1965

1966. 1967

TABLE II.-Preliminary projection of NSRDS program costs

[Millions of dollars]

Ultimate level.

1 Includes 330K OA funds.

2 Includes 380K OA funds.

BASIC ASSUMPTIONS USED IN DEVELOPMENT OF FIVE YEAR FINANCIAL PLAN FOR THE NATIONAL STANDARD REFERENCE DATA SYSTEM

In developing the five year projection for the needs of the National Standard Reference Data System the following assumptions were taken as basic ground rules:

(1) That competent individuals can be located for data compilation projects. Up to the present, the program has been limited by the availability of financial resources. At some stage the availability of competent persons will become the limiting factor on further expansion. There is no way to tell at present whether this limit will be reached at a level which is 2, 4, or 10 times greater than the present.

(2) That only higher priority projects would be supported during the next five years.

In each technical area it was assumed that 100% coverage of all types of quantitative data would probably not be justifiable on a cost benefit basis and therefore, that only the areas judged to be of urgent or high priority would be covered.

(3) That the technical scope of the program would remain essentially constant. No extension of the program to data on partially defined materials or on structure sensitive properties was envisioned in developing the 5-year projection.

(4) That user studies would show that sophisticated storage and retrieval systems are justified on a cost benefit basis.

A relatively large fraction of the estimate for office expense is associated with putting data into computerized storage and with the development of software for retrieving the data in a variety of ways. Costs for hardware and for communications equipment are also included.

(5) That information services would be established using appropriated funds, with user charges beginning to be an important factor only toward the end of the five-year period.

Under this assumption the costs associated with the development of information services would be borne by appropriated funds. User charges would cover only out-of-pocket costs associated with specific service requests.

ADDITIONAL QUESTIONS SUBMITTED TO DR. ALLEN V. ASTIN, DIRECTOR, NATIONAL BUREAU OF STANDARDS, DEPARTMENT OF COMMERCE, BY THE SUBCOMMITTEE ON SCIENCE, RESEARCH AND DEVELOPMENT

1. If an individual, organization, or agency does not consent to the Bureau's use of their reference data services or materials as provided in section 3 of the bill, what action will the Bureau take in such a case?

The National Bureau of Standards will promote and encourage voluntary participation by all concerned. If any individual, organization, or agency does not wish to make its data resources available to the National Bureau of Standards for incorporation into the National Standard Reference Data System, no attempt will be made to coerce them into doing so. Participation is entirely voluntary with the consent of all participants. If an industrial organization has a body of data which they do not wish to make available to the System, that is entirely their own choice. Again, if a data compilation does not meet the standards of the NSRDS but its compilers and users do not object because the work in its present form satisfies their requirements, this is their prerogative. If the topic is important enough to other users to justify additional action being taken, the operators of the National Standard Reference Data System would attempt to produce a compilation on the same subject that does meet standards.

2. Is it intended to "integrate" and "coordinate" privately prepared and published data without the participation of private concerns in the rulemaking process?

The intent of the Department of Commerce is that no action will be taken with respect to privately prepared and published data without the participation of and consent of individuals responsible for the data. The program is to be entirely voluntary and cooperative. In addition, since the whole system is user-oriented, every effort is being made to give users of data a chance to voice their interests, so as to provide guidance in selection of content and format of projects.

3. Since Standard Reference Data is international in scope, please describe the efforts going on in other countries, and in particular the work of the International Council of Scientific Unions and the Soviet Union's State System of Standard Reference Data.

The International Council of Scientific Unions has recently established an International Committee on Numerical Data for Science and Technology in order to assist the national efforts in several countries to maintain better contact with each other and to promote the expansion of this type of activity throughout the world. The Chairman of this Committee is Dean Frederick Rossini of Notre Dame University. Dean Rossini is also the Chairman of the general Review Committee for the NBS Office of Standard Reference Data and similarly is Chairman of the Executive Committee of the Office of Critical Tables of the National Academy of Sciences-National Research Council. He has been a leader in scientific data compilation work for many years.

Last year the Soviet Union established a State System of Standard Reference Data, patterned in many respects after the plan of the National Bureau of Standards. The Soviet plan of operation apparently envisions a much more extensive experimental program than does the NBS program. In the U.S.S.R. central direction is to be provided for experimental measurements of the properties of substances of all kinds, in addition to the central coordination and direction provided for data compilation activities. The Soviet program also apparently envisions coverage of a considerably wider scope of science and technology than does the present U.S. program.

Several of the more advanced countries of the world have active data compilation projects in various fields, some under government auspices and some private. In the United Kingdom, the Department of Education and Science has established an Office of Scientific and Technical Information, one of whose functions is to promote data compilation activities in the U.K. A French

organization has had a small but effective program for many years. A small group in Germany has organized the most effective and comprehensive privately sponsored program in the world. This program (The Landolt-Bernstein Tabellen) has been in existence since 1883. In Japan considerable interest has been expressed as a result of discussions held under the U.S.-Japan Scientific Exchange Agreement.

(a) How is coordination effected between such international organizations and the National Bureau of Standards?

The principal channel for coordination between multi-lateral international organizations and the National Bureau of Standards is through the National Academy of Sciences and its Office of Critical Tables. The Office of Critical Tables has been in existence since 1957 and has established relationships with most of the effective working groups throughout the world. The Director of the Office of Critical Tables has recently undertaken an additional responsibility to be part-time director of the staff of the new ICSU International Committee on Numerical Data for Science and Technology. The staff office is to be located in Washington, D.C., in housing provided by the National Academy of Sciences for the initial few years of existence of the Committee, after which time it is to be removed to a yet-to-be-determined location in Western Europe. Additional contact and coordination with the ICSU activity is provided through the activities of Dr. Frederick Rossini, Dean of Science at Notre Dame University, who as already mentioned, serves as Chairman of the ICSU Committee as well as Chairman of the NBS general Review Committee. Liaison with activities in other countries is also maintained through occasional visits of members of the staff of the National Bureau of Standards to important projects in other countries and similar visits by active workers in other countries to the National Bureau of Standards and to the individual data compilation projects under NBS auspices.

(b) What role does the State Department play in these international activities?

Foreign policy guidance would be expected from the State Department. The State Department Office of International Scientific Affairs has encouraged the development of international cooperative activities in this field. Members of the staff of this State Department Office are kept informed in full in the development of such activities but in general do not participate in the scientific meetings that are held. In addition, the Science Attaches of the United States, and those of other countries stationed in Washington, often provide communication channels between individuals and groups in the active countries. 4. To what extent have other Federal agencies financially supported the Bureau's standard reference data system and is such support expected to continue in the future?

First year support for the National Standard Reference Data System was provided by contributions of $200,000 each from four other government agencies-DOD, AEC, NSF, and NASA. During FY 1966 the Department of Defense provided an additional $300,000 and the NASA an additional $100,000 for general support of the program. Beyond that other government agencies have supplied smaller amounts of support for specific data compilation projects to the following extent: ARPA 150K and AEC 118K.

In addition, the Atomic Energy Commission, the Department of Defense, and the National Aeronautics and Space Administration have a number of high quality data compilation projects in special areas. This support has not gone through the National Bureau of Standards but does lead to the production of compilations of data which are part of the system.

5. Page 249 of the Appendix to the Federal Budget for Fiscal Year 1967 shows that $7.4 million was spent on Standard Reference Data in Fiscal Year 1965; an estimated $8.2 million will be spent in Fiscal Year 1966, and that the Bureau is requesting $9.1 million for Fiscal Year 1967. (a) Do the obove amounts represent the Bureau's funding of the system as envisioned in H.R. 15638? If not, please discuss where the dividing line exists in terms of work or functions performed?

The amounts listed in this question include two major categories of activities within the National Bureau of Standards: (a) the projects of the National Standard Reference Data System and (b) experimental measurements of properties of the kind that provide input to the compilation projects of the NSRDS.

Naturally these experimental measurements serve other functions as well as supplying input to the NSRDS. They provide key data for the measurement system, they are a vital part of programs for development of measurement techniques, and are part of the recognized mission of the NBS. For Fiscal Years 1965 and 1966 the funds appropriated by Congress that were employed in the activities of the National Standard Reference Data System (that is, those of the type envisioned in the proposed Standard Reference Data Act) were $1.16 million for each year. The balance, outside of the scope of the National Standard Reference Data System, was spent in activities of type (b) mentioned above, that is, experimental determination of important properties of substances.

The activities of the National Standard Reference Data System provided guidance for the experimental work in the determination of the properties which needed to be measured either because measurements were lacking or because higher precision was required.

6. Will the functions authorized by this bill be subject to the Administrative Procedure Act (5 U.S.C. 1002)? If not, why?

Yes-particularly with respect to the bill's functions relating to public information and rule-making.

(a) Is the purpose of section 4 of the bill in implementation of, rather than in lieu of, section 3 of the Administrative Procedure Act as modified by the House on June 20, 1966?

Section 4 of H.R. 15638 implements, but does not replace section 3 of the Administrative Procedure Act. The section also provides the Secretary with explicit authority to issue standards, criteria and procedures for the preparation and publication of Standard Reference Data.

7. In what form is it intended that the data be "published," and how frequently will it be updated or supplemented?

The compilations produced by the various components of the National Standard Reference Data System will appear in various forms: monographs, books. loose-leaf data sheets, decks of punched cards, magnetic tapes, microfiche, and perhaps other forms as well. Any form which the users of the data would find to be convenient for their purposes would be considered appropriate for the program. The frequency of updating depends upon the rate of activity in the area under consideration. In some areas it might be as often as every six months, in others perhaps a frequency of updating of once in three or four years might be quite adequate. This, too, depends upon the needs of the users.

(a) How does one publish computer tapes or cards and what section of the bill authorizes the sale of such materials?

Computer tapes are not published in the same sense that books are. Instead, a duplicate of a master tape is prepared by running the master plus a blank tape through an appropriate machine. Several slave-stations can be controlled from the one master station, so that up to ten copies could be made at once. Essentially, this is a custom service operation.

Card decks are duplicated similarly by running a master deck through a duplicator-puncher, which senses the holes in the master cards, and punches corresponding holes in the duplicate cards. The duplicator-puncher can also be controlled by a magnetic tape or a punched paper tape. It is intended that section 5 of the proposed Act cover standard reference data in these forms as well as in the form of a printed page.

8. Has any published material of the National Bureau of Standards ever been duplicated or reprinted by a commercial publishing organization?

(a) If so, what percentage would this be of the total material published by NBS?

The percentage of the Bureau's published material that has been reprinted by commercial publishing organizations is low, but this has occurred on occasions. The most recent examples we are aware of are the 1046 page "Handbook of Mathematical Functions" that was duplicated by Dover publications, Inc., and the 478 page "Tables of Thermodynamic and Transport Properties of Air, Argon, Carbon Dioxide, Carbon Monoxide, Hydrogen, Nitrogen, Öxygen, and Steam," copied by Pergamon Press Ltd. We understand that a commercial publisher currently is considering duplication of the 470 page monograph, "Inospheric Radio Propagation."

(b) Was the material duplicated in such a way that the public might believe it to be a publication of NBS rather than the publishing company? We are not yet aware of any problem from this source.

(c) If a commercial organization attempted to fradulently pass off its duplicated material as a publication of NBS, wouldn't the normal legal remedies be available to the government in such a case?

If it is clear that the method of sale of the duplicated material was intended to deceive the public, the matter could be referred to the Federal Trade Commission or the Attorney General for consideration. Additional remedies may also be available.

(d) In light of the above questions, why are the enforcement provisions of this bill necessary to protect the integrity of standard reference data system?

Up to the present time, we have had no authority to recover our costs through the sale of publications, and the low price of Government publications has discouraged duplication by commercial publishers. If higher prices are charged, as proposed under H.R. 15638, more frequent and perhaps irresponsible duplication can be expected. The enforcement provisions of the bill are intended as a safeguard against this eventuality.

9. In determining the price to be charged for standard reference data, does the Bureau intend to charge the same price per publication to both foreign and domestic users? If not, please explain.

Yes, except for any extra costs incurred in supplying them to foreign users. (a) What percentage of the overall sales do you expect to receive from foreign purchasers? Is this estimate based on past experiences?

Overall sales to foreign purchasers might total 30% to 50% of total sales. This estimate takes into account the experience of U.S. and foreign technical book publishers.

10. What percentage of the purchasers of standard reference data do you expect to be institutional purchasers (corporations, libraries, etc.) and what percentage do you expect to be individual purchasers?

Institutional purchasers, such as corporations, libraries, etc., would probably account for approximately 2 of the sale of standard reference data. In many cases, however, some of these institutional purchases will include multiple copy orders for the use of individuals within the organization.

11. Please discuss to what extent computers are currently being utilized in the National Standard Reference Data System and what the plans are for the future? At present computers are being used in the NSRDS in the operations of certain individual data centers for the storage, retrieval, and analysis of data of special interest to those centers. Plans for the future include the following considerations:

(a) The development of communicating networks of data centers in which a computer based storage system at one location could be directly interrogated by a scientist at another center or by the computer system or another center to obtain data, references, or interpretive programs of common interest;

(b) An information services operation of the National Standard Reference Data System based on a substantially automated data file which will be used for the storage and retrieval of all the critically evaluated data products of the system. This data file which will be used as the source of information for the various custom services which have been discussed previously before this committee. The entire content of this data file may not be located in any single computer or computer storage adjunct; however if portions of the file are stored in different locations, intercommunication among the computer components would be established for immediate access to the whole file;

(c) As our knowledge of the physical universe increases, it becomes more and more possible to calculate certain physical properties (especially of very pure materials) from a knowledge of basic physical laws and a few fundamental properties. Computer systems will be fully utilized to permit such calculations whenever the reliability of results is equal to that of the measured properties; (d) In certain cases experimental measurements of physical properties can be performed at widely spaced intervals of temperature, pressure, and other variables, and well-known mathematical expressions can be employed with the extrapolation or interpretation of values. Computer storage of the skeleton tables and the interpolating or extrapolating rules and formulas will be the most efficient way to store some standard reference data.