Dr. JONES. Mr. Chairman, we would go according to your will. I would suggest we go ahead through these recommendations and then open discussion. But, nevertheless, whatever you wish.

Mr. DAVIS. That is perfectly agreeable.

Dr. JONES. Thank you very much.

Mr. MOSHER. Mr. Chairman, just let me clarify this in my mind. I assume this discussion that Dr. Drucker has just given and his recommendation supports the item in the National Science Foundation budget for 1973 of around $20 million for experimenting to find innovative ways to attract the cooperation of industry and the private sector in general in research and development. Is that true?

Dr. DRUCKER. Yes, sir; it does, and I suppose the National Bureau of Standards is working for the same purpose.

Mr. MOSHER. But this backs up that whole new direction in the administration programs this year.

Dr. DRUCKER. Yes. We hope we have had some hand perhaps in fashioning this.

Mr. MOSHER. Good.

Dr. JONES. Mr. Chairman, I would like to turn now from the academic scientist-engineer-technologist, to an industrial scientist and technologist for a recommendation, too. I would like to present Dr. J. Ross Macdonald, of Texas Instruments, Inc., vice president of Corporate Research and Engineering and director of the Central Research Laboratory, to get a view from an entirely different sector of our

economy.

(The biographical sketch of Dr. Macdonald follows:)

DR. J. ROSS MACDONALD-TEXAS INSTRUMENTS, INC.

Dr. Macdonald, who was born in Georgia, received a B.A. in Physics (Williams College) and a S.B. in E.E. (M.I.T.) in 1944. He served as a Navy RadioRadar officer during World War II and returned to M.I.T. in 1946. He was awarded a S.M. degree in E.E. from MIT in 1947 and then studied in the M.I.T. Physics Department until he received a Rhodes Scholarship in 1948. Oxford University granted him a D. Phil. degree (solid state physics) in 1950 and the D.Sc. degree, based on published papers, in 1967. He worked at Armour Research Foundation and Argonne National Laboratory before joining Texas Instruments in 1953. He served as Director of the TI Physics Research Laboratory for several years and is currently Director of the Central Research Laboratories and Vice President, Corporate Research and Engineering. In addition, he has been an adjunct associate professor of biophysics at the Southwestern Medical School of the University of Texas since 1954 and became an adjunct professor in 1971. His experimental work has been primarily in the areas of ferromagnetic resonance, photo Hall effect, electrical behavior of solids, and electric circuits. He has made theoretical contributions to the areas of dielectric and mechanical relaxation, space charge behavior in solids and liquids, electrical response of electronic devices, electrolyte double layers, equations of state, and numerical analysis of experiments. He has published over 90 papers in more than 30 different scientific and engineering journals.

Main memberships and outside activities: Fellow, The American Physical Society; Fellow, Institute of Electrical and Electronics Engineers; Fellow, American Association for the Advancement of Science; Member, Phi Beta Kappa and Sigma Xi; Member, National Academy of Engineering; Member of NAE Council; Member, NAS-NRC Solid State Sciences Panel; Member, AIP Advisory Committee on Corporate Associates; Member, NSF Advisory Committee for Science Education; Chairman, NAS-NAE-NRC Numerical Data Advisory Board; Member, NAS-NRC Committee on Motor Vehicle Emissions; Member, Visiting Committee, M.I.T. Physics Department; and Achievement Award of the Institute of Radio Engineers.

STATEMENT OF DR. J. ROSS MACDONALD, VICE PRESIDENT OF CORPORATE RESEARCH AND ENGINEERING AND DIRECTOR, CENTRAL RESEARCH LABORATORY, TEXAS INSTRUMENTS, INC., DALLAS, TEX.

Dr. MACDONALD. Thank you, Tom.

Mr. Chairman and members of the committee, I am delighted to participate in this discussion on the Fourth National Science Board Report. It is a particular pleasure to be asked to speak on this report since I had a small part in working on it and its revisions.

Because of this experience, which gave me a close familiarity with the report, but of course no bias, I feel qualified to state that it is the best of its kind I have read, an eminently sensible and practical statement of important national needs and of how technology can be better brought to bear in helping achieve their solution.

Now let me make some specific comments on Recommendation II of the report, which deals with technological support for public goods and services.

The public goods and services sector of the economy presents an opportunity for major innovation and contribution by the Federal Government primarily because it involves needs crucial to modern society. The number of persons providing vital functions such as health care and education, coupled with goods and services "purchased" collectively (for example, clean air, water, fire protection, civic order, city sanitation) has grown to 65 percent of our labor force. As this service sector has become the dominant part of our economy, it has been criticized for achieving a much lower productivity increase than has the manufacturing of capital and consumer goods for the private market. Some of the main characteristics of the service sector, which are relevant to its alleged low productivity and low increase in productivity, are:

1. The structure of the sector consists of many small, uncoordinated, perhaps even uncooperative parts. Often a service is person to person. A good example is food service.

2. A service, such as solid waste disposal, is often taken for granted because its purchase is indirect or collective.

3. Capital costs for providing services-for example, health and education are often great, creating a high risk for free enterprise development of the service or contribution to it.

4. The indirect, "not for profit" character of a service often permits it to exist without effective resource allocation and without periodic evaluation of its performance and efficiency. The educational enterprise is a prime example.

5. Technology for improving productivity in some areas of the public goods and services sector, such as transportation and housing, already exists, but its application often encounters institutional, legal, or individual barriers.

6. Very few measures exist for quantifying the value of a service. Take local government for example. How can one evaluate the effectiveness of the services provided by local government?

One can identify at least three major problems in applying technology effectively to the public sector. The first problem arises from lack of problem definition and measurement.

Since one can control and improve only what one can measure, let me continue on the subject of measures for a moment to show its present status, the need, and the application of new measures. The President's National Commission on Productivity, headed by George P. Shultz, recognized the need for effort on measurement of productivity.

In a recent release, the Commission said, "adequate measurement of productivity is lacking for major and growing parts of the economy-such as government, the various services, construction, trade, finance, and real estate. Adequate measurement and better information are needed on actual productivity trends and developments in each sector of the economy so that lagging sectors can be more clearly identified and practical efforts can be made to improve their productivity growth."

Hatry and Fisk of the Urban Institute prepared a preliminary report for the Productivity Commission entitled, "Improving Productivity and Productivity Measurement in Local Governments." This report, covering 30 major cities, clearly demonstrates the general absence of measurement and the need for new measures, and it gives preliminary examples of the power of possible new measures in assessing the productivity of the service. Twenty-three out of 30 major U.S. cities record workload data such as tons of garbage collected per year. Only seven of the 30 cities used cost per unit of workload, however, and only four of the 30 had effectiveness data. The Institute found only one case of a city using effectiveness data to prepare time-series information on productivity.

Clearly, the measurement area for public services is both underdeveloped and complex-this complexity is one of the reasons for the need for Government-supported research into the development of measures of effectiveness for operations in the public sector of the economy.

The second problem area of the public sector is "customer" education, where the customer is the diffuse set of public agencies, regional authorities, governing boards, et cetera, who represent the public. In attempting to penetrate public sector markets, industry must attempt to sell to customers who themselves do not really know what their constituents want or need.

Further, companies often approach the customer with inadequate or nonexistent prototype hardware-nonexistent largely because the satisfaction of the need requires a system rather than a component; hence, there is usually an enormous cost and risk involved in engaging in an effective prototype demonstration of the kind really needed.

A second important role of Government-funded R.&D. for the public goods and services sector thus is the development and application of needs analysis and system studies, together with appropriate customer education.

Lastly, the role of Federal technology development for the public sector must include the responsibility of providing answers. The two previous topics I have touched on-measurement and educating the customer about needs-only identify problems and tell us when we are. working on the right ones.

80-251-72- -2

We still must get on with the job of developing answers. Because the magnitudes of many of the public sector problems are so large, the governmental role should extend further than R. & D. alone-on into federally sponsored demonstration projects.

In summary, the diffuse, taken-for-granted, not-for-profit, high-risk character of many parts of this sector has largely precluded entrepreneurs and innovators from applying their talents to filling perceived needs. Government-supported research, development, and demonstration could provide the innovative keys essential to providing these services to society in a productive, satisfactory way. I hope my comments show some of the reasons for Recommendation II: namely: "Key technologies essential to the attainment of societal goals, but not presently commercially viable, should be continually developed, strengthened, and renewed through Government-aided research and development." Thank you.

Mr. DAVIS. Thank you, Dr. Macdonald.

Dr. JONES. Thank you, Ross.

Mr. Chairman, let me just mention that Dr. Macdonald was also a member of the special committee for developing this report, and contributed very greatly thereto.

I would now like to present for Recommendation III Dr. William K. Linvill, executive chairman of the Department of EngineeringEconomic Systems of Stanford. And lest this committee become confused, I would like to tell you several things about Bill. He has an identical twin brother of similar accomplishments who is likewise a faculty member at Stanford. This creates no end of confusion for all kinds of people, but I understand this is Bill who is with us this morning and not John.

Bill comes to us with a very special background and expertise in that he has spent the last several years studying the Appalachian Regional Commission and its work trying to gain a better understanding of the way in which large-scale problems are attacked. And so in addressing Recommendation III, he will bring that special expertise to the committee.

(The biographical sketch of Dr. Linvill follows:)

WILLIAM K. LINVILL, EXECUTIVE CHAIRMAN OF THE DEPARTMENT OF ENGINEERINGECONOMIC SYSTEMS, STANFORD UNIVERSITY

William K. Linvill is concerned with problems of systems analysis and decision-making. The areas of application in which he has been involved range from computer-coordinated systems to long-range planning. Presently he is studying problems of societal choice and technological change.

Following receipt of a liberal arts degree from William Jewell College, he entered the cooperative course in electrical engineering at Massachusetts Institute of Technology where he received the joint Bachelor's and Master's degree in 1945 and the Doctor of Science degree in 1949. Following award of his doctorate, he was on the faculty at MIT from 1949 to 1958. His research interests were in computer control systems for air traffic control and air defense. In 1954, he became more specifically interested in the systems area and in 1956 took a two-year leave of absence from MIT to lead a project of the Institute for Defense Analyses on NATO Air Defenses. In 1958, he became a Senior Staff member of the RAND Corporation. In 1960, he joined Stanford University as a Professor of Electrical Engineering, and in 1963 he became Chairman of the Institute in EngineeringEconomic Systems there where he established and continues to develop a systems training and research program featuring field internships for graduate students. In 1967 he became Chairman of the Department of Engineering-Economic Systems. In 1972 he is on sabbatical from Stanford as a Senior Institute Fellow at Battelle Memorial Institute in Columbus, Ohio.

He is a Fellow of the IEEE, a member of the IEEE Systems Science Committee, and has been active in a number of groups and subcommittees including the Advisory Committee of the New Technical Activities Committee, the Systems Science and Cybernetics Group, the Power Engineering Education Committee, and the Awards Board of the Education Medal Committee. He was a member of the Committee of the National Academy of Engineering on the Interplay of Engineering with Biology and Medicine; a member of the National Research Council panel advisory to the Technical Analysis Division of the Institute for Applied Technology; and a member of the Advisory Committee for Engineering of the National Science Foundation.

STATEMENT OF DR. WILLIAM K. LINVILL, EXECUTIVE CHAIRMAN, DEPARTMENT OF ENGINEERING-ECONOMIC SYSTEMS, STAN

FORD UNIVERSITY

Dr. LINVILL. Mr. Chairman, members of the committee, societal growth and development in response to new technological opportunities is a promising prospect, but it involves large-scale interlocking societal efforts, substantial risks, and complicated transition both in institutional patterns and in individual lives. Many new technological opportunities involve large and diffuse problems, and our society really needs new processes to deal with them.

Recommendation III proposes a major effort to develop such a new process. Through such a process, the society can tackle very large and diffuse problems of national importance, explore their involved relationships to the society, develop alternative exploratory approaches, provide the various levels of societal choice, and consider stages of implementation of solutions.

The objective here is to develop a new process which allows us to launch new initiatives but to leave the men and companies and governmental institutions free.

The problem of the Appalachian Regional Commission represents the kind of societal choice problem with which we are concerned. In a problem such as developing an industrial capability in a given locality, often several separate States must deal jointly with several Federal agencies and several private companies.

The industrial development would influence the local and regional transportation needs, the drain on the region's water resources, its population migration, the health care needs, the vocational and educational needs, the region's resource base, its pollution problem, and so on. The various State and local groups must work jointly on the problem but still as independent agencies. Various elaborate brokerage processes have been established for dealing with such problems.

Just parenthetically, I studied the Appalachian Commission because of the fact that really it is an experiment in that brokerage process. I feel that it has much to do with where we go further.

Now, problems with regard to technological advances are equally complex in terms of the constituencies involved but often include as well uncertainty of technological outcome as well as uncertainty with regard to societal acceptance. Almost all technological change problems involve substantial dynamic effects. The society's problem, then, is to develop this process for societal choice in response to broad new technological responsibilities.

This process is to me a very key point. This process is now not available already designed. That is, we are not proposing a specific process.