|Developer||Royal Canadian Navy in partnership with Ferranti Canada|
|Release date||Prototype tested in 1953|
|Introductory price||CAN$1.9 million (almost CAN$15 million in year-2000 dollars)|
|Display||Adapted radar unit|
|Input||Trackball and trigger|
DATAR, short for Digital Automated Tracking and Resolving, was a pioneering computerized battlefield information system. DATAR combined the data from all of the sensors in a task force into a single "overall view" that was then transmitted back to all of the ships and displayed on plan-position indicators similar to radar displays. Commanders could then see information from everywhere, not just their own ship's sensors.
Development on DATAR was started by the Royal Canadian Navy in partnership with Ferranti Canada (later known as Ferranti-Packard) in 1949. The system proved too costly for the post-war Navy to develop alone, and when the Royal Navy and the United States Navy declined to share in the program it was ended. Both would then go on to start development of similar systems to fill this same role, the RN's Comprehensive Display System and the USN's Naval Tactical Data System.
In 1948, the Canadian Defence Research Board (DRB) sent a letter to various Canadian electronics firms informing them of their intention to start a number of projects that would partner the military, academia and private companies. A copy of the letter was sent to Ferranti Canada, then a small distributor of Ferranti's United Kingdom electrical equipment. The letter was forwarded to the then-CEO of Ferranti in the UK, Vincent Ziani de Ferranti, who became excited at the prospect of enlarging their Canadian operations largely funded by the government. At a meeting in October 1948 de Ferranti was disappointed to learn that while the DRB was equally excited, the amount of money they had to offer was basically zero.
Word of the meeting reached Jim Belyea, a researcher at the Navy's electrical laboratories outside Ottawa. Belyea had been developing the idea of an automated battlefield control system for some time, after having studied the problem of dealing with a coordinated attack by submarines on convoys. During World War II the slow speeds and short submerged range of the typical U-boat allowed the defenders to deal with them one-by-one, but as the capabilities of the newer Soviet designs improved it appeared that a coordinated all-underwater attack was a real possibility, one for which he felt an effective defence would require much faster reaction times.
Belyea's idea was to share radar and sonar data between ships, processing the data in order to present a unified view of the battlefield relative to any particular ship's current heading and location. Belyea had experience with naval training simulators, and thus knew that conventional electrical analogue computation and display would not be sufficient for DATAR.
Belyea's basic idea of sharing precise real time radar and sonar data between all ships in a convoy, compensating for ship movement and distinguishing between friendly and enemy ships was years ahead of its time. Indeed, it was a quantum jump into the future and although I am by no means up to date at the time of writing (September, 2002) I am virtually certain that all modern naval task forces basically incorporate the Belyea concepts.
However he had no good idea how to accomplish this, so he approached Ferranti, who had recently met with the DRB. Instead of the cash-strapped DRB, Belyea offered funding directly from the Navy itself. As Belyea was a lieutenant, he only had authority to approve contracts up to CAN$5,000. As a cunning solution, Belyea put out several contracts under different names all to Ferranti. This solution pleased everyone and the DATAR project was born in 1949, Ferranti setting up a new shop under the direction of Kenyon Taylor in Malton near the Avro Canada plants.
The DATAR prototype
By 1950 the small team at Ferranti Canada had built a working pulse-code modulation (PCM) radio system that was able to transmit digitized radar data over long distances. The opening of the Korean War dramatically shifted the government's spending priorities, and 100 new ships were ordered in 1951. Along with this came renewed interest in DATAR, and over the next two years they spent $1.9 million (almost 15 million in year-2000 dollars) developing a prototype. The prototype machine used 3,800 vacuum tubes (another source says 20,000, another 10,000) and stored data for up to 500 objects on a magnetic drum. The system could supply data for 64 targets with a resolution of 40 by 40 yards over an 80 by 80 nautical mile grid.
In a production setting, only one ship in a task force would carry the DATAR computer. The rest of the ships had computer terminals that allowed the operators to use a trackball and trigger to send position info over the PCM links to the DATAR. DATAR then processed the locations, translated everything into the various ship's "local view", and sent the data back to them over the same PCM links. Here it was displayed on another console originally adapted from a radar unit. In contrast with the United States Air Force's Semi Automatic Ground Environment (SAGE) system, DATAR did not develop "tracks" automatically, relying on the operators to continue feeding new data into the system by hand.
"Battleships" on Lake Ontario
The system was first tested in the fall of 1953 on Lake Ontario. A simulated convoy was set up, consisting of a shore station and two Bangor class minesweepers, the HMCS Digby and HMCS Granby. DATAR performed well, everyone being sent proper displays of the radar and simulated sonar "blips". The test was a complete success, and the Navy was apparently extremely pleased. The only serious concern was the failure rate of the tubes, which meant that the machine was non-operational for a considerable amount of time. Ferranti was extremely interested in adapting the DATAR system to a transistor-based design, which they believed would solve this issue.
However, equipping the entire Royal Canadian Navy's fleet would be extremely expensive. Production machines would likely be cheaper than $1.9 million prototype cost if produced in any sort of production run. In order to lower the overall cost, the Navy wanted to spread the development costs across a larger production line, and invited representatives of the Royal Navy and US Navy to view the system. They proved to be equally impressed; one US officer was too impressed and looked under the display console, believing the display was being faked. But no matter how impressed they were, it appears they felt they could do better on their own, and declined to get involved. The Royal Navy would start work on their Comprehensive Display System that year under the direction of Elliot Brothers, and the US Navy's Naval Tactical Data System in 1956.
The DATAR project thus ended on a somewhat sour note. The system had gone from concept to working prototype in less than four years, and was by any measure a complete success. Yet the cost of deployment was simply too much for the Royal Canadian Navy to bear alone, and they decided to simply do without.
Luckily the work did not go completely to waste. Ferranti Canada used the basic DATAR design on a number of projects, transistorizing it in the process. The system eventually led to both ReserVec and the Ferranti-Packard 6000 mainframe.
- Norman Ball and John Vardalas, Ferranti-Packard: Pioneers in Canadian Electrical Manufacturing, McGill-Queen's Press, 1993, ISBN 0-7735-0983-6
- David Boslaugh, When Computers Went to Sea, Wiley, 2003, ISBN 0-471-47220-4
- Arthur Porter, So Many Hills to Climb, Beckham Publications Group, 2004, ISBN 0-931761-18-2
- John Vardalas, "From DATAR To The FP-6000 Computer", IEEE Annals of the History of Computing, Volume 16, Number 2 (1994)
- The men who really invented the GUI By Clive Akass
- Richard Howard Gimblett, Michael Whitby, Peter Haydon, The Admirals: Canada's Senior Naval Leadership in the Twentieth Century, Dundurn Press, 2006, ISBN 1-55002-580-5