Uploaded April 2026 | Updated September 2026, 2 weeks ago
This is the final of a 3-part episode on Australia’s first wide area network (WAN). It is highly recommended that episodes 1 & 2 be viewed beforehand; however, there is a summary at the end of this episode that highlights key moments in the 3-decade history of CSIRONET.
In part 3, CSIRONET is quickly expanded, improved upon and evolves during the 1970s and 80s due to the influence of a new Chief of Division, Dr Peter Claringbold. Claringbold had replaced Dr Godfrey Lance, who was more of “a computer should do numbers man”, while Claringbold could see that computers could be used not just to crunch numbers, but a broader range of tasks that could help scientists with their work. This resulted in the acquisition of a lot of new equipment and peripherals for the CSIRO’s network.
The network was named CSIRONET by 1972 & grew to become a multi-host network with the acquisition of a CDC Cyber 70 model 76, known simply as the Cyber76, in 1973. The PDP-8s were replaced with PDP-11s in the same year.
Because there was a succession of Control Data Corporation (CDC) computers, much of the existing code base & libraries could be easily migrated to each new offering. This led to a long list of computers inducted into CSIRONET, including the Cyber 76, Cyber 730, Cyber 840, Cyber 845 and the Cyber 205. The Cyber 205 (introduced on ANZAC Day 1984) was Australia’s first true supercomputer.
By 1974, CSIRO had 30 sites with a total of 35 PDP-11s acting as nodes, connected with over 175 interactive terminals attached to CSIRONET. By June 1976, most CSIRO Divisions around Australia had a connection to CSIRONET and many new services were added.
CSIRONET also acquired a III COMp 80, which was a machine that could record back to Microfilm (Microfiche) from digital files. This saved the printing of tonnes of paper-based documentation every week and made distribution easier. Not only that, but it could also print on 11-inch-wide bromide proof sheets, which could then be used by professional printers, effectively creating a digital-to-paper workflow for publishing.
A high-speed local area network was installed in the early 1980s called the NSC Hyperchannel. It interconnected the CSIRONET mainframes at 50MB/sec as well as some host services, which allowed for the rapid transfer of data between systems. The Hyperchannel allowed Image processing and Searching to be carried out at far greater speeds than would otherwise be possible.
Fujitsu (IBM compatible) mainframes were purchased for CSIRONET in the late 1970s to enable business computing for CSIRO divisions and other government departments that needed it. These Mainframes were known as FACOM machines or “Fujitsu Automatic COMputer” and CSIRONET had purchased a FACOM M-150F, a FACOM M-180-N and a FACOM M-190 machines.
An automatic tape library was installed in the 1980s. This was controlled by the FACOM M-190 and consisted of a robotic tape recovery system, which could load and unload high-density tapes from the shelves to the tape read/write machines.
In the early 80s, CSIRONET had grown to a nationwide, multi-host X.25 packet switching network with 1000s of users ranging from scientists to Government organisations.
As the network expanded, the PDP-11-based nodes were seen as expensive and perhaps needed more IO for the network's continuation, so the CSIRO, along with a Canberra-based firm called Network Automation, designed and built a new node controller called a Micronode. This was based on a Motorola 68000 microprocessor, and 50-60 were manufactured for the network in 1983.
Peter Milne and Henry Hudson introduced CSIRO Mail to the network, having used and seen it on the USA’s ARPANET. It was implemented at a network level on CSIRONET, with the help of John Paine, allowing users to log in to any of the hosts, and mail would automatically be directed to the user. CSIRO Mail was very successful and popular.
The introduction and widespread adoption of Minicomputers created a very real threat to CSIRONET, as it became more desirable to keep computing local to the Divisions instead of having to rely on DCR and the expensive use of CSIRONET. A push from Government advisors to the Hawke federal government resulted in the breakup of some areas of the CSIRO. CSIRONET was privatised in 1985.
The CSIRO ceased using CSIRONET by 1990 as its computing needs were met by other means. CSIRONET faded away after 1991 with the public release of the Internet in the same year.
For more history on CSIRONET, please visit: https://csiropedia.csiro.au/csiro-computing-history-sidebar-2-the-start-of-the-on-line-network/
This is the final part of Episode 16 and it is also the final in the first Series on “The Computer History of Australia”. State of Electronics will return soon (2-3 months) with Series Two.
This is the final of a 3-part episode on Australia’s first wide area network (WAN). It is highly recommended that episodes 1 & 2 be viewed beforehand; however, there is a summary at the end of this episode that highlights key moments in the 3-decade history of CSIRONET.
In part 3, CSIRONET is quickly expanded, improved upon and evolves during the 1970s and 80s due to the influence of a new Chief of Division, Dr Peter Claringbold. Claringbold had replaced Dr Godfrey Lance, who was more of “a computer should do numbers man”, while Claringbold could see that computers could be used not just to crunch numbers, but a broader range of tasks that could help scientists with their work. This resulted in the acquisition of a lot of new equipment and peripherals for the CSIRO’s network.
The network was named CSIRONET by 1972 & grew to become a multi-host network with the acquisition of a CDC Cyber 70 model 76, known simply as the Cyber76, in 1973. The PDP-8s were replaced with PDP-11s in the same year.
Because there was a succession of Control Data Corporation (CDC) computers, much of the existing code base & libraries could be easily migrated to each new offering. This led to a long list of computers inducted into CSIRONET, including the Cyber 76, Cyber 730, Cyber 840, Cyber 845 and the Cyber 205. The Cyber 205 (introduced on ANZAC Day 1984) was Australia’s first true supercomputer.
By 1974, CSIRO had 30 sites with a total of 35 PDP-11s acting as nodes, connected with over 175 interactive terminals attached to CSIRONET. By June 1976, most CSIRO Divisions around Australia had a connection to CSIRONET and many new services were added.
CSIRONET also acquired a III COMp 80, which was a machine that could record back to Microfilm (Microfiche) from digital files. This saved the printing of tonnes of paper-based documentation every week and made distribution easier. Not only that, but it could also print on 11-inch-wide bromide proof sheets, which could then be used by professional printers, effectively creating a digital-to-paper workflow for publishing.
A high-speed local area network was installed in the early 1980s called the NSC Hyperchannel. It interconnected the CSIRONET mainframes at 50MB/sec as well as some host services, which allowed for the rapid transfer of data between systems. The Hyperchannel allowed Image processing and Searching to be carried out at far greater speeds than would otherwise be possible.
Fujitsu (IBM compatible) mainframes were purchased for CSIRONET in the late 1970s to enable business computing for CSIRO divisions and other government departments that needed it. These Mainframes were known as FACOM machines or “Fujitsu Automatic COMputer” and CSIRONET had purchased a FACOM M-150F, a FACOM M-180-N and a FACOM M-190 machines.
An automatic tape library was installed in the 1980s. This was controlled by the FACOM M-190 and consisted of a robotic tape recovery system, which could load and unload high-density tapes from the shelves to the tape read/write machines.
In the early 80s, CSIRONET had grown to a nationwide, multi-host X.25 packet switching network with 1000s of users ranging from scientists to Government organisations.
As the network expanded, the PDP-11-based nodes were seen as expensive and perhaps needed more IO for the network's continuation, so the CSIRO, along with a Canberra-based firm called Network Automation, designed and built a new node controller called a Micronode. This was based on a Motorola 68000 microprocessor, and 50-60 were manufactured for the network in 1983.
Peter Milne and Henry Hudson introduced CSIRO Mail to the network, having used and seen it on the USA’s ARPANET. It was implemented at a network level on CSIRONET, with the help of John Paine, allowing users to log in to any of the hosts, and mail would automatically be directed to the user. CSIRO Mail was very successful and popular.
The introduction and widespread adoption of Minicomputers created a very real threat to CSIRONET, as it became more desirable to keep computing local to the Divisions instead of having to rely on DCR and the expensive use of CSIRONET. A push from Government advisors to the Hawke federal government resulted in the breakup of some areas of the CSIRO. CSIRONET was privatised in 1985.
The CSIRO ceased using CSIRONET by 1990 as its computing needs were met by other means. CSIRONET faded away after 1991 with the public release of the Internet in the same year.
For more history on CSIRONET, please visit: https://csiropedia.csiro.au/csiro-computing-history-sidebar-2-the-start-of-the-on-line-network/
This is the final part of Episode 16 and it is also the final in the first Series on “The Computer History of Australia”. State of Electronics will return soon (2-3 months) with Series Two.
