Channels ▼


A Personal History of Systems and Computers: Part 5

Dan can be contacted at

It's December 1968 and I'm driving south on Interstate 81 between Harrisburg and Shippensburg, Pennsylvania. It's about 7:00 in the evening; it's dark and spitting sleet into the headlights of my old car. On the stretch between Carlisle and Shippensburg, I'm in the only vehicle on the road. I'm taking two programs to be tested on the RCA Spectra 70 at Shippensburg State College and wondering what in the world I'm doing here?

During the last half of the 1960s and the first few years of the 1970s, the demand for computer programmers was so great that there was a bidding war for our services. During the '60s, I changed jobs every year. Once TRW's conversion from unit record machines was completed, I began to look around town for something more profitable. Since Harrisburg is the capital of the Commonwealth of Pennsylvania, I checked into the opportunities available with state government. I found that the Commonwealth needed a lot more programmers than were then available. The problem was that in order to qualify as a Commonwealth programmer, you had to deal with the Civil Service examinations.

Naturally, I knew people in Harrisburg and they knew people. One of those people introduced me to the Director of Data Processing for the Civil Service Commission who was looking to fill an opening for the Chief of Programming in his own agency. I learned that, as in most things, there is a difference between knowing the Civil Service rules and understanding those rules. There were, indeed, examinations for computer programmers but there was no test yet available for the chief of those programmers. If I met the requirements for the job, I could be provisionally appointed to the job. I had almost three years programming experience, I managed a junior programmer during the conversion project, and I received a provisional appointment, with civil service benefits, as the Chief of Programming for the Civil Service Commission. Once a test was available for the position I had six months to take it and pass it. Of course, it was the Commission, itself, that created the examination.

So, during the fall of 1968 I began my chief-ship. The CSC had six programmers each with a specialty. One programmer was responsible for the scoring program, another programmer was in charge of the program that mailed notices to job candidates, and so on. The programs had few changes and the programmers had little work. I, on the other hand, had no work to do so I went exploring.

Like many of the smaller agencies, the Civil Service Commission did not have its own computer. Instead, we used the services of the Bureau of Centralized Electronic Data Processing (BCEDP). The BCEDP's primary responsibility was preparing the Commonwealths payroll and they had just completed converting the payroll programs from a Univac 1 (the last in operation in the U.S.) to an RCA Spectra 70. IBM was a New York company. Univac and RCA were from Pennsylvania and New Jersey and the Commonwealth had bowed to local pressure and purchased several RCA Spectra 70 computers.

I called my contact at the BCEDP and made an appointment to tour their facility. As we made our way to the computer room we passed cards. Think rooms of card files. Think stacks of card drawers. Some of the drawers held cards that had already been processed, other drawers had cards to be processed, while still others might some day be processed. We passed sorters and interpreters for the cards and bursters and decollators for the printed checks. And finally, there was the computer room itself. The Spectra 70 mainframe was placed at the end of two rows of tape drives. A card reader stood on one side of the mainframe, a printer on the other side and the console faced the front of the room. The only devices that I recognized were the tape drives.

I had decided on a "me, too, only cheaper" marketing strategy. The Spectra 70 used the same assembly language as the IBM 360 and it supported a similar COBOL compiler. You could take your existing IBM programs and by reassembling or recompiling them, use those programs almost unchanged on the Spectra 70. But, it was cheaper equipment. I didn't recognize the card reader because RCA had decided to use a tabletop design. The IBM 2540 reader engaged a gravity-feed hopper where the cards were stacked at a 60 degree angle and fed into the machine. The RCA reader had a perfectly level surface and the cards were loaded in rows on the tabletop, grabbed by a rubber conveyance and transported to the read head. It was loud, inefficient, and prone to jam. It was not, however, the worst of it. I'm convinced that the downfall of the Spectra 70 and, with it, RCA as a computer manufacturer was the RCA printer. The IBM 1403 printer was so reliable and produced such high-quality product that it lasted beyond the second generation and was attached to every 360 that IBM shipped during the '70s. The RCA printer, however, couldn't print a straight line. The first four or five words or numbers of a row were level and, then, two numbers or letters of the next field would drop below the horizontal. The quality was unsatisfactory and RCA seemed powerless to correct it. It was, however, acceptable for government work.

After the tour, my contact schooled me on testing procedures. He gave me sheets of paper to be completed and wrapped around our card decks. The sheets of paper identified what tapes and tape drives to use, they specified any operator input needed, and they provided a place for special instructions. Once our programs, properly wrapped, were delivered to the BCEDP they would be run sometime during the night. The output was returned to us the next morning. What my contact didn't tell me was that the operators weren't able to read. At least, that was the only conclusion that I could make after reviewing the delivered product. When you have only one compile or test each 24 hours, reruns are costly.

I kept exploring. I called the local RCA office and spoke to someone who told me that Spectra 70s had also been installed in each of Pennsylvanias state colleges. I investigated further and found that the closest state college was in Shippensburg, a comfortable 45 minute drive from our office. After several more calls, I contacted the data center manager at Shippensburg State College (now Shippensburg University) and discovered that their Spectra was used only eight hours a day, five days a week. Armed with this information, I made an appointment to see my boss, the Director.

The Director of Data Processing for the Civil Service Commission was Richard Coles. Unlike the rest of us in the department, he was always busy. In his mid-30s, he was already a bona fide bureaucrat. He was kind to me, but, he could be frustrating.

"Dick, I think I've come up with a way to really improve the turn-around of our tests," I told him.

"Really!" he said. He showed some excitement. Increased productivity was often a good thing, even in state government.

"I found out that Shippensburg State has a Spectra 70 that we can use at night and over the weekends. I can take our compiles and tests over there, run them, make changes on site, and probably do two or three runs at a time," I explained.

"What will they say at BCEDP?" he wondered.

"Dick, they will never miss us."

It took some doing but we worked it out. Our director had to send their director a letter. We had to exchange budget center numbers and agree on a fictitious hourly rate. Since I would be operating the machine myself, I had to agree to their rules and regulations. They would give us a few tapes and a small place to keep them. All that was left to do was to talk to the programmers.

I called a meeting of the programming staff and with no small amount of self-satisfaction, I reported my success. We would do in days what had taken weeks to accomplish in the past. Things would get done and projects would be completed. My announcement was greeted with a minute of silence.

"Did someone complain?" one of the programmers wanted to know.

"No," I answered, "no one complained."

"Why are you doing this to us?" our one female programmer protested. Tears were forming in her eyes.

"Weren't we all pulling our hair out with the mess at BCEDP?" I asked them.

The programming staff, it seemed, did not share my frustration. They liked a work schedule that could be completed in 30 minutes and they had no problem with finding things to fill the remainder of their day. As the meeting progressed, we came to a compromise. The programmers would continue to send their compiles to BCEDP. If they already had test tapes and other materials at BCEDP, those programs would be tested there, also. I would take only new projects to Shippensburg.

So, with little to gain, I found myself on a nearly deserted highway in the middle of a freezing rain storm taking the first two programs to the computer room at Shippensburg State College. Once there, I found the data center and things went as planned. I was able to run the Spectra with little instruction. I performed the two tests, made some changes, and ran them again. In less than two hours I was on my way back to Harrisburg.

The next morning, the last of my hopes for the department died. I delivered the test output to the programmers. After reviewing the work that I had done for her, our female programmer brought the sheet of paper with her test instructions back to my desk.

"You didn't sign it," she said. She waited patiently while I placed my name on the appropriate line.

Related Reading

More Insights

Currently we allow the following HTML tags in comments:

Single tags

These tags can be used alone and don't need an ending tag.

<br> Defines a single line break

<hr> Defines a horizontal line

Matching tags

These require an ending tag - e.g. <i>italic text</i>

<a> Defines an anchor

<b> Defines bold text

<big> Defines big text

<blockquote> Defines a long quotation

<caption> Defines a table caption

<cite> Defines a citation

<code> Defines computer code text

<em> Defines emphasized text

<fieldset> Defines a border around elements in a form

<h1> This is heading 1

<h2> This is heading 2

<h3> This is heading 3

<h4> This is heading 4

<h5> This is heading 5

<h6> This is heading 6

<i> Defines italic text

<p> Defines a paragraph

<pre> Defines preformatted text

<q> Defines a short quotation

<samp> Defines sample computer code text

<small> Defines small text

<span> Defines a section in a document

<s> Defines strikethrough text

<strike> Defines strikethrough text

<strong> Defines strong text

<sub> Defines subscripted text

<sup> Defines superscripted text

<u> Defines underlined text

Dr. Dobb's encourages readers to engage in spirited, healthy debate, including taking us to task. However, Dr. Dobb's moderates all comments posted to our site, and reserves the right to modify or remove any content that it determines to be derogatory, offensive, inflammatory, vulgar, irrelevant/off-topic, racist or obvious marketing or spam. Dr. Dobb's further reserves the right to disable the profile of any commenter participating in said activities.

Disqus Tips To upload an avatar photo, first complete your Disqus profile. | View the list of supported HTML tags you can use to style comments. | Please read our commenting policy.