The magazine of the Melbourne PC User Group
Computing History - The First ANCCAC Conference
 Ian McDowell

The Australian National Committee on Computation and Automatic Control held its First Conference on Automatic Computing and Data Processing in Australia at the University of Sydney and the University of NSW on 24-27 May 1960 under the chairmanship of Dr. J. M. Bennett of SILLIAC fame. The achievements of programmable digital computing are now, incredibly, more than forty years old.

The conference attracted learned papers as follows:
  • Commercial Applications - 43 papers in 12 sections
  • Technical Applications - 55 papers in 21 sections
  • Design, Construction and Programming Techniques - 41 papers in 12 sections
  • Equipment Offering in Australia - 9 papers in 3 sections
Professor J. P. Baxter of the Australian Atomic Energy Commission in his opening address warned of the coming information revolution, and warned that: "people working in these fields require above average knowledge and ability in mathematics, and a serious shortage of such people exists". The volume and diversity of learned papers presented at the conference shows, however, that the experts and enthusiasts capable of using computers far exceeded the amount and sophistication of equipment offering to them at the time.

CSIRAC at Melbourne University was in full swing, and most of the unit is thankfully displayed in the Museum of Victoria. Though it had roughly the computing power of an early Texas Instruments hand-held programmable calculator, we who used it thought it marvellous. Only SILLIAC and WREDAC (Weapons Research Establishment at Salisbury) offered similar established facilities. Yet the equipment explosion had begun, with IBM, Ferranti, Bendix, NCR, ICT, Burroughs, Emidec, and Leo machines described if not displayed. Scientific applications were beginning to take over from commercial needs as a driver for development. Barry de Ferranti did, however, make out a persuasive "case for a small machine".

Users programmed these early computers by cobbling together subroutines in their own versions of machine language. CSIRAC would add, subtract and multiply, but to divide (to say nothing of trigonometrical functions) it required lengths of paper tape. The experts predicted the proliferation of equipment, but not the rise and fall of programming languages such that the application software does all we require with a minimum of fuss, we trust. The 148 papers gave almost no hint of the transparency that was yet to come.

Probably most of the authors worked for scientific bodies or hardware vendors. The writer was one of the few independents. His paper described the use of CSIRAC to solve the problem of the economic optimisation of the extent of any capital work in terms of development and interest rates in relation to the construction cost equation. In later years he found that the behaviour of the optimisation function is fractal. The computer was needed to handle the successive approximations. Much of what the contributors offered was state of the art, and speakers revelled in the new ability to handle their mathematical modelling in other than physical ways. The writer's presentation was, in parts, uproarious. Looking back, he feels a sense of pleasure that he was part of the early days of modern computing, and a sense of wonder that he was involved at all.

  Reprinted from the March 2002 issue of PC Update, the magazine of Melbourne PC User Group, Australia

[About Melbourne PC User Group]