h************************************************** · 2014. 4. 9. · DOCUMENT RESUME. IR 015 075....

ED 336 068

TITLE

INSTITUTION

REPORT NOPUB DATENOTEAVAILABLE FROM

PUB TYPE

EDRS PRICEDESCRIPTORS

IDENTIFIERS

ABSTRACT

DOCUMENT RESUME

IR 015 075

Helping Amreica Compete: The Role of FederalScientific & Technical Information.Congress of the U.S., Washington, D.C. Office ofTechnology Assessment.OTA-CIT-454Jul 9078p.

Superintendent of Documents, U.S. Government PrintingOffice, Washington, DC 20402-9325 (GPO stock no.052-003-01196-4, $3.75),,Information Analyses (070) -- Reports -Research/Technical (143)

MF01/PC04 Plus Postage.Databases; *Federal Government; *G]obal Approach;*Government Role; *Information Dissemination;*Information Management; Information Technology;International Educational Exchange; Public Policy;Research and Development; Science Education;*Scientific and Technical Information*Information Policy

Prepared in response to a request from the HouseCommittee on Science, Space, and Technology, this report assesses howscientific and technical information (STI) developed by or for theFederal Government can contribute to a more competitive United Statesand what actions are needed to realize this potential. It is arguedthat STI is important because of its potential significant impact onglobal issues, as well as its critical role in the innovation processin education, basic research, applied research and development,product development and manufacturing, and the application of scienceand technology to meet the needs of the commercial, not-for-profit,and governmental markets. Topics of discussion include: (1) FederalScientific and Technical Information and the U.S. Competitive Edge,which includes science and technology policy, science education, andinternational competitiveness; (2) Reaching Consensus on Principlesof Federal Scientific and Technical Information Dissemination; (3)Reaching Consensus on the Open Flow and Government-wide Disseminationof federal STI, which focuses on the role of government widedissemination and archival agencies in STI; and (4) Framework for aPresidential Initiative on Scientific and Technical Information,which highlights the role of the Executive Office rl the President,including the Office of Management and Budget (OMB) and the Office nfScience and Technology Policy (OSTP), as well as other FederalGovernment agencies. A review of technological opportunities formanaging and disseminating federal STI and a list of acronyms areappended, and an index by major topics is provided. (DB)

********************h*************************************************** Reproductions supplied by EDRS are the best that can be made *

* from the original domment. *

***********************************************************************

r;..0

cNs

riELPI GCA

The RoleFe4ral

g'Scien

U.S. DEPARTMEra OF EDUCATION011tce 04 Educational Research and Improvement

EDUCATIONAL RESOURCES INFORMATIONCE'1TER (ERIC)

(3 This document has been reproduced asreceived from the person or organizationoriginating it

Cl Minor changes have been made to improverepioduchon quality

Points of view or opinions stated in this docu-ment do not necessarily represent officialOERI position or policy

tItk

fW4Ritrifia)441111t4

UKTED STATES CONGRESSOFFIcr OP-TEC OGY ASSESSMENT

,421

RFRT env AVAII ARI

Office of Technology Assessment

Congressional Board of the 101st Congress

EDWARD M. KENNEDY. Massachusetts. Chaurnan

CLARENCE E. MILLER, Ohio, Vice Chairman

Senate

ERNEST F. HOLLINGSSouth Carolina

CLAIBORNE PELLRhode Island

STEVENS

Alaska

ORRIN G. HATCHUtah

CHARLES E. GRASSLEYIowa

CHASE N. PETERSON. ChairmanUniversity of Utah

JOSHUA LEDERBERG, Vice ChairmanRockefeller University

CHARLES A. BOWSHERGeneral Accounting Office

LEWIS BRANSCOM,3

Harvard University

House

MORRIS K UDALLArizona

GEORGE E. BROWN. JRCalifornia

JOHN H. GIBBONS(Nonvoting)

Advisory Council

MICHEL T. HALBOUTYMichel T. Halbouty Energy Co.

NEIL E. HARLIowa State University

JAMES C. HUNTUniversity of Tennessee

HENRY KOHLERUniversity of Arizona

Director

JOHN H. GIBBONS

JOHN D. DINGEL L

Michigan

DON SUNDQUISTTennessee

AMO HOUGHTONNew York

SALLY RIDECalifornia Space Institute

JOSEPH E. ROSSCongressional Research Service

JOHN F M. SIMSUsibelli Coal Mine. Inc

MARINA v N WHI1MANGeneral Motors Corp

The Technology Aisessment Board approves the release of this report. The views expressed in tnis report are not necessariII

those of the Board. OTA Advisory Council. or individual members thereof

nELPINGMERICA

COMPETEThe Role ofFederalScientific & TechnicalInformation

UNITED STATES CONGRESSOFFICE OF TECHNOLOGY ASSESSMENT

4

Recommended Citation:

U.S. Congress, Office of Technology Assessment, Helping America Compete: The Role ofFederal Scientific and Technical Information, OTA-CIT-454 (Washingwn, DC: U.S.Government Printing Office, July 1990).

For sale by the Superintendent of DocumentsU.S. Government Printing Office, Washington, DC 20402-9325

(order form can be found in the back of this report)

f7;

ii

Foreword

This Special Report assesses how Federal scientific and technical information (STD cancontribute to a more competitive America and what actions are needed to realize this potential.The report was prepared in response to a request from the House Committee on Science,Space, and Technology.

Global change is a fact of contemporary lifewhether in the political, economic, ortechnological spheres. U.S. leadership in all of these areas is being challenged. We need to takeactions that can help renew the U.S. competitive edge in the worldwide marketplace of ideas,products, and services, and to provide leadership on global issues such as the environment.

A key area of U.S. strength covld and should be our scientific and technical information.The U.S. Government is the largest single source of in the worldranging from technicalreports on aerospace propulsion and solar thermal electric systems to satellite data on oceanicand atmospheric trends to bibliographic indices on medical and agricultural research.

Yet the United States is not taking full advantage of opportunities to use Federal STI aspart of a strategy to renew the U.S competitive edge. STI is very important to scientists andengineers in a wide range of research, development, and commercial activities. They spenda lot of time on STItime that is valued, conservatively, at several bit'ons of dollars per yearjust for federally funded researchers. When used efficiently, Federal STI pays off handsomely.

The Special Report has benefited from dibcussion et an August 1989 OTA workshop,several rounds of comments on earlier drafts, and debate on related topics at recent executivebranch meetings and congressional hearings. OTA appreciates the participation of the Officeof Science and Technology Policy, Office of Management and Budget, and Federal agencyofficials and members of the scientific, academic, library, business, and consumercommunities, among others, who provided useful comments and information. The report is,however, solely the responsibility of OTA and not of those who assisted us.

JOHN H. GIBBONSDirector

f;

ill

Reviewers and Contributors

August 8, 1989, OTA Workshop Participants

Gerald Barton, NOAA National Oceanographic Data CenterPatricia Berger, National Institute of Standards and TechnologyJane Bortnick, Congressional Research ServiceJoseph Caponio, National Technical Information ServiceJoseph G. Coyne, U.S. Department of EnergyRobert Dugan, National Commission on Librtuies and

Information ScienceLarnont Eltinge, Office of Science and Technology PolicyDon Fossedal, U.S. Government Printing OfficeLarry Herrmann, U.S. General Accounting Office

August 9, 1989, OTA Workshop Participants

Jerry McFaul, U.S. Geological SurveyKurt Molholm, Defense Technical Infonnation CenterFrank Reeder, Office of Management and BudgetHarold Relyea, Congressional Research ServicePhil Thibideau, National Aeronautics and Space AdministrationMary Ann Wallace, National Archives and Records AdministrationForrest Williams, Bureau of the CensusHelen Wood, National Oceanic and Atmospheric AdministrationRon York, Office of Science and Technology Policy

Prue Adler, Association of Research LibrariesJerry Berman, The Benton FoundationWilliam Carpenter, Martin Marietta Energy Systems, Inc.Bonnie Carroll, CFINIDIJoseph Coyne, U.S. Department of EnergyAI Domff, Institute for Scientific Information, Inc.Don Fossedal, U.S. Government Printing OfficeSteve Frantzich, U.S. Naval AcademyWarren Glimpse, U.S. Statistics, Inc.Charles McClure, Syracuse University School of Information

StudiesOther Review

Sandy Morton, Special Libraries AssociationRobert Park, American Physical SocietyDavid Plocher, OMB WatchMarc Rotenberg, Computer Professionals for Social

ResponsibilityDrhorah Runkle, American Association for the akderzzement of

ScienceMark Scully, U.S. Government Printing OfficeHarold B. Shill, West Virginia University LibraryDaniel Wilde, NERAC, Inc.

ers or Contributors

Andrew A. Aines, ConsultantKenneth B. Allen, Information Industry AssociationJack Arthur, Office of Management and BudgetTom Bold, National Technical Information ServieeCynthia Carter, National Research Council INametical Data

Advisory BoardMike Chinnery, NOAA National Geophysical Data CenterDaryl Chubin, Office of Technology AssessmentJoe Clark, National Technical Information ServiceJohn Clement, National Research Council Math Sciences

Education BoardJoseph W. Duncan, The Dun & Bradstreet Corp.Fred Durr, National Information Systems CorporationD. Kaye Gapen, University of Wisconsin at Madison LibrariesBob Gellman, House Committee on Government OperationsSteve Gould, Congressional Research ServiceAnn Heanue, American Library AssociationLisa Heinz, Office of Technology AssessmentPeter Hemon, Simmons CollegeChristopher Hill, Congressional Research ServiceJoseph E. Jenifer, U.S. Government Printing OfficeBrian Kahin, Harvard UniversityJoseph King, NASA National Space Science Data CenterThomas M. Kurihara, U.S. Department of DefenseBarbara Lawrence, American Institute of Aeronautics and

AstronauticsRalph E. Lewis, Lockheed Missiles & Space Co., Inc.Katherine F. Mawdsley, University of California a: Davis

LibraryGaiy Metz, USGS EROS Data CcnterGeorge Minot, Applied Information Technologies Research

Center

Mike Nelson, Senate Committee on Commerce, Science, andTransportation

Eni Njoku, NASA Earth Sciences and Applications DivisionJames Paul, House Committee on Science, Space, and TedmologyHenry H. Perrin, Jr., Villanova Law SchoolThomas E. Pinelli, NASA Langley Research CenterDouglas R. Posson, U.S. Geological SurveyJoe Price, Library of CongressRussell S. Rice, NASA Scientific and Technical Information

DivisionC.D. Robbins, Martin Marietta Energy Systems, Inc.Kenyon C. Rosenberg, NrIonal Technical Information ServiceAl Saley, Society of School LibrariansAndrew Sherman, U.S. Government Printing OfficeElliot R. Siegel, National Library of MedicineJean Smith, Office of Technology AssessmentKent Smith, National Library of MedicineTim Sprehe, Office of Management and BudgetNick Sundt, Office of Technology AssessmentAl Teich, American Association for the Advancement of

ScienceKenneth Thibodeau, NARA Center for Electronic RecordsGene Thorley, USGS National Mapping DivisionMitchel B. Wallerstein, National Research Council International

AffairsLisa Warnecke, Syracuse University School of Information

StudiesChristine Wegman, House Committe.: on Science, Space, and

TechnologyAlan R. Wenberg, National Technical Information ServiceRoxanne Williams, U.S. Department of AgricultureJoan Winston, Office of Technology Assessment

NOTE: Om appreciates the valuable assistance of outside reviewers and contributors. These individuals do not, however, necessarily

approve, disapprove, or endorse this Special Report. The report is the sole responsibility of OTA, not of those who so abiy

assisted us.iv

OTA Project StaffHelping America Compete

John Ande lin, Assistant Director, OTAScience, Information, and Natural Resources Division

James W. Cur lin, ManagerCommunication and Information Technologies Program

Fred B. Wood, Project Director

Administrative Staff

Liz Emanuel, Office Administrator

Karolyn St. Clair, Secretary

Jo Anne Price, Secretary

Contents

Page

Chapter 1. Summary 1

Chapter 2. Federal Scientific and Technical Information and the US. CompetitiveEdge 7

STI and Science and Technology Policy 7sn and Science Education 9STI and Research an" Development 10STI and Product Development and Manufacturing 13

STI and International Leadership on Global Issues 15

sn and International Competitiveness: A Summation 17

Chapter 3. Reaching Consensus on Principles of Federal Scientific and TechnicalInformation Dissemination 19

Federal Role in Information 1,..ssemination 19Principles of 511 Dissemination 23

Chapter 4. Reaching Consensus on the Open Flow and Governmentwide Disseminationof Federal STI 33

Open Flow of STI 33Role of Governmentwide Dissemination and Archival Agencies in STI 36

Chapter 5. Framework for a Presidential Initiative on Scientific and TechnicalInformation 43

Strengthening the OSTP Role 43Establishing Advisory Committees on STI 46Redefining OSTP-OMB Working Relationships on STI 47Upgrading Agency STI Management 49Developing Standards and Directories for STI 49Launching an 511 Education Initiative 52Improving International Exchange of STI 55

Appendix A. Technological Opportunities for Managing and Disseminating FederalScientific and Technical Information 57

Appendix B. List of Acronyms 67

Index 68

TablesTable Page1. Cost Breakdown for Ellustrutive National Geophysical Data Center CD-ROM Products . 282. Illustrative Use of Federal STI as Scienc_ Education Course Supplements 533. Illustrative Weaknesses in Current Global Change Data Management 56

!J

vi

Chapter 1

Summary

The United States must make berer use of itsscientific and technical information (STI) re-sources, if it wishes to be competitive in worldmarkets and maintain its leadership. STI is anessential ingredient of the innovation processfrom education and research to product develop-ment and manufacturing. It is a major product ofthe $65 billion per year the U.S. Governmentspends on research and development (R&D);researchers need ready access to STI if they areto stay at the cutting edge,' Many issues of ourtimehealth, energy, transportation, and cli-mate changerequire STI to understand thenature and complexities of the problem and toidentify and assess possible solutions. SU isimportant not only to scientists and engineersbut to political, business, and other leaderswho must make decisions related to scienceand technology, and to the citizens who mustlive with the consequences of these decisions.

The electronic collection, storage, and dis-semination of STI is a vision of the future thatis rapidly becciling reality. Electronic STIoffers the prospect of fast, efficient, and inex-pensive access to databases and documents.Scientists now use online computer networks totransmit STI around the nation and throughoutthe world. Others are experimenting with com-pact optical disks that can store a quarter millionpages of text on one disk.2

The Federal Govc:rnment has a golden oppor-tunity to help the United States sustain acompetitive position in scientific and technicalinformation. The United States has, at themoment, the necessary information and technol-

ogy base on which to build a strong nationaleffort. Congress intended that the President'sOffice of Science and Technology Policy (OSTP),established in 1976,3 provide executive brawlleadership on STI; OSTP has thus far failedthis mission.4

During the 1980s, STI was subsumed in thelarger debates over national information policyand science and technology policy. The Officeof Management and Budget (OMB) dominatedexecutive branch information policymaking andshowed little interest in STI. OSTP failed torecognize STI as an integral part of overall S&Tpolicy, and did not assert itself in many of thepolicy issues that affected STI. Federal STIprograms suffered as a result.

Executive branch leadership is imperativebecause STI is generated by many Federal R&Dagencies that must be coordinated if Lie govern-ment's STI efforts are to be successful. Agen-cies have set up a variety of ad hoc coordinatingmechanisms for specific aspects of STI; but anoverall, integrated approach is lacking. One ofthese existing committees could be expandedand chartered to serve a broader purpose.Alternatively, a new high-level interagencycommittee on STI could be established, withrepresentatives from the R&D programs thatgenerate STI, the agency data centers andtechnical document distribution offices, andgovernmentwide dissemination agencies suchas the Governm -:nt Printing Office (GPO) andNational Technil al Information Service (NTIS).

Whether through an interagency committee,OSTP and OMB guidance, or other means, the

!See, for example, U.S. Congress, House, Committee on Science and Technology, The Impact of Thformation Technology on Science, Science PolicyStudy, Background Paper No. 5 prepared by the Congressional Research Service, 99th Cong., 2nd sess. (Washington, DC: U.S. Government PrintingOffice, September 1986); and National Academy of Sciences, Committee on Science, Engineering, and Public Policy, htformation Technology and theConduct of Research (Washington, DC: National Academy Press, 1989).

2See the appendix for a discussion of technological opportunities.

3Public Law 94-282, the "National Science and Technology Policy, Organization, and Priorities Act of 1976," May 11, 1976.

40STP is in the process of deciding how to address 511 issues in the Bush Administration; see remarks of OSTP Director, D. Allan Bromley, beforea March 21, 1990, forum of the Federal Library and Information Center Committee, Washington, DC.

-1-

2 Helping America Compete: The Role of Federal Scientific and Technical Information

,,'I'Vertioill4 WI!

Photo credit: EROS Data Center, USGS

The Earth Resources Observation Systems (EROS) Data Center in Sioux Falls, SD, receives and stores data from Landsat andother Earth-observing satellites. The total earth sciences data volume managed by Federal agencies (primarily NASA, USGS, andNOAA):s projected to increase over two orders of magnitude by the year 2000 to about 10,000 terabytes. NASA's Earth Observing

System alone will generate an additional terabyte of data every day; this is equivalent to 10,000 Washington, DC telephone books(white pages) or 520,000 text books (at 200 pages each) per day.

success of the Federal STI program will dependon progress in four key areas:

1. technical standards for databases and doc-uments (graphics as well as text), so thatSTI can be electronically moved amongagencies and users with ease and effi-ciency;5

2. indexing of databases and documents, sothat STI users in and out of the governmentknow whet and where STI exists:6

3. funding for basic STI activities in agencyR&D budgets, to ensure the quality of STI,its proper storage, and dissemihation tousers:7

4. end-user involvement in all agency STIprograms, so that Federal STI is dissemi-nated in user-friendly formats that meetuser needs and are compatible with theequipment and technical capabilities of theusers.

Electronic media offer the only way to man-age the massive volume and complexity ofFederal STE; yet Federal agendes must avoid"technophoria," i.e., unrealistically optimis-tic expectations of the technology.8 The transi-tion to electronic formats, whilt- inevitable, willbe difficult for many users.9

5The standards-setting effort would heavily involve the National Institute of Standards and Technology, the designated lead standards agency for

the Federal Government, and rely to the maximum extent possible on standards developed by private sector and international standards-setting

organizations,

'Indexing would be coordinated with related activities by the National Technical Information Service and Government Printing Office; however,

preparation of keywords and abstracts could, in any event, be the responsibility of the R&D agencies and their contractors and grantees.

7For a discussion of the severe problems that result from uncle runding of agency STI activities, see U.S. General Accounting Office, Space

Operations: NASA Is A'ot Properly Safeguarding Valuable Data From Past Missions, Report to the Chairman, Committee on Science, Space, and

Technology, U.S. House of Representatives, GAO/IMTW-90.1, March 1990. GAO is conducting similar audits of NOAA and USGS da;a archives.

'Tenn coined by C R. McClure of Syracuse University in testimony before an Oct. 12. 1989, hearing of the House Committee on Science, Space,

and Technology, Subcommittee on Science, Research, and Technology.

9While OTA projects a domina -it role for electronic forrnats, paper(and to a lesser extent microfiche) formats will be heavily used for the foreseeable

future. But most paper documents wiil be produced by electronic printing from computerized databases; the same electronic database can be used to

disseminate STI online over networks, on magnetic tape or diskette, or on compact optical disk, as well as on paper or microfiche, In this way, it will

be possible to accommodate both high-tech and low-tech needs of STI users.1

Chapter 1Summary 3

t 'eammate...4e

it

A ,

tz.

Ar 'g Yei

Photo credit: Natinnal Space Science Data Center, N4S,4

The National Space Science Data ChAter in Greenbelt, MD, is the largest space data-afchive M the woild, whh about 120,000magnetic computer tapes of digital data currently on file. The computer tape is still the dominant medium for storing space data, butthe tapes are difficult and expensive for many researchers to use. New technologies make it possible to car ry out a gradual transition

from magnelic tapes to higher density storage meda such as optical disks or tapes and digital tape cartridges.

Progress on STI also depends on resolvinggovernmentwide information dissemination pol-icy issues. During the 1980s, OMB used itsauthority under the Paperwork Reduction Act tofavor private-sector responsibility for Federalinformation dissemination. The OMB view wascontroversial and sent mixed signals to theFederal R&D agencies about whether electronicSTI should be aggressively pursued. Legislationpending before Congress would rebalance gov-ernment policy to emphasize that Federal agen-cies (including the R&D agencies) have the

primary responsibility for dissemination of in-formation generated for agency missions, withan important supplementary or complementaryrather than preemptiverole for the privatesector.10 This legislation also addresses infor-mation management, pricing, public access, dueprocess, and other policy matters that woulddirectly affect STI.11

The U.S. scientific and technical enterprisedepends on the open exchange of STI. Until the1980s, the premise of openness was generallyviolated only in narrowly defined areas of

toSee S. 1742, the "Federal Information Resources Management Act of 1989," 101st Cong., 1st sess., Oct. 6, 1989; and H.R. 3695, the "Papery/odaReducation and Federal Information Resources Management Act of 1989," 101st Cong., 1st seu., Nov. 17, 1989, ordered to bt reponed by the HouseCommittee on Governmast Operations, Mar. 13, 1990. Also see Office of Managmnent and Budget, "Second Advance Notice of Further PolicyDevelopmert on Disse:umation of Information," Federa/ Register, vol. 54, No. 114, June 15, 1989, pp. 25554-25559; U.S. Congress, Hou.se, Commateeon Government Operation!, Subcommittee on Government Information, Justice, and Agriculture, Federal Information Dissemination Policies andPractices, Hearings, 101st C.ong., ht sess., Apr. 18, May 28, and July 11, 1989 (Washington, DC: U.S. Government Printing Office, 1989); and U.S.Congress, Seuate, Committee on Governmental Affairs, Subcommittee on Government It fonnation and Regulation, Reauthorization of the PaperworkReduction Act, Hearings, 101st Cong., 1st sees., June 12 and 16, 1989 (Washington, DC: U.S. Government Printing Office, 1989).

"See OTA comments on S. 1742, prepared for a Feb. 21-22, 1990, hearingff6e Senate Committee on Governmental Affairs.


e 7

Plato cm& National Smog Science Data Canter, NASA

NASA and other Federal science agencies are currentlyexperimenting with optical disksprimarily the 12-InchWORM (Write Once-Read Manytlmes) and 4.75-inch

CD-ROM (Compact Disk-Read Only Memory). A typicalCD-ROM can store up to about 600 megabytes. This Is

equivalent to roughly 300,000 text pages (at 250 words petpage), 1,650 floppy diskettes, 30 of the 20-megabyte hard

disks, or 15 of the 1,600 bits-per-inch 9-track magneticcomputer tapes. For many applications, CD-ROM Is muchless expensive than computer tapes, and r Nukes only arNcrocomputer and CD-ROM reader rather than a more

expensive mini- or main-bap:a computer needed for tapes.

national security. In recent years, the bases forrestrictions on open dissemination of informa-tion have been extended to: a) so-called "un-classified but sensitive" STI that might compro-mise national security; b) the transfer of controlover federally fundel technical data and tech-nology from the government to the privatesector to promote commercialization; and c)limitations on access by foreign governmentsand companies to Federal STI to maintlin theeconomic competitiveness of the United States.

Globalization of the economy means +hat agrowing fraction of U.S. domestic R&D compa-nies operate under foreign ownership or withforeign partners, just as many U.S. corporationshave their own foreign subsidiaries or partners.Similar trends are evident in the commercialinformation sector, to the point where one

cannot assume that a U.S. information vendoroperates under domestic rather than foreignownership, and vice versa. Under these condi-tions, th- old approaches to controlling the flowof STI do not work and need to be revisited.Many of them may not be needed at all.

Another vexing issue is the role of thegovernmentwide dissemination and archivalagencies in the decentralized, increasingly elec-tronic environment of Feckral STI. The creation,storage, and dissemination of electronic STI isdecentralized within the R&D agencies forseveral reasons:

The volume of STI is vast. Centralizing allSTI in one databank is not technically oradministratively fusible.The technical systems for creating, storing,and disseminating STI are typically closelytied to agency automation systems. Cen-tralizing STI could foreclose innovationand opportunities for improving productiv-ity in the agencies.The diversity of STI uses spans a numberof disciplines and research areas. Central-izing STI would complicate communica-tions between the STI process and theusers.

A key challenge is how to preserve andstrengthen the indexing, archiving, and distribu-tion roles of the: 1) GPO:12 2) DepositoryLibrary Program (DLP); 3) NTIS; and 4)National Archives and Records Administration(NARA). These agencies realize the need forchange, but have thus far failed to developworkable strategies for electronic STI. If theyare to flourish in the unfolding electronicenvironment, GPO, DLP, NTIS, and NARAmust become innovative, flexible, and competi-tive in anticipating and meeting electronicinformation needs.°

%eluding the Superintendent of Documents (SupDoes), who administers cataloging, sales, international exchange, and depository library programs,

among others.

13For background discussion, see U.S . Congress, Office of Technology Assessment, Informing the Nation: Federal Information Dissemination in an

Electronic Age, OTA-CIT-396 (Washington, DC: U.S. Government Printing Office, October 1988). For pending legislation, see H.R. 3849, the

"Government Printing Office Improvement Act of 1990," 101st Cong., 2d. sess., Ian. 23, 1990, which centrally addresses GPO and DLP; S. 1742, op.

cit., footnote 10, and H.R. 3695, op. cit., footnote 10, which touch on GPO, DLP, and NARA; H.R. 4329, the "American Technology Preeminence Act,"1 9

Chapter 1-Summary 5

A Presidential STI initiative could focusattention on these important issues. The list ofdesignated presidential science and technol-ogy priorities, such as science education,14technology transfer,il high-performance com-puting and netwoddne internatiwni wmpelitive-ness," and global change,n justifies addi-tional emphasis on STI. STI is crucial to thesuccess of each of these initiatives, starting withthe role of STI in science education.

Low-cost, user-friendly electronic STI couldstimulate computer-based science, mathemat-ics, and engineering -ducation. Pilot projectshere and abroad indicate that junior and seniorhigh school students (and even some in the

elementary grades) can handle electronic data-bases as part of the science curriculum. Computer-based STI can help capture the interest, imagina-tion, and enthusiasm of students through "hands-on" science that could improve the quality ofscience education.

Improving the "information literacy" ofscientists and engineers must go hand-in-handwith upgrading STI; otherwise, the best STIsystems will fall short. By integrating STIaccess, retrieval, and use into science educationat all levels, the research skills and productivityof U.S. scientists and engineers could be strength-ened in the long-term.

footnote 13 continued101st Cong., 24. sess., introduced and ordered to be reported by the House Committee on Science, Space, and Technology, Mar. 21, 1990, whichaddresses NTIS modernization, indexing, and electronic dissemination; statements of Fred B. Wood, OTA, and other witnesses before a Mar. 7-8, 1990,hearing on Hit. 3849 by the House Committee on Administration. Subcommittee on Procurement and Printing; statements of Red B. Wood, OTA, andother witnesses before a Mar. 8, 1990, hearing on NT1S modernization by the House Committee on Science, Space, and Technology, Subcommittee onScience, Research, and Technology, OTA and other comments provided on S. 1742 in connection with a Feb. 21-22, 1990, bearing of the SenateCommittee on Govennental Affairs; statements V' Joseph E. Jenifer, Acting Pill tic Printer, and other witnesses before a Feb. 7, i989, hearing of theHouse Committee on Appropriations, Subcommittee on the Legislative Branch, 4 May 23, 1989, hearing of the Committee on House Administration,Subcommittee on Procurement and Printing, and a July 11, 1989, hearing of the House Committee on Government Operations, Subcommittee onGovernment thformation, Jusfice, and Agriculture; and the statement of Robert Houk, Public Printer, before au Apr. 6, 1990, hearing of the SenateCommittee on Appropriations, Subcommittee on the Legislative Branch. Also see U.S. Congress, House, Committee on House Administration,Subcommittee on Procurement and Printing, Title 44 U.S.C.-Review, Hearings, 101st Cong., 1st sess.. May 23 and 24, and June 28 and 29, 1989(Washington, DC: U.S. Government Printing Office, 1989), and U.S. Congress, House, Committee on Science, Space, and Technology, AmericanTechnology Preemintatce Act, Report 101-481, Part 1, to accompany H R. 4329, 10Ist Cong., 24 sess. (Washington, DC: U.S. Government PrintingOffice, 1990).

14For background discussion, see American Association for the Advancement of Science, Science for All Americans: Project 2061 Rer rt on LiteracyGoals in Science, Mathematics, and Technology (Washington, DC: 1989); U.S. National Research Council, Everybody Counts: A Rep( to the Nationon the Future of Mathematics Education (Washington, DC: National Academy Press, 1989); and U.S. Congress, Office of Technology Assessment,Educating Scientists and Engineers: Grade School to Grad School, OTA-SET-377 (Washington, DC: U.S. Government Printing Office, June 1988),and Power On! New Fools for Teaching and Learning, OTA-SET-379 (Washington, DC: U.S. Government Printing Office, September 1988).

1:17o1 related discussion, see U.S. Congress, Office of Technology Assessment, Technology and the American Economic Transition: Choices for theFuture, OTA-TET-283 (Washington, DC: U.S. Government Printing Office, May 1989), Holding the Edge: Maintaining the Defenze Technology Base,OTi 1-ISC-420 (Washington, DC: U.S. Government Printing Office, April 1989), and Arming Our Allies: Cooperation and Competition in DefenseTec ltnology, OTA-ISC-449 (Washington, DC: U.S. Government Printing Office, May 1990). Also see H.R. 4653, the "Export Facilitation Act of 1990,"Wist Cong., 24 sess., Apr. 26, 1990, ordered to be reported by the House Committee on Foreign Affairs, May 10, 1990.

16For discussion, see S. 1067, the "High-Performance Computing Act of 1990," 101st Cong., 1st sess., May 18, 1989, ordered to be teported by theSenate Committee on Commerce, Science, and Transportation, Apr. 3, 1990; H.R. 3131, the "National High-Performance Computer Technology A ctof 1989," 101st Cong., 1st sess,, Aug. 3, 1989; H.R. 4329, Title V11, the "National High-Performance Computer Technology Program Act of 1990,"101st Cong., 24 s033., introduced and onlered to be reported by the House Committee on Science, Space, and Technology, Mar. 21, 1990; U.S. Congress,Office of Technology Alsessment, High Performance Computing & Networking for Science, OTA-BP-CIT-59 (Washington, DC: U.S. GovernmentPrinting Office, September 1989); and Executive Office of the President, Office of Science and Technology Policy, The Federal High PerformanceComputing Program, Sept. 8, 1989.

17For related discussion, see U.S. Congress, Office of Technology Assessment, Making Things Better: Competing in Manufacturing, OTA-ITF.-443(Washington, DC: U.S. Government Printing Office, February 1990) and Critical Connections: Communication for the Future, OTA-CIT-407(Washington, DC: U.S. Government Printing Office, January 1990).

18See, for example, the discussion of global chahge data management needs in U.S. Federal CeordinAting Council for Science. Engineezing, andTechnology, Committee on Earth Sciences, Our Changing Planet: The FY 1990 Global Change Resee ch Plan (Washington, DC: Office of Science

and Technology Policy, July 1989), pp. 91-99, and U.S. Natiocal Aeronautics and Space Mministe don, Earth Systems ScienceCommittee, Earth

Systems Science: A Closer View (Washington, DC: NASA, January 1988). 1 A


,

aripas.

kwierree«...#10`

k

4

*0110110Maie

..411111hmsfillig

Photo credit: Office of Scientific and Technical Information, DOE

The Department of Energy's Office of Sdentific andTechnical Information Is implementing Information systemsthat!. se magnetic, optical, and online electronic technolo-gies. DOE alone generates about 30,000 technical docu-ments and articles per year; the governmentwide volume isabout 200,000 items annually. New technologies areessential to cop wfth the burgeoning scientific andtechnical literatur for example, one double-sided 12-InchWORM can store about 1.2 million text pages or 6,000technical documents (at 200 pages each).

Photo credt: Natloral Library of Medicine

Tbis "Electronic Cardiology Textbook" represents thestate-of-the-art in the use of electronic imaging to commu-nicate scientific and technical information. The "textbook"stores Images and sounds of the human heart on an opticaldisk; the user turns the pages electronically with a mouseand microcomputer.

Photo credit: Government Prindrig Office

Online information networks SONO several important needsof the scientific and technical community. GPO uses anonline system, shown here, to receive documents fromremote locations; Users simply "dial-up" the GPO systemand transmit their material to GPO for processing. Onlinenetworks are used by several Federal science agencies totransmit documents, data, and messages; search biblio-graphic databases; transfer large streams of data; andremotely access large-scale high-performance computers.

1 5

Chapler 2

Federal Scientific and Technical Informationand the U.S. Competitive Edge

The drumbeat of political, economic, and environ-mental change around the world presents the UnitedStates with perhaps its greatest challenge sinceWorld War IL The global society is more competi-tive with respect to scientific and technologicalachievement, educational attainment, market devel-opment, and political leadership in addressing inter-national issues. Thi? Special Report examines indetail one key element in restoring U.S. competitivestrengththe role of suentific and technical infor-mation (ST) developed by or for the FederalGovernment.

The importance of STI stems from its critical rolein all phases of the innovation process. Theseinclude education, basic research, applied researchand development, product development and manu-facturing, and the application of science and technol-ogy to meet the needs in the commercial, not-for-profit, and governmental markets.

STI and Science and Technology Policy

STI policy is a component of overall Federalscience and technology (S&T) policy. The latterincludes the range of Federal actions that caninfluence the conduct of U.S. research and develop-ment (R&D) and conversion of R&D results intoproducts and sei vices to satisfy domestic needs andcompete with foreign suppliers. Federal S&T policyis diverse, and includes: direct Federal funding (e.g.,for basic research); the conduct of research inFederal laboratories; tax incentives for private sectorPAD (e.g., accelerated depreciation and tax cre lit.;fol. technology investments); and rules or guidelines

to waive antitrust laws for industry R&D consortia,among others; as well as policies and actions forcollecting and disseminating STI.

This Special Report focuses primarily on the STIcomponent of Federal science and technology pol-icy. Related OTA studies address other aspects ofS&T policy.' STI is an indispensable part of theR&D infrastructure. But more than that, it is anational asset that can contribute to strengtheningthe technological foundation of the U.S. economy.Debates may rage over the role of the FederalGovernment in promoting industrial competitive-ness. But clearly it is incumbent on the governmentto improve the ST1 base on which many public andprivate R&D decisions are made. The challenge is tohelp STI more fully serve national priorities.

STI frequently has been lost in the broader debatesover: U.S. technology policy; the role of scienceadvice in the White House; and, in recent years, theneed for and shape of a national information policy.2During the 1980s, the case for STI has beenbolstered by considerable research (discussed laterin this chapter) that has documented the role of STIin priming the pump of R&D and innovation. STI is,indeed, at the heart of the process by which sciencegenerates new ideas that in turn fuel technologicalinnovation. The concern over U.S. competitivenessgives new impetus to the need for a sound ST1policy.

Discussions of U.S. competitiveness typically aredominated by the international economic dimen-sion, i.e., the ability of U.S. companies to compete

:See the ongoing U.S. Congress, Office of Technology Assessment studies, "Basic Research for the 1990s," scheduled for completion in winter199'., and "Information Research and Technology: High Performance Computing and Networking for Science," scheduled for completion in fall 1990.Also see, for example, Office of Technology Assessment, The Regulatory Environment of Science, OTA-TM-SET-34 (Washington, DC: U.S.Government Printing Office, February 1986); Technology and the American Economic Transition: Choices for the Future, OTA-TET-283 (Washimon,DC: U.S. Government Printing Office, May 1988); Educating Scientists and Engineers: Grade School to Grad School, OTA-5ET-377 (Washington,DC: U.S. Government Printing Office, June 1988); Holding the E4e: Maintaining the D#'enseTechnology Base, OTA-ISC-421 (Washington, DC: U.S.Government Printing Office, April 1989); High Performance Computing & Netwo4ing for Science, OTA-BP-CIT-59 (Washington, DC: U.S.Government Printing Office, September 1989); Computer Software and Intellectual Property, OTA-BSCIT-61 (Washington, DC: U.S. GovernmentPrinting Office, March 1990); and Making Things Better: Competing 'n Manufacturing, OTA-ITE-443 (Washington, DC: U.S. Government PrintingOffice, February 1990).

2See, for example, J.M. Logsdon, 'Ibward a New Policy for Technology: The Outline Emerges," Technology Review, October/November 1972,pp. 36.42; U.S. Department of Commerce, Office of the Assistant Secretary for Science and Technology, U.S. Technology Policy: A Draft Study(Washington, DC: National Technical Informatijn Service, March 1977); U.S. Congress, Office of Technology Assessment, Computer-Based NationalInformation Systems: Technology and Public Policy Issues, OTA-CIT-146 (Washington, DC: U.S. Government Printing Office, October 1981); C.R.McClure and P. Hernon United States Scieruific and Technical Information: Views and Perspectives (Norwood, NJ: Ablex Publishing Corp., 1989).

71 CI

8 Helping America Compete: The Role of Federal Scientific and TechnicalInformation

MIONMINNIMIIIMM=MEMINEIIIMP` Ammo*

STI is an indispensable part of the R&Dinfrastructure. But more than that, it isa national asset that can contribute tostrengthening the technological found&tion of the U.S. economy.

effectively in the international marketplace. STI ismultidimensional and involves a "life cycle" ap-proach to competition ranging from education toresearch to manufacturing to marketing and publicpolicy. When viewed in a total competitive context,STI is the backbone of our competitive edge.

STI is linked with several national goals. STIresulting from Federal R&D is intended to promotethe advancement of scientific knowledge and techni-cal applications of that knowledge. It also servesother national goals, including: improving the abil-ity of U.S. industrial firms to compete in theinternational economy; strengthening the U.S. de-fense and civilian technology base; improving U.S.science and engineering education; promoting inter-national cooperation on global science and technology-related problems; and enhancing the free flow of STIrequired by an open, democratic society.

America's ability to achieve these national goalsin part through STI has been limited by our inabilityto clearly define the contribution of ST1 to thesegoals and to reconcile the conflicts over competinggoals that inevitably arise. The policy framework forST1 dissemination must recognize and spell out therole of STI at each stage of education, research, andapplication. For example, STI about solar pho-tovoltaic energy can be structured in terms of whatis needed for: educating future solar energy scien-tists and engineers; supporting basic research on thephysics and electronics or photovoltaic energy;facilitating applied research on photovoltaic cells;enhancing the development of prototype and com-mercial photovoltaic energy systems, and the manu-facturing technology for production of such sys-tems; encouraging the integration of photovoltaicsinto U.S. commercial and defense energy applica-

tions; and informing the national and internationaldebate on alternative energy and environmentalpolicies.

The U.S. competitive challenge is epitomized bythe so-called "technology-intensive" industrial sec-tors, such as computers, telecommunications, elec-trical machinery, instruments, chemicals, and trans-portation. These sectors have been the mainstays ofthe U.S. post-World War 11 economy, due to highrates of .growth in real output, productivity, andemployment, and for many years contributed to apositive trade balance. Recently, even the strongestU.S. industrial sectors have come under intensifiedcompetition. This is due in part to the rise of theglobal economy and dominance of multinationalcompanies (that operate across national boundaries),the continuing Federal budget deficit and negativetrade balance and resulting effects on internationalmonetary exchange rates, and the partially offsettinggrowth of the service sectors (where the UnitedStates competes strongly in some areas, notablyinformation services). But the new competitiverealities have spurred attention to other root causes.3

In a world of rapid technological change, success-ful competition is driven as much by the informationskills of the work force and by the timing ofinformation access, as it is by the brute intellectual,financial, and natural resources of the competitors.Scientific and technical advancements are information-intensive, and those who know how to obtain and useSTI will have a competitive edgewhether thecompetition is over market share or over intellectualleadership on global issues.

In this context, the role of electronic technologiestakes on significance, since the generation, location,

STI is multidimensional and involves a"life cycle" approach to competitionranging from education to research tomanufacturing to marketing and publicpolicy. When viewed in a total competi-tive context, STI is the backbone of ourcompetitive edge.

3For a comprehensive analysis, see U.S. Congress, Office of Technology Assessment, Technology and the American Economic Transition: Choices

I'm. the Future, OTA-TET-283 (Washington, DC: U.S. Government Printing Office, May 1988); and OTA, Interna'anal Competition in Services,OTA-1TE-328 (Washington, DC: U.S. (Jovernment Printing Office, July 1987).

1 P,

Chapter 2Federal Scient9lc and Technical Information and the U.S. Competitive Edge 9

Scientific and technical advancementsare information-intensive, and those whoknow how to obtain and use STI willhave a competitive edgewhether thecompetition is over market share or overintellectual leadership on global issues.

and retrieval of STI can be vastly speeded up.4 STIusers need to be better educated about how Federal(and other) STI may help, how and where Federal(and other) STI may be accessed, and use of the toolsthat facilitate access (e.g , online databases, compactoptical disks, and bibliographic and search andretrieval software). And science agency managersneed to do a better job at including STI dissemina-tion and use as integral parts of agency R&Dprograms.

STI and Science Education

Science and technical education is the foundationfor the technological and economic competitiveposture of the United States. Congress in enactingthe "Education and Training for a CompetitiveAmerica Act of 1988" noted:5

"our [Nation's] standing in the internationalmarketplace is being further eroded by thepresence in the workforce of millions of Amer-icans who are functionally or technologicallyilliterate or who lack the mathematics, science,foreign language, or vocational skills needed toadapt to the structural changes in the globaleconomy;"our competitive position is also being erodedby declines in the number of students taking

advanced courses in mathematics, science, andforeign languages and by the lack of moderntechnical and laboratory equipment in oureducational institutions;"restoring our competitiveness and enhancingour productivity will require that all workerspossess basic educational skills and that manyothers possess higbly specific skills in mathe-matics, science, foreign languages, and voca-tional areas."

Recent OTA studies have identified a wide rangeof actions to improve science education: upgrade thequality of elementary, secondary, and higher educa-tion with respect to science and engineering; in-crease student interest in science and engineering;and expand the number of science and engineeringstudents (and ultimately the pool of trained scientistsand engineers).6

Several of these actions relate directly to FederalS For example, OTA found that "hands-on"computer-based science learning can increase stu-dent interest in the subject matter and enhancestudent learning. OTA also noted the growing role ofcomputer-based science in science museums, sci-ence centers, and science fairs around the country.Overall, availability of Federal STI in low-cost,user-friendly electronic formats could add an impor-tant dimension to computer-based mathematics,science, and engineering education. School librariescan serve as a focal point for teacher and studenttraining in the use of online and compact optical diskmedia, and can provide a shared computer resourceavailable to support the science curriculum.7 Thiscould be an extension of the role already performedby library staff at many college and universitylibraries and at some of the larger and better-fundedpublic libraries. In general, strong library mediaprograms at the elementary and secondary levels

4See, for example, J. Bortnick and N.R. Miller, The Impact ofInformation Technology on Science (Washington, DC: Congressional Research Service,July 1985); National Academy of Sciences, Committee on Science, Engineering, and Public Policy, Information Technology and the Conduct of Research(Washington, DC: National Academy Press, 1989); and U.S. Congress, Office of Technology Assessment, High Performance Computing andNetworking for Science, OTA-BP-CIT-59 (Washington, DC: U.S. Government Printing Office, September 1989).

5Puhlic Law 100-418, the "Omnibus Trade and Competitiveness Act of 1988," 100th Cong., 2d sess., Aug. 23, 1988, Title VIEducation andTraining for American Competitiveness, sec. 6002(a)(3-5).

6See U.S. Congress, Office of Technology Assessment, Educating Scientists and Engineers: Grade School to Grad School, OTA-SET-377

(Washington, DC: U.S. Government Printing Office, June 1988); Power On! New Tools for Teaching and Learning, O1A-SET-379 (Washington, DC:U.S. Government Printing Office, September 1988); Elementary and Secondary Education for Science and Engineering, OTA-TM-SET-41(Washington, DC: U.S. Government Printing Office, December 1988); and Higher Education for Science and Engineering, OTA-BP-SET-52(Washington, DC: U.S. Government Printing Office, March 1989).

7See, for example, J.W.Ieisner, "Learning at Risk: School Library Media Programs in an Information World," School Library Media Quarterly,Fall 1985, pp. 11-20; B.K. Stripling, "Rethinking the School Library: A Practitioner's Perspective," School Library Media Quarteri., , Spring 1989, pp.

136-139.1 CD


School libraries can serve as a focalpoint for teacher and student training inthe use of online and compact opticaldisk media, and can provide a sharedcomputer resource available to supportthe science curriculum.

correlate with improved student skills in use oflibrary and information resources, and in studentachievement both overall and in specific sciencesubjects.8

Pilot projects have shown that junior and seniorhigh school students can readily handle (:,omputer-based bibliographic 2earches as an aid to course-work.9 Teachers concluded that database searchingenhanced student thinking and research skills. On-line searching was also used to augment the scienceeducation curriculum. For example, students con-ducted online searches on topics such as theArmenian earthquake, space sickness, and the cli-matic effects of tropical deforestation.

Integratin STI access, retrieval, and use intoscience education at all levels could improve theresearch skills and productivity of U.S. scientistsand engineers in the long term. Various studies.havehighlighted the "inadequate information gatheringand management skills of the R&D community"and the lack of skills and/or motivation to useavailable bibliographic 1ools.10 Electronic dissemi-

nation of Federal STI could assist in attacking thisproblem.

Improving the "information literacy" of U.S.scientists and engineers must go hand-in-hand withupgrading ST'. Even the best STI system would fallshort if the users lack the skills to search biblio-graphic databases, retrieve and manipulate data, andscan documents. In many fields of science andtechnology, STI developed by other countries isincreasingly important. Foreign patents now accountfor about 50 percent of all U.S. patents. The numberof foreign scientific journals and articles is growingmuch faster than those published in the UnitedStates." U.S. researchers must learn to utilizeforeign STI, while making better use of domesticSTL The experience with Japanese STI suggests thatU.S. researchers are, by and large, not well-trainedin foreign languages and, generally, in techniquesfor accessing and utilizing foreign STI, and largelyfail to recognize the need for doing so.12

STI and Research and Development

The creation of new information and knowledgeis the major objective of R&D. This informationtakes many forms: information from basic researchon AIDS conducted by Federal laboratories; designand testing of prototype photovoltaic solar energycells by the Department of Energy (DOE); or thesynthesis of satellite data collected by the NationalOceanic and Atmospheric Administration (NOAA)to impmve understanding of the interaction of tneatmosphere and oceans in climate change.

Scientists and engineers involved in R&D oftenspend between one-quarter and one-half of their time

8See J.C. Mancall, "An Overview of Research on the Impact of School Library Media Programs on Student Achievement," School Library MediaQuarterly, Fall 1985, pp. 33-36.

9See, for example, MM. Bailey, J. Wieman, J. Newman, and N. Motomatsu, Research Goes to School II: How To Go Online to the InformationDatabases (Olympia, WA: Know-Net Dissemination Project, 1985); and N. Motomatsu and J.A. Newman, Research Goes to School III:Going Online

With Students (Olympia, WA: Office of the Superintendent of Public Instruction, 1986). In Australia, a recent survey identified 20 schools (2 primary

and 18 secondary) using online systems ''as a source of up-to-date infonnation for teachers and students and as a means of helping students to acquireinformation skills." Sec L.A. Clyde and J. Kiik, "The Use of Electronic Information Systems in Australian Schools: A Preliminary Survey," SchoolMedia Library Quarterly, Summer 1989, pp. 193-199. In the United States, the Oakland County (Michigan) School District has successfully pilood the

use of online bibliographic databases av tillable from a commercial vendor. Students, teachers, and administrators from the participating schools (3 juniorhigh and 3 high schools) enthusiasticoly embraced online searching. See Oakland County Schools, Educational Resource Center, "Database Searching

Pilot Project," Pontiac, MI, Nov. 9, 1989,1°See C.R. McClure, "Increasing Access to U.S. Scientific and Technological Information: Policy Implications," ch. 12 in C.R. McClure and P.

Hernon, United States Scientific and Technical Information Policies: Views and Perspectives (Norwood, NJ: Ablex Publishing Corp., 1989), pp.

319-354.

l I See D.W. King, D.D. McDonald, and N.K. Roderer, Sciennfic Journals in the United States: TheirProduction, Use, and Economics (Stroudsburg,

PA: Hutchinson ROSS Publishing Co.. 1981).

12See C.T. Hill, Japanese Technical Information: Opportunities To Improve U.S. Access, Report No. 87-818 (Washington, DC: Congressional

Research Service, Oct. 13, 1987); C.T Hill, "Federal Technical Information and U.S. Competitiveness: Needs, Owtortunities, and Issues," Government

Information Quarterly, vol. 6, No. 1, 1989, pp. 31-38. 1 (4

Chapter 2Federal Scientific and Technical Information and the U.S. Competitive Edge 11

on information-related activities that include bothanalyzing and reporting on one's own research andsearching for and applying the research results ofothers. Researchers in most disciplines spend about15 to 20 percent of their time just on reading the STIliterature, including scholarly joumal articles, con-ference proceedings, and technical reports.13 Re-searchers also find relevant ST1 through participa-tion in technical conferences and activities ofprofessional and scientific societies, and throughinformal letters, meetings, conversations, and, re-cently, electronic mail and bulletin boards.

Roughly three-quarters of researchers access STIliterature in order to apply other research fmdings toa current project and/or for professional develop-ment or current awareness of STI trends. About halfof the researchers read the S7'I literature to helpprepare an article, book, or report, and abouttwo-fifths to help prepare a lecture or presentation."Not surprisingly, empirical research in a range ofgovernment and private-sector settings has foundthat reading STI is positively correlated with pro-ductivity (e.g., number of reports or publicationswritten and presentations delivered).15

DOE estimates that federally funded energyresearchers spend, collectively, die equivalent ofover $1 billion per year of their time on STI (out ofan annual R&D budget of over $5 billion). This issplit about equally between generating new STI(e.g., writing technical reports) and reading otherST1 (e.g., journal articles), and amounts to roughlyone-fifth of the total DOE R&D budget. In addition,DOE spends about $250 million annually on STIinformation management, libraries, technical infor-mation centers, and other STI-related activities. Ifthe DOE estimate is extrapolated to the entireFederal R&D efforr, then Federal researchers spendroughly 12 billion dollars' worth of time each yearon STI.

The Department of Defense (DoD) has not madecomparable estimates. But assuming that the roughly

Scientists and engineers involved in R&Doften spend between one-quarter andone-half of their time on information-related activities that include both ana-lyzing and reporting on one's own re-search and searching for and applyingthe research results of others.

180,000 scientists and engineers doing DoD-relatedresearch work spend an average of 10 hours per weekon STI, the annual time investment is 144 millionhours. If time is valued, conservatively, at $30 perhour, then DoD researchers spend at least $2.7billion per year worth of their time on STI.16 Theactual figure is likely to be double or triple (e.g., $5to $7 billion per year, which would be consistentwith the DOE estimates), and even this would notinclude S71 time spent in connection with DoD test,evaluation, maintenance, and operational activities.

This kind of investment in STI is essential toscientific advancement and technical innovation thatare, in large measure, built on the cumulativeknowledge base of the scientific and technicaldisciplines. Breakthroughs may come slowly or, onoccasion, may occur quickly as a result of ground-breaking research, a new interdisciplinary synthesis,or a "paradigm shift" where the cumulative knowl-edge leads scientists to revise their basic hypothesese.g., with respect to the susceptibility of the Earth toglobal change, and the role of the oceans, land,glaciers and ice sheets, biota, and the atmosphere inclimate change. Geology, glaciology, oceanogra-phy, and climatology are among the several scien-tific disciplines that benefit from and contribute toFederal R&D and STI. Likewise, rapid advances inour understanding of human health depend on theextensive exchange of STI among researchers in

"See N.K. Roderer, D.W. King, and S.E. Brouard, "The Use and Value of Defense Technical Information Center Products and Services," contractorreport prepared by King Research, Inc. for the Defense Technical Information Center, Mar. 31, 1983, p. 20 and references cited therein. Also see E.R.Siegel, "transfer of Information to Health Practitioners," in B. Dervin and MI. Voight(eds.), Progress in Communication Sciences,Vol. III (Norwood,NJ: Ablex Publishing Corp., 1982), pp. 311-334.

14See, for example, Roderer et al., op .cit., footnote 13, p. 34.

15King Research Inc. has obtained similar results for numerous Federal and private-sector organizations. For a summary, see J.M. Griffiths and n.W.King, "Evaluating the Perfonnance and Effectiveness of Information Services," paper prepared for the Mid-Atlantic Chapter, Medical LibraryAssociation, Rockville, MD, 1989.

16180,000 persons X 10 hours spent on STI/week X 50 weeks/year (assuming 2 weeks vacation) X $30/hour = $2.7 billion. DoD officials haveconfirmed the 180,000 persons as a reasonable estimate. n


disciplines such as biology, physiology, psychol-ogy, medicine, and nutrition.

Improving the use of ST1 could increase the returnon the Federal Government's substantial investmentin R&D, which is 'currently about $65 billion peryear and represents roughly one-half of the total U.S.investment in R&D. Assessing the value of informa-tion dissemination services and products, whetherFederal or otherwise, is obviously difficult. Onetechnique is to estimate the savings (benefits)resulting from using an ST1 service or product.Using this approach, each dollar spent on FederalST1 dissemination generates a a estimated directbenefit of at least $2 to $5 to users in the researchcommunity (e.g., in terms of time saved, duplica-tions avoided, etc.) and on occasion can reach intoth c. hundreds to thousands of dollars." Onlinedatabases are especially highly leveraged. If the costof originating the information is not included(presumably funded out of Federal R&D funds),some databases generate an estimated value (savingsor benefit in the eyes of the user) of $15 to $25 foreach dollar spent." This helps explain why manyonline users readily pay $15 to $25 per hour foronline access to government databases (and up to$150 to $200 per hour for commercial databases).

Users of technical reports from DOE's Office ofScientific and Technical Infornation indicate signif-icant savings for each report ir;ed, and that about 75percent of reports used yield some savings.° Sometypical examples of specific savings are:

A basic energy sciences researcher saved over$50,000 by obtaining STI that eliminated theneed to do a complete design from scratch of adouble-effect absorption cooling system.A health and environment researcher saved$5,000 through STI that mooted the require-

ment for certain tests on disposal of wastewaterfrom coal conversion.A fusion researcher saved a person-year ofeffort through ST1 that summarized prior re-lated research on ion beam propagation andfocusing.A nuclear researcher saved $1,000 through S7'Ithat provided calculationsthat would other-wise have had to be redoneon steam elec-tronic plant construction.

The benefits and savings from effective use ofFederal S71 include:

time saved in locating other researchers doingrelated work;time and money saved in minimizing duplica-tion of research effort;new insights or breakthroughs resulting frommore complete awareness of related research;new information not available elsewhere;better understanding of relevant Federal R&Ddirections; andtime and money saved in writing researchreports, papers, and articles.

Federal science agencies face a major challenge inmanaging the already immense and rapidly increas-ing volume of Federal SI'l so that it is accessible anduseful to researchers. For example, over 200,000new technical documents are generated each year asa result of Federal R&D, adding to the base of anestimated 4 million existing documents.2° Satellitedata and imagery are contributing to an S71 explo-sion in the space and earth sciences. The total earthsciences data volume managed by Federal agencies(primarily NASA, USGS, and NOAA) is roughly100,000 gigabytes.21 The total volume is projectedto increase by two orders of magnitude over the next5 to 10 years to 10 million gigabytes (or 10,000

"Sec Roderer et al., op. cit., footnote 13; and D.W. King, J.M. Griffiths, N.K. Roderer, and R.R.V. Wiederkchr, "Value ofthe Energy Data Base,"

contractor report prepared by King Research, Inc. for the U.S. Department of Energy, Mar. 31, 1982.

18For a good summary of relevant research issues and results, see B.C. Carroll and D.W. King, "Value of Infcnnation," Drexel Library Quarterly,

vol. 21, No. 3, Summer 1985, pp. 39-60.

19Ibid. The average savings per report was $1,300 (1982 dollars).

2°The Department of Energy (DOE) has generated a cumulative total of about 800,000 technical documents that are estimated to repiesent about

one-fifth of the governmentwide Mal. The National Technical Information Service (NTIS) clearinghouse includes about 2 million technical reports,

estimated to represent about one-half of the governmentlaide total. DOE generates about 30,000 new technical documents each year, estinuited to be

15 percent of the governmentwide total; NTIS adds about 65,000 new documents to its clearinghouse each year, estimated to represent about one-third

of the governmentwide total. There estimates are for technical documents and anicles published in the technical litemture, but exclude papers delivered

at technical conferences. For DOE, the annual volume of technical articles equals that of technical documents (about 15,000 each).

2)One gigabyte is equivalent to the volume of information contained in about 450,000 double-spaced typed pages of text. One terabyte equals 1,000

gigabytes or 1 trillion bytes; 100,000 gigabytes equals 100 terabytes. The current and projected earth sciencei, lata volumes are based on estimates by

the Interagency Working Group on Data Management for Global Change.

Chapter 2Federal Scient(c a. id Technical Information and the U.S. Competitive Edge 13

terabytes). When launched in the late 1990s, NASA' sEarth Observing System (EOS) will generate in afew months more data than the total U.S. archive ofLandsat satellite data collected over the last 18 years.

Electronic technologies can help the Federalscience agencies manage STI and ensure thatFederal data and documents are made available tousers in cost-effective, timely, and usable form. Thepotential for electronic STI dissemination is espe-cially great becausewhether data, documents, ordirectories to data or documentsit is generallywell suited to electronic formats. Electronic dissem-ination makes it possible to provide STI to research-ers in forms that are more convenient to retrieve andeasier to manipulate. This could enable manypotential new kinds of research and analysis. (Seethe apperdix for a detailed discussion of technologi-cal opportunities.)

STI and Product Developmentand Manufacturing

STI is also a key element in the transfer oftechnology from the laboratory to the productionline. The aerospace industry is a case in point. It issupported by a substantial Federal R&D investment,it has close collaboration with Federal agencies(civilian and military), industry, and academia, andit has a tradition of aggressively using the results ofFederal aerospace R&D in commercial applications.A recen, survey of aerospace engineers sheds lighton the dominant role of STI in an industry with asuccessful track record of commercialization andinternational competition.22

Ninety percent of the aerospace engineers identi-fied technical communication as very important. Onthe average, respondents spend about 35 percent oftheir workweek communicating technical informa-tion to others and about 31 percent of their weekworking with technical information received fromothers. Based on a 40-hour workweek, they spendroughly 26 hours on STI-related activities, a findingconsistent with other studies.23 These engineers

produce ou the average 1.6 government technicalreports and 1.9 other technical reports every 6months and use roughly 52 technical reports (abouthalf generated from Federal R&D) during that time.

The aerospace industry has been successful atcommercial utilization of Federal R&D because, formany decades, both government and industry haverecognized the importance of Federal R&D and thehighly leveraged role of STI and technology transferMechanisms in the commeicialization process. TheNational Aeronautics and Space Administration(NASA) has long-established relationships withacademia and industry to encourage the use of snNASA has established a network of IndustrialApplication Centers as part of its TechnologyUtilization Program. The centers provide technicalinformation to industry so that aerospace technologycan be used in commercial applications. rie effec-tiveness of this approach is illustrated by theseexamples:24

A Western Springs, Illinois, firm specializingin high-resolution, oblique, aerial photographyrequested that NERAC research the NASAdatabase for available information on film andcameras. Using the technology provided byNERAC, the firm improved the quality of itsaerial photographs, which now sell for upwardsof $2,000 each.A New York firm designed a computer-controlled robot using NERAC technologyfrom its NASA database. NERAC rapidlygathered information on robot off-line pro-gramming methodology so that the firm's R&Dstaff could implement the concept by using amicrocomputer system and graphics display.A firm dedicated to the development andmanufacturing of testing equipment requestedthat NERAC research noise control technol-ogy. The search identified technical informa-tion that led to the development of a veryhigh-performance hearing protector (with over34 dBA of insulation) that will be marketed for

T.E. Pinelli, M. Glassman, W.E. Olin, and R.O. Barclay, Technical Communications in Aeronautics : Results of an Exploratory Study, NASATechnical Memorandum 101534, Part 1 (Hampton, VA: U.S. National Aeronautics and Space Administration, Langley Research Center, February1989).

The survey instrument was sent to 2,000 randomly selected aerospace scientists and engineen (from the membership of the American Institute ofAeronautics and Astronautics). The response rate was 30.3 percent (606 out of 2,000). The affiliations of respondents were distributed es follows:academic (7%); industry (62%); not-for-profit (3%); NASA (12%); other government (16%). The professional duties of the respondents were. research(20%); administration/management (24%); design/development (37%); teaching/academic (6%); marketing/sales (6%); and other (7%).

23Sec Pinelli et al., op. cit., fcotnote 22; R.M. Davis, Technical Writing: Its Place in Engineering CurriculaA Survey of the Experience and Opinions

of Prominent Engir 7, Air Force Institute of Technology Technical Report 75-5 (Wright-Patterson Air Force Base, OH: September 1975).

24Provided by one of the oldest Industrial Application Centers, NERAC, Inc.

14 Helping America C .e: The Role of Federal Scientific and Technical Information

use on aircraft flight lines, in airports, atrifle/pistol ranges, etc.anyplace where fullnoise protection is required.A fffm designed a water distillation systemusing solar energy, based in part on technicalinformation provided by NERAC. The concen-tration of the Sun's rays causes a thermalreaction which initiates a distillation procRssthat results in water vaporization. The watervapor is distilled and collected as a usableproduct. The solar system is expected toprovide low-cost, fresh water supplies to re-mote, arid, and coastal towns.

Engineers rely more on their own knowledge andcontacts with colleagues and inhouse experts tosolve technical problems than on technical reports,journals, libraries, technical information centers, oronline technical information databases. The 1989NASA survey of aeronautical engineers confirmedthe same pattern revealed in numerous other studiespersonal, informal sources take precedence over themore formal, organized information sources.25

Engineers are likely to continue to rely in the firstinstance on personal, informal information sources.But there are significant opportunities to improvethe effectiveness of technical reports, libraries,information centers, and computerized databases.For example, at present, less than half of aerospaceengineers (44 percent, based on the NASA survey26)use electronic databases at all, and less than half (46percentn) use a library or technical informationcenter more than once a month. When electronicdatabases are searched, about two-thirds of theengineers use intermediaries (e.g., reference librari-ans) to perform the search. Engineers are generallyopen to the use of new information technologies.The four technologies receiving the highest percent-age of "I don't use it, but may in the future" in theNASA survey are:28

laser disk/videodisk/compact optical disk (65percent);videoconferencing (62 percent);electxonic bulletin boards (54 percent); andelectronit networks (53 percent).

Data on physical and chemical properties ofmaterials are another form of STI that is importantto industrial technology. The technical design ofautomobiles through electronic equipment mustcomport with a wide range of so-called "standardreference data" on the basic properties of materialsand industrial processes used in manufacturing. Theaccuracy of this data is essential. Engineers includea margin of safety in product and process designs,but if faulty design data are used, the product orprocess could fail under certain operating condi-tions.

The Federal Government plays a major role indeveloping, maintaining, and updating standardreference data. This is accomplished through re-search conducted in Federal laboratories, Federalsupport for university and inck_stiial Tx:search, government-industry cooperative initiatives, and Federal partici-pation in a wide range of professional and technicalstandards activities. Topics range from radiationchemistry to thermodynamics to metallurgy toelectronic properties to microwave spectral data.29

Private-sector managers recognize that the "costof not knowing" can be very high. Employees canwork hard, but if they do not work "smart," theresult can be losses instead of profitswhetherconducting research, competing for a Federal R&Dcontract, or selling a safe, reliable, and competitiveproduct in the marketplace.

STI plays a crucial role in the commercializationprocess by which the results of Federal r privateR&D are translated into marketable pr_ wets. Thechallenge is to increase the return on the Federal

25Sec Pinelli et aL, op.cit., footnote 22, p. 56; Ti. Allen, Managing the Flow oj fechnology: Technology Transfer and the Dissemination of

Technological Information Within the R&D Organization (Cambridge, MA: MIT Press, 1977; ILL. Shuchruan, Information Transfer in Engineering

(Glastonbtuy, CT: The Futures Group, 1981); S. Ballard, CR. McClure, TI. Adams. M.D. Levine, L. Ellison, TE. James, Jr., L.L. Malysa, and M. Meo,

Improving the Transfer and Use of Scientific and Technical Information: The Federal Role (Norman, OK: University of Oklahoma, Science and Public

Policy Program. 1986).

26Pine1li etal., op.cit., footnote 22, p. 66.

p. 65.

28Ibid., p. 73.

29Tbe National Research Council's Numerical Data Advisory Board (recently trimmed the Scientific and Technical Information Board) has issued

many relevant reports. For references, see National Research Council, Numerical Data Advisory Board, Improving the Treatment of Scientific and

Engineering Data Through Education (Walhington, DC: National Academy Press, 1986). Also see minutes of the Sept. 21, 1989, NDAB meeting

available from C. Carter, staff director, NRCNDAB, 2101 Constitution Ave., N.W., Washington, DC 20418.

p 3

Chapter 2Federal Scientific and Technical Information and the U.S. Competitive Edge 15

R&D investment through more effective utilizationof the STI resulting from Federal R&D. This can beachieved by several means, including: improving theusability of government technical reports (e,g.,formats, indexing, electronic retrieval); strengthen-ing the capabilities of libraries and informationcenters to meet STI needs; shaipening the skills ofscientists and engineers in using these resources; andcontinually upgrading the ability of technology-enhanced STI systems (e.g., online, compact opticaldisk) to provide affordable, user-friendly search andretrieval service. This is a challenge demanding thecombined efforts of government, industry, acade-mia, and the broader scientific and technical com-munity.

STI and International Leadership onGlobal Issues

Another part of the competitive edgein additionto education, R&D, and commercializationis theability of the United States to provide internationalleadership on a wide range of global problems.Providing and exchanging STI are important compo-nents of such leadership. The challenge is tomaintain and strengthen the open flow of relevantSTI in the face of greatly intensified global eco-nomic competition.

The United States has substantial informationassets, and these are being augmented by use ofelectronic avenues of dissemination. A case in pointis the NiEDLINE (MEDLARSMedical LiteratureAnalysis and Retrieval Systemonline) databasedeveloped by the National Library of Medicine andoffered online by NLM and several commercialvendors to the U.S. and foreign medical communi-ties. MEDLINE is used for a wide variety of patientcare, research, teaching, and administrative pur-poses.

Surveys indicate that MEDLINE is having asignificant effect on medical decisions." Physiciansuse MEDLINE information to select the mostappropriate tests and diagnose a wide range ofmedical problems in order to prescribe a treatmentplan. MEDLINE' s successes are well-known.31 Forexample, a pathologist examining a supposed " wart"

MM1111,

STI plays a crucial role in the commer-cialization process by which the resultsof Federal and private R&D are trans-lated into marketable products. Thechallenge is to increase the return on theFederal R&D investment through moreeffective utilization of the STI resultingfrom Federal R&D.

used MEDLINE to confirm a diagnosis of skincancer (polypoid melanoma) and develop a treat-ment plan. A phy 'In treating an adolescentpatient who collap during a foot race usedMEDLINE to rule out exercise-induced pancreatitisas a possible cause, and prescribed rest and absti-nence from food (which worked). And a physicianused MEDLINE to locate information in a Swissjournal about a new treatment for aplastic anemia.

Medical problems and rsearch lmow no nationalboundaries, and the effectiveness of databases suchas MEDLINE depends on international collabora-tion in the collection and exchange of medicalinformation. Computerized databases have becomeessential to this process, with both NLM and privatevendors making global electronic access possible.User-friendly microcomputers are bringing access toMEDLINE to the grassroots. In the United States,medical personnel can access MEDLINE using"Grateful Med," a search and retrieval softwarepackage developed by NLM and sold by theNational Technical Information Service (NTIS) for$29.98 per copy. Grateful Med runs on IBM-compatible and Apple Macintosh personal comput-ers.

Another area of intensive Federal informationactivity with significant international implications isgeographic information. Computerized geographicinformation systems ("GIS") make it possible tcaccess and manipulate large volumes of naturalresource, environmental, geologic, and other spa-tially referenced data. A 1988 survey32 identified 35

30S.R. Wilson, N. Starr Schneidkraut, and M.D. Cooper, "Use of the Critical Incident Technique To Evaluate the Impact of MEDLINE," contractorreport prepared by the American Institutes for Research for the National Library of Medicine, Betheida, MD, Sept. 30, 1989.

31Ibid., see app. G, "Impact of the Information Obtained From MEDLINE on Medical Decision-Making."

32Federal Interagency Coordinating Committee on, Digital Cartography, Reports Working Group, A Summary of GIS Activities in the Federal

Government, August 1988, pp. 10-12.

16 Helping America Compete: The Role of Federal Scientific and Technical Wonnation

Federal agencies with GIS applications, including,for example: the Agency for International Develop-ment (famine early warning, forestry); U.S. ForestService (forest planning, gypsy moth suppression,fffe behavior modeling); Soil Conservation Service(soil survey database, river basin and watershedplanning, farm and ranch conservation planning);Census Bureau (all 1990 Census activities); NOAA/National Environmental Satellite Data InformationService (atlases showing geographic Jistribution ofice, drought, temperature, precipitation, sunshine,length of day, etc.); Bureau of Reclamation (landclassification, irrigation monitoring, baseline habi-tat); and U.S. Geological Survey (earthquake hazardassessment, national mapping program water qual-ity monitoring). Transportation is another emergingarea of major GIS application, especially withrespect to renewing the U.S. surface and air transpor-tation infrastructure.

GIS-based activities require much greater coordi-nation among Federal, State, local, and inte! aationalgovernment agencies. While Federal agencies col-lect and/or develop a large amount of geographicinformation, State and local governments are amongthe heaviest users and also generate geographicinformation as well. The same is true for foreigngovernments and international intergovernmentalbodies (e.g., the United Nations Environment P '-grain and Food and Agriculture Organization). lnerange of international GIS applications is shownbelow:

1. Preparation of thematic maps, with data on thesocioeconomic, demographic, soil, water, andother characteristics of defined geographicareas.

2. Preparation of base maps, including the plot-ting and revision of quadrangle maps, aeronau-tical charts, marine navigational charts, oceansurveys, and the like.

3. Preparation of terrain maps, including topo-graphic, elevation, slope, relief, and perspec-tive maps, among others.

4. Data display and analysis, including the pres-entation and manipulation of map data and themerging .and integrating of map databases.

5. Environmental assessment and monitoring,including the use of geographic information hipreparing environmental impact assessments

and studies of irrigation, pollution, soil conser-vation, flood potential, and the like.

6. Land and water resource planning and man-agement, including site and road designs,farm, forest, and wetland surveys, habitat andwater quality studies.

7. Mineral resource assessment, including geo-graphic maps, fuel and resource inventories,and geologic hazard analyses.

8. Navigational systems, including air trafficcontrol systems and flight simulators.33

The role of STI and its dissemination variesdepending on the area of science or technology.Historically, the Federal Government has encour-aged the open exchange of Federal STI as afoundation (P.': science and technology. Until re-cently, acass to STI has been restricted only innarrowly defined areas of national security. This hasbeen especially tnie in areas such as medicine andthe environment, where health and safety considera-tions are paramount. But even in these areas, thechanging competitive environment has led to in-creased sensitivity about open, international accessto Federal STI (e.g., with respect to its use incommercialization of biotechnology, medical drugsand devices, or environmental mitigation tech-niques).

Over the last decade or so, intensified interna-tional technical and economic competition has led toadditional restrictions on access to federally spon-sored STI. These restrictions are based primarily onreasons of national security, foreign policy, andinternational competitiveness. Electronic technolo-gies speed the transfer of information on nationaland global scales. Concern over this rapid, uncon-trolled dissemination has fueled a debate overrestrictions on access to STI.

This debate involves the balancing of competinginterests. For example, in the area of export controls,the need to protect against export of militarilysensitive technologies and technical data directly orindirectly to U.S. adversaries must be balanced withthe need to minimize adverse effects on internationalscientific exchange and on international trade oppor-tunities. In domestic technology transfer, the need toencourage the transfer of technology (and relatedtechnical data) from the Federal Government to theprivate sector must be balanced with the need to

331bid.

Chapter 2Federal Scientific and Technical information and the U.S. Competitive Edge 17

minimize restrictions on access to unclassifiedFederal STI and promote a competitive marketplace(even though foreign companies may also benefit).Thus, the short-term interest of a solar energycompany conducting Federal R&D must be weighedin the context of the long-term development needs ofthe U.S. solar energy industry as a whole and theinterests of information vendors and users (e.g.,librarians, entrepreneurs, policy analysts) who thriv eon the open exchange of Federal STI.

In light of the political, military, and economicchanges occurring in Europe and the Soviet Union,perhaps U.S. policies limiting the open flow ofFederal STI should be reevaluated. It may be that thejustifications for the restrictive approaches of the1980s are less valid as the world reaches the lastdecade of the 20th century.34 A key step in restoringthe U.S. competitive edge is to build on the strengthsof the U.S. governmental, academic, and commer-cial information sectors. Federal STI must play animportant part in the overall U.S. competitivenessstrategy.

STI and International Competitiveness:A Summation

On the one hand, the world is becoming muchmore competitive in political, economic, and tech-nological terms. The moves toward political democ-racies and market-f sed economies will open upmany new opportu.aties for U.S. firms tradingoverseas and for U.S. Government leadership on keyinternational issues. However, these same opportu-nities will similarly be available to other nations.Markets and competition rely heavily on science,technology, and innovation.

Evidence shows that STI is very important toscientists, engineers, and managers in technology-intensive government agencies and industries.35Why? Because maintaining an information advan-tage is crucial to achieving a competitive edge. In therapidly changing global marketplace of ideas andproducts, information has become an essential

competitive resourcealong with technology, capi-tal, labor, and management.

The challenge for the United States is how tostrengthen and deploy our own competitive STIassets. The Federal Government supports the largestR&D complex in the world, and generates the largestvolume of STI. The United States has a strongeducaonal and library infrastructure, and the U.S.commercial information industry is foremost in theworld: Also, the United States is highly competitivein assembling the technical infrastructure (e.g.,online and optical disk systems) necessary to deliverinformation products and services, including STI.

Yet the United States does not have an overallstrategy to capitalize on these substantial assets, andto overcome its weaknesses, e.g., the training ofscientists and engineers in STI search-and-retrievalskills, or the consideration of STI user's needs inscience agency planning. To realize the potential ofU.S. leadership in STI will require reaching a strongconsensus on overall Federal STI policy.

The stakes are high, as measured by market sizeand private-sector and governmental activity:

The Western European database services mar-ket is expected to double in the next 5 years, toover $7 billion, with the online portion pro-jected to increase from 60 to 70 percent;The European Economic Community is spon-soring a wide range of pilot projects for theEuropean information services market, includ-ing, for example, a multimedia atlas of theMediterranean regioli on compact optical disk(that combines data, images, digital maps, andgraphics on Mediterranean geography, agricul-ture, environment, and industry);Two private companies (one U.S., one foreign)are planning cooperative sn projects with theU.S.S.R. Academy of Sciences and State Com-mittee for Science and Technology (includingthe establishment of training centers to teachSTI online search skills to Soviet officials andscientists);

34The Federal Govermrent is already reevaluating the need for controls on export of a variety of computer and telecommunications equipment and

systems to the Eastern hloc; proposals to relax export restrictions are being discussed with the Coordinating Committee for Multilateral Export Controls(COCOM). Also see H.R. 4653, the "Export Facilitation Act of 19%, ' 101st Cong., 2d sess., Apr. 26, 1990, ordered to be reported by the House

Committee an Foreign Affairs, May 10, 1990.

35For recent results on ST1 in the aerospace industry, see TE Pinelli, M. Glassman, P..0. Barclay, and W.E. Olin, Technical Communication in

Aeronautics: Results of an Exploratory StudyAn Analysis of Managers' and Nonmanagers' Responses,NASA Tech. Memo. 101625 (Hampton, VA:

NASA Langley Research Center, August 1989), and Technical Communication in Aeronautics: Results of an Exploratory Study An Analysis of Profit

Managers' and Nonprofit Managers' Responses, NASA Tech. Memo. 101626 (Hampton, VA: NASA Langley Research Carter, October 1989). 26


U.S.-origin databases still dominate the infor-mation markets of the industrialized nations,but many (especially the EEC and Japan) haveexplicit strategies to develop their own domes-tic information industries.

Perhaps the single most important step the U.S.Govern:nent can take is to recognize the importantrole of STI in strengthening U.S. competitiveness. Inthe immediate post-World War II years, the com-manding across-the-board lead in science and tech-nology meant that we could directly control thecreation and dissemination of STI and needed to payscant attention to foreign STL Now, U.S. science andtechnology are under competitive pressure in manyareas, and the U.S. lead in STI is no longer secure.Other developed countriessuch as Japan and theEuropean Communityare targeting STI as a keyelement of a national strategy, and seem committedto aggressively develop their own STI capabilities.36The imperative for a reinvigorated U.S. STI strategyis strong. As summed up by Dr. Lewis M. Bran-

Perhaps the single most important stepthe U.S. Government can take is torecognize the important role of STI instrengthening U.S. competitiveness.

scon ). of the J.F. Kennedy School of Government atHarvard University:

Most othcr industrial democracies take informa-tion policy very seriously as an element of sciencepolicy and of strategies for competitiveness. In theU.S., for reasons I have never fully understood,information policy is the stepchild of economicpolicy and has lost its place in science policy. Wespend our efforts discussing what information tokeep, sell, or give away. The better question is howto create it, acquire it, and use it.37

2 7

34See McClure and Hernon, op. cit., footnote 2.

37Statement of Lewis M. Branscomb, Director, Science, Technology, and Public Policy Progran J.F. Kennedy School of Government, Harvqrd

University, before a hearing of the U.S. Senate Committee on the Judiciary, Subcommittee on Tochno ogy and the Law, Mar. 16, 1988. Dr. Branscoulb

formerly served as Chief Scientist of the 1BM Corp., and as Director of the National Bureau of Standards.

Chapter 3

Reaching Consensus on Principles of FederalScientific and Technical Information Dissemination

Many STI managers in the Federal agencies,along with scientists, engineers, scholars, librarians,and vendors who specialize in STI, recognize thehighly leveraged role of Federal STI in renewing theU.S. competitive edge. However, during most of the1980s, sharp debate over several key elements ofFederal information policy and the resulting lack ofconsensus have prevented the STI community fromsending a clear message to top congressional andexecutive branch policymakers. The most contro-versial aspects of STI policy have been:

the Federal role in information dissemination;principles of STI dissemination;policy on the open flow of STI; androle of the governmentwide dissemination agen-cies.

In all these areas, electronic technologies aggravateold issues or create new ones.

During the last year anc, a half, the debate inCongress has advanced to the point where a greaterdegree of consensus and, thus, legislative action ispossible. Unanimous consent may be unlikely onsome iss .ts, but if the potential of STI is to berealized, a working consensus is needed. Thischapter discusses the debate over principles of STIdissemination, including the Federal role. Chapter 4covers the policy debates on the open flow of STI,and on the role of the governmentwide dissemina-tion agencies.

The ongoing information policy debates aredirectly relevant to efforts by the 101st Congress toupdate public laws on Federal information dissemi-nationincluding the Paperwork Reduction Act,

Printing Act, Depository Library Act, and Freedomof Information Act.'

Federal Role in Information Dissemination

STI has been caught up in the philosophicaldebate over the role of the Federal Government indisseminating Federal information to the public. Allsides of the debate agree on the need for someFederal role, but agreement on specifics, especiallywith respect to the relative roles of the governmentand private sector in dissemination, is more elusive.Federal STI is relevant to both the missions of theresearch and development agencies and to govern-mentwide dissemination objectives. In the absenceof a governmentwide strategy or policy for STIdissemination, the development of a comprehensiveinformation dissemination policy under the auspicesof the Office of Management and Budget (OMB) isof greater importance.

OMB and Circular A-130

OMB is the dominant force in shaping FederalSTI dissemination policy.2 Its role was strengthenedby die Paperwork Reduction Act of 1980,3 whichestablished an Office of Information and RegulatoryAffitirs (01RA) within OMB. The Act was amendedin 1986 to explicitly include information dissemina-tion within its scope.4 The Act assigns the OIRADirector broad responsibilities to minimize the costand maximize the usefulness of information col-lected, maintained, and disseminated by the FederalGovernment. Further, the Act requires the OIRADirector to develop and implement Federal informa-tion policies, principles, standards, and guidelineswith respect to information collection and dissemi-nation. The Act also requires each Federal agency to

1For a detailed discussion of how technology has outpaced the law, see U.S. Congress, Office of Technology Assessment, Informing the Nation:Federal Information Dissemination in an Electronic Age, OTA-CIT-396 (Washington, DC: U.S. Government Printing Office, October 1988).

2See, for example, C.R. McClure ,ttki P. Hernon, U.S. Scientific and :rechnical Information Policies: Views and Perspectives (Norwood, NJ: AblexPublishing Corp., 1989); C.R. McClure, P. Hernon. and H. Relyea (ed.), United States Government Information Policies: Views and Perspectives(Norwood, NJ: Ablex Publishing Corp., 1989); statement of Harold B Shill, Associate Professor, West Virginia TJniversity, on behalf of the WestVliginia Library Association and West Virginia Univers'ty Libraries, before a May 23, 1989, hearing of the House Government OperationsSubcommittee on Government Information, Justice, rid Agriculture; statement of Harold B. Shill, on behalf of the American Library Association,Legislative Assembly, before a July 14, 1987, hearing of the House Committee on Science, Space, and Technology, Subcommittee on Science, Researchand Technology.

3Public Law 96-511, Dec. 11, 1980.

4Public Law 99-500, Oct. 18, 1986, and Public Law 99-591, Oct. 30, 1986.

19

20 Helping America Compete: The Role of Federal Scienttfic and Technical Infornwtion

STI has been caught up in the philo-sophical debate over the role of theFederal Government in disseminatingFederal information to the public.

IM=OMIIMIRiMMI

designate a senior official to be responsible foragency compliance with OIRA policies, principles,standards, and guidelines on information collectionand dissemination.5

While the authority of OIRA clearly extends toinfonnation dissemination, Congress did not-atleast in the Paperwork Reduction Act-provideguidance on the shape, direction, or even basicphilosophy of information dissemination policiesthat might be promulgated by OIRA. Part of thereason for this omission is that, at the time thePaperwork Reduction Act was being debated andenacted, other committees were considering legisla-tion on the printing chapters of Title 44 of the U.S.Code (chs. 1-19; the PRA is ch. 35) that would haveaddressed key aspects of information dissemina-tion.6 This parallel legislation was not enacted. AndCongress has not yet provided explicit statutoryguidance to OIRA on information disseminationp, icy, although the 101st Congress is considering avariety of legislative proposals to amend variouschapters of Title 447

OMB's efforts during the 1980s to promulgategovernmentwide information dissemination policyproved to be controversia1.8Much of the controversyfocused on the role of the private sector in informa-tion dissemination and charges to be levied for useof Federal information dissemination. Both the draftand final versions of OMB Circular A-130 on"Management of Federal Information Resources"emphasized that Federal agencies place "maximumfeasible reliance" on the private sector for informa-tion dissemination, and that costs be recoveredthrough user charges where appropriate.9

The fmal December 1985 version of OMB Circu-lar A-130 gave more explicit recognition to theimportance of government information, but stillemphasized the role of the private sector. Thus,Federal agency dissemination must be either "spe-cifically required by law" or "Necessary for theproper performance of agency functions," providedthat the information products and services dissemi-nated "do not duplicate similar products or servicesthat are or would otherwise be provided by othergovernment or private sector organizations."1° Ineffect, in the absence of statutory guidance to thecontrary, OMB applied the philosophy of OMSCircular A-76 regarding contracting out of commer-cially available services in general to informationdissemination in particular."

However, A-76 does not address or define whatdissemination functions are "inherently" govern-mental, that is, are "so intimately related to thepublic interest so as to mandate performance by

5Public Law 96-511 as amended, sec. 3501, 3504, 3506,

8H.R. 5424, the "National Publications Act of 1980," 96th Cong., 2d sess., Sept. 27, 1980.

7See, for example, H.R. 3695, the "Paperwork Reduction and Federal Information Resources Management Act of 1989," 101st Cong., 1st sess,, Nov.

17, 1989; S. 1742, the "Federal Information Resources Management Act of 1989," 101st Cong., 1st sess., Oct. 6, 1989; and H.R. 3849, the "GovernmentPrinting Office Improvement Act of 1990," 101st Cong., 24 sess., Jan. 23, 1990. For related discussion, see OTA comments on S. 1742 prepared for

a Feb. 21-22, 1990, hearing of the Senate Committee on Governmental Affairs, and the statement of Fred B. Wood, OTA, on MR. 3849 before a Mar.

7, 1990, hearing of the Committee on House Administration, Subcommittee on Procurement and Printing.

8See OM, Informing the Nation, op. cit., footnote 1, ch. 11; H.C. Relyea, J. Bormick, and R.C. Ehlke, Management of Federal Information

Resources: A Gerw-al Critique of the March 1985 OMB Draft Circular (Washington, DC: Congressional Research Service, Library of Congress, July

5, 1985); and P. Hernon and C.R. McClure, Federal Information Policies in the 1980s: Conflicts and Issues (Norwood, NJ: Ablex Publishing Corp.,1987). Also see "Librarians Fight Government Plan," New York Times, Feb. 21, 1989, p. Al7; J. Markoff, "Giving Public U.S. Data: Private PurveyorsSay No," New York Times, Mar, 4, 1989, pp. Al, 47; J. Markoff, "Policy Shift on Access to U.S. Data," New York Times, Apr. 10, 1989, pp. D1, D8;D. Sherwood, "Data Wars," Government Executive, April 1989, pp. 24 ff; C. Webb, "Government Databases: Competing With Private Services?"

Presstitne, April 1989, pp. 18-20; TI. McIntosh, "Electronic Age Offers Promises, Problems for Government Information," FWA Daily Report for

Executives, Aug. 11, 1989, pp. C-1 to C-17; and WI. Moore, "Access Denied," National Journal, Jan. 20, 1990, pp. 121-1'

90ffice of Management and Budget, draft, "Management of Federal information Resources," FederalRegister, vol, 50, 1s,.. 51, Mar. 15, 1985, pp.

10734-10747; Office of Management and Budget, Circular A.130, "Management of Federal Information Resources," vol. 50, Dec. 24, 1985, pp.

52730-52751.

loOMB Circular A-130, secs. 9(a) and (b). 2 9"J. Timothy Sprehe, "Developing Federal Information Resources Management Policy: Issues and Impact for Information Managers," Information

Management Review, vol. 2, No. 3, 1987, see pp. 33-41; and OMB Circulars A-76, Aug. 4, 1983, and A-130, Dec. 12, 1985.

Chapter 3Reaching Consensus on Principles of Federal Scientific and Technical Information Dissemination 21

Government emp1oyees,"12 OTA's prior analysis ofNational Technical Information Service (NTIS) andGoverment Printing Office (GPO) privatizationproposals suggested that many NTIS and GPOdissemination functions are nsx suitable for privati-zation. Many other agency information dissemina-tion functions arguably are vital to agency perform-ance of statutory missions. There have been fewcredible analyses of the factors that make contract-ing out of Federal information dissemination cost-effective. Such analyses are difficult."

OMB Circular A-130 has been widely interpretedby agencies as strongly encouraging, if not requir-ing, user charges for information dissemination.However, a careful reading of A-130 indicates thatpricing decisions, unless specifically prescribed bystatute, are left up to the discretion of agency heads,who may set charges no greater than that required torecover the cost of dissemination and who maywaive or eliminate charges if necessary to carry outmission objectives.

STI Agencies and Circular A-130

The net effect of Circular A-130 has been topolarize views on Federal infomiation dissemina-tion policy, divert significant time and resources intodebate over what A-130 is or should be, and createuncertainty or risk aversion among Federal agencieswith respect to dissemination. Federal science agen-cies were not immune from this policy environment.Some STI agencies, notably NTIS and variousagency information clearinghouses and libraries,had to defend their programs against privatizationproposals. In the case of NTIS, OMB's insistence onprivatizationwhich was later overruled by Congressmight have resulted in a 2- or 3-year delay in itsmodernization. Some STI agencies have adopted adefensive, low-profile attitude toward informationdissemination, as a way of coping with the A-130environment.

OMB's privatization policy could have acceler-ated if A-130 went unchanged and Federal agenciesissued their own departmental regulations to imple-

ment A-130. The Department of Commerce is a casein point: it is particularly important because severalQ mmerce agencies have significant STI functions

NTIS, National Oceanic and AtmosphericAdminisiration (NOAA), National Institute of Stan-dards and Technology (MST), and the Patent andTrademark Office (PTO)).

In August 1988, the Department of Commerceissued a draft policy on elecironic informationdissemination.14 Commerce was the fffst and, as yet,4:114 Federal agency to develop a proposed ccmpre-hensive policy. The draft was prepared by a depart-mental task force and was intended to carry out therequirements of the Paperwork Reduction Act andA-130. The draft policy was circulated for commentand revised several times, but was never publishedin the Federal Register and has since been put onindefinite hold, due to the change in administrationand more recently to the subsequent changes inOMB policy direction. Nonetheless, it is useful toreview the original Commerce draft policy as anexample of what might emerge as agency implemen-tation of A-130 if left unaltered.

The basic thrust of the draft Commerce policy wasthat "[o]perating units will use private sector firmsto develop, manage, and operate electronic dissemi-nation activities to the maximum extent possible,"and that, "before initiating electronic informationdissemination, operating units will conduct a priva-tization analysis." The proposed policy placed theburden of proof on the agency to "justify anyproposed direct Federal role in disseminating elec-tronic information in terms of overriding publicneed, law, and/or program mission." The directivewas particularly burdensome with respect to thedevelopment and dissemination of value-added elec-tronic information products and services, and in themarketing and dist.ribution of agency information,all functions which the Department felt should becarried out primarily by the private sector. TheDepartment, in its own "highlights" sheet, notedthat, as a standard of performance, Commerce's

120MB Circular A-76.

BOTA, Informing the Nation, op. cit., footnote 1, ch. 12. Also c F.B. Wood, "Propos& for Privatization of the National Technical InformationService: A Viewpoint," Government Publications Review. vol. 15. 1988, pp. 403-409.

14U.S. Department of Commerce, Draft Department Administrative Order on "Electronic Information Dissemination," Aug. 5, 1988, published inpart as "Draft Policy of the U.S. Department of Commerce on the Dissemination of Information in Electronic Format," Government Information

Quarterly, vol. 6, No. 1, 1989, pp. 89-96. 3 0


Federal information dissemination pol-icy development appears to be movingtoward a compromise on two of the mostcontentious issues: the roles of the gov-ernment and the private sector; and theapplication of user charges.

electionic dissemination activities should "[o]fferno value-added features." Likewise, the draft policyplaced the burden of proof on the agency to justifywhy fees to recover the actual costs of disseminationshould not be applied.

Overall, the proposed policy placed so manysubstantive and procedural hurdles in the path ofagency electronic dissemination activities that inno-vation and creativity could have been seriouslyimpaired. Even though the policy stipulated proce-dures by which agency components could havejustified government .electronic dissemination and/or fee waivers, the procedural burden was highenough to discourage agency initiatives.

An Emerging Consensus?

In Informing the Nation, OTA reviewed a largenumber of agency-specific and governmentwidestatutes with regard to congressional intent oninformation dissemination. While the PaperworkReduction Act itself provides little direct guidance,talmg as a whole the body of public law, OTAconcluded that congressional intent is clear:

In general, unimpeded dissemination of andaccess to Federal information is encouraged orfrequently required and is vital to performance ofagency and programmatic missions established bystatute as well as to the principles of open govern-ment and a democratic society.15

OTA suggested that Congress consider making arenewed commitment to the overriding principle ofpublic access established by Congress in otherstatutes, but updated to reflect the increasinglyelectronic nature of Federal information. In particu-lar, OTA suggested that Congress consider enactinga congressional version of the information dissemi-nation principles addressed in OMB's CircularA-130.16

Since publication of Informing the Nation, anumber of other key reports, OMB draft policies,and, recently, congressional testimony and billshave been issued.17 Federal information dissemina-tion policy development appears to be movingtoward a compromise on two of the most contentiousissues: the roles of the government and the privatesector; kuid the application of user charges.

The shift in OMB thinking is illustrative. InJanuary '989, OMB issued an "Advance Notice ofFather Policy" to revise A-130 that was interpretedas favoring private sector over government dissemi-nation of Federal information, limiting agencydissemination to basic and not value-added elec-tronic information products, and requiring user feesto recover the costs of dissemination, absent compel-

150TA, Informing the Nation, op. cit., footnote 1, p. 259.

16Ibid., p. 260.

17See, for example, 1.1. Berman, "The Right to Know: Public Access to Electronic Information," paper prepared for the Markle Foundation, in P.R.

Newberg (ed.), New Directions in Telecommunications Policy, vol. 2, Information Policy and Economic Policy (Durham, NC: Duke University Press,

1989); statements of J.J. Berman, Director, Information Technologies Project, American CivilLiberties Union, before an Apr. 18, 1989, hearing of the

House Committee on Government Operations, Subcommittee on Government Information, Justice,and Agriculture, and a Feb. 22, 1990, hearing of the

Senate Committee on Governmental Affairs; G. Bass and D. Pincher,Strengthening Federal Information Policy: Opportunities and Realities atOMB,

Benton Foundation, Project on Communications and Information Policy Options (Washington, ')C: The Benton Foundation, 1989); statement of DavidPincher, Staff Attorney, OMB Watch, before a May 24, 1989, hearing of the House Committee on Administration, Subcommittee on Procurement and

Printing; statements of Nancy Kranich, Director of Public and Administrative Services, New York University Libraries, on behalf of the American

Library Association, and D. Kaye Gapen, Dean of Libraries, University of Wisconsin, on behalfof the Association of Research Libraries, before a May

23, 1989, hearing of the House Committee on Government Government Operations, Subcommittee on Government Information, Justice, and Agriculture;

statements of Alan F. Westin, President, Reference Point Foundation, and Professor of Public Law and Government, Columbia University, and Kenneth

B. Allen, Senior Vice President for Government Relations, Information Industry AsAciation, before an Apr. 18, 1989, hearing of the House Committee

on Goveminent Operations, Subcommittee on Government Information, Justice, and Agriculture; H.H. Perrin, Jr., Electronic Acquisition and Release

of Federal Agency Information,Report to the Administrative Conference of the United States, Oct. 1, 1988; statement of Henry H. Perritt before a July

11, 1989, hearing of the House Committee on Government Operations, Subcommittee on Government Information, Justice, and Agriculture; and

Administrative Conference of the United States, Recommendation 88-10 on "Federal Agency Use of Computers in Acquiring and Releasing

Information," adopted Dec. 8-9, 1988.

Chapter 3-Reaching Consensus on Principles of Federal Scientific and Technical Information Dissemination 23

ling reasons to the contrary.18 The public commenton the January OMB notice was overwhelminglycriticaL 19 OMB concluded that the January daft didnot accurately communicate OMB's policy viewsand had further confused and polarized the debate.As a consequence, on June 15, 1989, GMB issued a"Second Advance Notice of Further Policy Devel-opment on Dissemination of Information" thatformally withdrew the January 4 notice, summarizedthe comments received, and presented OMB'sreactions and preliminary conclusions.20 On June16, OIRA Administrator Jay Plager announced thewithdrawal in testimony before the Subcommitteeon Government Information and Regulation of theSenate Committee on Governmental Affairs,21

The June 15, 1989, OMB notice deserves carefulscrutiny by the STI community, because OMBintends to prepare a new d-it policy consistent withthe discussion in the June 15 notice and with anyrelevant legislation that may be enacted. (The newdraft will also incorporate information collection,based on a 1987 draft and comments receivedthereon.22) If history is any guide, the penultimateOMB policy can be expected to have a significantimpact on Federal STI dissemination.

The essence and significance of the June OMBnotice is captured in the following quotation:23

OMB wishes to make clear that its fundamentalphilosophy is that government information is apublic asset; that is, with the exception of nationalsecurity matters and such other areas as may be

prescribed by law, it is the obligation of thegovermnent to make such information readily avail-able to the public on equal terms to all citizens; thatto the extent the flow of information from thegovernment to the public can be enhanced by theparticipation of the private sector, such participationshould be encouraged; and that participation by theprivate sector supplements but does not replace theobligations of government. Thcse principles applywhatever the form, printed, electronic, or other inwhich the information has been collected or stored.OMB did not intend that either OMB Circular A-130or the January 1989 notice should have the effect ofdissuading agencies from canying out activities theybelieve are necessary for the proper performance ofagency functions . .. or that Federal agencies or thepublic should be made to rely primarily on theprivate sector for the dissemination of governmentinformation.

Principles of STI Dissemination

Converging views on the Federal role in informa-tion dissemination has made legislative actionpossible. Various congressional committees aredeveloping legislative proposals to provide OMBand Federal agencies with cpecific statutory guid-ance on information dissemination.24 Legislationand related OMB policy can be expected to have asignificant impact on Federal STI dissemination.

The STI community needs to monitor, carefullyreview, and participate in the development of theseinitiatives to ensure that governmentwide dissemi-nation principles are consistent with those appropri-

18Office of Management and Budget, "Advance Notice of Further Policy Development on Dissemination of Information," Federal Register, vol. 54,No. 2, Jan. 4, 1989, pp. 214-220,

oSee summary of comments in Office of Management and Budget, "Second Advance Notice of Further Policy Development on Dissemination ofInformation." Federal Register, vol. 54, No. 114, June 15, 1989, pp. 25554-25559.

20Ibid.; also see J. Markoff, "O.M.B. Proposes Switch in Information Policy," New York Times, June 10, 1989, p. A-28; and U.S. Office ofManagement and Budget, "Siirnmary of Comments on OMB's Second Advance Notice of 'Anther Policy Development on Dissemination ofInformation," Oct. 19, 1989.

BTestimony of Jay Plager, Administrator, OMB Office of Information and Regulatory Affairs, before a June 16, 1989, hearing of the SenateCommittee on Governmental Affairs, Subcommittee on Government Information and Regulation. Also see testimony of Jay Plager before a June 28,1989, hearing of the House Committee on Administration, Subcommittee on Procurement and Printing; see U.S. Congress, House, Committee on HouseAdministration, Subcommittee on Procurement and Printing, Title 44 U.S.C-Review, Hearings, May 23, 24 and June 28, 29, 1989, 101st Cong., 1stsess. (Washington, DC: U.S. Government Printing Office, 1989), pp. 152-159.

22See Office of Management and Budget, "Policy Guidance on Electronic Collection of Information," Federal Register, vol. 52, No. 152, Aug. 7,1987, pp. 29454-29457; and OMB, "Summary of Comments on Policy Guidance on Collection of Information," Nov. 17, 1987.

BOMB, "Second Advance Notice," op. cit., footnote 19, p. 25557.

24See H.R. 3695 and S. 1742, op. cit., footnote 7. 9 9


ate for STI, and, if not, to make sure that separateguidance is provided for STI.25

Strengthening Public Dissemination ofValue-Added Federal STI

Most agree on the need for public disseminationof STI, but there are differences on how this shouldbe achieved. One serious complication for STIoccurs when unclassified information is deemed tobe sensitive for reasons of national security, foreignpolicy, or competitiveness. In these cases, the goalof public access may conflict with other policyobjectives. Policy on the open flow of STI is treatedas a separate issue area and discussed in chapter 4.The Federal science agencies emphasize that theprimary puipose nf Federal STI is to support agencyR&D missions, that public dissemination is animportant but secondary objective.

A further complication occurs when value-addedFederal STI is involved. Some government andinformation industry officials have argued thatFederal agency electronic dissemination of raw datawas acceptable, but government disseminationvalue-added information was not an appropriategovernmental function and should be the province ofprivate industry.26 In this view, dissemination by theU.S. Geological Sum, y of STI on magnetic com-puter tape would have been appropriate, but USGSdissemination of value-added or enhanced infomm-tion would note.g., a compact optical disk withdata on earthquake monitoring that also included thesearch software for retrieving and manipulating thedata.

Value-added is not the best determinant to distin-guish between government and private-sector roles.

Many Federal science (and other) agencies havelegislative responsibilities to develop and dissemi-nate value-added information, and have been doingso for decades. Restricting the Federal agencies fromproviding value-added information, or from provid-ing information available on paper in electronicform, would prevent some Federal agencies frommeeting statutory obligations. Value-added restric-tions could prevent agencies from providing thebenefits of electronic technologies through auto-mated data services to taxpayers who collectivelypaid for the development of the information in thefirst place.

Federal agencies should be able to providevalue-added information that furthers agency mis-sions, but they should carefully consider private-sector capabilities, so that contracting out andmarketplace alternatives are utilized when appropri-ate. Private information vendors (commercial andnot-for-profit), on the other hand, should be encour-aged to repackage and resell Federal information(that is not classified or otherwise restricted), and toadd further value to create enhanced informationproducts and services where the market exists.Whether government or private dissemination ispreferred, however, should not be ba sed on ideology,but on which mode(s) can best serve national needs.

Improving Cost-Effectiveness andDiversity of Federal STI

OMB has long supported agency automationprograms in the belief that automation will becost-effective in ths long tenm The judicious use ofelectronic technologies could lead to more timely,complete, and accurate Federal information dissemi-

25For historical perspective on the development of information dissemination principles, see, for example, U.S. Executive Office of foe President,Domestic Council, National Information Policy, report to the President of the United States (Washington, DC: Natonal Commission on Libraries andInformation Science, 1976); J.H. Yurow, R. 7. Aldrich, 1LR, Eclair, Y.M. Braunstein, D.Y. Peyton, S. Pogrow, L.S. Robertson, and A.B, Wildavsky,Issues in Information Policy, report prepared for tbe National Telecommunications and Information Administration, U.S. Department of Commerce,NI1A-SP-80-9 (Washington, DC: U.S. Government Printing Office, February 1981); U.S. Congress, House, Committee on Government Operations,Subcommittee on Government Information and Individual Rights, Government Provision of Information Services in Competition With the PrivateSector, Hearing, 97th Cong., 2d sass. (Washington, DC: U.S. Government Printing Office, Feb. 25, 1982); Rep. Glenn English, "Electronic Filing ofDocuments With the Government: New Technology Presents New Problems," Congressional Record-House, Mar. 14, 1984, H1614-1615; U.S.Congress, House, Committee on Government Operations, Subcommittee on Government Information, Justice, and Agriculture, Electronic Collectionand Dissemination of Information by Federal Agencies, Hearings, Apr. 29, June 26, and Oct. 18, 99th Cong., 1st sess. (Washington, DC: U.S.Government Printing Office, 1986); U,S. Cc agress, HOUR, Committee on Government Operations, Subcommittee on Government Information, Justice,and Agriculture, Electronic Collection Ind Dissemination of Information by Federal Agencies: A Policy Ove7view, House Report 99-560, 99th Congress,2d sess. (Washington, DC; U.S. Government Printing Office, Apr. 29, 1986); U.S. Congress, Hoese, HR. 2600, Securities and Exchange CommissionAuthorization Act of 1987, 100th Cong., 1st sess., June 4, 1987; U.S. Congress, House, Committee on Energy and Commerce, Securities and ExchangeCommission Authorization Act, report to accompany HR. 2600, 100tb Cong., 1st sess Rep. No. 100-296 (Washington, DC: U.S. Government PrintingOffice, Sept. 9, 1987); and U.S. Congress, House, Committee on Government Operations, Federal Information Dissemination Policies and Practices,Vlearings, Apr. 18, May 23, and July 11, 1989, 101st Cong., 1st sess. (Washington, DC: U.S. Government Printing Office, 1989).

26This view was reflect-A in the U.S. Departmet r of Commerce, Draft Administrative OrcIti on "Electronic Information Diner dilation," Aug. 5,1988, and the Office of Management and Budget ' Advance Notice of Further Policy Development on Dissemination of Information," Jan. 4, 1989.

0 14


Federal agencies should be able to pro-vide value-added information that fur-thers agency .missions, but they shouldcarefully consider private-sector capa-bilities, so that contracting out andmarketplace alternatives are utilized whenappropriate.

illININIIIIIIM11111111111111111111111

nation. However, the 1980s offer several examplesof agency electronic dissemination projects thatwent astray or suffered serious and sometimes costlyproblems. In part, this is the price of innovation andprogressand neither is private sector R&D im-mune from "wrong tracks," "blind alleys," and"learning the hard way." Nonetheless, this pointsup the need for better ways for Federal agencies toshare learning among themselves and the privatesector. The most cost-effective route may sometimesbe primarily an agency initiative, at other times deferentirely to the private sector, or develop collabora-tively by the agency and a private firm. There isroom for more creative approaches in optimizingFederal investment in information dissemination.While cost-effectiveness is an important criterion, itmust be balanced with the principal goals offulfilling the statutory R&D requirements of thescience agencies and promoting public dissemina-tion of STL

It is important to maintain and broaden theavenues used for dissemination of Federal informa-tion, includi74 STI. For 5T1, these avenues are:

the Federal science agencies themselves;the governmentwide dissemination agenciessuch as NTIS and GPO;the press (including print and electronic mediaand a wide r-ige of specialized scientific andtechnical journals and newsletters);commercial information vendors (ranging fromsmall companies that specialize in a few areasof STI, to very large corporaii6ls with entiredivisions devoted to STI publishing, data5ases,etc.);not-for-profit information vendors (includinguniversity and foundation-based providers);

researchers and scholars who collect, analyze,and synthesize Federal STI and disseminate theresults through multiple channels (rangingfrom conference presentations, to congres-sional testimony, to technical reports);professional, consumer, and trade associationsthat specialize in areas relevant to STI (andprocess and redisseminate STI to their ownconstituencies);the library community (including public, pri-vate, special, academic, research, and schoollibraries throughout the Nation);State and local governments and associations;andforeign countries and companies that use Fed-eral STI for policy or commercial purposes.

Involving Users and Providers in STI Planning

Planning for Federal information (including STI)dissemination should provide opportunities for theusers and the public te participate in the process, aswell as the appropriate agencies. Inadequate in-volvement of the potential users has led to pastfailures in new information services. User involve-ment is especially important for STI, because usergroups are often highly specialized and sophisti-c ated.27

Some agency officials are concerned that publicparticipation in STI planning could become cumber-some and slow down or discourage agency innova-tion. On one hand, the use of public funds forinformation systems to carry out public purposessuggests the need for an open process. On the otherhand, procedural red tape could chill agency innova-tion, as it sometimes has in the private sector. Thekey is to match the procedural requirements to thepurpose, nature, and scale of the project. Forexample, multi-million dollar systems like theNational Aeronautics and Space Administration'sEOS (Earth Observing System) or the Securities andExchange Commission's EDGAR (Electronic DataGathering and Retrieval) may be required to followrigorous public notice and participation procedures.At the other extreme, small pilot or demonstrationprojects may be required to include public notice butnot to use a formal comment period, meetings, andapproval procedures that may be needed for largeoperational projects.

27See statements of Charles R. McClure, Syracuse University, Red B. Wo, OTA; and Joseph G. Coyne, U.S. Department of Energy before a hearing

of the House Committee on Scieoce, Space, and Technology, Subcommitee on Science, Research, and Technology, Oct. 12, 1989.

26 Helping America Compete: The Role of Federal Scientific and Technical It:formation

ININIIIIIIP

Most user and provider groups supportthe alternative concept of marginal costrecoverymeaning that user chargesfor Federal information disseminationwould not exceed the marginal cost ofdissemination and would not includecosts of collecting or creating the infor-mation.

Determining User Charges for Federal STI

User charges continue to be controversial. Someat OMB have advocated full cost recovery forFederal information dissemination. Under this pol-icy, user charges for Federal information could havebeen set to recover the entire costs of collecting,processing, and maintaining as well as disseminat-ing. This proposal was opposed by both user andprovider groups, on the grounds that much Federalinformationincluding STI---would be priced outof reach, and that the taxpayer would effectively beasked to pay twice.28

Experience with Landsat STI suggests that theacademic research community is particularly bur-dened by full cost pricing. Responsibility for pricingof Landsat imagery and digital data has moved in thepast from the U.S. Geological Survey (USGS) to theNational Oceanic and Atmospheric Administrationand fmally to EOSAT, a commercial companyestablished under the Land Remote-Sensing Com-mercialization Act of 1984.29

During the 1980s, Landsat STI prices have beenincreased to recover a greater portion of full costs tothe point where 1989 EOSAT prices are about nine

times higher than 1980 prices for imagery and threetimes higher for digital data." This has reduced salesto academia by more than half.31

Transition from manual imagery interpretation todigital data analysis explains part of the reduction inimagery sales, but examination of worldwide Land-sat sales for 1981-88 shows that users are payingmuch more for much less. For example, between1981 and 1988, the volume of data digital salesincreased by only 10 percena, while the revenue fromdigital data sales increased by about 600 percent.During this same period, the volume of imagerysales decreased by about six times, while thecorresponding revenues increased by 10 percent.32Full cost prices are affordable by some largegovernment agencies and private corporations (U.S.and foreign), but these prices have squeezed researchactivities performed by academia, State/local gov-ernments, small business, individuals, and someFederal agency programs that are faced with tightbudgets (including, ironically, some USGS pro-grams).33

Most user and provider groups support the alter-native concept of marginal cost recoverymeaningthat user charges for Federal information dissemina-tion would not exceed the marginal cost of dissemi-nation and would not include costs of collecting orcreating the information. The definition of "mar-ginal cost" is ambiguous. Three definitions havebeen suggested:

1. Marginal cost is the incremental cost ofproducing the n+1 unit of a specific informa-tion product or service Thus, the cost per copyof a printed report would be the direct cost ofproducing one more paper copy; the cost of adatabase would be the direct cost of one more

281%11 COSt recovery has also been opposed on legal grounds. The courts have ruled that, under the User Fee Act of 1952 (31 U.S.C. 9701), user fee

chargcu by Federal agencies must be reasonably related to the direct and indirect costs of providing a product or service. For relevant decisions, see 846

F.2d 765 (D.C. Cir. 1988); 777 F.2d 722 (D.C. Cir. 1985); 554 F.2d 1109 (D.C. Cir. 1976); and 554 F.2d 1094 (D.C. Cir. 1976).

U.S. Congress, Public Law 98-365, July 17, 1984. For general discussion of Landsat commercialization, see U.S. Congress, Office of Technology

Assessment, Remote Sensing and the Private Sector: Issues for Discussion, OTA-TM ISC-20 (Washington, DC: U.S. Government Printing Office,

March 19E4); U.S. National Commission on Libraries and Information Sci4nce, Information Policy Implications of Archiving Satellite Data: To Preserve

the Sense of Earth From Space (Washington, DC: 1984); and National Research Council, Space Applications Board, Remote Sensing of the Earth From

Space: A Program in Crisis (Washington, DC: National Academ ir Press, 1985).

30G. Metz, "Landsat Product Price Examples, 1980-1989," EROS Data Center, Sioux Falls, ND.

31The peirent of total EROS sales to academia declined from 10% in FY1979 to 5% in FY1988 for Landsat image.y, and from 14% in FY1979 to

5% in FY1988 for Landsat data. In FY1988, only 408 frames of imagery and 379 data items were sold to academic users.

320. Austin, R. Pohl, and G. Metz, A Summary of Worldwide Landsat Sales: 1988 (Sioux Falls, SD; EROS Data Center, U.S. Geological Survey, May

30, 1989).33Also see G. Austin, Annual Report of Landsat Sales for Fiscal Year 1988 (Sioux Falls, SD: EROS Data Center, U.S. Geological Survey, 1989).

Chapter 3Reaching Consensus en Principles of Federal Scientific and Technical Information Dissemination 27

electronic copy (e.g., on raagnetic tape) or onemore hour of online access time.

2. Marginal cost is the average cost of producinga specific product or service. Here, the cost percopy of a printed report would be the totalcosts of producing n copies divided by thenumber of copies. The cost of a databasewould be the total costs of providing thedatabase divided by the number of serviceunits (e.g., magnetic tape copies, hours ofaccess time).

3. Marginal cost is the average cost of a group ofproducts or servic.ls (i.e., a product line). Thus,the cost per copy of a printed report would bethe total costs of producing n1+n2+; copies ofm1+m2+my reports divided by the total numberof copies. The cost of a database would be thetotal costs of providing n databases divided bythe total number of service units for alldatabases combined.

Definition 1 is the true economic marginal cost.But if the intent is to recover the cost of dissemina-tion but not the costs of collecting or creating theinformation, then definitions 2 and 3 could apply.True marginal pricing reflects only direct variablecosts (e.g., labor and materials used in printing),whereas average costs typically also cover directfixed costs (e.g., production line supervision, elec-tricity) and some share of indirect costs (e.g.,building rent, marketing, general management, capi-tal investment).

A major policy question is whether the priceformula should apply to an individual informationproduct or service or to a line of products andservices; what costs elements should be included(variable, fixed, direct, indirect); and how muchflexibility agencies should have in pricing. Thedemand for Federal information varies widely, andper-unit costs for the high sales volume items will berelatively low, while per-unit costs for the low-volume items will be relatively high.

For example, the user charge for a compact opticaldisk could vary from $5 to $500 depending on thepricing formula and volume of demand. At a salesvolume of 500 copies, the true marginal cost(defmition 1) would be about $5 per copy and theaverage cost (definition 2) typically $50 to $100 percopy. At a sales volume of 50 copies, the marginalcost might increase to $10 to $20 per copy and theaverage cost to $500 per copy. Thus true marginal

cost yields the lowest price, but leaves much of thecost of dissemination uncovered. The uncoveredcosts would have to be paid from appropriated funds.The average cost formula covers the cost of dissem-ination, but is very sensitive to total volume. Forhigh-volume items, average cost is low and viceversa.

As an illustration, NOAA appears to use defmi-tion 2 above, the average cost of producing a specificproduct or service, as the basis for pricing. NOAAincludes both direct and indirect costs in its calcula-tions. Typical direct costs are labor, supplies,printing, and computer resources. Indirect costscover a portion of NOAA and U.S. Department ofCommerce overhead and rent. The costs of collect-ing or creating the data are not included. NOAAcalculates the total direct and indirect cost ofproducing each product or service, and divides thetotal cost by the quantity produced to determine aper-unit cost. Assuming that estimated demandmeets or exceeds the quantity produced, the unitprice is usually set to equal the unit cost.

The cost breakdown for several NOAA NationalGeophysical Data Center CD-ROM products isshown in table 1. The Geophysics of North AmericaCD-ROM was relatively expensive to produce, butthe unit cost was kept down due to the higherestimated sales volume and quantity produced. TheGloria CD-ROM was a pilot project subsidized byUSGSthus the low price. And the Deep SeaDrilling CD-ROM was inexpensive to produce, witha low unit price even with modest estimamd demand.

Many STI items have low total sales, and thus theprice could be prohibitive if calculated on anaverage-cost-per-product basis (definition 2). Low-demand STI items might be best suited for either truemarginal cost pricing (with the rest of the costscovered out of appropriated funds) or average-cost-per-product-line pricing (definition 3). Both theNational Teclmical Information Service (NTIS) andNational Library of Medicine (NLM) use product-line pricing, which in effect results in a cross-subsidy between the high-demand and low-demanditems. NTIS and NLM believe that use of truemarginal cost pricing (defmition 1) would threatentheir viability, unless appropriated funds were pro-vided to cover the rest of the costs; and that use ofavera; cost pricing per product (definition 2) wouldfurther curtail demand for many of the lower volumeinformation products and services, since prices for


Table 1Cost Breakdown for Illustrative National Geophysical Data CenterCD-ROM Products

cost elements

QuantityUnitcostProduct

Directlabor

Otherdirect

Indirectcosts

Geophysics of North America: $66K $58K $40K 700 $235

CD-ROMDocumentation 231< 8K 14K 700 65

Software 53K 44K 32K 700 185

Gloria Side Scan Sonar:CD-ROM 6Ka 3Ka 200 45

Deep Sea Drilling ProjectCD-ROM 61< 9K 3K 200 90

aUSGS paid for the development of this pilot CD-ROM project.

SOURCE: National Geophysical Data Center, 1989.

these items would likely rise out of reach of manyusers.

NTIS must operate its clearinghouse on a break-even basis with no appropriated funds. NTIS usesrevenues from brokerage fees and services to otheragencies, along with product-line pricing, to helpoffset the losses that would otherwise occur due tothe many NTIS documents that register no or verylow sales volume. The sale or lease of Federal STIin electronic formats is now the fastest growingmarket segment for NTIS, increasing at 5 to 10percent annually. "Electronic" sales account forabout one-quarter of total NTIS revenues.34

NTIS sales of paper or microfiche documentscontinue to decline, with annual sales of indices,newsletters, published searches, and technical docu-ments reaching all-time lows in fiscal year 1989. Theaverage total demand for NTIS documents is about10 copies over the life of a document, and one-quarter to one-third of the documents never sell acopy.35

Financing the NLM dissemination program ismore complicated since (unlike NTIS) NLM doesreceive appropriated funds for creation of databases.This means that NLM must determine where tax-supported information collection or creation endsand user-financed information dissemination be-

gins. According to NLM, online database prices areset to recover only the cost of dissemination (exceptfor foreign users, who pay full cost since theypresumably have not paid taxes). NLM uses averagecost product-line pricing, which means that userspay the same average price for all databases ($271hour during peak periods). The most heavily useddatabase (MEDLINE) absorbs much of the overheadcosts and helps keep prices down for the otherdatabases (e.g., TOXLINE, AIDSLINE, CANCER-LIT).36 However, vendors are concerned that NLMcombines both offline products (e.g., magnetictapes) and online services when estimating costs,and cross-subsidizes not only from MEDLINE toother databases, but from offline products to onlineservices through the use of royalty fees.

The NLM example raises several pricing ques-tions. How should agencies distinguish amongcollection, creation, maintenance, and disseminationcosts? What costs should be included in determiningaverage or marginal cost? What products and/orservices should be included in determining costs?Under what circumstances should products andservices be kept separate or combined, for pricingpuzposes? Should agencies cross-subsidize differentproducts and services, and if so, to what degree?How should product lines be defined, when calculat-ing average costs over a range of products and/or

34FY1989NTIS sales of software and data sets from the NTIS inventory were $2.59 million, leasing of NTIS and other agency databases $2.35 million,

and brokerage of other agency electronic items $1.41 million for a combined "electronic" sales of $6.35 millionabout one-fourth of the $24.4 million

total revenues,35For further discussion, see OTA. Informing the Nation, op. cit,, footnote 1, chs. 5 and 12; and testimony of Fred B. Wood of OTA before a Feb.

24, 1988, hearing on NITS privatization and a Mar. 8, 1990, hearing on NTIS modernization held by the House Committee on Science, Space, and

Technology, Subcommittee on Science, Research, and Technology.

36Chairman, Board of Regents, National Library of Medicine, memorandum to the Assistant Secretary for Health, U.S. Department of Health and

Human Services, "Response to Systems Review Board Recommendations on the Pricing of NLM Products and Services," May 29, 1984; and K.A.

Smith, "Government Databases: The NLM Philosophy," Database, vol. 11, No. 3, 1988, P. 58. 37


services? How should agencies set prices for dissem-ination to special user groups such as foreign userswho do not pay taxes or not-for-profit users whocannot afford even the average cost? Should agen-cies be able to retain sales revenues to help offsetdissemination costs? (NLM retains about one-half or$6 million/year in an NTIS deposit account, andreturns the rest to the U.S. Treasury.)

Whatever one's views on pricing formulae, thereis a general consensus that user charges should notexceed the cost of dissemination, and that agenciesshould be able to reduce or waive user charges ifneeded to carry out agency missions. Should thispricing philosophy Oecome governmentwide policy,reconciliation of other statutes might be necessary.For example, Title 44 of the U.S. Code requires thatthe Superintendent of Documents (SupDocs) setprices for government publications at cost plus 50percent." However, as a practical matter, in recentyears the House and Senate Appropriations Commit-tees have transferred net revenues from the SupDocssales program to support the Depository LibraryProgram (and thereby correspondingly reduce theneed for DLP appropriated funds). Also, in 1988,Congress authorized the National Oceanic andAtmospheric Administration to assess fees based on"fair market value" for commercial users of certainNOAA information products and services (govern-mental, university, and not-for-profit users wouldpay only marginal costs).38 NOAA officials havefound it difficult to determine fair market value, andboth agency and industry officials question whetherthis is a viable basis for setting user charges.

Defining Intellectual Property Rightsin Federal STI

STI developed by or for the Federal Government,like other types of Federal information, by law maynot be copyrighted. Some researchers and vend;irsinclude Federal information in scholarly works orcommercial products that are copyrighted (e.g., a

vendor who copyrights a new compact optical diskthat includes bibliographic STI from multiple sources,one of which is the Federal Government).

The major issue concerus the use of so-called"copyright-like" devices by Federal agencies. Sev-eral science agencies use licensing agreements intheir dissemination programs. NLM makes its onlinedatabase MEDLINE available to both online andcompact optical disk vendors, through a licensingagreement that levies charges estimated to equal theaverage per-unit cost for a user of the NLM line ofdatabases.

NTIS similarly licenses its bibliogral 'hie databaseto private vendors through a licensing agreementand also serves as licensing agent for other agencies'databases.

The National Agricultural Library (NAL) distrib-utes its AGRICOLA bibliographic database tovendors via NTIS. NTIS charges vendors $2,000/year for the yearly AGRICOLA data on magnetictape, and $200/year for back files. NTIS retains allof this revenue. NTIS also charges online vendors $6per AGRICOLA connect hour and $0.05/"hit" (abibliographic citation on the desired subject), andCD-ROM vendors a fee equal to 25 percent of thedisk sales price. These online and CD-ROM userfees are split 20 percent to NTIS and 80 percent toNAL.39

Some private vendors view such licensing ar-rangements as restrictive and illegal.° These ven-dors believe that agency licensing agreements dis-courage competition among commercial servicesand/or inhibit demand, and have the effect ofrestricting access to Federal STI. Other vendors fmdlicensing agreements acceptable so long as they arenonexclusive and fairly priced. The NLM and NTIS

, licensing agreements appear to be nondiscriminl-tory in that any vendor can be licensed, and the feesare set to recover the cost of databases and related

3744 U.S.C. 1708.

S. 2209, Tide IV, sec. 409.

39Gary K. McCone, National Agricultural Library, U.S. Department of Agriculture, letter to Fred B. Wood, Office of Technology Assessment, U.S.

Congress, Dec. 28, 1989.40For discussion of concerns about the NLM MEDLINE database, see U.S. Congress, House, Committee onGovernment Operations, Electronic

Collection and Dissemination, Oct. 18, 1985, Hearings, and Apr. 29, 1986, Report, op. cit., footnote 25; and statement of P. James Tenagno, President,

Maxwell Online, Inc. before a July 11, 1989, bearing of the House Committee on Government Operations, Subcommitee on Government Information,

Justice, and Agriculture. For the NLM view, see NLM, "Comments on the Twenty-Eighth Report by the Committee on Government ()mations," June

5, 1986; and "NLM Policy on Database Pricing," December 1989.

38

30 Helping America Compete: The Role of Federal Scientific and Technical Wormation

operations.4i Nonetheless, snme vendors questionwhether the online connect hour charges accuratelyreflect identifiable government costs that vary as afunction of the level of use or number of subscribers.

Some users and vendors believe that it would bebetter for agencies to provide information free ofcharge or charge only the true marginal cost. This,according to these vendors, would reduce or elimi-nate cost as a barrier to access, and presumablyeliminate concerns about fees in licensing agree-ments. NASA's National Space Science Data Center(NSSDC) operates largely without fees. NSSDCdisseminates computer tapes at no charge (if thetapes are returned after copying) and allows limitedaccess to online databases, also at no charge, forscientific or educational use by:

NASA installations;NASA contractors and grantees;other Federal agencies, contractors, or grantees;colleges and universities;State or local governments; andnot-for-profit organizations.

NSSDC charges other users the marginal cost of $45per magnetic computer tape (or $25 if the usersupplies the tape) and direct processing costs forlarger amounts of online database use. Online userspay their telecommunication charges, whether ac-cess is direct or over networks. NSSDC does not yethave a policy for high-density storage media such asCD-ROM.

NAL's AGRICOLA currently recovers about$60,000 per year of online revenues. At NLM, NTIS,and other agency data centers (e.g., NOAA' s NGDC),revenues based on average cost pricing (and licens-ing agreements) comprise a much larger part of theiroperating budgets. The impact of changes in pricingpolicy would vary widely among agencies. Detailedfinancial analyses would be required to estimaterevenue shortfallsand the necessary compensatingappropriation increasesunder various pricing andlicensing scenarios. And even if price was not anissue, some kind of agreements could be needed tomaintain quality control and protect the integrity ofagency databases. An agency has a valid interest in

assuring that quality standards are met, a statedpurpose of NLM's licensing agreements (along withcost recovery).

The de facto copyright of Federal informationthrough the transfer of patent rights or rights intechnical data from the Federal Government tocontractors, employees, or pri-ate parties (e.g.,companies, universities) presents another problem.Both Congress and the President have encouragedcloser collaboration between the government andprivate sector to facilitate the commercialization oftechnology developed by or for the Federal Govern-ment. The transfer of patent rights and rights intechnical data can encourage technology transfer,but both might restrict access to Federal information.High-tech companies and universities benefit fromthis policy, but the information industry, librarians,and the general public are concerned that access toFederal STI could be impaired if the policy is carriedtoo far (see ch. 4 discussion).

Enhancing the Role of the Private Sector

The Federal Government can encourage theprivate sector in several ways: First, the governmentshould ensure open and equitable access for thosewho seek Federal information regardless of cost.Second, the government is expected to assist thelibrary and educational institutions distribute Fed-eral information through technology-enhanced dis-semination. This will require rethinking the futureroles of libraries and schools in the information age,including new arrangements with the governmentand commercial sectors:*

Third, the commercial information industry ex-pects the government to provide equitable, competi-tive conditions for contractors and vendors involvedin Federal information dissemination. The Securitiesand Exchange Commission and Patent and Trade-mark Office have proposed "exchange agreements"whereby private contractors would provide theagencies with "free" automation services in returnfor exclusive rights for redissemination of agencyinformation. These agreements were bitterly con-tested by Congress and the information industry asanticompetitive and have since been modified or

oFor a recent debate, see R.C. Atkinson, "A Question of Information Policy," editorial, Science, vol. 246, Nov. 10, 1989, p. 733; and DA.B.Lindberg, "Information Policy," letter to the editor. Science, vol. 246, Dec. 22, 1989, pp. 1547-1548.

42Sec OTA, Informing the Nation, op. cit., footnote 1, chs. 6 and 7; Association of Rescind' Libraries, Technology and U.S. Government InformationPolicies: Catalysts for New Partnerships (Washington, DC: October 1987); U.S. Congress, Office of Technology Assessment,Linking for lzarning:A New Course in Education, OTA-SET-430 (Washington, DC: U.S. Government Printing Office, November 1989); and U.S. Department of Education,

Office of Library Programs, Rethinking the Library in the Information Age (Washington, DC: U.S. Government Printing Office, October 1988).

0 (I


Increased availability of Federal STI inelectronic formats should stimulate andstrengthen the private-sector role in STIdissemination.

terminated. The industry insists that, when contrac-tors disseminate Federal information, the agenciesshould be obligated to provide the same informationon equal terms to any interested vendors.

The information industry is also sensitive to theprospect of direct competition between Federalagencies and commercial vendors. The industry nowrecognizes the legitimacy of direct governmentdissemination. The views of the Information Indus-try Association have changed from opposition to anydirect electronic dissemination by government, toopposing agency dissemination of value-added butnot basic or raw government information. Theindustry now supports a partnership or complemen-tary relationship between government and industry.For example, improvements in agency dissemina-tion of STI could stimulate new opportunities forprivate sector development of innovative STI Prod-ucts and services that cut across agency and discipli-nary lines.

Some vendors now offer a variety of bulk rate,off-peak, and discount products and services togovernmental and not-for-profit customers. Theindustry opposes any copyright-like restrictions onFederal agency information, and prefers that licens-ing or other agreements be offered on a nondiscrimi-natory basis to all competitors. The industry benefitsfrom obtaining Federal information in electronicforms, since the cost of converting electronic infor-mation to commercial applications is typically less

than working from paper formats. It follows that ifthe benefits of electronic formats are available to thecommercial sector, they should also be available tothe not-for-profit sector (e.g., libraries, universities,and noncommercial companies such as OCLC, Inc.and Reference Point, Inc.43).

Increased availability of Federal STI in electronicformats should stimulate and strengthen the privat--sector role in STI dissemination. This has beenshown to be true with online and compact opticaldisk formats. Collection and creation of the FederalSTI databases and documents are paid for by thetaxpayers. The development cost of many of thesedatabases is beyond what most private organizationscould afford or would risk on such a venture. Thesedatabases are a shared national resource. Newelectronic technologies enable the Federal scienceagencies to prepare and maintain these databasesand distribute them to the publicincluding theprivate sector. Private vendors are thus assisted bythe government in their business of redisseminating,repackaging, and enhancing Federal STI and con-verting it into marketable products and services.

Electronic Foderal STI should also benefit com-mercial telecommunication companies." As elec-tronic Federal STI is accepted by users and demandfor online services increases, the use of telecommu-nication gateway services should likewise increase.Market stimulation should extend to the Bell operat-ing companies, long distance telephone carriers,commercial value-added networks, and also not-for-profit networks. The latter include the OnlineComputer Library Center network, Research Librar-ies Information Network, Western Library Network,and scientific networks such as Bitnet and NSFnet(operated by Educom and the National ScienceFoundation, respectively).

43keference Point has recently initiated a project to develop a global environmental information network for the exchange of information on climatechange, deforestation, loss of biodiversity, and other global environmental challenges.

"For general discussion of the U.S. communications infrastructure, see U.S. Conyess, Office of Technology Assessment, Critical Connections:Communication for the Future, OTA-C1T-407 (Washington, DC: U.S. Government Printing Office, January 1990).

Chapter 4

Reaching Consensus on the Open Flowand Governmentwide Dissemination of Federal STI

Open Flow of STI

The U.S. scientific and technical enterprise ispremised on the open exchange of STI. The basicpremise of openness has generally been modifiedonly in narrowly defined areas of STI relating tonational security. Recently several trends haveconverged to raise questions about the need torestrict the flow of Federal STI for other reasons.

First, the United States is no longer a leader inmany areas of science and technology. The U.S.advantage that existed during the post-World War IIyears, through the 1950s and 1960s, has evaporated.Second, the global economy is more competitive,with foreign countries and companies challengingU.S. dominance in several economic sectors. Third,the U.S. military industrial advantage is undercompetitive pressure from foreign manufacturers.Fourth, electronic technologies vastly speed up thecollection, storage, dissemination, and use of SI'land thus accelerate the rate of inform4tion transferwithin the global scientific t aid technic al commu-nity.

Several efforts to restrict access to Federal STI foreconomic or security reasons emerged in the 1980s.The Department of Defense (DoD) generally sup-ports an open exchange of basic research informa-tion to promote scientific progress in defensetechnology. However, some DoD agencies and

1,services (e.g., especially the Air Force and NationalSecurity Agency (NSA)) favor restrictions on accessto applied research and technical information. Thisled to proposals to give NSA the lead in ensuringgovernment computer security and to extend NSA'sauthority to so-called "sensitive but unclassified"Federal information.2

"Sensitive but unclassified" was to includeunclassified information that becomes sensitive tothe national security when, for example, it isaggregated in electronic form and available overonline databases. Opposition to this proposal by thecommercial information industry, academia, scien-tific and library associations, civil liberties groups,and Congress led to enactment of the ComputerSecurity Mt of 1987. This act assigned the NationalBureau of Standards (now the National Institute ofStandards and Technology (N1ST))rather thanNSAthe lead role for civilian computer security,and limited the role of DoD with regard to unclassi-fied, civilian Federal information. Information in-dustry and civil liberties representatives, amongoth, rs, are still concerned about the NSA role incivilian information systems, and its potential tointerfere with the free flow of unclassified Federalinformation.3

Congress seeks to ensure that the flow of scien-tific and technological information is equitable and

'See U.S. Congress, Office of Technology Assessment, Federal Government Information Technology: Management, Security, and CongressionalOversight, OTA-CIT-297 (Washington, DC: U.s. Government Printing Office, February 1986); Commercial Newsgathering From Space,O1A-TM-ISC-40 (Washington, DC: U.S. Government Printing Office, May 1987); Marine Minerals: Exploring Our New Frontier, OTA-0-342(Washington, DC: U.S. Government Printing Office, July 1987), esp. ch. 7 on "Federal Programs for Collecting and Managing Oceanographic Data;"The Regulatory Environment of Science, OTA-TM-SET-34 (Washington, DC: U.S. Government Printing Office, February 1986); InternationalCompetition in Services, OTA4M-328 (Washington, DC: U.S. Governmem Printing Office. July 1987); Defending Secrets, Sharing Data,OTA-CIT-310 (Washington, DC: U.S. Government Printing Office, October 1987); Science, Technology, and the First Amendment, OTA-CIT-369(Washingtvn, DC: U.S. Government Printing Office, January 1988); Holding the Edge: Maintaining the Defense Technology Base, OTA-ISC-420(Washington, DC: U.S. Government Printing Office, April 1989).

2U.S. Congress, Office of Technology Assessment, Defer ding Secrets, op. cit., footnote 1, chs. 1, 6, and 7; also see W.R. Blados, "ControllingUnclassified Scientific End Technical Information," Informai.on Management Review, vol. 2, No. 4, 1987, pp. 06-60.

3Public Law 100-235, the "Computer Socurity Act of 1987," Jan. 8, 1988. Also see testimony of Kenneth Allen, Senior Vice President,Information

Industry Association, and Marc Rotenberg, Director, Washington Office, Computer Professionals for Social Responsibility, before a May 4, 1989,hearing of the House Committee on Governmeat Operations, Subcommittee on Izgislation and National Security. The House Committee on GovernmentOperations and industry and public-sector representatives art still not satisfieu with the working relationship between NIST and NSA, and seek further

assurances that N1STnot NSAwill be in charge of civilian computer security.

3--1 1

34 Helping America Compete: The. Role of Federal Scientific and Technical Information

111111=111

Too much emphasis on short-term com-mercializatien of technology and relatedtechniciti data could actually impair theU.S. long-term competitive posture.

I MENIMIEn=r-

reciprocal among nations.4 T F-r:r0tary of State isdirected to consider several a negotiatinginternational scientific agreemnts:J

scientific merit;equity of access by U.S. public and privateentities to public (and publicly supported pri-vate) research and development (R&D) oppor-timities and facilities in each country which isa major trading partner of the United States;possible commercial or trade linkages with theUnited States which may flow from the agree-ment or activity;national security concerns; andany other factors deemed appropriate.

The "Stevenson-Wydler Technology InnovationAct of 1980"6 and the "Federal Technology Trans-fer Act of 1986"7 are efforts to reinforce the U.S.position in international competition by facilitatingthe transfer of technology from Federal laboratoriesto the private sector. These acts authorize Federallaboratories to cooperate with other governmental(Federal, State, local) entities and with the privatesector (including universities and commercial firms)in R&D, and to license, transfer, or waive patentrights resulting from cooperative R&D. However, if

exclusive rights in technical data are given by thegovernment to the private sector, this could result inconstraints on the dissemination of much unclassi-fied Federal STI.

A 1987 executive order directs agencies totransfer tix.hnical data by allowing Federal contrac-tors and grantees to own rights in computer software,engineering drawings, and technical data funded byFederal contract or grant.8 This executive order andother proposals by the Office of Federal Procure-ment Folic? caused a vigorous debate over how totransfer government-funded technology and stillpreserve the public value of knowledge producedwith taxpayer money.° Agencies such as the De-partment of Energy (DOE)and National Aeronauticsand Space Administration (NASA) consider theopen exchange of technical information to befundamental to their research missions. A blankettransfer of rights in technical data could impairresearch in fields such as energy and space thatgenerate technologies that are valuable assets withcommercial potential. Too much emphasis on short-term commercialization of technology and relatedtechnical data could actually impair the U.S. long-term competitive posture."

In many fields of science and technology, STIdeveloped by other countries is increasingly impor-tant. Policies that severely restrict public access tounclassified Federal STI might encourage similarrestrictions by other countries and frustrate theinternational exchange of SU. The thrust of DOEpolicy in energy research is to increasenot decreasethe equitable exchange of international energy STI.The DOE Office of Scientific and Technical Infor-

4See, ',7or example, H.C. Relyea, Striking A Balance: National Security and Sciennfic Freedom, American Association for the Advancement ofScience, Washington, DC, 1985; U.S. Congress,Office of Technology Assessment, Science, Technology, and the Fir., imendment, op. cit., footnote1, ch. 4; and National Academy of Science, Panel on the Impact of National Security Controls on InteruAtional Technology Transfer, Balancing theNational Interest: U.S. National Security Export Controls and Global Economic Competition (Washington, DC: National Academy Press, 1987).

5Public Law 100-418, the "Omnibus Trade and Competitiveness Act of 1988," Aug. 23, 1!..S8, Part IISymmetrical Access to TechnologicalResearch, sec. 5171 (a) and (d).

8Pirblic Law 96480, Oct. 21, 1980.

7Public Law 99-502, Oct. 20, 1986.

8Executive Order 12591, Apr. 10, 1987.

U.S. Office of 1 ederal Procurement Policy, "Intellectual Property Rights Policy," draft, February 1989.

10For discussion of proposals to establish and transfer copyright in Federal computer software, see U.S. General Accounting Office, TechnologyTransfer: Copyright Law Constrains Commercialization of Some Federal Software, GAO-RCED-90-145 (Washington, DC: U.S. General AccountingOffice, May 1990), and testimony of James W. Curlin, OTA, and other witnesses before an Apr. 26, 1990, hearing of the House Committee on Science,Research, and Technology, Subcommittee on Science, Research, and Technology. For general discussion of computer-related intellectual propertyissues, see U.S. Congress, Office of Technology Assessment, Computer Software & Intellectual Property, OTA-BP-CIT-61 (Washington, DC: U.S.Government Printing Office, March 1990), and Intelkctual Property in an Age of Electronics and Information, OTA-CIT-302 (Washington, DC: U.S.Government Printing Office, April 1986).

11See, for example, the special issue, "Symposium on the Impact of Competitiveness," Government Information Quarterly, vol. 6, No. 1, 1989.rt

Chapter 4ReacAing Consensus on the Open Flow and Governmentwide Dissemination of Federal STI 35

mation manages the Energy Technology Data Ex-change (ETDE) under the auspices of the Interna-tional Energy Agency. Canada, Denmark, Finland,France, the Federal Republic of Germany, Italy,Japan, the Netherlands, Norway, Spain, Sweden,Switzerland, and the United Kingdom participatealong with the United States.12

Participating countries send summaries of energy-related STI to DOE on a monthly or biweekly basis.DOE transt its them to participating countries fordisseminatioa to their own researchers and poli-cymakers. The ETDE includes about 7,500 biweeklyupdated STI entries and over 2 million entries in theretrospectiv. ..ile. The latter is available by onlinecommercial vendors to research organizations, uni-versities, and libraries within the participating coun-tries. Online usage is divided roughly as follows:industry (71 percent); academia (15 percent); andgovernment (14 percent).13

Numerous vendors sell or resell Federal STIdatabases, or :r..:-.1ude significant Federal ST1 in morecomprehensive ,latabases, to both domestic andinternational cvstomers. Reduced availability ofFederal STI to L.ommercial vendors (and for thatmatter, not-for-profit vendors as well), coupled withreciprocal restrictions by other countries, wouldreduce the utility and value of comprehensive,subject-specific databases.

The challenge is to develop an STI disseminationpolicy that:

1. encourages U.S. researchers to employ allmeans, including electronic where appropri-

ate, to facilitate access to and use of domesticand foreign S11; but at the same time

2. protects U.S. national security interests bycontrolling access to classified or narrowlydefined militarily sensitive STI; and

3. encourages U.S. international competitivenessthrough:

a. the open, reciprocal international exchangeof STI,

b. domestic transfer of federally funded tech-nology from the Federal Government to theprivate sector whore appropriate,

c. protection of private-sector proprietaryrights in technology and data (to the extentnon-Federal funds are used), and

41. domestic transfer of rights in technical datadeveloped by or for the Federal Government(with Federal funding) to the private sectorin narrowly defined areas where the benefitssubstantially outweigh the costs.14

Congress, the Office of Science and TechnologyPolicy (OSTP), and the Office of Management andBudget (OMB) must reconcile their philosophicaldifferences about the open flow of ST1 and provideguidance to the agencics. A balance is needed. Thisbalancing should consider legislative proposals thatfocus on the open, unrestricted flow of Federalinformation's as well as legislation that wouldtransfer federally supported technology and infor-mation to the private sector.16 A balance must alsoconsider statutes that promote information access

12International EnergyAgency , Energy Technology Dora Exchange, 1989 Annual Report, ETDE/OA-37 (Oak Ridge, TN: U.S . Department of Energy,

Office of Scientific and Technical Information, 1989); International Energy Agency, Introducing ETDE: An !EA Multiloteral Information Program,

E1DE/OA-06-Rev. (Oak Ridge, TN: U.S. Department of Energy, Office of Scientific andTechnical Information, June 1989).

13Ibid.

I4For some proposed policy statements, see "Changing Federal Relationships in Intellectual Property," February 1989 draft, provided to OTA by

CENDI, and "Policy Diractions (in New Regulations on Patents and Copyrights]," May 1989 draft, pmvided to OTA by NASA.

15U.S. Congress, House, H.R. 2381, the "Information Policy Act of 1988," 101st Cong., 1st sess., May 16, 1989; H.R. 3695, the "Paperwork

Reduction and Federal Information Resources Management Act of 1989," 101st Cong., 1st seas., Nov. 17, 1989; and S. 1742, the "Federal Information

RCSOUIDCS Management Act of 1989," 101st Cong., 1st sess., Oct. 6, 1989; also see U.S. Congress, House, HR. 2773, the "Freedom of Information

Public Improvements Act of 1989," 101st Cong., 1st sess., June 28, 1989, that would redefme government records for FOIA purposes to cover all

"computerized, digitized and electronic information."

16See U.S. Congress, Senate, S. 550, the "Department of Energy National Laboratory Couperative Research and Technology Competitiveness Act

of 1989," 101st Cong., 1st seas., Mar. 9, 1989, as amended Aug. 4, 1989, and included as the "Department of Energy Natkoal Competitiveness

Technology Transfer Act of 1989," in Title mon, Part C of S. 1352, the "National Defense Authorization Act for Fiscal Years 1990 and 1991," Aug.

4, 1989. Also see U.S. Congress, Senate, Committee on Armed Sesvices, National Defense Authorization Act for Fiscal Years 1990 and 1991, Report

No. 101-81, lelst Cong., 1st seas. (Washington, DC: U.S. GovernmentPrinting Office, July 19, 1989); and U.S. Congress, Senate, Committee on EnergY

and Natural Resources, Department of Energy National Laboratory Cooperative Research and Technology Competitiveness Act of 1989, Report No.

101-108, 101st Cong., lst seas. (Washington, DC: U.S. Government Printing Office, Aug. 4, 1989).


(such as the Freedom of Information Act (FOIA)17)and those statutes that tend to limit access.

The Defense Authorization Act of 1984 author-izes DoD to withhold certain unclassified butmilitarily sensitive and export-controlled scientificand technical information developed by or for DoDthat would otherwise be accessible under FOIA.118NASA sought similar authority to withhold techni-cal information about NASA-funded technologies.NASA policies also limit the dissemination oftechnical information to U.S. industry only, if it islikely to give the United States a competitive edgein commercializing NASA technology. But thisinformation is currently available through FOIArequests, thus undermining NASA's policy. NASAhas therefore sought to establish "significant poten-tial for commercial use" as a statutory basis forFOIA exemption.°

A 1988 FOIA proposal supported by the Office ofScience and Technology Policy and U.S. Depart-ment of Justice would have provided exemption forany ST1 that: 1) "was generated in a labora-tory. . .owned Lind operated, in whole or in part, bythe Federal Government"; 2) "has commercialvalue"; and 3) if disclosed under FOIA, "could bereasonably expected to cause harm to the economiccompetitiveness of the U.S."2° This proposal wascontroversial and was challenged on severalgrounds, including:21

the need for such a blanket exemption has notbeen established, since only a very smallpercentage of STI is commercially sensitive;such an exemption could set a dangerousprecedent for undermining FOIA in other

subject areas and by other kinds of agencies;andan exemption could encourage reciprocal ac-tions by other countries that would underminethe international exchange of ST1 and hurt theU.S. R&D effort in the longer term.

In reviewing Federal policy, Congress needs totake into account the changing economic realities.The globalization of the economy means that anincreasing percentage of U.S. domestic R&D com-panies operate under foreign ownership, just asmany U.S. corporations now have their own foreignsubsidiaries. Most of the largest U.S. companiesoperate globally, with research, manufacturing, andmarketing distributed over many countries. Similartrends are evident in the commercial informationsector, to the paint where one cannot assume that aU.S. information vendor operates under domesticrather than foreign ownership, and vice versa. Underthese conditions, the old approaches to controllinginformation access do not work.

Role of Governmentwide Dissemination andArchival Agencies in STI

As information changes from paper (and to alesser extent microfiche) to electronic formats, theroles of the agencies with governmentwide dissemi-nation and archival responsibilities require reconsid-eration. This is especially true for scientific andtechnical information, much of which is in digitalform and may only be usable in electronic formats.

The major governmentwide agencies are: theGovernment Printing Office (GPO), responsible forprinting, sales of selected documents by the Superin-

17For a detailed discussion of issues concerning an electronic FOIA, see J. Grodsky, "The Freedom of Information Act in an Electronic Age," in U.S.

Congress, Office of Technology Assessment, Informing the Nation: Federal Information Dissemination in an Electronic Age, OTA-CIT-396(Washington, DC: U.S. Government Printing Office, October 1988), pp. 207-236; also see Jerry J. Berman, "The Right to Know: Public Access to

Electronic Information," in P.R. Newberg (ed.), New Directions in Telecommunications Policy, vol. 2, Information Policy and Economic Policy(Durham, NC: Duke University Press, 1989), pp. 39-69; H.H. Penitt, Jr., Electronic Acquisition and Release of Federal Agency Information, Report

to the Administrative Conference of the United States, Oct. 1, 1988 (also see the related article by H.H. Penitt, Jr., in Administrative Law Review, vol.

41, 1989, pp. 253 ff.); and Thomas L. Susznan, Chairman, American Bar Association, Committee on Government Information and Privacy, "Access

to Electronic Information Under the Freedom of Information Act," draft report, Feb. 28, 1989. Also see statements of Ronald Plesser, Esq., Piper &Marbuy, and Patti A. Goldman, Esq., Public Citizen, Inc., before a July 11, 1989, hearing of the House Committee on Government Operations,

Subcommittee on Government Information, Justice, and Agriculture.

11U.S. Congress, Public Law 98-94, "Department of Defense Authorization Act of 1984," Sept. 24, 1983; also see W.R. Blados, "Controlling

Unclassified Information," op. cit., footnote 2.

11See statement of Kenneth S. Pederson, Associate Administrator for External Relations. U.S. NationalAeronautics and Space Administration, before

a hearing of the House Committee on Science, Space, and Technology, Subcommittee on International Scientific Cooperation, July 19, 1989.

2°U.S. Department of Justice, Office of Legal Policy, Office of Information and Privacy, "New FOIA Legislation Proposed to Promote U.S.

Competitiveness," FOIA Update, vol. IX, No. 1, Winter 1988, pp. 1-2.

215ee U.S. Congress, Senate, Committee on the Judiciary, Subcommittee on Technology and the Law, Information Policy and Competitiveness,

Hearing, 100th Cong., 2d seas. (Washinbton, DC: U.S. Government PrintingOffice, Mar. 16, 1989).

1414

Chapter 4Reaching Consensus on the Open Flow and Governmentwide Dissemination of Federal S7'I 37

The question is how to preserve andstrengthen the ability of the government-wide agencies to carry out their func-tions in a decentralized electronic envi-ronment.

tendent of Documents (SupDocs), and distributionof documents through the Depository Library Pro-gram (DLP); the National Technical InformationService (NTIS), the clearinghouse and sales outletfor technical documents; and the National Archivesand Records Administration (NARA), concernedwith archiving and long-term preservation of docu-ments.22

Decentralized Nature of STI

It is clear that the creation, storage, and dissemi-nation of STI is decentralized within the scienceagencies. This is because:

1 STI is voluminous, and agencies have diffi-culty in managing their own information basesmuch less another agency's data;

2. centralizing all STI in one data bank is neithercost-effective nor technically feasible at thistime;

3. technical systems for creating, storing, anddisseminating STI are typically closely tied toagency automation programs;

4. centralizing STI dissemination, even if techni-cally feasible, could slow innovation and limit

opportunities for improving efficiency in theagencies;

5. the diversity of STI needs and users among theFederal science agencies includes many varieddisciplines and research areas;

6. the decentialized approach brings agency STIofficials and the scientists and researchers whocreate and use the STI closer together; and

7. the economies-of-scale for electronic formatsare ixchieved at lower levels of demand than forink-on-paper printing.

Several agencies have data centers that areresponsible for collecting, archiving, and dissemi-nating databases, and much of these data are alreadyin electronic formats. The major centers include: theNational Space Science Data Center, National Cli-matic Data Center, National Oceanographic DataCenter, National Geophysical Data Center, EarthScience Information Center, and Earth ResourcesObservation Systems Data Center. Several of thescience agencies have their own central STI office(e.g., at NASA and DOE23) for STI documents andbibliographies, and most have infrastructure forhandling STI, though it varies among the agencies(e.g., in terms of resources, staffing, visibility).

The question is how to preserve and strengthenthe ability of the governmentwide agencies to carryout their functions in a decentralized electronicenvironment. Alternatives were considered b, OTAin Informing the Netion,24 by various congressionalcommittees in hearings on NTIS, GPO, and theDLP,25 and at a NARA conference on electronicrecordkeeping.26

=The implications of electronk information for GPO, SupDocs, DLP, and NTIS are discussed in U.S. Congress, Office of Technology Assessment,Informing the Nation, op. cit., footnote 17, see esp. chs. 4-7, and 12. The implications for NARA are considered in National Academy of PublicAdministration, The Effects of Electronic Recordkeeping on the Historical Record of the U.S. Government(Washington, DC: National Archives andRecords Administration, January 1989),

23Soe National Aeronautics and Space Administration, The NASA Scientific and Technical Information System and How to Use It, NASA SP-7073,Washington, DC, 1989; and Department of Energy, The Role of the Office of Scientific and Technical Information in DOE' s Scientific and TechnicalInformation Program, November 1988,

24U.S. Congress, Office of Technology Assessment, Informing the Nation, op. cit., footnote 17.

23See, for example, U.S. Congress, House, Committee on Science, Space, and Technology, Subcommittee on Science, Research, and Technology,National Technical Information Service, Hearing, 100th Cong., 2d sess., U.S. Government Printing Office, Washington, DC, Feb. 24, 1988; U.S.Congress, House Committee on Science, Space, and Technology,National Bureau of Standards Authorization Actfor Fiscal Year 1989, Report 100-673,Part 1, 100th Cong., 2d sess., U.S. Government Printing Office, Washington, DC, June 3, 1988; U.S. Congress, House, Cotnmittee on Energy andCommerce, National Bureau of Standards Authorization Act for Fiscal Year 1989, Report 100-673, Part 2, 100th Cong., 24 sess., U.S. GovernmentPrinting Office, Washington, DC, July 8, 1988; U.S. Congress, House, Committee on Administration, Subcommittee on Procurement and Printing,hearings on "Review of 'be Printing Chapters of Title 44 of the U.S. Code Due to the Changes in Electronic Information Format, Distribution, andTechnology During the IA :t Decade," May 23-24 and June 28-29, 1989; U.S. Congress, House, Committee on Goverment Operations, Subwmmitteeon Government Information, Justice, and Agriculture, hearings on "Federal Information Dissemination Policies and Practices," Apr. 18, May 23, and

July 11, 1989.26National Archives and Records Administration, "Electronic Records: A Strategic Plan for the 1900s," Conference Summary and

Recommendations, June 21-23, 1989, see especially the recommendations of the working group on information collection and dissemination.

A t

38 Helping America Compete: The Role of Federal Scien:ific and Technical Information

Under the OTA scenario, the Federal scienceagencies retain primary responsibility for the storageand dissemination of STI collected by each agency.The science agencies would be governed by:

their enabling statutes regarding STI;OSTP guidance provided under the NationalScience and Technology Policy, Organization,and Priorities Act of 1976, as possibly amendedto give further congressional direction on STIpolicy;OMB guidance promulgated under the Paper-work Reduction Act (ch. 35 of Title 44 of theU.S. Code, as pnssibly further amended toprovide congressional statutory direction onoverall dissemination policy");GPO (and Joint Committee on Printing) guid-ance under the printing chapters of Title 44, aspossibly amended,28 to ensure that the integrityof the GPO printing procurement program,SupDocs sales program, and DLP is main-tained;NTIS guidance promulgated under the "Na-tional Technical Information Service Act of1988"29 to ensure that the integrity of the NTISclearinghouse is maintained; andNARA guidance promulgated under the archi-val chapters of Title 44, as possibly amended,to ensure long-term preservation and access toSTI.

This scenario is predicated on the assumption thatOMB, GPO, NTIS, and NARA guidance would begenerally consistent and compafible.

Roles of Science Agencies, NTIS, and GPO

One possible division of effort between themission agencies and governmentwide agencies isoutlined below using a hypothetical example of anelectronic producthydrology information of theU.S. Geological Survey (USGS) (e.g., trends instream flows and reservoir and lake levels) issued onCD-ROM:

USGS would notify GPO, NTIS, and NARA inadvance of production and supply productinformation (e.g., size of the hydrology data-

base, type of search-and-retrieval software,estimated cost and demand).GPO would decide whether the CD-ROMshould be included in the SupDocs salesprogram, based on an estimate of demandbeyond that being met by USGS direct sales.USGS could opt to use SupDocs as the primarysales outlet if the CD-ROM qualified.GPO also would determine whether the CD-ROM should be offered to depository libraries,and if so, how many libraries desired a copy ofthe CD-ROM.NTIS would decide whether the CD-ROMshould be included in the NTIS clearinghouseand sales program.GPO and NTIS would, on a coordinated basis,make sure that the CD-ROM is cataloged andlisted in appropriate governmentwide directo-ries and bibliographic databaseswhether ornot it is sold by GPO and/or NTIS."NARA would review the CD-ROM to deter-mine long-term archival needs.GPO and NTIS would, again on a coordinatedbasis, advise USGS of their need for copies ofthe CD-ROM (to meet estimated SupDocssales, depository library distribution, and NTISsaks needs).USGS would obtain CD-ROM production serv-ices in the manner that best meets its cost,quality, and turnaround requirements. Thiscould be through an agency contractor, GPOcontractor, GPO itself (if an inhouse service isoffered), or NTIS contactor (if NTIS offersCD-ROM services).Wherever the USGS CD-ROM is produced,GPO and NTIS would ride the order for thenumber of additional copies required.

This example could apply to all offline electronicproducts, including optical disks, magnetic tapesand cassettes, and diskettes (hard and floppy). Thelarge online electronic STI databases would bemaintained by the agency data centers. But onlinedirectories and possibly subsets of data might beha -died similarly to the CD-ROM illustration above.Some directories also could be disseminated onCD-ROM or other offline electronic foimats.

27See H.R. 3695, op. cit., footnote 15; S. 1742, op. cit., footnote 15.

23See H.R. 3849, the "Government Printing Office Improvement Act of 1990," 101st Cong., 2d se.ss., Jan. 23, 1990.

29See U.S. Congress, Public Law 100.519, Subtitle B--National Teclmical Information SaViCe, rqdifICA at 15 U.S.C. 3701 et. soq.

30See discussion of STI directories in ch, 5.

4 u

Chapter 4Reaching Consensus on the Open Flow and GovernmentwideDissemination of Federal 571 39

The future of NTIS and GPO will be influenced bythe increasingly decentralized, competitive environ-ment of the eleztronic information marketplace.Federal science agencies are rapidly installing elec-tronic systems for their activities, including thecollection, processing, and dissemination of STI(see the appendix for illustrations). NTIS and GPOwill have to adapt to the reality that technology haschanged and sometimes eliminated the distinctionsbetween reports, publications, databases, and thelike, and has blurred the distinctions between theirroles and those of the agencies.

Most Federal STI will likely exist in electronicform as computerized electronic databases. Userswill have a wide assortment of formats available,from printed reports to online information retrieval,printing-on-demand, and compact optical disk. NTISand GPO will have to become more flexible,adaptive, creative, competitive, and user-orientedthan they currently are.31 Many users may continueto prefer the convenience of "one-stop" informa-tion shopping at NTIS or GPO, especially forhard-to-find documents (or their electroric equiva-lents). But the governmentwide dissemination pro-grains will need to complement, not preempt,individual agency activities.

GPO and NTIS appear to be philosophicallyaccepting this reality. The former Acting PublicPrinter has stated GPO's preference for the "Elec-tronic GPO-Decentralized" approach.32 In thisscheme, GPO would continue its centralized con-

ventional printing functions, but would aggressivelyplan for and implement electronic printing anddissemination services, working through a decen-tralized Federal electronic information environ-ment.33 Centalized conventional ink-on-paper print-ing would continue, with about three-quarters of allFederal printing done by or through GPO (althoughthree-fourths of this is contracted out by GPO tocommercial printing companies), and the rest atauthorized agency printing plants. Decentralizedagency electronic information dissemination wouldcontinue, with GPO offering a variety of electronicservices to the agencies, but on a competitive,discretionary basis (in contrast to conventionalink-on-paper printing where CPO services must beused, unless an explicit exemption or exception isgranted).34

The NTIS Director and Deputy Director havestated th commitmerd in principle to the "Elec-tronic NTIS" alternative.35 After years of decliningdemand for paper and microfiche products and thedebate over privatization, Congress has directedNTIS to modernize. NTIS has developed a prelimi-nary plan to increase its use of electronic formats,including CD-ROM, electronic bulletin boards, and,ultimately, an electronic documeat system thatcould accept electronic input from the sourceagencies and support electronic printing-on-demand. To be successful, NTIS will need to reduceper-unit costs, decrease the time delays between theexistence of a document and its availability via

31For further discussion, see U.S. Congress, Office of Technology Assessment, Informing the Naticn, op. cit., footnote 17, ch. 12.

32U.S, Congress, Office of Technology Assessment, Informing the Nation, op. cit., footnote 17, ch. 4. Also see FB. Wood, "Title 44 and Federal

Information DissaninationA Technology and Policy Challenge for Congress: A Viewpoint," Government Publications Review, vol. 17, 1990, pp.

1-5.

33See statement of Joseph E. Jenifer, Acting Public Printer, Government Printing Office, before a May 23, 1989, hearing and statement of Samuel

B. Scag8s, Assistant Public Printer, Operations and Procurement, Government Printing Office before a June 29, 1989, hearing, Committee on House

Administration, Subcommittee on Procurement and Printing. Also see statements of Joseph E. Senifer, Acting Public Printer, before a Feb. 7, 1989,

hearing of the House Committee on Appropriations, Subcommittee on the Legislative Branch, a July 11, 1989, hearing of the House Committee on

Government Operations, Subcommittee on Government Information, Justice, and Agriculture, and a Mar. 7, 1990, hearing of the Committee on House

Administration, Subcommittee on Procurement and Printing. The new Public Printer stated a position on some of these issues before an Apr. 6, 1990,

hearing of the Senate Committee on Appropriations, Subcommittee on the Legislative Branch. For other views and general discussion, see U.S. Congress,

House, Committee on House Administration, Subcommittee on Procurement and Printing, Title 44 U.S.C.Review, Hearings, 101st Cong., 1st sess.,

May 23, 24, and June 28, 29, 1989 (Washington, DC: U.S. Government Printing Office, 1989); statements of Fred B. Wood, OTA, before May 23, 1989,

and Mar. 7, 1990, hearings of the Committee on House Administration, Subcommittee on Procurement and Printing; and OTA comments on S. 1742,

die "Federal Information Resources Management Act of 1989," prepared for a Feb. 21-22, 1990, hearing of the Scnate Committee on Governmental

Affairs.34This general approach is consistent with that of other countries such as New Zealand and Canada. See Canadian Communications Savices

Directorate, "Electronic Publishing Information Center," Electronic Publishing Bulletin, October 1989.

35U.S. Congress, Office of Technology Asmsment, Informing the Nation, op. cit., footnote 17, chs. 5 and 12. 4 7


NTIS, and increase user awareness of NTIS serv-ices.36

Beyond this, NTIS must develop a clear strategicvision of its future, and a realistic, detailed implem-entation plan for getting there. During the 1980s,user demand for paper and microfiche documents inthe basic NTIS archive steadily dropped; agencycooperation in providing documents to NTIS alsodeclined. NTIS sales of electronic formatsthe onebright spotare likely to feel increasing pressurefrom agency, private sector, and GPO competition.37

Roles of NARA and DLP

The roles of NARA and the DLP deserve specialattention. NARA might find that agency data centerscan efficiently archive STI databases, releasingNARA from the need to retain physical control.Even if an agency or data center serves as thearchive, NARA would help ensure that the system iscost-effective and meets data and technical stan-dards (e.g., regarding longevity of storage media,conversion from one storage medium to another, andportability among different media and equipment).NARA could also assist the data centers in determin-ing what should and should not be retained inhouse,with permanent STI archives retained by NARA.NARA needs to develop clear and workable agree-ments with the science agencies, and with NTIS andthe Library of Congress (LOC), to ensure thatarchivable STI does not fall through the cracks.

Machine-readable materials are included withinthe legal definition of "record."38 NARA hasinitiatorl a program for archiving electronic recordsthat is now being extended to Federal r r Perma-

nent electronic records identified by NARA include,for example:39

unique and important scientific and technicaldata resulting from observations of naturalevents or phenomena or from controlled labora-tory or field experiments;natural resources data related to land, water,minerals, or wildlife; andgeographic data used to map the surface of theearth.

NARA will need to assess the vast store of geo-graphic, space, and earth sciences data with respectto archival needs and requirementsa task thatbecomes even more challenging with the rapidevolution of electronic storage and retrieval technol-ogies and the poor condition of current data archives.Technological change means that large amounts ofarchived STI will be inaccessible and/or unusable tofuture generations of researchers unless standardinformation formats are developed and mandated.The long-term utility of STI requires that today'sdata and documents be retrievable with tomorrow'stechnologies.40

As for the Depository Library Program, thereappears to be a consensus that electronic formatsshould be included, although there are differences ofopinion over implementation. For several years now,the congressional Joint Committee on Printing(JCP), Depository Library Council, and the majorlibrary associations have argued that, as the FederalGovernment makes increasing use of electronicinformation, the DLP must also include electronicinformation, lest the integrity of the program be

36See National Technical Information Service, Annual Report to the Congress from the Secretary of Commerce, The National Technical Information

Service: Operations, Audit, and Modernization, January 1989; also see U.S. Congress, Office of Technology Assessment, Informing the Nation, op. cit.,

footnote 17, chs. 5 and 12; C.R. McClure, P. Hernon, and G.R. Purcell, linking the U.S. National Technical Information Service With Academic and

Public Libraries (Norwood, NJ: Ablex Publishing Corp., 1986). Also see Statement of Joseph F. Caponio, Director. National Technical InformationService, before a July 13, 1989, hearing of the National Commission on Lib! aries and Information SCieDee.

37For further discussion, see the statements of Fred B. Wood, OTA, Harold H. Shill, West Virginia University, and Jean Mayhew, United TechnologiesCorp., before a Mar. 8, 1990, hearing of the House Committee on Science, Space, and Technology, Subcommittee on Science, Research, and Technology.Also see C.F. McClure, "The Future of the National Technical Information Service: Issues and Options," Jan. 20, 1990, contractor paper prepared forOTA. For background discussion of the NTIS privatization debate, see F.B. Wood, "Proposals for Privatization of the National Technical InformationService: A Viewpoint," Government Publications Review, vol. 15, 1988, pp. 403-409 (which is based on testimony presented at a Feb. 24, 1988, hearingof the House Committee on Science, Space, and Technology, Subcommittee on Science, Research, and Technology).

3844 U.S.C. 3301.39U.S. National Archives and Records Administration, "Managing Electronic Records: An Instructional Guide," draft, no date, pp. 15-17; also see

Michael L. Miller, "Appraisal and DEposition of Electronic Records," National Archives and Records Administration,March 1988 draft; and June 13,

1989. cooperative agreement between NARA and NOAA.

40rnle U.S. General Accounting Office is conducting audits of the major data archives maintained by Federal science agencies. See U.S Government

Accounting Office, Space Operations: NASA Is Nut foperly Safeguarding Valuable Data From Past Missions, Report to the Chairman, Committeeon Science. Space, and Technology, U.S. House of Representatives, IMTEC-90-1 (Washington, DC: GAO, March 1990). Subsequent reports will aaress

NOAA and USGS data archives. 4 8

Chapter 4Reaching Consensus on the Open Flow and Governmentwide Dissemination of Federal ST1 41

The long-term utility of STI requiresthat today's data and documents beretrievable with tomorrow's technolo-gies.

-1111MINMI

eroded.41 In May 1989, GPO's General Counselruled that it has legal authority to distribute agencypublications in electronic format to depositorylibraries, thereby clarifying a 1982 opinion that waswidely interpreted as limiting the DLP to traditional(paper and microfiche) formats.42 This apparentlyended a long conflict between GPO and the JCPabout whether the depository library provisions orTitle 44 apply to government publications regardlessof format.°

The differences between the JCP, OMB, andInformation Industry Association (IIA) appear tohave narrowed. OMB supports the voluntary partici-pation of agencies in DLP electronic dissemination,and is willing to consider requiring that some agencyelectronic information products be provided todepository libraries.44 The IR now supports the

inclusion of some electronic formats in the DLP, butwith reservations about online dissemination andfinancing, and suggests testing alternative mecha-nisms such as vouchers, bulk rate and off-peakcontracts, user charges, and cost-sharing.45 In Con-gress, the House Appropriations Subcommittee onthe Legislative Branch has supported distribution ofCD-ROMs to depository libraries, and may be opento distributing other electronic formats, includingonline services, although questions of cost, demand,technical feasibility, and administrative responsibil-ity have not been resolved.46 These questions,among others, are being addressed through the DLPelectronic pilot projects now being implemented.'"

Two of the DLP pilot projects cover Federal STLThe first involves the distribution of the Environ-mental Protection Agency's "Toxic Release Inven-tory (TRI)" to depository libraries. TRI includesdetails on the location, storage, emissions, and wastetreatment and transfer for over 300 toxic chemicals.EPA is disseminating TR' to the public online viathe National Library of Medicine computer center,in magnetic tape format via the NTIS and GPO salesprograms, and in computer output microfiche andCD-ROM fonnats through selected libraries, incluc -ing all 1,400 depository libraries.48 The second

41See statements of D. Kaye Gapen, Dean of Libraries, University of Wisconsin (on behalf of (he Association of Research Libraries), and SandraMcAnich, Head, Government Documents, Univenity of Kentucky Libraries (on behalf of the Government Documents Roundtable, American LibraryAssociation), before a May 24, 1989 hearing of the House Administration Committee, Subcommittee on Procurement and Printing. Also see the statementof D. ICaye Gapen, on behalf or are American Library Association and Association of Research Libraries, before a Feb. 7, 1989, hearing of the HouseCommittee on Appropriations, Subcommittee on the Legislative Branch.

42Manorandum from GPO General Counsel to Acting Public Printer, "GPO Dissemination of Federal Agency Publications in Electronic Format,"May 22, 1989.

435ee U.S. Congress, Joint Committee on Printing, Provision of Federal Government Publications in Electronic Format to Depository Libraries,Report of the Ad Hoc Committee on Depository Library Access to Federal Automated Databases (Washington, DC: U.S. Government Printing Office,1984); U.S. Congress, Joint Committee on Printing, An Open Forum on the Provision of Electronic Federal Information to Depository Libraries, 99thCong., 1st sess. (Washington, DC: U.S. Government Printing Office, 1985); Joint Committee on Printing resolutions of Apr. 8, 1987, June 17, 1987,

and June 29, 1988 regarding GPO, depository libraries, and electronic formats; and letter from Honorable FrankAnnunzio, Chairman, Joint Committee

on Printing, to Honorable Ralph E. Kennickell, Jr., Public Printer, Mar. 25, 1988.

44See Office of Management and Budget, "Second Advance Notice of Further Policy Development on Dissemination of Information,"FederalRegister, vol. 54, No. 114, June 15, 1910, pp. 25554-25559.

45See statement of Kenneth B. Allen, Senior Vice President, Government Relations, Information Industry Association, accompaniedby Peyton R.

Neal, Jr., Clair, HA Government Printing Office Committee, before a May 24, 1989, hearing of the House Committee on Administration, Subcommittee

on Procurement and Printing. Also see a somewhat more critical statement of Paul P. Massa, Presidtnt, Congressional Information Services, Inc., before

a July 13, 1989, hearing of the National Commission on Libraries and Information Science. One private vendor, Legi-S late, Inc., has offered to provideelectronic online dissemination of selected congressional information to depository libraries at bulk ratediscounted prices, based in part on the results

of a successfid 5 112 month pilot test with 51 depository libraries. The same concept could be used by othervendors with respect to other types of Federal

information, including ST1. See Legi-Slate, "Pilot Project Evaluation Preliminary Summary," Jan. 8, 1989.

46U.S. Congress, Committee on Appropriations, Legislative Appropriations Bill. 1989, Report to accompany H.R. 4487, Report No. 100-621, 100th

Cong., 2d sess., 1988. Also see statement of Honorable Viz FaZiO, Chairman, House Committee on Appropriations, Subcommittee on the Legislative

Branch, before a June 28, 1989, hearing of the House Committee on Administration. Subcommittee on Printing and Procurement.

47The U.S. General Accounting Office is conducting an evaluation of the research methodology of theelec tronic pilot projects. See May 8, 1989, letter

from Donald E. Fossedal, Assistant Public Printer, U.S. Government Printing Office, to Richard Fogel, Assistant Comptroller General, O.S. General

Accounting Office.

41Statement of Edward J. Hanley, Director, Office of Information Resources Management, U.S. Environmental Protection Agency, before a hearing

of the Subcommittee on Government Information, Justice, and Agriculture, House Committee on Government Operations, Apr. 18, 1989.

4 Q


project involves the U.S. Department of "EnergyData Base" (EDB). DOE has proposed to providedepository libraries with online access to the EDB.49(See the appendix for further discussion of the EDBpilot project.)

In addition, GPO is seeking suggestions fromprivate vendors on how they might participate inelectronic dissemination to depository libraries.Industry interest appears to be high. Finally, Federalagencies also seem generally supportive of anelectronic role for the DLP, but have unansweredquestions and concerns about selection procedures,financing, and user support for electronic formatitems included in the DLP."

The remaining DLP issues concern cost andfinancing, especially for online dissemination. CD-ROM and offline formats are gaining acceptance ascost-effective alternatives to paper and microfiche.It is likely that most depository libraries would selectonly a relatively small portion of total FederalSTIas is the case with Federal information ingeneraland would more typically refer users toSTI data centers and existing archives. This wouldrequire that depository libraries have efficient accessto directories, indices, and bibliographies of theFederal STI, rather than to the STI itself.

Alternatively, a small number of depositorylibraries could be designated as STI depositories.These libraries would include a large amount of STIin their collections, and would serve as a sharedresource for local and regional libraries. STI deposi-tories could be strategically located in areas ofconcentrated scientific and technical activity wherethe local community is committed to building itsR&D base. STI depositories would have to have the

technical capabilities to use all electronic formatsonline, CD-ROM, and diskette. This possibilitycould be explored in depth as part of an overallreexamination of the DLP.

Funds for STI dissemination at depositories couldcome from suveral sources, with a portion fundedthrough the DLP direct appropriation, a portion bythe Federal science agencies (e.g., for free copies ofselected agency CD-ROMs and fe reductions orwaivers for online access to selected agency data-bases), a portion by the depository libraries (e.g., formicrocomputers, CD-ROM readers, and modems),and a part by the library users (e.g., for telecommuni-cation line charges). The libraries could have discre-tion over how the appropriated funds are spent. Forexample, libraries could be issued vouchers foraccess to online STI bibliographic databases. Thesefunds could be expended on a mix of government,commercial, and not-for-profit databases, dependingon user needs. Vouchers might also be used forlibrary purchase of equipment needed to supportelectronic dissemination, and for subsidy cf tele-communication or electronic printing charges in-curred by students or others with limited means.51

Overall, an estimated 9 to 10 million persons usedepository libraries each year. Academic librariesrepresent about 55 percent of all depository libraries.Students and faculty account for 85 percent ofacademic library users. Students and professional,

. technical, and managerial persons together represtatabout 77 percent of public depository library users.52Thus, depository users are likely to be a readymarket for Federal STI in electronic formats, andopen to technical and institutional innovation ininformation dissemination. The electronic pilot proj-ects will shed more light on this prospect.53

49U.S. Congress, Joint Committee on Printing, "Dissemination of Information in Electronic Format to Federal Depository Libraries: Proposed ProjectDescriptions," Arne 1988.

50See statement of Forrest B. Williams, Branch Chief, Data User Services Division, U.S. Bureau of the Census, before a July 13, 1989, hearing ofthe National Commission on Libraries and Information Science.

51Representatives of library associations are concerned about proposals for sharing costs of online or other electronic dissemination. Depositorylibraries already spend sdveral dollars (in building, equipment, and staff costs) for every dollar spent by the Federal Government on documents, andoppose shifting more costs of dissemination to the libraries. See statenrnts of Cheryl Rae Nyberg, American Association of Law Libraries, MerrilyThylor, Association of Research Libraries, and Katherine F. Mawdsley, American Library Association, before a Mar. 8, 1990, hearing of the Committeeon House Administration, Subcommittee on Procurement and Printing.

52C.R. McClure and P. Hernon. Users of Academic and Public GPO Depository Libraries (Washington. DC: U.S. Government Printing Office, 1989).

53For further discussion of depository library alternatives, see U.S. Congress, Office of Technology Assessment, Informing the Nation. op. cit.,footnote 17, ch. 7; and Association of Research Libraries, Technology and U.S. Government Information Policies: Catalysts for New Partnerships(Washington, DC: October 1987). Also see statements of D. Kaye Gapen before the House Committee on Administration and House Committee onGovernment Operations, op. cit., footnote 41; and statements of Vicki W. Phillips, Chair, Depository Library Council to the Public Printer, Patricia GlassSc1.7.man, President, NezI-Schuman Publishers, Inc. (on behalf of the American Library Association), and Bruce M. Kennedy, Head, ReferenceDepartment, Georgetown University Law Center (on behalf of the American Association of Law Libraries) before a July 13, 1989, hearing of the NationalCommission on Libraries and Information Science.

5 0

Chapter 5

Framework for a Presidential Initiative onScientific and Technical Information

An important part of a strategy for scientific andtechnical information (ST1) is leadershipleadershipfrom the science and technology community, Con-gress, Federal science agencies, and the ExecutiveOffice of the President, including the Office ofManagement and Budget (OMB) and the Office ofScience and Technology Policy (OSTP). Leadershipis necessary to reach a workable consensus on theoutstanding issues of ST1 dissemination and accessdiscussed in chapters 3 and 4. Leadership is neededto improve interagency coordination and agencyorganization for STI. How can this leadership beprovided? The key is presidential leadership on STI.This can be done in several ways:

1. strengthen the OSTP role;2. establish an OSTP advisory committee and an

interagency coordinating committee on Fed-eral STI;

3. redefme OSTP-OMB working relationshipsregarding STI;

4. upgrade STI dissemination functions withinagency R&D and Inform lion Resources Man-agement programs;

5. develop technical standards and directories forSTI dissemination;

6. launch an STI education initiative; and7. improve international STI exchange programs.

Strengthening the OSTP Role

Congress intended that OSTP be the focal pointfor STI leadership in the executive branch, and thatthe OSTP Director (who serves as the President'sScience Advisor) designate STI as a priority concernof OSTP.1 The "National Science and TechnologyPolicy, Organization, and Priorities Act of 1976,"2OSTP'S organic statute, addresses STI in the decla-ration of congressional policy. Congress was con-cerned that ST1 had received little attention.3 TheAct recognizes that "effective management anddissemination of scientific and technological infor-mation" is part of the U.S. science and technologybase. It states that "Federal departments, agencies,and instrumentalities should establish procedures toensure among them the systematic interchange ofscientific data and technological fmdings developedunder their programs."4 The legislative intent was toinclude STI in the OSTP mission implicitly.5 STI ismentioned in the charter of a President's Committeeon Science and Technology that was to consider,among other things, "improvements in existingsystems for handling scientific and technical infor-mation on a governmentwide basis, including con-sideration of the appropriate role to be played by theprivate sector in the dissemination of such informa-

1For a discussion of legislative history and options, see U.S. Congress, House, Committee on Science and Technology, Subcommittee on Science,Research, and Technology, Optimizing the Value of U.S. Scientific and Technical Information: Legislative Options, report prepared by the CongressionalResearch Service (Washington, DC: October 1978). For general discussion of science advice in the White House, see WG. Wells, Jr,, "Science Advice

and the Presidency, 1933-1976," dissertation, School of Government and Business Administration, The George Washington University, Washington,

DC, 1977; W.T. Golden (ed.), Science Advice to the Pres-Vent (New York: Pergamon Press, 1980); WT Golden (ed.), Science and Technology Advice

to the President, Congress, and Judiciary (New York: Pergamon Press, 1988); G./. Knew, "Suggestions for Collection of Archival InformationPertaining to Presidential Science Advisory Bodies Before 1976," memorandum, Congressional Research Service, Nov. 15, 1989; and statements ofFred B. Wood, OTA; Joseph G. Coyne, U.S. Department of Energy; and Castles R. McClure, Syracuse University, before an Oct. 12, 1989, hearingof the House Committee on Science, Space, and Technology, Subcommittee on Science, Resench, and Technology.

2U.S. Congress, PubF Law 94-282, May 11, 1976.

3According to most observers, the peak of White House interest in occurred in the Kennedy and Johnson Administrations, during which timepresidential science advisory bodies issued several landmark studies on Federal STI. See, for example, J.H. Crawford, Jr., G. Abdian, W. Frazer, S.

Passman, RB. Stegmaier, Jr., and I. Stern, Scientific and Technical Commwtications in Government, Task Force Report to the President's SpecialAssistant for Science and Technology (Washington, DC: U.S. Department of Commerce, April 1962); and Federal Council for Science and Technology,

Committee on Scientific and Technical Information, Recommendations for National Document Handling Systems in Science and Technology

(Washington, DC: U.S. Department of Commerce, November 1965).

4Public Law 94-282, sec. 102(a)(5)(c) and sec. 102(cX10).

5Earlier legislative proposals addressed STI in more devil. The House committee reports made clear that STI was to have a high priority. See, for

example, U.S. Congress, House, Committee on Science and Technology, National Science and Technology Policy and Organization Act of 1975,Report

No. 94-595, 94th Cong., 1st sess. (Washington, DC: U.S. Govermnent Printing Office, Oct. 29, 1975).

5


The low profile of OSTP with respect togovernmentwide ST! policy has, in ef-fect, ceded the dominant executivebranch policy role to the Office ofManagement and Budget. .11:

tion."6 This provision of the law has not beenimplemented.7

DSTP has provided a modicum of staff attentionto STI matters, and has encouraged the FederalCoordinating Council for Science, Engineering, andTechnology (FCCSET) in some STI matters. TheCouncil, established under Title IV of the Act, ismade up of the OSTP Director (chairman), andrepresentatives of the Federal science and technol-ogy agencies. The Council created the Committee onEarth Sciences, which has endorsed the work of theInteragency Working Group on Data Managementfor Global Change. This working group is address-ing some of the STI technical and policy issues asthey relate ) earth sciences and global change.FCCSET also has supported work in the areas ofhigh-performance computing and networking, whichrelate to STI dissemination.8 But, neither OSTP northe FCCSET has given much attention to thedissemination of STI documents or bibliographicdatabases, to issues involving agencies like theNational Technical Information Service (NTIS) andGovernment Printing Office (GPO) that are respon-sible for disseminating such materials, or to govern-mentwide information dissemination issues thatrelate to STI.

The low profile of OSTP with respect to govern-mentwide STI policy has, in effect, ceded thedominant executive branch policy role to the Officeof Management and Budget. OMB has used its

authority under the Paperwork Reduction Act topromulgate governmentwide information policy thatcovers STI as well as most other types of Federalinformation (see ch. 3). OSTP has only minimallyused its authority under the National Science andTechnology Policy Act to get involved in STIpolicy. Thus the activities of OMBnot OSTPhave had by far the largest impact on STI, andespecially on dissemination.

A strengthened OSTP role would help ensure thatthe special needs and problems of STI are consid-ered, and that the contribution of STI to broadernational goals is identified and realized. A strongerrole should also improve interagency coordinationon STI. The OSTP director may, on his owninitiative, give a higher priority to STI matters. Thismight involve the assignment of OSTP staff to STIissues, and the formal recognition of STI functionswithin each of the major OSTP programmatic areas.But even so, Congress could seriously consideramending the law to provide stronger congressionalguidance. This could be done by adding STI as anexplicit, required area of OSTP responsibility and toFCCSET's charter, and perhaps by authorizingOSTP funds specifically for STI activities.9

OSTP could prepare and issue a strategic plan onSTI, with the advice and assistance of advisorycommittees and agency officials, as was done inhigh-performance computing. This recently issuedcomputing plan° states the goals, rationale, actions,responsibilities, and budget for implementing theU.S. high-performance computing and networkingprogram. Program leadership is assigned to OSTP,assisted by an FCCSET Committee on ComputerResearch and Applications and an advisory panelselected by and reporting to the OSTP Director. TheFCCSET Committee is responsible for interagencyplanning and coordination, technology assessment,and preparation of policy recommendations andaimual progress reports to OSTP. The advisory panel

6Public Law 94-282, sec. 303(a)(2).

7For a general review of OSTP performance, see G.J. Knew, Analysis of the Office of Science and Technology Policy, CRS Report No. 88-205 SPR

(Washington, DC: Congressional Research Service. February 1988) and White House Office of ScienceandTechnalogy Policy: An Analysis, CR.S Report

No. 89-689 SPR (Washington, DC: Congressional Research Service, November 1989).

5S ee U.S. Office of Science and Technology Policy, Executive Office of the President, A Research and Development Strategy for High Performance

Computing, Committee on Computer Research and Applications, Federal Coordinating Council on Science. Engineering, and Technology (Washington,

DC: Executive Office of the President, Nov. 20, 1987): and U.S. Office of Scienceand Technology Policy. Executive Office of the President, The Federal

High Performance Computing Program (Washington, DC: Executive Office of the President, Sept. 8, 1989).

9Congress is consideriog this approach for high-performance computing. S. 1067, the "Higb-Performariee Computing Mt of 1990," would mandate

and authorize funding for OSTP and FCCSET activities in this area.

loU.S. OSTP, High Petformanee Computing Program, 1989, op. cit., footnote 8.r114.0

Chapter 5Framework for a Presidential Initiative on Scientific and Technical Information 45

Strategically, the most important STIrole for OSTP may be its visible leader-ship on STI issues coupled with theassessment of STI issues from an inte-grated cross-cut perspective across agen-cies and disciplines.

4111111111111111111"

will include scientific, academic, and industry ex-perts, and will provide the OSTP Director andFCCSET with independent assessments of programprogress, relevance, and balance, A similar organiz-ational approach could be used for "a FederalProgram."

An STI strategic plan could, even in its earlystages, serve as a focal point for involving OSTP inthe ongoing legislative efforts to amend the Paper-work Reduction Act, Printing Act, DepositoryLibrary Act, and other statutes lat affect FederalSTI. OSTP leadership could help develop a strategicvision of: 1) the role of the Federal R&D agencies,NTIS, and GPO in STI dissemination; 2) principlesof STI dissemination that encourage use of Federal511; and 3) updated policies on the open flow ofFederal STI that reflect rapidly changing globaleconomic, political, and technological realities.

An STI strategic plan also could integrate ST1activities across the several existing Federal policiesand programs to encourage technology transfer andindustrial innovation. These include, for example:"

the Federal Laboratory Consortium for Tech-nology Transfer, in which about 300 Federallabs participate, that promotes utilization oftechnical knowledge developed by or for Fed-eral labs;

e The Offices of Research and Technology Ap-plications, located at each Federal lab, thatidentify technologies and ideas with potentialoutside application; and

the Small Business Innovation Developmentprogram, that encourages technology develop-ment by small companies, including use offederally developed technology and STI incommercial applications.

Other programs encourage a variety of joint venturesand cooperative R&D agreements between theFederal Government, universities, and/or privateindustry. The proliferation of technology transferactivities has made the need for an STI cross-cuteven greater.

Strategically, the most important STI role forOSTP may be its visible leadership on STI issuescoupled with the assessment of STI issues from anintegrated cross-cut perspective across agencies anddisciplines. OSTP leadership would require itscollaboration with various STI constituenciesinthe science agencies, in Congress, in academia andprivate industryfor ideas, feedback, and dialog.The last time this happened on STI was in the1960s.12

The National Science Foundation (NSF) sup-ported several STI studies during the 1980s thatidentified 511 problems and possible policy solu-tions.13 A. that time, OSTP lacked the interest, staff,and high-level support to followup on the STIstudies of NSF and other groups. OSTP could notitself perform major STI policy research for lack ofresources. But its active involvement can go a longway toward supporting the efforts of others. OSTPcertainly can be expected to conduct policy planning

"For a general overview, see W.H. Schacht, Technology Transfer: Utilization of Federally Funded Research and Development, 1B 85031, and

Industrial Innovation: Debate Over Government Policy, 113 84004 (Washington, DC: C.. ingressional Research Service, Aug. 7, 1989).

12Presentation of A.A. Aines, former Acting Chairman, Committee on Scientific and Technical Information, White House Office of Science and

Technology, at a CENDI meeting, Dec. 12, 1989.

"NSF-sponsored studies include: A.H. Teich and J.P. Wei:A/erg, Issues in Sciennfic and Technical Information Policy (Washington, DC: American

Association for the Advancement of Science, Dm. 28, 1982); TX. Bikson, B.E. Quint, and L.L. Johnson, Scientific a nd Technical Information Transfer:

Issues and Options (SantaMonica, CA: Rand Corp., March 1984); S. Ballard, C.R. McClure, TI. Adams, M.D. Devine, L. Ellison, TE. James, Jr., L.L.

Malys& and M. Meo, Improving the Transfer and Use of Scientific and Technical Information: The Federal Role (Norman, OK: Science and Public

Policy Program, University of Oklahoma, September 1986); J.D. Eveland, Scientific and Technical Information Exchange: Issues and Findings(Washington, DC: Divisitn of Policy Research and Analysis, NSF, March 1987); NSF Divisionof Policy Research and Analysis,Scientific Information

Exchange: A Status Report on Converting New Fundamental Knowledge Into Competitive Products (Washington, DC: NSF, April 1987); and NSF

Division of Policy Research and Analysis, Federal Technology Transfer: kleti(atlirms and Agency Practices (Washington, DC: NSF, May 1987).t t)

46 Helping America Compete: The Role of Federal Sciennfic and Technical Information

and assessments based on the best available STI*research."

The extensive, multi-year debate leading up to theestablishment of OSTP in May 1976 reflected astrong consensus among leading scientists andengineers on the importance of these OSTP respon-sibilities to STI.15 This was followed 1976-77 bya vigorous debate over OSTP'e ;unctions in theCarter Administration. Few of che numerous innova-tive proposals brought forward16 were implementeddue to President Carter's decision to downsize theentire Executive Office of the President, includingOSTP.17 The Bush Administration (and the appoint-ment of Dr. D. Allan Bromley as the Director ofOSTP) is the first real opportmity in 12 years forOSTP to carry out the congressional intent ofOSTP's organic Act and fulfill the vision of thescientific and technical communityincluding aleadership role in Federal STI.

Establishing Advisory Committees on STI

The success of the Committee on Scientific andTechnical Information (COSATI) is frequently citedas evidence of the potential effectiveness of high-level advisory bodies. COSATI was formed in 1963by the former Office of Science and Technology(created in 1962 by executive order) and its Presi-dent's Science Advisory Committee (PSAC). COSATIand PSAC provided high-level executive branchleadership on STI.18 With a change of administra-tions, COSATI was transferred from the Office ofScience and Technology to NSF in 1971 andabolished in 1972. The Office of Science and

OSTP could use FCCSET to help agencySTI managers get higher priority forinformation dissemination and utiliza-tion as part of agency R&D programsthat collect or create the STI.

Technology itself was abolished in 1973.19 OSTPwas established by statute in 1976. The new OSTPDirector has recently created a President's Councilof Advisors on Science and Technologythe equiv-alent of PSACunder the President's statutoryauthority. Functions of the new President's Councilof Advisors could be extended to STI and thecreation of advisory subgroups such as COSATI.

Two STI advisory bodies are justified. A COSATIof advisors and experts could report to the OSTPDirector. This group might include representativesfrom major segments of the !zience and technologycommunity concerned with STI: scientists, scholars,information specialists, large and small businessleaders, librarians, State/local government officials,consumer and labor leaders, and the like. A secondadvisory body comprised entirely of agency STIofficials could be established under FCCSET. Thisgroup could include representatives from a cross-section of Federal science agencies, including themajor Federal science data centers and documentclearinghouses, and the governmentwide dissemina-tion and archival agencies.

14Both the outreach and policy assessment roles of OSTP arc addressed at a general level throughout the enabling statute and legislative history. SeePublic Law 94.282, op. cit., footnote 2; U.S. Congress, House, Committee on Science and Technology, National Science and Technology Policy andOrganization Act of 1975, Report, 94th Cong., 1st sess., Rep. No. 94-595 (Washington, DC: U.S. Government Printing Office, Oct. 29, 1975); U.S.Congress, House, Committee on Science and Teclmology, Science and Technology Policy, Conference Report, 94th Cong., 2d sess., Rep. No. 94-1446(Washington, DC: U.S. Government Printing Office, Apr. 26, 1976).

15See, for example, U.S. National Academy of Sciences, Science and Technology in Presidential Policymaking: A Proposal (Washington, DC:National Acadetny Press, June 1974).

165ee, for example, statements of Lewis M. Branscomb, "Science and Technology Issues: A Framework," June 14, 1976; Harold Brown, "Scienceand Technology Organization iu the Executive Office of the President," Aug. 23, 1976; and F.B. Wood, V. Coatcs, J. Coates, R. Ericson, and J. Logsdon,"Early Warning and Policy Assessment Capability lb Support Presidential Policymaking/Decisionmaking," Jan. 3, 1977; prepared for the Jimmy CarterPresidential Transition Team.

170STP was reduced to a minimal staff level of about 15 persons. However, it could have been worse. For example, the White House Office ofTelecommunications Policy was abolished, and its functions transferred to the Departments of Cominerre and State and the Federal CommunicationsCommission.

leSee, for example, President's Advisory Committee, Science, Government, and Information: The Responsibilities of the Technical Community andthe Government in the Transfer of Information (Washington, DC: U.S. Govermnent Printing Office, Jan. 10, 1963),

15Ibotnas E. Pinch. "Chronology of Selected Reports, Related Studies, and Significant Events Concerning Scientific and Technical Information inthe United States," May 1989 draft. For other historical perspectives, see A. Bishop and M.O. Fellows, "Descriptive Analysis of Major Federal Scientificand Technical Information Policy Studies," in C.R. McClure and P. Hernon, United States Scientific and Technical Information Policies: Views andPerspeaives (Norwood, NJ: Ablex Publishing Corp., 1989), pp. 3-55; and A.A. Aims, "A Visit to the Wasteland of Federal Scientific and InformationPolicy," Journal of the American Society of Information Science, vol. 35, May 1984, pp. 179-184.

P"' A


OSTP could ensure that Federal scienceagencies have a role in the STI poli-cymaking process at OMB. OSTP couldcollaborate with OMB on major initia-tives to improve the management ofFederal information systems, includingagency STI systems.

MEMMEN110111

The lack of an equivalent to COSATI, or a formalFCCSET advisory body on STL in part led to thecreation of CENDI (Commerce, Energy, NASA,NLM, Defense Information). CENDI is an intera-gency group established by several Federal scienceagencies (NTIS, DOE, NASA, DTIC, and NLM) toaddress STI issues. The CENDI agencies representover 90 percent of the Federal R&D budget. CENDIsupports a strong OSTP and FCCSET role in STI.

Compared to CENIX, an FCCSET committee onSTI could be upgraded in several ways. First, itsscope could be expanded to include the data side ofSTI as well as the bibliographic and document sideon which CENDI now ..oncentrates. Second, theFCCSEr committee's membership could be ex-panded to include other Federal agencies with majorST1 functions (e.g., USGS, NOAA, USDA, ar.EPA) that are not presently included in CENDI.Third. staff support and funding could be expandedbeyond that now available to CENDI. CENDI hasundertaken several new projects in the areas ofstandards, cataloging, indexing, and technologyassessment, but has no regular means of support(participating agencies make voluntary contribu-tions). Fourth, the FCCSET committee could assertleadership in educating government executives onthe importance of STI dissemination and govern-mentwide STI strategies, in a much more vigorousmanner than appears possible through CENDI. Fifth,the FCCSET committee could establish strongworking relationships with other interagencygroups.

Improved coordination is urgently needed amongthe interagency groups involved in Federal STI,including:

CENDI;Interagency Working Group on Data Manage-ment for Global Change;

Interagency Coordinating Committee on Digi-tal Cartography;Special Interest Group on CD-ROM Applica-tions and Technology;Federal Publishers Committee;Interagency Panel on Numerical Data;Interagency Advisory Council on Printing andPubli shing ; andFederal Library and Information Center Com-mittee.

OSTP could take a leadership role on an FCCSETsn committee, to help further offset the naturaltendency of all interagency groups to reflect agency-specific rather than governmentwide concerns. OSTPalso could help ensure, through FCCSET, that thevarious interagency groups have adequate adminis-trative and financial support, balanced membership,and an audience for the fruit ;f their labors. If theFCCSET committee is effective, some of the otherinteragency groups may no longer be needed.

OSTP could use FCCSET to help agency STImanagers get higher priority for information dissem-ination and utilization as part of agency R&Dprograms that collect or create the STI. R&Dmanagers have a strong tendency to emphasize theconduct of the research itself, rather than theeffective Jse of research results. OSTP could workwith FCCSET to help individual agency STI pro-grams contribute to governmentwide priorities, suchas the global change program.

Redefining OSTP-OMB WorkingRelationships on ST1

OMB has a dominant role in executive branchinformation policy and oversight. The OMB Officeof Information and Regulatory Affairs has devotedlittle attention specifically to sn, and some withinOMB strongly support a reactivated STI role forOSTP. But, even if OSTP gives priority to STI,OMB will continue to be a major player for tworeasons: first, OMB guidance on general govern-ment information policy will also apply to STI (e.g.,Circular A-130), unless 511 is granted a blanketatmption, an unlikely prospect; second, OMB willstill be the primary decisionmaker on budgets forFederal science agencies--including resources allo-cated to STI.

A new OSTP-OMB working relationship onFederal STI is necessary. OSTP could activelyparticipate in the drafting and public comment

48 Helping America Compete: The Role of Feder l Sciennyk, and l'echnical I ormation

process for revisions to OMB Circular A-130 andother circulars that affect STI. OSTP could ensurethat Federal science agencies have a role in the STIpolicymaking process at OMB. OSTP could collab-orate with OMB on major initiatives to improve themanagement of Federal information systems, in-cluding agency STI systems. Many criticisms ofFederal information systems apply to STI as well.Federal agencies have been criticized for not payingenough attention to the users of Federal informationand involving users from the outset of projectplanning. OSTP and OMB could encourage useroutreach activities and provide guidance to theagencies on how to hnprove outreach.

OMB issues an annual bulletin on "Federalinformation systems and technology planning" ,thatdirects agencies in the preparation of strategic plans.These are developed as part of agency and govern-mentwide 5-year plans. OSTP could suggest topicsfor special attention. In 1988, OMB asked agenciesto provide details on electronic mapping databases(otherwise lmown as digital cartographic, geo-graphic, or land information systems). In 1989,OMB asked agencies to provide information onimage processing systems and electronic data inter-change.20 These topics all relate to STI. Otherpossible STI-related topics include: high-densitydata storage systems; expert systems for informationretrieval; machine translation (of foreign languagepublications); and gateway technologies for multi-ple remote database access.

OSTP and OMB could help ensure that eachFederal science agency is aware of and carefullyexamines state-of-the-art activities of other agen-cies. For example, the Defense Technical Informa-tion Center (DTIC) prepared a year 2000 strategicplan and is implementing it.21 DTIC is the clearing-house for STI developed by or for the Department ofDefense (DoD). DTIC operates: an online researchdatabase (DROLS = Defense Research On-Line

Search); an intelligent gateway to DoD and someother online databases (that eventually will beextended to many Federal agency and commercialdatabases); and a prototype electronic documentsyst (that uses scanners, optical disks, supermi-crocomputers, intelligent work stations, and l' serprinters for storing and disseminating DoD technicaldocuments).

Other OSTP-OMB joint activities might include:

cosponsorship of ad hoc interagency commit-tees on specific priority topics, such as im-proved indexing of Federal STI and other typesof information (OMB already has proposed acommittee on this topic), management of verylarge databases (e.g., the Earth ObservingSystem), and quality control of standard refer-ence data (on physical, chemical, and engineer-ing properties).cosponsorship of a continuing dialogthroughmeetings, committees, conferences, and othermeansbetween agency R&D and STI manag-ers to ensure that the Federal investment in STIbest serves the R&D user community;coordination on appointments to any OSTP andOMB outside advisory committees that may beestablished on STI or Federal information;cofunding, directly or with agency support, ofresearch projects in targeted cross-cut areassuch as user training and STI education;OSTP participation in 0MB-sponsored intera-gency groups (e.g., the Interagency Coordinat-ing Committee on Digital Cartography22) andvice versa; andcosponsorship of conferences that bring to-gether all elements of the STI community, fromagencies to libraries to vendors.

The reentry of OSTP into STI activities wouldopen new possibilities for cooperation with OMB in

20U.S. Office of Management and Budget, Bulletin 89-17, "Federal Information Systems and Technology Planning," Aug. 22, 1989.

21For the original plan, now being upda:ed, see U.S. Department of Defense, Defense Logistics Agency, Defense Technical Information Center, D77C

2000 : A Corporate Plan for the Future, DTIC/TR-8413, July 1984. Also see, for example, T. Lahr and D. O'Connor, An Evaluation of D77C' s Prototype

CD-ROM (Alexandria, VA: Defense Technical Information Center, August 1989); C.W. Shockley, D.F. Egan, CH. Groth, Jr., and Di. O'Connor,Meeting the Scientific and Technical Information Challenge, Report DL605R2, contractor report prepared for DTIC (Bethesda, MD: LogisticsManagement Institute, October 1988); Aerospace StrUCtUrtil Information Analysis Center, Application of New Technologies to D77C Document

Processing, contractor report prepared for DT1C, August 1987; and GA. Cotter, The DOD Gateway Information System: Prototype Experience,

DTIC(TR-86/6, April 1986.

22Tbis committee was rechartered by OMB in 1989. See memorandum from Richard G. Darman, OMB Director, to Headsof Executive Departmentsand Independent Establishments, "Coordinanon of Federal Digital Cartographic Data Program," Feb. 28, 1989; also see memorandum from Lowell

E. Starr, Chairman, to Participants, Federal Interagency Coordinating Committee on Digital Cartography Governmentwide Forum, "FCCDCRecommendations for an Improved Federal Spatial Coordination Process," Dec. 5, 1989.


Technical standards can bridge amongdifferent formats so thpt once the infor-mation is in the system, it can beprocessed, edited, revised, stored, anddisseminated in electronic, paper, ormicrofiche formats.

jointly carrying out executive branch STI poli-cymakiug and oversight.

Upgrading Agency ST1 Management

Agency management of STI needs to be strength-ened, and .OSTP-OMB cooperation could help.23Information dissemination should have a higherpriority. Most agencies give scant attention todissemination, even though dissemination was in-cluded in the original Information Resources Man-agement (IRM) program concept, and is referred toin the Paperwork Reduction Act (as amended in1986). IRM officials and activities are mostlyoccupied with computers, telecommunications, man-agement information systems, and procurementactivities. Job definitions, career paths, and trainingprograms for information dissemination profession-als and IRM officials could be revised and strength-ened to reflect the importance of STI.

STI dissemination should have higher prioritywithin agency R&D programs as well. STI is theprimary product of R&D and is central to agencyR&D missions. Several possible actions to upgradeSTI deserve consideration:

the direct participation of STI staff in agencyR&D planning and decisionmaking;the assignment of technical information offi-cers to major science agency operating units;the separeon of dissemination as a line itemwithin agency R&D budgets;the allocation of at least some minimumpercentage of R&D grants, contracts, and

operating budgets to STI dissemination, datamanagement, and related areas;the participation of R&D program officials inselected interagency STI groups and activities;the participation of R&D grantees, contractors,and the like in agency innovation centersdesigned to share new information about STIdissemination, among other topics;.he involvement of R&D and STI managers infocus group discussions with and surveys ofSTI users; andthe joint sponsorship of independent researchon STI dissemination and use (perhaps withcooperation from NSF).

Further research on STI use needs to emphasizethe barriers as well as opportunities presented byelectronic formats. For example, what conditionsequipment, software, training, experiencecontribute to successful use of electronic formats? Isthe research on use of online formats applicable tooffline formats like compact optical disk? Howeffective are end users in conducting their ownsearches of STI databases compared with usingintermediaries (e.g., librarians, commercial ven-dors)? Is existing search-and-retrieval software suf-ficiently user-friendly to make widespread, decen-tralized use a reality? Are users able to atiapt to theavailability of STI in multiple and changing for-mats? In sum, agencies need to guard against"technophoria."24 While electronic formats arewell-suited to STI, disseminating agencies shouldnot adopt electronic formats uncritically without agood understanding of the impact on STI users.

Developing Standards and Directories for STI

Technical standards are essential if the govern-ment is to make improvements in cost-effectivenessand productivity and assist the private sector to useFederal STI. Technical standards can bridge amongdifferent formats so that once the information is inthe system, it can be processed, edited, revised,stored, and disseminated in electronic, paper, ormicrofiche formats. Standards developed for FederalSTI should be compatible with those adopted by theprivate sector and the international standards-setting

23For a general critique of agency infoimation management as it relates to ST1, see C.R. McClure, A. Bishop, and P. Doty, "Federal Scientific andTechnical Information (STD Policies and the Management of Information Technology for Dissemination of ST1," in Information Technology: Planningfor the Second 50 Years, Proceedings of the 51st Annual Meeting of the American Society for Information Science, Christine L. Borgman and EdwardY.H. Pai (eds.) (Medford, NJ. Learned Information Press, 1989). Also see U.S. Congress, Office of Technology Assessment, Informing the Nation:Federal Dissemination in an Electronic Age, OTA-CIT-396 (Washington, DC: U.S. (iovernment Printing Office, October 1988).

24Terin coined by C.R. McClure, Syracuse University School of Information Studies.#A,


organizations. Priority areas for standards-settinginclude;

STI indexing and cataloging (standard formatsare needed, so that NTIS, GPO, and missionagencies are using compatible approaches);STI quality control (especially for preventingor minimizing errors in collecting data andcreating documents, and for maintaining dataand document integrity throughout the infor-mation life cycle);STI security (technical and administrative stan-dards for preventing unauthorized use or altera-tion of Federal STI);text markup and page/document descriptionlanguages (e.g., Standard Generalized MarkupLanguage, which has been issued as an interna-tional standard and as a Federal InformationProcessing Standard (FIPS));optical disks (there has been significant pro-gress on CD-ROM standards, e.g., for master-ing, formatting, and reading, but not yet forsearch and retrieval software; standards forWORM, Erasable, and CD-I disks are in earlierstages of development); andelectronic data interchange (EDI), including theopen systems interface (OSI) concept (e.g., anOSI procurement standard has been issued as aFlPS and becomes mandatory in late 1990; aproposed EDI standard has been issued forcomment).

STI managers, users, and private vendors gener-ally agree on the need for interoperability amongvarious systems and equipment. The Federal Gov-ernment can accelerate the development and adop-tion of the standards needed to ensure interoperabil-ity. The National Institute of Standards and Technol-ogy (NIST), working with GPO, NTIS, and theFederal science agencies, could help in this standards-

setting effort. DoD is important in this process,because it and the defense industry together accountfor two-thirds of the Federal R&D budget and haveinvested hundreds of millions of dollars in CALS(Computer-Aided Acquisition and Logistical Sup-port). CALS is designed as a standardized system forthe electronic exchange of technical data, drawings,and documents.

Large STI databasessuch as in the geographic,spade, and earth sciencesmust have technicalstandards for data archiving and exchange, if theseresources are to be managed and used effectively.Geographic information systems (GIS) will permitgreater data exchange among the Federal scienceagencies. CIS require the integration of multipledata setsfrequently originating from several dif-ferent agencies. Most Federal agencies with GISapplications are using data sets from several otheragencies.25 GIS must have standards to ensureinteroperability among users in these agencies. Mostagencies using GIS have not yet developed standarddefinitions and/or classifications for the majorthematic data categories used in GIS applicationsand do not have an operational program to collectand manage standardized data.26 The OMB-chartered Federal Interagency Coordinating Com-mittee on Digital Cartogaphy (chaired by the U.S.Geological Survey) has made progress in developinga standard format for Federal geographic informa-tion storage and exchange.27

NASA is active in standards for space sciencedata. The Science Data Systems Standards Office (atNASA's National Space Science Data Center (NSSDC))is responsible for standards development. It workswith the national and international standards organi-zations, validates standards, and disseminates infor-mation about standards that are important to spacescience data collection, storage, and dissemination.

23U.S. Interagency Coordinating Committee on Digital Cartography, Reports Working Group, "A Summary of GIS Activities in the FederalGovernment," August 1988, pp. 16-18.

26Ibid., pp. 13-15.

27See, for example, U.S. Federal Interagency Coordinating Committee on Digital Cartography, Standards Working Group, "Federal GeographicExchange Format: A Standard Format for the Exchange of Spatial Data Among Federal Agencies," Dec. 15, 1986, U.S. Interagency CoordinatingCommittee, "Coordination of Digital Cartographic Activities in the Federal Government," Third Annual Report to the OMB Director, 1988. For

discussion of the need for a directory to GIS activities and improved Federal/State/local cooperation on GIS, see Lisa Warnecke, "Georgraphic/LandInformation Development Coordination Clearinghouse and Network," Syracuse University, School of Information Studies, January 1989, and"Geographic Information Coordination in the Stater Past Efforts, Lessons Learned, and Future Opportunities," in Piecing the Puzzle Together: AConference on Integrated Data for Decisionmaking, proceedings, National Governors Association, Center for Policy Research, May 27-29, 1987. For

recent updates on GIS standards and related topics, see U.S. Department of the Interior, Studyof Land Information, prepared in accordance with Public

Law 100-409, November 1989 draf4 Dec. 5, 1989, memo from Lowell E. Starr, U.S. Geological Survey, on "FICCDC Recommendations for anImproved FAeral Spatial Data Coordination Process," and agency responses thereto. For general background on the Interagency CoordinatingCommittee, see Memorandum from Richard O. Darman, Director, Office of Management and Budget, to Heads of Executive Departments,Establishments, and Independent Agencies, "Coordination of Federal Digital Cartographic Data Programs," Feb. 28, 1989.


The NSSDC has a generic data storage standard,known as the Common Data Format, that is beingbeta-tested by NASA laboratories and others.28

The standards-setting effort in the earth sciencesis being led by the Interagency Working Group onData Managenrent for Global Change, whose mem-bers include N, .3A, NOAA, NSF, USGS, the U.S.Navy, and the Departments of Energy, Agriculture,and State. The working group has emphasizedtechnical standards to facilitate the exchange of datadirectory information and data sets. Standards areneeded to enable users to access earth sciences dataon a variety of computers, over a range of electronicnetworks. This includes the need for standards ondata quality. The working group has involved NISTin its standards-se ling activities. The NationalResearch Council's Scientific and Technical Infor-mation Board (formerly the Numerical Data Advi-sory Board) also emphasizes the role of NIST indeveloping governmentwide standards for a varietyof large-scale scientific and technical databases.

Directories to Federal STI are also needed to helpusers find the information they seek. Some areconcerned that a directory or index might be used byOMB to thwart rather than encourage agencyinformation dissemination. But OMB has takensteps to quiet this concern. Under the OMB plan,each agency would maintain a current, comprehen-sive inventory of information dissemination prod-ucts and services, including: periodicals, nonrecur-ring publications, machine-readable datafiles (in-cluding compact optical disks), software, onlinedatabases, and electronic bulletin boards. Eachinventory would serve as an index to agencyinfonnation and would be submitted to a centralcollection point and compiled into a governmen-twide index." NTIS and GPO could collaborate onpreparation of a governmentwide directory, and startby collecting and consolidating available agency-specific directories. OMB intends to establish aninteragency group to develop an improved structureand content for agency inventories.

Directories to large-scale, scientific databases aswell as STI d mments should be included in theseefforts. The proliferation of space science electronicdatabasesoffline and onlineis an example of theimportance of directories to users seeking specificinformation. NASA's Master Directory offers onlineaccess to a directory of NASA and other space andearth science data sets and related information. Foreach data set, the directory includes a descriptivetitle, abstract, references, contact persons, archivalinformation, storage media, and technical details(e.g., parameters measured, scientific discipline,spatial coverage, time period). The directory allowsconnection to other information systems or databasedirectories." The NASA directory concept may beapplicable to other Federal science agencies, andcould be made available to the Federal depositorylibraries and other Federal information dissemina-tion facilities. NASA is also developing expert"data navigation" systems: software to help usersrapidly search, access, manipulate, and display data.

The Interagency Working Group on Data Man-agement for Global change is developing andadapting NASA's master directory into an "interop-erable directory" that will provide access to infor-mation about global change data. Earth sciences datawill be maintained by each agency on a decentral-ized basis, along with detailed catalogs or invento-ries of its data sets. Summaries of the data sets willbe in a central directory that can route inquiries to thedetailed catalogs located at individual data centersand can also transfer data among the various datacenters and users. Both online and offline electronicservices will be available.31

The operational version of the directory willinclude the following Federal earth sciences datacenters or systems: NASA (National Space ScienceData Center including the NASA Climate, Ocean,and Land Data Systems); NOAA (National Oceano-graphic Data Center, National Geophysical DataCenter, National Climatic Data Center); and USGS(Earth Science Information Center, Earth ResourcesObservation Systems [EROS] Data Center, National

23U.S. National Aeronautics and Space Administration, Goddard Space Flight Center, National Space Science Data Center, NSSDC Data Listings,

NSSDC-88-01, January 1988.

29See Office of Management and Budget, "Second Advance Notice of Further PolicyDevelopment on Dissemination of Information," Federal

Register, vol. 54, No. 114, June 15, 1989, pp. 25554-2.5559.

30U.S. National Aeronautics and Space Administration, Goddard Space Flight Center, The National Space Science Data Center, NSSDC-88-26,

January 1989, pp. 5-6.

31U.S. National Aeronautics and Space Administration, Goddard Space Flighzt Center, National Space Science Data Center, "Report on the Third

Catalog Interoperability Workshop, Nov. 1648, 1989," James R. Thieman, Mary E. James, and Patricia A. Bailey (eds.), March 1989."


Water Data Exchange [NAWPEX], and Earth Sci-ence Data Directory, among others).32 For example,USGS has an Earth Science Data Directory that canbe queried from remote computer terminals toidentify and locate over 2,000 databases in fieldssuch as geology, hydrology, and cartography. Thedirectory includes: a description of each database;time and geographic coverage of the data; frequencyof data updating; type of computer; and person(s)wsponsible. (USGS does not charge for onlineaccess, although users pay their own telecommuni-cation charges.)

The working group and participating Federalagencies are supporting the development of anArctic environmental data directory to further testthe directory concept on a small scale. Arctic climateis thought to be a sensitive indicator of globalchange. Thus the arctic data directory should havedirect utility to the global change research program,and it can also serve as a prototype for a larger earthsciences data directory. CD-ROM is being used fordisseminating the Arctic data directory, selecteddata sets, and reference and bibliographic materialsrelevant to polar regions.33

Launching an STI Education Initiative

Improving U.S. science education is important torenewing U.S. competitiveness. Federal STI can beused to teach students about science and technologyand assist then! in acquiring basic informationsearch and retrieval skills that are applicable to manycareers in the information age.

STI data sets could be usedeither online or ondiskfor computer models and simulations inscience laboratories. Students could use computer-based references and data in their work on topics like

Federal science agencies could sponsorpilot projects in local elementary andsecondary schools to demonstrate theuse of Federal STI in the science curricu-lum.

energy, environment, health, and space. It is possibleto design computer-based enhancements to thescience and math curricula that are matched tostudent skill levels for each year in school.34

Schools and colleges have already made a signifi-cant investment in microcomputers; but as yet, asidefrom the major research universities, STI is rarelyused in the classroom. Federal agencies, libraries,and private vendors have limitless opportunities toprovide accessible and affordable STI to elementaryand secondary schools as well as colleges anduniversities.35 There is a pressing need to break theroutine of science education, bring more excitementinto the program, and involve the students directly.36Computer-based STI might help capture the interestand enthusiasm of elemeatary and intermediatestudents through more "hands-on" science, andstrengthen the quality of science education as well.The implications for high school and collegiatescience could be profound.

"Hands-on" science means emphasis on observ-ing, critical thinking, and doi,-:z rather than rotememorization of facts. Some science educationmaterials already include computer software andcould be extended to Federal STI databases online or

325 ee, for example, U.S. Interagency Working Group on Data Management for Global Change, "Interagency Session on Data Management for GlobalChange," meeting minutes dated Sept. 18, 1987, and Mar. 3, 1989.

335ee Aug. 8, 1988, memo from Thomas L. Laughlin, Coordinator, Arctic Environmental Data Workshop, National Oceanic and AtmosphericAdministration, to Arctic Environment Data Directory Working Group; Douglas R. Posson, "Arctic Environmental Data System: Results from theBoulder, Colorado, Workshop," Arctic Research of the United States, Fall 1933, vol. 2; and Feb. 3, 1989, mcmo from Douglas R. Posson, Mamma,Arctic Environmental Data Directory Working Group, USGS, to Working Group Members.

14For a detailed discussion of opportunities for computer-based mathematics education, see National Research Council, Mathematical Sciences

Education Board, Reshaping School Mathematics: A Philosophy and Framework /or Curriculum (Washington, DC: National Acackmy Press, 1990),and Everybody Counts: A Report to the Nation on the Future of Mathematics Education (Washington, DC: NationalAcademy Press, 1989); and National

Council of Teachers of Mathematics, Commission on Standards for School Mathematics, Curriculum and Evaluation Standards for School Mathematics

(Washington, DC; National Council of Teachers of Mathematics, March 1989).

33Some private vendors already offer various Federal Sfl databases at educational or school library discoun prices that rangefrom 25 to 40 percent

off the list prick

36A1 Saley, I Jent, Society of School Libran, us International, telephone conversations with F.B. Wood of OTA, October and November 1989.0 1 1


on disk; many more curricular materials are wellsuited to the use of Federal STI as course supple-ments. The range of possibilities is illustrated intable 2.

The degree of difficulty in the course materialscould be scaled to the educational level. But theimportant point would be to include Federal (andother) STI as a component, where appropriate, 2.1student workbooks, teacher supplements, subjectmatter overviews, and even "take-home" softwarepackages for use on the family microcomputer or atthe local library or science museum.37

Several recent studies have focused on the prob..lems and challenges of U.S. science education,38 butfew have considered the role of Federal STI or STIgenerally. OSTP could provide leadership in thisarea, perhaps working with FCCSET or otheradvisory bodies, and launch a science educationinitiative based on Federal STI, or include FederalSTI as part of a broader science education progam.An ST1 education initiative could encompass thefollowing kinds of major activities:

1. Federal science agency STI pilot projects inlocal schools. Federal science agencies could spon-sor pilot projects in local elementary and secondaryschools to demonstrate the use of Federal STI in thescience curriculum. The various Federal agency datacenters could make copies of prototype Federal `..alCD-ROMs available at no or nominal charge andperhaps provide start-up training on a pilot basis incollaboration with the educational community.39Th: would help teachers and students better under-stand the potential of Federal STI in user-friendlyelectronic formats. Local schools could also experi-

Table 2illustrative Use of Federal 511 as ScienceEducation Course Supplements

Topic Applicadonlmedia

Earthquakes: Land In Motion Sbidents could analyze the 1889California earthquake in per-spective of long-term trends,other malor quakes and theirgeographicdIstrbition, usingdata from NOM $ NationalGeophysical Data Center (CD-ROM).

Earth: The Water Planet Students could exanine currentstream flows, lake levels, andpredOtation (rain, snow) forregional variations and long-term tends, using USGS waterdata and NOAA climatic data(online, CD-ROM).

Space: The Last Frontier Students could explore the solarsystem through the eyes ofspace probes such as Mari-ner and Voyager, using Im-agery from the NASA Na-tional Space Sdence DataCenter (videodisk, CD-ROM),

Toxic Waste: Silent Danger Students could identify loxic wastedumps in their vicinity, deter-mine the chemimls twolved,and analyze the toxicologicaland envirotal effects, wingdatabases from the EPA andNLM (online, CD-ROM).

SOURCE: Office Technology Assessment, 1990.

ment with online access to Federal data centers, andwith electronic networking for both data (anddocument) transfer and distance learning.40

2. Federal educational programs with STI applica-tions. The Department of Education, Department ofDefense, and National Science Foundation havemajor programs in science, engineering, and mathe-

37The National Science Teachers Association and National Council of Teachers of Mathematics both have developed extensivecurricular materials

that could be reviewed for potential Federal ST1 applications.

38See American Association for the Advancement of Science, Science for All Americans: Project 2061 Report on Literacy Goals in Science,

Mathematics, and Technology (Washington, DC: 1989); U.S. National Research Council, Everybody Counts: A Report to the Nation on the Future ofMathematics Education (Washington, DC: National Academy Press, 1989); and U.S. Congress, Office of TechnologyAssessment, Educating Scientists

and Engineers: Grade School to Grad School, OTA-SET-377 (Washington, DC: U.S. GovernmentPrinting Office, June 1988).

39The U.S. Geological Survey (in collaboration with NOAA and NASA) is implementing project JEDIJoint Earth Sciences Educational

Initiativoto bring earth sciences information in CD-ROM format to high school students and teachers in northern Virginia. The primary JEDI goal

is "to invigorate the teaching of earth science studies in our primary and secondary schools throughoutthe country. . .through the creation of a stimulating

and innovative set of teaching materials." And NASA is implementing project LASER--Learning About Science, Engineering, ar tl Researchto bringNASA science and technology resources, including ST1, to K-12 teachers and students. The project uses teacher workshops, pub!, ,Iraries, audio/visual

materials, and mobile laboratories (equipped with computer access to NASA's Spacelink ST1 system) to enrich science and rnaiuematics education anddevelop "hands-on" activities that reinforce student interest in science and math.

40For general discussion, see U.S. Congress, Office of Technology Assessment, Critical Connections: Communic ation for the Future, OTA-C1T-407

(Washington, DC: U.S. Government Printing Office, January 1990); and OTA, Linking for Learning: A New Course for Education, 01A-SET-430

(Washington, DC: U.S. Goverurnentpripriug Office, November 1989). I? I


matics education.4' Many of these programs permitand sometimes require the use of computer technol-ogy as part of teacher training, curriculum develop-ment, and instructional support activities.

The Hawkins-Stafford School Improvements Actof 198842 authorizes the use of funds for trainingmath/science teachers in computer use, and forpurchase of computer hardware and software. TitleVI of the Omnibus Trade and Competitiveness Actof 198843 authorizes demonstration programs intechnology education to:

inform students about technology applications;develop student skills in using technology;prepare students for life-long learning in atechnological society; andimprove teacher competency in technologyeducation.

Under OSTP coordination, these programs could bereviewed for opportunities to include Federal STI.

Technology,4%hanced use of Federal STI is anappropriate topic for teacher training and studentprojects. The Nat onal Science Teachers Associa-tion has endorsed science education initiatives todevelop curricula to instruct WA, chers on the use oftechnology in the classroom, am provide electronictechnologies to science teachers at all grade levels."

3. Federal science agency collaborative projectswith science museums, associations, and high-techinformation companies. Science museums are verysuccessful in the "hands-on," interactive approachto science education. Most science museums usemicrocomputer-based displays, games, or tutorials,and some provide microcomputer laboratories forintensive computa experience. Computer-based demon-strations of Federal STI applications would be adirect extension of current activities. Federal scien-

Improving the information literacy ofU.S. scientists and engineers is one of themost highly leveraged ways to increasethe return on the U.S. R&D investment.

tific and bibliographic databases could be operatedon a stand-alone basis (e.g., with a dedicatedmicrocomputer using diskette, hard disk, or CD-ROM formats). Science museums with modemscould access online Federal STI databases directlyfrom the government and/or private vendors.

Federal STI also could be included in scienceeducation programs sponsored by scientific associa-tions and/or private companies. The AmericanAssociation for the Advancement of Science (AAAS)and a telephone company cosponsor a program tohelp middle d high school science teachers learnabout new communications and information tech-nologies, and how these technologies can be used inscience classes. Federal STI would be a naturaladdition to this type of program.° Several privatevendors offer substantial educational discounts foroff-peak online access to various STI databases,including some Federal STI.

4. Federal collaboration with library and ii(orma-tion science professionals. Libraries and the profes-sional library and information science schools offeruntapped potential for improving the use of FederalSTI. Libraries at the major research universities arewell-versed in Federal STI and electronic databasesgenerally. But in many public and school libraries,the use of electronic databases is just beginning Inelementary and secondary schools, the problem iscompounded because the role of librarians in facili-tating electronic access is only dimly understood.

41See C.M. Matthews, Science, Engineering, and Mathematics PreCollege and College Education, 1B 88068 (Washington, DC: CongressionalResearch Sezvice, Nov. 3. 19b9) and Science and Engineering Education: The Role of the Departmenr of Defense,Report 89-256 SPR (Washington,DC: Congressional Research Service, Apr, 18, 1989); U.S. Congress, Office of Technology As..,ssment, Educating Scientists and Engineers, op.cit.,footnote 39, and Power On: New Tools for Teaching and Learning, OTA-SET-379 (Washington, DC: U.S. Government Printing Office, Septembe.g1988),

4211.S. Congress, Public Law 100-297, the "Augustus F. Hawkins and Robert T. Stafford Elementary and Secondary School Improvements Act of1988."

45U.S. Congress, Public Law 100-418, Title Vi, Subtitle 2, "Instructional Programs in Technology Education"; see J.B. Stedman,Computers inElementary and Secondary Schools: An Analysis of Recent Congressional Action, Report 88-419 EPW (Washington, DC: Congressional ResearchService, June 9, 1988).

"National Science Teacheis Association, "Science Education Initiatives for the 1990s," position paper, Sept. 7, 1988.

45See W. Worthy, "Diverse, Innovative Programs Revive Piecollege Science Math Education," Chemical and Engineering News, Sept. 11, 1989,pp. 7-12.

9


Professional grovps such as the Society of SchoolLibrarians are attempting to bring new informationtechnologies into the school library setting, andrecognize the relevance of Federal STI. Jnder OSTPleadership, Federal agencies could collaborate withthe school librarians to help make this vision areality.

OSTP and the Federal agencies could reacn out tothe Nation's schools of library and informationscience and initiate a dialog on how to improve thecollegiate cnrriculum on STI, working closely withthe schools A science, engineering, and technology.The objective would be to educate more librariansand information scientists with a specialty in STI,and upgrade courses on information skills in theacademic science and engineering programs. Im-proving the information literacy of U.S. scientistsand engineers is one of the most highly leveragedways to increase the return on the U.S. R&Dinvestment. Business leaders and academic scholarsincreasingly recognize this need.46

Improving International Exchange of STI

U.S. scientists and engineers are generally notconversant in foreign languages and do not readmany foreign language documents. Only a smallpercentage of foreign language material is translatedinto English, and even here, U.S. demand for suchtranslations has been weak. The problem is two-fold:many U.S. researchers do not sense the need toconsider foreign STI, and do not have the skillsneeded to do so even if they wanted.47

Congess enacted the Japanese Technical Litera-ture Act of 1986 to improve U.S. access to JapaneseSTI. NTIS is responsible for implementing the act,has agreements with about 50 Japanese informationsources, and offers online access to some Japanesedatabases. OSTP could review how well the act isworking, and whether the concept should be ex-tended to other foreign countries. Computer-aidedtranslation offers great promise for enhancing U.S.access to foreign STI. OSTP could examine howprogress in this area can be accelerated.

OSTP also could review U.S. bilateral and multi-lateral science and technology (S&T) agreements toensure that STI is sufficiently covered. STI is anexpt eit U.S. objective in implementing the U.S.-Japal S&T Agreement (i.e., to improve the flow ofJapai.ese STI to the United States), and an ST! TaskForce is focusing on computer-assisted translationof Japanese literature for private industry, academic,and government laboratory users in the UnitedStates.48 OSTP is taking a lead in the U.S.-Japanagreement, a role that could be extended to manyother S&T aptements, and to other aspects of U.S.access to foreign STI. These include education andexchange programs for U.S . and foreign researchers,and cooperative agreements between U.S. and for-eign SIT agencies.

The consensus seems to favor open, reciprocalexchange of SIT, with restrictions on access kept tothe minimum. OSTP should take the lead in balanc-ing the open flow, national security, and competi-tiveness concerns that arise in dealing with interna-tional STI issues.

The principle of open, reciprocal STI access hasbeen accepted for years in the civilian scientificresearch community (as contrasted with military orcommercial research). Global change research ex-emplifies the importance of international STI collab-oration and the complexities involved. The U.S.Interagency Working Group on Data Managementfor Global Change recognized from the outset thatearth sciences data must be collected and dissemi-nated globally to foster research on global change.The Federal earth science agencies have dozens ofinternational agreements for information exchange,and these could be the basis for an international datanetwork, if data systems are made compatible. Theworimg group is coordinating with several nationaland international scientific organizations on earthsciences data management, including:

National Research Council Space Science Board,Committee on Data Mf nagement and Compu-tation;

46See statement of C.R. McClure, Professor of Infonnation Studies, Syracuse University, before a hearing of the House Committee on Science, Space,

and Technology. Subcommittee on Science, Research, and Technology, Oct. 12, 1989.

41See C.H. Hill, "Enhancing U.S. Access to Foreign STE What Should be the Federal Role," in McClure and Hernon (eds.), Federal Scientific and

Technical Information, op. cit., footnote 19, pp. 172-192.

aSee memorandum from D. Allan Bromley, Director, OSTP to U.S. Members of the Joint HighLevel Committee on the U.S.-Japan Science ard

Technology Agreement, Nov. 20, 1989.


A common frustration to those workingon international data systems is the lackof a receptive audience at the seniorlevels of the government. This is begin-ning to change with regard to STI forglobal change.

National Research Council, Numerical DataAdvisory Board (recently renamed the STIBoard);

National Research Council, Committee onGeophysical Data;

International Geosphere/Biosphere Program,Data Management Working Group;

International Council of Scientific Unions,Panel on World Data Centers;

Committee on Earth Observation Satellites,Working Group on Data;Committee on Data for Science and Technol-ogy (CODATA); and

World Climate Data Program.

A common frustration to those working oninternational data systems is the lack of a receptiveaudience at the senior levels of the government. Thisis beginning to change with regard to STI for globalchange.

The challenge of managing global change data ismammoth. NASA's Earth Observing System alonewill generate an additional terabyte (1012 bytes) 4ofdata every day. This is equivalent to 10,000 Wash-ington, DC telephone books (white pages) or 520,000text books (at 200 pages each) per day.49 Electronictechnologies offer the only hope for managing thisdata (see appendix). The Interagency WorkingGroup has concluded, after several years of effort,that the size and geographic scale of global datarequire new approaches to data management andinternational cooperation if the potential of thesetechnologies is to be realized.5°

Table 3Illustrative Weaknesses In Current GlobalChange Data Management

Weakness Explanation

Data quality

Data managementprocedures

Data managementtechnologies

Data managementinfrastructure

Global change datasets

Satellite datacalibration

Data archives

Data standards

Many data sets lack credibility due toinconsistent or poor documentation andquality control.

There are no established criteria or policiesfor evaluating, archiving, and updatingglobal data sets.

New technologies are not applied in aconsistent or coordinated manner toglobal data sets.

The data systems to handle increasedobservational data ere not yet in place.

Very few data sets have been compiled andprocessed for the specific purpose ofmonitoring and detecting climatechange.

Despite 25 years of satellite observations,only one satelNte data set is sufficientlywell-calibrated to document globalchange.

Retrospective data sets are poorly cata-loged, inconsistently documented, inac-cessible, and subfect to an undisci-plined publication process.

Data formats and exchange mechanismsare inadequately standarched. Stan-dards that exist we not uniformly ad-hered to.

SOURCE: Committee on Earth Sciences, 1989.

Data management is critical to the success ofglobal change initiatives. The U.S. Global ChangeResearch Program now includes data managementin the overall plan, and presents a detailed datamanagement strategy.51 However, the problems thatneed attention are daunting, as highlighted in table3.

While some might question the severity of re-ported data management problems, the need forinternational cooperation is compelling. As theCommittee on Earth Sciences concluded:

Data management requires global and interna-tional cooperation.. .No one nation, agency, orinstitution will be able to gather the appropriate datawithout cooperation from other nations, other agen-cies, and other institutions. Individual agencies willneed the cooperation of others to collect, manage,and preserve data sets systematically for globalchanJ and make them accessible across the tradi-tional discipline and agency bouadaries.52

49See R. Kahn, "Coping Wzth All the Earth Science Data," EOS, vol. 69, No. 21, May 24, 1988, pp. 609, 612.

50U.S. Interagency Working Croup on Data Management for Global Change, "Interagency Session, Minutes," June 2, 1989.

31U.S. Federal Coordinating Council for Science, Engineering, and Technology, Committee on Earth Science, Our Changing Planet: The FY 1990

Global Change Research Plan (Washington, DC: OSTP, July 1989), pp. 91-99.

521bid., p. 94.

Tpl

Appendix A

Technological Opportunities for Managing and DisseminatingFederal Scientific and Technical Information

Introduction to thc Electronic STI Revolution

Dissemination of Federal STI is being transformed bythe ongoing rev lution in electronic information andtelecommunication technologies. The scientific and tech-nical community is one of the heaviest and most advancedusers of computers. The vast majority of U.S. scientistsand engineers have a microcomputer at work and/or athome, and some have access to mainframe and high-orformance computer resources either onsite or throughtelecommunication networks. The microcomputer orworkstation provides the scientist or engineer with aversatile tool. Continuous, steady improvement in theprice/performance of microcomputers has resulted in thepower of a 1970s-vintage mainframe computer now beingon the desktop of the typical scientist. The microcomputercan be used to search, recover, and store STI on magneticor optical media, manipulate and analyze ST1 using avariety of software, and access SIT remotely via onliiiebulletin boards, computer conferences, and databasenetworks.1

Online information networks serve at least threeimportant needs of the scientific and technical commu-niv. First, they are used for the transfer of very largestreams of STI, for example, from a central repository ofdata collected by Earth-observing satellites to regionaldata repositories and to individual research institutions oruser groups. Second, online networks are used US searchSTI bibliographic databases and to remotely accesslarge-scale high-performance computers. Third, onlinenetworks are used for informal exchange of STI amongresearchers, for example, an electronic bulletin board onresearch in progress or upcoming key events, a computerconference for exchanging working notes and ideasamong scientists conducting related research, and elec-tronic mail for submission of manuscripts and reviewcomments to scientific and technical journals and tofunding agencies.2 Online STI dissemination benefitsfrom both a proliflation of online gateways that providechannels for electronic information exchange (offered bytelecommunication common carriers, value-added carri-ers, and not-for-profit and governmental systems), and agrowing variety of STI services (especially bibliographicand reference services offered by commercial and not-fur-

Dissemination of Federal STI is beingtransformed by the ongoing revolutionin electronic information and tele-communication technologies.

profit organizations as well as some government agen-cies). Advances in online STI gateways and informationservices are made possible in part by progress inunderlying digital telecommunication technologies (e.g.,packet switching, fiber optics, and satellite networking).The net result is that online is feasible over a broader rangeof 511 dissemination applications than ever before.

The package of online and optical disk technologiesoffers a powerful combination. Online can be effectivelyused when time or geographic factors are most important(e.g., bibliographic updates on just-published research,access to remote computing resources or to internationalSTI databases) and offline optical disks can be used forlarge data sets and/or extensive data manipulation andanalysis requirements that are not time-sensitive andwould be much more expensive online (even at off-peakrates).

The future of STI dissemination will be dominated byelectronic formats. Some major types of ST1--e.g.,satellite remote sensing data or the results of large-scalecomputer modelsare created, stored, transmitted, andused in electronic form. These data are rarely, if ever,converted to paper or microfiche, except when summa-rized and analyzed in technical reports and scientificpapers. By comparison, STI bibliographic and referencematerials are currently offered and used in paper, micro-fiche, and electronic formats (principally online andCompact Disk-Read Only Memory (CD-ROM)). Full-length reports and documents are still largely distributedon paper or microfiche. However, electronic publishing israpidly taking over the document preparation and produc-tion process. Must STI documents are created electroni-cally with word processing systems or software, eventhough the output is still on paper or microfiche.

!See U.S. Congress, Office of Technology Assessment, Informing the Nation: Federal Information Dissemination in an Electronic Age,OTA-CIT-396 (Washington, DC: U.S. Government Printing Office, October 1988).

2See National Academy of Sciences, Conunittee on Science, Engineering, and Public Policy,Information Technology and the Conduct of Research:

The User' s View (Washington, DC: National Academy Press, 1989); U.S. Congress, Office of Technology Assessment, High Performance Computing

and Networking for Science, OTA-BP-CIT-59 (Washington, DC: U.S. Government Printing Office, September 1989).

-57--

58 Helping America Compete: The Role of Federal Scientific and Thchnical Information

The package of online and optical disktechnologies offers a powerful combina-tion.

Electronic publishing makes it possible to cany headvantages of elettronic word processing through allstages of document preparation and information dissemina-tion. Electronic publishing creates an electronic documentdatabase that can be accessed online, stored on magneticor optical media, or printed out in whole or in part onpaper or microfiche. The major barrier to realization of the"intelligent database" is standardization of data struc-tures and file formats for graphics and datasets as well astext.

The price/performance of all electronic publishingcomponents continues to impmve. This is resulting in acontinued narrowing of the gap between relativelyinexpensive desktop systems and expensive, high-endelectronic publishing and phototypesettneg systems. Desk-top systems can be linked to very fast, very high-qualityphototypesetters and printers.

Desktop publishinn and dissemination functions bene-fit from steady piogress in development of expertsystems. The expert systems applicable to STI dissemina-tion are no different in principle from the systems thathave been successfully applied to other scientific, indus-trial, and educational areas. Expert systems with sophisti-cated search strategies can be used to retrieve and deliverbibliographic or full-text STI from offline (e.g., CD-ROM) or online information systems. Expert systems canimprove the dissemination process by accounting for suchfactors as: the profile of the information product (numberof pages, layout, type style, use of graphics, etc.),anticipated user needs (e.g., size of demand by format),and the modes of dissemination (press run, provisions fordemand printing in paper or microform, online databaseaccess, optical disk distfxbution, etc.). Expert systems canalso assist SDI (selective dissemination of information)by matching user interest profiles with available data-bases, and, potentially, in translation of STI from foreignlanguages to English (and vice versa).

Over the next 3 to 5 years, use of printed Federal. SITis likely to decline modestly, while the use of electronicformats will likely increase dramatically. Some tiansi-tional effects are already evident. For example, theNational Technical Information Service (NTIS) experi-enced a roughly 50 percent reduction in sales of paper andmiciofiche copies of repons between 1980 and 1989. Thereduction is attributed in pag to the effectiveness of onlinesearehing of the NTIS bibliographic database (offered viaprivate vendors).3 The fagest growing NTIS product linenow is computer products. The Office of Scientific andTechnical Information at the Department of Energy hasnoted a similar declining demand for papaw and micro-fiche copiee over the past decade as reliance on computer-ized bibliographic databases increases.4

Surveys conducted by the General Accounting Officehave documented the plans of Federal agencies to increasetheir use of electronic fonnats for STI, and the gmwinedemaad of STI users for electronic formats. The surveyresules indicated a 50 percent or greater anticipatedincrease over a 3-year petiod in the number of civilianagencies using electronic mail, elecnorlic bulletin boards,floppy disks, and compact optical disks for I1 dissemina-tion. The results showed a doubling over the next 3 yearsin the nuniaer of scientific and twhnical associationsdesiring Federal ST1 in electronic formats. For Federaldepository hinnies, the results indicated, for example,about an eight-fold increase over the next 3 years indemand for Fedenti STI on compact optical disks. Incontrast, the results showed a ptonxted decline in demandfor paper and micmfiche fremats of about 15 to 20percent5

A key to redizing the potential for technology-enhanced dissene nation is the "information life cycle,"where STI dissemination is part of the larger process ofcollectionkreatiore storage, processing, and archiving.The stages in the STI process need to be integrated withinterconnected technologies to be cost-effective. Thus thecost and delays associated with rekeyboarding, incompat-ible equipment, and the like can be reduced.

Another key ia to substantially upgrade technologytraining for the scientific and technical community.Recent surveys have, concluded that many scientists andengineers are still not comfortable with online and ondisksystems.6 At present, user education and training receive

3U.S. Congress, Office of Technology Assessment, Informing the Nation, op. cit., footnote 1, pp. 112-114; U.S. National Technical InformationService, "Annual Report to the Congress on NT1S: Operatious, Audit, and Modernization," January 1989.

4Bonnie C. Carroll, Office of Scientific and Technical Information, U.S. Department of Energy, "DOE Reports Distribution Program: Curient Systemand Why Change Is Needed," Apr. 30, 1986.

5U.S. General Accounting Office, Federal Inf -mation: Agency Needs and Practices, Fact Sheet for the Chairman, Joint Committee on Printing, U S.Congress, GAO/GGD-88-115FS, September 19815, mid U.S. General Accounting Office, Federal Information: Users' Current and Future TechnologyNeeds, Fact Sheet for the Chairman, Joint Committee on Printing, U.S. Congress, GAO/GOD .89-20FS, `'-werriber 1988

6See statemmt of Charles R. McClure, Professor of Information Studies, Syracuse University, before a nearing of the House Committee on Science,Space, and Technology, Subcommittee on Science, Research, and Technology, Oct. 12, 1989. Cf') )

App. ATechnological Opportunities for Managing and Disseminating Federal Scientific and Technical Information 59

Over the next 3 to 5 years, use of printedFederal STI is likely to decline modestly,while the use of electronic formats willlikely increase dramatically.

minimal attention. In order to fully realize the potential ofelectronic technologies for STI, user training needs to beviewed and funded as an integral part of STI systemdevelopment. Once exposed to electronic ST1 and trainedin its use (whether formally or through self-or collegial-learning), the users frequently become technology enthu-siasts.7

The convergence of trends in technology and in userpreference for electronic data, combined with the emer-gence of systems integration and standards for the STI lifecycle, offer an almost limitless array of possibilities forS fI dissemination. Several of these arc highlighted belowin the context of Federal science agency applications.

CartographiclGeographic Information

Many aspects of science and technology depend ongeographic information- -frequently in the form of mapsthat show the location of transportation networks, naturalresources, climate regimes, environmental sources, andthe like. In the past, these maps were prepared by hand andprinted on paper. Over the last 15 years or so, mapmakinghas been computerized, and satellite imagery has beenincorporated along with data from field surveys and aerialphotogrammetry. But the final product was and still islargely printed on paper. Over the laet 5 years advances incomputer technologies have culminated in "nothing lessthan a cartographic revolution."

This revolution is being driven by digital cartographycombined with powerful hardware and software that canaceeas and manipulate geographic data from multiplesources. By collecting information in digital (as opposedto analog form), or by converting analog data (e.ga aerialphotoginphs) to digital form, the data can be readilyprocessed by computers to produce a vast array ofcomputer products. Digitized maps can be displayed oncomputer screens and recorded on magnetic and opticalmedia, for example, as well as used to produce traditionalprinted maps.8

The U.S. Geological Survey (USGS) expects that manyof these digitized maps will be produced in CD-ROMformat at a fraction of the cost of the equivalent magnetictapes or printed paper documents. USGS pilot tests ofCD-ROM indicate that it is likely to be an order ofmagnitude less expensive than computer tapes, and willrequire only a microcomputer and CD-ROM reader ratherthan a more expensive mini- or main-frame computerneeded for tapes.

Optical disks will revolutionize STI storage anddissemination. Optical disk technology uses a laser beamto record data on plastic disks by engraving pits in thesurface. Encoded disks can be read by a low-power laserbeam to retrieve the data. Other members of the opticaldisk family include: WORM (Write Once Ready Many-times); Erasable disks; Videodisk (for storing film or stillphotos); and CD-I (Compact Disk-Interactive) that com-bines text, data, video, audio, and software capabilities onone diale.

The CD, ROM is rapidly gaining acceptance, and thebasic technical standards are already in place. Themarginal cost of producing CD-ROMs is very lowcurrently about $2 per disk at volumes of several hundredor more. The full cost can be as much as $50 to $500 perdisk for several hundred, if the c asts of data preparation,premasta.14, and mastering are included. But even thiscompares favorably with other storage media. EachCD-ROM can store up to about 600 megabytes (millionsof bytes) of data. This is equivalent to about 300,000 textpages (assuming 250 words or about 2,000 bytes perpage). One CD-ROM can store the equivalent of about1,650 floppy diskettes, 30 of the 20-megabyte hard disks,15 of the 1,600 bits-per-inch 9-track magnetic computertapes, or 4 of the 6,250 bits-per-inch computer tapes. Thusa $2 CD-ROM can store as much az several hundreds ofdollars' worth of magnetic media. Microcomputer-basedCD-ROM authoring software now costs less than $1,000.Agencies with heavy CD-ROM activity might be able tojustify purchase of a premastering system ($50,000 to$100,000), but will need to contract out for mastering andduplication.

USGS issuing CD-ROMs with cartographic and geo-graphic information on a variety of topics, such as:

Gloria Sidescan Sonar Data contains data for theGulf of Mexico and parts of the eastern PacificOcean, produced by USGS, NOAA, and NASA, andavailable from USGS;

'Sex John RS, Clement, "Increasing Research Productivity Through Information Technology A User-Centered Viewpoint," ruanusaipt submitted

to "Research Reviews in Information and Documentation," October 1989; also see National Mademy of Sciences, Information Technology and the

Conduct of Reseatch: The User's View, op. cit., footnote 2.4U.S: Federal Interagency Coordinating Committee on Digital Cartography, "Coordinsting of Digital autographic Activities in the Federal

Government," Sixth Annual Report to the'OMB Director, 1988. fi 0.9


The U.S. Geological Survey expects thatmany of these digitized maps will beproduced in CD-ROM format at a frac-tion of the cost of the equivalent mag-netic tapes or printed paper documents.

AMINO=

111111111111111111&

Aerial Photography Recordscontains aerial photo-graphs from the USGS National Cartographic Infor-mation Center (recently renamed the Earth ScienceInformation Center);Joint Earth Sciencescontains sidelooking airborneradar data, prototype produced by and available fromUSGS, Bureau of Land Management, and SoilConservation Service;Hydrodata -contains daily measurement data forUSGS water gage stations, produced and sold byEarth Info., Inc. (for profit, formerly U.S. WestOptical Publishing); andUSGS Reference Materialscontains GEO Index (adatabase of geologic maps) and Earth Science DataDirectory, produced and sold by OCLC, Inc. (not-for-profit, Online Computer Library Center).

Space Science Data

The collection of scientific data by satellites androcketsalready very extensivewill increase furtherover the next few years, as a new generation of Earth- andspace-observing satellites, manned space missions, andinterplanetary and deep space probes is launched. Thestorage and dissemination of these da: A pose a majorchallenge to the Federal science agenciesand especiallyto the National Aeronautics and Space Administration(NASA). Several new electronic technologies have thepotential to avoid total systems overload from theexpected avalanche of space data.

NASA's primary institution for space data man-agement and dissemination is the National Space ScienceData Center (NSSDC) located at the Goddard SpaceFlight Center in Greenbelt, Maryland. NSSDC is thelargest space data-archive in the woad, with about 85,000magnetic tapes of digital data currently on file (along withanother 35,000 backup magnetic tapes). The NSSDCarchives only processed data, not the raw telemetry datareceived directly from space. The center also archives alarge volume of photographs and film taken by satellitesand space missions. Some data are maintained on

microfomi or hard copy. At present, the center archivesabout 4,000 different data sets, mostly from NASAmissions but with a few from Department of Defense orforeign missions. The center retains no classified data,and the primary users are researchers from the disciplinesof astronomy, astrophysics, lunar and planetay science,solar terrestrial physics, space plasma physics, and earthsciences.9

The opportuniees are substantial for use of opticaldisks to store and disseminate space science data. NASAis beginning to experiment with both 12-inch WORM and4.75-inch CD-ROM. One WORM product is currentlyavailable for dissemination (the data from 20 magnetictapes were transferred to one WORM disk). And fourprototype CD-ROM products are available: 1) a CD-ROM space science sampler that includes a cross-sectionof planetary, land, oceans, astronomy, and solar-terrestrialdata ($50 for the CD-ROM, software on floppy disk, anddocumentation); 2) a 3-disk CD-ROM set of Voyager/Uranus images ($100 for the disks, software, and docu-mentation); 3) a 5-disk CD-ROM set of Voyager/Jupiterand Saturn images ($175 for the disks, software, anddocumentation); and 4) a CD-ROM produced by theNOAA National Geophysical Data Center that includessolar wind and magnetic field data from NASA andvarious geomagnetic and solar data from NOAA (disk andbasic software free while they last; $100 for advancedsoftware and updates).

An understanding of the potential of optical disks canbe gained from the following hypothetical examples. TheApollo 17 lunar mission generated about 240 magneticcomputer tapes of digital data, 32,000 feet of 16 mm colorphotographs, and 39,000 feet of 16mm black-and-whitephotographs.19 These digital data could be stored on about4 double-sided 12-inch WORM disks. (One 12-inchWORM disk can store 1.2 gigabytes per sideequivalentto 30 of the 1,600 bits-per-inch mapetic tapes. Atwo-sided WORM disk can store 2.4 gigabytes or 60 tapesof data.) With 4:1 data compression, it would be possibleto store the Apollo 17 data on one WORM disk. The16mm photographic data, which in this example areequivalent to roughly 850,000 individual photographs,could be stored on about 17 analog videodisks (at thestandard 54,000 images per videodisk).

For some of the earlier missions, data for entire seriesof mission activities could be consolidated. For example,the Mariner interplanetary mission series generated thefollowing volumes of digital data in number of magnetictapes: Mariner 2 (5 tapes); Mariner 4 (10 tapes); Mariner9 (42 tapes); and Mariner 10 (184 tapes)." The total of

9See U.S. National Aeronautics and Space Administration, Goddard Space Flight Center, The National Space Science Data Center, NSSDC-88-26,January 1969.

1°U.S. National Aeronautics and Space Administration, Goddard Space FlightCcmter, NSSDC Data Listings, NSSDC-88-01, January 1988."mid,

App. ATechnological Opportunities for Managing and Disseminating Federal Scientific and Technical Information 61

286 magnetic tapes could be stored on about 5 double-sided 12-inch WORM disks (without data compression).The NSSDC archive provides clear evidence of theproliferation of space data over time, as the number andsophistication of space missions increased.

New optical and magnetic storage technologies makeit possible for NSSDC to carry out a gradual transitionfrom magnetic tapes and photographic film to higherdensity storage media such as optical disks or digital tapecartridges (not tape reels, see later discussion on earthsciences data) for digital data and videodisks for analogdata. This transition will be quickest for newly acquireddata, and for historical data that needs to be re-recordedon new media (i.e., due to deterioration of magnetic tapes,many of which are more than 10 years old and written onobsolete technology).

At the same time, demand for online data disseminationis also increasing. NSSDC is making more data setsavailable online either over networks or on a dial-up basis.Network options currently include: SPAN (the SpacePhysics Analysis Network) that links DECnet-basedcomputers in the United States, Canada, Europe, Japan,Australia, New Zealand, and South America; NSN(NASA Science Net work) that links with NSFnet and theARPANET-based Internet; BITNET that links variousuniversities and research organizations; and Telenet, apublic packet-switching data network.12

Second, technical evaluations and guidelines will needto be developed an when and how to use these media forstoring and disseminating data. How fast should high-density storage media be phased in, and what kinds of datasets are best suited for WORM, CD-ROM, videodisk,digital tape cartridge, and other storage technologies?These guidelines will need to take into account the abilityof users to accommodate high-density storage media; interms of training, equipment, and cost. What are thehighly leveraged data sets that are both best suited for thenew media and matched to user capabilities to handlehigh-density storage? And the guidelines will need toconsider the appropriate balance between offline high-density storage media and online dissemination.

At present, NSSDC includes only a small number ofdata sets in the online program, and generally limitsonline time to one-half hour or less. This restriction isbased in part on the limited transmission speeds (e.g., still

9.6 kilobits or occasionally 56 kilobits per second, formany universities) such that longer transmissions costmore than offline dissemination. However, online willbecome mote cost-effective as transmission speeds in-crease. NASA itself already has a 1-megabit/secondtransmission network for use by NASA laboratories ant'centers. And the proposed multi-agency national researchAnd education network (NREN) anticipates transmissionspeeds of 1-gigabit/second or more in the future.13 Somecurrent online space science data sets include'

International Ultraviolet Explorer Satellite, containsultraviolet spectral data, sponsored by NASA, Euro-pean Space Agency, and British Science and Engi-neering Council;Total Ozone Mapping Specut -teter, contains 120days of ozone data from the Nimbus 7 satellite,sponsored by NASA;Space Telescope Archive and Catalog, containscatalogs of astronomical data and various observinglogs from spaceborne astronomy mi qions, spon-sored by European Space Telescope and SouthernObservatory; andCrustal Dynamics Dau Information System, con-tains catalog of data from Satellite Laser Ranging,Lunar Laser Ranging, Very Long Baseline Interfer-ometry, and Global Positioning System experi-ments, sponsored by NASA, National GeodeticService, and various universities.

Farth Sciences Data

Over the last several years, the Federal scienceagencies, and the scientific community generally, havemade a significant effort to improve the collection,management, and dissemination of earth sciences data.This effort is driven by the widespread concern overproblems of global changeranging from climate changeand deforestation to air and water pollution to soil erosionand demineralization to droughtand thn recognitionthat better understanding of these global problems re-quires much better information. The concept of the Earthsystem has emerged as an important organizing principle,since global change involves all major earth subsystemsthe atmosphere, oceans, snow and ice, lakes and rivers,land formations, and the biosphere (e.g., trees, plants, andanimals) and can be affected by forces from deep within

'2U.S. NASA, Data Center, op. cit., footnote 9, pp.15-16; also see U.S. Congress, Office of Technology Assessment, High Petformance Computing,op. cit., footnote 2.

"See U.S. Office of Science and Technology Policy, Executive Office of the President, The Federal High Petformance Computing Program, Sept.

8, 1989; and U.S. Congress, Office of Technology Assessment, High Peiformance Computing, op. cit., footnote 2.


the Earth (e.g., volcanoes and earthquakes) and from farin space (e.g., changes in solar radiation).14

The earth system concept is being used to organize thevast array of data relevant to the disciplines that compisethe earth sciencesclimatology, oceanography, glaciol-ogy, hydrology, biology, biogeochemistry, geology, etc.In the U.S. Govermnent, the long-term objeciive is todevelop a "virtual" interagency information system forglobal change data. "Virtual" means that the informationsystem will be a family of decentralized data centers, mostof which already exist in some form, linked together bycommon directories, standards, and policies on access,user charges, quality control, and the like. The goal is tohave the system fully implemented by the time thatNASA's planned Earth-observing system is operationalin the late 1990s (and thus generating a large additionalvolume of earth sciences data).15

As is the ease for space science data, the most effectivetechnology for managing this massive volume of data ishigh-density storage. Some of the smaller data centerscould be converted entirely to a combination of WORMand CD-ROM. For example, the National OceanographicData Center, operated by NOAA, maintains about 12gigabytes of processed data in the following categories:chemical data (marine chemistry), pollutants/toxic sub-stances); biological data (e.g., fish/shellfish, marine birds,plankton); and physical data (e.g., wind/waves, current,subsurface temperCure). NODC also maintains about 12gigabytes of raw, unprocessed data. The entire NODCdatabase of 24 gigabytes would fit on about two to twelvedouble-sided 12-inch WORM optical disks, depending onthe data compression ratio. As new data accumulate, theWORM disks could be updated. Those portions of thedatabase in high demand could be extracted, mastered,and duplicated at very low cost on CD-ROM, and updatedCD-ROMs could be issued periodically.

Several Federal earth science data centers are experi-menting with CD-ROM. One is the National Snow and IceData Center, operated by the University of Colorado forNOAA's National Geophysical Data Center (NGDC).The Snow and Ice Data Center has issued a prototypeCD-ROM with data on Northern Hemisphere "brightnesstemperature grids," which are collected by a NASAsatellite and used to estimate the polar sea ice parameters.The CD-ROM disk comes with a software diskette and a

In the U.S. Government, the long-termobjective is to develop a "virtual" inter-agency information system for globalchange data. "Virtual" means that theinformation system will be a family ofdecentralized data centers.

user's guide, and is available free while supplies last. Thisis the first in what is planned as a series of CD-ROMs, andreflects a shift in data dissemination philosophy to offlinelow-cost optical disks for many research purposes.16 Ingeneral, the NGDC believes that CD-ROM will greatlyimprove the accessibility and usability of STI by theresearch community, as well as by governmental andprivate-sector organizations that depend on geophysicaldata.

The larger data centers are also considering high-density magnetic as well as optical storage. For example,the EROS (Earth Resources Observation Systems) DataCenter, operated by USGS, archives about 6 millionframes of aerial photographs and over 1 million Landsatand other remotely sensed satellite images. 'The Landsatimagery alone is roughly equivalent to 75 terabytes (or75,000 gigabytes) of digital data. Because of this largevolume, the EROS Data Center is considering the digitaltape cassette as the next generation high-density storagemedium. Each cassette can store up to 50 gigabytes ofdata, much more than either CD-ROM (about 0.6gigabyte per disk, or 4 gigabytes with 6:1 data compres-sion) or WORM (1.2 gigabytes per disk up to about 12gigabytes for a two-sided disk with 6:1 data compres-sion). Digital cassettes have a faster data transfer rate thanoptical disks. On the other hand, the digital tape cassetteis a magnetic medium that, like magnetic computer tapereels, deteriorates over time and needs a tape refresh every7 to 15 years. This compares with a projected lifetime of20 to 30 years or more for optical disks (the longevity ofoptical media is still uncertain). The cassettes andequipment cost considerably more than comparableoptical disk systems. Optical disks also have the advan-tage of random (as opposed to sequential) access and

14See J.A. Eddy, "The Earth As A System," Earth Quest, 1987, vol. 1, No. 1, pp. 1-2, available from the Office of Interdisciplinary Earth Studies,University Corporation for Atmospheric Research, Boulder, CO; U.S. National Aeronautics and Space Administration, Barth Systems ScienceCommittee, Earth Systems Science:A Closer View(Washing ton, DC: NASA, January 1988); F.B. Wood, Jr., "The Need for SystIms Research on GlobalClimate Change, " Systems Research,1988,vol. 5, No. 3, pp. 225-2A0; U.S. National Oceanic and Atmospheric Administration, Panel on Global ClimateChange, The Vision: A Rededication of NOM, Januaty 1989; and R. Corell, "A Paradigm Emerging," Earth Quest, 1990, vol. 4, No. 1, pp. 1-4.

155ee, for example, U.S. Interagency Working Group for Data Management of Global Change, "Interagency Session on Data Management for GlobalChange," minutes of meetings dated Sept. 18, 1981, Nov. 24, 1987, and Mar. 18, 1988.

16U.S. National Geophysical Data Center, National Snow and Ice Data Center, Data Announcement, "Scanning Multichannel Microwave Radiometer(SMMR) Brightness Temperatures for the Northern Hemisphere," June 1, 1989; also see R. Weaver, C. Morris, and R.G. Briny, "Passive MicrowaveData for Snow and Ice Research: Planned Products From the DMSP SSM/I System," EOS, Sep, . 29, 1987, pp. 776-777. '7 i

App. ATechnological Opportunities for Managing and Diss eminating Federal Sciennfic and Technical Irformation 63

microcomputer compatibility (with inexpensive, user-friendly software). Optical tape is another storage technol-ogy that warrants consideration. One 12-inch reel ofoptical tape can store up to a terabyte of data. Preparationand duplication cost, expected level of use, storagecapacity, data transfer rate, data access time, longevity,and equipment and training requirements are among thefactors that need to be considered in evaluating alternativestorage media.

Drought Monitoring InformationElectronic technologies open up new alternatives for

dissemination of time-sensitive Federal STI, such asdrought inf' -,"-m, that is widely used (contrasted withthe very h ,;:e and earth sciences data sets that areless tint. Ave and have fewer users). Droughtinformation is collected and disseminated by the U.S.Department of Agriculture and NOAA. The NOAA/USDA Joint Agricultural Weather Facility and NOAAClimate Analysis Center produce several drought-relatedreports and bulletins, such as the Weekly Weather :ndCrop Bulletin.

Should the government decide to prepare and distributea weekly or monthly electronic drought bulletin, it mightinclude: temperature and precipitation trends and fore-casts; streamflow, lake kand reservoir) level, and snowpack trends and forecasts; soil and plant (including forest)moisture conditions; soil quality conditions (e.g., mineralcontent, depth of topsoil); crop conditions; and overalldrought indices (e.g., the Palmer drought severity index).The information could be presented on a county, State,regional, and national (and international) level, and wouldbe ideally suited for use with analytical and presentationsoftware (e.g., using spreadsheet or graphics techniques).

Agency pilot tests and experience in other areas suggestseveral prototypes for electronic dissemination that couldbe applied to drought (or other) time-sensitive FederalSTI. The "best" approach depends on the type ofinformation, number and types of users, importance to theagency mission and statutory guidance, agency prece-dents, budgetary constraints, related private sector alter-natives, and the historical and political context.

A weekly or monthly drought bulletin could be madeavailable on an agency electronic bulletin board fordial-up access by users over commercial telecommunica-tion lines. This approach is used, for example, by the

111111M1111111111111111111111111, 711111,111111111

Electronic technologies open up newalternatives for dissemination of time-sensitive Federal ST!, such as droughtinformation.

National Science Foundation's "science indicators"bulletin board and the Department of Commerce's"economic bulletin board." Or the drought bulletin boardcould be disseminated online via the computer center ofa single agency contractor, an approach used by theSecurities and Exchange Commission (for corporatefinancial information) and USDA (for various agriculturalreports and bulletins). Alternatively, an agency computercenter could be employed. For instance, the Environ-mental Protection Agency is making its "toxic releaseinventory" available online via the National Library ofMedicine computer center. And the U.S. GeologicalSurvey provides data on earthquake epicenters onlinefrom the National Earthquake Information Center inGolden, Colorado. Finally, the drought bulletin boardcould be offered as a service of private sector commercialor not-for-profit value-added information gateways andvendors.17

Forest Monitoring InformationConcern about forest ecology is growing. The need for

monitoring forest healtl- reflects: 1) mounting evidence offorest decline due to changing environmental, climatic,and soil conditions, among other factors; and 2) growingappreciation of the key role of forests in stabilizingecosystems and climate on local to global scales.15

Because of the rural, remote location of most forests,monitoring in these areas has, until recently, beenlogistically difficult and expensive. Advances in informa-tion and telecommunication technologies now makz itpossible to provide for efficient and cost-effective moni-toring of remote locations. For example, before themicrocomputer revolution, temperature and precipitationtypically were measured at remote locations with ahy drothermograph that recorded the data on chart paper.In order to retrieve that data, someone had to periodicallyvisit the monitoring site, tear off the chart paper, and carrythe chart paper back to a central location where the data

"Sec statements of Edward J. Hanley, Director Office of Inferniation Resources Management, U.S. Environmental Protection Agency, John

Penhollow, Director, Office of EDGAR Management. U.S. SecuT :..es and Exchangc Commission, and John J. Franke, Assistant Secretary ofAdministration, U.S. Department of Agriculture, before a hearing of the Subcommittee on Government Information, Justice, and Agriculture, House

Committee on Government Operation.s, Apr. 18, 1989. Also sec U.S. Department of Commerce, Under Secretary for Economic Affairs, "Rtxpiest for

Comments on the Preliminary Implementation Plan of Subtitle E, Part I of the Omnibus Trade and Competitiveness Act of 1988, thc National Trade

Data Bank," Apr, 21, 1989,IsSce U.S. Department of Agricultum, Forest Service, "Forest Productivity and Health in a Changing Atmospheric Environment," Conceptual Plan

for the Fortst/Atmosphare Interaction Priority Research Program, 1988.17


Advances in information and telecom-munication technologies now make itpossible to provide for efficient andcost-effective monitoring of remote loca-tions.

was extracted and logged in. The data had to be typed inif computer processing (e.g., calculation of mew-9 andstandard deviations) was desired. This largely mechanicaland manual process was (and is, where still used)labor-intensive, expensive, and prone to data qualityproblems. Frequently, the data were collected onlysporadically (e.g., by forest rangers on trail maintenanceduty), anl was logged in and compiled on an erratic basis.As a result, temperature and precipitation records formany rural monitoring stations were incomplete ornonexi stent.

Microcomputers and satellites are transforming thenature of remote monitoring and providing a valuableinformation resource for both research and managementpurposes. Digital data recorders (at under $1,000 each) arereplacing the hydrographs. These simple devices recordenvironmental and climatic data on a removable digitalchip that stores 45 to 60 days of data. The chips still haveto be picked up by someone and carried to an office, butonce there, the data can be entered directly from the chipinto a microcomputer for data storage and manipulation.Daily, monthly, seasonal, and annual means, among otherstatistics, can be readily calculated. Remote automatedmonitoring /stations (costing $10,000 to $15,000 each,with a satellite dish) can be used to electronically collectand transmit the data via satellite to a central location. Forexample, the U.S. Interagency Fire Center in Boise,Idaho, collects climate data via a NOAA satellite fromabout 500 remote weather stations located on forestedlands (260 stations operated by the U.S. Forest Service,200 by the Bureau of Land Management, and 40 by theNational Park Service and Fish and Wildlife Service\

Both the Fotest Service and National Science Founda-tion are funding rural monitoring networks to meet forestecology and climate research needs. The Forest Servicehas a "long-term forest ecosystem monitoring program"to measure selected physical and biological parameters,establish baseline conditions, and detect changes overtime. The program builds on existini; stations located inexperimental forests, experimental rangelands, and re-

search natural areas. Most stations require equipmentupgrades and more consistent data collection and analy-sis. For example, only 16 of 83 experimental forests havelong-term data sets (i.e., more than 15 to 20 years), andsome of these sets are incomplete or currently inactive.

The monitoring information will help researchers andresource managers detern..'ne changes in forest composi-tion and species distribution resulting from air pollutioa,climate, and soil changes. This in turn should beimportant input to ecological models used to predict forestgrowth, commercial tree yield, forest and rangelandhabitat and carrying capacity, fire frequency and severity,recreational opportunities, and the like.

To the extent possible, experimental forest monitoringstations are being colocated with the "long-term ecologi-cal research" field sites funded by NSF. These sites aredesigned to study the ecology of a diversity of naturallandscapesincluding forest, prairie, desert, and aquaticenvironments. The intent is to better understand thepatterns of organic decomposition, primary photosyn-thetic production, food webs, biogeochemical cycling,nutrient movement through soils and groundwater, at-mospheric deposition, and climatic change." Standard-ized meteorological data collection is being implementedat the various ecological sites, so that baseline conditionsare adequate fe. detection (and documentation) of bothcyclic and loL-tenn climatic change." The ecologicalstations together with the more well-established experi-mental forest stations provide reasonably balanced geo-graphic coverage of the United States, with one or morestations in the following States: Alaska, Washington,Oregon, California, Idaho, Arizona, New Mexico, Kan-sas, Minnesota, Wisconsin, Michigan, Illinois, Missis-sippi, Georgia, South Carolina, North Carolina, Virginia,West Virginia, New York, Massachusetts, and NewHampshire.

The Forest Service and NSF include the disseminationof monitoring information to users as an importantobjective. Advances in information technology can makethis a cost-effective reality. For example, the monthly datafor each monitoring station could be sent via inexpensivefloppy disk or electronic mail or bulletin board to adesignated central location. This could be a Forest Serviceoffice, or a research university, or the National ClimaticData Center (operated by NOAA) or National TechnicalInformation Service. The central office or center wouldquality control and consolidate the data on one disk ormagnetic tape. Depending on the data volume anddemand, the consolidated data could be issued periodi-cally in floppy disk, magnetic tape, and/or CD-ROA.

19Soe James C. Halfpenny and Kathryn P. Ingraham (eds.), Long-Term Ecological Research in the United States: A Network of Research Sites,Forest

Sciences Laboratory, Corvallis, OR, 1984; James T. Callahan, "Long-Tenn EcologicalResearch," BioScience, vol. 34, No. 6, June 1984, pp. 363-367.

20David Greenland (ed.), "The Climate of the Long-Term Ecological Research Sites," Occasional Paper No. 44, Institute of Arctic and Alpine

Research, University of Colorado, 1987. 72

App. ATechnological Opportunities for Managing and Disseminating Federal Scientific and Technical Infonnation 65

fonnats. An electrow!c bulletin board could also becost-effective for disseminating monthly data sets.

In sum, electronic information technologies both: 1)help make the forest and ecological monitorbg system areality, and 2) help make it possible for the monitoringresults to be shared among the research community andother public- and private-sector users at little marginalcost to either the government or the users.

Energy Research DocumentsElectronic information technologies also open up new

possibilities for the dissemination of Federal scientificand technical documents that traditionally have beenmaintained in paper and microfiche formats. An estimated200,000 such documents are generated annually, withmore than half of the total originating from the Depart-nleigt of Energy, Department of Defense, or NASA.

Advancing technologies create new alternatives forelectronic dissemination of both Federal STI biblio-graphic databases and the STI documents themselves. Theactivities of the DOE Office of Scientific and TechnicalInformation are illustrative. DOWOSTI currently distrib-utes about 14,000 documents per year in paper ormicrofiche format to NTIS and in microfiche to theDepository Library Program (DLP). Abstracts of thedocuments are included in both the DOE bibliographicdatabase called "Energy Data Base" and the NTISbibliographic database. While the depository librariesreceive paper copies of Energy Research Abstracts, whichcontain abstracts of DOE-funded research, the librarieshave online access to the r OE and NTIS bibliographicdatabases only through private vendors at commercialrates.

To meet its own internal needs, DOE has implementedan Integrated Technical Information System (ITIS),which provides DOE employees and contractors Vthonline access to the most recent 14 months of the EnergyData Base. DOE has proposed a pilot test to offerdepository libraries similar online access. Besides timelyaccess to the Energy DatP., Base (compared with the paperformat Energy Research Abstracts), the pilot wouldprovide an electronic "gateway" to archival energyresearch summaries (maintained on a database by acommercial vendor), and "electronic cataloging" ofDOE documents in a format compatible with that used bydepository libraries (and the Library of Congress).21

Another aspect of the DOE pilot test is a study ofalternative formats for document distribution. Over thenext few years, DOE, like other Federal science agencies,has the opportunity to convert from paper and microfiche

Advancing technologies create new al-ternatives for electronic disseminationof both Federal STI bibliographic data-bases and the ST1 documents them-selves.

to .optical disk as the primary document format. Onepossibility is to require DOE research offices, laborato-ries, and contractors to submit all documents in anelectronic form (e.g., magnetic tape, online, diskette) thatcan easily be convened to high-density optical disks (e.g.,WORM or .1D-ROM). Since the demand for STI docu-ments is generally small, any desired paper copies couldbe printed on demand. (The more popular documentscould be printed in larger volumes with traditionalprinting processes.)

The study may show, as a hypothetical example, thatDOE could distribute copies of the documents via abimonthly CD-ROM, rather than on microfiche. Astandard double-sided CD-ROM can store about 300,000pages of material (double-spaced, typewritten) or about1,500 documents at 200 pages per document. Thus the14,000 documents could fit on about 10 CD-ROMs. TheCD-ROM cost probably would be significantly lowerthan microfiche (and much lower than paper). At present,DOE pays about $350,000 per year for microficheproduction of depository library materials, compared toan estimated $210,000 for mastering CD-ROMs andduplicating 1,400 copies of each (one per depositorylibrary). If DOE was able to piggyback depositoryCD-ROM duplication onto mastering and production forinternal and possibly NTIS needs, the cost could be evenlower (and savings greater). Compared to micmfiche,CD-ROM should be easier to use, permit full-textsearching, and provide higher quality document resolu-tion (on the screen or when printed out on demand).

One disadvantage of using a bimonthly CD-ROM is theup to 2-month delay in getting some energy researchdocuments to the depository libraries (and other users).This delay could be alleviated by maintaining the mostrecent 2 (or perhaps 4) months of documents online infull-text format, for retrieval and printing on-demand.Many private vendors are adopting a similar approach,whiut combines the strengths of online with CD-ROMformats. Another possible disadvantage is that all partici-pating depository libraries (and other wers) would needto have adequate CD-ROM facilities ;one or several

21U.S. Department of Energy, Office of Scientific and Technical Information, "DOE/Depository Library Gateway: Access tol.)0E R&D Results 4.n

Electronic Fe .n, A Pilot Project Proposal," August 1986; U.S. Congress, .1 at Committee on Printing, "Dissemination of Information in Electronic

Foimat to Federal Depository Libraries: Proposed Project Descriptions," hale 1988."


microcomputer, CD-ROM drive, and local printer set-ups, depending on the level of use). As CD-ROM readerscontinuc to drop in price and become standard equipmenton microcomputers, the availability of CD-ROM equip-ment will improve, at least in the larger research libraries.Special provisions may be neededwhether through theDI P or otherwiseto ensure that smaller, rural, oreconomically disadvantaged libraries have CD-ROMequipment.

An inherent advantage of electronic formats such asCD-ROM is that powerful bibliographic, retrieval, andeven expert search system software can be includeddirectly on the optical disk or loaded into the microcom-puter via diskette. CD-ROM or online versions of the"Grateful Med" user-friendly software developed by theNational Library of Medicine (NLM) will be common-place, whether developed by the government and/orprivate vendors. NLM developed "Grateful Med" tofacilitate user access to MEDLINE and other databases onthe NLM MEDLARS (MEDical Literature And RetrievalSystem). Tens of thousands of copies at $29.95 each havebeen sold through NTIS. The package includes 2 floppydisks, a user's guide, and an application for a MEDLARSaccess code. The capabilities of user-friendly softwaresuch as "Grateful Med" or numerous commercialsoftware packages can be easily replicated on CD-ROM.

In considering the appropriate role for Federal agenciesin online dissemination of STI bibliographic databases,three aspects warrant particular attention. First, most ofthe Federal scimtific and technical agencies have astatutory charter and/or mission objective to promote thewide distribution of information on the results of Federalresearch and development. Even agencies that operateunder restrictions (e.g., NASA) have a strong dissemina-tion mandate. Bibliographic databases are key tools Ltfacilitating access to information on R&D results, andonline databases (or for some purposes CD-ROM) offersignificant advantages in terms of timeliness and ease ofsearch and retrieval. Thus agencies need to be sensitive toequity of access to Federal STI, and ensure that, whatevermeans of online dissemination may be employed, certainuser groups are not disadvantaged. Students, teachers,retired scientists, small business persons, and the like mayneed special consideration.

Second, development and dissemination of onlinebibliographic databases (and now CD-ROM versions ofsame) are strengths of the private commercial andnot-for-profit information industry. A wide range ofexcellent STI bibliographic databases has been developedby private vendors that offer a portfolio of STI databases(including some from Federal agencies) over informationgateways and value-added networks. Again, equity of

access is a concern since full commercial onlhe rates canrange from $75 to $150 per hour or higher for privatelydeveloped databases, and commercial rates range fromabout $40 to $80 per hour for government databases (twoto four times the comparable government rate). On theother hand, commercial vendors increasingly are propos-ing or offering a variety of discounts for off-peak or bulkvolume use, that are more affordable for students,teachers, and the general public. Private sector not-for-profit vendors int providing some databases at ratesbetween full commercial and governmental levels.

Third, a Federal STI bibliographic database may ormay not be less expensive if offered online by thegovernment. There is no clear-cut answer. Each situationrequires individual ahalysis. For example, adding anonline database to an already existing online computercapability (e.g., at NLM) or providing expanded access toan existing online system (e.g., depository library accessto the DOE system) may have minimal marginal costs, ifthe existing computer center could handle the additionalfile and/or users without costly upgrades or expansion. Inthese situations, the incremental or marginal cost ofadditional computer use may be minimal, and competitivewith comparable private-sector costs. On the other hand,if this required an rpgrade of agency computer capability,the cost could be higher. For setting up a small electronicbulletin board, the cost of a new system is likely to bemodest, but for a large, heavily used bibliographicdatabase, the cost could be substantial. In makingdecfrions on online bibliographic (or other online)sys s, agencies will need to consider the quality ofset% , agency mission, equity of access, and relatedprivate-sector activities, in addition to cost-effectiveness.

With respect to CD-ROM (and other optical storagemedia), the situation is clearer. It seems likely that forsome types of Federal information, and especially variousSTI documents, high-density optical storage will largelysupplant paper and microfiche. It is not a question ofwhether this will happen, but when. Federal agencies will,in all probability, make this transition themselves in orderto Aneet their statutory mission and records managementresponsibilities. The agencies may employ any of severalmeans to make this transition, including private contrac-tors, NTIS, and/or GPO. But the end result is likely to bethe availability of many or most Federal STI documentson optical disk, at affordable prices, with powerfulbuilt-in search and retrieval capabilities, that will becost-effective compared to paper or microfiche. Thisupgrade may also offer many new opportunities for theprivate sector to develop more value-added applicationsand products.

74

Appendix B

List of AcronymsANMINImm

Acronyms

CALS Computer-Aided Acquisition andLogistical Support

CD-I Compact Disk-InteractiveCD-ROM Compact Disk-Read Only MemoryCENDI Commerce, Energy, NTIS, Defense

InformationCOSATI Committee on Scientific and Technical

InformationDLP Depository Library ProgramDoD Department of DefenseDOE Department of EnergyDTIC Defense Technical Information CenterEDB Energy Data BaseEDI Electronic Data InterchangeEDGAR Electronic Data Gathering and RetrievalEPA Environmental Protection AgencyEOS Earth Observing SystemEROS Earth Resources Observation SystemsEIDE Energy Technology Data ExchangeFCCSET Federal Coordinating Committee on

Science, Engineering, and TechnologyPIPS Federal Information Processing StandardFOIA Freedom of Information ActGIS Geographic Information SystemsGPO Government Printing OfficeIIA Information Industry AssociationIRM Information Resources ManagementJCP Joint Committee on PrintingLOC Library of CongressNAL National Agricultural LibraryNARA National Archives and Records

Administration

NASA

NGDCNIST

NLMNOAA

NODCNSANSFNSSDCNTISURA

OMBOSIOTAOSTPPSACPRANDR&DSDISupDocsSTIS&TTRIUSGSWORM

National Aeronautics and SpaceAdministration

National Geophysical Data CenterNational Institute of Stvlards and

TechnologyNational Library of bciedicineNational Oceanic and Atmospheric

AdministrationNational Oceanographic Data CenterNational Security AgencyNational Science FoundationNational Space Science Data CenterNational Technical Information ServiceOffice of Information and Regulatory

AffairsOffice of Management and BudgetOpen Systems InterfaceOffice of Technology AssessmentOffice of Science and Technology PolicyPresident's Science Advisory CommitteePaperwork Reduction ActPatent and Trademark OfficeResearch and DevelopmentSelective Dissemination of InformationSuperintendent of DocumentsScientific and Technical InformationScience and TechnologyToxic Release InventoryUnited States Geological SurveyWrite Once-Read Manytimes

75

Index

committees on STI, 1, 43-47CENDI (Commerce, Energy, NTIS, Defense Infortna-

tion), 4647FCCSET (Federal Coordinating Committee on Sci-

ence, Engineering, and Technology), 43-47ICCDC (Interagency Coordinating Committee on Dig-

ital Cartography), 47, 48, 50, 59IWGDMGC (Interagency Working Group on Data

Management for Global Change), 44, 47, 51-52,55-56

PCAST (President's Council of Advisors on Scienceand Technology), 43-47

competitiveness, 1, 3, 5, 7-18, 33-36, 45, 49-50, 52, 55-56Defense, Department of

open flow of STI, 33-36use of STI, 11, 48, 65

Depository Library Program, 4, 29, 36-42, 65-66earth sciences information, 2, 5, 11-12, 15-16, 26-27,

37-40, 44, 50-53, 55-56, 59-63drought monitoring, 63forest monitoring, 63-65

Energy, Department ofopen flow of STI, 33-36use of STI, 6, 11-12, 34-35, 37-42, 47, u.)-66

Federal science agenciesand OMB, 1, 3, 19-23, 47-49and OSTP, 1, 43-49and governmentwide dissemination agencies, 36-42open flow of STI, 33-36role in science education, 10, 52-55STI indices/directories, 51-52STI management, 46-49STI standards, 49-51

Freedom of Information Act, 19, 35-36Government Printing Office, 1, 3, 6, 36-42, 50indexing STI, 2, 38, 51-52information dissemi tation policy

Commerce, Department of, draft, 21-22diversity of STI sources, 24-25government/private-sector roles, 3, 15-18, 1925,- 30-

31, 41-42governmentwide dissemination agencies, 36-42intellectual property, 29-30, 34-36user charges, 26-29, 41-42user inw ivement, 25, 41-42value-added, 24, 30-31

information literacy, 5, 9-10, 52-55international exchange of STI, 8-9, 15-18, 33-36, 55-56,

61

=1.Landsat imagery, 2

pricing, 26storage technology, 62-63

libraries, 4, 6, 9-10, 14-15, 25, 30 .31, 40-42, 48, 52-55, 58National Aeronautics and Space Administration

National Space Science Data Center, 3, 4, 30, 37,50-51, 60-61

open flow of ST1, 36role in STI, 13-14, 37-40, 47, 50-51space science data, 60-61

National Agricultural Librarypricing, 29-30role in STI, 37-40

National Archives and Records Administration, 4, 38,4041

National Ubraty of MedicineMEDL1NE database, 15-16, 28-30, 66pricing, 27-30role in STI, 6, 37-40

National Institute of Standards and Technologycomputer security, 33standards, 50-51

National Oceanic and Atmospheric Administrationpricing, 27-28, 30role in STI, 37-40, 61-63

National Technical Information Servicepricing, 27-29role in 511, 4, 38-40

Office of Management and Budgetand OSTP, 1, 35, 47-49and science agencies, 1-3, 19-23, 41, 47-49Circular A-130, 19-23open flow of 511, 3-4, 16-17, 22, 33-36, 55-56

Office of Science and Technology Policyand OMB, 1, 35, 47-49and science agencies, 46-49open flow of STI, 33-36, 55-56role in STI policy, 1, 38, 43-56

online dissemination, 1, 5, 9, 14, 28, 29-30, 34-35, 40-42,51, 53-54, 57-59, 61, 63-66

optical disks, 1, 3, 4, 5, 9, 14, 27-28, 29-30, 38-39, 40-42,48-49, 51-52, 53-54, 57-61, 62-63, 65-66

Paperwork Reduction Act, 3, 19-23, 38, 45Printing Act, 19, 38, 41, 45science education, 5, 9-10, 52-55standards, 2, 14, 49-51technology transfer, 5, 13-11,, 33-36, 45users of STI, 2, 10-17, 24-29, 36-42, 47-52, 59-66

-68-U.3. GINETNENT PRINTING %ACE: MO 0-4$0.106

Ortier Proctivoq Code

Superintendent of Documents Publications Order FormCharge your order.

It's easy!To fax your orders and inquiries(202) 275-0019

6796YES, please send me the following indicated publications:

1.3

Helping America Compete: The Role of Federal Scientific and Technical Information (80 pages)GPO stock number 052-003-011M-4; price $3.75.

Please send me your Free Catalog of hundreds of bestselling Government books.

The total cost of my order is $ (International customers please add 25%.) Prices include regular domestic postage andhandling and are good through 12/90 After this date, please call Order and Information Desk at 202-783-3238 to verify prices.

Please Choose Method of Payment:

(Company or personal name) (Please type or print) ri Check payable to the Superintendent of Documents

n GPO Deposit Account 171771- Li(Additional address/attention line)

(Street address)

(City, State. ZIP Code)

(Daytime phone including area code)

VISA or MasterCard Account

ri Ihr I LL_I

(Credit card expiration date)Thank you for your order!

(Signature) 6/90

Mail To: Superintendent of Documents, Government Printing Office, Washington, DC 20402-9325

77

Office of Technology Assessment

The ()litre (il I rulinology Assessment (()T.N) \V IS (1. Cilled in 1972 as ananalytical arm of (:oligress. ()TA's basic ninction is to help legislative policy-makers anticipate and plan for the consequences of-technological changes andto examine the ('XI)(Tteti all(i 1.1Il(Xi)(11(qi, technologyaffects people's lives. The assessment of technology calls lOr exploration oltill 1)1lysi"11, cunnUnnu, snuiai, and l'Unntid inlnac" rcsldtfrom applications ol scientific kno\vledge. (YTA provides (:ongress \vitli III-(1(1)(11(1('I1t and timely infOrmation about the potential effectsboth benefi-cial and harmful technological applications.

Requests 1o...studies are made by chai,ailen st.inding committees of theI i()Ilse UI 1'.(1)lvtici)1;01 \'es ()I 1)1' 1111' 'Technology Assessment Board.

da' g"vcrning nnd). (n ()TA: Ur by On' 1 )irrutur "I ( )TA iii ronsultatinn \vidithr Board.

The Technology ,Assessment Board is composed of six members of the!louse, six memb(rs of the Senate, and tiw ( )TA 1 )irector, h(l is a non-voting member.

()TA ii:r; 5111(11c5 lindri. \VIIV III "'Mr ""(1,i15 : industry. techin hogy, and employment: international security and coin-

biological applications: u iod ;aid renewable resources:(Litimullication and information technologies-, oceans ;Ind environment; andsun nue, ednuation, and transportation.

78

h************************************************** · 2014. 4. 9. · DOCUMENT RESUME. IR 015 075....

Documents

Transcript of h************************************************** · 2014. 4. 9. · DOCUMENT RESUME. IR 015 075....