return to IBM 1401 1950s Team Bios
The 1401 Reference Manual Story
Jan Swanson Loper Barris
In 1956 I was in the process of graduating as a marketing major at Syracuse University when I was hired by IBM to be one of fifty-five women from all over the United States who made up the first major wave of female Systems Engineers. After three months of "machine School" in Buffalo, New York, I was assigned to the Rochester, New York branch office as part of the sales support team.
In those days Systems Engineers spent most of their time wiring control panels (plugboards) for customers’ electric accounting machines like the 602 and 604 calculators and the 402 and 407 printers. And, on a regular basis, we also taught customer classes in control panel wiring. This not only increased their self-sufficiency, but also helped to lighten our workload. I loved working with the customers, but I was not at all fond of wiring and testing control panels on tight schedules.
In late 1958 I was loaned to the Product Publications department in Endicott to solve a problem with a 305 RAMAC demonstration, which I did. The manager of that department, Noel Capettini (Cappy) asked me if I would be interested in joining his staff as a technical writer. I loved to write and I was eager to give this new job a try. (After all control panel wiring was not in the technical writer’s job description.) It happened; I was transferred to Endicott.
In early 1959 after I had successfully completed a couple of writing assignments, Cappy took me aside and asked me if I would like to lead a writing team in the preparation of the user manuals for an exciting new machine being developed in the Glendale Laboratory. It was to be the first stored program computer designed especially for the business community and it was going to revolutionize the world of billing, inventory control, accounts receivable, sales analysis, payroll, accounts payable and general ledger applications. What’s more, it didn’t have any vacuum tubes or control panels. I jumped at the chance!
Learning about the Machine
Clearly, I needed to get a "global" perspective on this engineering marvel if I was to plan the contents of a manual that would meet the needs of those customers I left behind in Rochester. So
The very next day I made an appointment to see the Chief Architect of the SPACE (Stored Programming Accounting Calculating Equipment) machine, Fran Underwood.
Fran and I hit it off immediately. We had a common purpose – to make this machine so attractive to its target audience that it would be a best seller for IBM and a joy for the customers to use. He knew the machine inside and out and I knew the audience and what they needed to know and understand. He was only too happy to regale me with stories about his elegant system design while I took copious notes. He had ready answers to most of my questions, and, if there was a detail missing, he knew who had the answer.
I remember the "learning sessions" as if they were yesterday. First came the Input Output devices, with heavy emphasis on the chain printer, followed by explanations of SPOOL (Simultaneous Peripheral Operations Online) techniques. Then came transistors, magnetic core storage, binary-coded decimal, and word marks. (I remember writing in my note pad in the shuttle back to my office the day Fran gave me the lecture on word marks: "How many words would a word mark mark, if a word mark could mark words?")
Writing the Manual
I remember thinking to myself. "If we (the publications team) can put all this knowledge in a form the customers can easily understand, they will be successful and so will the 1401."
Remember this was 1959 and there were no "application packages". Systems engineers and customers had to write stored programs for the 1401 from scratch in the field—at first in actual machine language or in a one-on-one symbolic language called Symbolic Programming System (SPS). Higher-level "macro" languages like Autocoder, Fortran and Cobol came a little later as did the Report Program Generator (RPG) and the Input/Output Control System (IOCS).
So we found lots of pictures, drawings and charts to illustrate new and/or difficult concepts. I put lots of examples in the manual (one for every instruction type) because I knew that examples were key to understanding how the instructions could be used. Fran, the engineers, the programmers, and other knowledgeable technicians were always available to critique our work to assure accuracy and make the manual just a little bit better.
Also in 1959 there were no word processors, Xerox machines, spell checks, grammar checks or desktop publishing programs. And, of course, there was no room in the computers for online information that could be quickly accessed through a Help command.
We wrote our manuals in longhand and secretaries typed them up. If we needed to make copies we used a ditto machine or mimeograph. When we had finished a section it had to be reviewed for technical accuracy by an engineer or programmer. Once it was technically correct it had to go through a literary edit. Then it had to be sent off to a typesetter and arranged on long "galley" sheets. When we were sure the galleys were correct, each section had to go to the art department where they pasted in photographs and charts and made up the camera-ready copy. Then, when the whole manual was ready, the camera-ready copy was sent to the printer. It’s a wonder we ever managed to get those manuals ready for announcement day!
The 1401 Announcement
On October 5, 1959 the 1401 was announced. The sales force and the customers were ecstatic. The machine they had been waiting for had finally arrived. They understood it and they loved it. Sales the first week exceeded all expectations. And back in the development lab we walked on air for months while we prepared the first systems for delivery.