• Computer development and applications, from computer-assisted design to computer operation of complex systems;

• Information theory-the general specification of conditions which must be met for messages to be transmitted and decoded with predictable probability of error.

What these cases show is the feedback that occurs-must occur between fundamental research and engineering design. Research in solid-state physics and information theory begets improved communications which, in turn, suggest new possibilities in research. Because of the importance of this synergism, Bell Labs has even located some of its groups-for example, the development-and-design-formanufacture group-in laboratories on Western Electric premises (ref. 137).

A second feature of the work undertaken at Bell Labs has been the importance attached to long-range planning. It is fundamental to define one's objective-to decide what the laboratory or a group within it should be doing now and perhaps a dozen years downstream. At Bell Labs, management has often been willing to wait for years before a commercial application becomes practical. Projects, even major ones, are small, rarely employing more than a dozen persons. What has characterized much of Bell Labs' work has been the persistence with which management has stuck to a certain objective. It will be interesting to see whether this long-term view of the world survives the changes in AT&T under the current order of divestiture. One of the advantages of the AT&T monopoly was that it could, and did, take the long view.

Take the development of electronic switching systems. The first studies for a “transistor switching system" were begun in 1951 at Bell Labs' New York offices and ran parallel to work being done by the electronic apparatus development organization. These studies, as well as other exploratory work, led to the conclusion that high-speed electronics could provide smaller and less expensive switching systems. At that point (1954), Western Electric authorized a field trial of an electronic system at company facilities in Morris, Illinois. This trial was a success, in the sense that it proved the soundness of stored-program control for telephone switching systems; but the Morris system as originally designed was not put into production, because "a revolution in technology had made obsolete all its major components." (ref. 138.) Instead, work began on a parallel effort, using many of the concepts proved at Morris, but incorporating magnetic memories, newer network elements, and silicon transistor and diode circuits. The Bell System introduced its first commercial system in 1963, an improved version in 1965, and an automated government network in 1966. In effect, the lead time from the first exploratory studies and first commercial networks took more than a

dozen years, involved parallel efforts at many of the Laboratories' facilities, led to several false starts and, once on the right track, resulted in new services and features.

If we seek to isolate the special features of the more important Bell Labs' projects, they would include the interdependence of fundamental and applied research, the broad base of knowledge in many disciplines, the continuity of teamwork among small groups of scientists and research engineers, the willingness of the sponsoring organizations to commit their resources to projects with very long lead times, and the close working relationships established with the ultimate user, Western Electric. Of these features, the most important was the interdependence of basic and applied work-and not at this laboratory only.

In this and the two preceding chapters, we have looked at technology development institutions as a whole-their common features, their problems, and the ways in which they interpret their charters. We now turn to the roles of the professional staff, the men and women who are the laboratory. Reversing the Marxist formulation, we can say that the overwhelmingly important job of the research administrator is to move from the administration of things to the management of people-or rather, to manage with them.

CHAPTER VII

The Management of the Professional Staff

Employment Patterns Among Federal Scientists and Engineers

The vitality of a technology development organization resides in its professional scientists and engineers. This group proposes the ideas that are starting points for development programs, examines their validity in the light of fundamental scientific principles, and devises strategies by which ideas are converted, step by step, into operating systems. Other elements of the organization are essential skilled technicians, shop machinists and electricians, administrative personnel. But unless the professional scientists and engineers function imaginatively and competently, the organization will lose momentum, stop generating worthwhile ideas, and either cease to achieve the goals set by the sponsoring agency or lower them in favor of what is considered attainable. In this chapter we will review the tactics of managers and senior officials in technology development to maintain high levels of staff performance. Specifically, we want to provide at least tentative answers to these questions: How does a laboratory acquire and retain a competent staff? What are the effects of aging on research? How do scientists and engineers make the transition to management? How important is professional mobility, both between divisions within a laboratory and between laboratories? Finally, how do managers and the officials to whom they report evaluate the quality of research within their organization?

Before these questions can be answered, we would like to make our assumptions about personnel management explicit. Our first assumption is that there are no personnel policies which are guaranteed to work across organizational lines. Continuing education may or may not work; indefinite employment or term employment may or may not work; rotation of staff between divisions may or may not work. It is as if, in the best laboratories, the organizational culture is strong enough to impose itself on any program of personnel management.

This brings us to our second point. Personnel issues are synonymous with the organization's goals. Arnold Deutsch notes that: "... the primary motivating factor in job selection by technical people is the

nature of the work itself. The interest, the challenge to the technical abilities, and the opportunities for significant professional achievement offered by the work to be done determine whether or not an engineer or scientist will consider a position." (ref. 139.) Consequently, nothing less than a thorough understanding of the functions of a given laboratory will enable one to decide which personnel policies are effective and why. Personnel management is affected by (for example) levels of agency funding and the discretion of laboratory directors in spending what is allocated; the diversification of laboratories into work for others; the advent of new technologies which suggest new programs; and the competition for the brightest graduates that government laboratories face from industry and the universities. Many personnel theorists have somehow forgotten that since the environment of any organization is constantly changing, no theory which ignores this truism will hold water.

Our third point has to do with nature of this environment, and may be summed up thus: Despite declines in funding for nondefense basic research, the aging of staff in Federal laboratories that have experienced manpower restrictions, and the tendency of agencies to locate their new programs in existing facilities rather than creating new ones, the functions of the Federal scientist and engineer are not likely to change greatly. As to the relative decline in Federal support for research and technology development, the evidence is clear enough since 1967. Only in the last few years have we started to reverse this trend, and it will take some time for the positive results to become apparent.

Consider, for instance, nine of the national laboratories operated under contract to the Energy Department (table 5).

These figures show that for seven of the nine laboratories there was little or no growth in staffing during the preceding decade and that the two exceptions -the Los Alamos and Lawrence Livermore National Laboratories were and are heavily involved in weapons development. At NASA the decline in the work force has been even steeper, with a drop of 30 percent, from over 30 000 to 22 000 between fiscal years 1971 and 1982. Possibly of more significance is the rise in the proportion of professional employees to the total workforce (fig. 42). From these figures it is apparent that scientists and engineers alone comprise just over half of NASA permanent employees and that, combined with professional administrators, they account for two-thirds of permanent employees. A further breakdown of the figures illustrates two important trends. The first is the increase in the number of supervisors as a percentage of total NASA employees. In fiscal 1982, there was approximately 1 supervisor for every 6.7 employees agencywide, while at Headquarters the ratio was almost 1 to 5. At many government laboratories there has also been a decline in the number of technicians available to support professional staff.

Table 5. Full-time Staff Equivalents from June 1967 to September 1982 for Nine National LaboratoriesTM