Roland F. Bryan
8 September 1995
Dedicated to Glen Culler
In 1961 the RW 400 Polymorphic Computer System was new and unique and was designed to service the first spy satellite by creating and displaying maps of the world's surface. It was unique in that it consisted of a giant switch with 16 computers on one side and 64 slave devices on the other. Any computer could reach any support device.
One of those devices was a large screen graphic display that could draw lines and curves, and which served to position symbols and characters on the surface of maps to record satellite findings. A person operating the display had two kinds of keyboards with which to enter information and commands. One was like a standard typewriter and the other a "function" keyboard where the keys could be programmed to enter any type of computer instruction.
It was this line drawing graphic display system that attracted Drs. Glen Culler and Burton Fried, who saw that the function keyboard could be programmed to enter mathematical equations and that the display could produce the curves and surfaces the mathematics described. With the help of the computer system, results would be immediate, producing x, y, and z positions on the screen and drawing the curves and surfaces generated from the equations that were entered.
By 1963 the Culler-Fried system, or the Interactive On Line System as it came to be called, was operational and allowed users to enter sequences of single key strokes to perform the most complex mathematical operations. The user could then store an entire sequence under a single key so as to repeat the function and then could string together any number of such functions and store those under another key as a User Program for later use.
The idea of having single keys and groups of keys manifest mathematical operations and then to have them stored under another single key as a user program were only two of the powerful new concepts of Culler and Fried. To this they added the visual display of curves and surfaces as quickly as the user could push the keys.
For the first time, a person at a display console could interact with the computer and see the results of every entry on the screen. With this system the user could bring intuition to bear so as to "steer" mathematical expressions toward solution by incremental steps, watching results of a single change in the equation and removing that change to try another, until finding the right combination. This was a true example of creative man-machine coupling to produce rapid resolution of mathematical problems.
First, they had to produce an inexpensive console because the large displays cost $100,000 or more. Next, those large displays needed the computer close by, so a method had to be devised to move the key codes and the graphics data over telephone lines to, and from, the computer system in Canoga Park. Keyboards existed but no small display consoles as we know them today.
Close-up photo of Teleputer keyboard
The solution was to take a pair of Teletype keyboards and a piece of laboratory equipment, a Tektronix Storage Oscilloscope, for the display. These were mounted in a console and the key tops were re-labeled for the mathematical operators. Electronics were added to take each key-push and send those over a telephone line to the computer system. A second telephone line carried the display information, in digital form, back from the computer to the remote console. Analog signals were produced from the digital information to drive the display. This remote console was dubbed the "Teleputer" by Culler and several years later was being produced in quantity for connection over telephone lines from locations around the USA.
After the Teleputer was developed, Culler returned with it to his campus at the University of California at Santa Barbara and operated by connection to Canoga Park. Fried, who also consulted to TRW, helped to specify another complete On Line System with 4-consoles for scientific application at TRW in Redondo Beach, California.
Bolt Beranek and Newman obtained the rights to build the remote systems and delivered them to UCLA, Livermore Labs, University of Kansas, and Harvard (for Burton Fried, Richard Post, James Rohlf, and Anthony Oettinger, respectively). Roving consoles were also built so that Culler and Fried could demonstrate the On Line System at conferences.
The next event for Glen Culler was the re-creation of the On Line System on an IBM 360 to operate classrooms full of Teleputer consoles for student use in Engineering, Sociology, and Physics. The remote users were moved to the IBM system, and in 1969 the system was connected to the ARPANET so that users could reach the On Line System over that network from numerous sites. The UCSB campus was served by the On Line System throughout 60s and 70s and students had access to its powerful approach which remained unlike anything available at any other university for many years.