migration of calculation software from 3rd-party mainframe to in-house PC saves time and money
One of my first programming opportunities came while working at National Valve (NAVCO) in 1984. One of our engineers had written a large program, called BORC (Branch Opening Reinforcement Calculation) designed to calculate reinforcement requirements for nozzles on pipe headers. It took into account things like pressures, temperatures, wall thickness of the pipes, and other factors. The program was large, complex, and it was written in BASIC.
In those days computing power was expensive. To reduce costs, NAVCO rented computer time on off-site computers. Writing and running the calculation programs at NAVCO were done using a teletype console, paper output, and an acoustic modem. A monitor was not part of the system! Since there wasn't a monitor, all program input was done in a batch mode and you had to be very careful when typing the input. Mistakes meant starting over from the beginning. Running the programs was slow and laborious.
At the time I was learning to use the teletype system, I was also learning CAD on one of our new Tektronix machines. The early Tektronix machines had CAD software (Technicad) built-in, unlike today where CAD software is purchased separately then installed on a computer. The Teknicad computers also came with a BASIC interpreter in an optional ROM pack and an integral cartridge tape drive. I soon learned to use both of these to my advantage.
My thought was to rewrite BORC for use on the Tektronix machine using its BASIC and tape drive to store the program. The problem was it would take several days to rewrite the program and test it. I wasn't sure management would agree to let a novice programmer undertake such a project, so I decided to do it on my own and arrange a demonstration when I was finished--if I succeeded.
I printed a copy of the original program from the teletype machine, then retyped it into the Tektronix machine where I made syntax changes as needed and saved the program to tape. I also created input screens and result reports to make the program interactive and easier to use. I rewrote the entire program, 100s of lines, on my own time, during lunch, before and after work. When the program was finished, I created a simple menu program which contained BORC and several smaller programs. I was ready for the demo.
The Tektronix's large screen was great for a demo. With the chief engineer, and several other engineers, looking over my shoulder I loaded the tape and let the Auto Start mechanism locate the tape's initial program, the menu, and run it. I selected BORC from the menu and started entering data in the screens. I could hear a chorus of "oohhs" and "aahhs" behind me. They got louder when the results displayed on screen with options to print them or reenter new branch data.
The demo was a huge success. It showed that programs could be developed and run in-house quicker, easier, and cheaper than before. NAVCO could stop paying others for computing time.
I was then given the assignment of converting the other half dozen or so programs for use on the Tektronix. I wasn't looking forward to retyping 100s of lines again so I explored another possibility.
The Tektronix also came with a modem jack. I found an old Bell 212 modem in another department and wrote a little communication program to initialize the modem and enable the Tektronix talk to it. Once connected to the remote computer, I was able to transfer the programs from the remote system and capture them to tape. After that, editing was a breeze. To cap it all off, I wrote a brief user guide for each program.
It was a memorable and worthwhile experience. I learned a lot, the company saved money, and future programming efforts were far more efficient and cost effective.