REPORT TO THE PRESIDENT - Networking and Information ... · technology research and development in...

REPORT TO THE PRESIDENT

PRESIDENT’S INFORMATION TECHNOLOGY ADVISORY COMMITTEE

Panel on Transforming Government

September 2000

Transforming Access To Government

Through Information Technology

Co-Chairs:

Raj ReddyIrving Wladawsky-Berger

Members:

Eric A. BenhamouVinton CerfChing-chih ChenDavid CooperSteven D. DorfmanDavid DormanRobert EwaldSherrilynne S. FullerHector Garcia-MolinaSusan L. GrahamJames N. GrayW. Daniel HillisRobert E. KahnKen KennedyJohn P. MillerDavid C. NagelEdward H. ShortliffeLarry SmarrJoe F. ThompsonLeslie VadaszAndrew J. ViterbiSteven J. Wallach

c/o National Coordination Office for Computing, Information, and Communications4201 Wilson Boulevard • Suite 690 • Arlington, VA 22230

(703) 292-4873

President's Information Technology Advisory Committee

August 31, 2000

The Honorable William J. ClintonPresident of the United StatesThe White HouseWashington, DC 20500

Dear Mr. President:

The President’s Information Technology Advisory Committee (PITAC) took special noteof your December 1999 executive memorandum promoting electronic government, aswell as recent announcements such as the launch of the FirstGov website for one-stopaccess to government information and services. We share your vision to create an“Information Age” government made more efficient, effective, and accessible throughinformation technology. In fact, our 1999 report, Information Technology Research:Investing in Our Future, identified the relationship between government and citizens asone of the vital areas of our national life where information technology offers thepotential to dramatically transform current practices in ways that will greatly benefit allAmericans.

Thus, we are pleased to enclose Transforming Access To Government ThroughInformation Technology, the first in a series of follow-ups to our 1999 report. This latestreport highlights our findings and recommendations on how the government can provideleadership by solving key IT technology challenges, improving public access to Federalresources as well as re-engineering and simplifying internal and external governmentaltransactions. Our goal is to define a program that will provide our citizens with full andeasy electronic access to their government regardless of their physical location, level ofcomputer literacy or physical abilities.

The report offers three key recommendations. First, we recommend that the Federalgovernment, as the world’s largest developer of, customer for, and user of informationtechnology, assume a leadership role in transforming government through the enhanceduse of information technology. The Federal government must define a coordinated andaggressive IT research program which addresses long-term technology challenges,including computer security and privacy, scalable information infrastructures, andstandards for data sharing and integration. Second, the Committee recommends that anOffice for Electronic Government be established within the Office of Management andBudget. The mission of this new office would be to promote innovative IT efforts andpolicies across the Federal government. A companion Government IT InnovationProgram (GITIP) should also be created to identify and fund high-risk, exploratory, andexperimental IT projects. Lastly, to bridge the gap between research and operationalsystems, the PITAC recommends establishing pilot projects and Emerging TechnologyCenters (ETCs) to encourage and promote information integration across the Federalgovernment and address the most immediate IT roadblocks.

Page 2August 31, 2000

We believe that adopting these recommendations will allow our nation to make significant strides towardsrealizing our shared vision. The Committee looks forward to working with you and the Congress to empowerAmerican citizens by providing them with access to their government through information technology. Thankyou for the continuing opportunity to advise you on these and other important issues.

Sincerely,

Raj Reddy, Ph.D. Irving Wladawsky-Berger, Ph.D.PITAC, Co-Chair PITAC, Co-Chair

Attachment

v

PRESIDENT’S IN F O R M A T I O N T E C H N O L O G Y A DVI S O R Y

C O M M I T T E E

President’s Information TechnologyAdvisory Committee

Co-Chairs

Members

Raj Reddy, Ph.D.Herbert A. Simon UniversityProfessor of ComputerScience and RoboticsCarnegie Mellon University

Irving Wladawsky-Berger,Ph.D.Vice President forTechnology and Strategy,Enterprise Systems Group

Eric A. Benhamou CEO and Chairman3Com Corporation

Vinton Cerf, Ph.D.Senior Vice President forInternet Architecture andEngineering MCI WorldCom

Ching-chih Chen, Ph.D.Professor, Graduate Schoolof Library and InformationScience Simmons College

David M. Cooper, Ph.D.Associate Director ofComputationLawrence Livermore NationalLaboratory

Steven D. DorfmanRetired Vice ChairmanHughes ElectronicsCorporation

David W. DormanCEOConcertRobert EwaldPresident and CEOE-Stamp Corporation

Sherrilynne S. Fuller, Ph.D.Head, Division of BiomedicalInformatics, Department ofMedical EducationUniversity of WashingtonSchool of Medicine

Hector Garcia-Molina, Ph.D.Leonard Bosack and SandraLerner Professor,Departments of ComputerScience and ElectricalEngineeringStanford University

Susan L. Graham, Ph.D.Chancellor’s Professor ofComputer Science,

vi

PA N E L O N T R A N S F O R M I N G G O V E R N M E N T

James N. Gray, Ph.D.Senior Researcher, ScalableServers Research Group, andManager, Bay AreaResearch CenterMicrosoft Corporation

W. Daniel Hillis, Ph.D.Applied Minds, Inc.

Robert E. Kahn, Ph.D.PresidentCorporation for NationalResearch Initiatives (CNRI)

Ken Kennedy, Ph.D.Director, Center forResearch on ParallelComputation, and Ann andJohn Doerr Professor ofComputer ScienceRice University

John P. Miller, Ph.D.Director, Center forComputational Biology, andProfessor of BiologyMontana State University

David C. Nagel, Ph.D.PresidentAT&T Labs

Edward H. Shortliffe, Ph.D.Professor and Chair,Department of MedicalInformaticsCollege of Physicians and

Surgeons, ColumbiaUniversity

Larry Smarr, Ph.D.Strategic AdvisorNational ComputationalScience Alliance, andProfessor of ComputerScience and Engineering University of California-SanDiego

Joe F. Thompson, Ph.D.William L. Giles DistinguishedProfessor of AerospaceEngineering, Department ofAerospace Engineering Mississippi State University

Leslie VadaszExecutive Vice PresidentIntel Corporation, andPresident Intel Capital

Andrew J. Viterbi, Ph.D.PresidentThe Viterbi Group

Steven J. WallachVice PresidentChiaro Networks


C O M M I T T E E

Table of Contents

Members of the President’s Information Technology Advisory Committee . . . . . . . . . . . . . . . . . . . . . . .v

Table of Contents . . . . . . . . . . . . . . . . . . . . . . . .vii

Panel on Transforming Government . . . . . . . . . . .viii

About This Report . . . . . . . . . . . . . . . . . . . . . . . .ix

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . .xi

A Vision of Government Access Transformed . . . . .1

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2

Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4

Finding 1: Technological barriers to citizen access 4

Finding 2: Information technology for organizationalefficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4

Finding 3: Structural impediments for Federal CIOs 5

Recommendations . . . . . . . . . . . . . . . . . . . . . . . . .6

Recommendation 1: The Research Program . . . . .7

Recommendation 2: The Office for ElectronicGovernmentand a Government Information Technology Innovation Program . . . . . . . . . . . . . . . . . . . .10

Recommendation 3: Bridging the Gap . . . . . . . .13

Note on Federal Workforce Issues . . . . . . . . . . . . .17

vii


Panel on Transforming Government

Co-Chairs

David M. CooperRobert H. Ewald

PITAC Members

Vinton CerfKen KennedyLeslie Vadasz

Ex Officio: Raj Reddy, Irving Wladawsky-Berger

Invited Participant

Herbert SchorrExecutive Director

University of Southern CaliforniaInformation Sciences Institute

viii


C O M M I T T E E

ix

A bout This Report

“Transforming Access to Government ThroughInformation Technology” is one in a series of reports to thePresident and Congress developed by the President’sInformation Technology Advisory Committee (PITAC) onkey contemporary issues in information technology. Thesefocused reports examine specific aspects of the near- andlong-term research and development and policies we needto capture the potential of information technology to helpgrow our economy and to address many of the importantproblems facing the nation.

The 24-member PITAC, comprising corporate andacademic leaders, was established by Executive Order ofthe President in 1997 and renewed for a two-year term in1999. Its charge is to provide the Federal government withexpert independent guidance on maintaining America’spreeminence in high performance computing andcommunications, information technology, and NextGeneration Internet R&D.

In February 1999, the PITAC issued an overview andanalysis of the current state of Federal informationtechnology research and development in a report entitled“Information Technology Research: Investing in Our Future.”That report set forth a vision of how informationtechnology can transform the way we live, learn, work, andplay, with resulting benefits for all Americans. But thereport warned that Federal information technology researchand development is seriously inadequate, given itseconomic, strategic, and societal importance. TheCommittee concluded that the government is funding onlya fraction of the research needed to maintain U.S.preeminence in information technology and propel thepositive transformations it enables.

The Committee identified 10 information technology“National Challenge Transformations” — including therelationship between government and citizens — that are

x


critical to America’s future. To meet these transformationchallenges, the PITAC recommended a strategic Federalinitiative in long-term information technology R&D andoutlined the research priorities that will drive the necessaryadvances in the new century.

To examine some of the transforming applications ofinformation technology in greater detail, the PITACsubsequently convened a group of panels led by Committeemembers and including invited outside participants withrelevant expertise. Three panels focused on informationtechnology national challenges: Transforming Government,Transforming Health Care, and Transforming Learning. TheCommittee also established several panels to report oncritical technology issues that span the transformations,including Digital Divide Issues, Digital Libraries, InternationalIssues, and Open Source Software for High End Computing.Over the past year, each of the panels has analyzedrelevant research data and documents; held workshopdiscussions and conducted interviews with experts in theirfields; and studied the fiscal, organizational, and economicimplications of strategies to generate necessaryinformation technology research and developmentadvances in these key areas of our national life. TheCommittee plans to convene additional panels in themonths ahead.

“Transforming Access to Government ThroughInformation Technology” and the succeeding reports in thisseries present targeted findings and recommendations tothe President and Congress designed to help the nationrealize the vision of these positive transformations. Theirbenefits for our future can be extraordinary, but they arenot guaranteed. To make the vision a reality, we need theresults of aggressive, well-funded, and well-managedFederal research programs.


C O M M I T T E E

xi

A cknowledgements

The Panel on Transforming Government appreciates theadvice and assistance of William Scherlis, Carnegie-MellonUniversity, who provided thoughtful and constructivereviews of drafts of this report. The Panel thanks thenumerous Federal employees who provided information,including Larry Brandt, Program Director for DigitalGovernment at the National Science Foundation, the ChiefInformation Officers Council and staff, and the managerswho briefed us. We also thank Tom Kalil, National EconomicCouncil, Lori Perine, Office of Science and TechnologyPolicy, and Jasmeet Seehra, Office of Management andBudget, for their assistance in uncovering the issues andexploring potential solutions.

The Panel would also like to acknowledge the work of theNational Coordination Office for Computing, Information,and Communications in supporting its efforts to producethis report. The Panel thanks Robert Winner, who kept thePanel on track and organized our ideas and analyses intothe final prose. We thank Yolanda Comedy, Sally Howe,Laurie Mitchell, and Kay Howell for supporting the Panel’sdeliberations, for their review of earlier drafts of thisreport, and for their helpful comments. We thank MarthaMatzke, who edited and formatted the final document. Andfinally, we are grateful to the entire staff at the NationalCoordination Office. Our meetings went smoothly becauseof their careful preparation.


C O M M I T T E E

1

A Vision of Government Access Transformed

Government services and information are easilyaccessible to citizens, regardless of their physicallocation, level of computer literacy, or physicalcapacity. Intelligent systems guide citizens byproviding a one-stop shopping experience for locatingrequested information. Documents and forms can beaccessed, completed, and submitted electronically.Automated business processes allow nearlyinstantaneous response to citizens’ requests. In timesof natural emergencies, emergency crews have instantaccess to three-dimensional building models, riskanalysis and assessment, high-resolution local weatherpredictions, stress analyses of damaged structures,rapid evacuation planning tools, and emergencyagency coordination.

PITAC Report,February 1999

IntroductionIn “Information Technology Research: Investing in Our

Future,” its 1999 report to the President, the President’sInformation Technology Advisory Committee articulated avision of the ways information technology will driveprogress in the 21st century. The Committee identified 10vital areas of our national life — including the relationshipbetween government and citizens — in which information


As a follow-up to its report, the PITAC established agroup of panels to examine the transformation challengesin greater depth and make recommendations for addressingthem to the President and Congress. This document detailsthe findings and recommendations of the Panel onTransforming Government, co-chaired by PITAC membersDavid M. Cooper and Robert H. Ewald.

O verview

The perspective of the Transforming Government Panel isshaped by a fundamental reality: The Federal governmentis the world’s largest and most complex developer of,customer for, and user of information technology. Evenindividual departments and agencies rank among theworld’s largest information technology users. In thatcontext, the PITAC charged the Panel with identifying thekey technical challenges and developing a long-rangetechnology-based strategy to harness the power ofadvanced information systems to make government’sstores of information and vital services easily accessible toand usable by all U.S. citizens “regardless of their physicallocation, level of computer literacy, or physical capacity.”The nation clearly should seek this goal, but currenttechnology is inadequate and difficult problems must be

Realizing the Vision: Technical Challenges andBenefits

There is a huge potential to make all governmentinstitutions both more efficient and more responsivethrough information technologies. Technical challengesinclude significant improvements in systems and methodsfor accessing data, including high performance datastorage and tools to locate and present information.Robust, reliable, and secure networks and software todeliver and protect critical information are important


C O M M I T T E E

To conduct its study of the challenges, the TransformingGovernment Panel met with several key Federal CIOs andother representatives, attended symposia, and reviewedstrategic planning documents and reports of previousstudies. The Panel considered only civilian applications inthe Federal administration; it did not consider Departmentof Defense applications or those of other national securityentities, the Congress, the courts, or state and localgovernments. In many cases, the needed technologies willbe the same across all these application levels, but in thenational security areas there are unusual requirements. ThePanel determined early in its deliberations that thesubstantial and distinct national security and DoD issueswarrant separate PITAC studies. The Panel urges PITAC toinitiate a panel focusing generally on national securityapplications of information technology and specifically onthose of the DoD.

The Transforming Government Panel agrees with thePITAC report’s finding that great opportunities exist toimprove public access to Federal information resources andto simplify transactions between the government andcitizens and within the government itself. The research andpilot projects the Panel proposes below have the dualbenefits of addressing government issues and attractingadditional, sorely needed workers to exciting projects. Thefollowing findings and recommendations are intended toassist the President and Congress to seize theopportunities and achieve the benefits.

Findings

Finding 1. Major technological barriers prevent citizensfrom easily accessing government information resourcesthat are vital to their well being. Today government

3


4

information is often unavailable, inadequate, out of date,and needlessly complicated.

The PITAC considered two forms of public access:convenient, easy-to-use access to well-managedinformation and convenient, easy-to-execute transactions.The government stores large amounts of importantinformation. However, finding that information in thegovernment’s many databases is difficult, and correlatingthe meaning of findings from a number of inconsistentlydefined databases requires deep knowledge of theexistence, contents, and management schemes of thosedatabases. Similarly, well-defined transactions often requireseveral sub-transactions with disparate agencies. Theseseparate transactions are often unlinked, requiring expertknowledge of a variety of agency locations, procedures,rules, and requirements.

Some progress can and should be made using technologyin its current state, but many end goals require substantialtechnological advances. For example, large-scale dataintegration across multiple independent databases is asignificant research area. The PITAC concludes thatprogress can best be made through a continuum of near-,mid-, and long-term efforts — all carried out concurrently— under the guidance and, in some cases, control ofstrong coordinating mechanisms.

Finding 2. Information technology can be used toincrease organizational efficiency and effectiveness andsave costs.

While the PITAC focused on issues surrounding theinteractions between government and the public, theCommittee remained mindful of the use of informationtechnology in internal government processes. Informationtechnology can increase the efficiency and effectiveness oforganizations of all sizes. Industry has amply demonstrated


C O M M I T T E E

over the last 20 years that information technology can beused — in combination with careful process redesign — asboth a forcing function and an implementation tool toincrease process and product efficiency and improveeffectiveness.

With some exceptions, the government has been slowerthan industry to take advantage of informationtechnology’s promise for several reasons. Among these arethe scale of the Federal enterprise, the lack of incentives(or the existence of powerful disincentives) to cut costs,the curtailment of capital investments, the cumbersomeFederal acquisition process, the risk-averse nature ofgovernment management, and conflicting and complexlegal requirements for both openness and privacy. Inaddition, the government faces a decreasing supply of

competent, up-to-date information technology workers toconceive, implement, field, and manage new systems.

Finding 3. The Federal CIO Council understandsinformation technology’s utility for improving governmentservices and processes, and the Council’s strategyestablishes appropriate and ambitious goals. But the CIOs’mandates require them to focus primarily on near-termoperational issues and acquisitions. Budget planningprocesses make it difficult to carry out effective cross-agency coordination and execution and the long-termresearch efforts that many of the goals require.

The Federal CIO Council and many agencies have setappropriate, ambitious goals for improvement ofgovernment functions through the use of informationtechnologies. Achieving these goals will enable cross-agency transactions and integration of information acrossagencies and departments — in essence, “one-stopshopping” for our nation’s citizens. Industry has shown that

5


6

large-scale solutions that cross many departments have tobe able to evolve and have to be based on uniform userinterfaces that result from strong cooperation amongdepartments. Even to approach the desired result willrequire substantial cross-agency planning and execution.Unfortunately, the current budget planning process doesnot encourage such cross-agency activities.

While the CIO Council has established mechanisms forsharing results and lessons, the process of creatingstandardized processes and information representations,eventually leading to cross-agency transactions andinformation federation and integration, is much harder andrequires cross-agency budget planning and execution.Creating cross-agency budgets requires substantial workand, therefore, is used only for large initiatives. Dependingon cross-agency plans is very risky because of theuncertainty that all participants will receive adequatefunding. Therefore, cross-agency projects and initiativescurrently have to be large enough to warrant the effort butpartitionable enough that no one really must depend onanyone else’s appropriations or performance. In addition,stovepiping of both congressional and executive reviewprocesses causes stovepiping of plans and programs. TheGovernment Performance Results Act (GPRA), for example,while valuable in requiring agencies to set goals againstwhich they can be held accountable, tends to hinderagency interdependencies in plans and programs becauseno agency will create a GPRA objective that depends onbudgeting and operational success in another agency.

The PITAC applauds recent e-government initiativesproposed by the Administration and Congress, such as thefirstgov.gov portal and those being considered in the Houseand Senate. However, many envisioned systems andservices depend on continued progress in informationtechnology research in the areas listed below. Currentprogram management models do not encourage R&D


C O M M I T T E E

7

program managers to take sufficient risks in pursuit oflong-term goals. At the same time, someone must beresponsible for technology transfer from successfulresearch efforts into operational systems, a process inwhich the government has traditionally been weak.

Recommendations

The PITAC recommends the following three actions:

.Establish a coordinated research program thataddresses Federal e-government requirements;

.Create an OMB Office for Electronic Government and aGovernment Information Technology InnovationProgram to promote, lead, and coordinate Federalefforts to improve citizens’ access to governmentinformation and improve government processes; and

.Establish pilot projects focused on connecting near-,mid-, and long-term Federal efforts.

Recommendation 1. The Research Program

Establish and coordinate an information technologyresearch program that addresses the Federal government’smost critical requirements for long-term technologydevelopment.

Having reviewed a range of Federal applications providedby the CIO Council and the NSF Digital GovernmentProgram, the PITAC has concluded that there are critical,long-term technical issues that need to be addressed tomake government services and information easilyaccessible to citizens. These are:

.Security and privacy — Federal systems often containinformation that should be available to authorized


parties but not to anyone else. As evidenced byrecent security breaches, the technologies toaccomplish both availability and protection are in theirinfancy. Approaches that work on a small scale rarelywork when applied to large systems, and envisionedFederal systems will be among the largest in existence.Critical information technology infrastructureprotection requires a focus on this area at a level thatmay dwarf current investments.

.Data integration — There are several technologicalissues here. First is how to present users a coherentview of information stored in radically varied ways onsystems that were created and have been optimizedfor a variety of purposes and base technologies.Second is how to make this coherent view both easyto use for non-technicians and adaptable to thevarious purposes that users might have. Third is howto do all of this efficiently. Each of these capabilitiesrequires research in information management, humanfactors, storage systems, middleware, and relatedtechnologies. Practical systems will have to be basedon strategies that combine integration, federation, andevolvability, and the systems’ development andoperation will require cooperation among departmentsand agencies at a level that is unusual for the Federalgovernment.

.Software development and quality — Softwaredevelopment is too difficult, risky, and costly, and itrequires highly trained professionals who are in shortsupply. Yet many Federal missions depend on softwaredevelopment and quality and, therefore, stand to gainenormous advantages from improvements in softwareand the processes required to create, test, deploy,and use it. To be most useful for government,improved software should be:

- developable or purchasable with predictable


C O M M I T T E E

behavior and performance

- maintainable at reasonable cost

- integrable, evolvable, extensible, and adaptablewith reasonable effort

- fail-safe

This is a long-term, difficult problem. However, thesoftware research areas described in the PITAC’s February1999 report are the ones the PITAC believes need to becentral to the research effort for Federal purposes. Severalof these, such as component technologies, requireapplication-specific demonstration efforts, and it would beappropriate and useful for the government to pursue someof those.

.Scalable information infrastructure (SII) — High-volume interactions with citizens and within thegovernment, especially those taking advantage of awide range of media such as full-motion video, requireadvances in the capability and reliability of ourinformation infrastructure. Just as the commercialworld has found thousands of useful applications forthe Internet, so are there many useful applications ingovernment. The government has been at the centerof ongoing Internet technology developments, and it isideally situated to try experimental and productionapplications now, as new Internet technologiesemerge. One goal should be to make the Internet soreliable that it can be depended upon for mission-critical applications.

The SII manages information as well ascommunications. Unlike physical libraries, digitallibraries manage information so as to provide apotentially infinite number of users with access toeach item in the library simultaneously. Many large-scale government applications require access to thevast stores of government information best managed


in digital libraries. Thus, the government needs toaggressively pursue the information managementtechnologies needed for large-scale digital libraries.The PITAC Panel on Digital Libraries is actively pursuingthis area and the Panel on Transforming Governmenthas coordinated with that Panel on its forthcomingreport.

The Next Generation Internet (NGI) is a fundamentallyimportant, ongoing SII project required for many of thetransformational efforts envisioned for thegovernment. This important Federal initiative deservesfull funding in all participating agencies, and manyother agencies should build applications upon it.

.Development and availability of high end systems —There are important governmental requirements foradvances in high end systems. The government mustcontinue to pursue research and experimentaldevelopment of these systems, software, andhardware components, including high performance,large-scale storage. Example applications includeintelligence in law enforcement, weather modeling andforecasting, large-scale planning, large-scaleinformation fusion in emergencies, geographicinformation systems, ecological modeling and science,and integrated product and process development.Fundamental government purposes are served throughthe widespread support of civilian science andengineering research, and advanced high end systemsare required for this. While there are commercialapplications for advanced high end systems, theFederal government remains the primary customer in ashrinking market niche.

.Socioeconomic implications of government uses ofinformation technology — The government shouldlaunch studies to measure and understand the effectsof new technologies on the relationship between


C O M M I T T E E

11

government and citizens, including businesses. Forexample, studies of experiments in on-line voting needto be carried out to illuminate both technical andsocio-political issues. Another important issue withpolitical, sociological, and technical aspects is securityand privacy. For example, there is an issue of whatgovernment databases should or should not befederated and how to “dither” information so thatinferences about individuals cannot be drawn, whilemaintaining the validity of statistical inferences.

Recommendation 2. The Office for ElectronicGovernment and a Government InformationTechnology Innovation Program

The Administration should establish an Office for ElectronicGovernment (OEG) within OMB with responsibility forpromoting innovative information technology efforts andpolicies that improve both citizen access to informationand government efficiency and effectiveness. The OEGwould (a) create incentives for, promote, facilitate, andprovide matching funds for innovative projects and bestpractices with cross-agency implications and (b) advise theDirector of OMB on funding of all e-government effortswithin agencies. The Administration should establish an OEGGovernment Information Technology Innovation Program tofund high-impact, innovative, or exploratory cross-agencyprojects executed and managed in various agencies.

The PITAC recommends creation of an Office forElectronic Government (OEG) within OMB led by a senior-level Presidential appointee. The OEG is needed tochampion the effective use of information technology and

12


the Internet as the best means to rapidly improve theaccess of citizens and businesses to government data, thequality of their interactions with the Federal government,and the government’s own business operations. The OEG’smission will be to promote innovative and cross-agencyefforts that serve long-term needs or that are tooexperimental or risky to be undertaken within the usualFederal information technology acquisition process. Thisincludes but is not limited to the Government InformationTechnology Innovation Program (GITIP). The OEG shouldalso advise the Director of OMB concerning the funding ofall e-government projects.

The OEG would be empowered to enable interagencyinformation technology research and development projects,managed within the agencies, aimed at improvinggovernment functions. The Office should be empowered tocreate incentives for strong cooperation where required todevelop standards and cross-agency systems, and GITIPshould enable the OEG to organize and add funds to intra-and interagency project and research budgets. The OEGshould promote coordination and cooperation amongagencies so as to establish process, interface, andinformation-representation standards as needed to reducecosts, increase effectiveness, and establish efficienciesacross the Federal government. Because it helps establishinteragency objectives at the OEG level, this approachaddresses the finding that government objectives andreviews tend to be stovepiped, for example, within theGPRA process. Rather than having the decision-makingauthority and control of an “information technology czar,”the OEG should have a service mandate that enablesagency and cross-agency innovation through incentives andbudget plus-ups.

The President and Congress should recognize and supportthe efforts of the CIO Council in implementing Federalinformation technology policies. The CIO Council’s efforts


C O M M I T T E E

13

so far are laudable, but the mechanism is not yet sufficientfor the task at hand. Incentives and mechanisms foraggressive innovation are particularly insufficient. Thecurrent acquisition system drives CIOs to satisfy near-termoperational requirements with minimum risk. Establishingthe OEG and GITIP will provide CIOs with increasedopportunity and means to undertake efforts that may notbe appropriate for the normal acquisition process becauseof their innovative, experimental, long-term, or exploratorycharacter, or because they involve multiple agencies. Acoordinated management mechanism, led by the OEG andimplemented by the Federal CIO Council with collaborationof the NSTC, would address this opportunity. Thus, theCIOs will have increased responsibility to innovate and toexpand their planning horizons based on visibility into andinfluence over mid- and long-term efforts.

Congress should empower and fund the OEG with R&Dfunds at a level annually justified within the authorizationand appropriations process to enable multiyear, cross-agency, and intra-agency projects. Based on our industrialexperience, cost projections of several governmentprojects, and experience with government-funded researchefforts, the PITAC recommends a starting point of about$100 million per year. This would allow 10-20 projects withtotal annual budgets of $1-20 million, including matchingfunds proportioned in accordance with the level and natureof risks taken.

The Panel reviewed several current commercial Internet-based development efforts and discussed the issue ofbudget size with venture capitalists. Venture capitalistsestimated that starting a new Web-based business costsabout $50-100 million, provided that it does not require asubstantial amount of new technology. The estimated$100 million-per-year funding line we recommend for GITIPwould provide for prototypes, experiments, and riskmanagement, not for acquisition of fieldable production


systems.

The governmentwide Y2K effort showed that centrallycoordinated budget plus-ups and management can beeffective when Congress is able to authorize funding basedon knowledge of project plans and selection criteria. Werecommend similar cooperation between the Administrationand Congress to review and agree upon project plans,criteria, and funding levels sufficient for success.

Individual efforts within GITIP would be created ascollaborations among client agencies (e.g., Agriculture,Census, EPA) and, where appropriate, technologycontributors (e.g., DARPA, NSF, NIH, NASA, NOAA, andothers). Program managers for the efforts would be drawnfrom a cadre of professional program managers selectedeither from CIO-led operational acquisition organizationsthat have suitable experience or from research agencies.Program managers would retain a primary affiliation withtheir home agencies.

The OEG, working in concert with the Federal CIO Counciland the National Science and Technology Council, should beresponsible for ensuring that results of long-term R&Defforts are appropriately exploited in advanced agencyprojects. The OEG should take advantage of the fact thatsome agencies — NASA and NOAA, for example — viewthemselves as particularly able to bridge from basicresearch to operational use.

In concert with and as an integral part of theestablishment of the technology efforts recommendedabove and pilot projects below, the OEG shouldimmediately establish an effort to measure improvementsand disseminate best practices and lessons learned.

The person appointed to lead the OEG should bring to thejob a combination of technical and operational vision, anunderstanding of how to get things done in thegovernment, and a strong understanding of information

15


C O M M I T T E E

technology research, development, and operational issuesand approaches. A small fraction of the $100 millionbudget would be required to operate a coordination office.As a point of clarification, the PITAC does not view theleader of the OEG as the senior person with oversightresponsibility for Federal information technology researchrecommended in the February 1999 PITAC report.

Recommendation 3. Bridging the Gap

Establish pilot projects and Enabling Technology Centers(ETCs) to extend technology and transfer it intooperational systems.

Pilot projects and ETCs should be budgeted and executedto increase the use and utility of information technologyfor Federal missions. Many useful mid- and long-termefforts are technically and operationally challenging. Anyeffort, for example, that requires the creation of a datamodel spanning multiple, currently independent governmentdata sources will be technically difficult and may raiseissues about polices and statutes that govern informationsharing among agencies. We encourage multiple pilotexperiments with information integration across manygovernment sectors.

Pilot projects are useful for many reasons. They enableteams of application experts and computer scientists towork collaboratively on a relevant problem. They formbridges among near-, mid-, and long-term e-governmentplans and programs. They can incorporate softwarevendors, systems integrators, and others at the beginning,so that implementing teams will be positioned andexpected to push the R&D products to operationalproduction mode. Pilot projects provide non-R&D agenciesaccess to the best information technology researchers andtheir facilities. They reinforce researchers’ understanding ofreal world needs while providing practitioners with a greater


vision of future possibilities. They assist in recruitment byexposing computer science students to very challengingproblems that might attract them after graduation. Perhapsmost importantly, pilot projects bootstrap theimprovement process and prove to practitioners thatrevolutionary changes can actually work.

Three areas are recommended for highest priority:

.Crisis Management — A government InformationTechnology for Crises Management Team has alreadybeen established that has planned an ETC on thistopic. The PITAC applauds this effort and encouragesthe Administration and Congress to establish aninteragency budget for it.

.Access for Disabled Citizens — There is a substantialrequirement for new technology here if thegovernment is to satisfy currently articulated accesspolicies.

.Extending firstgov.gov — This project currentlyenvisions a Yahoo-like portal to government Web sites.It seeks to organize the portal along several lines,including popular sites and an exhaustive taxonomy. Itdoes not provide for information integration orfederation, nor does it require standardization amongagencies. When taken only at this level, firstgov.gov isa near-term effort built with currently availabletechnologies. But we encourage the OEG,recommended above, to promote and support useful,extended visions requiring more advanced technology.The OEG project should develop advanced capabilitiesto provide Internet-based access to governmentservices, including transactions with governmentagencies, metadata representation in support ofintegration of information, and other advancedservices. Firstgov.gov should focus efforts ongovernment-specific capabilities such as transactionsupport, metadata creation, and comprehensive


C O M M I T T E E

17

searchable catalogs of information and services.

Other areas recommended for implementation:

.Information Technology for Information TechnologyTraining — This is potentially one part of the solutionto the Federal information technology workforceshortage. The efficacy of computer-based training hasbeen well documented. New techniques need to bedeveloped to take advantage of the Internet fordelivery, the Web as an organizing mechanism, and theincreasing availability of inexpensive computing powerto make systems adaptable to learners.

. Privacy, Confidentiality, Security, and Authenticity forInternet-based Government Interactions with the Public— Data integration of the types mentioned throughoutthis report require increased attention to privacy,confidentiality, security, and authenticity. As thedegree of integration increases, so does the likelihoodof inadvertent disclosure of information obtained undera pledge of confidentiality. With heightened concernover electronic medical records, tax records, and otherpersonal information documents, and the accumulationof large transaction databases by business enterprises,official data-collection agencies that depend onvoluntary responses are even more concerned aboutensuring the privacy and security of their respondentsand the information they provide. Agencies needmethods and resources to ensure that data are securefrom outside tampering. (Would we know if someonewent into a database of millions of numbers andchanged just one?) Tackling these security issuesrequires near-, mid-, and long-term activities that canaddress both technical and political facets of theproblem.

.Integrating Statistical Information — All levels of


18

government produce and use statistical information,but integrating statistical information in ways thatpolicymakers and researchers find useful is a verydifficult problem. Issues ranging from deciding whenand where to build a school, road, or business toinvestigating causes and potential remedies of air andwater pollution often can be focused on relativelysmall geographical areas. Issues related to publichealth, occupational safety and health, Social Security,Medicare, etc., may be national in focus, but variationsin policy implications for demographic or other sub-populations could differentially affect various state orlocal government policies. While there exist centerslike the California Environmental Resources EvaluationSystem and the National Center for GeographicInformation and Analysis, the first type of center isfocused on providing data for one state and themeand the second on research methodologies foranalyzing geographically based data. Research isneeded on how to create the first type of site moreeasily across a variety of geographies and themes and

providing tools for analysis that a school principal,farmer, health care provider, business entrepreneur, orstate office could use.

The pilot projects should be based on a strategy thatpursues near-, mid-, and long-term issues, yielding acontinual release of demonstration prototypes, an effectiveinfusion of emerging technologies, and a continuous pursuitof solutions to long-term research issues.

To execute such a strategy, the agencies will need toestablish partnerships among government, industry, anduniversities. Partnerships offer the potential benefit ofcreating a more efficient flow of new ideas into agenciesthat otherwise have not pursued long-term informationtechnology research in support of their missions. Oneapproach would be to cross-matrix application-based pilots

Ordering Copies of PITAC Reports

This report is published by the National CoordinationOffice for Computing, Information, and Communications.For additional copies or copies of other PITAC reports,please contact:

National Coordination Office for Computing,Information, and Communications4201 Wilson Blvd., Suite 690

Arlington, VA 22230(703) 292-4873

Fax: (703) 292-9097E-mail: [email protected]

PITAC documents are also available on the NCO Website:

Cover design by James J. Caras, National Science

Date post:	26-Jun-2020
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times