RAL had been doing a number of studies into various applications of office automation, including a pilot system on a PRIME computer. During 1981, DoI launched its Office Automation initiative and DoI were keen to get RAL expertise into this area. As a result, SERC put forward a proposal based on interconnecting together a number of systems from different manufacturers using the Cambridge Ring as the local area network base. The companies in the original proposal were:
The original proposal fell through due to the standard animosity between GEC and ICL. At that time, ICL were trying to get a wider base in the office automation area and kept coming forward with grandiose plans which would not fit with the already established facilities at RAL.
The original project was to be a fully distributed one with local area networks at SERC sites being linked by the wide area network.
Eventually, the original concept was scrapped and two separate but inter-related pilots were agreed with DoI giving a letter of intent on both projects in April 1982. The first was a project to take Data Recall stand-alone word processors and develop them so that they were connected via Cambridge Rings and the systems would be installed at both RAL and Swindon. The aim of the project was to quantify the advantages of inter-linking the word processors.
Some stand-alone Diamond word processors already existed at both RAL and Swindon so history of their use in a stand-alone environment already existed. The Office Pilot added 9 more Diamonds at RAL and 7 at Swindon. These would be installed as stand-alone systems initially.
Two Cambridge Rings were installed at RAL and Swindon. The RAL ring was the largest one in the world so far, covering a large part of the RAL site. The original plan was to connect the two rings by equipment developed under JNT contracts. The failure of this equipment to appear in time meant that the Rings were eventually connected using GEC systems as gateways. Data Recall provided the TSBSP/BBP protocols on the ring in the timescales defined and the project was quite successful. Documents could be transferred between the two sites and, today, this is still a mechanism for communication between the two establishments. The Logica Polynet Cambridge Rings have proved to be very robust and continue to work well in 1985.
RAL generate a great deal of scientific text which cannot be handled by conventional word processors. A second pilot was agreed which used the PERQ as a scientific word processor. DoI agreed to provide 9 PERQs which was, effectively, their contribution to the PERQ project as a pre-production purchase.
The aim of the project was to take conventional text developed on the Diamonds and add the scientific text on the PERQs. It was believed that a department might have a number of conventional word processors and perhaps a single scientific text one. Consequently, bulk text input would or could be done on the conventional word processors.
Agreed standards of interchange were required from ICL and Data Recall.
The project proved a disaster from start to finish. The lateness of UNIX being available on the PERQ was a significant problem. As a result, most of the scientific text production facilities had to be developed under POS. ICL eventually produced a usable system but it was very late and was never integrated with the Data Recall activities.
The work was started at Dalkeith while the PERQ developments were being done at Bracknell. When the PERQ developments moved to Dalkeith, it caused the office automation developments always to take the back seat. ICL had lost interest in Cambridge Rings so were never pushing to get the interfaces complete. Finally, ICL insisted on trying to mix the project with their general office automation developments. Attempts were made to move the project away from its original objectives.
In summary, the main advantage of the project was the arrival of 9 PERQs which were used for system development both at RAL and ICL.