In Sessions Five through Eight, Paul Ceruzzi, Robert Anderson, and Willis Ware interviewed thirteen participants to discuss RAND's role in the post-1945 development
of computers. Along with the University of California, Los Angeles (UCLA) and the Lawrence Livermore Laboratory in Livermore, California, RAND pioneered computer engineering
and programming on the West Coast. The Air Force initially funded this work as a means of accelerating systems analysis and missile and nuclear weapon development, but became
somewhat more reluctant to underwrite RAND's research when staff there began to develop more interactive applications for computers in the 1960s. Sessions were shot at RAND
Corporation headquarters in Santa Monica, California. Besides the listed participants for each session, other session participants often contributed as members of an audience.
Participants for this first session contributed to the development of computer hardware between 1945 and 1965. Paul Armer joined Project RAND in 1947 as a mathematician
and desk calculator operator. Five years later he became head of the Computer Sciences Department. In 1968 Armer left RAND to direct the Stanford University Computation Center,
and since then has also headed the Charles Babbage Institute for the History of Information Processing in Minneapolis, Minnesota, and presided over the American Federation
of Information Processing Societies.
William F. Gunning started working for Douglas Aircraft in their Flight Test laboratory in 1941. In 1947 he transferred to Project RAND where he worked as an electronic
engineer on the development of the Random Number Generator, the modification of the REAC (Reeves Electronic Analog Computer), the Williams memory of the SWAC (Standard Western
Analog computer), and the JOHNNIAC.
William P. Myers began working at RAND as Tabulating Department shift leader in 1948. With the arrival of the IBM 604 programmed calculator in 1951, he headed the Operations
Group of the Numerical Analysis Department before moving to the Systems Development Department. In 1958 Myers returned to the Numerical Analysis Department in various managerial
capacities; he switched to the Computer Science Department in the 1970s and retired in 1984.
Robert T. Nash worked at RAND from 1948 to 1957. He rose from IBM Cards Processor to Administrative Assistant of the Numerical Analysis Department by 1955. Nash then filled
the same position for the Programming Department of the Systems Development Division, where he arranged for the settlement of SDD staff at air bases of the SAGE (Semi-Automatic
Ground Environment) defense system. In 1956 he became manager of Field Services for the SDD and kept the position in 1957 when RAND spun off the Division into an autonomous
corporation.
Keith W. Uncapher also arrived at RAND in 1950 and focused on computer engineering. He began with designing components for the REAC, and was responsible for development
of the Selectron memory store of the JOHNNIAC. In the late 1950s he oversaw the RAND contract with Telemeter Magnetics for the first 4,096-word, 40-bit, magnetic core store;
in the 1960s he managed the development of JOSS (JOHNNIAC Open Shop System), the RAND tablet, the GRAIL system, the acquisition by RAND of larger computers, and the development
of a hardened communications system for the Air Force. Uncapher left RAND in 1972.
Willis H. Ware received his B.S. in electrical engineering from the University of Pennsylvania in 1941 and his S.M. from the Massachusetts Institute of Technology one year
later. After spending World War II at Hazeltine Electronics Corporation, he joined John von Neumann's Electronic Computer Project at the Institute for Advanced Study in Princeton,
New Jersey. He completed his Ph.D. at Princeton in 1951 and immediately joined RAND, where he helped lead the engineering on the JOHNNIAC, particularly its Selectron and magnetic
core memories. Between 1964 and 1971 Ware headed the Computer Science Division, during which time he initiated debate on computer security as a technical subject. Ware joined
the Corporate Research Staff in 1973 and advised various agencies on the applications of large computer systems. Throughout his career, Ware also helped found and support
professional groups associated with computing.
Participants for Session Six contributed to the development of computer software between 1945 and 1965. Morton I. Bernstein joined RAND in 1954 as an assistant mathematician.
For three years he wrote programs for the IBM 701 computer, the Linear Programming group, the Logistics Department, and consulted on various war games. In 1957 Bernstein assumed
responsibility for the JOHNNIAC, concentrating on programming language design and implementation. Between 1961 and 1963, when he left RAND, Bernstein worked with the engineering
staff on the RAND tablet.
Irwin Greenwald worked for RAND from 1950 to 1960 and from 1964 to 1969. During the 1950s he was responsible for programming JOHNNIAC's first assignment, the computations
on UNIVAC I for the first H-bomb test, RAND's system and utility programming of the IBM 701 and 704 computers, PACT-II (Project Advanced Coding Technique), and the formation
of the first users' group, SHARE (Society to Help Avoid Redundant Effort). In the 1960s Greenwald masterminded the major components of the JOSS-II system and the initial software
for the RAND Videographic System.
J. Clifford Shaw spent twenty-three years at RAND, beginning in 1950. He specialized in systems software, beginning with the IBM Card-Programmed Calculator, the IBM 701,
and the JOHNNIAC; and in artificial intelligence, where he worked with Allen Newell and Herbert Simon on various AI programs and IPL's (Information Processing Languages) I-VI.
In the early 1960s he developed JOSS and its language and the system and demonstration software for the RAND tablet. Shaw died in March 1991.
Paul Armer and Willis Ware also appeared in this session.
Participants in Session Seven contributed to the development of computer graphics hardware between 1960 and 1965. Raymond W. Clewett began working in 1937 as a machinist,
model builder, and laboratory foreman for Douglas Aircraft Company. After World War II he joined Lear Incorporated as machine shop foreman and design engineer for six years,
returning to Santa Monica as shop manager and design engineer for RAND in 1951. His contributions there included design and construction of JOHNNIAC, other computer hardware,
nuclear reactor test equipment, the RAND tablet, and closed circuit television reading devices for the visually impaired. After 1977, Clewett was an independent design consultant
and owner of HY-TECH Engineering and Development Lab.
Thomas Ellis joined RAND in 1953 after graduate research on computer engineering at UCLA, and worked on most of the projects discussed in these sessions before he left
in 1972. He was responsible for the JOHNNIAC's input/output machinery, design of the JOSS (JOHNNIAC Open Shop System) terminals, the RAND tablet, GRAIL (Graphic Input Language),
and the RAND/IBM Videographic system.
Paul Armer, Morton Bernstein, and Willis Ware also appeared this session.
Participants in Session Eight contributed to the development of computer graphics software between 1965 and 1975. Robert Anderson received degrees in physics, philosophy,
and applied mathematics between 1962 and 1968 at the University of California at Berkeley and Harvard University. He joined RAND in 1973 as head of its Information Sciences
Department while holding teaching and research appointments at the University of Southern California. In 1981 Anderson founded his own computer systems consulting firm. After
1986 he was director of RAND's Institute for Research on Interactive Systems (IRIS) and resident consultant.
Barry W. Boehm received his degrees in mathematics from Harvard University and UCLA between 1959 and 1964. Concurrently with his graduate work at UCLA, Boehm headed RAND's
new Information Sciences Department. From 1973 to 1989 he was Chief Scientist of TRW's Defense Systems Group, after which he became Director of the Defense Advanced Research
Project Agency's Information Science and Technology Office. Boehm has written three books on software engineering.
Edward C. DeLand started working at RAND after finishing his Ph.D. in mathematics at UCLA in 1956. He spent the next six years managing the analog computers, REAC and TRAC,
and then began developing mathematical models for blood biochemistry. After switching to physiological research in 1963, DeLand helped construct the BIOMOD program. DeLand
left RAND in 1972 for UCLA's Department of Surgery where he continues to develop applications for computers in medical instruction and diagnosis.
Gabriel F. Groner was finishing a Ph.D. in electrical engineering at Stanford University when he joined RAND in 1964. His interests in character recognition, interactive
systems, computer graphics and medical applications of computers were manifested in his contributions to the GRAIL program; the BIOMOD simulation system; the study of computer
applications to industrial automation; and the CLINFO data system for medical research. Groner left RAND in 1978.
Thomas Ellis also appeared in this session.
The discussions detailed the development at RAND of computer hardware and software in the post-war era. Sessions provide visual documentation of early computer components.
Collection Citation:
Smithsonian Institution Archives, Record Unit 9536, The Research and Development (RAND) Corporation Videohistory Collection
