A look into the dawn of the computer age

Why are computers called “computers?” Because they were originally created to do computations faster, on a larger scale, and more efficiently than a person can do them with a pencil on a piece of paper. The math involved in something like determining what metals can withstand the physical stresses of an airplane wing is too complex—and mistakes are too dangerous and expensive—to leave to a mathematician. You don’t just take off into the sky with your fingers crossed to see if the wings hold up. Instead, you run simulations.

The NC State University Libraries Special Collections Research Center (SCRC) is excited to announce the donation of the Brian W. Hollocks Collection to the Computer Simulation Archive. Hollocks is a retired Professor from the Faculty of Management at Bournemouth University in the United Kingdom, where he taught and researched simulation, computer-supported cooperative work, knowledge management, and artificial intelligence. Before his teaching career, Hollocks worked in industry for the United Steel Companies and the British Steel Corporation in simulation, operational research, and software development.

The Hollocks collection offers a snapshot at the early years of the computer industry as it found large-scale applications in postwar industry and business, which led to the dawn of personal computers in the 1980s.

The materials in the collection reflect Hollocks’ work in operational research, including his early work on discrete-event simulation software. The collection encompasses materials he acquired during his career from the 1960s into the 1980s, including circuit boards, operation manuals, instruction sheets, and handbooks. “The items concerned were rescued in the U.K. from waste bins and demolition as departments closed or were relocated, or as equipment was replaced and junked,” Hollocks writes.

The collection also includes two circuit boards from a Ferranti Pegasus mk 2 computer, a first-generation valve computer, as well as its programming and operations manual, in addition to related materials from the Department of Operational Research and Cybernetics at the United Steel Companies, Ltd. Essential to the modular computing model of the Pegasus, these circuit boards each performed a different computational function and could be inserted into the mainframe in different configurations to run different simulations.

“It’s not the first, but it is the machine that really shows what it was like when this industry was first being built up and created,” says Tilly Blyth, Science Museum (London) Head of Collections and Principal Curator in a museum video about the Pegasus. “So this is the machine that was actually really important for users. This is the machine that tried to address many of the very early issues about how you actually worked with a computer.”

Some of the Hollocks collection touches upon his work at the United Steel Companies with Keith Douglas Tocher (1921–1981), a pioneer in computer simulation. Working on Tocher’s team, Hollocks helped develop the General Simulation Program (GSP), the first discrete-event simulation package that used modular computing in order to execute a range of different possible simulations. In 1970, Hollocks re-implemented the GSP in Fortran, an early computer language.

To access the collection, please contact the SCRC at library_specialcollections@ncsu.edu.