[House Hearing, 112 Congress]
[From the U.S. Government Publishing Office]



 
                    OVERSIGHT OF THE NETWORKING AND
        INFORMATION TECHNOLOGY RESEARCH AND DEVELOPMENT PROGRAM
                     AND PRIORITIES FOR THE FUTURE

=======================================================================

                                HEARING

                               BEFORE THE

             SUBCOMMITTEE ON RESEARCH AND SCIENCE EDUCATION

              COMMITTEE ON SCIENCE, SPACE, AND TECHNOLOGY
                        HOUSE OF REPRESENTATIVES

                      ONE HUNDRED TWELFTH CONGRESS

                             FIRST SESSION

                               __________

                     WEDNESDAY, SEPTEMBER 21, 2011

                               __________

                           Serial No. 112-37

                               __________

 Printed for the use of the Committee on Science, Space, and Technology


       Available via the World Wide Web: http://science.house.gov



                  U.S. GOVERNMENT PRINTING OFFICE
68-318                    WASHINGTON : 2012
-----------------------------------------------------------------------
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, SPACE, AND TECHNOLOGY

                    HON. RALPH M. HALL, Texas, Chair
F. JAMES SENSENBRENNER, JR.,         EDDIE BERNICE JOHNSON, Texas
    Wisconsin                        JERRY F. COSTELLO, Illinois
LAMAR S. SMITH, Texas                LYNN C. WOOLSEY, California
DANA ROHRABACHER, California         ZOE LOFGREN, California
ROSCOE G. BARTLETT, Maryland         BRAD MILLER, North Carolina
FRANK D. LUCAS, Oklahoma             DANIEL LIPINSKI, Illinois
JUDY BIGGERT, Illinois               GABRIELLE GIFFORDS, Arizona
W. TODD AKIN, Missouri               DONNA F. EDWARDS, Maryland
RANDY NEUGEBAUER, Texas              MARCIA L. FUDGE, Ohio
MICHAEL T. McCAUL, Texas             BEN R. LUJAN, New Mexico
PAUL C. BROUN, Georgia               PAUL D. TONKO, New York
SANDY ADAMS, Florida                 JERRY McNERNEY, California
BENJAMIN QUAYLE, Arizona             JOHN P. SARBANES, Maryland
CHARLES J. ``CHUCK'' FLEISCHMANN,    TERRI A. SEWELL, Alabama
    Tennessee                        FREDERICA S. WILSON, Florida
E. SCOTT RIGELL, Virginia            HANSEN CLARKE, Michigan
STEVEN M. PALAZZO, Mississippi
MO BROOKS, Alabama
ANDY HARRIS, Maryland
RANDY HULTGREN, Illinois
CHIP CRAVAACK, Minnesota
LARRY BUCSHON, Indiana
DAN BENISHEK, Michigan
VACANCY
                                 ------                                

             Subcommittee on Research and Science Education

                     HON. MO BROOKS, Alabama, Chair
ROSCOE G. BARTLETT, Maryland         DANIEL LIPINSKI, Illinois
BENJAMIN QUAYLE, Arizona             HANSEN CLARKE, Michigan
STEVEN M. PALAZZO, Mississippi       PAUL D. TONKO, New York
ANDY HARRIS, Maryland                JOHN P. SARBANES, Maryland
RANDY HULTGREN, Illinois             TERRI A. SEWELL, Alabama
LARRY BUCSHON, Indiana                   
DAN BENISHEK, Michigan                   
RALPH M. HALL, Texas                 EDDIE BERNICE JOHNSON, Texas
                            C O N T E N T S

                     Wednesday, September 21, 2011

                                                                   Page
Witness List.....................................................     2

Hearing Charter..................................................     3

                           Opening Statements

Statement by Representative Mo Brooks, Chairman, Subcommittee on 
  Research and Science Education, Committee on Science, Space, 
  and Technology, U.S. House of Representatives..................     9
    Written Statement............................................    10

Statement by Representative Daniel Lipinski, Ranking Minority 
  Member, Subcommittee on Research and Science Education, 
  Committee on Science, Space, and Technology, U.S. House of 
  Representatives................................................    10
    Written Statement............................................    12

                               Witnesses:

George Strawn, Director, National Coordination Office, Networking 
  and Information Technology Research and Development (NITRD) 
  Program
    Oral Statement...............................................    13
    Written Statement............................................    15

Edward Lazowska, Director eScience Institute/Bill and Melinda 
  Gates Chair, University of Washington
    Oral Statement...............................................    27
    Written Statement............................................    28

Robert Sproull, Director of Oracle Labs, retired
    Oral Statement...............................................    33
    Written Statement............................................    35

Robert Schnabel, Dean, School of Informatics, Indiana University
    Oral Statement...............................................    41
    Written Statement............................................    42

Discussion
  ...............................................................    47

              Appendix: Answers to Post-Hearing Questions

George Strawn, Director, National Coordination Office, Networking 
  and Information Technology Research and Development (NITRD) 
  Program........................................................    60

Edward Lazowska, Director eScience Institute/Bill and Melinda 
  Gates Chair, University of Washington..........................    66

Robert Sproull, Director of Oracle Labs, retired.................    70

Robert Schnabel, Dean, School of Informatics, Indiana University.    72


                    OVERSIGHT OF THE NETWORKING AND


        INFORMATION TECHNOLOGY RESEARCH AND DEVELOPMENT PROGRAM


                     AND PRIORITIES FOR THE FUTURE

                              ----------                              


                     WEDNESDAY, SEPTEMBER 21, 2011

                  House of Representatives,
    Subcommittee on Research and Science Education,
               Committee on Science, Space, and Technology,
                                                    Washington, DC.

    The Subcommittee met, pursuant to call, at 2:04 p.m., in 
Room 2318 of the Rayburn House Office Building, Hon. Mo Brooks 
[Chairman of the Subcommittee] presiding.


                            hearing charter

              COMMITTEE ON SCIENCE, SPACE, AND TECHNOLOGY

             SUBCOMMITTEE ON RESEARCH AND SCIENCE EDUCATION

                     U.S. HOUSE OF REPRESENTATIVES

  Oversight of the Networking and Information Technology Research and 
           Development Program and Priorities for the Future

                     wednesday, september 21, 2011

                          2:00 p.m.--4:00 p.m.

                   2318 rayburn house office building

Purpose

    On Wednesday, September 21, 2011, at 2:00 p.m. the Subcommittee on 
Research and Science Education will hold a hearing to review the 
networking and information technology research and development (NITRD) 
program to ensure U.S. leadership in networking and information 
technology and to discuss priorities for the future.

Witnesses

      Dr. George Strawn, Director, National Coordination 
Office, Networking and Information Technology Research and Development 
(NITRD) Program

      Dr. Edward Lazowska, Bill & Melinda Gates Chair in 
Computer Science & Engineering, University of Washington

      Dr. Robert Sproull, Director of Oracle Labs, retired

      Dr. Robert Schnabel, Dean, School of Informatics, Indiana 
University

Overview

      Advances in networking and information technology (NIT) 
continue to transform the world in which we live. We increasingly rely 
on the systems, tools, and services of this ever-growing and ever-
changing domain. It is not only as a matter of convenience in our daily 
lives, but critical to our future economic prosperity, health, and 
security.

      The Networking and Information Technology Research and 
Development (NITRD) Program is the federal government's mechanism for 
coordinating the Nation's unclassified NIT research and development 
(R&D) investments. NITRD's formal membership consists of 14 federal 
agencies while many additional agencies participate in program 
activities.

      NITRD was originally authorized in the High-Performance 
Computing Act of 1991 to help coordinate ongoing high-performance 
computing programs throughout the federal government. The Act was 
amended in 1998 and 2007. In the 111th Congress, the U.S. House of 
Representatives passed the Networking and Information and Technology 
Research and Development Reauthorization Act twice. The Senate did not 
take up H.R. 2020 and removed the language in the 2010 America COMPETES 
Reauthorization Act.

      As required by law, in December 2010, the President's 
Council of Advisors on Science and Technology (PCAST) released its 
report, Designing a Digital Future: Federally Funded Research and 
Development in Networking and Information Technology. The report finds 
that ``NITRD is well coordinated and that the U.S. computing research 
community, coupled with a vibrant NIT industry, has made seminal 
discoveries and advanced new technologies that are helping to meet many 
societal challenges,'' but also notes the need for more accurate 
accounting and additional investments in basic research. \1\
---------------------------------------------------------------------------
    \1\  President's Council of Advisors on Science and Technology, 
Report to the President and Congress December 2010,Designing a Digital 
Future: Federally Funded Research and Develpment in Networking and 
Information Technology, p.v.

      The Fiscal Year 2012 (FY 12) budget request for agency 
programs captured under NITRD is $3.9 billion, roughly $200 million 
---------------------------------------------------------------------------
more than the FY10 actual amount.

Background

    Federal support for research and development in networking and 
information technology (NIT) originally stemmed from an interest in and 
the challenge of developing computers capable of addressing complex 
problems, primarily those focused on national security and global 
competition. Now, several decades after the dawn of the digital 
revolution, NIT encompasses a broad array of technologies from smart 
phones to digital libraries and cloud computing. Having changed the way 
we listen to music, drive our cars, and communicate with each other, 
this ever-growing field has led to the creation of many of the 
technologies and systems we rely on daily.
    Additionally, research and development (R&D) in NIT provides a 
greater understanding of how to protect essential systems and networks, 
systems and networks that support fundamental sectors of our economy, 
from emergency communications and power grids to air-traffic control 
networks and national defense systems. NIT R&D works to prevent or 
minimize disruptions to critical information infrastructure, to protect 
public and private services, to detect and respond to threats while 
mitigating the severity of and assisting in the recovery from those 
threats, in an effort to support a more stable and secure Nation.

Networking and Information Technology Research and Development Program 
        (NITRD)

    The Networking and Information Technology Research and Development 
(NITRD) program is the main federal R&D investment portfolio in 
networking, computing, software, cybersecurity, and related information 
technologies. NITRD coordinates this unclassified R&D across 14 federal 
agencies (see Table 1).





    The NITRD program has played a role in several important 
technological advances including the computational decoding of the 
human genome; modeling and simulation of complex physical systems 
(aircraft, automobiles, power grids, and pharmaceuticals); unmanned 
aerial vehicles, search-and-rescue robots; and computer-based education 
and training.
    The Subcommittee on NITRD of the National Science and Technology 
Council (NSTC) is the internal deliberative organization for NITRD 
policy, program, and budget guidance. The NITRD Subcommittee includes 
representatives from each participating agency, as well as the Office 
of Management and Budget (OMB). The Subcommittee coordinates the 
planning, budgeting, implementation, and reviews of NIT R&D across the 
NITRD member agencies to help assure continued U.S. leadership, satisfy 
the needs of the federal government for advanced IT capabilities, and 
accelerate development and deployment of new technologies. \4\
---------------------------------------------------------------------------
    \4\  About the Subcommittee on Networking and Information 
Technology Research and Development (NITRD Subcommittee), http://
www.nitrd.gov/subcommittee/program.aspx.
---------------------------------------------------------------------------
    NITRD research activities are organized in eight Program Component 
Areas (PCAs). The PCAs also align the NITRD program budget categories. 
The eight PCAs include: Cybersecurity Information Assurance (CSIA); 
Human Computer Interaction and Information Management (HCI & IM); High 
Confidence Software and Systems (HCSS); High End Computing 
Infrastructure and Applications (HEC I&A); High End Computing Research 
and Development (HEC R&D); Large Scale Networking (LSN); Software 
Design and Productivity (SDP); and Social, Economic, and Workforce 
Implications of IT and IT Workforce Development (SEW). \5\ However, 
NITRD research areas and activities shift regularly as the NIT field 
creates and develops new R&D challenges.
---------------------------------------------------------------------------
    \5\  NITRD Program PCA Definitions, http://www.nitrd.gov/
subcommittee/pca-definitions.aspx.
---------------------------------------------------------------------------
    The NITRD National Coordination Office (NCO) provides staff support 
for the NITRD program. The NCO provides program and financial 
management services, technical and subject matter expertise in 
facilitation, strategic planning, technical writing, networking and 
information technology services, and administrative staff support for 
the NITRD Subcommittee and other NITRD subgroups. The National Science 
Foundation (NSF) serves as the host agency for the NCO. \6\
---------------------------------------------------------------------------
    \6\  About the Subcommittee on Networking and Information 
Technology Research and Development (NITRD Subcommittee), http://
www.nitrd.gov/subcommittee/program.aspx.

---------------------------------------------------------------------------
Legislative History

    Congress originally authorized NITRD in the High-Performance 
Computing Act of 1991 (P.L. 102-194), after recognizing that a number 
of federal agencies had ongoing high-performance computing programs 
without a coordinating body. The Act established that coordinating body 
to improve interagency coordination, cooperation, and planning among 
those agencies with high-performance computing programs. In addition, 
it authorized a multi-agency research effort, called the High-
Performance Computing and Communications program, to accelerate 
progress in the advancement of computing and networking technologies 
and to support leading edge computational research in a range of 
science and engineering fields. The statute established a set of 
mechanisms and procedures to provide for the interagency planning, 
coordination, and budgeting of the research and development activities 
carried out under the program. The Act has since been amended through 
the Next Generation Internet Research Act of 1998 and the America 
COMPETES Act of 2007.
    In 2007, the America COMPETES Act amended the existing statute in 
several ways:

      Specified that the external advisory committee for the 
program must carry out biennial reviews of the funding, content and 
management of the interagency R&D program and report its findings to 
Congress;

      Required the Office of Science and Technology Policy 
(OSTP) to develop and maintain a roadmap for developing and deploying 
high-performance computing (high-end) systems; and

      Clarified that grand challenge problems supported under 
the interagency program are intended to involve multidisciplinary teams 
of researchers working on science and engineering problems.

NITRD Reauthorization in the 111th Congress

    In the 111th Congress, the U.S. House of Representatives passed 
H.R. 2020, the National Information and Technology Research and 
Development Reauthorization Act. (See Appendix A for details.) The bill 
sought to prioritize and strengthen federal information technology 
activities across the federal government by:

      Improving program planning and coordination through 
strategic planning and an Advisory Council with appropriate policy and 
technical expertise;

      Rebalancing portfolios to focus less on short-term goals 
and more on large-scale, long-term, interdisciplinary research with the 
potential to make significant contributions to society and U.S. 
competitiveness;

      Requiring the program to support R&D in cyber-physical 
systems and human-computer interactions, visualization, and information 
management, including the convening of a university/industry task force 
to explore collaborative R&D activities with participants from 
universities, federal labs, industry and other partners; and

      Formally codifying the role of the NCO and specifying the 
source of funding for the office.

    The Senate did not act on this legislation. H.R. 2020 was also made 
a part of the House-passed America COMPETES Reauthorization Act of 
2010, but the language was removed by the Senate before enactment.

2010 PCAST Report on NITRD

    In December 2010, the President's Council of Advisors on Science 
and Technology (PCAST) completed a legislatively required report on 
NITRD. The report, Designing a Digital Future: Federally Funded 
Research and Development in Networking and Information Technology, 
found that ``NITRD is well coordinated and that the U.S. computing 
research community, coupled with a vibrant Networking and Information 
Technology (NIT) industry, has made seminal discoveries and advanced 
new technologies that are helping meet many societal challenges.'' \7\
---------------------------------------------------------------------------
    \7\  President's Council of Advisors on Science and Technology, 
Report to the President and Congress December 2010, Designing a Digital 
Future: Federally Funded Research and Development in Networking and 
Information Technology, p. v.
---------------------------------------------------------------------------
    The 2010 report made several assessments about the role of the NIT 
field in answering the Nation's challenges and priorities:

      Advances in NIT are a key driver of economic 
competitiveness. They create new markets and increase productivity.

      Advances in NIT are crucial to achieving our major 
national and global priorities in energy and transportation, education 
and life-long learning, health care, and national and homeland 
security.

      Advances in NIT accelerate the pace of discovery in 
nearly all other fields.

      Advances in NIT are essential to achieving the goals of 
open government. \8\
---------------------------------------------------------------------------
    \8\  President's Council of Advisors on Science and Technology, 
Report to the President and Congress December 2010, Designing a Digital 
Future: Federally Funded Research and Development in Networking and 
Information Technology, p. vii.
---------------------------------------------------------------------------
    Stressing the need that federal investments be in NIT basic 
research, since the private sector is heavily involved in the 
development side, the report suggests that an investment of at least $1 
billion annually will be required for new, potentially transformative 
research. The report also recognizes that in the current economic 
uncertainty, repurposing and reprioritization of funding will be 
necessary, but does not rule out new funding and indicates a lower 
level of investment ``could seriously jeopardize America's national 
security and economic competitiveness.'' \9\
---------------------------------------------------------------------------
    \9\  Ibid., p. x.
---------------------------------------------------------------------------
    The PCAST report includes recommendations for increased investments 
in long-term, multi-agency research initiatives in health, energy and 
transportation, and cybersecurity. It emphasizes, ``Where fundamental 
NIT advances are needed to support these initiatives, mission agencies 
should invest in fundamental research in NIT, either alone or in 
collaboration with NSF, and should not limit their programs to 
application-specific research.'' \10\
---------------------------------------------------------------------------
    \10\  Ibid., p. xiii.
---------------------------------------------------------------------------
    The report also calls for exercising leadership to bring about 
changes in K-12 STEM education; enhancing the effectiveness of 
government coordination of NIT research and development; and redefining 
NITRD budget categories to separate NIT infrastructure for R&D in other 
fields from NIT R&D.
    With specific regard to education, the report finds that ``NIT is 
the dominant factor in America's science and technology employment, and 
that the gap between the demand for NIT talent and the supply of that 
talent is and will remain large.'' \11\ The report recommends 
increasing the number of graduates in NIT fields at all degree levels 
and calls for the inclusion of computer science in K-12 education.
---------------------------------------------------------------------------
    \11\  Ibid., p. 85.

---------------------------------------------------------------------------
NITRD Fiscal Year 2012 Budget Request

    In February 2011, NITRD released its Supplement to the President's 
Budget request. The Supplement is a summary of the NITRD research 
activities planned and coordinated for Fiscal Year 2012 (FY 12) for 
each of the participating agencies. The NITRD request totals $3.9 
billion for FY 2012, a 1.9 percent increase from FY 10 expenditures, 
and reflects many spending priorities recommended in the PCAST report.
    The NITRD Supplement breaks down budget requests for each of the 14 
federal agencies involved in NITRD according to the PCAs. \12\ (See 
Appendix B for details.)
---------------------------------------------------------------------------
    \12\  The Networking and Information Technology Research and 
Development Program Supplement to the President's FY 2012 Budget, p. 
30.
---------------------------------------------------------------------------
    For agencies within the jurisdiction of the Committee on Science, 
Space and Technology, the budget request totals are reflected in Table 
3:






    Major changes in investments for agencies within the Committee's 
jurisdiction include a $152 million (14 percent) increase for the 
National Science Foundation (NSF). This amount includes $35 million for 
High End Computing R&D for nanotechnology research and the Science, 
Engineering, and Education for Sustainability (SEES) effort and 
investment in Cyberinfrastructure Framework for the 21st Century 
Science and Engineering (CIF21); $22 million for cybersecurity 
activities; $60 million primarily to support a National Robotics 
Initiative; $12 million for basic research in radio spectrum systems; 
and $24 million in SDP for new software centers, CIF21, and increased 
SEES investment.
    The Department of Energy request includes a $112 million (27 
percent) increase: $66 million for research and new partnerships to 
address the challenges of emerging disruptive computing technologies 
from the private sector; $30 million for cybersecurity research; and 
$16 million for installation and operation of an Energy Sciences 
Network (ESnet) dedicated optical network.
    The National Institute of Standards and Technology includes an 
increase of $53 million (65 percent), $25 million of which is to be 
used to support new cybersecurity initiatives. The remainder is spread 
across other PCAs for interoperability in emerging technologies 
activities.

Appendix A

H.R. 2020

The Networking and Information Technology Research and Development Act 
                    of 2009

SECTION-BY-SECTION ANALYSIS

Section 1. Short Title ``Networking and Information Technology Research 
        and Development Act of 2009.''
Section 2. Program Planning and Coordination
    Requires the NITRD agencies to periodically assess the program 
contents and funding levels and to update the program accordingly.
    Requires the NITRD agencies to develop and periodically update (at 
three-year intervals) a strategic plan for the program. The 
characteristics and content of the strategic plan are described, and 
include strengthening NIT education, fostering technology transfer, and 
encouraging innovative, large-scale, and interdisciplinary research.
    Encourages a more active role for OSTP in ensuring that the 
strategic plan is developed and executed effectively and that the 
objectives of the program are met.
    Ensures that the existing advisory committee for NITRD is closely 
linked to the President's Council of Advisors on Science and Technology 
while retaining the necessary breadth and depth of expertise in NIT 
fields.
    Specifies that the annual report now required for the NITRD program 
explicitly describes how the program activities planned and underway 
relate to the objectives specified in the strategic plan.
    Specifies that the annual report now required for the NITRD program 
include a description of research areas supported in accordance with 
Section 3, including the same budget information as is required for the 
Program Component Areas.
Section 3. Large-Scale Research in Areas of National Importance
    Authorizes NITRD agencies to support large-scale, long-term, 
interdisciplinary research with the potential to make significant 
contributions to society and U.S. economic competitiveness and to 
encourage collaboration between at least two agencies as well as cost-
sharing from non-federal sources.
    Characteristics of the projects supported include: collaborations 
among researchers in institutions of higher education and industry, and 
may involve nonprofit research institutions and federal laboratories; 
leveraging of federal investments through collaboration with related 
State initiatives, when possible; and plans for fostering technology 
transfer.
    Authorizes support of activities under this section through 
interdisciplinary research centers that are organized to investigate 
basic research questions and carry out technology demonstration 
activities.
Section 4. Cyber-Physical Systems and Information Management
    Requires the program to support research and development in cyber-
physical systems; human-computer interactions, visualization, and 
information management.
    Requires the NCO Director to convene a university/industry task 
force to explore mechanisms for carrying out collaborative research and 
development activities for cyber-physical systems with participants 
from universities, federal laboratories, and industry. The NCO is to 
report to Congress on any findings and recommendations from the task 
force on models for collaborative R&D.
Section 5. National Coordination Office
    Formally establishes the NCO; delineates the office's 
responsibilities; mandates annual operating budgets; specifies the 
source of funding for the office (consistent with current practice); 
and stresses the role of the NCO in developing the strategic plan and 
in public outreach and communication with outside communities of 
interest.


    Chairman Brooks. The Subcommittee on Research and Science 
Education will come to order. Good afternoon. As a very quick 
preamble, we are expecting votes at any point in time. If we 
are called for votes, we will have to break temporarily until 
such point as the votes are concluded and then if it is all 
right with you, Congressman Lipinski, be back five minutes 
after the last vote? We will try to reconvene then five minutes 
after the last vote.
    That out of the way, welcome to today's hearing entitled 
``Oversight of the Networking and Information Technology 
Research and Development Program and Priorities for the 
Future.'' Today we are presented with the opportunity to review 
the Networking and Information Technology Research and 
Development Program, abbreviated NITRD, and to discuss 
priorities for the future.
    The NITRD Program is the main federal research and 
development investment portfolio in unclassified networking, 
computing, software, cybersecurity, and related information 
technologies. It also serves as the mechanism for interagency 
coordination of this research and development. Fourteen member 
agencies, including the National Science Foundation, National 
Aeronautics and Space Administration, the Department of Energy, 
NOAA, and the Department of Homeland Security provide budgets 
for NIT research and development. Numerous other agencies are 
also actively engaged in the coordination.
    Networking and information technology includes an array of 
technologies from smart phones to cloud computing. 
Multidisciplinary innovations include computational decoding of 
the human genome, modeling and simulation of complex physical 
systems for aircraft, automobiles, power grids and 
pharmaceuticals, near-real-time weather forecasts and climate 
models, and unmanned aerial vehicles and search-and-rescue 
robots. Among its many goals, NIT research and development in 
this field works to minimize and prevent disruptions to 
critical infrastructures like power grids and emergency 
communication systems. These investments are necessary not only 
to help maintain world leadership in science and engineering 
and strengthen U.S. competitiveness, but they also grow the 
economy through the creation of NIT jobs and enhance national 
security.
    For instance, cybersecurity is one of the biggest security 
challenges facing our Nation today. It permeates through all of 
our federal agencies and even into our private computer 
systems. This is just one area that the NITRD Program helps to 
coordinate our federal research and development. It indicates 
how imperative it is that we continue to support critical and 
collaborative research efforts such as this.
    Today, our witnesses will share with us their insights on 
the current state of the program and future priorities. It has 
been several years since the NITRD program was last reviewed by 
this Subcommittee. The program was reauthorized by the House in 
the last Congress on two occasions, only to languish in the 
Senate. Hopefully, input from our experts today will help 
inform this subcommittee's current work and bring to light new 
advances and challenges for the NIT R&D since the last bill's 
introduction.
    Part of this Subcommittee's role is to ensure that federal 
dollars are being spent on the best research and development. 
At a time when American competitiveness and national security 
are at risk, it is important that we maintain our lead in the 
development of these crucial technologies.
    I look forward to hearing from each of our witnesses on 
this important topic. Thank you for joining us.
    [The prepared statement of Mr. Brooks follows:]
                Prepared Statement of Chairman Mo Brooks
    Good morning, and welcome to each of our witnesses. Today, we are 
presented with the opportunity to review the Networking and Information 
Technology Research and Development Program (NITRD) and to discuss 
priorities for the future.
    The NITRD program is the main federal R&D investment portfolio in 
unclassified networking, computing, software, cybersecurity, and 
related information technologies. It also serves as the mechanism for 
interagency coordination of this R&D. Fourteen member agencies, 
including the National Science Foundation, NASA, the Department of 
Energy, NOAA, and the Department of Homeland Security provide budgets 
for NIT research and development. Numerous other federal agencies are 
also actively engaged in the coordination.
    Networking and information technology includes an array of 
technologies from smart phones to cloud computing. Multidisciplinary 
innovations include computational decoding of the human genome, 
modeling and simulation of complex physical systems for aircraft, 
automobiles, power grids and pharmaceuticals; near-real-time weather 
forecasts and climate models; and unmanned aerial vehicles and search-
and-rescue robots. Among its many goals, NIT research and development 
in this field works to minimize and prevent disruptions to critical 
infrastructures like power grids and emergency communication systems. 
These investments are necessary not only to help maintain world 
leadership in science and engineering and strengthen U.S. 
competitiveness, but they also grow the economy through the creation of 
NIT jobs and enhance national security.
    For instance, cybersecurity is one of the biggest security 
challenges facing our nation today. It permeates through all of our 
federal agencies and even into our private computer systems. This is 
just one area that the NITRD program helps to coordinate our federal 
R&D, but it indicates how imperative it is that we continue to support 
critical and collaborative research efforts such as this.
    Today, our witnesses will share with us their insights on the 
current state of the program and future priorities. It has been several 
years since the NITRD program was last reviewed by this Subcommittee. 
The program was reauthorized by the House in the last Congress on two 
occasions, only to languish in the Senate. Hopefully, input from our 
experts today will help inform this Subcommittee's current work and 
bring to light new advances and challenges for NIT R&D since the last 
bill's introduction.
    Part of this Subcommittee's role is to ensure that federal dollars 
are being spent on the best research and development. At a time when 
American competitiveness and national security are at risk, it is 
important that we maintain our lead in the development of these crucial 
technologies.
    I look forward to hearing from each of our witnesses on this 
important topic. Thank you for joining us.

    Chairman Brooks. At this time I defer to the Ranking Member 
on the Democrat side, Mr. Lipinski, for his remarks.
    Mr. Lipinski. Thank you, Chairman Brooks. As the Chairman 
noted it has been more than two years since this Committee 
developed and passed bipartisan legislation to reauthorize and 
update the NITRD Program. I was a co-sponsor of Chairman 
Gordon's bill in 2009, and as the Chairman said, the Senate 
never acted on it, of course, that is not the first time it has 
been said in this Subcommittee. It seems like almost every 
hearing. I am hoping that this hearing is the first step 
towards action in this Congress.
    Networking and information technologies are developing 
quickly. In the last two years not only has the NIT landscape 
changed but a committee of experts, PCAST, has delivered a new 
set of recommendations and priorities to Congress. The previous 
PCAST report was very helpful in developing our last bill, so I 
am looking forward to hearing from the witnesses about what has 
changed in the last two years and how that bill can be updated 
and improved.
    The NITRD Program evolved from a federal program 
established under the High Performance Computing Act of 1991. 
That act provided the funding that led to the development of 
Mosaic in 1993, the Worldwide Web browser that made the 
Internet user-friendly and led to its explosion in the 1990s. I 
am proud to note that Mosaic was created by a team of 
programmers at the federally funded National Center for Super 
Computing Applications at the University of Illinois.
    Netscape founder Mark Andreesen, who was the leader of the 
Illinois team before launching his own company, was quoted as 
saying, ``If it had been left to private industry, it wouldn't 
have happened. At least not until years later.''
    It was an unfortunately worded reference to the High-
Performance Computing Act of 1991, by its author and champion 
Al Gore, which turned into the punch line that Al Gore invented 
the Internet. But it is without question that the act did set 
the stage for coordinated federal R&D and investment strategy 
that has underpinned U.S. leadership in networking and 
information technology over the past two decades.
    Today we find ourselves in a different world in which U.S. 
leadership in NIT can no longer be taken for granted, and we 
need to think carefully about how we set priorities under 
difficult budget conditions. PCAST recommended three areas for 
priority investments; NIT for health, NIT for energy and 
transportation, and cybersecurity. This third area, 
cybersecurity, has been one of my highest priorities this 
Congress, and I joined Mr. McCaul in introducing the 
Cybersecurity Enhancement Act earlier this year.
    I look forward to hearing from witnesses about priorities 
and future directions of NITRD's cybersecurity component.
    Finally, I want to say a few words about NIT education and 
workforce issues. In 2009, SRI International produced a report 
on NIT workforce at the request of the NITRD Program office. In 
that report the analysts at SRI found that the NIT landscape is 
more complicated than just the ``more jobs than skilled 
workers'' mantra we sometimes hear. The supply and demand curve 
really depends on the sector within NIT and the level of 
education skills that we are talking about. I watch with 
growing concern as some of our leading IT companies have 
outsourced increasing numbers of jobs, following a disturbing 
pattern that has decimated manufacturing in our country.
    Over the past decade since, IBM's domestic workforce has 
suddenly shrunk while its overseas workforce has grown. As of 
last year it was down to one-quarter of the total, and for the 
first time ever, the company stopped providing breakouts of the 
number of employees that it has in the U.S. While the company's 
Project Match Program is offered to help workers laid off from 
domestic sites obtain travel and visa assistance for jobs in 
countries like India, China, and Brazil, and even though IBM 
has withdrawn its patent application for a ``Method and System 
for Strategic Global Resource Sourcing,'' I am worried that we 
could be training students for jobs that end up having the jobs 
being outsourced.
    At the same time I know that we have a real need for 
cybersecurity professionals who can help protect our most 
sensitive networks, informaticians who can discover new ways to 
deal with the exponentially growing amount of data we produce.
    So I want to hear from the witnesses today about how we can 
be confident we are training students for jobs that will be 
available here in the U.S. and how we can focus education and 
training resources within NITRD on those job skills.
    I thank the witnesses again for taking the time to appear 
before us and to help educate us about the NITRD Program, and I 
look forward to your testimony.
    Thank you.
    [The prepared statement of Mr. Lipinski follows:]
          Prepared Statement of Ranking Member Daniel Lipinski
    Thank you, Chairman Brooks. [As the Chairman noted] It has been 
more than two years since this Committee developed and passed 
bipartisan legislation to reauthorize and update the NITRD program. I 
was a cosponsor of Chairman Gordon's bill in 2009, and while the Senate 
never acted on it, I hope that this hearing is the first step towards 
action in this Congress.
    Networking and Information Technology are developing quickly, and 
in the last two years not only has the NIT landscape changed, but a 
committee of experts, PCAST, has delivered a new set of recommendations 
and priorities to Congress. The previous PCAST report was very helpful 
in developing our last bill, so I am looking forward to hearing from 
the witnesses about what's changed in the last two years and how that 
bill could be updated and improved.
    The NITRD program evolved from a federal program established under 
the High Performance Computing Act of 1991. That Act provided the 
funding that led to the development of Mosaic in 1993, the World Wide 
Web browser that made the Internet user-friendly and led to its 
explosion in the 1990s. I am proud to note that Mosaic was created by a 
team of programmers at the federally funded National Center for 
Supercomputing Applications at the University of Illinois. Netscape 
founder Marc Andreeson, who was a leader of the Illinois team before 
launching his company, was quoted as saying, ``If it had been left to 
private industry, it wouldn't have happened, at least, not until years 
later.''
    It was an unfortunately worded reference to the High Performance 
Computing Act of 1991 by its author and champion, Al Gore, which turned 
into the punchline that Al Gore invented the Internet. But it is 
without question that that Act set the stage for a coordinated federal 
R&D and investment strategy that has underpinned U.S. leadership in 
networking and information technology over the past two decades, But 
today we find ourselves in a different world, in which U.S. leadership 
in NIT can no longer be taken for granted, and we need to think 
carefully about how we set priorities under difficult budget 
conditions. PCAST recommended three areas for priority investments: NIT 
for health, NIT for energy and transportation, and cybersecurity. This 
third area, cybersecurity, has been one of my highest priorities this 
Congress, and I joined Mr. McCaul in introducing the Cybersecurity 
Enhancement Act earlier this year. I look forward to hearing from 
witnesses about priorities and future directions of NITRD's 
cybersecurity component.
    Finally, I want to say a few words about NIT education and 
workforce issues. In 2009, SRI International produced a report on the 
NIT workforce at the request of the NITRD program office. In that 
report, the analysts at SRI found that the NIT landscape is more 
complicated than just the ``more jobs than skilled workers'' mantra we 
sometimes hear. The supply and demand curve really depends on the 
sector within NIT, and the level of education and skills that we are 
talking about.
    I have watched with growing concern as some of our leading IT 
companies have outsourced increasing numbers of jobs, following the 
disturbing pattern that has decimated manufacturing in this country. 
Over the past decade, for instance, IBM's domestic workforce has 
steadily shrunk while its overseas workforce has grown. As of last year 
it was down to one quarter of the total, and for the first time ever 
the company stopped providing breakouts of the number of employees it 
has in the U.S.
    While the company's ``Project Match'' program has offered to help 
workers laid off from domestic sites obtain travel and visa assistance 
for jobs in countries like India, China, and Brazil, and even though 
IBM has withdrawn its patent application for a ``Method and System for 
Strategic Global Resource Sourcing,'' I worry that we could be training 
students for jobs that end up being outsourced. At the same time, I 
know that we have a real need for cybersecurity professionals who can 
help protect our most sensitive networks and informaticians who can 
discover new ways to deal with the exponentially growing amount of data 
we produce.So I want to hear from the witnesses today about how we can 
be confident we are training students for jobs that will be available 
here in the U.S., and how we can focus education and training resources 
within NITRD on those job skills. Thank you again for taking the time 
to appear before us today to help educate us about the NITRD program, 
and I look forward to your testimony.

    Chairman Brooks. Thank you, Mr. Lipinski.
    If there are Members who wish to submit additional opening 
statements, your statements will be added to the record at this 
point.
    At this time I would like to introduce our witnesses for 
today's panel. First we have Dr. George Strawn. He is the 
director of the National Coordination Office for the Networking 
and Information Technology Research and Development Program. 
Prior to his appointment as Director of NITRD Dr. Strawn served 
as the Chief Information Officer at the National Science 
Foundation. Dr. Strawn holds a Bachelor of Arts in mathematics 
and physics from Cornell College and a Doctorate in mathematics 
from Iowa State University.
    Next we have Dr. Edward Lazowska. He holds the Bill and 
Melinda Gates Chair in Computer Science and Engineering at the 
University of Washington and is the Founding Director of the 
University of Washington eScience Institute. He also serves as 
a board member or technical advisor to a number of high-tech 
companies and venture firms. Dr. Lazowska holds a Bachelor of 
Arts from Brown University and a Doctorate from the University 
of Toronto.
    Dr. Robert Sproull recently retired as Vice President, 
Director of Oracle Labs. He is a member of the National Academy 
of Engineering, a Fellow of the American Academy of Arts and 
Sciences, and had served on the United States Air Force 
Scientific Advisory Board. Dr. Sproull holds a Bachelor of Arts 
in physics from Harvard and a Doctorate in computer sciences 
from Stanford.
    Dr. Robert Schnabel is the Dean of the School of 
Informatics at Indiana University. From August 2009 to July 
2010, he served as Interim Vice President for Research for 
Indiana University. Dr. Schnabel holds a Bachelor of Arts in 
mathematics from Dartmouth College and a Master of Science and 
a Doctorate from Cornell University, both in computer science.
    As our witnesses should know, spoken testimony is limited 
to five minutes, after which the Members of the Committee will 
have five minutes each to ask questions.
    At this point I recognize our first witness, Dr. George 
Strawn. Dr. Strawn, you are recognized for five minutes.

STATEMENT OF DR. GEORGE STRAWN, DIRECTOR, NATIONAL COORDINATION 
  OFFICE, NETWORKING AND INFORMATION TECHNOLOGY RESEARCH AND 
                      DEVELOPMENT PROGRAM

    Dr. Strawn. Good afternoon. As you have noted, I am George 
Strawn, the Director of the National Coordination Office for 
the NITRD Program, and along with Farnam, Dr. Farnam Jahanian 
of NSF, I also co-chair the NITRD Interagency Committee. I 
would like to thank Chairman Brooks, Ranking Member Lipinski, 
and Members of the Subcommittee for the opportunity to discuss 
the role of the NITRD Program in helping the United States 
maintain its leadership in networking and information 
technology.
    The NITRD Program provides for the coordination of the 
government's portfolio of unclassified investments in research 
and development in NIT. The National Coordination Office hosted 
at NSF facilitates the various activities of the NITRD Program. 
My written testimony describes NITRD in some detail, and I will 
now highlight three of the topics discussed there: first, some 
of our newest activities; second, why federal investment in NIT 
R&D remains crucial; and third, one of the reasons why NITRD 
collaboration is successful.
    First, our newest NITRD activities include five new 
coordination groups, one in NIT education, a second in 
cybersecurity, and in fact, it is our second coordination group 
in cybersecurity, one in health IT, one in wireless spectrum 
efficiency, and one in what we are calling big data. 
Cybersecurity is important enough that this subcommittee held a 
separate hearing on it last May. Given my time limits I will 
just say a word about big data as one of our new activities.
    Federal agencies in the NITRD Program are generating 
exabytes of research data annually. An exabyte is a billion, 
billion bytes, and society at large is generating a similar 
amount. Our ability to create and store data has greatly 
outpaced our ability to preserve, manage, access, and make 
effective use of it. The agencies participating in our big data 
group have identified four initial focus areas; core 
technologies research, big data projects, education and 
training for big data scientists, and competitions and prizes 
to stimulate general activity in the area.
    I anticipate important advances in the science and practice 
of big data.
    Next, the importance of federal NIT investments is 
illustrated by the many federally-supported projects that 
helped spawn the information age, going all the way back to the 
19th century. The U.S. Congress supported Samuel F.B. Morse's 
development of the telegraph, for example. The U.S. Census 
Bureau supported the development of an innovative punch-card 
technology to store and process census data, which became the 
IBM Corporation. More recent examples include the U.S. Army's 
development of the electronic digital computer, DARPA's 
development of the ARPAnet followed by NSF's development of the 
NSFnet, followed by the Internet industry. And as previously 
mentioned, NSF's support for the research that led to the first 
Web browser and then to the Google search engine.
    The multiplier effects of federal NIT R&D investments is 
widely documented, perhaps most famously in the National 
Research Council's tire tracks diagram, which shows the 
interplay of federal and private sector investments that has 
turned many NIT research projects into billion dollar markets.
    Finally, one reason for the effectiveness of NITRD 
collaboration model is that it involves both mission agencies 
and agencies who are focused on the use of IT and agencies 
focused on the theory of IT. As political scientist Donald 
Stokes showed in his 1997 book, ``Pasteur's Quadrant,'' R&D is 
better described as a two-dimensional activity rather than the 
usual linear description where research always precedes 
development.
    His dimensions were use value and theory value, and his 
point was high-use value can be concurrent with or even 
occasionally precede high-theory value. For example, the Army's 
need for and development of the digital computer preceded the 
development of most computing science theory. NIT R&D can be 
especially fruitful when high use and high-theory value are 
brought together, and that is exactly what the NITRD Program 
seeks to accomplish.
    Thank you, again, for the opportunity to provide this 
testimony, and I would be happy to answer any questions you may 
have.
    [The prepared statement of Mr. Strawn follows:]
                PREPARED STATEMENT OF DR. GEORGE STRAWN,
                DIRECTOR, NATIONAL COORDINATION OFFICE,
 NETWORKING AND INFORMATION TECHNOLOGY RESEARCH AND DEVELOPMENT PROGRAM
    I am George Strawn, Director of the National Coordination Office 
(NCO) for the Networking and Information Technology Research and 
Development Program (NITRD). With my colleague, Dr. Farnam Jahanian of 
the National Science Foundation (NSF), I co-chair the NITRD 
Subcommittee of the National Science and Technology Council's Committee 
on Technology. I want to thank Chairman Brooks, Ranking Member 
Lipinski, and Members of the Subcommittee for the opportunity to come 
before you today to discuss the role of the NITRD Program in helping 
the United States maintain leadership in networking and information 
technology (NIT).
    Prior to coming to the NCO for NITRD in 2009, I was at the National 
Science Foundation (NSF) and had taken part in NITRD activities as an 
NSF staff member since 1995. The positive impression I formed about 
NITRD during those years led me to apply for my current position and 
adds to my appreciation of this opportunity to testify before you 
today.

NITRD Overview

    The NITRD Program has been authorized under three legislative acts. 
The first, the High Performance Computing Act of 1991 (Public Law 102-
194), established the Program, setting forth a framework that combined 
research goals with specific provisions for interagency cooperation, 
collaboration, and partnerships with industry and academia. Two 
additional acts-- the Next Generation Internet Research Act of 1998 
(Public Law 105-305) and the America COMPETES Act of 2007 (Public Law 
110-69)--reauthorized the Program and extended its scope in various 
ways.
    As the NITRD Program this year celebrates its 20th anniversary, I 
hope the Members of this Subcommittee and its parent Committee share 
the NITRD community's pride that our multi-agency framework has truly 
met the test of time. Over the course of two decades, the authorizing 
legislation has enabled the NITRD enterprise to evolve and expand to 
address increasingly rapid technological shifts and new 
responsibilities.
    The framework of the NITRD Program provides for coordination across 
the government's portfolio of unclassified investments in fundamental, 
long-term research and development (R&D) in advanced networking and 
information technology (NIT). NITRD research supports both the missions 
of our federal agencies and the Nation's broader goals such as homeland 
and national security, economic competitiveness, energy independence, 
environmental stewardship, affordable health care, and science and 
engineering leadership.
    All of the research reported in the NITRD portfolio is managed, 
selected, and funded by one or more of the 18 NITRD member agencies 
(listed on page 14) under their own individual authorizations and 
appropriations. The Program's major research areas (termed Program 
Component Areas [PCAs] in the 1998 reauthorization legislation) 
currently include:

      1. Cyber security and information assurance;

      2. High-confidence software and systems;

      3. High-end computing;

      4. Human-computer interaction and information management;

      5. Large-scale networking;

      6. Social, economic, workforce implications of IT and 
workforce development;

      7. Software design and productivity.

    We have also launched some exciting new ventures that are 
highlighted below.
    NITRD research is performed in universities, federal research 
centers and laboratories, federally funded R&D centers, private 
companies, and nonprofit organizations across the country. The synergy 
exhibited by the NITRD member and participating agencies (listed on 
page 15)--is accomplished through interaction across the government, 
academic, commercial, and international sectors using cooperation, 
coordination, information sharing, and joint planning. Collaborative 
activities are focused on selected areas where the agencies can 
identify technical challenges that multiple agencies face and address 
them together to leverage each other's activities. These targeted 
collaborations enable the agencies to maximize resource sharing, 
minimize duplication of effort, and partner in investments to pursue 
higher-level goals.

Structure of NITRD Coordination

    The Subcommittee on Networking and Information Technology Research 
and Development of the National Science and Technology Council's (NSTC) 
Committee on Technology (CoT) serves as the internal deliberative 
organization for NITRD policy, program, and budget guidance and 
direction within the Executive Branch. The NITRD Subcommittee interacts 
with federal agencies that need advanced networking or information 
technology to identify networking and information technology research 
and development needs. Its high-level goals are to help assure 
continued U.S. leadership in networking and IT, satisfy the needs of 
the federal government for advanced IT capabilities, and accelerate 
development and deployment of new technologies. Subcommittee members 
include senior R&D managers from each of the member agencies and 
representatives from the Office of Management and Budget (OMB), Office 
of Science and Technology Policy (OSTP), and the NCO. The Subcommittee 
interacts with Congress, OMB, OSTP, the NSTC, the CoT, other Federal 
agencies, and private-sector and international organizations on behalf 
of the NITRD Program.
    The NCO facilitates the activities of the NITRD Program (see list 
immediately below), and the office serves as the hub of public 
information about the Program. The NCO's technical, administrative, NIT 
services, and administrative support staff provide program and 
financial management services; technical and subject-matter expertise 
in facilitation, strategic planning, technical writing, networking and 
IT services; and administrative staff support for the NITRD 
Subcommittee and the IWGs, CGs, SSGs, Teams, and other NITRD subgroups. 
The cost of operating the NCO is shared by the NITRD member agencies in 
proportion to their NITRD budgets. The NCO also supports the 
President's Council of Advisors on Science and Technology (PCAST), 
under Executive Order 13539. The NCO Director reports to OSTP, works 
closely with OSTP and OMB, and attends OSTP technical-staff meetings. 
NSF serves as the host agency for the NCO. The NCO maintains the NITRD 
Web site (www.nitrd.gov) and prepares and archives NITRD publications.
    Supported by the NCO's staff and services, the NITRD Program uses 
the following general mechanisms to pursue its coordination mission:

      Monthly meetings of its Interagency Working Groups 
(IWGs), Senior Steering Groups (SSGs), Coordinating Groups (CGs), 
Teams, and Subgroups (SGs). These regular interactions among 
representatives from many agencies enable participants to exchange 
information and collaborate on research plans and activities such as 
standards development, testbeds, research workshops, cooperative 
solicitations, and sharing operational best practices for federal NIT, 
such as in the annual DOE High Performance Computing (HPC) Best 
Practices Workshop. \1\ Also as a result of these exchanges, NITRD 
representatives frequently serve on grant review panels and participate 
in principal investigator meetings of other agencies.
---------------------------------------------------------------------------
    \1\  http://www.nersc.gov/events/hpc-workshops/

      Formation of new coordination activities as needed to 
address national priorities. These are described under "New NITRD 
---------------------------------------------------------------------------
Ventures" below.

      Workshops, which typically include academic and industry 
participants from across the country as well as federal 
representatives.

      Formal reports, including the annual NITRD Supplement to 
the President's Budget, strategic planning documents, and workshop 
reports. These documents all play an important role in helping set 
national agendas in research areas of critical interest.

      Support for external studies and assessments.

      Outreach to the federal and private sectors.

New NITRD Ventures

Cybersecurity SSG

    In 2008, as part of a mandate to improve coordination of federal 
cybersecurity R&D under the Comprehensive National Cybersecurity 
Initiative (CNCI), the NITRD agencies developed a plan calling for the 
establishment of a new kind of coordinating group under NITRD. An 
interagency Senior Steering Group (SSG) for cybersecurity R&D was 
established whose members included federal managers with budgetary 
responsibilities. This group was in addition to our cybersecurity PCA. 
A key goal of the SSG concept was to facilitate moving strategic 
cybersecurity R&D approaches developed by the NITRD agencies into 
programmatic activities. The Cybersecurity SSG has proven effective and 
we have adopted the same model to establish other NITRD SSGs.
    Like the Cybersecurity SSG, the new SSGs enable NITRD to broaden 
its focus from established NIT R&D categories to ones that address new 
opportunities and critical national priorities for the United States. 
These SSGs, and a new venture in the education arena, are described in 
the following sections. The participating agencies for each of the SSGs 
can be found on pages 16-17.

Big Data SSG

    Most of the world's information is now ``born digital,'' and legacy 
texts, images, sounds, videos, and films as well are being digitized 
around the clock. Typical estimates put the amount of digital data 
generated annually at many orders of magnitude greater than the total 
amount of information in all the books ever written, and the total is 
expected to continue growing exponentially. In the sciences alone--a 
central concern of our federal science agencies --the proliferation of 
ultra-powerful and distributed data-collection instruments and 
experimental facilities has turned the conduct of leading-edge research 
into a global-scale, data-intensive enterprise. Together, the federal 
agencies in the NITRD Program generate exabytes of research data 
annually. Financial, commercial, communications, and Web-based 
enterprises likewise generate vast amounts of new digital information 
on a moment-by-moment basis. However, our capacity to create electronic 
data is outpacing advances in the technologies needed to enable us to 
preserve, manage, access, and make effective use of society's data 
resources--the highly complex, ultra-large-scale data sets that we in 
NITRD refer to as ``big data.''
    NITRD has a PCA for human-computer interaction and information 
management, but big data offers new possibilities and new challenges. 
Responding to a request from OSTP, NITRD formed the Big Data Senior 
Steering Group in January 2011. The Big Data SSG is charged with 
identifying current big data R&D activities across the Federal 
Government, offering opportunities for coordination and collaboration, 
and considering what national initiatives on big data would be most 
useful. The science of big data begins with issues of scale, 
complexity, and heterogeneity encompasses the many significant 
challenges in turning data into knowledge, including search, discovery, 
mining and visualization of ultra-scale data; interoperability; and 
semantics. In their first months, the agencies participating in the Big 
Data SSG have identified four focus areas for their initial activities: 
core technologies, data projects, training, and competitions.

Health IT R&D SSG

    The formation of NITRD's Health IT R&D SSG is our response to the 
American Recovery and Reinvestment Act of 2009 (Public Law 111-5), 
which called on NITRD to ``have programs in Health IT R&D.'' We have 
always had mission agency members, but this is the first time that a 
formal NITRD activity is devoted to a mission programmatic goal--
improving health and health care. Health IT R&D includes fundamental 
research, applied R&D, technology development and engineering, 
demonstrations, testing and evaluation, technology transfer, and 
education and training.
    The Health IT R&D SSG was launched in January 2010 after an initial 
NITRD planning activity on the topic. The agencies participating in the 
group are working towards a next-generation health information 
infrastructure that will provide universal, interoperable information 
systems for U.S. health care. R&D challenges include, to name just a 
few: universal data exchange language; security, privacy, and identity 
management; interoperable electronic health records (EHRs); personal 
health records (PHRs); devices/sensors; and further empowering of ``e-
patients.'' The SSG's interests also encompass: coordination of 
standards, implementation specifications, and certification criteria; 
maintaining frequent communication with, and serving as the liaison 
among, the SSG agencies, academia, and industry; and responding to U.S. 
national goals for health IT R&D and the health IT recommendations of 
PCAST in its 2010 report Realizing the Full Potential of Health 
Information Technology to Improve Healthcare for Americans: The Path 
Forward. \2\
---------------------------------------------------------------------------
    \2\  http://www.whitehouse.gov/sites/default/files/microsites/ostp/
pcast-health-it-report.pdf
---------------------------------------------------------------------------
    The Health IT R&D SSG is developing a health information technology 
recommendations document that provides research directions for health 
IT R&D. In addition, the SSG has formed the Health Information 
Technology Innovation and Development Environments (HITIDE) subgroup, 
which is working on policy and governance issues for experimental 
testbed activities. Testbeds will enable our agencies to try out new 
health technology prototypes in realistic, real-time environments.

Wireless Spectrum R&D (WSRD) SSG

    The Wireless Spectrum R&D SSG was established in November 2010 to 
coordinate spectrum-related research and development activities across 
the Federal Government. WSRD's purpose is: to help coordinate and 
inform ongoing activities across federal agencies; and to facilitate 
the identification of gaps in the government's R&D portfolio with 
respect to technologies that allow a more efficient use of spectrum. 
These activities are consistent with the guiding principles of WSRD, 
which are transparency, smart investment, and the expansion of 
opportunities for technology transfer across and beyond the Federal 
Government.
    The WSRD members have developed a preliminary inventory of some 670 
federal wireless spectrum R&D activities and are preparing a gap 
analysis from the inventory and recommendations on federal research 
that could advance the goals of the June 28, 2010, Presidential 
Memorandum: ``Unleashing the Wireless Broadband Revolution,'' Section 
3. In addition, WSRD will work with academia and the private sector to 
develop priorities, encourage private investment, and develop public/
private partnerships when appropriate.

Cyber-Physical Systems

    H.R. 2020 calls out cyber-physical systems (CPS) as a new area of 
national priority for NITRD activity. Cyber-physical systems are real-
time, networked computing systems--interconnected software, 
microprocessors, sensors, and actuators--deeply integrated within 
engineered physical systems to monitor and control capabilities and 
behaviors of the physical system as a whole. Such systems have become 
increasingly important to our society and are essential to the 
effective operation of: U.S. defense and intelligence systems; critical 
civilian infrastructures (e.g., air traffic control, power grid, and 
water supply systems), industrial-process control systems, and other 
large-scale civilian systems; as well as to smaller-scale systems that 
are vital for U.S. economic competitiveness (e.g., in airplanes, cars, 
robotic devices, and medical instruments and devices).
    Much of the work of NITRD's High Confidence Software and Systems 
(HCSS) PCA over the past decade has been focused on CPS research and 
building a national CPS research community that engages multiple 
sectors and disciplines. Currently, we are considering augmenting this 
work with an SSG devoted specifically to CPS R&D. In addition to H.R. 
2020, the last two PCAST reviews of our Program (Leadership Under 
Challenge: Information Technology R&D in a Competitive World, \3\ 
August 2007, and Designing a Digital Future: Federally Funded Research 
and Development in Networking and Information Technology, \4\ December 
2010) concluded that improving the quality, capabilities, and 
trustworthiness of our life- and safety-critical information 
technologies, including cyber-physical systems, should be a key focus 
of federal research.
---------------------------------------------------------------------------
    \3\  http://www.whitehouse.gov/sites/default/files/microsites/ostp/
pcast-07-nitrd-review.pdf
    \4\  http://www.whitehouse.gov/sites/default/files/microsites/ostp/
pcast-nitrd-report-2010.pdf

---------------------------------------------------------------------------
Education and Workforce Activities

    As employers with needs for all kinds of highly skilled scientific 
and technical NIT personnel, the NITRD agencies are acutely aware of 
some of the problems in our formal education system that limit the 
number of graduates adequately prepared to become part of the NIT 
workforce. To underscore their concern, in the draft strategic plan for 
NITRD the agencies highlight development of a ``cyber-capable'' U.S. 
population as one of three critical foundations for a bright national 
future. One immediate outcome of the strategic planning discussions has 
been the establishment of a new Team (called SEW-Education) under the 
auspices of NITRD's Social, Economic, and Workforce Implications of IT 
and IT Workforce Development (SEW) Program Component Area.
    The SEW-Education team is seeking ways to promote the integration 
of instruction about the science of computing throughout the K-12 
curriculum. Indeed, the former co-Chair of the NITRD Subcommittee, Dr. 
Jeannette Wing (now back at the computer science department at 
Carnegie-Mellon University), introduced the concept of ``computational 
thinking for everyone.'' She spearheaded NSF initiatives to support 
development of innovative ways to familiarize students at all levels 
with the fundamental concepts of computation, such as algorithms, and 
how they can be applied to solve problems in every domain--just as 
students now learn fundamental concepts in mathematics and other 
sciences in grade-appropriate curricula starting at the elementary 
level.
    In national public forums we held in 2008 and 2009 to inform NITRD 
strategic planning, academic computer scientists and K-12 educators 
told us that few, if any, K-12 schools had a curriculum in computer 
science (CS). According to these experts, computer science teaching was 
limited to an introductory high school course in programming, offered 
by only 65 percent of high schools in 2009 and taken by a small 
percentage of students. In lower grades, they said, teachers informally 
helped students use computer applications but there was virtually no 
instruction about the science of computation. In a society increasingly 
dependent on complex digital systems, the NITRD agencies believe, the 
gaps in K-12 students' knowledge and experience, and in the 
availability of skilled CS teachers, are worrisome and need to be 
addressed through grade-appropriate computer science curricula.
    At the same time, the demand for NIT workers is growing. A 2009 
study conducted for NITRD by SRI&otes that two NIT-related 
occupations--network systems and data communications analyst, and 
computer applications software engineer--are among the five fastest-
growing in the U.S. economy, and are the only two of the five to 
require a college degree. 1A\5\ According to U.S. Bureau of Labor 
Statistics projections in 2010, there were 7.6 million STEM workers in 
the United States, representing about one in 18 workers. STEM 
occupations are projected to grow by 17.0 percent from 2008 to 2018, 
compared to 9.8 percent growth for non-STEM occupations. STEM workers 
command high wages, earning 26 percent more than their non-STEM 
counterparts, more than two-thirds of STEM workers have a least a 
college degree, compared to less than one-third of the non-STEM 
workers, and STEM degree holders enjoy high earnings, regardless of 
whether they work in STEM or non-STEM occupations.
---------------------------------------------------------------------------
    \5\  Commissioned by the NITRD Subcommittee in response to 
Recommendation 2.1 (page 23) of Leadership Under Challenge, the 2007 
PCAST report on its review of the NITRD Program.
---------------------------------------------------------------------------
    Furthermore, labor-market projections for the NIT workforce do not 
capture the reality that a very broad range of occupations increasingly 
involves applications that require NIT knowledge and skills. Nor can 
statistical projections serve as a guide for assessing the adequacy of 
the educational system to prepare a workforce that leads the world in 
advanced innovation. The managers of the NSF programs targeting the K-
12 problem participate in the SEW-Education group and are helping 
develop its action plan. The first NSF effort, Computing Education for 
the 21st Century (CE21), is focusing special attention on the middle 
school through early college levels, with the goals of: increasing the 
number and diversity of students and teachers who develop and practice 
computational competencies in a variety of contexts; and increasing the 
number and diversity of postsecondary students who are engaged and have 
the background in computing necessary to successfully pursue degrees in 
computing-related and computationally intensive fields of study. \6\
---------------------------------------------------------------------------
    \6\  http://www.nsf.gov/funding/
pgm-summ.jsp?pims-id=503582
---------------------------------------------------------------------------
    The second NSF activity, CS 10K (which stands for 10,000 Computer 
Science teachers in 10,000 high schools), aims to increase the 
effectiveness of computing education in high school through the 
introduction of an entirely new curriculum (based on a proposed, new 
Advanced Placement course) concomitant with the preparation of teachers 
prepared to teach it by 2015. \7\
---------------------------------------------------------------------------
    \7\  http://www.computingportal.org/cs10k
---------------------------------------------------------------------------
    A three-year, $14.2 million effort initiated in 2010 by SEW-
Education participant and NITRD member DARPA is also directed to the 
middle- and high-school levels of the K-12 system. Citing the decline 
in CS college graduates and the growing need for computer scientists 
and engineers, the agency's solicitation asked for innovative proposals 
to: combat young people's misperception that the ``dot.com'' bust 
eliminated all CS jobs; excite middle- and high-school students about 
CS and STEM careers; provide means of retaining the excitement of 
extracurricular activities, such as NASA's Space Camp, in the regular 
curriculum; and offer plans for institutionalizing the new approaches 
over the long term. \8\
---------------------------------------------------------------------------

    \8\ http://www.darpa.mil/Our-Work/I2O/Programs/
Computer-Science-in-Science-
Technology-Engineering-and-Mathematics-
Education-(CS-STEM).aspx
---------------------------------------------------------------------------
    These efforts are complemented by the National Initiative for 
Cybersecurity Education (NICE) led by NIST. This comprehensive program, 
to which many NITRD agencies are contributing, includes activities in 
four component areas: national cybersecurity awareness; formal 
cybersecurity education; cybersecurity workforce structure; and 
cybersecurity workforce training and professional development. \9\
---------------------------------------------------------------------------
    \9\  http://csrc.nist.gov/nice/
---------------------------------------------------------------------------
    SEW-Education also plans to coordinate its efforts with those of 
the new NSTC Committee on Science, Technology, Engineering, and Math 
Education (CoSTEM). The COMPETES Act of 2010 directed OSTP to set up 
this committee, which is co-chaired by OSTP and NSF, and gave it the 
following responsibilities:

      Coordinate the STEM education activities and programs of 
the federal agencies;

      Coordinate STEM education activities and programs with 
the Office of Management and Budget;

      Encourage the teaching of innovation and entrepreneurship 
as part of STEM education activities;

      Review STEM education activities and programs to ensure 
they are not duplicative of similar efforts within the federal 
government;

      Develop, implement through the participating agencies, 
and update once every five years a five-year STEM education strategic 
plan.

Other Recent NITRD Highlights

    In addition to the recent NITRD developments described above, I am 
pleased to report that we have welcomed four new agencies as members of 
the NITRD Program. Last spring, the Office of the National Coordinator 
(ONC) for Health Information Technology of the Department of Health and 
Human Services became a NITRD member. The ONC representative also 
serves as a co-chair of the Health IT R&D Senior Steering Group. We are 
also delighted that the Department of Defense's Service research 
organizations--the Air Force Research Laboratory (AFRL), the Army 
Research Laboratory (ARL), and the Office of Naval Research (ONR) have 
come onboard as NITRD member agencies. The Service labs have long been 
active participants in the Program's research groups, so it is 
gratifying to have the benefit of their contributions as Subcommittee 
members as well.
    I want also to note briefly that the National Coordination Office 
is working on a prototype R&D dashboard that will provide greater 
access to NITRD funding data, enabling the public to explore the 
Program's research activities in greater depth. You can find our 
initial conceptual dashboard on the NITRD Web site. \10\
---------------------------------------------------------------------------
    \10\  http://itdashboard.nitrd.gov/

---------------------------------------------------------------------------
Maintaining U.S. Leadership in Networking and Information Technology

    Networking and information technology--computers, wired and 
wireless digital networks, electronic data and information, IT devices 
and systems, and software applications--today provide the indispensable 
infrastructure for activities across all facets of our society and our 
economy. Throughout the IT revolution, the United States has led the 
world in the invention and applications of these technologies. For well 
over six decades, ongoing federal research and development to supply 
advanced IT capabilities for government missions has fueled the 
creation of new ideas, innovators, and innovations addressing key 
national priorities, such as those cited above and repeated here for 
emphasis: homeland and national security, economic competitiveness, 
energy independence, environmental stewardship, affordable health care, 
and science and engineering leadership. In fact, the 2010 PCAST review 
of NITRD noted that ``the federal investment in NIT research and 
development is without question one of the best investments our Nation 
has ever made.''
    There is no doubt that the historic U.S. supremacy in NIT is under 
global challenge from aggressive competitors. We no longer manufacture 
all the components for the NIT products we use, and that pipeline is 
something we need to monitor carefully. However, we need to be mindful 
that the U.S. companies that so successfully built our country's 
multibillion-dollar NIT commercial marketplace are also becoming global 
enterprises. I believe that we remain the world's NIT leader, but 
continued innovation leadership is required to maintain our position.
    I come to the NIT leadership question with the perspective of a 
computer scientist who has for many years been a student of the history 
of U.S. information technologies. Technology innovation proceeds in 
extended cycles. Big--usually unexpected--scientific discoveries come 
first, followed by long periods of incremental innovations and 
commercialization of products developed out of the initial fundamental 
advances. In the field of networking and information technology in 
particular, the Federal Government has historically been the sponsor of 
the fundamental scientific breakthroughs that spawned the information 
revolution. This history dates back to the 1800s. In the 1830s, the 
U.S. government supported Samuel F.B. Morse's development of the 
telegraph. In the 1890s, the U.S. Census Bureau supported the work of 
its employee, Herman Hollerith, in developing an innovative punch-card 
technology to record and store census data. Several decades later, the 
company Hollerith subsequently started became the International 
Business Machines Corporation (IBM). More recent examples include the 
Army-supported development of the electronic computer; DARPA's support 
for the ARPAnet followed by NSF's support for the NSFnet, which became 
the Internet; and NSF's support for the research that brought us the 
first Web browser and the Google Web search engine.
    The NITRD Program sustains this historic federal role in discovery, 
in the 21st century's far more complicated global technological and 
economic environment. As noted above, our collaborative multi-agency 
framework has enabled the NITRD agencies to keep evolving their NIT 
activities to keep pace with the increasing pace of emerging 
technologies and applications. The portfolio of research and 
development activities sponsored by the NITRD agencies grows ever 
broader. I would argue that this portfolio is an invaluable resource 
for maintaining U.S. leadership in NIT because it is the Nation's only 
full-spectrum NIT R&D enterprise.
    Thus the NITRD portfolio serves a unique purpose in what many term 
``the U.S. innovation ecosystem.'' Over the decades, the United States 
has developed a fluid, information-rich research and innovation 
environment that stretches from federal programs and laboratories, 
across university campuses and research centers, to industrial R&D 
facilities and small business start-ups. As the National Academies and 
others have noted, \11\ there are innumerable feedback loops in this 
ecosystem through which ideas and concepts travel, get transformed, 
fuel new directions, turn student experimenters into skilled 
technologists and keen entrepreneurs, and ultimately produce path-
breaking innovations. The NITRD research performed in universities, 
federal research centers and laboratories, federally funded R&D 
centers, and in partnerships with private companies and nonprofit 
organizations across the country generates continuous interaction, 
information exchange, and feedback in the ecosystem, providing new 
perspectives and insights to both federal and private-sector 
stakeholders. Through its broad reach across the ecosystem, NITRD 
funding also supports the education and training of the Nation's next 
generations of NIT researchers, technical experts, entrepreneurs, and 
IT industry leaders.
---------------------------------------------------------------------------
    \11\  See, for example, Evolving the High Performance Computing and 
Communications Initiative to Support the Nation's Information 
Infrastructure, the National Research Council, National Academy Press, 
Washington DC, 1995; and Assessing the Impacts of Changes in the 
Information Technology R&D Ecosystem, National Research Council, 2009.
---------------------------------------------------------------------------
    The NITRD Program thus supports not only the vitality of the 
innovation ecosystem as a whole but the national NIT talent pool it 
nurtures. We are pleased that the PCAST, in its 2010 review of the 
Program, concluded that NITRD is widely and correctly viewed as 
``successful and valuable,'' and we are working, as noted throughout 
this testimony, to address PCAST recommendations for ways to improve 
our efforts.
    I will now turn to the other questions the Subcommittee posed that 
I have not yet addressed.

NITRD Objectives and Critical R&D Issues

    While maintaining their mission focus, the NITRD agencies make 
every effort through their NITRD Program activities to grapple with the 
most critical NIT R&D problems. Although the NITRD collaborative 
umbrella enables only coordination--not prioritization--of agencies' 
mission activities, each agency faces and responds to the challenge of 
pushing the cutting edge of technological change. To cite just a few 
areas, the NITRD agencies individually and together are investigating 
the implications of cloud computing for data-intensive science and 
high-end computation. Our agencies are also leading the government's 
major research effort to change the balance of power in cyberspace, so 
that legitimate uses are secure and malefactors can no longer attack at 
will. NITRD members are working on critical technical challenges at the 
upper limits of computing power and speed, such as energy conservation, 
nanoscale materials and techniques, and software architectures and 
applications for machines with hundreds of thousands of processors. The 
agencies are also pressing forward with improving software engineering 
for the long-lived, ultra-scale software-based systems that are the 
work horses providing many of the Nation's most vital capabilities 
across all sectors.
    Amid the relentlessly accelerating rate of technological change of 
recent years, we in NITRD are also learning how to be more adept in 
adjusting our coordination emphases to be more responsive. The new 
ventures described above were created and became productive in record 
time for NITRD. Each of these is addressing significant national issues 
that require intellectual contributions from the NIT research community 
and advances in NIT R&D. The new SSGs and subgroups represent a 
different, more flexible model for NITRD collaboration--one in which 
collaborative groups are quickly formed to focus on emerging issues, do 
their work, and then may disband as their topical tasks conclude and 
new issues arise that need attention. NITRD's underlying PCAs will 
continue to exist, because they provide continuity in budget reporting 
over time. But shifting opportunities for short-term coordination 
activities are likely to be the new NITRD norm. These shifts align with 
the recommendations in the 2010 PCAST review of NITRD.
    In my view, the value of the collaboration model in the NITRD 
Program, which involves both mission agencies focused on the ``use 
value'' of their missions and science agencies focused on ``theory 
value,'' is illustrated by the political scientist Donald Stokes in his 
1997 book Pasteur's Quadrant. Stokes defined a two-dimensional array of 
four types of R&D. His dimensions were ``use value'' and ``theory 
value,'' and his point in making four quadrants was that high use value 
does not need to imply low theory value. And that high use-value 
science can generate high theory-value science, just as high theory-
value science can generate high use-value science. Stokes pointed to 
the French scientist Louis Pasteur's groundbreaking work on causes and 
prevention of diseases as having high use value and high theory value 
(such research is said to lie in Pasteur's quadrant).
    Stokes's concept usefully describes the essentially 
multidimensional nature of activities such as NIT research. That is, 
NIT R&D properly involves high use and high theory value. Computer 
science theory can arise from applied science, just as science research 
can arise from computer science research results; it is the constant 
interplay between the pure and the applied R&D sectors that generates 
many of the innovations that astonish us. We need pure research (such 
as computational complexity) and we need use-inspired research (such as 
arises when a mission agency seeks solutions to its science problems).

Research Opportunities and Academic and Industry Inputs

    The research communities in academia and industry contribute to 
NITRD activities in a variety of forms. For example, in the past 12 
months the NITRD agencies have finished their work on two major 
strategic plans--a five-year strategic plan for NITRD and a strategic 
plan for game-changing R&D to secure cyberspace. We are pleased that 
our cybersecurity R&D strategic plan, Trustworthy Cyberspace: Strategic 
Plan for the Federal Cybersecurity Research and Development Program 
will soon be released to the public and the NITRD plan is under review. 
Throughout both development activities, the NITRD agencies reached out 
extensively to engage the private sector in workshops, Requests for 
Input, wikis, and other forums.
    One of the results in moving to include flexible, topic-focused 
coordination groups is that the NITRD Program can more conveniently 
draw upon academic and private-sector expertise across disciplines, 
sectors, and research and engineering domains in order to turn research 
results into practical applications.
    A different form of outreach to the private sector takes place 
under NITRD member DOE/SC's Innovative and Novel Computational Impact 
on Theory and Experiment (INCITE) program. Since 2003, INCITE has 
promoted transformational advances in science and technology by 
competitively awarding large allocations of time on the agency's most 
powerful computing platforms (``leadership class'' systems), as well as 
supporting resources and data storage, to industrial, federal, and 
academic researchers nationwide who lack access to such resources. For 
2011, 57 INCITE awardees received a total of 1.7 billion processor 
hours. \12\
---------------------------------------------------------------------------
    \12\ http://www.doeleadershipcomputing.org/wp-content/uploads/2011/
07/INCITE-IR-FINAL-7-19-11.pdf

---------------------------------------------------------------------------
Importance of Federal NIT Investments

    As discussed above, the history of NIT development has demonstrated 
the crucial role of federal investments. The results of these 
investments have spawned a myriad of technological innovations, novel 
products and communications capabilities, and an entirely new, 
multibillion-dollar economic sector in NIT that has been responsible 
for significant expansion in well-paying job opportunities. The 
multiplier effects of Federal NIT R&D--on both innovation and 
employment--are widely documented, perhaps most famously in the 
National Research Council's ``tire tracks'' graphic illustrating how 
the feedback loops mentioned above operate over time to move research 
discoveries into the marketplace. \13\ There is little doubt that our 
country and the world are moving rapidly into an increasingly digital 
future. We in NITRD concur with PCAST that federal research leadership 
will continue to be an imperative if we are to sustain our preeminence 
in the networking and information technologies that we invented and 
developed.
---------------------------------------------------------------------------
    \13\  In 2003, the National Research Council's Computer Science and 
Telecommunications Board (CSTB) updated the original 1995 tire tracks 
figure from the Evolving the High Performance Computing and 
Communications report in a new report, Innovation in Information 
Technology that summarized eight prior CSTB studies on the subject.

---------------------------------------------------------------------------
Comments on H.R. 2020

    The NITRD Program has benefited for 20 years from Congressional 
authorization and we look forward to this reauthorization. I believe 
that this draft legislation from the last Congress is well focused and 
will continue to aid our activities. Two small changes would, I 
believe, increase its value and focus.

      A. The draft legislation currently calls for a three-year 
cycle for updating the NITRD strategic plan and a two-year cycle for 
advisory committee review. I respectfully suggest that if both 
activities were put on a three-year cycle, there would be better 
linkage between them. The same might be true of a two-year cycle, but 
the 50% increase in reporting activity would not, in my opinion, be 
offset by gains in value to the program.

      B. A new section in the current draft legislation 
highlights cyber-physical systems and information management. As 
discussed above, we affirm the emphasis on cyber-physical systems. We 
respectfully suggest that, as also discussed above, ``big data'' is a 
better phrase than ``information management'' to characterize the 
advances required at this time.

    These comments, I believe, are consistent with the findings of 
PCAST's 2010 review of NITRD.
    Thank you very much for affording me the opportunity to provide 
testimony before the Subcommittee on Research and Science Education of 
the House Committee on Science, Space, and Technology. I will be happy 
to answer any questions you may have.

NITRD MEMBER AGENCIES

    The following federal agencies, which conduct or support R&D in 
advanced networking and information technologies, report their IT 
research budgets in the NITRD crosscut and provide support for program 
coordination:

Department of Commerce (DOC)

      National Institute of Standards and Technology (NIST),

      National Oceanic and Atmospheric Administration (NOAA),

      Department of Defense (DoD),

      Defense Advanced Research Projects Agency (DARPA),

      National Security Agency (NSA),

      Office of the Secretary of Defense (OSD) and Service 
Research Organizations:

        Air Force Research Laboratory (AFRL),

        Army Research Laboratory (ARL),

        Office of Naval Research (ONR).

Department of Energy (DOE)

      National Nuclear Security Administration (DOE/NNSA),

      Office of Science (DOE/SC).

Department of Homeland Security (DHS)

Department of Health and Human Services (HHS)

      Agency for Healthcare Research and Quality (AHRQ),

      National Institutes of Health (NIH),

      Office of the National Coordinator for Health Information 
Technology (ONC).

Environmental Protection Agency (EPA)

National Aeronautics and Space Administration (NASA)

National Archives and Records Administration (NARA)

NITRD PARTICIPATING AGENCIES

    Representatives of the following agencies with mission interests 
involving networking and IT R&D and applications also participate in 
NITRD activities:

Department of Commerce (DOC)

      National Telecommunications and Information 
Administration (NTIA)

Department of Defense (DoD)

      Defense Information Systems Agency (DISA)

Department of Energy (DOE)

      Office of Electricity Delivery and Energy Reliability 
(DOE/OE)

Department of Health and Human Services (HHS)

      Food and Drug Administration (FDA)

Department of Interior

      U.S. Geological Survey (USGS)

Department of Justice (DOJ)

      Federal Bureau of Investigation (FBI)

Department of State (State)

Department of Transportation (DOT)

      Federal Aviation Administration (FAA)

      Federal Highway Administration (FHWA)

Department of the Treasury (Treasury)

      Department of Veterans Affairs

Director of National Intelligence (DNI)

      Intelligence Advanced Research Projects Agency (IARPA)

National Transportation Safety Board (NTSB)

Nuclear Regulatory Commission (NRC)

U.S. Department of Agriculture (USDA)

NITRD SSGs Participating Agencies

Cybersecurity

Department of Commerce (DOC)

      National Institute of Standards and Technology (NIST)

Department of Defense (DoD)

      National Security Agency (NSA)

      Office of the Secretary of Defense (OSD)

Department of Homeland Security (DHS)

Director of National Intelligence (DNI)

Executive Office of the President

      Office of Science and Technology Policy (OSTP)

National Science Foundation (NSF)

Big Data

Department of Commerce (DOC)

      National Institute of Standards and Technology (NIST)

      National Oceanic and Atmospheric Administration (NOAA)

Department of Defense (DoD)

      Defense Advanced Research Projects Agency (DARPA)

      National Security Agency (NSA)

      Office of the Secretary of Defense (OSD) and Service 
Research Organizations

      Air Force Research Laboratory (AFRL)

      Army Research Laboratory (ARL)

      Office of Naval Research (ONR)

Department of Energy (DOE)

Department of Health and Human Services (HHS)

      National Institutes of Health (NIH)

Director of National Intelligence (DNI)

Executive Office of the President

      Office of Science and Technology Policy (OSTP)

National Aeronautics and Space Administration (NASA)

National Science Foundation (NSF)

Health IT

Department of Commerce (DOC)

      National Institute of Standards and Technology (NIST)

Department of Defense (DoD)

      Telemedicine and Advanced Technology Research Center 
(TATRC)

Department of Health and Human Services (HHS)

      Agency for Healthcare Research and Quality (AHRQ)

      Assistant Secretary for Preparedness and Response (ASPR)

      Centers for Disease Control and Prevention (CDC)

      Food and Drug Administration (FDA)

      Indian Health Service (IHS)

      National Institutes of Health (NIH) Office of the 
National Coordinator for Health Information Technology (ONC)

Department of Veterans Affairs (VA)

National Science Foundation (NSF)

Executive Office of the President

      Office of Science and Technology Policy (OSTP)

Wireless Spectrum Technologies

Department of Commerce (DOC)

      National Institute of Standards and Technology (NIST)

      National Telecommunications and Information 
Administration (NTIA)

Department of Defense (DoD)

      Defense Advanced Research Projects Agency (DARPA)

      National Security Agency (NSA)

      Office of the Secretary of Defense (OSD) and Service 
Research Organizations

      Air Force Research Laboratory (AFRL)

      Army Research Laboratory (ARL)

      Office of Naval Research (ONR)

Department of Energy (DOE)

      Idaho National Laboratory (INL)

      Oak Ridge National Laboratory (ORNL)

Department of Homeland Security (DHS)

Department of Justice (DOJ)

Department of Transportation (DOT)

      Federal Aviation Administration (FAA)

Executive Office of the President

      National Economic Council (NEC)

      Office of Science and Technology Policy (OSTP)

Federal Communications Commission (FCC)

National Aeronautics and Space Administration (NASA)

National Science Foundation (NSF)

    Chairman Brooks. Thank you, Dr. Strawn.
    We next recognize Dr. Lazowska for his five minutes.

               STATEMENT OF DR. EDWARD LAZOWSKA,

   DIRECTOR, ESCIENCE INSTITUTE, BILL AND MELINDA GATES CHAIR

    Dr. Lazowska. Thank you, Chairman Brooks, Ranking Member 
Lipinski, the other Members of the Subcommittee for the 
opportunity to speak with you today. My name is Ed Lazowska. I 
am a long-time faculty member at the University of Washington. 
I recently co-chaired the working group of the President's 
Council of Advisors on Science and Technology, charged with 
reviewing the NITRD Program, but I am speaking today as an 
individual and endorsed by the Computing Research Association.
    So I have 10 points that I would like to make in 30 seconds 
apiece, and here they are. First, information technology R&D is 
something that changes the world, and you see this in your own 
lives. You shop through Amazon, you get your movies from 
Netflix, you get books on your Kindle, you get the world of the 
Internet on your iPhone, you learn from Khan Academy, you have 
maps and directions and navigation and routing from Google and 
GPS, you have a Roomba robot vacuum cleaner, you have adaptive 
cruise control on your car. We all benefit from national 
security through information superiority. We all benefit. Your 
committee focuses on dramatic advances in science and 
engineering discovery that are enabled by information 
technology.
    All of these are the results of IT R&D. I just spent the 
previous hour up the hall at a session called ``Deconstructing 
the IPad,'' and it involved computer scientists, physicists, 
device engineers talking about the role of federal research and 
development in all of the technologies that underlie the IPad. 
Apple has done some amazing engineering to produce this 
miraculous device, but every piece of it, the GPS chip, the 
touch screen interface, the CPU, every aspect of that device 
can trace its roots to federally-funded research, and that is 
what your committee is focused on.
    Second, information technology R&D drives our prosperity. 
It is not just the information technology industry. It is 
productivity gains in other sectors because of the use of IT. 
Economists agree that information technology has boosted U.S. 
productivity more than any other set of factors in the recent 
past.
    Third, IT is the dominant factor in American science and 
technology employment. I hope we'll talk later about workforce 
issues, and offshoring is certainly a significant issue, but 
accounting for that, the Bureau of Labor Statistics projects 
that 60 percent of all new jobs in all fields of science and 
engineering in this decade will be for computing specialists. 
That's in this country. All right. So more than all the 
physical sciences, all of the life sciences, all of the social 
sciences, and all other fields of engineering combined. That's 
what they project for computer specialists in the next decade.
    Fourth, federal support is a key part of this, and you 
heard from Dr. Strawn about this. You will hear more from 
others. Every major sector of the IT industry traces its roots 
to the federally-sponsored fundamental research program. It's 
the role of federally-sponsored research to take the long view. 
Industry R&D is the vast majority of it focused appropriately 
on the engineering of the next release for product.
    Fifth, there's huge potential and huge need for further 
breakthroughs. I will defer talking about the specifics of 
that.
    Sixth, many areas of IT R&D are crucial to national 
priorities and national competitiveness, and I am going to give 
you two local examples. Yesterday my young University of 
Washington colleague, Shwetak Patel, won a MacArthur genius 
award for work using machine learning to tell you exactly which 
devices in your home are consuming exactly how much electric 
power from a single device that you plug into a wall outlet 
anywhere in your house. Okay. It is just sort of miraculous in 
terms of incenting better use of energy, better energy 
efficiency.
    A few years ago another young University of Washington 
colleague, Yoky Matsuoka, won a MacArthur genius award for work 
on prosthetics that couple directly to the nervous system. So 
this gives new hope to thousands of returning veterans who are 
impaired, disabled in various ways.
    If you want breakthroughs in energy, in national security, 
in health care, in scientific discovery, in transportation, 
then you need breakthroughs in computer science. That's what 
powers those other examples. There is a set of federal agencies 
that understands this well, and there is a set that needs a 
better appreciation of it, and that is part of Dr. Strawn's job 
in the NCO--to bring those agencies together and cause them to 
pull together.
    Seventh, the Nation is investing far less in IT R&D than is 
shown in the federal budget. Many agencies report the 
acquisition of computer technology to support research in other 
disciplines as part of their NITRD crosscut. This is a 
completely appropriate research expenditure, but it's not IT 
R&D, and we need to improve the reporting so that we know how 
much we are actually spending on this field.
    Eighth, PCAST urged and I personally urge that the Federal 
Government needs a high-level, sustained expert strategic 
advisory committee for information technology R&D, something 
perhaps analogous to the former President's Information 
Technology Advisory Committee.
    Ninth, computer science must be viewed as an essential 
component in STEM education. Every child and every adult needs 
fluency in computing and what we call computational thinking.
    Finally, last, no other field comes close. Here is what 
PCAST said. ``As a field of inquiry, Networking and Information 
Technology has a rich intellectual agenda, as rich of that as 
any other field. In addition, NIT is arguably unique among all 
fields of science and engineering in the breadth of its 
impact.'' That's why your work matters.
    Thanks for the opportunity to share my view, and I, too, 
look forward to your questions.
    [The prepared statement of Dr. Lazowska follows:]
          Prepared Statement of Dr. Edward Lazowska, Director,
            Escience Institute, Bill and Melinda Gates Chair
    Thank you, Chairman Brooks, Ranking Member Lipinski, and the other 
Members of the Subcommittee, for this opportunity to discuss the 
Federal Government's Networking and Information Technology Research and 
Development program. I am pleased to add my perspective on the 
Committee's questions, drawn from nearly 40 years in academia as a 
member of the computing research community, my experience as the 
current Chair of the Computing Research Association's (CRA) Computing 
Community Consortium (CCC), and as a member and chair of many federal 
IT advisory committees--including, most recently, as the co-Chair of 
the Working Group of the President's Council of Advisors on Science and 
Technology (PCAST) to review the NITRD program. However, I present this 
testimony as an informed individual and not as a representative of any 
particular organization, although my comments have the endorsement of 
the Computing Research Association.

Information Technology R&D Changes the World

    The importance of this hearing's topic is hard to overstate. 
Advances in information technology are transforming all aspects of our 
lives. Virtually every human endeavor today has been touched by IT, 
including commerce, education, employment, health care, energy, 
manufacturing, governance, national security, communications, the 
environment, entertainment, science and engineering. We have the 
world's products available to us with the click of a mouse, instruction 
tailored to individual students and delivered from hundreds or 
thousands of miles away, the ability to be productive and connected 
regardless of location, doctors empowered by virtual agents that can 
help navigate subtle drug interactions or diagnose with data rather 
than gut feelings, an emerging intelligent power grid working together 
with smart structures to more effciently utilize power resources, 
advanced robotics that will enable the nation to retain a competitive 
manufacturing sector, government that works more transparently, a 
military that achieves dominance through information superiority, a 
network of friends reachable instantly anywhere around the globe, a 
planet wired with sensors feeding us real-time information about its 
health, movies and music and games that engage all our senses and take 
us to places no previous generation has ever seen, and a science and 
engineering enterprise primed with all the tools and data to enable 
discovery at a pace never before seen--all because of advances in 
computing systems, tools and services enabled by information technology 
research and development.

Information Technology R&D Drives Our Prosperity

    Advances in information technology are also driving our economy--
both directly, in the growth of the IT sector itself, and indirectly, 
in the productivity gains that all other sectors achieve from the 
application of IT. IT R&D creates new industries that create new jobs, 
and transforms existing industries in ways that increase their 
productivity and make them more competitive. In fact, it is this latter 
effect that has had the most profound impact on the economy and the 
Nation's competitiveness. Across every sector of the economy, 
businesses large and small have used IT systems, tools, and services to 
improve their productivity, boost their effciency, and increase their 
economic output to an unprecedented extent. Large companies like 
Walmart and United Parcel Service have used advanced IT tools to track 
and manage inventory on a minute-by-minute basis. Companies like Boeing 
and Procter & Gamble use high-performance computing in applications 
ranging from designing super-effcient airframes to modeling the airflow 
over potato chips on a production line to minimize breakage and loss. 
Small manufacturers use IT to do virtual prototyping, avoiding costly 
prototype construction and allowing them to compete with much larger 
firms for lucrative manufacturing contracts. And sites like Etsy and 
eBay allow individual artists or entrepreneurs to set up virtual 
storefronts and sell to the world. Advances in IT empower U.S. 
businesses, augment their competencies, and enable them to compete in 
an increasingly global economy. The development and application of IT-
related systems, services, tools and methodologies have boosted U.S. 
labor productivity more than any other set of forces in recent decades.

Information Technology Is the Dominant Factor in American S&T 
        Employment

    Given information technology's influence in so many sectors of our 
lives, it should not be surprising that demand for IT workers is 
strong. Indeed, as the PCAST review of the NITRD program released last 
year noted, ``All indicators--all historical data, and all 
projections--argue that [Networking and Information Technology (NIT)] 
is the dominant factor in America's science and technology employment, 
and that the gap between the demand for NIT talent and the supply of 
that talent is and will remain large.'' \1\ Bureau of Labor Statistics 
projections indicate that more than 60% of all new jobs in all fields 
of science and engineering in the current decade will be for computer 
specialists. Increasing the number of graduates in IT fields at all 
levels should be a national priority; the NITRD program should increase 
its focus on computer science education, from kindergarten through 
higher education, as one way to help meet that goal.
---------------------------------------------------------------------------
    \1\  Designing a Digital Future: Federally Funded Research and 
Development in Networking and Information Technology. Report to the 
President and Congress, President's Council of Advisors on Science and 
Technology, December 2010.

Federal Support Is a Key Part of the Vibrant Ecosystem that Drives IT 
---------------------------------------------------------------------------
        Innovation

    The advances in IT that have had such a profound effect on every 
aspect of our lives are driven by innovation that itself is the product 
of a vibrant research ecosystem--an ecosystem comprised of university 
research in academic departments, industrial research facilities, 
federal research labs, industrial development organizations, and the 
people and ideas that flow between them. The National Research Council 
has called this ``an extraordinarily productive interplay'' and the 
President's Information Technology Advisory Committee (PITAC) 
emphasized the ``spectacular return'' on the federal investment made as 
part of this ecosystem. \2\
---------------------------------------------------------------------------
    \2\  Information Technology Research: Investing in Our Future. 
Report to the President, President's Information Technology Advisory 
Committee, February 1999.
---------------------------------------------------------------------------
    The federal role in this system is largely limited to investments 
in long-term, early stage scientific research, typically at U.S. 
universities. This research often occurs many years,or even decades, 
before a product is developed for the marketplace.
    The great majority of industry-based research and development is of 
a fundamentally different character than university-based research. 
Industry-based research and development is, by necessity, much shorter 
term than the early-stage research performed in universities. It tends 
to be focused on product and process development, areas which will have 
more immediate impact on business profitability. Industry generally 
avoids long-term research because it entails risk in several 
unappealing ways. First, it is hard to predict the outcome of 
fundamental research. The value of the research may surface in 
unanticipated areas. Second, fundamental research, because it is 
published openly, provides broad value to all players in the 
marketplace. It is difficult for any one company to ``protect'' the 
fundamental knowledge gleaned from long-term research and capitalize on 
it without everyone in the marketplace having a chance to incorporate 
the new knowledge into their thinking. Those companies that do make 
significant fundamental research investments tend to be the largest 
companies in the sector. Their dominant position in the market 
increases the likelihood that they benefit from any market-wide 
improvement in technology basic research might bring. For example, IBM 
and Microsoft are among the companies that invest the largest 
proportion of their R&D expenditures on research looking out more than 
one product cycle, but at Microsoft, as reported by PCAST, it is 
estimated that this still constitutes less than 5% of total R&D. At 
most other companies, it is far less. University research does not 
supplant industry research, or vice versa.
    An example might be instructive here. Apple's IPad is a seemingly 
miraculous little machine. Available for about $500, it's a sleek, thin 
little slab of glass and metal that sits darkly in a purse or a pocket, 
then comes to life with a button push and a swipe of a finger, quickly 
figures out where it is, and then connects itself to the largest 
collection of humanity's knowledge ever assembled. It's a remarkable 
confluence of technologies--processing capability powerful enough to 
have appeared on the list of the world's fastest supercomputers as 
recently as 1994, a sensor suite (global positioning system, compass, 
accelerometer, microphone, camera, light sensor) robust enough to allow 
it to know where it is and what it's looking at, and an interface 
revolutionary in its ease of use. These technologies have enabled some 
truly game-changing capabilities--applications that allow turn-by-turn 
directions, or the ability to translate signs in a foreign language 
just by pointing its camera at them, or truly high-speed, ubiquitous 
connectivity to the power of the Internet, instantly, almost anywhere 
in the world.
    What Apple has managed to do to bring these technologies together 
and meld them in a seamless way to enable these applications has been 
nothing short of remarkable. Without exception, however, all these 
technologies have their roots in early-stage scientific research, and 
all bear the stamp of federal government support.
    Take, for example, the revolutionary multi-touch IPad interface--
the pinch-to-shrink, swipe-to-scroll, twist-to-rotate gestures that 
make a tablet like the IPad intuitive and very easy to use. All were 
born out of university research, largely funded by the Federal 
Government, conducted as early as the late 1960s and early 1970s. In 
fact, in 1998, researchers at the University of Delaware, whose work 
had earlier been enabled by research funding from the National Science 
Foundation, established a company called FingerWorks to market an early 
touch-screen keyboard based on their research. In 2005, Apple bought 
the company and its technology, then spent over two years adapting it 
for the first iPhone.
    A similar case can be made for the processor--the brain of the 
device--which has its roots in the design of the original integrated 
circuit back in 1958, by a young Texas Instruments engineer named Jack 
Kilby. But it's a far cry from that original design to the modern chip 
that powers the IPad. Industry research at TI and Fairchild, and later 
at IBM, Intel and others was obviously important in moving development 
along, but just as important was research at U.S. universities, on 
Reduced Instruction Set Computing (RISC) and Microprocessor without 
Interlocked Pipeline Stages (MIPS) technologies, as well as Very-Large-
Scale Integration (the process of creating integrated circuits by 
combining thousands of transistors into a single chip)--technology that 
put computer design in the hands of computer system architects (and 
graduate students) rather than only in the hands of engineers and 
technicians in costly chip fabrication plants. Federal investment in 
research (through DARPA and increasingly NSF) and government-industrial 
partnerships like SEMATECH were crucial in catalyzing research across 
institutions, accelerating the pace of innovation--and work at 
universities in particular helped generate the people and ideas that 
fueled industry's advancements.
    It is possible to draw similar timelines for all the other key 
technologies in the IPad. This is not to diminish the accomplishment of 
Apple--on the contrary, what Apple has done has been to blend these 
technologies into a harmonious whole in a way that maybe only Apple 
could do. But it highlights the crucial role of early-stage research, 
in many cases supported by the Federal Government (and often only by 
the Federal Government), in enabling world-changing innovation. And it 
shows that federal support for early stage research is truly an 
investment--an investment that has a history of demonstrating 
extraordinary payoff in the explosion of new technologies that have 
touched nearly every aspect of our lives, and in economic terms--in the 
creation of new industries and literally millions of new jobs.

There Is Tremendous Potential--and Tremendous Need--for Further 
        Breakthroughs

    The history of innovation in computing is impressive, but the 
future opportunities are even more compelling. Research in the future 
of networking, revolutionizing transportation, personalizing education, 
powering the smart grid, empowering the developing world, improving 
health care, enabling advanced manufacturing and driving advances in 
all fields of science and engineering are all compelling challenges 
well suited to advancements in IT. Indeed, without continued progress 
in computing research, our ability to address key national and global 
priorities in energy and transportation, education and life-long 
learning, health care, and national and homeland security will be 
seriously constrained.

Many Areas of IT R&D Are Crucial to National Priorities and National 
        Competitiveness

    In its 2010 report Designing a Digital Future, PCAST identified 
three areas of research that the Council felt were ``particularly 
timely and important.'' I support the Council's recommendations. They 
called for:

      A national, long-term, multi-agency research initiative 
on NIT for health that goes well beyond the current national program to 
adopt electronic health records.

      A national, long-term, multi-agency, multi-faceted 
research initiative on NIT for energy and transportation.

      A national, long-term, multi-agency research initiative 
on NIT that assures both the security and the robustness of cyber-
infrastructure.

    In addition, the Council identified a broader set of research 
frontiers of the field that require increased focus from NITRD 
agencies, including:

      A broad multi-agency research program on the fundamentals 
of privacy protection and protected disclosure of confidential data.

      A collaborative research program that augments the study 
of individual human-computer interaction with a comprehensive 
investigation to understand and advance human-machine and social 
collaboration and problem-solving in a networked, on-line environment.

      Fundamental research in data collection, storage, 
management, and automated large-scale analysis based on modeling and 
machine learning.

      Research in advanced domain-specific sensors, integration 
of NIT into physical systems, and innovative robotics in order to 
enhance NIT-enabled interaction with the physical world.

    It is critical to recognize that many areas of IT are now equal in 
importance to high performance computing (HPC) as measures of our 
nation's competitiveness. Twenty years ago, at the time of passage of 
the High Performance Computing and Communications Act of 1991 (which 
established the modern NITRD program), it was appropriate that much of 
the focus of the federal effort in computing was on the importance of 
HPC to scientific discovery and national security. Today, many other 
aspects of IT have risen to comparable levels of importance. Among 
these are the interactions of people with computing systems and 
devices; the interactions between IT and the physical world (e.g., 
robotics); large-scale data capture, management and analysis (critical, 
today, to scientific discovery and national security); systems that 
protect personal privacy and sensitive confidential information; 
scalable systems and networking; software creation and evolution; and 
critical infrastructure protection (e.g., the financial system, the 
power grid, the air traffic control network). World leadership in all 
of these areas is crucial to our nation's security and prosperity.

The Nation Is Investing Far Less on IT R&D than Is Shown in the Federal 
        Budget

    One of the difficulties of assessing the adequacy of the federal 
investment in various areas of IT R&D is the ambiguity of the data 
about IT R&D investments reported by the various agencies participating 
in NITRD. PCAST found that much of what gets reported by NITRD agencies 
represents spending on IT that supports research in other fields--such 
as computing clusters for scientists in other fields--and not spending 
on researchin information technology. In some cases, the discrepancy in 
reporting leads to a dramatic over-reporting of IT R&D investments by 
the agencies: at one major NITRD agency, PCAST estimated that only 
between 2 percent and 11 percent of reported NITRD expenditures truly 
represented investments in IT R&D. I share PCAST's concern that ``by 
leading policymakers to believe that we are spending much more on such 
activities than is actually the case, this discrepancy contributes to a 
substantial, systematic underinvestment in an area that is critical to 
our national and economic security.''

The Federal Government Needs High-Level, Sustained, Expert Strategic 
        Advice on ITR&D

    Another key recommendation contained in the PCAST report with which 
I concur is the call for the establishment of a ``high-level standing 
committee of academic scientists, engineers, and industry leaders 
dedicated to providing sustained strategic advice in NIT.'' Given the 
pace of innovation and change within the field, the challenge of its 
multi-disciplinary, problem-driven research, and the size and scope of 
the federal investment, having sustained guidance from a free-standing, 
independent advisory committee seems crucial to NITRD's success. I was 
pleased to see recognition of this in H.R. 2020, and I feel it is 
imperative that the recommendation of the PCAST report be implemented.

Computer Science Must Be Viewed as an Essential Component of Science, 
        Technology, Engineering and Mathematics (STEM) Education

    As I noted above, the workforce needs of the IT fields going 
forward demand a sustained effort to increase the number of students 
going into computing fields. National security needs will require that 
a large number of those students be American citizens. In addition, 
participants in many other workforce fields will need IT knowledge and 
skills. Making progress on this effort will require reversing trends 
not just in computing, but across the STEM disciplines. I am pleased 
that PCAST has called for the National Science and Technology Council's 
Committee on STEM Education to exercise strong leadership to bring 
about fundamental changes in K-12 STEM education in the U.S. Among 
these changes has to be the incorporation of computer science as an 
essential STEM component. As they note, ``fluency with NIT skills, 
concepts and capabilities; facility in computational thinking; and an 
understanding of the basic concepts of computer science must be an 
essential part of K-12 STEM education.'' Groups like ACM's Education 
Policy Committee have expended great effort to get computer science 
recognized as a key part of the K-12 curriculum, but must be met with 
more acceptance if we are to meet the needs of our information-driven 
economy now and in the future.

In Some Areas, H.R. 2020 Did Not Go Far Enough

    As co-Chair of the Computing Research Association's Government 
Affairs Committee back in 2009, I joined in endorsing the passage of 
H.R. 2020, the Networking and Information Technology Research and 
Development Act of 2009. I believe the Act would make the NITRD program 
stronger by enacting several of the recommendations of PCAST. In 
particular, I was pleased that the NITRD Act included a requirement 
that the NITRD program undergo periodic review and assessment of the 
program contents and funding, as well as develop and periodically 
update a strategic plan--both necessary in helping ensure the 
significant federal investment in IT R&D is used as effectively as 
possible. This review and assessment is best done by an independent 
standing advisory committee composed of experts from academia, industry 
and government. As noted earlier, the creation of such a committee is 
essential.
    I do not believe the Act went far enough in addressing the nation's 
IT workforce and education needs. As CRA noted in a joint letter with 
the Association for Computing Machinery and the National Center for 
Women and Information Technology back in March 2009, we felt it is 
critical that federal efforts to educate young people in computer 
science improve, and that investments recognize that all racial, gender 
and socioeconomic groups are crucial to the continued health of and 
future innovations in the computing field. The three organizations made 
four specific recommendations for the bill, which I support:

      Promote computing education, particularly at the K-12 
level, and increased exposure to computing education and research 
opportunities for women and minorities as core elements of the NITRD 
programs;

      Require the NITRD program to address education and 
diversity programs in its strategic planning and road-mapping process;

      Expand efforts at NSF to focus on computer science 
education, particularly at the K-12 level through broadening the Math 
Science Partnership program; and

      Enlist the Department of Education and its resources and 
reach in addressing computer science education issues.

    Conclusion: Federal Investment in Information Technology R&D Has 
Yielded, and Will Continue to Yield, Extraordinary Payoff

    Computing research--networking and information technology R&D--
changes our world, drives our prosperity, and enables advances in all 
other fields.
    The Federal Government has played an essential role in fostering 
these breathtaking advances. The federal investment in computing 
research is without question one of the best investments our Nation has 
ever made. The payoff has been an explosion of new technologies that 
have touched nearly every aspect of our lives, and the creation of new 
industries and literally millions of new jobs.
    The future is bright. There is tremendous opportunity--and 
tremendous need--for further breakthroughs. The Federal Government's 
essential role in fostering these advances--in supporting fundamental 
research in computing and other engineering fields--must continue.

    Chairman Brooks. Thank you, Dr. Lazowska.
    Our next witness is Dr. Sproull.

    STATEMENT OF DR. ROBERT SPROULL, DIRECTOR OF ORACLE LABS

    Dr. Sproull. Good afternoon. I am Robert Sproull, recently 
retired as Director of Oracle Labs. I want to thank Chairman 
Brooks and Ranking Member Lipinski and Members of the 
Subcommittee for an opportunity to appear before you today to 
offer an industrial perspective. While I do not represent any 
specific company, most of my career has been doing research or 
managing research with industrial labs; Xerox Palo Alto 
Research Center, Sun Microsystems Labs, and most recently 
Oracle. But I have also been a university researcher on 
federally-funded projects, an advisor to high-tech venture 
capital investors, and a researcher in a federal laboratory.
    My main point is that industrial research and innovation 
alone will not sustain the extraordinary advances of the IT 
economy that the two preceding witnesses have described. This 
growth and advance over the last 50 years have depended on 
high-risk, high-reward, long-term research, mostly performed in 
academia and funded by the U.S. Government. Industry works 
closely with academic research so as to harness their findings 
and expertise as essential ingredients in its offerings.
    The NIT economy is a complex ecosystem in which government, 
industry, and academia interact closely. Industry excels at 
developing and producing complex products, incremental 
innovations, compelling product designs such as the IPad, and 
global markets. Academia excels at high-risk research on 
fundamental problems with uncertain economic payoffs, and 
venture capital funding propels to market product-ready 
technologies that might be ignored or even fought by large 
companies. The biggest payoffs emerge from interactions among 
all of these groups.
    The ecosystem has produced extraordinary results. Its 
complex behavior is sketched in the tire tracks diagram that 
Dr. Strawn referred to, and I will present it to you for your 
enjoyment. This is a slide that explains how to interpret it. 
Time will go from left to right. This is the portable 
communication technology characterized by cell phones. 
University research activity is depicted with a red horizontal 
line. Industrial research activity is shown with blue lines, 
and notice interactions between them. You will see even more 
interactions in a moment, depicted by the black arrows. And 
emerging products are depicted with dashed black lines, and 
when product streams reach a billion dollars, they are shown by 
solid green lines.
    So there are 19 sectors of the IT economy. This was 
published in the diagram that was produced in 2003. Each of 
these sectors yields revenues of over a billion dollars each. 
We have a new update underway to illustrate even more recent 
successes such as Internet search and others in social 
networking.
    Please note that the path to a billion dollar business is 
not a simple progression from fundamental research to applied 
research to development to delivery. Nor does a single idea or 
breakthrough suffice to build an industry. Rather, dramatically 
new capabilities build on an accumulation of many varied 
research and innovation results. It may take 15 years or more 
to develop the technologies and markets of a billion dollar 
business.
    The ecosystem depends on research, especially long-term 
fundamental research. This research is high risk and 
unpredictable. It is impossible to predict the degree of impact 
of a research result or how it ultimately may be used in 
products. The long diagonal lines in the diagram show only some 
of the cases where work in one area became essential in another 
area.
    Government-funded long-term fundamental research has played 
an essential role in each of the trajectories depicted in this 
diagram. For example, ARPA in the '60s recognized that 
information technology could address many defense problems and 
undertook programs of long-term research to improve its 
effectiveness, and they are responsible for many, especially on 
the left-hand side, of the early technologies.
    Today, solutions to many more problems facing the 
government will depend critically on NITRD techniques. National 
priorities in energy, transportation, health, and cybersecurity 
all depend on NIT and will benefit from long-term research. 
These priorities will also require short and medium-term 
investments, as well as a great deal of routine IT, but we must 
not let these components undercut long-term investments. As 
ARPA and other agencies have shown, it's the long-term research 
that leads to extraordinary advances.
    As recommended in the PCAST report, the NITRD Program 
should be expanded as necessary to match the broadening scope 
of NITRD investments made by the Federal Government. Although 
the program was started to coordinate high-performance 
computing investments, the newer priorities dramatically 
increase the scope of federal NRD, NITRD investments and 
coordination requirements.
    I have been extremely fortunate to have been part of the 
research community of the NITRD ecosystem. It's been exciting 
and rewarding, and it remains so. The fact that we have come so 
far does not reduce the challenge or potential impact of 
research problems we face today.
    Thank you.
    [The prepared statement of Dr. Sproull follows:]
   Prepared Statement of Dr. Robert Sproull, Director of Oracle Labs

    Thank you, Chairman Brooks, Ranking Member Lipinski, and the other 
Members of the Subcommittee for this opportunity to discuss the Federal 
Government's Networking and Information Technology Research and 
Development program. I am pleased to offer my perspective on your 
questions based on more than 40 years of experience doing or managing 
computing research in academia and industry and also advising high-
technology venture capital investors. Among other roles, I currently 
serve as the Chair of the National Research Council's Computer Science 
and Telecommunication Board (CSTB), and recently retired from Oracle as 
the Director of Oracle Labs. This is an applied research laboratory 
first started by Sun Microsystems in 1990 and retained by Oracle when 
they acquired Sun in 2010. I present today's testimony as an informed 
individual and not as a representative of any particular organization.

Introduction

    Extraordinary economic and societal benefits have exploded from the 
U.S. NITRD ecosystem, which is a complex interplay of government, 
academia, and industry that dates back more than 40 years. Some of the 
technologies themselves have improved extraordinarily, such as the 
price/performance of microprocessors; equally, new markets have grown 
explosively as networking infrastructure and low-cost electronics have 
enabled innovative products and businesses. I will describe below some 
of the aspects of this ecosystem, especially the importance of 
fundamental research and the interplay of government, academic, and 
industrial roles.
    I wish to stress at the outset, however, that this ecosystem would 
not have been born, nor would it be successful today, without a 
vigorous, thoughtful strategy of federal investment in fundamental 
research in NIT. Especially important in the early days were programs 
of long-term research sponsored by NSF and ARPA. An important milestone 
was the High Performance Computing Act of 1991, which recognized the 
importance of high-performance computing to federal missions, 
especially those of Defense and Energy. But as IT technology itself 
became more pervasive in the U.S., signaled most vividly by the 
blossoming of the World Wide Web in 1993, a wide class of NIT 
technologies became critical to short- and long-term requirements of 
many more federal agencies. The Act and its research coordination role 
were appropriately extended to address the expanded set of challenges. 
This extension in scope must continue: today, NIT's role in national 
security, national competitiveness, and national priorities is far 
broader than high-performance computing alone.

NITRD Goals

    The nation has identified advances in energy, transportation, 
health, and cybersecurity as important national priorities. I concur 
with the PCAST NITRD Working Group, on which I served, that these are 
important drivers where NIT research and innovation can make enormous 
contributions, and with the PCAST report recommendation to expand 
NITRD's purview as necessary to address these areas. H.R. 2020 is an 
excellent first step, identifying cyberphysical systems in particular 
for more attention. The recently announced National Robotics Initiative 
is a concrete example of investing in cyberphysical systems research. 
But there is an even wider need for cyberphysical systems in achieving 
national priorities, for example as part of controllers and systems 
that achieve efficiencies in energy and transportation, and for 
monitoring patient health. Indeed, the national priorities show a broad 
panorama of areas, including high-performance computing, in which NITRD 
investments will be essential.

Sun Microsystems' Research Lab, an Industrial Contributor to the NITRD 
        Ecosystem

    Sun Microsystems was founded in 1982 to build advanced computer 
workstations, based on results of research conducted primarily at 
Stanford, Berkeley, and Bell Laboratories. In 1990, Sun created a 
research laboratory. I was a founding member and eventually became its 
director. When Oracle acquired Sun in 2010, they retained the lab as a 
way to start Oracle Labs. I retired from Oracle earlier this year.
    I characterize the lab as an ``applied research lab,'' in that most 
of its research projects, though risky, have medium-term objectives 
(e.g., less than three years) that, if successful, would have a 
significant impact on a Sun product or product line. Our job is to 
selectively explore risky ideas and reduce their risk to a level that 
would be acceptable to an engineering team. Ideally, our research team 
would then transfer to the engineering organization, carrying its ideas 
and insights into a larger engineering team. We like to say that 
``technology transfer is a contact sport,'' meaning that the most 
effective transfers from research to engineering are those that 
transfer people.
    The lab was deliberately kept small, with a budget of about 2% of 
Sun's total R&D budget. SunLabs hires mostly Ph.D.s in computer science 
and engineering fields, but also high caliber college graduates in 
those fields. When the lab started, our CEO, ScottMcNealy, explicitly 
asked us to be ``eyes and ears'' for Sun, to participate in the global 
IT research community, to learn from it, and to contribute to it. Our 
researchers are nationally and internationally known, attending and 
presenting papers at international conferences.
    SunLabs does very little fundamental or long-term research. An 
applied research project might develop broadly applicable results, but 
that is not its principal objective. In order to import a broad range 
of fundamental new ideas, we pay careful attention to academic 
researchers and their results, as McNealy requested.
    Sun evolved a system of ``collaborative research'' with academic 
partners. We would contribute money or equipment to an established 
university research project that we judged might be able to contribute 
to Sun's technologies. Then our researchers would interact closely with 
those in the university. We encouraged academic researchers and 
graduate students to work with us at Sun, as consultants or student 
interns, to learn from their ideas--again through people. For example, 
Sun's embracing of Reduced Instruction Set (RISC) processor 
technology--a technology behind most computer processors in use today--
was accelerated by collaborating with the RISC research group at U.C. 
Berkeley and by consulting help from its principal investigator, Prof. 
David Patterson. This model served Sun well, and helped us sustain 
innovation at a time of rapid technological change. These collaborative 
interactions with academia also allowed us to present challenging Sun 
problems to academics and thus influence academic research agendas.
    Though Sun Labs managed almost all the research projects at Sun, it 
was responsible for only a fraction of Sun's innovations. The product 
engineering organizations, developing both hardware and software, 
routinely innovate. For example, Sun is famous for introducing in 1984 
the ``network file system'' (NFS), which allowed computers to share 
files over a computer network. Though innovative, its development was 
not the direct result of research.
    Incidentally, I dislike the word ``breakthrough,'' because it is 
too often assumed that breakthroughs are the only objective of research 
and stem only from research, especially fundamental research. To the 
contrary, high-impact innovations can emerge in many ways, and 
sometimes the principal reason for the high impact--and thus perhaps 
the perception of a ``breakthrough''--may simply be a sharply lower 
price or rapid market penetration. But these dramatic advances usually 
depend on much varied research, much incremental, perhaps some 
revolutionary, and often far earlier than the apparent 
``breakthrough.''

The NITRD Ecosystem--The Big Picture

    As part of an early assessment of the High-Performance Computing 
Act of 1991, a study by the National Research Council developed a 
graphic presentation known as ``the tiretracks diagram'' to illustrate 
some of the features of the complex interactions among government, 
academic, and industrial players that lead from early research to 
several billion-dollar subsectors of the IT economy. The graphic is 
attached below.
    The graphic charts the development of technologies from their 
origins in industrial and federally-supported university R&D, to the 
introduction of the first commercial products, through to the creation 
of billion-dollar industries and markets. The principal features of the 
NITRD ecosystem that this diagram illustrates are:

      Contributions are made by universities (usually federally 
funded) and industry, in varied orders and magnitudes. Ideas and people 
often contribute to different paths; there are frequent flows from 
academia to industry and vice versa. There is no direct path from 
research to impact.

      Initial research often takes a long time to pay off; 15 
years is typical.

      Research often pays off in unanticipated ways: 
developments in one sector often enable advances in another, often 
serendipitously.

      Innovations occur at all points along technology 
trajectories, not only in research settings.

      University and industry research are different: 
university research favors long-term fundamental problems, while 
industry generally focuses on the next product cycle or two (at most a 
few years). Results of university research are public and available to 
all, creating a challenge for industry uptake.

    The original diagram produced by the NRC in 1995 identifies nine 
billion-dollar sectors. The updated diagram produced in 2005 shows 19 
billion-dollar subsectors of the IT economy, each of which bears the 
clear stamp of federal investment, usually in high-risk research with 
uncertain commercial application or payoff. The Council is at work now 
producing the next version of the chart, and they are likely to 
identify several new billion-dollar subsectors--search and social 
networking, for example--that have emerged just since 2003.

The NITRD Ecosystem--a Java Example

    In the late 1990s, Sun Microsystems introduced the new Java 
programming language. Although new programming languages are rarely 
adopted widely, Java became popular because of its ability to run 
robustly on many different computer types and because of its modern 
design, especially features that reduced some of the tedious chores of 
programming; that is, it increased programmer productivity. Many IT 
staffs and product developers embraced Java to program their products 
and services. Today, Java is often taught to high school students as 
their first programming language. One of the reasons Oracle acquired 
Sun is that much of Oracle's product suite had come to depend on Java.
    Java was designed by James Gosling in 1991 as part of a research 
project exploring ways to use graphical point-and-click user interfaces 
to control televisions, set-top boxes, kitchen appliances, and other 
consumer gear. This product objective did not succeed, but Java found a 
foothold in the mid `90s as a way to program Web browsers to create 
animated and interactive experiences. Early releases for this purpose 
reached a large number of programmers, the language became quite 
popular, and Sun went on to develop versions for conventional computer 
systems (as opposed to browsers).
    Java's design and implementations draw heavily on preceding 
research in many areas. Object-oriented programming languages had long 
been studied by industry and academia. Especially important was the 
SmallTalk language, developed by Xerox researchers in the 1970s, 
inspired by a language named Simula, developed by Norwegian researchers 
in the 1960s. Research to speed up execution of SmallTalk programs 
became a popular focus of university research on a wide range of 
fundamental language implementation problems. For example, a graduate 
student at Stanford, Urs Holzle, developed a revolutionary way to 
generate fast code for the Self language, a close kin of SmallTalk. He 
and others founded a startup, Animorphic, to exploit this technology in 
a commercial SmallTalk system to compete with other SmallTalk offerings 
from a small group of startups (none of which survive today).
    When Java became popular, the Animorphic team quickly retargeted 
their work to a Java implementation, judging that it would have greater 
commercial value than SmallTalk. Sun, looking for ways to speed up its 
own Java implementations, bought Animorphic, and the team incorporated 
their technology into Sun offerings, where it became known as ``HotSpot 
technology.''
    This is but one of many threads from research to product that 
contributed essential components to Java technology.
    This detailed glimpse of one of Java's technology paths shows the 
NITRD ecosystem at work. The players are global; there are complex 
interactions among industry andacademic researchers; people and ideas 
flow rapidly; startups play an important role whether or not they 
ultimately succeed as standalone businesses; fundamental innovations 
may take a long time to reach mainstream products; a commercial success 
will track back to countless research projects and results, many of 
them funded by the Federal Government. The ecosystem collapses without 
federal support of fundamental research.

Characteristics of the Ecosystem

    Using the term ``ecosystem'' to describe the complex interactions 
among participants in NITRD activities may seem a stretch, but the term 
is apt. There are many distinct players, with varied but blurry roles, 
and complex dependencies. As we've seen, an IT product depends on other 
NITRD activities in complex ways akin to the dependencies in a 
biological system. Different players perform complementary roles. Long-
term academic research provides new results whose impact cannot be 
predicted at the time. Industry amplifies these results through its own 
applied research and product development processes. The overall health 
of the system depends both on funding from government and from revenues 
received for products and services offered by healthy IT businesses.
    Like a biological ecosystem, the NITRD ecosystem could be disrupted 
or damaged inadvertently. The NRC report Assessing the Impacts of 
Changes in the Information Technology R&D Ecosystem addressed exactly 
this concern in 2009. It concludes that federal investment in research 
is essential and dangerously thin. It points to the importance of 
venture funding. It also points out that the ecosystem includes 
customers: ``The most dynamic IT sector is likely to be in the country 
with the most demanding IT customers and consumers.'' Thus, for 
example, improving U.S. broadband networking is essential to creating 
the demand to develop world-class innovative services.
    The most dangerous and least visible threat to the ecosystem is 
that we all focus on short-term research and payoffs, thus 
underinvesting in the long-term research that may lead to extraordinary 
technical advances and returns.
    Investing in fundamental research is risky, and the amount and 
character of payoff cannot be predicted. But federal sponsors have an 
excellent record of directing fundamental research, in concert with the 
research community itself. DARPA, for example, pursues military needs, 
and its long-term vision and investments have resulted in fundamental 
and high-payoff results, such as interactive computing, networking, and 
RISC microprocessors. NSF's recognition that digital libraries would 
become important led to high payoff in search engines, which can be 
seen in today's search offerings from Google, Yahoo, and Microsoft. The 
wisdom of long-term federal research investments is evident in the 
productive ecosystem they have spawned.
    As I remarked in my introduction, the national goals in energy, 
transportation, health, and cybersecurity are excellent guides for 
today's NITRD research investments. Who knows what billion-dollar NIT 
industries may emerge from research toward these goals?

The Research Workforce

    I want to offer a few comments on the workforce available to 
industrial research groups. Note that this is a small subset of the 
overall IT workforce. I offer these comments to emphasize the varied 
nature of skills and training in the workforce.
    At SunLabs, we hired mostly Ph.D.s, many fresh out of graduate 
school. Candidates come from all over the world. In most cases we know 
of students finishing their degrees because we have ongoing 
collaborations with their professors or the students themselves. In all 
cases, we seek candidates who have demonstrated research skills in 
areas aligned with the research project we are staffing. For example, a 
project to explore new ideas in building Java ``virtual machines'' 
seeks candidates who have built virtual machines, garbage collectors, 
or other programming-language artifacts as part of their academic 
research. Consistent with our objective of transferring technology by 
transferring people, we seek researchers adept at building systems and 
willing to join engineering teams.
    Although we expect staff to work from one of our two lab locations 
in the U.S., this is not always possible. Some candidates have family 
constraints that prevent a move. Some foreign nationals cannot obtain 
visas, or must work from abroad until a visa can be obtained. The 
international Internet makes remote work (``dispersed R&D'') possible, 
but not preferable. Location still matters, but as networking improves, 
it matters less.

Understanding Federal Research Investments

    The PCAST Working Group that examined the NITRD program had trouble 
determining the levels of research investment in different areas 
because of difficulties in labeling and measuring expenditures. In 
industry, we make clear distinctions between different kinds of 
investment in IT, in part so that the investments can be balanced 
appropriately.
    First, support of fundamental and applied research. The goals of 
this work are too risky to depend on results to meet customer or market 
needs.
    Second, investments to develop new IT products and services, some 
for sale and some for internal use. These developments may be routine 
or highly innovative, but the development itself is not very risky: 
schedules, milestones, tests, and periodic releases characterize the 
work.
    Finally, investments in NIT infrastructure that support all parts 
of the business, including the two items mentioned above, NIT research 
and NIT development. These investments are usually the least risky and 
innovative of all, and are usually driven by estimates of computing and 
networking capacity needed. As NIT infrastructure becomes necessary to 
support almost all business activities, these expenses are similar to 
those for space and utilities, and are accounted as an overhead for the 
activities they support.
    Federal budget reporting makes it difficult to distinguish these 
three classes of investment. Infrastructure, in particular, should not 
be characterized as an NIT R&D investment unless it supports NIT R&D 
itself. For example, a Web server that provides citizen access to an 
agency's database is not an NITRD investment, though it is a use of 
NIT. While distinctions between research and development (the first two 
categories) are sometimes blurry, the appropriate measure is one of 
risk and reward. It is the risky but potentially broadly valuable 
investments that should be classified as research.
    NITRD program coordination would be improved if the participating 
NITRD agencies were required to report their R&D expenditures more 
clearly. To coordinate research activities, actual research investments 
must be reported. Either better categories such as the ones I've 
outlined or more thorough line-item reporting would help. This is an 
area where a bill such as H.R. 2020 could contribute.

Conclusion

    The NITRD program has demonstrated an ability to coordinate federal 
investments in essential research, starting with high-performance 
computing and now extending to a broader set of national goals. The 
challenge now, for sponsors and researchers alike, is to make the case 
to an increasingly broad set of NITRD mission agencies that long-term 
investments in fundamental NITRD research lead to large rewards for 
their missions and for the nation.


    Chairman Brooks. Thank you, Dr. Sproull.
    Next our final witness is Dr. Robert Schnabel.

               STATEMENT OF DR. ROBERT SCHNABEL,

                  DEAN, SCHOOL OF INFORMATICS,

           INDIANA UNIVERSITY, SCHOOL OF INFORMATICS

    Dr. Schnabel. Chairman Brooks, Ranking Member Lipinski, and 
distinguished Members of the Subcommittee, thank you so much 
for the opportunity to speak with you today. I'm Bobby 
Schnabel. I'm Dean of the School of the Informatics at Indiana 
University. In this testimony I'll represent both my university 
role and also my national roles in computing education with ACM 
and with the industry non-profit coalition Computing in the 
Core.
    I will speak primarily to the education and workforce 
issues that have been mentioned and which are essential to 
keeping our Nation's innovation and economy strong.
    My fundamental points are these. The workforce demand in 
computing and IT is high, it is growing, and it greatly exceeds 
our current and projected capacity. To meet that demand we will 
need to educate both a greater number and a greater--more 
diverse set of people in computer science and IT disciplines. 
And to reach the required enrollments at the university levels 
we really need to bolster computer science education at the K-
12 level, where, unfortunately, the current delivery of 
computer science education is meager and actually diminishing.
    I strongly support the inclusion that you have made of 
workforce education and diversity issues in the NITRD 
legislation and urge you to assure that NITRD agencies are 
accountable to report back what they are doing to improve K-12 
education and diversity.
    And finally, I strongly encourage you bringing the 
Department of Education back into the NITRD Program.
    Now, I could stop there and possibly set the record for 
brevity, but I will elaborate a little bit. First, I want to 
express my sincere appreciation for your work on the NITRD 
legislation overall. NITRD is at the core of what we do in 
computing research at U.S. universities and has been spoken by 
many here, the research advances under the NITRD Programs have 
been the lifeblood that have fueled much of our Nation's 
economic growth.
    As the scope of computing continues its marvelous 
expansion, it's crucial that the scope of the NITRD Programs 
expand as well. A particularly important area in my estimation 
is health IT.
    Now, returning to workforce, as the SRI study that has been 
alluded to validated, the demand for IT professionals is much 
greater than the supply, both in total and in almost all of the 
subcategories of IT. Universities have actually done a very 
good job of evolving with the times in turning out graduates 
that industry really values. They are simply not turning out 
nearly enough of them, and when you look at the reason for 
that, a good part of the problem is the lack of rigorous 
computer science at the K-12 level.
    A recent report called, ``Running on Empty,'' that was 
issued by the ACM and the Computer Science Teachers' 
Association showed that computer science at K-12 really faces a 
triple whammy. It is the lack of computer science standards 
that have been implemented by the states, the lack of rigorous 
computer science courses that count as core graduation 
requirements, and the lack of computer science teachers. An 
explanation I sometimes get for that is rather simple. Computer 
science wasn't around 50 years ago when much of this curriculum 
got solidified, and since then there has simply been no room at 
the end.
    So for states to strengthen computer science K-12 
education, I believe that the encouragement and support at the 
federal level is essential, and NITRD can help significantly 
with this. In my written testimony I discuss a number of 
helpful steps.
    One important example of that is support for the very 
exciting new computer science advanced placement course that is 
being developed and the accompanying CS 10K Project, which is 
attempting to train 10,000 new teachers to be able to deliver 
that course.
    Another simple but really key component is just assuring, 
as has been said, that all federal STEM legislation clearly and 
explicitly state that computer science is part of the scope of 
that legislation.
    So to conclude and to come back to the comments about 
workforce, if this hearing had been held yesterday, it would 
have been difficult for me to attend. Our School of Informatics 
was holding its annual career fair where we had 80 companies in 
the largest space that our campus can accommodate, interviewing 
our students. The message that we got from those companies is 
the same one that I hear every year, and that is that the three 
to four hundred students who will graduate this year is a 
fraction of what they would like to hire from us.
    As I look to my colleagues and they are nodding, that 
message could be repeated at virtually every university in this 
country. I really appreciate your dedication to helping our 
Nation solve that problem and will look forward to responding 
to your questions.
    Thank you.
    [The prepared statement of Dr. Schnabel follows:]
               Prepared Statement of Dr. Robert Schnabel,
                      Dean, School of Informatics,
               Indiana University, School of Informatics
    On behalf of Indiana University, its School of Informatics, the 
Association for Computing Machinery, its Education Policy Committee, 
the members of the Computing in the Core Coalition and myself, thank 
you, Chairman Brooks, Ranking Member Lipinski, and Members of the 
Subcommittee, for the opportunity to share comments on the Networking 
and Information Technology Research and Development (NITRD) program 
with you.
    I have been involved in computing and the computing community for 
nearly 40 years. Prior to assuming my current dean position in 2007, 
this includes 30 years at the University of Colorado at Boulder as a 
professor of computer science, and service as chair of Computer 
Science, associate dean for academic affairs in the College of 
Engineering and Applied Science, Vice Provost for Academic and Campus 
Technology and CIO, and founding director of the Alliance for 
Technology, Learning and Society (ATLAS) Institute. I also am a co-
founder and executive team member of the National Center for Women & 
Information Technology.
    Computing is transforming our world--driving innovation in numerous 
fields, leading to entirely new multi-billion dollar industries 
creating thousands of new jobs, and transforming how we live, work, and 
socialize. Fueling this engine of innovation are the investments that 
various agencies have made in the computing research enterprise and the 
workforce that supports it. The Networking and Information Technology 
Research and Development program (NITRD) plays a key role in 
coordinating and focusing these federal programs.

Summary of Recommendations Concerning Education and Workforce

    H.R. 2020, as passed the House in the 111th Congress, proposes 
enactment of the President's Council of Advisors for Science and 
Technology (PCAST) recommendations for assessment and strategic 
planning by the NITRD program. These elements will strengthen the 
overall NITRD program. We particularly appreciate and strongly support 
the committee's inclusion in H.R. 2020 that NITRD address both 
education and workforce issues, including the diversity of the IT 
student and workforce population, as part of its strategic planning 
process.
    If we are to continue to discover and develop the innovations that 
have created new industries and transformed others, we need to ensure a 
healthy IT workforce that is skilled and large enough to meet the 
nation's growing IT needs, and reflects the gender and racial diversity 
of our nation. While university computing and IT education and research 
programs have done a good job of changing with the times to meet 
current needs, the education pipeline feeding our workforce is not 
producing enough graduates in IT fields to meet the growing needs of 
the computing industry, let alone the other industries that rely on 
computing and the public agencies that need computing professionals. In 
addition, women and many minority groups are greatly underrepresented 
among computing and IT students and in the IT workforce, depriving the 
nation both of potential skilled workers and of the innovation that 
results from diverse teams.
    A key element of this pipeline is in crisis and is directly related 
to the insufficient number of students in university computing and IT 
programs: K-12 computer science education. If we do not address the 
issues in K-12 computer science education, students will have few 
opportunities to experience this critical discipline or its concepts 
before higher education and our computing pipeline will continue to 
suffer. NITRD and the National Coordinating Office (NCO) can play a key 
role in addressing obstacles standing in the way of strengthening K-12 
computer science education. As the committee works toward considering a 
new NITRD reauthorization, we recommend Congress add additional 
provisions for NITRD programs to specifically address the systemic 
issues facing K-12 computer science education, namely:

      NITRD programs should report to NCO what steps they are 
taking to address K-12 computer science education reform.

      Include the Department of Education in the NITRD program.

      Include and clearly define computer science in federal 
education programs.

      Create state planning and implementation grants for 
computer science K-12 curriculum and build national networks of support 
for K-12 computer science education.

      Create pre-service and professional development 
opportunities for K-12 computer science teachers.

    The remainder of this testimony expands upon the preceding points.

NITRD's Important Role in Sustaining Innovation

    Information technology, driven by public and private research 
funding, has transformed our society and our economy. As amazing as the 
progress of the last 20 years is in this regard, the future can be even 
more amazing, if public and private players sustain our IT research 
ecosystem. Historically, the diversity of our NITRD agencies has been a 
major strength, fostering multiple approaches to complex problems. The 
Internet began as a Defense Advanced Research Projects Agency (DARPA) 
project, grew with National Science Foundation (NSF) support and 
blossomed with commercial funding. The Human Genome Project was a 
triumph of biomedicine and IT, building on National Institutes of 
Health, DARPA, NSF and Department of Energy research and birthing 
personalized medicine.
    A key element of the NITRD program involves fostering communication 
and coordination across 13 federal agencies where IT is relevant. This 
creates a rich ecosystem for information technology research and 
development, spanning many programs. The legislation proposed in the 
111th Congress strengthens the program by addressing several key 
recommendations forwarded in 2007 by the President's Council of 
Advisors on Science and Technology.
    As the National Coordinating Office (NCO) begins to develop 
strategic plans for computing research, it also should consider how 
agencies are meeting the ongoing challenge of supporting the continual 
broadening of the field of computing and information technology. I am 
closely acquainted with this broadening as the Dean of the School of 
Informatics at Indiana University, which offers a variety of 
undergraduate and graduate degrees in both computer science and 
informatics to meet the growing needs of the NIT workforce. These 
programs include research ranging from the foundational aspects of 
computer science to a wide range of applications and human and societal 
implications of computing and IT. It is important that NITRD programs 
embrace this breadth of research areas, as well as the growing 
diversity of university departments and schools that are part of the 
computing and IT field.
    One area of particularly great and increasing national importance 
in both research and education is health IT. The challenges that this 
area addresses range from assuring that the federal government and the 
country's health care system meets the needs of modernizing and 
standardizing health records, to providing powerful and easy-to-use 
information technology systems that support health care providers, to 
creating tools and systems that allow individuals to monitor and 
improve their own health independently. It is clear there are 
tremendous needs and opportunities in Health IT, and this area should 
be considered as a strategic focus for NITRD.

Addressing Our Workforce and Education Needs

    While everyone is talking about jobs these days--where to find 
them, how to create them--the computing industry is clamoring for the 
talent it needs to fill thousands of vacancies. The U.S. Bureau of 
Labor Statistic projects that the computing sector will have 1.5 
million job openings over the next 10 years, making this one of the 
fastest growing economic fields. There are many pathways into these 
jobs, but a deeper look at the fastest growing occupations within this 
field (such as computer software engineers or computer and network 
systems analysts) shows they either will require a computer science or 
related degree or greatly benefit from the knowledge and skills 
imparted by computer science courses. It is gratifying to see that the 
report "Networking and Information Technology: Workforce Study" 
presented to NITRD in May 2009 by SRI, corroborated these widely used 
workforce projections.
    Further, CNN's Money and PayScale.com ranked the ``Best Jobs in 
America,`` and the number one job is Software Architect. Other computer 
science career paths also were high on the list, including Database 
Administrator at number 7, Information Systems Security Engineer at 17, 
Software Engineer at 18, and at least 10 other computing careers 
ranking in the top 50. I commonly forward articles about the jobs that 
are most in demand to our school's career services office; computing 
and IT jobs are virtually always on these lists.
    During the past several decades, computing and IT has grown to 
address these needs. We have moved from a field focused on the 
foundational systems that make computers run (e.g., operating systems, 
programming languages) and applications in scientific computing and 
business data processing, to also encompass a wide array of general 
purpose computer applications (e.g., databases, computer graphics, 
robotics, computer security, graphical user interfaces) and discipline-
oriented applications (e.g., bioinformatics, health informatics). 
Higher education has adapted both by greatly broadening the scope of 
computer science at many universities to embrace this breadth and by 
adding new schools of computing, informatics and information that 
enlarge or complement them.
    In general, the students that are being produced by university 
computing and IT programs are meeting the needs of the IT workforce 
well; there are just far too few of them. Despite the tremendous job 
opportunities that computer science knowledge offers:

      Participation in AP Computer Science has been flat for a 
decade; \1\
---------------------------------------------------------------------------
    \1\  Growth in AP Computer Science tests taken has remained flat 
for the past decade while AP tests in other STEM fields have grown 
rapidly; see http://www.acm.org/public-policy/AP.jpg.

      Interest in majoring in computer science among incoming 
freshman is at an all-time low; \2\ and
---------------------------------------------------------------------------
    \2\  Source: UCLA Higher Education Research Institute Survey of 
Incoming Freshmen.

      here is little ethnic and gender diversity among those 
who take computer science courses. \3\
---------------------------------------------------------------------------
    \3\  According to the National Center for Women and Information 
Technology, computer science education has significant equity barriers. 
In 2008, only 17 percent of Advance Placement (AP) computer science 
test takers were women, and only 4 percent (784 students) were African 
American.

    This relates to insufficient exposure to computer science in K-12. 
We regard this as a fundamental issue that federal, state and local 
governments need to address to achieve its workforce needs.
    ACM has been on the forefront of efforts to strengthen K-12 
computer science education for years. Last year it spearheaded the 
formation of the Computing in the Core coalition to raise the national 
profile of K-12 computer science education. The founding members of 
this coalition are major stakeholders in the field of computing ranging 
from industry--Microsoft, Google, and SAS--to non-profit organizations, 
including the Association for Computing Machinery, Computer Science 
Teachers Association, National Center for Women and Information 
Technology, Computing Research Association, and Anita Borg Institute. 
Recently, the Coalition has grown to include the College Board, the 
National Council of Teachers of Mathematics and the National Science 
Teachers Association. Computing in the Core is united in our commitment 
to improving computer science education, which we strongly believe is 
marginalized in K-12 classrooms nationwide today.The marginalization of 
K-12 computer science education is a result of numerous federal, state 
and local education policies that do not make room for K-12 computer 
science education, coupled with deep confusion about what computer 
science education is in elementary, middle and secondary schools. A 
recent study, Running On Empty: The Failure to Teach K-12 Computer 
Science in the Digital Age, \4\ revealed K-12 computer science 
education is currently focused on basic skills, which teach students 
how to consume technology, versus acquiring deeper knowledge and skills 
which teach them to create new technologies. Further, only nine states 
``count'' computer science courses toward a core academic graduation 
credit. Finally, few states have robust teacher certification programs 
for K-12 computer science teachers.
---------------------------------------------------------------------------
    \4\  This study can be found at http://www.acm.org/runningonempty.
---------------------------------------------------------------------------
    The systemic absence of rigorous and engaging computer science in 
K-12 education starts at the local level, but there is a set of 
recurring policy issues that the Federal Government and the NITRD 
program can take strides to address:

      There are few states that have standards for computer 
science education and there are virtually no assessments for computer 
science education.

      Professional development for computer science teachers is 
limited as resources are focused away from this area.

      Computer science courses typically do not count toward a 
student's core graduation credit requirements.

    While decisions on these issues are often vested at the state and 
local level, NITRD and the NCO can address obstacles in federal STEM 
education and workforce-related programs computer science faces to help 
creating breathing room for state-led reforms of K-12 computer science 
education. We make the following specific recommendations for the 
committee to consider:

      NITRD programs should report to the NCO what steps they 
are taking to address K-12 computer science education reform.

       NITRD has a Program Component Area (PCA) that includes 
education activities and specifically mentions the 21st Century 
workforce and K-12 education as strategic priorities. However, there is 
little specific attention to these issues within the PCA or 
prioritization within the NITRD program in general. Most education 
funding within the NITRD program is from the National Science 
Foundation (NSF), while the Department of Education does not 
participate in the NITRD program at all. Of the NSF activities, there 
appears to be little to no involvement with some of the key programs 
within NSF's Education and Human Resources Directorate focused on 
strengthening K-12 science, technology, engineering and mathematics 
education, including the Math Science Partnership program. We encourage 
greater ties with these programs, particularly MSP.

       We note that the CE-21 program within the Computing and 
Information Science and Engineering Directorate at NSF is one program 
focused on addressing K-12 computer science education. It has invested 
in the development of a new AP Computer Science: Principles \5\ course 
intended to be broadly engaging and appealing to students, as well as 
other initiatives focused on reviving K-12 computer science education. 
The effort also rightly focuses on inclusion--making sure that the AP 
test and the computer science discipline appeal to a population of 
students diverse in race, ethnicity, socioeconomic status and gender. 
We support this program and point to it as model for addressing some of 
the key challenges in K-12 computer science education.
---------------------------------------------------------------------------
    \5\  See http://www.collegeboard.com/html/computerscience/
index.html.

---------------------------------------------------------------------------
      Include the Department of Education in the NITRD program

       As previously mentioned, the Department of Education is 
not one of the agencies currently participating in the NITRD program. 
Considering the key linkage between education and workforce, it is 
difficult, if not impossible, to address the workforce needs and the K-
12 education issues, without having the Department of Education at the 
table. We urge you to ask the agency to return to NITRD.

      Include and clearly define computer science in federal 
education programs.

       Computer science means the study of computers and 
algorithmic processes, including their principles, their hardware and 
software designs, their applications, and their impact on society. 
Computer science education includes the following elements: design 
(both software and hardware), creation of digital artifacts, 
abstraction, logic, algorithm development and implementation, 
programming paradigms and languages, theoretical foundations, networks, 
graphics, databases and information retrieval, information security and 
privacy, artificial intelligence, the relationship between computing 
and mathematics, the limits of computation, applications in information 
technology and information systems, and social impacts of computing. 
\6\
---------------------------------------------------------------------------
    \6\  ACM and CSTA have a four-part, grade-appropriate framework 
describing the standards for computer science education in K-12; see 
http//www.csta.acm.org/Curriculum/sub/ACMK12CSModel.html.

       As schools have increasingly stepped up the integration, 
use, and teaching of information technology as tools that support 
learning, distinctions between these areas that involve the use of 
computing and IT as learning tools, and genuine computer science 
education have blurred. Educators and policy makers consistently 
confuse the use of technology and teaching of technology literacy with 
teaching computer science as a core academic discipline within the STEM 
fields. PCAST recognized this issue in their 2010 report, Prepare and 
Inspire: K-12 Education in Science, Technology, Engineering and Math 
---------------------------------------------------------------------------
(STEM) for America's Future:

        ``Computer-related courses should aim not just for 
technological literacy, which includes such utilitarian skills as 
keyboarding and the use of commercial software packages and the 
Internet, but for a deeper understanding of the essential concepts, 
methods and wide-ranging applications of computer science. Students 
should gain hands-on exposure to the process of algorithmic thinking 
and its realization in the form of a computer program, to the use of 
computational techniques for real-world problem solving, and to such 
pervasive computational themes as modeling and abstraction, modularity 
and reusability, computational efficiency, testing and debugging, and 
the management of complexity.''

       Federal programs exacerbate this confusion with vague 
terminology, as well as simply including ``STEM'' as eligible subjects. 
This often does not translate into computer science programs being 
included in the scope of the programs when they are implemented at the 
state and local levels. Relying on ``STEM'' as the foundational 
definition can inadvertently set up barriers for computer science. For 
example, NSF's Math and Science Partnership program specifically states 
that it is open to all ``STEM'' proposals; however, a closer review 
shows that grants must focus on improving ``math and science'' scores. 
Any proposal focused on computer science must show gains in math and 
science, not actually on computer science.

       For these reasons, it is crucially important that 
federal STEM workforce and education programs explicitly state that 
they include computer science. This recommendation is consistent with a 
recent report on PCAST that said computer science must be part of STEM 
education programs. As a coordinating body, NITRD should work with 
participating agencies to explicit include computer science as an 
eligible discipline within STEM education programs.

      Create state planning and implementation grants for 
computer science K-12 curriculum and build national networks of support 
for K-12 computer science education

       States should be developing specific, thorough plans to 
improve computer science education. Few states are deliberately 
integrating computer science into their K-12 offerings at elementary 
schools or ensuring its place in the high school curriculum. A broader 
capacity initiative focused on improving curriculum, outreach and 
evaluation would build support for the goals and efforts of state 
planning and implementation of grants.

       As we previously recommended, bringing the Department of 
Education back into the NITRD program could create additional resources 
for such plans, but other NITRD agencies (such as the Department of 
Defense, NSF and the Department of Energy, which all house formal and 
informal education programs) should work directly with States to ensure 
state workforce and education needs are met. Establishing these plans 
or pilots for reforms within the States is a step toward addressing the 
deeper policy issues in K-12 computer science education.

      Create pre-service and professional development 
opportunities for K-12 computer science teachers

       Very few schools of education are focused on preparing 
computer science teachers, and because of a focus on ``core'' courses, 
there is limited professional development funding for computer science 
teachers. Federal agencies have numerous professional development and 
pre-service programs; however, we have consistently found little 
support for K-12 computer science education teachers within them. As 
course offerings in computer science grow, particularly with the new 
Advanced Placement Computer Science Principles course being introduced 
into schools, a program that specifically addresses the shortage of 
certified computer science teachers at the K-12 level is imperative, as 
are investments in professional development for those already teaching. 
Again, NITRD can play a role in raising this issue within agencies that 
have STEM or general education professional development or pre-service 
teacher programs.

The Computer Science Education Act

    Provisions of the No Child Left Behind Act (NCLB) have also 
contributed to computer science's marginalization. Because of NCLB's 
accountability provisions and its definition of ``core'' disciplines, 
states have put resources toward investments in curriculum, pedagogy 
and professional development related to ``core'' courses. Furthermore, 
high school graduation requirements are tied to core courses. There are 
countless stories of teachers being pulled out of computer science 
courses to support the mathematics proficiency goals of NCLB. While you 
and your colleagues consider the future of NITRD, the House Education 
and the Workforce Committee is considering reauthorization of the 
Elementary and Secondary Education Act. Computing in the Core is 
working to ensure that a revised education law accommodates computer 
science in its provisions related to STEM education and ensures that 
computer science educators have access to the professional development 
and supports their colleagues do. The Computer Science Education Act 
from the 111th Congress represents our priorities related to programs 
administered by the Department of Education.

Conclusion

    The NITRD program plays a crucial role in the development and 
health of the country's networking and information technology 
capabilities, and we strongly support the program. To meet the large 
and growing needs of this industry, the nation will require a much 
larger and more diverse array of computer science and IT professionals 
than it currently is producing. We welcome and applaud the inclusion of 
workforce, education and diversity issues in the NITRD program. We 
particularly encourage the NITRD program to play an active role in 
strengthening K-12 computer science education, as this is the 
foundational issue that needs to be address to bolster the population 
of students focusing on computing and IT at the university level, and 
entering the IT workforce.
    Thank you again for the opportunity to appear before the 
Subcommittee today and for your attention. The groups I represent today 
stand ready to work with the committee to address our recommendations 
as NITRD reauthorization moves forward in this Congress. I'll be 
pleased to address any questions you have.

    Chairman Brooks. Thank you, Dr. Schnabel, and thank you the 
other panel members for your testimony.
    Reminding Members of the Committee that Committee rules 
limit questioning to five minutes.
    The Chair at this point will open the round of questions, 
and the Chair recognizes himself for five minutes.
    This question is with respect to each of the witnesses. 
Each of you discuss the importance of federal investments in 
networking and information technology research and development, 
I am sure you are all also aware of the budget and deficit 
decisions facing the United States Congress. In looking at the 
FY 12 budget and what is already a finite amount for federal 
investment and will likely be even smaller next year, how would 
you prioritize federal NITRD investments?
    Whoever wishes to go first.
    Dr. Strawn. Chairman Brooks, I will speak from the NITRD 
Coordination Program, for example. We certainly will have 
discussions with our agencies asking them how they prioritize 
their individual activities and then we will seek to mold that 
into a unified whole, looking for gaps that might have occurred 
as people prioritize away important activities.
    So our goal of coordination may help in that activity.
    Chairman Brooks. Will you please report your results to 
this committee, submit a written report?
    Dr. Strawn. Yes, sir.
    Chairman Brooks. Thank you.
    Dr. Lazowska. Mr. Chairman, I think that America needs to 
decide in which areas of R&D it absolutely must be the world 
leader, and it needs to make extra investments in those fields. 
And I think this is the number one field in which America has 
to be the world leader. It has to be the world leader because 
this enables advances in all other fields because it drives our 
economy forward, because it is the largest source of science 
and technology employment, because it is essential to our 
national security and because if you want advances in areas 
like energy and transportation and health, then you need 
advances in this field. This is really the cornerstone of our 
economic success, our national security, and our discovery in 
all other fields.
    So I honestly think that the question you need to ask is 
how to make very difficult choices among fields of 
prioritization, and I want to emphasize that the NITRD crosscut 
budget dramatically overstates the amount of funding that the 
Federal Government is actually spending on NIT R&D because of 
the categorization issues I addressed before.
    The one other thing I would like to add is it is often 
tempting to confuse industry R&D for research that looks out a 
long way, and in truth the Federal Government is by far the 
most significant investor in research that looks out more than 
one product cycle. I am from Seattle. Microsoft is one of the 
computing companies that has a significant investment in 
research that looks out more than one product cycle. That is 
called Microsoft Research. It is about 900 people around the 
world, most of them in the United States, and it represents 
about four percent of Microsoft's R&D budget. The rest is very 
talented engineers producing, in caricature, the next version 
of Office and Windows. All right. It is R&D, but it is not what 
is going to lead to the next generation of breakthroughs. That 
is our job.
    Dr. Sproull. Mr. Chairman, not surprisingly, my counsel 
would be to be sure that the long-term investment remains as 
healthy as it can. Indeed, perhaps the entire investments need 
to be modulated, but I think we have shown that it is the 
long--as Dr. Lazowska was saying, the long-term has huge payoff 
and may lead to a brighter economy for all of us in that long-
term.
    And the cliche, I am afraid, is apt, which is let us not 
let the urgent drive out the important.
    Dr. Schnabel. Mr. Chairman, I will just add briefly. I 
think a strategy always has to be a combination of things that 
are focused and things that are general. We have heard about 
some focused priorities in areas as health, energy, and 
security, but as we travel around the world, and all of us do 
that, one still hears in the countries that we now see as our 
growing competitors a great envy for the culture of innovation 
in this Nation that none of them can replicate. And to be able 
to sustain that we also have to leave room for things that are 
not as focused but will lead to that next round of innovation 
or otherwise we will be killing the goose that lays the golden 
egg.
    Chairman Brooks. Well, if I could just add a comment, 
Chairman's prerogative for a second, most everywhere I have 
been in Congress this year, everybody I talk to says their 
program is number one, and obviously, we can't fund everyone's 
program at number one given the financial circumstances we face 
unless we want to risk the Federal Government's solvency and 
bankruptcy, in which case every program would be last because 
we wouldn't have enough.
    To give an example of the severity of the problem, last 
Thursday the Chairman of the House Armed Service Committee gave 
us a briefing on information he had received on the impact of 
some of these potential cuts on a national defense. We were 
looking at hundreds of thousands of military uniformed 
personnel, DOD civilian personnel, support contractors in the 
private sector that are going to be laid off or the positions 
will no longer exist. Talking about mothballing one carrier 
battle group, two nuclear submarines, 10 percent of our fighter 
aircraft and strategic bomber aircraft.
    So if you can share with us any insight at some point in 
the future by submitting in writing to this committee a 
supplemental statement in which you share with us what you 
think your priorities for funding ought to be within NITRD, 
that would help us. If we have the amount of funding we have 
right now and we can fund everything at the same level, well, 
then that is fine. We don't have to get into that 
prioritization process.
    But this is an opportunity for you to share with us your 
expertise about where the money ought to be spent if we are 
forced to deal with less overall. Otherwise, thank you for your 
comments.
    Next I recognize Mr. Lipinski.
    Mr. Lipinski. Thank you, Mr. Chairman.
    This past Monday I was speaking to a group of constituents, 
and I was going through a job--five-point jobs plan I had put 
out a couple of months ago, and I said in there we need to 
invest in education, invest in innovation, and I think that is 
what you are talking about here.
    And I think all of your testimony sort of really supports 
what I said why we have to do that is not only do we grow the 
jobs here in the United States, but you look at the 1990s, and 
we were able to grow out of the budget deficit to a budget 
surplus. You know, some of that came because we were able to 
grow, and you know, certainly was connected up with the really 
the explosion of the World Wide Web.
    And so I think everything that all of you had said really 
supports that idea, and I think it is really critical as we do 
prioritize what really is--what really are good investments. I 
think the Chairman is correct. We really need to prioritize, 
but here I think that we are in the right place in terms of 
prioritizing this research.
    I want--I was going to ask a question of workforce, but I 
think I am going to go instead because something Dr. Lazowska 
had said prompted me to go in a--to my second question first, 
and this is a question for everybody.
    One of PCAST's recommendations, which Dr. Sproull in 
particular reiterated in his testimony, is for the NITRD 
Program to better account for the part of its budget that 
supports actual NIT R&D as opposed to NIT infrastructure that 
enables R&D and other science and engineering fields.
    Should we be removing the budget for such infrastructure 
from the NITRD Program altogether? Are there downsides in 
narrowing the scope of the NITRD Program in this way? And 
finally, do we have a good estimate of what we are actually 
spending in NIT R&D if the NITRD budget is currently 
overcounting what we are spending?
    Who wants to start on that? Dr. Lazowska.
    Dr. Lazowska. I will begin, and I am sure Dr. Strawn will 
have something to add. I want to say that I am speaking for 
myself and not for PCAST, and it is important to understand 
that.
    My view and I think the view of our working group on the 
report at least was that it is entirely appropriate to continue 
to include those funds as part of the NITRD crosscut, simply to 
categorize them more carefully. There are already parts of the 
NITRD crosscut budget that account for infrastructure. For 
example, the acquisition of high-performance computing 
equipment.
    But suppose that the National Institutes of Health, for 
example, spends substantial amounts of funding on databases for 
biomedical research. All right. Those are very, very important 
expenditures. They are crucial to driving biomedical research 
forward, they are IT, they belong in the NIH research budget, 
it is fine to have them as part of the NITRD Program, but they 
should be identified as the use of advanced information 
technology infrastructure to drive biomedical research forward.
    As you perhaps know from reading the report, we asked an 
independent organization to look at several agencies, and the 
accuracy of the categorization varies widely across agencies, 
and I want to emphasize that no one is actively misreporting. 
What is needed is simply a more accurate characterization. Dr. 
Strawn and the National Coordinating Office have already taken 
steps in that direction, and I want to emphasize that this is a 
coordinating process that works.
    And two quick examples. The previous PCAST report 
recommended increased emphasis on cyber physical systems, and a 
crosscutting program across multiple agencies was launched very 
quickly. The most recent PCAST report said increase the 
emphasis on large-scale data analysis, which is necessary for 
commerce, for scientific discovery, for national security, and 
we already have a big data senior steering group.
    So it is really a very responsive coordinating process.
    Mr. Lipinski. Who wants to go next? Dr. Strawn.
    Dr. Strawn. Yes, sir. Let me just say--thank you. The only 
thing I would refer back to is my use of the ``Pasteur's 
Quadrant,'' book reference in my oral testimony about R&D being 
a two-dimensional structure, and it is--that sort of means it 
is not always clear what will produce the most important long-
term research.
    The project orientation, which has been effectively used by 
ARPA and DARPA and now by energy and by information and so 
forth where we have advanced research projects, may look like 
development or even look like infrastructure, but, in fact, if 
something entirely new comes out of it, it can spawn whole new 
research areas as well as whole new areas of activity.
    As Dr. Lazowska mentioned, in one of our coordinating areas 
on high-performance computing, we have two sub-areas that we 
report separately, high-performance computing R&D and high-
performance computing infrastructure and applications. This has 
been our largest area because there has been a large federal 
investment in high-performance computing, so we thought it was 
appropriate some time ago to break out those two distinctions. 
And it is certainly conceivable that we can break out in some 
of our other coordinating areas of similar situation, and I 
only caution that we shouldn't look for a sharp line between 
what is information science and supportive science and what is 
information science theory.
    Dr. Lazowska. Could I add a comment? Through the wonder of 
information technology, I was able to learn that the Navy's 
newest submarine is going to cost $13 billion, a $4 billion 
budget overrun. Just a news post on my iPhone. I want to 
emphasize that the National Science Foundation Computer 
Information Science and Engineering annual budget is in the 
order of half a billion dollars per year, and the corresponding 
DARPA Information Technology Investment is on the order of half 
a billion dollars per year.
    These are very significant amounts of money, but they are 
rounding errors in terms of the federal budget, in terms of the 
cost overrun of a single submarine, and they are what power our 
Nation forward. Okay. It is those NSF and DARPA investments in 
information technology that make possible the prosperity we 
enjoy today.
    So it is important to keep in perspective the relatively 
small amount of money that the Federal Government is investing 
in this field and the billion dollar industries, the many 
billion dollar industries that it creates.
    Dr. Strawn. We like to think our leverage factor or our 
multiplier factor is very great compared to many other federal 
investments.
    Mr. Lipinski. It would be wonderful to see what you can--
what can be shown as much as you can, that is not easy to do, 
but always getting back to, you know, what the Chairman 
emphasizes is, you know, prioritizing and anything that you can 
do to show the results is--would be very helpful.
    So thank you. I yield back. Thank you for the extra time.
    Chairman Brooks. Thank you, Mr. Lipinski, and members of 
the panel, I would like to echo some of your comments. It would 
be wonderful if we would implement economic policies that will 
deal with some of the structural issues that have inhibited our 
country's economic growth. That is the best way to get out of 
this, and we all understand the pivotal role your sector has 
played in the economic growth that we have enjoyed in past 
years. Let's just pray that Congress and the White House 
collectively will do the right thing.
    Next the Chair recognizes Mr. Bucshon of Indiana.
    Mr. Bucshon. I thank everyone for attending, and mine is 
going to be more focused on--less on budget and more on what 
type of research that we can be doing.
    I am in health care. I was a heart surgeon before this. As 
you know there is a big push nationally, and this may not be as 
important as some of our national security issues, but it is 
important to our country, there is a big push, of course, for 
electronic medical records and patient--there is more and more 
patient data being stored permanently not on, for example, X-
rays that are not on actual film but that are on hard drives 
around the country and around the world.
    And from a medical standpoint right now, we have a system, 
as you probably know, which is a hodge podge of a multitude of 
different electronic systems, most of which are proprietary and 
don't communicate between each other, which just to give you an 
example in my hometown, Evansville, Indiana, we have two 
hospitals, and they both have different systems, completely 
different systems, no way to communicate. My medical practice 
had a system of electronic medical records with all kinds of 
data, no way to communicate with either hospital, and then a 
couple of the other practices now are putting in their own 
systems.
    So I guess my question would be to the panel, you know, 
from a medical standpoint what type of, you know, what--is 
there anything going on out there to try to figure out how to 
coordinate this type of information, which is, you know, which 
is private medical information to make it more accessible to 
medical professionals, to make it more easy for the public to 
maybe even gain access to their own data? And because my view 
is if unless we can solve the security issues and the 
communication issues, actually electronic medical records are 
going to cost us money, not save us money. That is my view.
    So I would be interested in anyone's comments.
    Dr. Strawn. Thank you for the question, Mr. Bucshon. In 
response to a Congressional legislation in the Auto Program, 
the NITRD Program stood up a senior steering group in health IT 
research and development. We have been working for 
approximately--more than a year now. We have attracted pure 
computing folks such as those at NSF and NIST, and we attracted 
many of the applied computing folks across HHS and DOD and 
other areas. We have a large senior steering group that has a--
is looking at a large portfolio of possibilities and certainly 
electronic health records is one of those items.
    We are moving toward focusing on, again, long-term issues, 
the creation of a health information infrastructure. Then the 
next step is how do we turn that information infrastructure 
into knowledge and action, how--while we are doing all this how 
do we empower the patient, the physician, everybody involved 
with access to that information turned into knowledge, and the 
devices, whether it is robotics or assistive devices in homes 
or electronic communication.
    Those are at least four of the areas that we are looking at 
as we initially begin our dialogue with the health IT 
community, and my understanding is that there are more than 600 
existing EHR systems, which just shows the size of the problem 
that you have alluded to.
    One potentially small step in that direction is a high-
level agreement that I understand has been reached between VA 
and DOD to interoperate their two electronic health systems. 
Our goal is to make that a prototype for something that will 
produce interoperability hopefully among all systems.
    We certainly think that interoperability is more of a way 
to go than forcing one single standard. Just like the Internet 
itself was sort of a software network based that interconnected 
many networks that operated at a hardware level.
    Mr. Bucshon. Okay. Thank you.
    Dr. Lazowska. I was just going to add that this is clearly 
a place where the government can play a role, that is nudging 
industry to adopt standards and interoperability. There was a 
PCAST report in the past year on health IT that focused on 
precisely this issue; this is separate from the PCAST NITRD 
report.
    You are absolutely right about the privacy and security 
issues, and that is something that really underpins all 
civilian, and of course, military use of information 
technology. It is an area where every federal agency needs more 
than we have today. We have not focused enough on long-range 
approaches to privacy and security. Our approach has been Band-
aids rather than something that is going to get us out of this 
sort of rat race in which we are trying to keep ahead of the 
bad guys, and it is really going to impede adoption of this 
important technology.
    The final thing I want to mention is that there are aspects 
of information technology and health that go beyond electronic 
medical records and will be in the future as important. The 
question I always ask is why my body is so less well 
instrumented than my automobile. You know, I bring the 
automobile into the dealership, and the mechanic sticks a jack 
in under the dash and reads out the last six months of data and 
tells me what the problem is, and when I go to see my 
physician, she hits me on the knee and says, ``Where does it 
hurt?'' to first approximation.
    Okay. This has got to change, and there are many other 
areas where we will see change. For example, the genotype, 
phenotype correlation that is going to use big data to 
transform medicine in the future. So we need to invest in that 
entire spectrum.
    Mr. Bucshon. All right. Thank you. I think one more comment 
from a fellow Hoosier.
    Dr. Schnabel. Yeah, and if I may, too, just briefly. First, 
to reinforce your point, in fact, the Executive Associate Dean 
of our medical school made the comment that he can go to India, 
put his bank card in an ATM machine, and it works, but if he 
walks across the street to a new medical provider, he can't get 
his records to follow him.
    And we heard some comments about that, and it is a huge 
problem, but I do want to reinforce. I think it is really 
important for this subcommittee as we hear about health IT from 
each witness to realize the breadth of that field and that it 
includes clinical things which are much more than medical 
records themselves, also many things that assist physicians and 
doctors. It includes a whole consumer space of devices that we 
can use as individuals outside of medical care and even a 
population space of modeling of influenza and other things.
    So it is a very rich space of research.
    Mr. Bucshon. Thank you very much. My time has expired.
    Chairman Brooks. Thank you. What we are going to try to do 
is get the next two Congressmen in. If we have time before 
votes are called, apparently there are some issues counting up 
the right totals, giving us a little bit more time; we might 
have a second round of questions.
    But with that, Mr. Bartlett from Maryland, you are 
recognized.
    Mr. Bartlett. Thank you very much. We clearly are the most 
creative, innovative society in the world. We lead the world in 
computers and information technology. We can't even turn out 
people fast enough to fill the jobs available in these areas.
    And in spite of that every 12 hours we have another billion 
dollar trade deficit. I am told that only three things 
ultimately produce wealth: mining, manufacturing, and farming.
    Now, a lot of people have gotten very wealthy with 
computers and information technology, but, you know, you can't 
eat those electrons. They won't keep the rain off your head, 
they won't take you anywhere. Ultimately, at the end of the 
day, aren't these technologies simply enablers that help us to 
do other things better, that really create wealth? It isn't 
somebody else doing most of these other things better. To the 
extent that we continue to develop these technologies, aren't 
we just enabling our competitors?
    What do we have to do so that we start doing the things 
that ultimately really create wealth, because are not these 
things simply tools that help us do these other things better 
and now somebody else in another part of the world is doing all 
these things better. What do we have to do so that we are 
encouraging the technologies that ultimately create wealth so 
that this billion dollars every 12 hours doesn't continue, 
because that is not sustainable?
    Dr. Strawn. Mr. Bartlett, I think you are certainly 
correct, and I think that is one of the reasons that one of the 
current focuses is on advanced manufacturing, which I interpret 
to mean the continual inclusion of additional IT services and 
capabilities into the manufacturing process. Farming, I might 
also say, I was last weekend at my brother-in-law's farm in 
Illinois where he is now farming by the foot with GPS 
technologies and so forth, so I think that we are beginning to 
use these advanced technologies in important ways and 
applications.
    And my view is certainly these applications are the end 
result. It also turns out that the more theory we have the more 
applications we can serve. So both a balance between the theory 
and the application seems to me to be the most efficacious in 
terms of the long-term focus.
    Dr. Lazowska. I think our standard of living depends on our 
workers being more productive than workers anywhere else in the 
world, and information technology contributes enormously to 
that productivity in all sectors. Dr. Strawn addressed farming, 
and that is an important one. One hundred fifty years ago, if I 
recall correctly, something like 98 percent of employed 
Americans worked on the farm, and 100 years ago it was perhaps 
50 percent, and now I believe it is 1-1/2 percent, and they 
produce enough food to feed our Nation and much of the world, 
and there are, of course, many contributors to that, including 
new crops and new fertilizers, but GPS and information 
technology plays an important role. That is why our service 
sector is more effective. Again, it is these productivity gains 
in the economy.
    So our standard of living depends on us being more 
productive and more efficient, and that is what information 
technology brings to us.
    Mr. Bartlett. But we still have that billion dollars every 
12 hours of trade deficit, and my question is being preeminent 
in these technologies, which I think are simply tools to help 
us do those things better that would free us from this 
dependence on foreign goods, how do we get from where we are 
with our clear superiority in computers and information 
technology area to where we are manufacturing, mining, the 
kinds of things that will free us from this intolerable trade 
deficit?
    Dr. Sproull. Sir, I would suspect that a lot of your 
billion dollars a day is energy costs, and all of the new 
energy sources, renewables and so forth, depend heavily on NIT 
for control, for production. As an enabler surely, yes, of the 
control system for both generating the power, distributing it, 
and improving the technologies as they go forward.
    For example, one of the reasons high-performance computing 
was initially focused on--and the Act of 1991--was as a design 
aid to be able to model complex energy-producing technologies 
more accurately so that they would be more efficient. It has 
been used, for example, in things like how you burn coal more 
efficiently.
    So that all goes to, it seems to me, helping reduce your 
deficit.
    Mr. Bartlett. Thank you very much, Mr. Chairman.
    Chairman Brooks. The Chair next recognizes Mr. Hultgren of 
Illinois.
    Mr. Hultgren. Thank you, Mr. Chairman. I apologize. It is 
kind of a crazy afternoon, so we are kind of coming in and out, 
have a couple different meetings at the same time but really do 
appreciate so much you being here. Chairman, thank you for 
hosting this meeting as well.
    I do have a question. I had an interesting lunch today. We 
pulled a group of people together talking about the iPad, 
something that is very important in me getting through my day 
as far as keeping my calendar and information and things, but 
talking about the technology that led to the iPad and how so 
much of that was really started by basic scientific research 
and a commitment to physics and even the GPS. It was so 
interesting to hear how we got--and the requirement of the 
ultimate precision clocks that we have got that really allow 
for GPS to work as it does. It was amazing, so interesting.
    But I just want to follow up, and I will start with Dr. 
Sproull if this is all right, in the testimony for today's 
hearing I know you all have talked about key technology 
advances, JAVA being one, iPad being one. Having been part of 
an industry applied research laboratory for over 20 years, what 
would you say are the three most critical and interesting 
changes or advances in the NIT industry since the early 1990s, 
and how did federal investment play a role in those advances 
that you would say?
    Dr. Sproull. Thank you. So the dominant one has to be the 
explosion of the World Wide Web, and the reason that exploded, 
the idea, as you know, came from a physics researcher in 
Europe, but the reason it exploded and the reason it exploded 
in the U.S. is the Internet was already there, and the basic 
communication protocols were in place, the switches were being 
built, things were starting to be deployed. It had migrated 
from a Defense Department prototype into the National Science 
Foundation where it was spread to wider academics, and then 
became available for commercial use at about the same time that 
the Web was invented, if you will.
    And then--that is right, but what happened--but you asked 
what the development was. The explosion came because a usable 
browser was developed, and that was developed here as was 
discussed in your meeting this afternoon on your turf.
    Mr. Hultgren. Yes.
    Dr. Sproull. And moreover, the U.S. venture community 
figured out a way to form a company to capitalize on it, it 
became very valuable, at least on paper very quickly, because 
the market exploded. And that transforms so many other things. 
The other witnesses here today have pointed out how important 
infrastructure capabilities enable still other developments, 
not just new end products or new specific services but further 
developments. So we are doing things with the Internet today 
that Andreessen didn't dream of even in 1993.
    So I have to say that that is one, two, and three, and the 
two and three if you really need two and three are things that 
came from one.
    Mr. Hultgren. Good. How about the rest of you? Are there--
would you agree with that? Are there other things that you have 
seen very specifically that you would say, hey, this is--we 
need help in telling the story of how, what we are doing is 
vital, and our failure for basic scientific research for really 
what we talk about on this subcommittee and this committee in 
this Congress is it is not a zero sum game. If we don't do it, 
somebody else will, and I think Congressman Bartlett talked 
about that as well, and, you know, some of the care that we 
have to take in this.
    But I would just be interested in hearing from the rest of 
you of what you have seen in your experience that maybe is most 
startling or most influential as far as advances.
    Mr. Strawn. Well, Dr. Sproull mentioned the Internet and 
things that derive from it. One of the things I believe that is 
deriving from it right now and is at early stages but may turn 
out to be of considerable importance is cloud computing. Cloud 
computing is in some sense a recentralization of computing 
where all of your software and your data are on centralized 
servers rather than on your own PC.
    Now, obviously there are tremendous security and privacy 
issues associated with that, but some people have described 
this as the industrialization of computing, and it may turn out 
to be of extreme importance, and it derives directly from the 
availability of the Internet.
    Dr. Lazowska. Let me mention a few things that are entirely 
different today than they were 10 years ago, and I will mention 
a couple of civilian things and some military things.
    One Dr. Strawn just touched on and that is the way we build 
these extremely large-scale systems. It is just totally 
different, totally different. What we do now is use unreliable 
hardware components because you can't possibly build systems as 
large as we need with reliable hardware. You couldn't afford 
to. You couldn't physically do it. So we use algorithms to make 
them reliable. That is a total change.
    Second, search. Ten years ago you used to actually file 
stuff away. Now you just search for it, you know, and I can't 
remember the last time I put something in an electronic folder.
    A third is mobility, the fact that you carry your whole 
life around with you, and again, all of these technologies 
trace themselves right back to the fundamental research 
program.
    A fourth one is digital media. The fact that photographs 
and videos and audio, all of that material today is created and 
edited and consumed in a digital world rather than an analog 
world. You know, when was the last time you saw 35 millimeter 
film as an example, but, you know, I don't even have CDs and 
DVDs anymore. My music is on my Mac Mini, and my movies come 
over Netflix, over the Internet.
    So these are things that have changed our lives and are 
totally different now than they were just 10 years ago, and you 
can trace every one of them back to the research program.
    If you look at America's military situation today in 
logistics, our ability to deploy troops appropriately around 
the world, in robotics, the drone aircraft, for example, that 
are used around the globe, large and small, the small-scale 
robots for investigating areas, search and rescue, exploration 
where you can't send individuals. Even things like natural 
language translation, the military for a number of years now 
has used artificial intelligence systems to simply determine 
which five percent of the documents in foreign languages are 
worth having some human look at, which is a 20-to-one reduction 
in the number of translators you need.
    So, you know, our military competitiveness today really 
depends on information technology in every imaginable way.
    Dr. Schnabel. And if I may, I will just remind us that the 
visual that Dr. Sproull used, that tire tracks diagram is, 
indeed, an answer to this question, because it actually traces 
back nearly 20 industries in most cases to university NITRD-
sponsored research groups.
    Mr. Hultgren. That is great. Well, again, thank you all. 
Appreciate the work that you have done. It is interesting, I 
think, even with my own kids, I kid them sometimes when they 
are playing with our family camera and just say, don't waste 
the film, because I remember my mom always saying that. Mom and 
dad, don't waste the film. Now you don't even think about that, 
but I also think back to my high-computing days in college of 
having a Commodore 64, and that was cutting edge. So it is 
amazing how far we have come.
    So, anyhow, thank you all so much. Appreciate it and yield 
back, Chairman. Thank you.
    Chairman Brooks. Thank you, Mr. Hultgren.
    The Chair recognizes Mr. Lipinski of Illinois for a brief 
period.
    Mr. Lipinski. Thank you, Mr. Chairman. I just wanted to--
I'm going to put this question for the record. I just want to 
throw it out there and sort of give you a heads up. I think the 
question of we come back to jobs, and we have the issue here--I 
think there are two things that make a career in NIT 
particularly difficult in the private sector.
    First, the fields move so quickly that it takes constant 
education and training to stay in the game, and second, some of 
the jobs are very easily outsourced. You can move almost 
anywhere if you have a computer and you have the network 
connection.
    So we don't have time to go into the--an answer, so I am 
just--but I think that is an issue. I am just--would like to 
have a--if we get a written response following up on, you know, 
what we can do to address this problem, what can we do to avoid 
training people for jobs that might not exist, is it something 
we can do, are there specific areas or programs that we should 
focus on, what is the best way for us to try to address this?
    But like I said, I don't think we have time here. I will 
formally give that as a question for a written response. I want 
to thank all of you for your testimony here today, and this is 
something that I am hopeful that we can have legislation on in 
this Congress.
    So thank you, Mr. Chairman, for the hearing.
    Chairman Brooks. I would like to thank the witness panel 
for the insight that you have shared with us. Quite frankly I 
think both Mr. Lipinski and I wish we had more time for follow-
up, but as you have heard the bells and buzzers go off in this 
place, that means that we have got to get on the House Floor 
shortly to vote.
    With that having been said, the Members of the Subcommittee 
may have additional questions for the witnesses, and we will 
ask you to respond to those in writing. The record will remain 
open for two weeks for additional comments from the Members. 
The witnesses are excused, and this hearing is adjourned.
    [Whereupon, at 3:13 p.m., the Subcommittee was adjourned.]
                                Appendix

                              ----------                              


                   Answers to Post-Hearing Questions




                   Answers to Post-Hearing Questions
Responses by Dr. George Strawn, Director, National Coordination 
        Office,Networking and Information Technology Research and 
        Development (NITRD) Program

Questions submitted by Chairman Mo Brooks

Q1.  As discussed at the hearing, Congress recognizes the value and 
importance of networking and information technology (NIT) research and 
development (R&D) funding; however, our existing budget constraints 
make prioritizing a requirement. Please detail how you would prioritize 
federal NIT R&D funding, what the top priorities would be, and how 
savings can be achived within the NITRD portfolio.

;A1.  As noted in my written testimony, all of the research reported in 
the NITRD portfolio is managed, selected, and funded by one or more of 
the 18 NITRD member agencies under their own individual authorizations 
and appropriations. Each NITRD agency engages in ongoing internal 
strategic planning and research prioritization activities to focus 
funding resources on those efforts most essential to carrying out its 
federal mission. NITRD agencies' published strategic plans typically 
identify their priority objectives in NIT R&D--the technical advances 
they need to meet mission needs. \1\ These research objectives are 
refined and adjusted in the annual discussions within the Executive 
Branch and then with the Congress.
---------------------------------------------------------------------------
    \1\  For example: NSF 2011-2016 Strategic Plan, http://www.nsf.gov/
news/strategicplan/
nsfstrategicplan-2011-2016.pdf; DOE 2011 
Strategic Plan, http://science.energy.gov//media/bes/pdf/
DOE-Strategic-Plan-2011.pdf; DoD 
Information Enterprise Strategic Plan 2010-2012; http://cio-
nii.defense.gov/docs/DodIESP-r16.pdf.
---------------------------------------------------------------------------
    Since the NITRD Program serves as a coordinating organization, it 
does not manage the portfolio of federal NIT investments; each agency 
manages its own NIT R&D investments. However, the NITRD Program's 
research framework represents the agencies' shared mission R&D 
priorities, as well as broad federal priorities and basic research to 
support the longer-term goals of the Federal Government and to develop 
technologies that promote U.S. economic, scientific, and technological 
leadership. By collaborating on NIT R&D where it makes sense-- such as 
in sharing high-end computing resources and codes for weather 
prediction--the agencies reap economies of scale and effort. As 
reported to Congress annually in the NITRD Supplement to the 
President's Budget, the NITRD research budget crosscut reports the 
agencies' NIT R&D spending and priorities. These priorities often also 
respond to directives of Congress and recommendations of the 
President's Council of Advisors on Science and Technology (PCAST) on 
technologies and capabilities deemed critical to the national interests 
of the United States. Examples include high-efficiency, ``smart'' power 
distribution systems; technologies to improve the quality, efficiency, 
and effectiveness of the U.S. health care system; next-generation tools 
for maintaining and working with ``big data''; radio spectrum 
efficiency and broadband access; fundamental advances in cyberphysical 
systems in such domains as industrial process control, transportation, 
and medical devices: and advances in cybersecurity with Domain Name 
System Security Extensions (DNSSEC) to secure key aspects of the 
Internet's infrastructure.
    In their NITRD Program activities, agencies identify shared high-
priority technical challenges and address them together to leverage 
each other's research efforts. For example: NITRD's networking agencies 
collaborate to maintain a unified high-speed network infrastructure for 
federally sponsored scientific research; and the high-end computing 
agencies cooperate in a single benchmarking activity to evaluate new 
supercomputer systems--a labor-intensive, time-consuming effort 
previously conducted by each agency. By joining together to tackle key 
R&D issues they face in common, the NITRD agencies are able to leverage 
resources, minimize duplication of effort, and partner in investments 
to pursue shared goals. I would also note here that the unique and 
single most significant result of NITRD coordination is broadly 
applicable technologies and capabilities--NITRD advances often yield 
new, open technologies that can be adopted across the commercial 
landscape. A current example is PerfSONAR, a suite of network-
monitoring tools developed by the NITRD agencies that enables managers 
for the first time to analyze network performance across multiple 
links; the tools are now being adopted by international research 
networks and the private sector.
    To summarize, I believe the NITRD portfolio appropriately reflects 
current Federal NIT R&D priorities and budget constraints, and will 
continue to reflect that balance in the future.

Q2.  In your testimony you mentioned four specific component areas for 
the National Initiative for Cybersecruity Education (NICE): awareness, 
education, workforce structure, and workforce training and professional 
development. Are all NICE-funded activities captured under the agency 
budgets for NITRD? How much are we investing in NICE activities in 
total and in each component area?

A2.  The NITRD Program is not tasked with tracking investments in NICE 
activities, either as a whole or by component area. NICE is an 
interagency effort led by NIST in which agencies identify common goals 
and milestones, commit their own resources toward achieving those 
goals, and align their respective implementation plans and activities. 
Implementation details for the goals and objectives outlined in the 
NICE strategic plan are currently under development and will be shared 
based on the policies of the agencies responsible for the execution of 
those details. NICE was formed under CNCI initiative 8, and budget 
figures are maintained by the Office of the Director of National 
Intelligence's Coordinate & Monitor team in the Office of National 
Intelligence Manager for Cyber.

Q3.  A primary objective of the NITRD program is to support 
foundational computing research to drive innovation, productivity, and 
ultimately, economic growth. With that in mind, do (or should) agencies 
evaluate the degree to which a research proposal considered under NITRD 
may be connected to actual innovation and economic growth, as opposed 
to discovery science or climate science? Put another way, how does (or 
how should) the reality of limited federal support for computing R&D 
impact cross-discipline prioritization and project selection?

A3.  The portfolio of research and development activities sponsored by 
the NITRD agencies constitutes this country's primary full-spectrum NIT 
R&D enterprise. I mean this in several senses:

      The member agencies of NITRD constitute the only U.S. 
research endeavor that funds investigations across the broad range of 
networking and information technologies. The Program's breadth is a 
vital characteristic because NIT technologies are uniquely 
interdependent and are developed from an inherently multidisciplinary 
basis in the sciences and in engineering. Collaboration among the NITRD 
agencies models the multidisciplinary nature of NIT challenges, and 
NITRD-funded projects often require multidisciplinary collaboration 
among performers.

      NITRD research is performed throughout the Nation--in 
universities, federal research centers and laboratories, federally 
funded R&D centers, private companies, and nonprofit organizations 
across the country. As noted in my testimony, the broad reach of NITRD 
activities generates continuous interaction, information exchange, and 
feedback, which provides new perspectives and insights to federal and 
private-sector stakeholders alike.

      Through its national scope, NITRD funding is the primary 
source of support for the education and training of the Nation's next 
generations of NIT researchers, technical experts, entrepreneurs, and 
NIT industry leaders.

      NITRD investments also span the NIT research spectrum, 
from fundamental inquiry to applied development. The balance varies 
from agency to agency, with NSF and DOE/SC emphasizing foundational 
research and agencies such as DARPA and DHS leaning more toward applied 
development. As noted in my testimony, research in networking and 
information technologies requires both theoretical investigations and 
``use-focused'' applied engineering--and NIT innovations depend on the 
back-and-forth flow of ideas between the theoretical and the practical.

      Although not reported in the NITRD crosscut, NITRD 
agencies are increasingly focusing on the transition from laboratory to 
marketplace, using Small Business Innovation Research (SBIR) and Small 
Business Technology Transfer (STTR) grants to speed the transition to 
practice of NITRD-developed technologies. This important stage of 
advanced development also is explicitly addressed in ``Trustworthy 
Cyberspace,'' the forthcoming strategic plan for R&D in cybersecurity 
developed by NITRD's Cyber Security and Information Assurance (CSIA) 
agencies. Transition to practice is also a focus of such new efforts as 
DHS cybersecurity R&D solicitations and NSF's Innovation Corps. \2\
---------------------------------------------------------------------------
    \2\  http://www.nsf.gov/news/
news-summ.jsp?cntn-id=121879.

    In combination, these characteristics make the NITRD Program a key 
national resource for seeding U.S. innovation of all kinds. If NITRD 
did not exist, today we would be inventing it. Per my response to your 
first question, I believe that the NITRD enterprise reflects 
---------------------------------------------------------------------------
substantial prioritization of federal efforts.

Q4.  Given that participation in AP Computer Science has been flat for 
a decade, as we heard during the hearing, please explain how a new AP 
Computer Science curriculum will be any different. How will it not only 
increase the number of college Computer Science majors, but also 
promote greater ethnic and gender diversity?

A4.  The new AP course--Computer Science (CS) Principles--has been 
designed from the beginning to be engaging, challenging, inspiring, and 
relevant for all students. It is better than the existing CS AP course, 
which had been designed to mimic what colleges had been doing in their 
first course for CS majors. (Many colleges are rethinking their 
introductory sequences, especially their introductory courses for non-
majors, and a number are looking at teaching CS principles.)
    The number of academic computing courses taken by U.S. high school 
students is very low: The percentage of U.S. students taking science, 
technology, engineering, and mathematics (STEM) courses has increased 
over the last 20 years for all STEM disciplines except computer 
science, where participation dropped from 25% to 19%. \3\ High school 
computer science teachers report teaching 8% fewer CS advanced 
placement (AP) courses in 2009 than just two years earlier. \4\ In most 
high schools in the U.S., there is no academic computing course that 
carries college preparatory credit. (Often the only courses offered 
focus on keyboarding or application proficiency in Microsoft Word or 
PowerPoint, for example.) As a result, most U.S. students arrive at 
college with little or no understanding of what computing is as a 
scientific discipline; they know what chemistry is and what physics and 
math are, but they have no idea about computing.
---------------------------------------------------------------------------
    \3\  2009 NAEP High School Transcript Study; http://
www.nationsreportcard.gov/hsts-2009.
    \4\  2009 Teacher Survey, Computer Science Teachers Association; 
http://csta.acm.org/Research/sub/HighSchoolSurveys.html.
---------------------------------------------------------------------------
    The lack of high school experience in computing differentially 
affects women and minorities. Women have few female role models to 
counter the popular images of computing as a singularly male-oriented 
endeavor. Minorities are affected because they are more likely to be at 
low-resourced schools that provide fewer opportunities to study 
computer science.
    How can we get more academic computing courses into our high 
schools? One key has to be Advanced Placement. AP classes--regardless 
of area--carry college preparatory credit in most schools. In addition, 
AP is rigorous, has fidelity of replication (all students must be 
readied for the same test), and is popular with students, their 
parents, school administrators, and college admissions offices. Thus, 
AP provides a single point of leverage to begin curricular change.
    Why hasn't the current AP CS course succeeded in attracting 
students? The current course is a year of Java programming. Students, 
especially those with no prior experience with computers, see little 
reason to take it. It is a low-level, detailed introduction to 
programming. What students need, I believe, is a course that is 
engaging, rigorous, and inspiring. They should learn to program, but 
the course should not be programming-centric. Instead, it should cover 
the design of algorithms, the potentially transformative role of 
technology, the breadth of its applications, its role in enhancing 
creativity and augmenting human capabilities, the uses of big data, 
issues of complexity and computability, and societal impact and ethical 
considerations. In short, the course should be challenging but also 
exciting and relevant to students' lives, no matter whether they end up 
in a NIT career or not.
    CS Principles has been designed to meet these criteria. Its initial 
reception has been positive. It was piloted last year at five 
universities and this year is being taught in roughly 20 universities/
colleges (including some of the top CS departments in the country) and 
40 high schools. More than 200 high schools vied to be part of the 
official College Board pilot, which was restricted to 10 schools.
    In addition to CS Principles, a pre-AP course, called Exploring 
Computer Science (ECS), has been developed with NSF funding and uses a 
similar philosophy. Both CS Principles and ECS appear to be successful 
in attracting women and underrepresented minorities. \5\ For example, 
in four years within the Los Angeles Unified School District, ECS 
enrollment has jumped 553% and now tops 2,000 students, 80% of whom are 
African American or Latino, and 40% are female. In the CS Principles 
course taught at the University of California at Berkeley in the fall 
of 2010, 60% of the best-performing students were female (including the 
top student); in a predecessor course the year before, women had been 
only 40% of the most successful students.
---------------------------------------------------------------------------
    \5\  From reports of Principal Investigators to Program Officers of 
NSF's CE21 Program.

Q5.  The U.S. Department of Education is not currently a member of 
NITRD. Why are they no longer participating in the program, and why 
---------------------------------------------------------------------------
should they be at the table?

A5.  The Department of Education has not been a formal participant in 
the NITRD Program for well over a decade, although ED representatives 
now and again join in NITRD activities. The NITRD Program would welcome 
the Department's re-engagement, and we have sent a formal letter of 
invitation to the Department inviting their participation.

Questions submitted by Ranking Member Daniel Lipinski

Q1.  As I mentioned during the hearing, the fast-moving nature of the 
fields and the ease with which some jobs can be outsourced can add 
uncertainty to careers in NIT, especially in the private sector.

      Do you see the outsourcing of NIT jobs as a problem, and 
if not, why?

      Are there things we can do to avoid training people for 
jobs that might not exist?

      Are there specific areas or programs on which the federal 
agencies should focus?

A1.  These are difficult issues, about which there are undoubtedly 
widely divergent and strongly held viewpoints. I stated in my testimony 
that I believe our global leadership in NIT is under challenge from 
many competitors. The NITRD agencies are greatly concerned about the 
development of the U.S. workforce, given that we are moving at an ever-
increasing pace into a pervasively digital future. Our agencies see the 
need to better prepare our population to live and work successfully in 
the digital world as a national imperative. The U.S. must maintain the 
skilled workforce needed to compete in the global economy. Moreover, 
some outsourcing may directly affect national security--for example, 
the maintenance of the NIT infrastructure of the U.S. industrial base.
    Strengthening the foundations of learning in computer science--
currently woefully inadequate--at every educational level should be a 
national priority. The best way to prepare for change is to provide a 
broad, fundamental education in this subject, beginning at the 
precollegiate level and continuing throughout college. If the education 
system provided that kind of foundational academic grounding, follow-on 
training and retraining activities would then become relatively easier, 
and Americans would be better prepared to adapt to shifting 
technologies.
    Rather than trying to anticipate where employment opportunities 
will lie over the long term (Bureau of Labor Statistics studies \6\ and 
others predict that skilled NIT-related jobs will be among the Nation's 
fastest-growing over the next five years), I believe we need to make a 
commitment as a society to integrate computer science into STEM 
education as vigorously as we set about improving math and science 
curricula after Sputnik. The generations of scientists and engineers we 
produced in that effort are the world's best. We should aim for nothing 
less in computer science and engineering. If we educate for NIT 
leadership, the challenges you cite will eventually recede in 
importance.
---------------------------------------------------------------------------
    \6\  http://www.bls.gov/emp/ep-table-103.htm.

Q2.  The percentage of women obtaining degrees in computer science is 
particularly low, and even more troubling, began to decrease around 
2001 even as female participation in other STEM fields continues to 
slowly increase. The apparent rebound (as of 2009) in the number of 
women obtaining computer science Master's degrees appears to be 
entirely due to an increase in the number of temporary residents 
obtaining such degrees; the number of U.S. citizens and permanent 
residents continues to decrease. Do we understand why American women 
are turning away from computer sciences in such high numbers? Are there 
any data since 2009 to indicate that this trend may be changing? What 
additional steps could we take to increase the recruitment and 
retention of women in computer sciences? How can federal agencies such 
as the National Science Foundation and other NITRD agencies help with 
---------------------------------------------------------------------------
these efforts?

A2.  There is no one reason for the low enrollments of women in 
computing. The lack of engaging, inspiring, and relevant computing 
courses in high schools, as discussed above, is certainly one of them, 
but popular culture perpetuates many false, negative images of 
computing as well. Women say, for example, that they are not as 
interested in computing because it lacks societal impact. \7\ Computer 
scientists are often portrayed as quirky loners who work 24-7 with 
little human interaction. In addition, the male-dominated environment 
often found in computing labs may make women feel unwelcome or 
dismissed. Women (and minorities) see very few role models in the world 
of NIT.
---------------------------------------------------------------------------
    \7\  Recent national data on the status of women in computing are 
available from the National Center for Women & Information Technology, 
http://ncwit.org (e.g., http://ncwit.org/resources.scorecard.html).
---------------------------------------------------------------------------
    The data do not show a substantive change since 2009. \8\
---------------------------------------------------------------------------
    \8\  2009-2010 Taulbee Survey, Computing Research Association; 
http://www.cra.org/resources/crn-archive-view-detail/
undergraduate-cs-degree-production-
rises-doctoral-production-steady/.
---------------------------------------------------------------------------
    How do we fix this? There is no one answer. High school, as 
discussed above, is certainly key, but more is needed. As a society, we 
need to change the popular media's image of computing, and encourage 
high school teachers and guidance counselors to avoid perpetuating old 
stereotypes. Likewise, new best practices for increasing diversity in 
computing at the high school and college levels need to be developed 
and assessed, and guidance counselors and computer science educators 
need to make more of an effort to deploy solid recruitment and 
retention practices--for example, by improving outreach and 
communication with the female parents of female students to make them 
feel welcome at the CS table. The CS education community also needs to 
ensure that the learning environment in our computer science classes 
and departments is more welcoming of the contributions of minorities. 
Female CS undergraduates need to be provided with research experiences, 
internships, and mentoring, and women in NIT-related jobs from all age 
groups and professional levels need to be brought together for 
networking and mentoring.
    How could federal agencies such as NSF help? The Computer and 
Information Science and Engineering (CISE) Directorate at NSF has long 
been committed to reducing the underrepresentation of women in 
computing. Currently, that effort falls under its Computing Education 
for the 21st Century (CE21) Program. CE21 recognizes that efforts to 
reform computing education and efforts to broaden participation must go 
hand-in-hand: It will not be sufficient to ``fix'' computing education 
if we continue to leave out close to 70% of the population (women, 
minorities, and persons with disabilities), and it will not be 
sufficient to engage women and minorities if we do not also build their 
competencies and skills. Both efforts must inform each other. 
Currently, CISE funds a number of projects aimed at increasing the 
participation of women. Two examples include:

      NCWIT (National Center for Women and Technology), which 
functions as a clearing house, resource center, and convener of people 
and events for the whole community, including in particular its 
Academic Alliance (more than 100 university departments), K-12 
Alliance, Social Science Network, and Industry Alliance.

      CRA-W/CDC (Computing Research Association's Committee on 
the Status of Women in Computing Research (CRA-W) and the Coalition to 
Diversify Computing (CDC)), which focuses on research experiences and 
mentoring for undergraduates, and recruitment and retention for 
graduate school and successful early research careers.

    Other NSF efforts include the support of the Grace Hopper 
Celebration of Women in Computing national and regional conferences 
(which bring students and professional women together for technical 
talks and mentoring), the support of the Dot Diva website created by 
WGBH for girls aged 13-17, and work with the National Girls 
Collaborative Project on dissemination of informal education activities 
for girls.

Q3.  Does the National Coordination Office intend to implement the 
PCAST recommendation that a distinct presidential advisory council on 
networking and information technology (NIT), which existed as PITAC 
until 2005, be reconstituted as a standing committee? If not, why not? 
How do you respond to the specific justifications for this 
recommendation that are described in the PCAST report--namely that 
federal NIT investments require continuous attention by a focused 
committee of experts who can provide predictive rather than reactive 
advice?

A3.  This recommendation is under discussion within the Executive 
Branch and among the NITRD agencies.

Q4.  In your testimony, you mentioned the National Initiative for 
Cybersecurity Education (NICE), which is coordinated by NIST, but you 
did not discuss how NICE is coordinated with the education and 
workforce development program component area under NITRD. Can you 
elaborate on how these programs fit together, and in addition how the 
NITRD education programs are coordinated with the broader effort to 
coordinate federal STEM programs? What is the rationale for having 
parallel coordinating structures for NIT education broadly, and 
cybersecurity education specifically? Are there any disadvantages to 
folding NICE activities, including coordination activities, into the 
NITRD program and coordination activities?

A4.  The NITRD Program's Social, Economic, and Workforce Implications 
of IT and IT Workforce Development (SEW) Program Component Area 
supports research on the co-evolution of IT and social and economic 
systems; innovative applications of IT in education; and the 
implications of IT for education and training overall. In recognition 
of the importance of the education and training component, the SEW 
agencies formed a SEW-Education Team to consider ways to help foster 
improved education and training in computer science. These NITRD 
activities are closely coordinated with the NICE program at NIST 
through the appointment of the NICE Lead as the co-chair of the SEW-
Education Team in NITRD. This is enabling SEW-Education members to 
shape broader but complementary activities to help address the systemic 
education problems discussed above. The Team is monitoring the start-up 
activities of the National Science and Technology Council's new 
Committee on Science, Technology, Engineering, and Math Education 
(CoSTEM), called for by the America COMPETES Reauthorization Act of 
2010, and anticipates working to align its activities with the 
directions identified in the CoSTEM five-year strategic plan now under 
development.
    Thank you again for affording me the opportunity to address the 
important questions you raise on a topic so vital to the future of our 
country. On behalf of the NITRD Program, I look forward to working with 
you to sustain our Nation's leadership in networking and information 
technologies in the years to come.
Responses by Dr. Edward Lazowska, Director, eScience Institute, Bill 
        and Melinda Gates Chair, University of Washington

Questions submitted by Chairman Mo Brooks

Q1.  As discussed at the hearing, Congress recognizes the value and 
importance of networking and information technology (NIT) research and 
development (R&D) funding; however, our existing budget constraints 
make prioritizing a requirement. Please detail how you would prioritize 
NIT R&D funding, what the top priorities would be, and how savings can 
be achieved within the NITRD portfolio.

A1.  As you know, I co-chaired the Working Group of the President's 
Council of Advisors on Science and Technology for the recent PCAST 
review of the NITRD program. One of the key recommendations in that 
report was a call to establish ``a broad, high-level standing committee 
of academic scientists, engineers, and industry leaders dedicated to 
providing sustained strategic advice in NIT.'' We view the 
establishment of such a group (really, the re-establishment of such a 
group--it would be analogous to the President's Information Technology 
Advisory Committee which existed under the Clinton and Bush 
administrations) as essential. Such a group would be perfectly 
positioned to answer the question you have posed, and many other 
important questions.
    With that said, the PCAST report did attempt to address this 
question. The report includes recommendations for research areas of 
particular importance to national priorities--health information 
technology, energy and transportation, and the security and robustness 
of cyber-infrastructure--as well as a call for increased investment in 
a number of fundamental NIT research areas that will accelerate 
progress across a broader range of priorities, including: privacy and 
security; human-computer interaction; data analytics, including data 
collection, storage, management, and automated large-scale analysis 
based on machine learning and predictive modeling; and computing in the 
physical world, through advanced sensor and control networks, 
innovative robotics and other means. I personally agree with these 
recommendations and would offer them to the committee.
    It is, however, difficult to discuss the size of the investments 
required, if only because it is very difficult to get an accurate 
understanding of the size of the current investment. As I reported at 
the hearing, and as PCAST mentioned in its report, agency reporting of 
NIT R&D investments is, in places, very questionable. While investments 
reported by the National Science Foundation and DARPA are reasonably 
accurate portrayals of NIT R&D spending by those agencies, it appears 
that much of the spending reported by other agencies is more accurately 
characterized as ``NIT investments in support of other areas of 
research'' rather than ``NIT research funding.'' In fact, PCAST found 
that between 89 and 98 percent of the reported investment in NIT R&D by 
the National Institutes of Health was not NIT R&D, but rather the use 
of IT in support of other areas of research--for example, public 
databases of research-related information. While this is valuable and 
appropriate research spending, it's not NIT R&D, and calling it NIT R&D 
leads to the belief that we are spending far more on NIT R&D than we 
really are.
    This leads to another key point that I raised in response to a 
question during the hearing. You of course appreciate the role of NIT 
R&D in driving our economic competitiveness, in achieving our major 
national and global priorities, in accelerating the pace of discovery 
in all other fields, and in achieving the goals of open government. 
Quoting PCAST, ``As a field of inquiry, NIT has a rich intellectual 
agenda--as rich as that of any other field of science or engineering. 
In addition, NIT is arguably unique among all fields of science and 
engineering in the breadth of its impact.'' Despite this, the federal 
investment in NIT R&D is exceedingly modest by any measure. It is 
important for the nation to identify those fields of science and 
engineering in which we must lead the world. It is impossible to 
imagine that any field would have higher priority than NIT. Compare the 
federal NIT R&D investment, though, to that of other fields!
    A final point: While high performance computing remains a critical 
area of focus for the NITRD program: (1) other areas of NIT--for 
example, robotics, and large-scale data analysis--have risen to equal 
levels of importance as measures of our international competitiveness; 
(2) this is true even for applications in national security and 
scientific discovery, traditionally the bastion of HPC, where large-
scale data analysis (which requires significantly different 
architectures and algorithms) is of great and growing importance; and 
(3) even within numerical computing, we need to rely on a better metric 
than ``FLoating-point Operations Per Second'' (FLOPs)--the default 
measure of supercomputing ``power'' on lists of the world's most 
powerful supercomputers--to assess our progress and leadership in the 
space; landing at the top of the Top 500 List is exceptionally 
expensive and does not necessarily guarantee the nation will build a 
machine that's particularly useful; far more valuable a priority is to 
invest in research that could allow for a leap frog of current high 
performance computing technology.

Q2.  In your testimony, you quoted from the PCAST report, restating 
that, ``All indicators --all historical data, and all projections--
argue that NIT is the dominant factor in America's science and 
technology (S&T) employment, and that the supply of that talent is and 
will remain large.'' When did NIT become the dominant factor in S&T 
employment? Are other S&T sectors dwindling, or is this due to the 
growth in the NIT field?

A2.  The data on the growth of the IT workforce need comes from 
projections developed by the U.S. Department of Labor's Bureau of Labor 
Statistics. These are 10-year projections released every two years. For 
at least a decade, these projections have shown that employment growth 
in the ``computer science and mathematics'' sector will far outstrip 
growth in all other science and technology fields combined (and this 
growth is almost entirely in the computer science fields). In fact, the 
individual formerly in charge of these forecasts for BLS famously once 
said ``All other fields of science and engineering are hiding behind 
information technology'' by which he meant that the vast majority of 
the overall STEM (Science, Technology, Engineering, and Mathematics) 
workforce gap and job growth was in the computing fields.
    My read of the statistics is that this represents the pervasiveness 
of NIT throughout the U.S. economy. The growth in employment is not 
just due to the growth of the IT sector, though that's certainly an 
element, but in the use of NIT across industries, from health care to 
banking to transportation to energy and beyond, as a means of improving 
productivity and gaining other efficiencies. So it's not that these 
other fields have become less important, it's that NIT and NIT workers 
have become increasingly important to these other fields.

Q3.  Your testimony stated that ``The workforce needs of the IT fields 
going forward demand a sustained effort to increase the number of 
students going into computing fields.'' Why aren't students entering 
the computing fields today? What can academic institutions and industry 
do to encourage student involvement and engagement?

A3.  The good news is that, after a period of declining enrollments 
that began at around the time of the "Dot-com bust," data from the 
Computing Research Association's Taulbee Survey tracking enrollments 
and graduation rates indicates that the trend has reversed and that 
student interest in computing majors has increased in each of the last 
two years. Anecdotal evidence from my own department and from 
colleagues around the country suggests that this year's survey results 
will likely show much larger increases in enrollments. At the stronger 
programs across the nation, enrollment is booming.
    Enrollments in computer science are somewhat cyclical and do 
correlate to some degree with the overall state of the IT economy. In 
the "Dot-com boom" times, computer science enrollments increased faster 
than many university programs could handle. The bust that followed 
decreased enrollment, but it appears the highly visible success of many 
NIT-related companies like Facebook, Google, and Apple may be 
motivating large numbers of new students to pursue computing-related 
careers, resulting in the positive numbers we see now.
    While enrollments are important, it's also important that we retain 
an adequate number of those students interested in research careers 
through their Ph.D.s. In those cases, federal support for university 
research plays a crucial role in supporting the researchers who will 
employ those students as graduate researchers. These graduate 
researchers are like the lubrication in the innovation ecosystem, 
enabling the transfer of ideas gleaned from fundamental research into 
industry and the marketplace. Federal support for research not only 
enables that fundamental research, it helps train the students who will 
take that fruits of that research into industrial research labs with 
them or into their own startup companies, such as Google, which, only a 
dozen years after its emergence from Stanford, has a market 
capitalization of $190,000,000,000, employs 32,000 people, and is a 
verb.

Questions submitted by Ranking Member Daniel Lipinski

Q1.  As I mentioned during the hearing, the fast moving nature of the 
fields and the ease with which some jobs can be outsourced can add 
uncertainty to careers in NIT, especially in the private sector.

      Do you see the outsourcing of NIT jobs as a problem, and 
if not, why?

      Are there things we can do to avoid training people for 
jobs that might not exist?

      Are there specific areas or programs on which federal 
agencies should focus?

A1.  Let me say at the outset that the Bureau of Labor Statistics 
workforce projections cited in my testimony and in response to a 
previous question--which indicate that domestic workforce demand in the 
computing sector during this decade will far outstrip demand in all 
other fields of science, technology, engineering, and mathematics 
combined--account (as best BLS can) for offshoring. Computing is a 
field of enormous opportunity for well-prepared people in our country, 
and will continue to be so.
    The issues of outsourcing within the IT sector are complex, and 
determining the impact is made difficult by the uneven quantity, 
quality and objectivity of the data available. But the computing 
community has tried to understand some of the issues involved. In a 
report entitled ``Globalization and Offshoring of Software,'' a task 
force of the Association for Computing Machinery concluded the 
following about potential increases in ``offshoring'' of NIT work:

      Globalization of the software industry is likely to 
continue to grow.

      Both anecdotal evidence and economic theory indicate that 
offshoring between developed and developing countries can, as a whole, 
benefit both.

      Because of the lack of good data, skepticism is warranted 
regarding claims about the number of jobs to be offshored and the 
projected growth of software industries in developing nations.

      Standardized jobs are more easily moved from developed to 
developing countries than higher-skill jobs, but competition in higher-
end skill jobs is increasing.

      To stay competitive in a global IT environment and 
industry, countries must adopt policies that foster innovation. 
Policies that improve a country's ability to attract, educate, and 
retain the best IT talent are critical. Educational policy and 
investment is at the core.

    The bottom line is that information technology remains a sector in 
which the nation must retain leadership in order to be globally 
competitive--competitive economically and technologically superior for 
our national defense. The ease with which some aspects of NIT can be 
outsourced only heightens the need to ensure that the ecosystem for 
research in the U.S. remains strong. As a nation, we can afford to 
offshore technical support. We cannot, however, afford to offshore 
innovation. There must be a far greater emphasis on Bachelors, Masters, 
and Doctoral education at strong institutions which prepare students to 
be innovators.

Q2.  The percentage of women obtaining degrees in computer science is 
particularly low, and even more troubling, began to decrease around 
2001 even as female participation in other STEM fields continues to 
slowly increase. The apparent rebound (as of 2009) in the number of 
women obtaining computer science Master's degrees appears to be 
entirely due to an increase in the number of temporary residents 
obtaining such degrees; the number of U.S. citizens and permanent 
residents continues to decrease. Do we understand why American women 
are turning away from computer sciences in such high numbers? Are there 
any data since 2009 to indicate that this trend may be changing? What 
additional steps could we take to increase the recruitment and 
retention of women in computer sciences? How can federal agencies such 
as the National Science Foundation and other NITRD agencies help with 
these efforts?

A2.  This is an issue of enormous concern. It is not just a matter of 
``equity'' or of ``workforce''--it is an issue of ``quality,'' since 
computer systems intended for use by the full breadth of our population 
must be designed by individuals who reflect the full breadth of our 
population.
    This is also an area of intense focus, over many years. This is, in 
fact, bad news: The decrease in female computer science enrollment 
began in the 1980s, not in 2001, and we have been working to reverse 
the trend for many years. The discouraging numbers that you see today 
are despite decades of serious effort at assessing and addressing the 
problem.
    The good news is that recent trends are positive. For example, 
between 2009 and 2010, the percentage of computer science Bachelor's 
degrees from research universities received by women increased, and the 
number of U.S. citizens and permanent residents enrolled in mathematics 
and computer science grew faster than the number of temporary 
residents.
    I personally feel that the greatest problem is persistent 
stereotypes (think ``Dilbert'' or ``geek''). The bad news about 
stereotypes is that they have at least a grain of truth in them, and 
thus they die hard.
    What can federal agencies do? Support a greater number of graduate 
fellowships for women, because role models (as faculty members and 
researchers) are important. Support the Computing Research Association 
Committee on the Status of Women in Computing Research (CRA-W), which 
runs a number of highly effective programs. Support the National Center 
for Women in Information Technology (NCWIT), which also runs a number 
of highly effective programs. Support CRA-W and NCWIT so that they can 
focus on their work, rather than on their survival. Finally, support 
current efforts by the National Science Foundation's CISE Directorate 
to revamp the Advanced Placement curriculum in computer science, and to 
train 10,000 teachers in the next few years to teach this new 
curriculum--the CS 10K effort. All of these efforts are starving.

Q3.  In your opening statement, you said that 60 percent of the new 
jobs created in this country over the next decade will be related to 
NIT. Could you explain the research or studies behind this figure? Will 
traditional courses of study (e.g., computer science) effectively 
prepare students for these sorts of careers? Are new pedagogical 
approaches needed? Do we need to revisit how we teach other areas, like 
engineering?

A3.  The Department of Labor's Bureau of Labor Statistics every two 
years assembles workforce projections for a broad range of fields over 
the next decade. Over the last several cycles--going back more than a 
decade--BLS projections have consistently shown that projected 
increases in the computing fields completely outstrip all other fields 
of science, technology, engineering, and mathematics (STEM) combined. 
From what I can see, this is due primarily to two factors: the growth 
of the IT sector--that is, the producers of NIT hardware, software and 
services--and the increased use of IT across all other sectors. In 
fact, it is the latter factor which undoubtedly accounts for the 
largest share of the overall increase. NIT is being used to enable all 
other sectors to work more productively--to produce more while using 
fewer resources. Each of these sectors will require skilled workers to 
incorporate these new technologies effectively, creating literally 
millions of new opportunities for NIT professionals.
    ``Computational thinking'' is transforming all other disciplines, 
and they are somewhat slow to respond. We need to integrate 
``computational thinking'' into K-12 STEM curricula, and we need to 
integrate computation and computational thinking into other fields of 
science and engineering. Graduate programs in "eScience" (data-
intensive science) are emerging today, and are an important trend that 
should be supported. eScience is the future of all science.
Responses by Dr. Robert Sproull, Director of Oracle Labs, retired

Questions submitted by Chairman Mo Brooks

Q1.  As discussed at the hearing, Congress recognizes the value and 
importance of networking and information technology (NIT) research and 
development (R&D) funding; however, our existing budget constraints 
make prioritizing a requirement. Please detail how you would prioritize 
federal NIT R&D funding, what the top priorities would be, and how 
savings can be achived within the NITRD portfolio.

A1.  I recommend that long-term research investments be protected, that 
is, that they not be reduced more than other spending. True, long-term 
NITRD research is risky, but it has consistently produced results with 
huge economic benefits to the nation, including growingjobs and federal 
tax revenues.

Q2.  As part of the PCAST working group assessing the NITRD program, 
you said the group ``had trouble determining the levels of research 
investment in different areas beecause of difficulties in labeling and 
measuring expenditures.'' You went on to say that industry makes 
``clear distinctions between different kinds of investment in IT, in 
part so that the investments can be balanced appropriately.'' Why can't 
the federal government do the same?

A2.  As I detailed in my written testimony, private sector accounting 
distinguishes several categories for IT expenditures: expense of 
running IT services (including cost of depreciation of computers and 
other capital equipment); expense of development of IT services (e.g., 
software engineers developing ``routine IT''); and expense of NIT 
research to explore new and risky innovations in NIT products and 
services. These categories clearly distinguish operating, development, 
and research expenses.
    It would be helpful to distinguish these categories in federal 
spending as well. However, federal accounting is different--there is no 
equivalent of depreciation, for example. It would be helpful, however, 
for the NITRD agencies to report their investments in categories 
similar to these, as recommended in the PCAST NITRD report.

Questions submitted by Ranking Member Daniel Lipinski

Q1.  As I mentioned during the hearing, the fast moving nature of the 
fields and the ease with which some jobs can be outsourced can add 
uncertainty to careers in NIT, especially in the private sector.

Q1a.   Do you see the outsourcing of NIT jobs as a problem, and if not, 
why?

A1a.  I distinguish between ``outsourcing,'' in which a firm contracts 
with another firm to perform work, generally resulting in fewer jobs in 
the first firm and more in the second; and ``offshoring,'' in which a 
firm in one country (the U.S.) pays for work performed in another 
country, whether using its own employees or those of a contractor. The 
principal concern is with offshoring, which shifts jobs out of the 
United States.
    In my experience working with highly trained and innovative 
researchers, we seek people with the best and most appropriate talent, 
regardless of their nationality or location. We prefer to attract 
researchers to move to a domestically-located laboratory, but if they 
cannot we will hire them to work in their preferred location. A clear 
step to insure domestic hiring is to have strong graduate education in 
the United States and provision for its graduates, of whatever 
nationality, to be able to work in the U.S. Easing immigration of 
excellent researchers trained abroad so they can work in our domestic 
labs is also important.
    This same approach of hiring the best wherever they are has caused 
foreign firms to open NIT research labs in the United States, where 
they employ U.S. citizens.
    In NIT research, ``offshoring'' finds the strongest talent. If our 
nation can educate and retain the best NIT talent, domestic and foreign 
companies both will employ it in the U.S.

Q1b.   Are there things we can do to avoid training people for jobs 
that might not exist?

A1b.  Because of the time lag between a student's selection of study 
areas or majors and their entry into the workforce, typically after 
college, there is an unavoidable possibility of cyclic workforce 
surpluses and deficits. Students are quite shrewd in judging future 
demand; it seems to me that it would be hard to devise a better scheme.
    Short-term training is less problematic: the job market changes 
slowly enough that a shrewd student will not embark on fruitless 
training.

Q1c.  Are there specific areas or programs on which the federal 
agencies should focus?

A1c.  To ensure the best NIT talent is developed in the United States 
rather than offshore, we must build strong STEM education programs in 
K-12 as well as in colleges. While a STEM background is not absolutely 
essential for NIT careers (philosophers who love logic can become 
exceptional software engineers!), STEM education must be strong if the 
NIT workforce is to be strong.

Q2.  The percentage of women obtaining degrees in computer science is 
particularly low, and even more troubling, began to decrease around 
2001 even as female participation in other STEM fields continues to 
slowly increase. The apparent rebound (as of 2009) in the number of 
women obtaining computer science Master's degrees appears to be 
entirely due to an increase in the number of temporary residents 
obtaining such degrees; the number of U.S. citizens and permanent 
residents continues to decrease. Do we understand why American women 
are turning away from computer sciences in such high numbers? Are there 
any data since 2009 to indicate that this trend may be changing? What 
additional steps could we take to increase the recruitment and 
retention of women in computer sciences? How can federal agencies such 
as the National Science Foundation and other NITRD agencies help with 
these efforts?

A2.  I do not feel qualified to answer this question. I defer to my 
academic colleagues.
Responses by Dr. Robert Schnabel, Dean, School of Informatics, Indiana 
        University

Questions submitted by Chairman Mo Brooks

Q1.  As discussed at the hearing, Congress recognizes the value and 
importance of networking and information technology (NIT) research and 
development (R&D) funding; however, our existing budget constraints 
make prioritizing a requirement. Please detail how you would prioritize 
federal NIT R&D funding, what the top priorities would be, and how 
savings can be achived within the NITRD portfolio.

A1.  I agree that the NITRD program should prioritize funding for 
research and development. As I mentioned in my testimony, health IT is 
an area of particularly great national importance. The challenges that 
this area addresses range from assuring that the Federal Government and 
the nation's health care system meets the needs of modernizing and 
standardizing health records, to providing powerful and easy-to-use 
information technology systems that support health care providers, to 
creating tools and systems that allow individuals to monitor and 
improve their own health independently. I also agree with the recent 
PCAST report that recommended prioritizing funding toward IT 
applications in security, energy and transportation.
    As Dr. Lazowska noted in his testimony, it is difficult to assess 
the overall size of research funding focused on computing since it is 
co-mingled with funding for IT infrastructure; therefore, it is 
difficult to assess potential savings. Getting a better handle on what 
NITRD funding actually is going toward research versus funding for 
information technology infrastructure that supports research in other 
areas would help better assess where savings could be achieved. The 
highest priority of the NITRD portfolio should be funding focused on 
high-risk, possibly high-reward IT research.
    I also support the recommendation of the recent PCAST report for a 
standing committee of networking and IT specialists to oversee the 
federal IT research portfolio. This committee could continually review 
the program and help drive the establishment of priorities for funding.

Q2.  You indicate that the Federal Government needs to include and 
clearly define computer science in federal education programs and 
create pre-service and professional develpment opportunities for K-12 
computer sceince teachers. Computer science is included in many federal 
education programs. Are there any you know of that specifically 
prohibit computer science as an eligible field? If so, please describe.

A2.  I am not aware of any programs intended to address K-12 
educational needs that explicitly prohibit (by law or regulation) 
computer science as an eligible discipline; however, there are many 
instances where the additional restrictions of the programs implicitly 
rule out or discourage computer science or put computer science 
programs at a severe competitive disadvantage. Education programs--such 
as the ``STEM'' education programs within the Committee's 
jurisdiction--often erect barriers that either bias against or exclude 
computer science because of where computer science currently exists 
within the U.S. K-12 system. For example, as described further below, 
many federal programs expect proposals to build up either a strong 
assessment base in the discipline and/or certification of teachers, or 
are linked to ``core academic subjects.'' Computer science teachers 
face much different issues in the K-12 environment than teachers of 
mathematics, and existing programs often do not take this into account. 
These conditions create a chicken-and-egg problem that makes it 
difficult for computer science to be included.
    Because of the accountability provisions in No Child Left Behind 
and the focus of States on that Act's ``core'' disciplines in 
developing high school graduation requirements, investments in 
curriculum, pedagogy and professional development very often are 
focused on ``core'' courses. There also has been a pronounced shift 
towards presuming that States will adopt the work of the ``Common Core 
Standards Initiative'' and its ``college and career ready standards'' 
in both the competitive grant guidance for the Race to the Top program 
and in the President's proposed Fiscal Year 2011 budget. In practice, 
this means schools, States and federal programs emphasize mathematics, 
reading, and natural sciences. Therefore, well-meaning federal 
legislation intended to improve STEM education broadly often does not 
include computer science at the state and local levels, since it is not 
typically considered part of this ``core.''
    This same issue plays out in programs authorized by the COMPETEs 
Act and the Elementary and Secondary Education Act, putting K-12 
computer science education in a classic ``Catch-22.'' Because computer 
science is not part of the core, it does not have the same level of 
assessments or teacher support as core programs that education policy 
makers seek to improve course offerings, but it is difficult to develop 
these without being in the core.
    An important example is the NSF's Math and Science Partnership 
program. This program has five types of awards, including Targeted 
Partnerships intended for ``a specific disciplinary focus in 
mathematics or the sciences.'' At a high level, the program is broadly 
STEM focused seemingly to be a ``big tent'' for all STEM-related 
disciplines. In fact, the COMPETEs Act amended what was then current 
law to clarify the scope of the program to include all of the STEM 
disciplines. However, guidance to grant applicants asks specifically 
for baseline data on how the proposals will improve student achievement 
in mathematics and/or science standards. However, the significant 
public investments in mathematics and science assessments rarely 
address computer science. Therefore, computer science proposals have 
difficulty meeting the baseline data requirements. This puts computer 
science proposals at a distinct disadvantage relative to mathematics 
and science proposals, which deters would-be applicants and creates a 
barrier for the computer science field.
    This same type of Catch-22 pertains to multiple COMPETEs Act 
programs that would require ``highly qualified'' computer science 
teachers. For example, the MSP award category Teacher Institutes for 
the 21st Century was created by the COMPETEs Act. The scope of the 
program is to serve STEM teachers who ``are considered highly 
qualified.'' The COMPETEs Act references the underlying definition of 
``highly qualified'' in the Elementary and Secondary Education Act 
(ESEA). One of the requirements for teachers to be highly qualified is 
certification and demonstrated knowledge in the subject area in which 
they teach. (Other programs in COMPETEs that rely on the highly 
qualified criteria include Teachers for a Competitive Tomorrow and the 
Robert Noyce Teacher Scholarship Program.) Very few States have 
certification programs for computer science teachers. Thus, although 
these programs do not explicitly prohibit computer science, their 
requirements often have the effect of making computer science 
ineligible. These same issues appear as barriers for schools 
distributing Title II funding under the Elementary and Secondary 
Education Act.
    In summary, the ``STEM'' designation in numerous federal programs 
often does not recognize and account for the reality that the computer 
science discipline faces in the schools. Many federal STEM programs are 
designed for the core academic structure in schools across the country. 
At the state and local level computer science courses can be classified 
as mathematics or science courses, but in 35 states it is simply an 
elective. Effectively this means that computer science teachers and 
courses are treated differently than mathematics and science teachers 
and courses. Clearly including computer science in STEM education 
programs, and clearly defining that this means the conceptual aspects 
of computing (as opposed to basic technology literacy) as the 
President's Council of Advisors on Science and Technology recommended, 
would help address the lack of applicability of federal STEM programs 
to computer science in numerous states across the nation.

Q3.  Given that participation in AP Computer Science has been flat for 
a decade, as we heard during the hearing, please explain how a new AP 
Computer Science curriculum will be any different. How will it not only 
increase the number of college Computer Science majors, but also 
promote greater ethnic and gender diversity?

A3. The new computer science AP course, Computer Science Principles, 
follows the model of several other redesigns of AP courses by focusing 
on the fundamental aspects of the field in a way that is engaging, 
relevant to the real world and inspiring for all students. It seeks to 
show how important computer science is to many areas of our society 
while giving a broad introduction to the subject area. Thus, this 
largely a ``breadth'' approach. The increased appeal of the course is 
tied to this approach.
    In contrast, the current AP Computer Science A course is largely a 
``depth'' approach that goes deeply into the specifics of the Java 
computer programming language and gives little sense of the broad 
applicability of computer science. Students without prior programming 
knowledge often find the course difficult or unapproachable, and the 
course makes no attempt to appeal to students who are attracted more by 
the broad societal applicability of computer science than by the 
technical material for its own sake. The demographics of the current AP 
CS A course are very clear; the AP test data shows that largely white 
males take it and in small numbers relative to other AP ``STEM'' 
disciplines. Teachers often have pointed out that the current course is 
not an ideal first course for students new to computer science, and 
because computer science courses have little room in the curriculum, 
the current AP computer science course may be the only opportunity 
students have to get exposure to this field in secondary K-12 
education.
    Computer Science Principles contains rigorous content and includes 
a programming component, but uses programming as a way of exploring the 
broader concepts of computer science. Unlike AP CS A, it is more about 
learning core computer science concepts and much less about learning 
the nuances and syntactical specifics of a particular language.
    One reason to expect that this new approach to AP computer science 
will attract more students and a more diverse set of student is the 
positive results of other ``breadth'' approaches towards teaching 
computer science. For example, a new high school level course, 
Exploring Computer Science, has also adopted this foundational, problem 
solving approach to introducing computer science to great success. The 
availability of this NSF-funded course in urban, public schools has led 
to rapid and dramatic results. Over the past four years, over 4,000 Los 
Angeles public high school students across 25 high schools have taken 
this college preparatory course--of which 40 percent of enrolled 
students were girls and 80 percent were students of color. This type of 
foundational and contextual approach to computing around which the new 
AP Computer Science Principles course also is framed holds great 
promise in drawing students into computer science, particularly 
students who have been historically underrepresented in computer 
science.

Q4.  Many of your recommendations fall outside of the jurisdiction of 
this Committee and, in some cases, the scope of the Federal Government. 
However, I am curious. I understand how you could teach middle and high 
school computer science skills like programming, software development, 
and the use of algorithms; but exactly how would you teach a first or 
second grader computer science beyond basic skills?

A4.  Like all academic disciplines, computer science involves the 
development of knowledge and skills that are best introduced and 
mastered incrementally as students move through their education 
experience. For this reason, the Computer Science Teachers Association 
has developed the K-12 Computer Science Standards which provided 
learning outcomes keyed to students' intellectual development at 
several milestones throughout their schooling experience. These 
outcomes are organized into two levels at elementary school, grades K-3 
and 3-6. Elementary school students are introduced to foundational 
concepts in computer science by integrating basic skills in technology 
with simple ideas about computational thinking. For example in grades 
K-3, a student begins to develop an understanding of sorting by 
beginning to arrange information into a useful order. In grades 3-6 
students can begin to develop a simple understanding of an algorithm 
that includes sequencing of steps and sorting of information. In Grades 
3-6 students can also begin learning the foundations of algorithmic 
problem solving starting with breaking down a larger problem into 
smaller steps that are easily solved.
    To illustrate an even more specific answer to the question, CSTA's 
new draft standards for K-12 computer science reference the resources 
at Computer Science Unplugged (http://csunplugged.org/), which include 
some specific approaches to meeting the learning objectives of sorting 
or how computers represent numbers as 1s and 0s in grades K-2.

Questions submitted by Ranking Member Daniel Lipinski

Q1.  As I mentioned during the hearing, the fast moving nature of the 
fields and the ease with which some jobs can be outsourced can add 
uncertainty to careers in NIT, especially in the private sector.

Q1a.   Do you see the outsourcing of NIT jobs as a problem, and if not, 
why?

A1a.  In short, no, this does not appear to be a major problem. We 
increasingly operate in a global economy and major firms increasingly 
employ a global workforce. As long as trade barriers generally are low, 
markets and talent exist in other countries, and our nation's standard 
of living and wage scales are higher than those in the developing 
world, it will be attractive for firms to locate some jobs overseas, 
particularly lower-level jobs. But if in conjunction, the employment 
situation in NIT continues to thrive in the United States, which it 
does, this is a ramification of a global economy but not a sign of 
trouble for the U.S. IT industry.
    Every indication, both from government studies and data from 
corporations and universities, is that the employment situation in NIT 
fields in the U.S. shows a considerably greater demand for workers than 
the current supply, and this it will continue this way into the 
foreseeable future. The latest Bureau of Labor Statistics projections 
for 2008-2018 predict a 22.3% growth in employment in the computing 
sector (the ``computer specialists'' category 15-1000 in the report) 
with 1,384,600 job openings over this 10-year period out of a projected 
4,187,000 total jobs, meaning that 33% of the jobs will need to be 
filled with new workers. Those of us who work in computing fields in 
academia see this phenomenon up close, as the corporate demand for our 
students always exceeds the graduates we can supply. In the last two 
years, this situation has heated up even further. There also is 
verification of this same excess of NIT jobs vs. supply of workers from 
the corporate sector; for example, in July 2011, Microsoft general 
counsel Brad Smith reported in testimony before the Senate Judiciary 
Subcommittee on Immigration, Refugees and Border Security that as of 
May 2011, Microsoft had over 2,600 vacant computer science positions. 
In addition, the recent Dice report ``America's Tech Talent Crunch'' 
identified shortages of information technology talent in virtually all 
of the key IT markets in the nation including Silicon Valley, Seattle, 
Dallas, Boston, Atlanta, New York, and the DC/Virginia/Baltimore 
region.
    Finally, it should be emphasized that in general, it is the lower-
level, more commodity computing jobs that go overseas. Of the 10 BLS 
subcategories under computer specialists, the only one that shows a 
small projected decrease (2.9%) is ``computer programmers,'' but this 
is far more than offset by a projected 30.4% increase in ``computer 
software engineers, systems software,'' a category that is also 
considerably more highly paid. In many ways, a computer software 
engineer is an upgraded version of a computer programmer. This 
illustrates the general point: as long as we train U.S. citizens for 
the higher-level computing jobs, the ones that not only require 
technical skills but also applications knowledge and the ability to 
work with clients and customers, there will be plentiful job demand for 
them.

Q1b.   Are there things we can do to avoid training people for jobs 
that might not exist?
A1b.  There definitely are steps that we should take, and that many 
U.S. universities are taking, to make sure that the education and 
training that students receive prepare them well for the needs of the 
U.S. workforce. The main one is consistent with the points mentioned 
above: we need to prepare people for the modern, quickly-evolving world 
of computing technology where computing is applied in a huge variety of 
business, scientific, social and other contexts. Industry increasing 
looks for employees who combine technical expertise with the ability to 
interface with applications that range from health care to media, and 
who have the ability to work on diverse teams and with clients. U.S. 
programs in computer science, information technology, informatics and 
related fields increasingly are taking this orientation and preparing 
students well, although the number of students specializing in these 
fields remains insufficient.
    The other crucial consideration is to realize that the computing 
world evolves so quickly that no training can prepare people 
sufficiently for a 10-year career, to say nothing of a 40-year career; 
students and workers need to be prepared to constantly learn new areas 
and skills. An education that goes beyond the purely technical to 
combine the applications knowledge and communication skills mentioned 
above also prepares the student well for lifelong learning. 
Universities will continue to need to evolve their curricula to meet 
the demands of a fast-moving industry, and corporations will need to 
compliment this education with more specialized training that keeps 
employees current and teaches skills that are particular to that 
company.

Q1c.  Are there specific areas or programs on which the federal 
agencies should focus?

A1c.  The federal agencies will be best served by supporting computing 
education broadly within the U.S. at the K-12 and higher education 
levels. The agencies should encourage educational approaches that 
increase the quantity and diversity of students who are attracted to 
NIT fields. These approaches include imparting not only the technical 
content of computing but also a sense of the wide range of applications 
and situations where computing leads to a better world. Applications in 
fields including health care, media and communications, energy, 
transportation, arts and entertainment should be made apparent to the 
students.

Q2.  The percentage of women obtaining degrees in computer science is 
particularly low, and even more troubling, began to decrease around 
2001 even as female participation in other STEM fields continues to 
slowly increase. The apparent rebound (as of 2009) in the number of 
women obtaining computer science Master's degrees appears to be 
entirely due to an increase in the number of temporary residents 
obtaining such degrees; the number of U.S. citizens and permanent 
residents continues to decrease. Do we understand why American women 
are turning away from computer sciences in such high numbers? Are there 
any data since 2009 to indicate that this trend may be changing? What 
additional steps could we take to increase the recruitment and 
retention of women in computer sciences? How can federal agencies such 
as the National Science Foundation and other NITRD agencies help with 
these efforts?

A2.  There is no easy answer as to why girls and women are opting out 
of computing; that is an important reason why, in 2004, the Computer 
and Information Science and Engineering directorate of the National 
Science Foundation provided funding to start NCWIT, the National Center 
for Women & Information Technology, whose mission is to significantly 
increase girls' and women's meaningful participation in computing. 
Since then, over 300 organizations (universities, corporations and non-
profits) have worked together to understand the underlying causes and 
possible solutions of the low participation of women in computing. 
Causes include:

      The lack of rigorous, relevant and inclusive computer 
science instruction in K-12 education; curriculum needs to be 
formulated in a manner that attracts all students and not just 
predominantly males, by combining exposure to the applications and 
societal implications of computing with purely technical content.

      The lack of relevant and inclusive computer science 
introductory instruction at the post-secondary level.

      The pervasive image of computing as a ``white geeky 
male'' endeavor, and a lack of understanding that most computing 
professionals apply computing to a variety of fields ranging from 
health care to media to entertainment.

      The lack of exposure for computing educators to research 
concerning unintended bias and stereotype threat.

      The lack of understanding about computer science, and 
hence encouragement to pursue computer science education, by adult 
stakeholders.

    While women's overall participation in university computer science 
education has not yet turned the corner, members of NCWIT's Academic 
Alliance reported recently that the percentage of female enrollments in 
their majors has increased, according to a 2010 annual survey conducted 
by NCWIT's external evaluator. Sixty percent of the survey respondents 
reported increased enrollment of women and 39% reported increased 
graduation rates. It is expected that growth in graduation rates will 
continue, since students take four to five years after enrolling to 
graduate. National data are beginning to corroborate members' reports. 
Although women's share of all computer and information science 
Bachelor's degrees awarded in the U.S. declined slightly between 2007 
and 2010, two NCWIT-only datasets as well as the Computing Research 
Association dataset (one-third of which are NCWIT members) show an 
increase in women's share of degrees awarded. Furthermore, as the NCWIT 
academic alliance membership grows, so does NCWIT's influence on the 
overall percentage of computing degrees awarded in the U.S. In 2010, 
NCWIT academic alliance member organizations awarded 21% of the nearly 
41,000 Bachelor's degrees awarded in computer and information sciences, 
an increase of 8% from 2007, when NCWIT members graduated 13% of all 
CIS degrees.
    Increasing women's participation in information technology 
education and workforce requires a systemic approach that gives 
attention to multiple factors: recruitment of women by making them 
aware of the opportunities, particularly for helping society, that a 
computer career provides; development of computing curricula that 
combine technical skills, applications, and communication and teamwork 
skills; support for women students that recognizes that women often 
enter computing programs less confident of their abilities than men; 
and support for women in the technical workplace that provides the 
flexibility to balance careers and lives. Ultimately this requires 
high-level commitment by universities and employers to the importance 
of this issue to our society's economic competitiveness. Organizations 
taking this systemic approach do show results; as just one example, my 
own School of Informatics and Computing at Indiana University 
Bloomington has succeeded in doubling the number of women undergraduate 
majors, from 75 to 150, in less than two years by taking such a 
comprehensive approach.
    Three key things that federal agencies can do to help with these 
efforts are to include programs that support diversification of the NIT 
student body and workforce in their funding portfolios, to support 
organizations that are producing successes in these areas, and to 
promote the importance of a diverse NIT workforce for our nation's 
economic health and competitiveness.
                                   
