, and I think I certainly share with my colleagues the
anticipation that the cyber coordinator will be announced soon.
But I think you can boil down the thrust of that review
into three things. First, that we have got to take action to
enhance the security of our Federal Government systems; second,
that we need to continue to enhance the public/private
partnership to make sure our infrastructure is secure; and
third, that we need to partner effectively with the
international community since this is a global problem, not
just a U.S. problem.
And in our view, the challenges, these three challenges,
mean that NIST and thereby the Secretary and Department of
Commerce have an absolutely essential and critical mission and
contribution to make to seeing where the 60-Day Review goes.
To be more precise--and I welcome Ms. Furlani's update on
what is going on with the ITL. I have known her for many years
and look forward to working with her on where this could
possibly go once they have stepped back from this program. The
reality is that this change in NIST review comes at a very
critical time about the direction I think we are going to take
with the cybersecurity review, and one of the key questions is
whether its implementation is going to be informed
predominantly by the military intelligence framework on
cybersecurity or whether it is going to be able to adapt across
a wide-variety of sectors and parts of our economy.
Our view, based on the experience so far, is that you have
got to have that blend of perspectives for it to work, and if
it is going to be effective, it means that NIST must be
enhanced and reinvigorated in its role, and thereby the
Department and the Secretary must play a leadership role in
where the 60-Day Review is going to be carried out.
So therefore, we think rather than looking at what are the
merits or non-merits of the ITL reorganization, this is a great
time to look at really where the future of NIST and its
cybersecurity activities need to go.
Mr. Chairman, in our testimony we make a number of
recommendations and ask some key questions which I assume have
been submitted for the record. Let me try to summarize those
here.
First, we urge the Committee as it has done for decades to
make sure that NIST does not become a regulator of private-
sector actions. You all have been very consistent in making
sure that NIST remains a first-class laboratory, not a fifth-
class regulator. NIST does best and carries out its mission
when it collaborates with the private sector, not try to impose
government-defined standards or technologies on the private
sector, and my testimony goes through some examples where they
have come very close to that line without a great deal of
success, and in my view some negative consequences.
The second thing we would urge, and is consistent with some
of the other panelists, is that we would urge serious
consideration to making the Computer Security Division a stand-
alone laboratory. We have heard three key challenges facing the
Computer Security Division. One is funding, one is staffing and
recruiting and retaining good staff, and the third is enhancing
and reinvigorating the global brand. We think that currently
the CSD, being one of six divisions inside of ITL, and ITL
being one of ten laboratories inside of NIST, is not really the
right framework in which that can occur. And so again, we know
that there are issues involved in doing any reorganization, but
we think that there needs to be serious consideration given to
this. Creating a cyber information security, information
assurance lab, I don't want to get hung up on the name. We
think it would send a very important signal to the private
sector and to the world that the United States Government is
taking its role very seriously in this regard.
The third recommendation we would make is that NIST needs
to make sure that its primary customers, agencies of the
Federal Government, are the focus of its efforts, and the
committee is well-aware of its responsibilities in that regard.
The fourth recommendation we would make is that NIST needs
to continue to work with the private sector and the political
leadership of the Commerce and USTR (United States Trade
Representatives) among others as we work to roll back some of
the ridiculously stringent regimes that we are seeing from
other governments which are trying to impose indigenous or
unique standards in this area.
I was not able to appear in the June hearing, Mr. Chairman,
because I was in China personally working to try to roll those
back, and while we certainly depended on the leadership of
Ambassador Kirk and Secretary Locke in getting that done, NIST
was an absolutely essential partner because of its perceived
global reputation as an independent assessor, independent
evaluator, credible place where we could talk about legitimate
ways of approaching these issues globally. That is going to
become more important as we see countries like India and Russia
also beginning to take on those efforts.
So with that, Mr. Chairman, I just want to say that we
think that NIST and the Department of Commerce have an
absolutely essential role. We are very pleased to see that
Secretary Locke in particular has brought some terrific people
in who are really beginning to focus on these issues. We
commend those steps. We commend this hearing, and we look
forward to working with you and the executive branch to carry
out these goals. Thank you.
[The prepared statement of Mr. Bohannon follows:]
Prepared Statement of Mark Bohannon
Chairman Wu, Ranking Member Smith, Members of the Committee, on
behalf of the more than 500 members of the Software & Information
Industry Association (SIIA), the principal association of the software
and digital content industry, we appreciate the opportunity to discuss
the current cyber and information security activities of the National
Institute of Standards and Technology (NIST) and how they fit into the
action plan of the Cyber Space Policy Review (60-Day Review). As the
Committee is aware, I also served as an official at the Department of
Commerce during the 1990's working with NIST on computer security
issues.
The 60-Day Cyber Space Review was an extraordinarily comprehensive
document, recognizing that ``cyberspace touches practically everything
and everyone.'' \1\ We are not alone in awaiting the appointment of a
White House coordinator to undertake the many and varied `next steps'
that the Review identified.
---------------------------------------------------------------------------
\1\ Preface, Cyberspace Policy Review, p. i.
---------------------------------------------------------------------------
Among the central thrusts of the Review is that action must be
taken, first, to enhance the security of the Federal Government's
systems; second, to continue and enhance the public private-partnership
that is essential to securing our nation's infrastructure; and, third,
to partner effectively with the international community.
In each of these vital challenges, NIST--and thereby the Secretary
and Department of Commerce--has an essential and critical mission and
contribution to make.
We read news reports of a possible reorganization of NIST's
computer security areas of competence. I must emphasize that I am
relying entirely on published reports on this matter. However, we are
concerned about these reports regarding the future of NIST's Computer
Security Division (CSD).
If this proposed reorganization would separate--some would say
bifurcate, some would say disperse--the activities of NIST's basic
research functions from those of its applied-external activities (which
include its evaluation processes and engagement internationally), this
would be in our view a serious detriment to the ability of NIST and the
Department to step up to the plate if and when the Cyberspace Review is
undertaken systematically.
This potential change in NIST computer security functions is taking
place as the 60-day Review--and the direction it will take--remains a
work in progress. One key question is whether its implementation will
be informed predominantly by a defense-intelligence framework and the
related assumptions about cybersecurity. If the follow-on to the 60-day
Review is going to be meaningful across a variety of commercial sectors
and viable economically, there must be strong leadership from the
Department of Commerce--and that cannot occur without an effective and
enhanced role of NIST.
It is also occurring as we face mounting global challenges, which
include efforts by other governments to undertake stringent
cybersecurity regimes outside of global norms. There are also important
efforts underway to focus on the next generation of international
frameworks for assuring cross-border analyses of vulnerabilities and
bases for product evaluation.
Therefore, it is an opportune time to look at how to make sure
NIST--and the Department--are prepared and ready to engage the
interagency process, the public and our international partners with a
view to the future.
In Appendix A, we outline a number of questions that we believe are
timely and essential to NIST's role in cyber and information security,
and very relevant to the 60-day Review objectives. Let me summarize
them here.
First, we urge the Committee, as it has consistently done by
decades, not to make NIST a ``regulator'' of private sector actions.
NIST has effectuated its mission best through long-standing
collaboration with the private sector. This collaboration, which is not
replicated to the same degree by any other agency of the Federal
Government, has benefited not only government agencies (which are the
first line customers of NIST's work), but also our nation's
infrastructure, innovation environment and competitive strength.
When NIST has ventured away from this mission and collaborative
approach, the result has been injurious. For example, in undertaking
Federal Information Processing Standards for federal agencies, NIST has
recognized (including making mandatory) controversial cryptographic
implementations like Clipper Chip and Skipjack (which are still
identified for government use). The controversies around these
approaches are enormous.\2\ NIST is not equipped to become a regulatory
body which proscribes specific standards for the private sector, nor
would it be desirable to make it such, as it would inherently distract
from its core competencies and mission. Instead, it is critical to look
ahead to the next generation of challenges, which require NIST to
remain the globally recognized forum for reaching consensus on key
issues (as it did with the highly successful competition to identify
the Advanced Encryption Standard), and reinvigorating its recognition
as a world-class laboratory.
---------------------------------------------------------------------------
\2\ See ``The Clipper Chip'' (http://www.epic.org/crypto/clipper).
Second, we would strongly urge consideration to making the Computer
Security Division a separate lab within NIST should be a priority. The
CSD is one of currently six Divisions within the Information Technology
Laboratory (ITL), which is itself one of 10 laboratories within the
NIST organization. This action--creation of a stand alone Cyber and
Information Security Lab--would send an important signal, both to
Government agencies and to the private sector, and enhance the NIST
`brand' in this important area. As a Division within one of 10
competing Labs at NIST, the Division is, for example, handicapped in
its recruiting and retention of quality employees. For example, the
Division Chiefs are not Senior Executive Service (SES) position.
To state the obvious, this recommendation is in direct contrast to
any suggestion of dispersing or bifurcating the computer security
functions of NIST, which would present serious risks to the funding and
global branding of NIST in cybersecurity work. It would also compound
the problems that NIST has been facing in recent years.
On the one hand, NIST--specifically the Computer Security
Division--has been handed in recent years a number of legislative
mandates, including some that have not been funded.\3\ This compounds
the on-going funding paradigm of the Division (which is shared by other
NIST Labs) that requires it, except in rare years, to get up to 40
percent of its funding from other agencies (or engage in cost-
reimbursement work through CRADAs), since appropriation funds may
account for as little as half of the year's program.
---------------------------------------------------------------------------
\3\ See, e.g., Cybersecurity R&D Act (2002).
---------------------------------------------------------------------------
On the other, the work of the Division on broad-based research,
including those initiatives that benefit both the public and private
sectors, is increasingly under pressure due to the demands of other
agencies, including the Office of Management and Budget (OMB), for
assistance to other Federal agencies in computer security. These
demands are compounded by the growing mandatory imposition of NIST
work--whether in the form of FIPS or guidance--on government agencies
(a consequence of OMB implementing the requirements of FISMA, and no
longer allowing ``waivers'').
These conflicting pressures--as well as the challenge of keeping
quality staff--have impacted a number of key areas of work that NIST
collaborates on with the private sector, particular improvements in
conformity assessment.
Third, make sure that NIST's primary customers--agencies of the
Federal Government--are the focus of its efforts through effective
implementation of NIST's mandated responsibilities which include:
Raising awareness of IT risks, vulnerabilities and
protection requirements, particularly for new and emerging
technologies;
Researching, studying, and advising agencies of IT
vulnerabilities and devising techniques for the cost-effective
security and privacy of sensitive federal systems;
Developing standards, metrics, tests and validation
programs:
to promote, measure, and validate security in
systems and services
to educate consumers and
to establish minimum security requirements for
federal systems
Developing guidance to increase secure IT planning,
implementation, management and operation.
Fourth, work with the private sector and the leadership of the
Department of Commerce and other agencies of the Federal Government in
taking on the global challenge of other governments' stringent
cybersecurity regimes. We were very pleased to see the recognition in
the 60-day Review that it will be essential to partner effectively with
the international community. We are seeing efforts in several
countries--China, Russia, India, just to name a few--to impose
stringent, potentially trade-restrictive frameworks that require
mandatory evaluation of U.S. IT products against locally developed,
indigenous information security standards. This is not only bad
security practice; it is potentially adverse to our nation's technology
base and economic security.
As we have worked to roll back these regimes, the U.S. Government
has been a critical partner. NIST, in particular, has played an
essential role based on its status as a world class laboratory that is
respected for its independent assessments and solid work. There is no
other entity like NIST anywhere in the world. When we engage other
governments, the officials sitting on the other side are almost
entirely from their defense, intelligence and national security
operations.
In closing, Mr. Chairman, I reiterate the need for an engaged and
prepared Department of Commerce in taking up the challenge of our
nation's cybersecurity strategy, and playing a key role in the
direction of the 60-day Review. NIST is essential to that role, and the
recommendations and questions we have posed here chart what we believe
is a path for a renewed and reinvigorated cyber and information
security function of NIST. We also note that, in the few short months
since Secretary Locke has taken over the leadership of the Department,
we are seeing a more focused and engaged team at the top levels of the
Department. This is a very positive development which we commend and
look forward to working with.
Again, thank you for the opportunity to appear today. I will be
glad to take any questions from the Committee.
APPENDIX A
In the context of NIST's overall mission and its
existing paradigm for research, what is the most effective way
to ensure that the CSD is able to carry out its mission and
work collaboratively with the private sector to achieve its
goals?
What is the process for developing a strategic plan
for CSD to carry out its mission?
Is the current budgetary process for CSD--which
relies on appropriate monies, but also requires each group
within CSD to contract for specific monies with particular
agencies--consistent with CSD's mission and consistent
execution of long-term programs?
In a highly competitive environment for skilled
talent in this area, how is NIST supporting the CSD in this
regard and what can be done to both attract and keep these
individuals to the CSD?
The Cybersecurity Research & Development Act included
a number of ``grand challenges.'' How has NIST/CSD responded
and what can be done to enhance the capacity of the agency to
carry out these challenges?
What has been the experience with the National
Infrastructure Assurance Program (NIAP) and should NIST
continue to have a key role in its implementation?
With the Common Criteria now a broadly accepted basis
for conformity assessment, how is the CSD looking to ensure its
continued effectiveness and relevance to the dynamic challenges
of combating information security?
How is NIST preparing to support, working with the
private sector, the development of the next generation of
Common Criteria arrangements, including improvements in the
development of protection profiles?
Has the Special 800 series been effective in
providing guidance, and how can the process be updated and
improved? How is NIST working to avoid inappropriate use of the
Special 800 series which are now being used as legal standards
imposed on private sector companies when they were never
designed to be used in that way?
With the adoption of data encryption playing a larger
role in data security, is NIST's FIPS 140-2 validation program
effective at ensuring timely and effective evaluations? Does
the program encourage use of validation?
There are several efforts to redefine what are
``national security'' and ``non-national security systems.''
How does this discussion affect NIST's role and what are can be
done to avoid unnecessary duplication and complexity?
How can the work of the CSD in implementing FISMA be
highlighted and reinforced and how can its role be made more
effective?
Biography for Mark Bohannon
Mark Bohannon is the General Counsel and Senior Vice President
Public Policy for the Software & Information Industry Association
(SIIA).
As the principal trade association of the software code and
information content industry, the more than 500 members of the Software
& Information Industry Association (SIIA) develop and market software
and electronic content for business, education, consumers and the
Internet. SIIA's members are software companies, e-businesses, and
information service companies, as well as many electronic commerce
companies. Its membership consists of some of the largest and oldest
technology enterprises in the world as well as many smaller and newer
companies.
Mr. Bohannon is responsible for the legal and public policy agenda
of SIIA. Mr. Bohannon's experience includes engagement with hundreds of
companies developing online services for the business, consumer and
government markets.
Prior to joining SIIA, Mr. Bohannon was a senior official of the
U.S. Department of Commerce where he served as Chief Counsel for
Technology and Counselor to the Under Secretary for Technology. During
his tenure, his responsibilities included a number of technology policy
initiatives, fighting against technical barriers to trade, and
promoting effective e-Commerce, intellectual property and Internet
policies. Mr. Bohannon also served on numerous USG delegations to
bilateral talks and negotiations with major trading partners and
multilateral fora such as the Hague, OECD, UNCITRAL and WTO.
A native of Austin, Texas, Mr. Bohannon is a graduate of the Edmund
A. Walsh School of Foreign Service at Georgetown University and of the
George Washington University Law School in Washington, D.C.
Discussion
Chairman Wu. Thank you very much, Mr. Bohannon, and now it
is in order for the panel to ask questions, and the Chair
recognizes himself for five minutes. I hope to be able to
address both the international cooperation issues and also the
reorganization issues in five minutes, but it may stretch out a
little bit.
Mr. Bohannon, you cited the Chinese incidents, and I would
like to expand on that a little bit, and also I would like to
ask the rest of the panel about the appropriate role that you
see for NIST playing in development of international standards,
what has gone well and what can be improved going forward. Mr.
Bohannon, would you care to go first?
Mr. Bohannon. Sure. The developments in China are complex.
Let me try to summarize them the best I can.
Starting several years ago, China began to develop a series
of standards for evaluating IT products in a wide range of
areas. Not surprisingly they include areas in which the U.S. IT
industry is dominant or has very superior products to the rest
of the world. Those standards are based in large part on
indigenous standards that were developed by indigenous
standards organizations without really input from anyone
outside of China, to be honest with you. They would require
evaluation of products through laboratories that are at best
loosely associated with the Chinese government.
These kinds of evaluations are very sensitive. NIST has
handled these issues very carefully and has a long-time history
of working in this area. Those Chinese standards would have
prevented market access for many IT products. Working with the
Secretary of Commerce and USTR, we have been able to roll them
back. They are still quite broad in scope, however, and we are
looking forward to continuing to work with the United States
Government.
But the implications were that China would develop very
indigenous, very unique standards for security in products that
are widely used and that would be detrimental not only to the
security of China in terms of its practices but also our
technology base and our innovation base.
Chairman Wu. Mr. Bohannon, I think we are familiar with the
scope of the problem, but if there are any further comments you
want to make about NIST's role and why that was important.
Mr. Bohannon. Sure. NIST's role--they were very important
because the Chinese see NIST as a truly world-class laboratory
from which NIST can provide an independent view about things.
Its role in developing advanced encryption standard is well
known, but that is only the tip of the iceberg. It is not only
China but other governments see NIST as a place where it can go
for unbiased, professional independent assessments of what are
good security practices and how they can be implemented in a
meaningful way.
And so we were very pleased when the NIST team were willing
to have late-night videoconferences with their counterparts
explaining to them why the United States Government doesn't do
things like ask for source code or why other governments don't
ask for source code. That was a very important message. It was
a different message than could come from the trade route with
the Secretary of Commerce. It came from the best in world-class
experts in this area to explain why that is not good security
practice based on global norms. And those are making a big
difference.
Chairman Wu. Thank you very much, Mr. Bohannon. Would any
of the other witnesses like to comment on what has been
occurring well or not well and what could be improved going
forward on international standard setting?
Dr. Landau. I would just like to say, and this is an old
example but it contrasts with a previous failed example or an
example that was not so successful. The advanced encryption
standard was done extremely openly, extremely transparently.
Not only were the submissions open but in fact, the comments on
the proposed specifications were given out, that is, the
proposed specs were put out and NIST asked for comments, and
then the proposed specs were changed in accordance with
comments it received internationally. The result was a very
open competition, and when the standard was chosen, and it was
a standard designed by two Belgians, the acceptance was
immediate internationally which created a much better situation
for industry, it created a much better situation for security,
and it created a situation in which the United States
Government has approved the use of the advanced encryption
standard for top-secret implementations. The NSA has approved
of it, and I think it is a tremendous success and it has to do
with the transparency of the process, the scientific integrity
with which it was carried out.
Chairman Wu. Mr. Starnes.
Mr. Starnes. Thank you. A couple things on the
international standards front that relate to this discussion
about brand, the primary brand at NIST is NIST, and it is a
significant brand. And as NIST doesn't have regulatory
authority, they are very good in these kinds of technology and
standards discussions across border.
There is a movement, a broad movement that we are watching
carefully to this notion of trusted platform. How do I know
that this device is trusted? That involves both hardware and
software systems. So there is deep concern that our definition
of trust is not consistent. We can describe technical trust,
but social trust is a little bit more interpretive. So there is
work being done against some particular elements, fundamental
elements in the platform, things called Trusted Platform
Modules (TPMs) where China, for example, would like to build
their own based on their definition of trust. And I think we do
have to look carefully at some of these formative issues, and
NIST can play an extremely important role in creating an
adopted international standard at the core basis of the
evolution of this trusted platform movement.
Chairman Wu. Thank you very much, Mr. Starnes, and my time
is expired, but Dr. Schneck, perhaps we can come back in the
next round. Mr. Smith, five minutes.
Mr. Smith. Thank you, Mr. Chairman. Dr. Schneider, you
stated in your testimony the need to revise FISMA. Could you
elaborate? What do you see as the problems and what you believe
should be done about them especially as it relates to NIST?
Dr. Schneider. Yes, thank you. So I suppose I am an
outsider. I am not working for a federal agency, and therefore
I don't have to follow FISMA guidelines periodically to
establish the security of my computing systems.
But I have heard people who do this in my capacity on the
ISPAB, and it strikes me as a very expensive madness, an annual
ritual where IT managers have to compile an enormous amount of
paper certifying a number of things that is only loosely
correlated with the security of their systems.
When the Federal Government didn't require our agency
computer systems to be very secure, there was much distance to
cover, and the sort of initial inventorying that FISMA reviews
are about were a very good way to get started. We are now way
down that path, and we understand much better about
vulnerabilities and about how to address them, and the current
FISMA requirements are not about that. They should be
continuous, they should involve monitoring, they should be
focused much more on technical issues and much less on
inventory-style documentation. They should be much less
legalistic exercise between some sort of auditor and agency
management, and I think that lots and lots of resources are
being spent trying to accommodate a set of guidelines without
getting much security gain.
Mr. Smith. Okay. Thank you. Ms. Furlani, you noted in your
testimony about OMB talking about outcome-focused rather than
compliance-focused metrics in cybersecurity. What type of
products do you expect to emerge from that effort and what is
the timeline associated with that?
Ms. Furlani. The effort has just begun, so I am not
prepared--I really don't know a timeline. But there is an
energy assigned to it to try to make the changes as quickly as
possible.
The focus is to understand some of the issues that Fred--
Dr. Schneider has mentioned and how the changes might be
implemented that metrics could be more realistic in today's
environment.
Mr. Smith. Okay. Thank you. Dr. Landau, you mentioned in
your testimony the emerging security needs in the area of cloud
computing. Could you explain exactly what that is and how it is
used on federal computer networks and what unique security
needs accompany it?
Dr. Landau. When you have a system that is in your office
or in your IT center, you own it, you manage it. When the data
is instead held somewhere else on Google documents, or Gmail
should be examples that people tend to be familiar with, then
you are no longer managing the security of your system or your
IT managers are no longer managing the security of your system.
I don't know exactly how the Federal Government is using cloud
computing. I know that NIST has been preparing documents about
security risks and security definitions for cloud computing,
and I would defer to Cita for that.
But you raise a whole set of security risks and a whole set
of policy risks and legal risks when you move to cloud
computing, and those have to be addressed, whether it is in
business, whether it is in government, whether it is in
education. As you shift where the data is being held, what is
the backup policy, who has access to it, what are the legal
policies? If the data is being held in the United States, that
is one thing. Is the data being held in Canada? Is the data
being held in the UK? What is the backup policy? So it is a new
set of security risks that are being introduced.
Mr. Smith. Thank you very much.
Dr. Landau. Sure.
Chairman Wu. Thank you very much, Mr. Smith. Ms. Edwards?
Ms. Edwards. Thank you, Mr. Chairman, and thank you to each
of our witnesses. I am in the 4th Congressional District in
Maryland. We are really proud to be the home of the NIST labs.
I know I have had a chance to visit and meet with all of our
partners, friends in NIST and am incredibly impressed by the
work that is done there, and I appreciate your testimony.
Ms. Furlani, I have a question because I don't quite
understand the argument around concerns raised about
reorganization if there isn't a deep impact on the actual work
that takes place and NIST's responsibilities. And so I wonder
if you have any comments about some of the testimony that you
have heard here today regarding NIST's capacity to take on
these responsibilities and also maintain what I think is a
really high standard for cooperation and work with private
industry and trusted work with private industry in so many
other areas. And why would a reorganization actually impact
that trust that has been well-established?
Ms. Furlani. Thank you for the opportunity to speak. The
energy that was applied to rethinking how we could better use
the resources that are ours to manage to address all the
incoming opportunities to succeed was the driver behind the
proposal--the initial thinking of how we might consider
restructuring to be better prepared to address the future, the
perception that somehow we would be diminishing what we were
already--the great things that are already being accomplished
was misplaced. And so what we were trying to do is make sure
that we could address the new requirements with the resources
that we have and bring the broader perspectives that are
available across the laboratory to that focus.
Ms. Edwards. Thank you, and I have a couple of other
questions that are actually related more to this concern that
the absorption of cybersecurity responsibilities and standard
setting in the Homeland Security, national security arena,
apart from NIST's role--and I wonder if any of the witnesses
have some thoughts about as to the value of maintaining a
somewhat independent standard setting for cybersecurity that
isn't completely folded into a national security framework. I
am thinking about areas like healthcare and, you know, some
things that seem a little bit of a distance from national
security concerns.
Dr. Schneck. Thank you and thank you for the opportunity to
address that. As a McAfee employee and as a citizen with a
background of high-performance computing and actually a founder
of the Georgia Tech Information Security Center, I look at the
development of cybersecurity standards as a collaborative
effort, a necessarily collaborative effort with academia, with
private sector and with NIST's scientific guidance as has been
mentioned by the other panelists. And we look at that because
we are up against an enemy, an international enemy. We are all
connected, and we all face the same threat. And this enemy is
collaborative, and this enemy works fast. So if we were to have
an only-government or a very regulatory standards body for
cybersecurity, you not only stifle the market or innovation as
we have mentioned, but you set back the implementation of
standards of stronger cybersecurity for two to three years, and
by the time we are able to meet those standards in the networks
that keep the lights on, we are three years behind what the
market has developed to do better than the enemy, and we lose
that war.
So I feel strongly that cybersecurity standard setting
needs to be a very collaborative exercise with the private
sector, with academia, with many experts from government with
different agencies and certainly with NIST's scientific
guidance bringing crucial guidance into that process.
Ms. Edwards. And does that mean, in your view does that
mean that the coordination for that has to take place out of
the White House or is there some sort of other interagency
coordination at the federal level with private industry and
academia that should be set up that is at the whims of one
administration's focus or not?
Dr. Schneck. The focus is how we set standards for
cybersecurity, not overall cybersecurity strategy but simply
standards. The view I would put here today and on behalf of BSA
is it is collaborative. It is private sector and academia but
with strong respect for and inclusion of that crucial role that
NIST plays, and the China example is a great point of bringing
the science back into the equation because the science is what
will help us win that war against that threat.
Ms. Edwards. Thank you, Mr. Chairman.
Chairman Wu. Thank you very much. I believe we have just
commenced a series of nine votes, and I think it would be
inhumane to ask the panel to wait that long through the votes,
so it is my intent to move as expeditiously as possible. And we
probably have ten more minutes for questions.
Dr. Schneck, I know that you wanted to make a comment about
international issues, but perhaps we could submit a series of
written questions and look forward to your response.
Dr. Landau, you distinguish between security issues,
identity issues and privacy issues. How does that affect the
framework of security standards that we should be developing?
Dr. Landau. Well, up until now, NIST has focused on the
security standards, and anytime that I was on the ISPAB and we
discussed NIST addressing privacy standards, NIST had stayed
very far away. I am delighted to hear that it is beginning to
move in that direction. I would urge the Committee to give NIST
even more authority to do so because I think there is a crying
need as we see the accumulation of data in private hands and
the need for a good set of standards.
Identity management is a very complicated issue, and we
have seen some fledgling efforts. I think that NIST has a very
good understanding of the difficulties of doing identity
management, and I am sorry that NIST was not pulled more into
the discussions earlier this summer as well. It produced the
levels of assurance document, that is part of the 800-series,
but it wasn't as involved as I think it ought to have been in
the policy implications of making decisions about identity
management systems for different levels of assurance. And that
is a place where I had said earlier I thought that NIST should
be providing more policy guidance and should be somewhat more
independent.
Chairman Wu. So if I am simplifying this incorrectly,
please correct me, that there has been a fair amount of
activity on the privacy side and that more activity is needed--
I am sorry, on the security side and more activity is needed on
the privacy and identity side?
Dr. Landau. More activity is needed on the privacy side and
I would say on the policy side, on the policy side where it is
closely allied to technical issues, and one particular example
of that is the identity management.
Chairman Wu. Very good. And I wanted to ask the panel
whoever wants to respond that, you know, we have been talking
about standards and focused on that. Is there some low-hanging
fruit here if one of the federal entities, NIST or otherwise,
developed better education programs so that people up and down
the food chain, but especially end-users, became more aware of
what they could do. Would that help the overall privacy
assurance security issues, you know, outside of standard
setting?
Dr. Landau. So I would like to say here that while I think
a lot of the Computer Security Division, the one place that I
think it has not handled things well is in outreach and in
particular inability to find the information. If you know the
information is there and you look for it, you can find it. But
if you are not determined, it is somewhat hard to do. And I
would like to see better outreach, better development of its
website, more usable access to information.
Chairman Wu. What kinds of mechanisms could we use to push
that out as opposed to having it as a pool that people reached
into?
Dr. Landau. I know that NIST had a program in which it
advised small business, but it was a very small program. I
mean, there is the answer, it is a very small program. It
doesn't reach very many people. I think NIST should be doing
that work and not the FBI.
Chairman Wu. Would the ag extension or manufacturing
extension sort of mechanism or model apply in this case?
Dr. Landau. I don't know what the ag extension model is
well enough. I am sorry.
Chairman Wu. Mr. Furlani and then Mr. Starnes, we will come
back to you.
Ms. Furlani. Yes, we have been planning and working
historically with our Manufacturing Extension Partnership and
with our new YouTube video. We are hoping to leverage that
capability to get to the small manufacturers. Of course--it is
addressing all small businesses. We were hoping that maybe some
of your newsletters might refer back to our YouTube video and
make it more available to your constituents as well.
Chairman Wu. Mr. Starnes.
Mr. Starnes. Yes, traditionally in information security, we
have, to your observation, Representative Edwards, pulled the
collective knowledge and talent across multiple parts of
industry and government, and I think this is a clear case where
we need to do that again and are doing that. So we have to
differentiate between standards, which NIST is very good at,
and methods and best practice which are putting standards into
action. And there are some very good technical solutions that
are coming, multilaterally as I mentioned, from government that
move us from just certification and accreditation. We spent
$1.31 billion on certification and accreditation last year.
Many of those dollars should be spent with these new tools and
techniques for continuous monitoring of information technology
systems using all of the intelligence of all of our federal
agencies and commercial entities.
Chairman Wu. Thank you very much, Mr. Starnes. My time has
about expired. Mr. Smith, further questions?
Mr. Smith. I think just briefly. Mr. Bohannon, you
emphasized in your testimony that Congress should avoid making
NIST a regulator of private-sector actions. Could you elaborate
and maybe touch on how the government procurement is de facto
an approach to regulation, whether on purpose or not?
Mr. Bohannon. That is a very good question. Obviously
government procurement is where the rubber hits the road, when
it comes to NIST work. The approach with NIST, though, is
consistent with trying to figure out how to walk that fine line
because with a few exceptions, some of which I talk about in my
testimony, on the whole where NIST has developed federal
information processing standards, it has done so in an open,
transparent and collaborative way so that when FIPS (Federal
Information Processing Standard) are in fact referenced for
government use, they are the product on the whole of working
with the stakeholders, the technology providers, and the users
to make sure there is a standard that as much as possible
conforms to general commercial practice. And notably the
government has its needs. Those are taken into account, but
that is a fine line and one that, going back to Congresswoman
Edwards' question, you know, some of us have short memories.
But it was just seven years ago that when the Department of
Homeland Security was going to be created, the proposal was to
move the Computer Security Division to DHS. Thanks to the
leadership of this Committee, on both sides of the aisle, that
did not happen. I think we would be in a very different
situation today if the Computer Security Division had moved. I
think its work on the special 800-series, I think its work on
AES (Advanced Encryption Standard), I think the work where it
needs to go would have been fundamentally different because it
would have come out of an agency that had very specific law
enforcement and regulatory mission stakes, and the credibility
of that work would have been dramatically differentiated.
So Mr. Smith, you are absolutely right. It is a fine line,
but as we pointed out, the way NIST does its business in a
collaborative way means that on the whole, it doesn't always
get it perfect, but on the whole, the results are consistent
with commercial goods, commercial practice, taking into account
stakeholders, and try to reflect the best of what should be in
that standard.
Mr. Smith. Thank you. Thank you, Mr. Chairman.
Chairman Wu. Thank you very much, Mr. Smith. Ms. Edwards?
No further questions. Okay.
Ms. Furlani, there have been numerous suggestions for about
what you might do, what your agency might do and so on. I
wanted to give you an opportunity to respond to any of the
suggestions that you want to respond to, but in particular, I
would very much like you to respond to--I mean, it is not as
simple as should CSD become a laboratory on its own, but that
is--let us reduce it to that simplicity, and could you respond
to that and any other comments, suggestions that you would like
to respond to?
Ms. Furlani. Thank you, Chairman Wu. Certainly we have had
a lot of input, and as I have said, both support and concerns.
We are going to go back to the drawing board essentially and
revisit what might make the best next proposal. The idea of
separating cybersecurity from information technology is
difficult for me to understand because of the intertwined
nature of the two, but the decision of course would be Dr.
Gallagher's, not mine. So we have lots to consider, many from
the panel members which I greatly appreciate and others. And we
have a lot of rethinking--and of course the original goal,
which I want to go back to and make sure that we have the full
input from my staff which is where we all started with just
trying to get the staff's input. We are back to revisiting the
entire setup, and we will hopefully come out with something
that enables us to move forward in the future, meet our new
opportunities and challenges in a much more robust capable way.
Chairman Wu. Thank you very much, Ms. Furlani. And for the
entire panel, we will submit some additional written questions.
But you all have put a lot of work into the prepared written
materials, into preparing for the oral testimony, and some of
you have traveled a decent distance to get here. So at the risk
of shortening up my thank-yous in person at the table, I want
to give each of you who has something that you want to
contribute to this discussion but you haven't had an
opportunity to put that either in your oral testimony or you
haven't been asked that question. Please, at this time, for as
much time as we have, if you want to add that last point, this
is your chance.
Dr. Landau. I would like to just make a brief comment which
I do have in my written testimony about the importance of
usability work in security, and I know that the Computer
Security Division has begin work on this, and I think it is an
important, new direction. I would like to see the Committee
strongly support that work because of course, it increases
security. Thank you.
Chairman Wu. Thank you. Anyone else?
Dr. Schneck. Thank you. One opportunity--we talk a lot
about cybersecurity and the threats and the scariness of it and
the work that we need to do. The issue of awareness was raised
before, and that is a very positive point, and I think there is
a huge opportunity for NIST to work with the National
Cybersecurity Alliance. Part of this is in my written
testimony, but when you visualize that, this group, what they
do is they take the message and they bring it to the street,
from the federal to the State, local, tribal community level
and to homeowners' associations and to schools so that our
youngest citizens all the way on up are learning not just what
to be careful of but how to responsibly build security and
privacy as Dr. Landau has referred to today into their daily
lives and to our use of cyber, because it is going to affect
our entire way of life forward.
Chairman Wu. Thank you very much. Anyone else?
Mr. Starnes. Thank you, Chairman, but I would like to put
just a punctuation mark on my C&A (Certification and
Accreditation) comments of earlier. If we took just 30 percent
of the C&A dollars that were spent in 2008, that would be more
than we spent on cybersecurity research in the entire year. So
I encourage the Committee to focus legislatively on these
processes as well and help government agencies and industry do
zero-based implementation of important new methods around
continuous monitoring.
Chairman Wu. Thank you very much, Mr. Starnes. And since
Mr. Smith and I are at risk of missing some of these votes, Dr.
Schneider, Mr. Bohannon, your indulgence in perhaps providing
us comments and answering other inquiries as we go. I want to
again thank you all very, very much for your testimony, and no
guarantees in life, but I think there is a high probability
that we will try to pitch in with relevant legislation to try
to improve the situation, and we look forward to your comment
on that effort also. So thank you very much. The record will
remain open for two weeks for additional statements from
Members and for answers to follow-up questions. The witnesses
are excused, and the hearing is now adjourned. Thank you.
[Whereupon, at 3:17 p.m., the Subcommittee was adjourned.]
Appendix:
----------
Answers to Post-Hearing Questions
Answers to Post-Hearing Questions
Responses by Cita M. Furlani, Director, Information Technology
Laboratory, National Institute of Standards and Technology
Questions submitted by Chairman David Wu
Q1. What are the current limitations and flaws of FISMA and what parts
of FISMA policy must change to improve the security of federal
information technology systems? What role should NIST play in an
effective FISMA framework?
A1. The Federal Information Security Management Act (FISMA) Act of 2002
tasked the National Institute of Standards and Technology (NIST) with
the responsibility to establish security standards and guidelines for
the Federal Government and charged the Office of Management and Budget
(0MB) with enforcement of FISMA.
NIST developed two standards, Federal Information Processing
Standard (FIPS) 199, Standards for Security Categorization of Federal
Information and Information Systems and FIPS 200, Minimum Security
Requirements for Federal Information and Information Systems and
associated guidelines including Special Publication 800-53, Recommended
Security Controls for Federal Information Systems and Organizations to
provide a foundation for federal agency security.
Since FISMA's release, agencies' security capabilities have been
maturing, and it is time to shift the focus from compliance to
improving the implementation of their enterprise security. The existing
NIST work in developing standards and guidelines and in creating tools
for monitoring the status of security settings enables this shift. NIST
is prepared to develop additional security automation tools to further
optimize system security configurations and report status of system
components. NIST is poised to ensure its standards and guidelines
address new security technologies that can be used to mitigate the ever
changing threat environment. In addition, NIST is working with 0MB and
others to develop security metrics that will better quantify the
improvements that agencies make to their security implementations and
provide more robust methods for assessment of agencies' security
posture.
Answers to Post-Hearing Questions
Responses by Susan Landau, Distinguished Engineer, Sun Microsystems,
Burlington, MA
Questions submitted by Chairman David Wu
Q1. What are the current limitations and flaws of FISMA and what parts
of FISMA policy must change to improve the security of federal
information technology systems? What role should NIST play in an
effective FISMA framework?
A1. I am speaking from my experience on ISPAB; since I retired from the
board in January 2008, this information is a bit dated. I have just
three points to make.
In its early years, FISMA increased security
awareness. However, after multiple times of agencies filling in
the FISMA reports, it appears--at least from the outside--that
FISMA has become more of an exercise in paperwork than a schema
for enforcing good security practices.
The problem is incentives and this is not a NIST
issue, but a Federal Government one. Unless the cost for
failure to have a good security posture and a good recovery
plan is high, it is difficult to incentivize the agencies to
treat cybersecurity with the appropriate attention.
Backup and disaster recovery are two issues not
covered by FISMA; they should be part of any cybersecurity plan
(and continuity of operations should be updated with each
technology enhancement).
I hope this is useful to you.
Answers to Post-Hearing Questions
Responses by Phyllis Schneck, Vice President, Threat Intelligence,
McAfee Corporation
Questions submitted by Chairman David Wu
Q1. What are the current limitations and flaws of FISMA and what parts
of FISMA policy must change to improve the security of federal
information technology systems? What role should NIST play in an
effective FISMA framework?
A1. We believe Congress needs to reform FISMA, to close the gap between
compliance and security. Congress needs to legislate to empower
officials in charge of the security of agencies' computer networks:
* First, they need authority to actually enforce security
requirements over their agencies' networks and systems. It would be
appropriate for OMB to develop some additional incentives to push
agencies to comply with their FISMA requirements, including having some
percentage of cyber budgets of agencies withheld in the event that they
do not show good progress toward meeting their compliance obligations.
Alternatively, agencies could be rewarded with larger budget growth
rates for their cyber security programs when they show actual progress
in improving the security postures of their operations.
* Second, they need the technical and human resources necessary to
perform these tasks, such as network monitoring and automated security
policy compliance monitoring and enforcement capabilities. This in
particular is where NIST efforts will be most needed. Network
monitoring and automated security policy compliance monitoring should
be done across the government on the basis of common standards. This
would allow a government-wide security center to have a consistent view
of federal networks' security.
We also need the legislation to ensure these officials are
accountable for identifying and addressing the threats and
vulnerabilities that their networks actually face. We can do this in
particular by having ``red teams'' test the effectiveness of the
security measures in place against real-life attacks, and by having
this serve as a feedback loop that leads to system and network security
improvements.
Answers to Post-Hearing Questions
Responses by William Wyatt Starnes, Founder, CEO, and President,
SignaCert, Inc.; Founder, Tripwire, Inc.
Questions submitted by Chairman David Wu
Q1. What are the current limitations and flaws of FISMA and what parts
of FISMA policy must change to improve the security of federal
information technology systems? What role should NIST play in an
effective FISMA framework?
A1. FISMA is a broad methodology that seeks to normalize the IT
compliance and reporting for Federal IT infrastructure. Generally the
method encourages ``periodic testing'' of IT devices and infrastructure
against a range of configuration, vulnerability and usage best
practices.
There are several problems with this approach.
Specifically, it is questionable whether the FISMA
report card actually properly and correctly reflects the actual
security, compliance and readiness of the Civilian Agency
reporting the results.
Also, FISMA largely is viewed as a Certification and
Accreditation (C&A) process, and the C&A processes are ``point
in time'' current state of the IT devices. This ``IT Audit''
mentality:
creates periodic ``peak load'' human resource drain
by the Agencies to do the FISMA reporting
generates tremendous paperwork, much of which goes
largely unused past the summary reporting for the
actual roll-up to the OMB
as it is point in time and periodic, there is large
time gap between the audits where a actual and
problematic security and compliance issues can emerge,
causing increased risk and disruption.
There is a lot of emerging consensus that we should change or
update the methodologies and technologies used for FISMA, as well as
changing driving legislation, to deal with the aforementioned risks and
weaknesses.
NIST/NSA/DHS and others have been in close collaboration for
several years on best practices, method and technologies that address
many of the FISMA gaps. Called the Security Content Automation
Protocol, or SCAP, these methods are very well suited to all IT
management needs, and we recommend that the full extended version of
SCAP be establish as the technical pillar for FISMA 2.0 usage.
In concert, we strongly urge Congress, through all of the
committees activities, to begin to shift ``C&A'' requirements (and the
thus the dollars allocated for C&A), to SCAP CONTINUOUS MONITORING.
IT compliance, done well, should be largely transparent to the
users and even the IT staff. Good systems hygiene should be complete,
intrinsic and continuous, not just scorecard driven period process.
Legislated FISMA processes should fully embrace this concept.
With the SCAP framework, the following key IT issues can be
continuously addressed in a consistent form at all Agencies, and
ultimately across the broader DOD and even commercial IT enterprise:
1. It devices are configured with the right software
components (including supply chain provenance) not only at the
time and point of deployment, but actively and continuously
across their usage lifetime.
2. The deployed software can be configured correctly at point
of deployment, and maintained in the correct, secure and most
stable configuration throughout their usage lifetime.
3. The presence of vulnerabilities can be actively tested and
validated in a consistent and complete way across the entire IT
infrastructure on an active, real time and continuous basis.
See the diagram below for a simple view of the core SCAP test
areas.
Another benefit of these methods is that we can use the SCAP
protocol to aggregate and automate ``best practices'' knowledge against
all three of the areas above so that IT operational readiness (AS A
NATION) gets better based on the collective knowledge and experience of
the best IT expertise that we have, and we can immediately apply that
knowledge--reducing our cyber vulnerabilities across all industry
sectors.
Answers to Post-Hearing Questions
Responses by Fred B. Schneider, Samuel B. Eckert Professor of Computer
Science, Cornell University
Questions submitted by Chairman David Wu
Q1. What are the current limitations and flaws of FISMA and what parts
of FISMA policy must change to improve the security of federal
information technology systems? What role should NIST play in an
effective FISMA framework?
A1. I do not have direct experience with FISMA and I have not read the
legislation. But I am a member of NIST's Information Security and
Privacy Advisory Board (ISPAB), and our board has heard a good deal
from officers at civilian Federal agencies that must comply with FISMA
as well as from the Inspector General (IG) community, which is
responsible for auditing FISMA compliance. These comments are based on
what I have heard from those communities.
I believe that we should strive to have FISMA compliance for an
agency mean that the agency's computing systems are secure enough,
given the tasks they perform, the data they store, and the information
and services they can access. I fear that the way FISMA is interpreted
today does not succeed at this.
FISMA compliance should embody a philosophy of risk management
rather than one of absolute security. Risk management requires
understanding the consequences of system compromise, including loss of
functionality, ex-filtration of confidential data, corruption of
information, and even possible use by an attacker as a stepping-stone
to other systems. This is multi-dimensional and, therefore, attempting
a simple categorization of all systems within an agency or across
agencies is unlikely to be useful. Only with richer kinds
characterizations, can we portray system weaknesses in a sufficiently
useful way for decision-makers. And only richer characterizations will
incentivize corrective measures that address the real problems in
context (as opposed to incentivizing measures that merely sound
impressive on paper).
There needs to be a strong coupling between FISMA compliance and
security of a system in its deployed context. Today that coupling is
weak. A system that has been deemed compliant today might still be easy
to attack; a system that today implements sufficient defenses for its
role will not necessarily be deemed FISMA compliant.
This disconnect between FISMA compliance and real security partly
results from
an absence of good metrics for security,
FISMA compliance being dominated by documenting
defenses rather than by exercising them, and
FISMA compliance being seen as a periodic obligation
discharged by negotiating with an auditor rather than a
continuous one concerned with eliminating system
vulnerabilities as they become known.
The first of these--the absence of metrics--is an open research
question; the other two are inherent in the way FISMA compliance is
interpreted and evaluated.
NIST is an obvious place to undertake research in security metrics.
That said, I am doubtful that anyone will ever devise a way to measure
whether a system is secure (because security is relative to attacks,
and new attacks are being discovered every day). But it does seem
reasonable to expect better ways than practiced today for evaluating a
system and ascertaining whether it is secure against some set of known
attacks. And NIST is a reasonable place to develop and codify as
metrics these better ways; FISMA compliance assessments should adopt
such improved metrics as they become available.
NIST has in the past done a good job of developing and documenting
security best practices for civilian Government agency computing
systems. Best practices bring good security, so we should want NIST to
continue that work. And a security evaluation of a system for FISMA
compliance should ascertain whether current best practices are being
followed. I would urge, though, that ``best practices'' be expanded to
include the obligation that a system is checked against lists of known
vulnerabilities. That is, we need to check that certain desirable
features and processes are present but also check that undesirable ones
are absent.
Finally, FISMA compliance needs to require more than documenting
what a system is. We don't evaluate the efficacy of a weapons system or
a military unit only by evaluating metrics--we run exercises in the
field and force engagement with realistically simulated attackers.
FISMA compliance needs to adopt that approach for our computing
systems. Some of this can be accomplished with existing automated
tools, but some will require building new tools. We should also
contemplate requiring periodic random surprise red-team attacks and
simulated natural disasters, because this evaluates system
trustworthiness in a way that incentivizes continuous readiness. The
key point is to promote the view that system defense a continuous
obligation and is results-oriented, rather than being documentation-
oriented. Documentation is a useful basis for determining
accountability after a system is found wanting, but documentation does
little to defend against attacks.
Answers to Post-Hearing Questions
Responses by Mark Bohannon, General Counsel and Senior Vice President
for Public Policy, Software & Information Industry Association
(SIIA)
Questions submitted by Chairman David Wu
Q1. What are the current limitations and flaws of FISMA and what parts
of FISMA policy must change to improve the security of federal
information technology systems? What role should NIST play in an
effective FISMA framework?
A1. As the Committee is well aware, the Federal Information Security
Management Act (FISMA), enacted in 2002, sets forth a comprehensive
framework to ensure the effectiveness of security controls over
information resources that support federal operations and assets. FISMA
assigns specific responsibilities to federal agencies, the Office of
Management and Budget (OMB), and the National Institute of Standards
and Technology (NIST). It also requires agencies and OMB to annually
report on the adequacy and effectiveness of agency information security
programs and compliance with the provisions of the Act. To help meet
these requirements, OMB established a uniform set of information
security measures that all federal agencies report on annually. NIST
produces important guidance and publications related to FISMA
implementation.
In reviewing the current limitations and flaws of FISMA, recent
investigations by the General Accounting Office (GAO) are useful
inputs. As the GAO has stated,\1\ leading organizations and experts
have identified different types of measures that are useful in helping
to achieve information security goals. While it found that officials
categorized these types using varying terminology, GAO concluded that
they generally fell into three types: (1) compliance, (2) control
effectiveness, and (3) program impact. These types are consistent with
those laid out by NIST in its information security performance
measurement guide.\2\ The GAO found that, while information security
measures can be grouped into these three major types, organizations and
experts reported that all such measures generally have certain key
characteristics, or attributes. These attributes include being (1)
measurable, (2) meaningful, (3) repeatable and consistent, and (4)
actionable.\3\
---------------------------------------------------------------------------
\1\ See GAO report number GAO-10-159T, entitled `Information
Security: Concerted Effort Needed to Improve Federal Performance
Measures' which was released on October 29, 2009.
\2\ National Institute of Standards and Technology, Performance
Measurement Guide for Information Security, NIST Special Pub. 800-55
Revision 1 (Gaithersburg, Md.: July 2008).
\3\ Although we focused on identifying attributes and practices for
measuring the performance of information security programs, our
findings conformed closely to our prior work on effective performance
measurement and reporting practices for the Federal Government in
general. See, for example, GAO, Managing for Results: Enhancing Agency
Use of Performance Information for Management Decision Making,
available at www.gao.gov/cgi-bin/getrpt?GAO-05-927, Sept. 9, 2005.
---------------------------------------------------------------------------
Using this framework, GAO determined--and we concur--``that federal
agencies have not always followed key practices identified by leading
organizations for developing information security performance measures.
While agencies have developed measures that fall into each of the three
major types (i.e., compliance, control effectiveness, and program
impact), on balance they have relied primarily on compliance measures,
which have a limited ability to gauge program effectiveness. Agencies
stated that, for the most part, they predominantly collected measures
of compliance because they were focused on measures associated with
OMB's FISMA reporting requirements. In addition, while most agencies
have developed some measures that include the four key attributes
identified by leading organizations and experts, these attributes were
not always present in all agency measures. Further, agencies have not
always followed key practices in developing measures, such as focusing
on risks.''
GAO focused on the inadequacies of OMB's measures which ``did not
address the effectiveness of several key areas of information security
controls, including, for example, agency security control testing and
evaluation processes. There is no measure of the quality of agencies'
test and evaluation processes or results that demonstrate the
effectiveness of the controls that were evaluated.''
As a starting point, the most recent five recommendations GAO made
to OMB to assist federal agencies in developing and using measures that
better address the effectiveness of their information security programs
are worth considering:
``issue revised guidance to chief information
officers for developing measures,'' which we would add should
follow and build on the relevant work and publications produced
by NIST;
``direct chief information officers to ensure that
measures exhibit key attributes'';
``direct chief information officers to employ the key
practices for developing a measure as identified by leading
organizations,'' again taking into account the work and
publications produced by NIST;
``revise annual FISMA reporting guidance to
agencies''; and:
``revise the annual FISMA report to Congress to
provide better status information on the security posture of
the Federal Government.''
In addition, we would note that implementation of FISMA, with the
continued leadership of NIST working with OMB, would benefit from:
Requiring that federal agency CIOs and CISOs are
appropriately positioned within their agencies management
structure to promote ``top down'' priority of information
security.
Agencies sometimes use FISMA compliance as an excuse
to reject innovations simply because they are new and not
explicitly reflected in the FISMA checklists. FISMA should
actively encourage government agencies to be more open to
deploying cutting edge solutions.
Audit and oversight methods should be harmonized to
the greatest degree possible using NIST work and publications.
There also needs to be work to establish consistency in IG
examinations, recognizing that IG offices are not necessarily
staffed with requisite skill sets.
Agencies should conduct at least annual risk
assessments that incorporate classified information and input
from the private sector. Those risk assessments should also
incorporate the work and outcome of NIST as well as other
sources, including the Department of Homeland Security's US-
CERT.