Guest blog: Barbara Ainsworth – Monash Museum of Computing History, Faculty of Information Technology, Monash University

CSIRAC Computer, Plaque for Opening Ceremony

CSIRAC Computer, Plaque for Opening Ceremony, 14 June 1956

On June 14, 1956 the computer CSIRAC was officially recommissioned at the new Computation Laboratory at the University of Melbourne. The name CSIRAC or CSIR Automatic Computer was new as well. The name acknowledged the beginnings of the digital computer at the CSIRO Division of Radiophysics and its function as an automatic computer rather than the more common use of the word computer at the time, which was applied to human computers rather than machines. The CSIRO computer had a variety of names during its lifetime.

The name acknowledged the beginnings of the digital computer at the CSIRO Division of Radiophysics and its function as an automatic computer

In the late 1940s, Trevor Pearcey and Maston Beard built an electronic, digital, general purpose, stored-program computer at the CSIR/CSIRO Division of Radiophysics in their facilities on the campus of the University of Sydney. The computer had several official names and an affectionate nickname before it was finally designated CSIRAC in 1956. The Australian Government research body the Council for Scientific and Industrial Research (CSIR) operated from 1926 to 1949 when it changed its name to the Commonwealth Scientific and industrial Research Organisation (CSIRO). Working at the CSIR Division of Radiophysics, Trevor Pearcey proposed building an electronic computer in a paper in late 1946. He described such a machine as a “computor of the successive principle”. (The spelling of the word computer was still varied at this date.) The CSIR administration were interested in Pearcey’s proposal and agreed for Radiophysics to start a project to design and then build a digital computer. The team under Pearcey (logic) and Maston Beard (engineering) worked on the machine, then called CSIR Mark 1. This was a standard naming convention using the institution and the stage of the project. For example, the early US electro-mechanical computer, the IBM Automatic Sequence Controlled Calculator (ASCC), was also known as Harvard Mark 1. The name implies that further developments are envisaged in the project’s life. Indeed, the Harvard Mark 1 was succeeded by several more versions with incremental numbering.

In November 1949, the team at Radiophysics ran the first successful program on CSIR Mark 1 and it provided a computing service for CSIRO and other users. In August 1951, Pearcey demonstrated the computer at the first Australian computer conference. The computer’s name used in the Conference’s Proceedings varied from Radiophysics Mark 1 Automatic Computer (p.42) in a paper by Pearcey and Beard to CSIRO Mark 1 Electronic Computer in the table of items displayed for the conference. Pearcey used several versions of the name in his correspondence and publications but all included the term Mark 1.

For Pearcey, the inclusion of the name Mark 1 was significant as he envisaged further developments in the project. Maston Beard used the term Mark 11 in his thesis in 1958 to describe developments in the hardware. The computer was in regular use from the early 1950s. In 1952, the CSIRO was considering building a more advanced version of their computer. They sent tenders to several local electronics firms but found the supplied quotes to be very expensive. The CSIRO reviewed the situation and came to the conclusion that the CSIR/CSIRO Mark 1 had been an interesting project but circumstances had changed. They considered the cost of developing a second stage to be too expensive and would consume a large amount of the Division’s budget. Also, new overseas technical developments had moved on from the first-generation computers and commercial machines were becoming available. They decided to end the CSIRO’s future involvement with computer development in April 1954, and find a new home for CSIR/CSIRO Mark 1.

After some time, the CSIRO eventually offered their computer to the University of Melbourne. Frank Hirst was sent to Sydney to supervise the removal of all of the components. The machine was crated up and sent by truck to Melbourne. During his visit, he noticed that the Radiophysics staff called the computer “Cicero Mark 1” which he took as a play on the word CSIRO.

Naming computers in the 1950s reflected their unique designs and origins but contemporary machines overseas had increasingly strange names. A US machine built at Los Alamos Scientific Laboratory in California in 1952 was given the name Mathematical Analzyer Numerical Integrator and Automatic Computer Model 1 which was reduced to the acronym MANIAC 1. Not a name to instil confidence but it was the first computer to defeat a human being in a chess-like game in 1956. The name was probably quite intentional to highlight the strange but often affectionate acronyms appearing on machines across the world. In England, a computer was installed in 1951 at the Atomic Energy Research Establishment at Harwell in Oxfordshire with the name the Harwell Dekatron Computer. It was later transferred to a college, now Wolverhampton University, and given the name WITCH, an acronym for Wolverhampton Instrument for Teaching Computing from Harwell. (This machine is now on display at the NMOCH in the UK.)

Meanwhile, in Australia during 1955, the University of Melbourne received the CSIRO computer in crates and proceeded to put it back together in the Physics Building. It was ready by mid-1956 to be recommissioned and the University was aware of the need to mark the occasion with a formal handing over ceremony. However, what should they call it? The correspondence between the CSIRO and the University of Melbourne during 1955 used the name CSIRO Electronic Computer. The University thought CSIRO Mark 1 no longer suitable and Cicero inappropriate. Suggestions were rife amongst the staff including ABAC (Australian Binary Automatic Computer) or something using Melbourne University’s initials such as MUSE or MUSTARD. In March 1956, Professor Cherry wrote to Sir Ian Clunies Ross, Chairman, CSIRO and proposed to put a plate on the computer to record the work of Trevor Pearcey and Maston Beard in its design. The suggested wording for this plate had one word left out – the name of the computer! Cherry consulted the CSIRO on their preferred name choice. CSIRO felt that the name should include some acknowledgement of its origins. The name CSIREC or CSIRO Electronic Computer seemed a likely contender. Sir Ian Clunies Ross approved of this choice and described the term as “euphonious”. However, in following correspondence, Dr Pawsey from Radiophysics pointed out that as the EC letters would be pronounced in a sharp sound and this may imply that the computer was a ‘wreck’!

On 14 June 1956, Sir Ian Clunies Ross officially opened the new Computation Laboratory at the University of Melbourne and the Mark 1 was formally recommissioned and named CSIRAC.

Finally, the wording CSIRAC (CSIR Automatic Computer) was selected. On 14 June 1956, Sir Ian Clunies Ross officially opened the new Computation Laboratory at the University of Melbourne and the Mark 1 was formally recommissioned and named CSIRAC. The plate naming the computer CSIRAC and acknowledging Pearcey’s and Beard’s contribution was unveiled to commemorate the occasion. The computer provided a computing service until it was retired in 1964 and it processed more than 1000 projects over its 14 years of operation.

The dawn of Australia’s digital age will be celebrated at The Riding the Digital Wave Summit being held at the University of Sydney on 30 September 2021.

Image

Museums Victoria Collections https://collections.museumsvictoria.com.au/items/1686150. Accessed 15 April 2021