less-than-best laboratory might be closed: if it had served its initial purpose; if there was no likelihood that a new role for the laboratory could be found; and if the closing down of the laboratory would not leave a significant gap in the national capability to do technology development work. But most of the centers were adaptable and many had gone through at least one reorganization in the late 1950s, moving from aeronautics to launch-vehicle development, or from development work on guided missiles to lunar and planetary probes, as with the Jet Propulsion Laboratory. By 1969, another cycle of reorganization was underway. Yet the more subtle changes in a center's mission could only occur very gradually.

To summarize, NASA prospered during the early 1960s, because of its administrative flexibility, the political skills of its senior officials, the delegation of program management to the field, and the timeliness with which the important decisions were made. But the same elements were not enough to enable NASA to weather the severest test to which any mission-oriented agency can be put: namely, how to react to the completion of the mission. And this, as we shall see, is not NASA's problem alone.

Epilogue: The 1970s and After

In retrospect, the early 1960s were the golden age of Federallysponsored, mission-oriented technology development. Beginning in about 1967, the budgets of the three largest mission-oriented agencies — DOD, AEC, NASA accounting for 90 percent of all Federal research and technology development outlays, began to decline (ref. 48).

Part of this decline was due to specific developments such as the Vietnam War, skepticism over the future of commercial nuclear power (confirmed, for many, by the Three Mile Island incident in 1979), and the phasing down of Apollo. Yet the decline was general, part of systemic changes in the American economy which continue to the present. Where spending on all kinds of research and technology development from all sources amounted to 2.9 percent of the gross national product in the mid-1960s, it had fallen to 2.2 percent by 1978. During that period, Federal spending, especially on applied research, prototype, and demonstration programs fell from 1.9 percent to 1.1 percent of the gross national product (ref. 49). Clearly, such a continued trend has major implications for American economic growth or lack of it. We cannot enter into so enormously complex a subject here, but we would like to examine briefly the efforts of the technology-based agencies to adjust to cutbacks.

The fate of the AEC was the most dramatic. It disappeared. In 1974 to 1975, its regulatory functions were transferred to the newly-created

Nuclear Regulatory Commission and its other activities were given to the newly-created Energy Research and Development Administration. In 1977, the Department of Energy was created, assuming overall management of all Federal programs related to energy. The AEC laboratories tried to adjust. As early as 1964, the AEC had issued guidelines for laboratories wishing to perform non-nuclear work for other customers: The proposed work should not lead to an increase in the size of AEC or contractor staff, should not require new facilities, should not be subcontracted, and should be done only if the other sponsoring government agency could not get the work done as conveniently by private industry (ref. 50). Within these guidelines, the laboratories worked for a variety of clients; Oak Ridge has worked for DOD, NASA, the Office of Saline Water in the Interior Department, and the Public Health Service; while the Lawrence Livermore National Laboratory has carried out non-classified work, most recently in the area of superconducting materials. Whether such work is a holding action or will turn into a new mission remains to be seen.

Defense technology development was hit particularly hard in the late 1960s. Basic academic research — a precondition for certain kinds of defense technology — fell from $137 million in 1965 to $108 million in 1974 (ref. 51). Only 4 percent of the total was classified, as universities began to withdraw from certain defense projects. Nor were matters improved by the "Mansfield Amendment" attached to the 1970 defense authorization act, which forbade the use of DOD funds to finance any research unless, in the opinion of the Secretary of Defense, it had “a potential relationship to a military function or operation." (ref. 52.) The case of NASA is particularly interesting. Apollo and its successor, Skylab, were completed by 1974. Long before then, it was apparent to agency officials that to maintain a manned spaceflight program, NASA would have to confront the problem of deciding upon the next major thrust in space technology. The answer to that problem was the Space Shuttle. The main lines of the Shuttle program were drawn up by NASA task forces in 1968 and 1969, modified in 1971, and approved by President Nixon in January 1972. Briefly, the Shuttle is a partially reusable launch vehicle, consisting of an orbiter, two recoverable solid-fuel, strap-on boosters, and an external fuel tank, which is jettisoned shortly after the main engines have used all the fuel. The orbiter is designed to be launched like a rocket, operate in near-earth orbit for up to several days and, returning to Earth, land like a glider (fig. 22). It is designed for a variety of functions: to conduct experiments in zero gravity, place communications or weather satellites in orbit, and eventually, to supply an orbiting space station.

The Space Shuttle system, then, is radically different from Apollo. For all its sophistication, Apollo was simple and its principles were well

understood. No new technology was required in most instances. The theory of rocket propulsion was worked out in detail by Konstanin Tsiolkovsky, Hermann Oberth, and Robert Goddard between 1900 and 1920, and the atmospheric entry problem had been solved in the 1950s in connection with the ICBM program. Perhaps the only application of new technology in the Apollo program was in the area of guidance and control, since the lunar landing module required a completely automatic electronic control system. The Space Shuttle is very much more complex and did require the development of new technology. The orbiter must survive the shock of launch and reentry and then, given one opportunity, make a "dead stick" landing. Moreover, the management of the Shuttle program brings with it problems more like those of a commercial enterprise than of a government agency. These include how to market the Shuttle and attract paying customers; how to set user fees so that NASA will at least recover direct costs; how to screen proposals for manufacturing operations in a zero-gravity environment; how to improve the orbiter's ability to operate beyond near-earth orbit. With most expendable rockets to be phased out, NASA has staked a great deal some would say, almost everything on the success of the Shuttle.

In this chapter we have discussed the most important development in Federally-sponsored technology development over a span of 30 to 35 years. It would be presumptuous to extrapolate from that period to the major programs of the 1980s and beyond. But certain features of Federallysponsored technology development can be discerned. Three are especially important:

1. The role of new technology, especially in information processing, will continue to increase. Along with the revolution in

microelectronics, there has been a major change in how we process information. Small electronic components make it possible to construct compact and powerful computers that can now perform analysis and other decision-aiding functions which once required hours or even days of work, in a few seconds. Since the size of electronic components can be reduced further, up to a factor of a thousand, the upper limit of this technology has not yet been reached (ref. 53).

2. At the same time, the need for a standard programming language to replace the dozens of languages now used by DOD has never been greater. Thus the Army has a major project, VIABLE, to standardize all of its data processing activities, while the Air Force, in January 1983, awarded a contract to Sperry Corporation to replace all its computers. Most important, DOD has sponsored the development of a single language, ADA, which is intended to replace all of the languages now being used for military systems. This trend towards standardization is certain to continue and will spread to other areas for example, to the creation of knowledge-based systems, the automatic assembly of software parts, the development of programs which can be built from existing parts, and the creation of very high-level languages with their own control and data structures (ref. 54).

3. Finally, the role of the government laboratory is likely to change. Given the need of many laboratories to broaden their missions, to maintain their current sponsors or else find new ones, they will have to become more flexible than they now are. Consider that few laboratories do any production work and that increasing numbers of installations rely on contractors for base operations. The laboratory's function becomes, more than ever, that of generating new ideas leading ultimately to operating systems. In the remainder of this book, we will look at the problems involved in the absolutely crucial role of the research installation as generator of new ideas, and we will suggest some improvements in the way this role is handled.