Return to IBM 1401 1950s Team Bios
Excerpt of e-mail, tale about O M Scott from John, Dec 23, 2018
These are excerpts from my autobiography entitled "From Beer to Paternity" which I wrote for my children and their descendents. One chapter encompasses my four years in Endicott, NY (1959-1963) where I worked at the IBM development laboratory. Three years were spent on the 1401 project. This was my first professional job, having just received my BSEE. In addition to technical details, it includes stories (sometimes humorous) of a young man becoming exposed to the world. I will not edit or change my recollections of the days when I just turned twenty-two. I am too old for worrying about political correctness or inaccuracies. Just remember, the older I get, the better I was.
IBM 1401 Memories from Endicott
by John Pokoski
Table of Contents - added for accessability
Summer and Fall of 1959
- 1st Day
- 1st Supervisor, Fran Underwood
- IBM's Old Business Model
- New Leader, New Focus
- The Race for an Inexpensive Accounting Machine
- I was completely green
- 1401 passed A-Test, Closed Circuit TV Announcement
- Stress about the 1410
- I'm responsible for testing and debugging the B-test machine
- Bosses and top engineers lead by example. Log Book
- Core Memory Problems
- Core Memory Design Problem and Log Book
- Third shift was the toughest
- Second shift was an easier adjustment - Playboy Magazine
- Debugging the machine is like detective work
- First factory machine with new Multiply/Divide
- Making new Multiply/Divide work
- Emphasis Change, into Testing, Production, Field Service
- Noon Breaks are Serious! Bridge and Pinochle ;-)) The Camera
- Office Life, Poker Evenings
- Called "on the carpet" :-?
- I write Angry Letter !!
- I get promoted to Associate Engineer, Early :-))
- Tom Watson Jr. to show off a 1401, and it doesn't Multiply-Divide
- Happy Ending. Software in English?, Touring New York
- Automated Logic Diagrams, ALDs
- 1401 the Money Maker. Now produce them!!
- Clumsy Carp and the 1403 Printer
- World Trade and British Sterling units
- Have Oscilloscope, Will Travel ;-))
- Field Service is a different world. Chicago Adventures
- Office Christmas Party
On our first day at work, we had to initially go through the personnel bureaucracy. In addition to getting benefits information, we had to sign certain secrecy agreements and intellectual property agreements. This didnít bother me since I figured the company had the rights to what they paid me to do. But we also had to sign certain other legalistic agreements regarding conflict of trade. These documents seemed more appropriate for a salesperson. It turns out that IBM had gotten into anti-trust trouble with the federal government, and had signed a consent decree. Among other things, it required IBM to refrain from removing IBM equipment from customers who also deigned to rent or purchase other equipment from competitors. You see, IBM salesmen were often very cutthroat, and the monopolistic position of the company gave them a huge edge over their competitors. Certain pieces of equipment were made only by IBM, and a threat to refuse to rent them to a customer could be very persuasive. I was later astounded to learn that in certain situations, the blueprint designs of new IBM products had to be made available to key competitors.
After the paper work was completed, it was explained to me that during the summer, I would be assigned to the department I expected (Walt Schaefferís and Byron Ruckerís) full time. In September, all new engineering hires would go through a training program, each morning, for several months. This would be mostly technical, including everything from transistor electronics to switching theory to tape drives to programming. This was necessary, since we would be doing state of the art work, and most universities didnít cover this material. The course also covered IBM history and policies. Also in the fall, a rotation program would be begun. Each new hire would spend a couple of months in one department, and then be rotated into another, then another, so that the person could get a better picture of his interests and opportunities. However, you could choose to stay in one department if both sides were sure of the fit.1st Supervisor, Fran Underwood
With that, I was led down to a lower building, to my department. I would learn more about the engineering profession and computers in the next year than I would in the remainder of my life. I was brought into the office of my supervisor, a tall, dark-haired, rugged fellow in his forties named Fran Underwood. He was dressed in a dark blue suit, white shirt, and tie, the IBM uniform. No one ever told us to dress this way, but every male did, including clerks, technicians, and draftsmen. The suit wasnít necessarily blue, but it was generally dark. Usually the coat was draped over the back of your chair while working, but the tie always remained on. Fran sat me down and described what I was getting into. "You will be working on a machine which will turn this company on its ear. It will be the biggest money-maker IBM has ever had. This town will boom with more new jobs than ever before, and the economy will prosper." The air was electric!
And his words were prophetic. I had fallen into a dream. My first job was going to be involved with a wildly successful system. Many engineers never experience that thrill. I later found out that Fran was the "father" of the 1401, the computer the group was designing. I donít believe that he even had an engineering degree, but Iím not sure. He had been a draftsman, and his smarts, self-education, and drive had led him up the ladder. He had conceived this machine, and he was its driving force. Fran knew that this was my first engineering position, and he was nice enough to give me some advice. He leaned over, and pointed out the door to an engineer a few years older than me. "See that guy? His name is Paul Farbanish. Heís going places. Watch him, do what he does, and youíll be alright."
IBM's Old Business Model, Unit Record Equipment
At that time, computers were built of vacuum tubes and relays. They were huge and expensive, filling large rooms and costing millions of dollars. It was only about ten years earlier that Tom Watson Sr., the founder of IBM, had nixed the idea of investing in computer development, since he was advised that two computers, one on the east coast and one on the west coast, could solve all the problems which a computer was capable of solving. IBMís growth had been based on "unit record equipment". For example, a workerís weekly hours and pay-rate would be keypunched, each week, into an IBM card by a "keypunch" operator. The stack of punched cards for all employees of the company might then be run through a "sorter" which, by sensing the appropriate punched holes in the cards, could arrange them in some sort of order-say by last name or by social security number. Then they would be run through a "calculator", which might multiply the hourly rate, as punched in one field of the card, by the number of hours worked, as punched in another field, and punch out the weekly salary in a third field.
The stack of cards would then be run through a "printer" that would use the name and salary fields to print out paychecks. The paychecks were distributed to the workers and the stack of punched cards or "unit records" were stored away. They might be used later for further analysis of work patterns. Each of the pieces of unit record equipment was electromechanical and was controlled by a plug-board that was wired by the operator to select the appropriate fields and operations. IBM generally leased all of this equipment to each customer, and took responsibility for training the customer, maintaining the equipment, and selling the printer paper and punched cards. It is obvious how a customer could become completely dependent on IBM, thus making him vulnerable to IBMís price increases or "upgrading" to more expensive equipment.
New Leader, Continuing Benefits, New Focus, Computers
Soon after the founderís son, Tom Watson Jr., took over the companyís leadership, he became convinced that computers were the wave of the future, and that for business applications a single computer could process data more efficiently than unit record equipment. He moved the company in that direction, despite his fatherís opposition.
I should mention here that IBM was an unusually paternalistic company. During the depression, Tom Watson did not lay off employees, but he shortened workweeks and lowered salaries so that everyone could remain employed. He also instituted stock purchase plans which allowed employees to have money deducted from their paycheck at any rate desired until it reached 85% of the stock price at either the start or the completion of the purchase, whichever was lower. You couldnít lose. You had to make at least 15% on your investment. So, even the lowest level workers were owners of the company.
Each major IBM facility had a Country Club, which all IBMíers automatically belonged to. At Endicott, there were two beautiful golf courses, one eighteen holes, and one nine. I believe it cost ninety cents to play a round on the eighteen-hole course. If the janitor was signed up for a 9 AM tee time, even the head of the lab couldnít bump him. There was a game room with pool and ping-pong tables. There was an auditorium where free cultural events were regularly held. But there was no booze. It was verboten at IBM. Again, there was an unwritten dress code. That is it was unwritten for engineers. There may have been a handbook for salesmen.
There were many internal classes, as well as college classes taught on-site, free. In addition, low level people could take "IQ" tests, which might allow them to take courses and move up the ladder. Believe it or donít, there was an IBM songbook. It was an item from the past, but I saw one. It included such ditties as "IBM Blue" and "Hail to the Country Club". (IBMís color was blue, and its nickname was "Big Blue".) Once at the IBM school, where I was preparing to teach a class,
I heard an auditorium of salesmen warming up by vigorously singing "Stouthearted Men" with piano accompaniment. Much of this may sound quaint and silly, but the company had an extremely diligent and loyal group of employees. I knew some thirty-year vets, and the company could do no wrong in their eyes. Many times, unions tried to crack IBM, but to no avail.
Anyway, with the advent of Tom Jr., IBM quickly moved to the top of the computer market, despite the early lead of Remington-Rand who built the UNIVAC, the first commercial computer. IBMís computers were used on the DEW line to monitor possible USSR attacks over the North Pole. IBM had large computers used for scientific and data processing, like the 704. But smaller companies couldnít afford such expensive systems. Transistor technology gave the potential of greatly reducing size and cost, while increasing reliability.
The Race for an Inexpensive Accounting Machine
The 1401 was to be IBMís answer for an inexpensive accounting machine. It was known that a French company called BULL was working on a similar machine. The race was on as to who would be the first to announce, and to dominate the market. Such a machine would eventually eliminate the need for much of the unit record equipment, and it was imperative to IBM to win the race.
When I joined the group, much of the design of the first prototype was completed, and it was being assembled in the lab. After testing and debugging in our own lab, it moved on to a separate group called "product test" that had to independently check it for meeting specifications and for reliability. This was called "A-test". Temperature, humidity, and power variations were used in this testing. If it passed ok, and the market was still there, the machine was announced, and the salesmen went to work.
While this was going on, a second prototype was being built, incorporating everything learned, all engineering changes, some new technology, and optional features. This would be the prototype used for manufacture, and would also go through product test ("B-test"). The first machine manufactured at the factory went through "C-test". Once the machine was announced, training courses had to be set up for salesmen, programmers, customer engineers, factory engineers and technicians, and even customers. This was truly a big operation, and a lot was at stake.
The head of the operation, three levels above me was Chuck Branscomb, a youngish (38?), tall, handsome, bright engineer. He had two people directly report to him. Al Miller headed the input-output (I-O) group, which handled the printer and the card reader and punch. (Tom Schappe worked for him.) Jim Ingram headed the central processing unit (CPU) group, which also covered the tape drives. Jim was a southerner, and a shrewd guy. He "held my card". That is, he had control of my time card, which meant that he made final decisions on salary, promotions, etc. However, Fran Underwood really handled the technical decisions on the machine, and was almost equivalent to Jim.
As I mentioned, Fran was my first boss. There was a clerk, Ernie Dunbar, and two secretaries, Barbara Downs and Alice ?. There were several staff engineers, including Walt Schaeffer and Ken Bell, both middle-aged, bright, super-nice guys. Tom Cooper was a Staff Engineer heading up the "memory" section. (IBM later deleted all use of the word "memory" and replaced it with "storage" to reduce peopleís fears of computers being too human-like and taking over the world.) There were several younger Associate Engineers (ranked just below Staff) including Paul Farbanish, Jim Harvilchuck, and Byron Rucker. There were several Junior Engineers, including Jim Alstrom, Marty van der Pool, and me. However, Jim and Marty were in the reserves, and would be leaving at the end of the summer for the military. (I had a one-year critical skills draft deferment.) We were close to being professional staff, but not quite. For example, we could not work overtime without getting paid. There were also several technicians, including Joe Papa, Jud McCarthy, John Young, and Andy Kisacky, and a few modelmakers and draftsmen.
I was completely green
I was completely green about computers and industry, so Fran put me under John Youngís wing, and also arranged for Harvilchuck and Farbanish to tutor me on the circuitry and on the architecture of the 1401. I was given stuff to read and study, but the memos and drawings really werenít meant to be teaching tools.
My first real job was to learn the multiply-divide optional feature, and become expert on it. It was designed but not tested yet. My very first task was to make a "third level" logic drawing. This was a large schematic that I drew on a drafting board. So I was a draftsman after all. Fran, who was quite artistic, monitored my progress. Eventually, I worked on the debugging of the prototype, but that job was essentially completed by the time I got involved.
By the end of the summer, the prototype was in A-test and the second prototype was being assembled. I started going to my half-day classes for new hires, but declined being rotated to different groups. I could see that I was lucky to be on a really hot project, and I didnít want to lose time by temporarily working in another department.
1401 passed A-Test, Closed Circuit TV Announcement
That fall, the machine passed A-test, and a live, closed-circuit television network was set up to make a grand announcement at various IBM centers across the country. IBM bigwigs, salesmen, and potential customers attended. The broadcast itself was from the Glendale Lab, using the A-test machine, with Watson Jr., O.M. Scott (the head of the small systems division), and other big guns on camera.
Unfortunately, just before the broadcast was to begin, the system crashed.
Our top engineers were called upstairs in a panic. There was no time to really debug the system, so they faked it by rigging it to flash lights on the console and spew paper from the printer just as though a real job was being run. The people in the hinterlands didnít know the difference. The appearance and the bullshit were all that counted. I was amazed at how cool our engineers were and how little concern there was that the machine being announced wasnít working.
from e-mail, Dec 24, 2018
The only thing I remember about the announcement was a nervous tension the day before and that morning among the senior engineers. They were a bit dismayed that the system had to be moved from A-test (which I think hadn't quite finished) and were concerned about the disruption. However, they understood the significance of the event and were excited about it. The office seemed vacant while they were gone. After it was over, they came back relieved and happy, talking a mile a minute, laughing about faking it.
It is funny, but just writing about makes me shiver, thinking about the loyalty, teamwork, industry, dedication, and perseverance of the development team-Underwood, Shaffer, Murray, Bell, Ingram, and everyone, down to the technicians and secretaries. I have never again seen anything like it.
Stress about the 1410
About this time, work was begun at Glendale on a transistorized small scientific computer called the 1410. It was to be "compatible" with the 1401. Without getting into detail, this put restrictions on the 1410 designers that they resented.
There were some smart Young Turks working on it who saw better ways of doing things than as on the 1401 (of course, hindsight is 20-20), but their hands were tied, since the 1401 was so far along, it would be awkward to modify it. However, they continued to push, and really zeroed in on the multiply-divide scheme. This optional feature of the 1401 wasn't even built and debugged yet, and it was quite cumbersome, complex, and expensive.
Finally, the hierarchy of the lab decided to have the 1410 hotshot engineers make a formal presentation to the big-wigs as well as the 1401 people who were involved. Fran Underwood and Earl Bloom were the 1401 designers of multiply-divide. Ken Bell and Walt Schaeffer knew about it generally, but werenít experts.
I was surprised and a bit nervous when someone, I believe Jim Ingram or Chuck Branscomb, asked me to attend at the last minute. I sat in the rear of the auditorium, by myself. I immediately noticed the cockiness of the Young Turks as they presented their case. Equally apparent to me was the disdain expressed by the senior 1401 designers. They clearly had a "not invented here" attitude and certainly didnít want to be "shown up".
The basic proposed change involved the way that the multiplier and multiplicand fields would be addressed. Their approach, they claimed, would save a lot of hardware required for testing digits and shifting, and would also speed up the operation. The meeting ended with no decisions made. A few hours after the meeting, Jim Ingram quietly came up to me and softly asked, "Were they right? Could all that hardware be saved?" I said it could, if the programmers agreed to the new way of addressing the fields.
I was nervous about my comments, because I wasnít really an expert. I had only studied our system on paper. I figured that Fran and Earl were dragging their feet, and that there could be real turmoil if the change were made. My guess was that Bell and Schaeffer were more open-minded, but they didnít really know the feature that well. A few days later it was announced that a compromise decision was made.
I'm responsible for testing and debugging the B-test machine
The present system currently being assembled for the B-test machine would continue as scheduled. I would be primarily responsible for its testing and debugging. Concurrently, a design of the 1410 approach would be begun. This would be carried out by Steve Bespalko and me. Steve was an engineer from the factory who had been brought over to learn the 1401 in preparation for manufacturing. He was about fifteen years older than me, a Hunky (Eastern European heritage), a Catholic, and a nice guy. If the design went ok, it would be implemented on the first machine manufactured at the factory; that is, the C-test machine. I would debug it also. Well, I had opened my mouth, and now I had to put my money where my mouth was.
Steve was all excited about the decision. He saw it as a way to shine, and to show that he was as good as the design people at Glendale, despite being a manufacturing engineer. He offered to take the lead in the new design, but he would go over things with me and get my advice several times daily. This was fine with me. I had never designed anything, and would have been pretty unsure of myself. Besides, I would have my hands full testing and debugging the current B-system. The pressure mounted, but it was exciting!
This was all happening in the summer and fall of 1959, my first year in New York.
During that fall and winter, I was working shift work, since our B-test machine had to be running as soon as possible so we could go into production. We ran three shifts, around the clock, with two-man teams alternating shifts. Usually, there would be an engineer and a technician on a shift, and we would rotate from 8 AM-4:30 PM to 4 PM-12:30 AM to midnight-8:30 AM on a weekly basis, including weekends. The bosses and top engineers occasionally worked a regular shift, but were usually around their desk to do more advanced work, satisfy the bureaucracy, and put out fires.
Bosses and top engineers lead by example. Log Book
But they were inspirational, leading by example. Even after working a ten or twelve-hour day, some would often pop in at 8 PM or midnight, just to see how things were going, and if they could help. Bill Murray was about forty years old and in charge of the card reader-punch unit. I recall him once working two consecutive shifts, taking eight hours off, and then working two consecutive shifts again. Boy was it exciting, especially when we found a design bug, and installed an engineering change that fixed it. Then we would run more advanced tests until the next bug showed up.
Some problems were solved in ten minutes, while others took days. Continuity was made by an overlap between shifts, and by a detailed engineering log that we all had to fill out. I had never been exposed to the use of a log-book before, and I only considered it to be a convenience for continuity. I believe others felt the same way, since it was not unusual to find jokes, wisecracks, and even funny signatures. No one ever commented about that in a negative way, despite the fact that they knew that the engineering log is a critical and important document.
For example, it could be used to verify patent claims in court. It could record patterns of problems over months, which might enlighten people about basic design or parts flaws. Each morning, as the top bigwigs came in between 7 and 8 AM, the first thing they did was read the log to see how things went the previous night.
Core Memory Problems
In the early stages of the process, we had periodic problems with our magnetic core memory. The problems were intermittent, and they were eventually cured by replacing some of the printed circuit boards that drove the memory cores. I was suspicious of this "fix", since the boards that were replaced tested out ok off-line. However, I didnít push it, since I knew little about transistor circuits, and less about magnetic core memory.
However, on the graveyard shift later in the testing process, the problem started showing up again. I thought this was ominous, and determined to find the root of the cause that night. After a lot of board swapping, and some analysis of the circuits, I became pretty sure that the core-switching circuits were under-powered. The circuits were designed to produce a minimum amount of current output, and they all met that minimum. However, some exceeded that minimum by quite a bit, since the transistors on those particular boards happened to be more "powerful" than others.
This was no problem, since circuits are generally designed to work correctly if all the components meet some minimum standard. My theory was that the amount of current needed to switch the cores was underestimated or calculated incorrectly. What we had been doing was selecting boards that produced current well above the incorrect minimum which had been calculated. Thus they worked, while others, while within specification, didnít. I was pretty sure of my conclusion, but didnít have the time, background, or data to prove it.
However, I wrote a long story about it in the log-book in big, bold capital letters. I said that this was a major problem that would come back to haunt us if it wasnít addressed. This was sticking my neck out quite a bit for a junior engineer less than six months on the job. When Jim Ingram came in and read the log, he spoke to me about it and he had me talk to the circuit expert in our group. He was responsible for all circuitry on the machine, and interacted with the original designers of this family of circuits. He listened to me and said he would check it out. The next day, he reported to me and to Jim Ingram that the circuits were ok. I still didnít think so, but he was the expert, and I wasnít in a position to argue.
Core Memory Design Problem and Log Book
About six months later, the factory was beginning to manufacture 1401ís, and I heard that there was a major problem, requiring that the whole line had to be shut down for a week. You can imagine the cost involved in that, since about twenty machines at about $150,000 each were involved. It turned out that the memories were erratic, and a design flaw had been found. Of course, this reflected on the Glendale designers, not on the manufacturing plant downtown.
Jim Ingram and a couple of other managers remembered my warning, found the old log and looked it up. He called me in to let me know what had happened. He didnít have to say that he wished he had believed me more. My satisfaction was tempered by how we had signed the log that night. I had written Burt and Harry Piel. At that time Burt and Harry Piel were funny characters on Pielís Beer commercials. It was actually the famous comedians Bob and Ray, playing the supposed owners of the brewery.
Did I feel sheepish when I saw the signature. Jim didnít chastise me for it. He didnít have to. I imagined that many big-shots had seen it by now. I didnít know whether to feel bad if they didnít ask who it really was, depriving me of my coup, or if they did ask, and chalked me up as a jerk. I have told this story more than once to my students to illustrate the importance of an engineering log.
Third shift was the toughest
Working the night shifts was strange. Third shift was the toughest. Along about 4 AM, your body says, "Go to sleep!" We would get pretty groggy, especially on Mondays, when our bodies had not yet adjusted to the new hours. We always took our breakfast break about that time. We would drive in to Kenís diner in downtown Endicott. This was a "greasy spoon" where we would get black coffee, eggs, bacon or sausage, and home-fries drenched in orange grease. Perfect! Ken was a jovial sixty-year old, and we enjoyed each otherís company as he cooked and we ate. But we wouldnít be alone.
There would be an occasional cop, milkman, trucker, or other early bird. Kenís was the highlight of our night. If we got back soon enough, and we had reached dawn, we would sometimes throw a football around, or hit flyballs to each other in the empty parking lot for ten minutes to invigorate ourselves. Then around 8 AM, the regulars would come streaming in. It was really tough to go home and try to sleep all day. Then I couldnít really go out at night, because I would wake up feeling groggy, and I didnít know whether to eat breakfast (again) or dinner. After eating, I didnít dare go out, as I needed to be sober and alert for work.
It was strange, but even on third shift, people wore white shirts and ties, even though no one else was around, and the work was physical, including reaching and climbing inside and under the 1401 to replace boards and string electrical cables. I soon decided this was foolish, and started wearing an open necked sport shirt and less dressy pants. Boy, did people stare at me at eight oíclock when they arrived. I heard a few comments from workers. "Whereís your white shirt?" But the bosses didnít say anything.
Soon, most of the guys on the night shifts wore casual clothes. It only made sense, and I would have defended it on that basis if need be. A few times, when there were really grubby jobs coming up on the day shift, like stringing cable, I wore casual clothes. But I felt a bit uncomfortable, and didnít do it often. Perhaps this doesnít seem very dramatic to you. Thatís because you didnít know the IBM culture at that time.
Second shift was an easier adjustment - Playboy Magazine
Second shift was an easier adjustment as far as sleeping goes. We would go eat at about 8 PM, and after work I would go home, eat a snack, and be in bed by 1 AM or so. However, when I got up at nine, I had the whole day to look forward to with nothing to do. I would read, do laundry, write home, and kill time in general. Sometimes I played golf alone, but thatís another story.
One morning, I answered a knock at the door to find a young man about my age selling magazine subscriptions. He was really a good high-pressure salesman, the first I ever encountered. After about thirty minutes, he had me convinced that I needed a bunch of magazines, and I signed subscription forms for about five. Playboy, Sports Illustrated, True, etc.
As soon as he left, I began to come to my senses. I didnít really want those magazines. I didnít read magazines that often, and I didnít want to sit around reading all night and day anyway. Also, I didnít know how long I would live at that apartment, and I figured it would be a pain to change mailing addresses for all these magazines. I kept getting madder by the minute and was cursing myself aloud for being so stupid to fall for the sales pitch.
Then, about a half-hour later, there was another knock at the door. This time it was a well-dressed middle-aged man from the same subscription service. He said that he had been told that I had bought several subscriptions, and it was his companyís policy to double check all purchases, to allow the buyer to back out if he wished. I replied, "Of course I want these magazines. I know what Iím doing!" He apologized for bothering me, and left. For the next three years, I was paying for and receiving magazines that I really didnít need or want. I also changed addresses four times in that time frame. But, at least no one suspected that I didnít know what the hell I was doing. Until now!
Speaking of Playboy, I brought one in to work one day. All the guys paged through it, particularly studying the Playmate of the month, an especially buxom brunette, who was kneeling at a forty-five degree angle to the camera. This formed quite a titillating portrait. We all knew that Fran Underwood ogled women, especially those who were well-endowed. Harvilchuck came up with the idea of putting the centerfold on Franís wall while he was out of the office. He said Fran would get a kick out of it.
The idea made me a bit nervous, since strangers could walk by before Fran got back, and think he had hung the photo on his wall himself. I got the bright idea of taping the centerfold under the first sheet of the 3íx4í drawing pad standing on the easel in his office. Fran was quite artistic, and often used the pad to sketch out ideas as he thought about them. Our desks were just outside his door, and we figured that he would do some doodling on the first sheet, flip it over, and see the girl while we smirked.
Fran was gone all morning, but he returned after lunch. To our horror, he was leading two well-dressed, middle-aged men, obviously big-shots. Fran brought them into his office, continuing the conversation. Then he stepped up to the pad and started writing on it with his marking pen. Oh my God! Sure enough, he flipped the page, and stepped back, mouth agape. I started packing my briefcase, figuring that I was not long for IBM. But, true to form as predicted by Harvilchuk, Fran broke into laughter, as did the others, as they all admired the vision in front of them. I breathed again. Harvilchuk was never worried. Later, Fran guessed who had done it, and chuckled about it. He kept the centerfold.
Debugging the machine is like detective work
I really enjoyed debugging the machine. It was like detective work, searching for clues and discussing what might be the cause of the problem. Is it a wiring error? A bad circuit board? A design flaw? We had to progressively bring the machine "up" after its first power-up. In fact, the first test after power-up is to look for smoke. Nothing works at first, and you have to gradually fix the offending bugs.
In general, the people doing the testing had not designed the particular section currently under test, so there was always a fuzziness about how it should actually work. Despite various briefings ahead of time, we often had to figure out how it should work on the spot, sometimes at 2AM, when no designer help was available.
Even if the designer was debugging, there could be some very hairy problems, especially if they were intermittent. "Solid" problems werenít bad, since you could write a program loop to continually repeat the problem, which allowed you to effectively use the oscilloscope to trace the signals until the fault was deciphered. You could never be sure when an intermittent problem would occur, causing us to rely more on mental theorizing and logical guesswork than on probing with a scope.
I learned a lot, and became a whiz with the oscilloscope. My mind was geared to the logical thinking required to troubleshoot, more than it was geared for design work. I was mainly a left-brain guy.
When we got the basic machine running, we moved into the optional features, including multiply-divide. I was the debugging expert now, and it was gratifying to see that I knew more than the others now. Still, if I really got into a bind, I could ask Earl Bloom, who, with Fran, had designed the feature. The scheme was very complicated and confusing, and there were quite a few design bugs, resulting in many engineering changes. But eventually, we got it running. After a bit more debugging, the machine was turned over to the Product Test group for B-test.
First factory machine with new Multiply/Divide
Now the first factory machine was begun. This would be like the B-test machine, except that it would have the new multiply-divide scheme as designed by Steve Bespalko, with my help. The engineers and technicians who had temporarily worked at Glendale to learn the machine went back to the main plant on North Street to begin debugging it, the C-test machine.
The top people had been sent over, since they would be the leaders as the manufacturing line came up to speed. However, I noticed that they were much more adept at finding mechanical and electrical flaws than design errors. I think that was because by the time a machine got to the plant, almost all design errors had been eliminated. Thus they were less experienced with, or even cognizant of, this type of problem.
About late November, the basic C-test machine was pretty much operational, and I was sent to the factory to get the new multiply-divide system running. I was put on permanent second shift, so no one would interfere with me, or I with anyone else. I was the only one who knew the system there. Steve was available to me, but he had another assignment now.
I knew my neck was on the line a bit, since I had suggested that this approach to multiply-divide would be cheaper, and would work. I already knew that it was significantly cheaper, since it had been built. Now I had to make it work.
Making new Multiply/Divide work
An engineer named Jim Dawson was assigned to work with me. He was about fifteen years older than me, and was a country boy from the hills of Kentucky. He was an independent cuss, and I knew that a lot of people didnít like him. However, I like direct, straight-talking people, and we really hit it off well.
He knew nothing about multiply-divide, and little about the 1401, but he knew the ins and outs of the plant. He let me take the lead, and acted as my humble technician. Not only did I have to get the m/d option working, but I needed to get it working quickly.
If it did not work, it would have to be stripped out of the machine and be replaced by the old system. This would be time-consuming, and would hold up C-test, which would delay the full start-up of the manufacturing line. I think they expected it to be finished by the New Year. However, I had planned to fly home to St. Louis for a few days at Christmas. My tickets were already purchased, so I had a definite self-imposed deadline to meet, just before Christmas.
I wouldnít leave for vacation if the system werenít completed. I had too much pride and loyalty to do that. The pressure was very high, but I liked it. Time flew. By this time, I was a pretty experienced debugger, and I also understood the new design backwards and forwards. I worked on the option throughout the nights leaving the days to the testing of the basic machine and other optional features.
We kept moving forward, fixing one bug after another. If there were a major engineering change required, I would usually check with Steve to see if he had a better solution. After a while, I developed confidence that the new system would work, but the deadline was quickly approaching.
Every time you fix a bug, you have hopes that the test programs will run perfectly after that. In other words, you hope that each bug you fix will be the last. Two days before my scheduled departure, a major problem showed up, requiring an extensive change. I figured out how to fix it, and we put in some temporary "haywired" connections to try it out.
It all worked!!! Now Jim and I had to write up the engineering change in a formal way and officially install it. The next night, Jim read off the wire changes to me as I wire-wrapped them in. At midnight, we powered the system up and ran our test programs. They ran!
With about five hours to spare! Jim and I laughed, shouted, and danced with each other. Gruff old Jim told me that he worked a lot of extra hours to help me out, because he knew of my deadline and how important this was to me. He didnít have to tell me. I could see it. Without saying so, we had become friends. The next morning, I reported to my bosses at Glendale that the new multiply-divide option worked, and Tom and I went to the airport.
I thought that I might be treated as somewhat of a hero upon my return to Glendale. After all, I had been largely responsible for the successful implementation of a much cheaper unit, and I had easily beaten all deadlines. No one said anything about it. I was a little disappointed, but not upset. After all, I was only doing my job.
Emphasis Change, into Testing, Production, Field Service
The emphasis in the department was beginning to change. Everyone had been frantically working to design and re-design things, then debug and test engineering models. The A-test machine had completed product testing long ago, and the lengthier test of the B machine was nearing completion. B-test included running the machine for long periods under conditions of high and low heat and humidity, high and low voltage, vibration, etc.
Also, the input of the reliability and serviceability people had to be tended to more actively now. After all, IBM customer engineers would have to maintain these machines under difficult conditions. We really had to clean up our documentation, so that it would be in good shape for the factory and later, the field.
One of my first jobs had been to produce a written description of how the multiply-divide hardware worked. A "professional" writer used my description and my assistance to write the service manuals for the field. The very top engineers were beginning to move on.
Fran Underwood was still attached to the project, but he was primarily brainstorming new potential features and uses for the 1401. He wasnít a manager anymore, and I now worked for Bill Murray. This was fine with me. Bill was a great guy. Hard worker, smart, great sense of humor, Irish-Catholic, loyal to his workers and colleagues, and simply a nice man! Earl Bloom, who also had a reputation as being one of the brighter and more creative people, was also involved with new ideas.
Most of the rest of us were segueing into the testing, production, and field service end of things. As the factory and field people learned their skills with the 1401, the Glendale engineers would be rotated to newer projects that were in the growth stages.
Noon Breaks are Serious! Bridge and Pinochle ;-)) The Camera
Despite the hard-working professionalism of all the people in our department, when noon came, work stopped. Thatís because card games started. Most of the middle managers and engineers played bridge. Some of the younger guys played pinochle. In the former group was Bill Murray, previously mentioned. Joe Conzola was a short Italian engineering manager. Al Miller was a bit older, maybe fifty, and he headed the input-output group. (Tom Schappe worked for him.) Ken Bell was a staff engineer, a nice guy, and really smart. John Gandour was a thirty-five year old associate engineer of Lebanese descent. He was one of the nicest, sweetest people I ever met, and we became good friends and confidants. Jud McCarthy was only a technician, but he was very bright, and very brash and outgoing. He is also a wonderful person who helped me out many times, and he is my friend to this day. Earl Bloom, a bright, young staff engineer, played. There must have been at least one more, because at exactly noon, two bridge games usually started. Maybe it was Byron Rucker.
If some were out of town or on vacation, there might only be one game. These games were intense, and the players were excellent. I learned the game from watching them. There were always several onlookers. They would each have a brown bag sandwich that they munched during shuffles. Most had a cigarette circulating from their lips to the gap between the index finger and middle of their right hand to the notch on their personal ashtray.
We had a one-hour lunch period, and the games wrapped up about then. But not always! If a rubber was on the line for one more hand, a game might continue until as late as 1:15. But no one complained about wasting company time. These fellows were important to the project and they gave much of their own time and energy to it. Iím sure the big bosses knew that, and also knew that the card games were a perfect mental release, and also helped foster a spirit of camaraderie. Thatís why I liked the place.
There was usually a pinochle game or two, in addition. Paul Farbanish and I were the main players, but Andy Kisacky, Schappe, Harvilchuck, Rucker, and others often played. These games were less intense, with more joking and laughing. It was a poor man's card game. After we finished our lunch, several of us usually smoked a cigar. Paul, Byron, and I were the main culprits. For several years, I smoked a cigar or two each day. (But I never inhaled. I chewed though.) After all I was an adult, and I had to show it.
We didnít play pinochle every day. Sometimes we would choose to go to the upper building for lunch in the cafeteria. This provided a threefold opportunity: we got a hot lunch; we could ogle the secretaries; and we could get some outdoor exercise.
In the fall of 1959 I bought a new 35mm camera, the first camera I ever owned. It cost $60, a huge sum in those days. In October, as we walked to lunch, I was awe-struck by the scene on the hillside behind the upper building. It looked like a rainbow, with orange, red, yellow, green, and brown colors. I had never seen an autumn like those produced in the Northeast. This hill was particularly beautiful, with its many maples. I ran to my car for my new camera, and snapped picture after picture.
I still have the slides. They remain beautiful, but now my perceptions are jaded by the annual autumns in New Hampshire. Thatís the advantage of youth. So many experiences are new and exciting. That is the main thing I miss about youth; even more than non-achy bones and young love! But I am thankful that I have always been sensitive to unusual beauty, historical places, and geological phenomena. My heart leaps, and my eyes water even now as I reminisce about that October day on the Glendale parking lot. Some October, I may return to view it, even though IBM is absent now.
Office Life, Poker Evenings
A lot of good-natured by-play occurred in the office. One noon-time, John Young, Jud McCarthy, Armand Gavazzi, and I were shooting the breeze about songs and singers. Then Young asked us each who our favorite male singer was. As we each replied, John would pooh-pooh our choice. When we had finished, we angrily asked the smirking Young who his favorite was. Without discussing it, I am sure we would all have dissed his choice. But John couldnít think of his name. He could picture the guy, and tried to describe him. He was handsome, dark-haired, a baritone, sang show-tunes, and was in the movies Oklahoma and Carousel.
We guessed people like Jimmy Durante, Eddie Cantor, Al Jolson, Kate Smith, and other ridiculous choices, just to pull Johnís chain. He kept getting more and more frustrated, and finally said that he would call his wife at home. She would certainly know. As he left, Armand, a technician from the factory with a wry sense of humor said, "When he comes back with the name letís play dumb. Pretend that we never heard of the guy." Soon, John, all smiles, returned, and said triumphantly, "I knew she would remember. Gordon MacRae!"
We looked at each other, dumbfounded. "Who?" "Never heard of him!" "Are you sure thatís his name?" "What was it again? Gordy McCreedy?" John couldnít believe that we had never heard of Gordon MacRae, but none of us even cracked a smile. John stomped off, completely frustrated. We quietly roared, if thatís possible. For at least a year after that, one or the other of us would ask him, "Are you sure his name was Gordy McCreedy?" John never suspected. He just thought we were ignorant. Jud and I still laugh about it forty years later.
Someone came up with the idea of having a regular poker game. I donít recall if we played every Friday night or once a month. Regulars included Farbanish, Conzola, Murray, McCarthy, Kisacky, Dunbar, Gandour, and me. We usually had about six players, and rotated from house to house, with the host providing the beer and pretzels. We just played nickel, dime, quarter, and the games ran from about eight until about one in the morning.
What made this special was the group of people. They were all (except maybe Conzola) really funny, especially after a few beers. Also, the mix of people included managers, engineers, technicians, and a clerk. The chit-chat would vary, but it would always get around to work at some point. Bill Murray, especially, would open up about what was happening with the top brass. On the other hand, our clerk, Ernie Dunbar, had rabbit ears, and he knew everything about everybody.
I got more insight as to what was happening in our department and lab from those card games than I ever could at work. It showed me the importance of social contact at country clubs et al if you really had high goals for yourself. I wasnít devious. It was just fun for me. Andy Kisacky had a slow talking, Hunkie accent, and a droll sense of humor. Ernie was super funny. Many times my side hurt from laughing. His wise cracks were best, but I still laugh about him starting to deal a hand and saying, "Kyesackys are wild!" I guess you had to be there.
After the 1401 was released, it was displayed at various trade shows. Some engineers would go along, partly to keep track of the machine and fix it if need be, but also as a reward for good work. The top dogs got to go to the European debut, at a Paris show. Bill Murray said that they were treated like kings by the local IBM officials, and spent evenings at the Folies Bergere and the Crazy Horse. Some of the beautiful, half-naked dancers were brought to their table. Bill said that they spoke to each other in French, thinking that they couldnít be understood. But one of the Americans spoke French and quietly provided some amusing translations for the other guys.
Called "on the carpet" :-?
At work, I was a little disappointed that not much was happening with our B-test machine, after the heavy lifting I had put into it before Christmas. I had heard that it was plodding along through B-test ok.
One day in March, I was told that I was wanted "on the carpet". (This was code to see the boss. IBM had special bennies, starting with private offices for low level managers, private offices with carpets for the next level, private offices with carpets and a potted plant for the next level, and private offices with carpet, plant, and outside window for the next level. After that came private secretaries, good-looking since the boss picked them, and office suites.)
Jim Ingram called me into his office and showed me a letter that was sent to him from the management at the factory. It stated that the wiring changes that had been added to the multiply-divide feature during its debugging, did not meet specifications, and they had to be removed and replaced. For the readerís information, all wire-wraps on production machines needed to meet a certain standard with regard to number of wraps around the pin, length of wire, route, neatness, etc. In fact, the technicians had to be trained and certified to wire wrap.
In the lab, on our test machines, we generally made our wires rather loose and sloppy, partly because we were always in a rush, partly because many of the changes installed during the design phase would have to be re-designed anyway, and partly because A-test machines didnít go to a customer. I explained all this to Jim (he knew it anyway), and told him that I couldnít care less if they had to spend two hours replacing a few wires. I had bigger problems at the time.
He never said anything, and didnít chastise me. He could tell that I was mad, and I think he got a kick out of it. I believe I detected him chuckling to himself as I exited his office.
As I went back to my desk, I became even more angry. Not only was I not congratulated for doing such a good job on the multiply-divide system, but people grumbled about it. I knew that this system was going to mean a huge pure profit for the company, since I had learned that the feature was priced to the customer based on the company cost of the old system, which was much more expensive. I felt that I was largely responsible for its success, since in its early, tenuous stages,
I could have put the kibosh on it when my advice was asked. In contrast to the older 1401 engineers, I defended the new approach, and was a key guy in the design and the one who made it work.
I write Angry Letter !!
Farbanishís desk was behind mine and he could see that I was upset, and asked why. I told him my sad story. He said, "Jeez, let them know about it. They probably have no idea what has happened and donít really care. Thatís old history now. You should write a letter to Bill Murray (my boss), and copy all the bosses up the line. Explain what happened and what you did."
I never liked to toot my own horn, and I had figured that everyone was as knowledgeable as I about what I had done. But maybe Paul was right. I remembered Franís advice about emulating Farbanish. I started outlining my letter.
I counted up all the printed circuit boards in both the old and the new versions. Ditto for the interconnecting cables. I then found the estimate of 1401ís that were going to be built (10,000-more than all the computers currently in the world) and those that would have the m/d option (8000). Since I knew that the customer price was the same, all I had to do was multiply the difference in hardware costs by the number of machines carrying the option, and the result was pure profit for IBM. It came to $7,000,000.
(Years later, I found that the whole corporate net profit that year was only about seventy million. In my mind, I was responsible for ten per cent of IBM's profits that year.) Now, even I was amazed at what I had done.
I wrote my letter, emphasizing these facts, and also adding that the resultant fewer cables reduced general cabling problems that the full system was running up against due to lack of space. In addition, the new system was smaller and simpler, hence would be more reliable, and easier for factory technicians and customer engineers to learn and service. The manuals would be far easier to write, and local offices would have to stock fewer parts.
I mentioned my initial support of the new approach, but decided not to mention the foot-dragging of the senior 1401 engineers at that time. Some of the big-wigs knew, and I didnít want to put anything into writing that would make enemies. I was also careful to credit Steve Bespalko.
Then I copied the letter to Steve Bespalko, Bill Murray (my immediate boss), Jim Ingram (Billís boss), Chuck Branscomb (Jimís boss), and, I think, B.O. (Bob) Evans (Chuckís boss). Bob was really a high-powered guy, who later ran the 360 computer project, probably the biggest of all time for IBM. I got it typed, and mailed it out before I lost my nerve.
I think a copy is in the attic someplace. It is pretty crude, but I think I had been fairly accurate. The next day, I got a glance from Bill Murray, as though he wished to tell me something. I knew he would be on my side, but maybe he wanted to tell me that the shit hit the fan and I was going to be fired tomorrow. I went back to my job, and heard nothing.
I get promoted to Associate Engineer, Early :-))
About a month later, Bill called me into his office. He had a huge smile on his face as he shook my hand. He said that I had just been promoted to Associate Engineer, and congratulated me. He then took me around and told the others. Now, this was a big deal. The usual wait for promotion from junior engineer to associate engineer was two years. This was less than a year.
I was the first in my hiring class of twenty-two to get promoted. All promotions were announced all over the whole lab weekly by posting the announcement on all bulletin boards. They always contained your picture and a brief biography. When I would run into any of the new engineers, they gave me a "how did he do it" look. My guess is that the memo, especially the profit figure, caught everyoneís attention, and Bill used that opening to push hard for my promotion.
I was lucky enough to start on a hot project, get assigned a hot job, and succeed at it. Certainly, a number of my colleagues were smarter than I. But I was smart enough to take Fran Underwoodís advice, and listen to Paul. I sincerely thanked Paul for giving me the advice and gumption. He laughed and was happy for me. He was a nice guy, and still is.
My promotion gave me a fat raise, and also made me a bone-fide engineer. It also meant that I was switched from hourly to weekly salary. In other words, I didnít get paid for overtime, which had been at time and a half. There was an exception to this. If a lot of extra work was expected, the manager could "authorize" you for so many hours of overtime per week. However this was generally quite a bit less than we had to work to get our job done. In fact, as the production line started, much help was needed at the factory.
But I didnít care: I liked my job, my salary, and my new prestige. Very few engineers have a first year like mine. Many never see a design of theirs reach the real world. The multiply/divide feature had two special lighted push buttons on the front panel of the computer. I considered them to be mine, and for many years, when I would see a 1401, whether it was in California, or Arizona, or Montana, or New Hampshire, or Times Square, I looked at my buttons with pride. I knew what I had done!
Tom Watson Jr. to show off a 1401, and it doesn't Multiply-Divide
It was sometime in 1960, I believe in the spring, that one of the first 1401ís was shipped to the IBM showpiece central office in the Time-Life building at Rockefeller Center in New York City. Tom Watson was scheduled to show it to the IBM Board of Directors.
Our department got a call from the customer engineers there that there was a bug in the system. The multiply-divide tests didnít run. Grif Rees was a field engineer in our office, about twenty-seven years old. Field engineers were generally college graduates who fixed faulty systems in customersí facilities. They usually worked out of the factory, and worked closely with the design engineers, as necessary. Customer engineers, on the other hand, were not degreed people, and worked out of the local IBM office to fix local problems.
If the customer had a problem, he called the local IBM office, which sent a customer engineer to fix it. If he got stuck, he called the manufacturing plant for advice. If phone help didnít suffice, a field engineer from the plant would be sent to the site.
Early in the history of a new system, the customer and field engineers were inexperienced with it. Also, there might still be subtle design bugs. Thus, a design engineer might have to be called. Again, if he couldnít fix it by phone, he would go to the site. In this case, the customer engineer did not have a clue, since he had never touched a multiply-divide unit. The heat was on, since this was Tuesday morning, and the big demo was scheduled for Wednesday morning. Grif was a smart guy, and a competent, experienced field engineer, but he knew next to nothing about multiply-divide. After all, it was brand new.
So he got permission from the bosses for me to accompany him to the City. I had mixed emotions about the trip. It was exciting to be involved in such a high pressure, prestigious event, but I knew it would be up to me to fix the machine. Grif was there only because I was completely green, never having taken such a trip before.
We rushed home and packed our bags, and departed on a Mohawk flight that afternoon. By 6 PM, we were standing at the faulty system, surrounded by a sweating customer engineer and a couple of worried looking local managers. The 1401 was in a large, luxurious showroom, with a huge plate glass window opening to Fifth Avenue. The customer engineer described the problem, which could have been caused by anything; he was so vague and unknowledgeable.
They wondered how long it would take to fix it, since they needed to run more tests and prep the system for the big day tomorrow. Grif smiled and replied confidently, "It shouldnít take long. Why donít you fellows go out for your dinner? We should be done in an hour or two." I almost passed out. Grif had no idea how long it would take, and I recalled working on multiply-divide bugs in the lab and factory for several days before I fixed them. When they left, Grif asked me if I wanted to get something to eat now, or wait until later. I let him know how I felt. He just laughed and told me not to worry. He had confidence in me. I wish I did!
We rolled up our sleeves, with Grif acting as my technician. Fifteen minutes later it was fixed. It was a "solid" bug; that is, it was not intermittent. I had fixed the same problem once before, and the factory had neglected to install the engineering change. I went from feeling like Woody Allen to Superman. We sat around with our feet up until the others returned. We coolly told them all was well, and asked when the demo was scheduled. It was at 10 AM, so we said we would stop by before nine in case there were any last minute problems. They were thrilled and thankful. Grif and I could just envision them telling Watson about our heroics as we all watched the demo, and Tom shaking our hands and inviting us to lunch.
We had dinner and hit the sack in a luxury hotel. The next morning, we had breakfast, and arrived at about 8:30. The guys were nervous, but everything was cool so far. We just stood near the machine and watched as we awaited Tom and the Board. About 9:55 AM, someone excitedly entered and said they were on their way and would arrive in a few minutes. The manager immediately said, "You fellows will have to leave now", and ushered us outside the room. We watched through the window as a dapper T.J.Watson pointed at the machine and walked around it with the Board members, probably discussing the modern features and the money it was sure to make for the company. I say probably, because I canít read lips. Iíll bet old Tom couldnít have fixed it if it failed!
Happy Ending. Software in English?, Touring New York, Ike
We toured the facility, since it was a centerpiece of IBM action. I remember seeing a programmer writing code for the 1401 in what looked like English. The only instruction codes I had ever seen or heard about were in machine language; that is, ones and zeros. They would be shorthanded to binary coded decimal alphanumerics. For example, A100200 meant add the field located at 100 in memory to the field located at 200 in memory. This guy wrote statements like SUM=PRINCIPAL+INTEREST. I asked him what he was doing, and he said that a program in the 1401 would change his English instructions into 1401 machine language instructions. I didn't believe him. I asked Grif, and he verified it. This was the first hint that I had of a compiler language like Fortran.
We had a few hours until our plane left so we walked around to find a place for lunch, and came upon a large crowd lining the street three deep. We found out that President Eisenhower was to make a speech at the United Nations, and his route from the airport was to pass us. We decided to wait and watch. As we waited, watching the crowd was great fun. As usual, every type of character was present; from executive to wino. As I looked up at the windows of the buildings, I recognized some faces above us. Among them were the singers Fabian and Frankie Avalon, and Shelly Berman the comedian. Frankie Avalon was acting like a big-headed jerk, and eventually, Berman started trying to be funny. He did a Julius Caesar imitation from the window. He was moderately humorous and got a few chuckles.
Then one New Yorker excitedly ran into the middle of the street, pointed at Berman, and yelled, "Hey look, its Jerry Lewis!" That brought the house down, much to Bermanís chagrin. I watched the newspapers that week and finally noticed that these fellows were appearing on the Perry Como show Saturday night. I guess we were standing in front of a rehearsal studio.
Finally the open limousines came, moving fast, with Ike sitting on the top of the back seat of the second, with his famous smile, waving to everyone as the people cheered wildly. He looked ruddy and healthy as usual. Ike exuded goodness and confidence as always, and everyone felt proud to be an American. This was the first and only time that I have laid eyes on a president. We returned to Endicott, our trip a success. It was important for me, as I learned how to handle myself on a business trip.
Automated Logic Diagrams, ALDs
During that summer of 1960, one of Harvilchuck's jobs was to route our engineering designs and changes through "design automation". The computer logic would first be sketched out in a "third-level" drawing, which would give the big picture, but no details as to packaging or cabling. After these designs had been agreed to, a "second-level" drawing would be made. It would include details as to which logic circuit would be implemented with which circuit board, and how the wires would connect to the rest of the machine.
This stage was less creative, but still might involve design changes to achieve more efficient packaging. Finally, these drawings would be broken down into sub-sections which would be drawn on special, standardized, large, grid-lined, "vellum" pages by lower level engineers or technicians, or even young, high school educated girls hired for this purpose. We had two of the latter in our department: Barbara Schaffer and Carol Bilski. We sometimes had other, part-time help at this task.
Particularly in the summers, high-school girls would be hired. Finally, these "first-level" vellums would be taken to the design automation department, where women would transcribe the vellums into punched cards. The punched cards would be fed to a computer program which would check for proper connections (e.g. are outputs going to inputs?), circuit loading, multiple use (by mistake) of the same circuit, and the like.
The results would be returned to the engineers, who would then make corrective changes to the vellums. They would then be sent back through design automation. When the design finally went through the process without errors, the program would be told to generate a "wire list". This list would efficiently connect all the circuits together, and generate a stack of punched cards for use in the Gardner-Denver wire-wrap machines.
This machine would interpret the cards to wrap wires on the pins of the chassis into which the circuit boards would be plugged. The machine would wrap a maximum of only two wires on a pin, to facilitate easy manual alteration for potential engineering changes. It would also route the wires in a neat, efficient manner.
Anyway, the interface between our department and design automation was Jim Harvilchuck, assisted by our clerk, Ernie Dunbar. I recall that one time there was some sort of a screw-up in the transfer process, causing a two day project delay at a critical time. Jim jumped on Ernie for it, and Ernie yelled back. It got hot, and eventually they had to be pulled apart. They would have definitely come to blows on the office floor.
Many people would think this was awful, but I thought it was great. Both guys had a personal pride, and were both dedicated to our project, and keeping the machine on schedule. Everyone had the same kind of pride and dedication, leading to a slight natural tension. This was great, as it meant that people really cared, they weren't only putting in their time. That made the project fun for me-and successful for IBM. I had seen the same spirit on successful athletic teams on which I played. You kept each other on your toes. A day later, Jim and Ernie shook hands.
1401 the Money Maker. Now produce them!!
The 1401 had become a big money maker as Fran Underwood had predicted. The factory was producing them in large numbers now. There were typically about twenty machines on the manufacturing line around the clock. Each had a tester who was debugging the machine as fast as possible.
These fellows did not have college degrees, but were IBM employees who had passed I.Q.tests which qualified them to take IBM courses on electronics, computers, and debugging. Of course, some had experience on older machines. All had classroom courses on the 1401, using manuals such as the ones I helped write.
Floating around (floaters) were experienced technicians, with either two year degrees or military technician experience. These people bailed out the testers when they got stuck. Most had worked at Glendale in the latter stages of development.
All were overseen by the line managers. Of course, when the line first started up, the testers were often getting stuck, since the machine was new to them. Since new development had slowed, the lower level Glendale engineers and technicians were able to spend much of their time helping get the manufacturing line to flow. I was one of those.
I was surprised to find that I liked it. Troubleshooting bugs is like diagnosing an illness or solving a mystery or a puzzle. You put all the clues together, do things to develop more clues (such as lower voltages to force faults or writing simple programs to loop on an error), make educated guesses, and try a fix-e.g. replace a board or shorted wire. I knew the central processing unit (CPU) very well now, and was the number one expert on multiply-divide.
I was also beginning to understand the card reader and punch, the printer, and even the magnetic tape units. It was quite an ego trip to tackle a problem that someone had been working on for many hours, and solve it in minutes. Multiply-divide was particularly obtuse, and I would amaze very confused and frustrated testers by looking at the failure, entering a simple multiply program from the front panel, picking a few combinations to multiply, checking the products, and then telling them which board to replace. I wouldn't even look at an oscilloscope.
The problem was usually a bad board, since the wiring was automated, and the design bugs were all gone. Although the boards were pre-tested, the tests weren't always accurate. When I fixed a problem that rapidly, the tester would be pleased, since he could move on. (His speed at completing a machine test was monitored.) However, some would feel very stupid and inadequate. I would take great pains to explain that I was only able to do it because of my lengthy experience on the feature, and they would learn it in time. I would also explain my methods to them. I liked these guys. They had pulled themselves up by the bootstraps. And I liked IBM because it gave them the chance to improve themselves.
Of course, not all problems were easy to solve. I spent many hours on some. Consistent problems generally weren't bad. Intermittent problems were headaches. Usually they would be caused by a transistor on the "edge" which would work correctly 99.999999% of the time. You would think the system was working, when the same failure would pop up again for some strange reason. These weren't fun for me, since their solution required more guesswork and less logic. I recall working with a tester of Chinese descent all night on an important machine. He grew increasingly nervous and I could see him losing confidence in me. Finally, he looked at me and said, "Well John, only one thing to do. Caw faw helrp!" I almost hit him. I was help.
Each 1401 system on the line was separated from the others by a fancy rope mounted on pedestals and surrounding the unit. The rope and pedestals were similar to those uses to line up people waiting at movies. Inside was a signboard with the name of the eventual owner: e.g. First National Bank of Chicago or Ford Motor Company. Regularly, there would be tours of the factory, and future machine owners would often search out their machine and proudly gawk at it. (Not many people had computers in those days.)
One time, a tourist noticed a tester cussing in frustration. The guy asked what was wrong. The tester blurted, "This God-damned machine is a lemon. It's one problem after another. It will never work right." Of course the tourist happened to be the future owner of that particular machine. The next day, all signs were removed from all machines. I don't know what happened to the tester, but all the technicians and testers couldn't stop laughing. (On the sly.)
Clumsy Carp and the 1403 Printer
Endicott was the home of Johnny Hart, creator of the B.C. comic strip. He drew the strip in his home. He used his friends and neighbors as models for his B.C. characters. One of them, Clumsy Carp, was patterned after a friend of John Gandour. The fellow was an IBM'er, a mechanical engineer who was the originator of the famed 1403 "chain printer".
This was a unique design, with three sets of type slugs of the alphanumeric characters stuck to a chain, similar to a bicycle chain, which continually rotated at high speed in a horizontal position a short distance from the print paper, with the inked ribbon sheet in between. When the moving character was lined up with the right position at which to print, an electrical signal fired a "hammer" which struck the ribbon, pressing it against the paper and printing that character.
It always amazed me that the print quality was so outstanding, since the type never stopped moving. The hammer struck quickly enough, I guess. The machine could print 600 (later 1000) lines of 132 high quality characters per minute. It was a key part of the 1401 system, since it was to be used to print such things as paychecks at high speed. The printer was a big moneymaker for IBM for years after the 1401 was obsolete, as it was attached to newer systems.
Clumsy Carp had worked on developing it in the lab, but when problems seemed to keep occurring, the company cut off development, so he took his work home and worked on it on his own time, in his basement.
Eventually, he was able to demonstrate a working model to management. I'm sure he was richly rewarded. One day, John Gandour told me about him and took me down to meet him. It was a thrill for me, although he was a down to earth, perfectly normal guy.
I asked him why Hart named his character Clumsy Carp, and he replied that he was a klutz, always dropping things. He couldn't have been too much of a klutz! I don't know if B.C. is still popular, but a big pro golf tournament in Endicott, the B.C. Open, is still played each year. Arnold Palmer and the other top stars played in it. Anyway, Carp is the only real cartoon character I have met.
World Trade and British Sterling units
The branch of IBM which handled its international operations was called World Trade. This group had offices and plants all over the world. For the 1401 to be used effectively in Great Britain, modifications had to be made to the hardware. The reason was that the 1401 was an accounting machine, and the arithmetic for handling sterling units (pounds and shillings) is different (and more complicated) than for dollars and cents.
The design modifications were to be made by a young English engineer, with the assistance of our group. I forget this fellow's name, but he was a brilliant guy, genteel and well mannered, a very handsome blonde of about twenty-six. He really picked things up quickly, redesigning the whole system.
He was assisted by two Scottish technicians who did the blueprints, packaging, design automation, and eventually, the debugging. These fellows had thick brogues, and were very small and skinny like most Scots. (The English had seen to that two hundred years earlier.)
The final sterling systems were to be manufactured in Sindlfingen, West Germany. Thus, there were two German manufacturing engineers assigned to help with paperwork and debugging, and to meet the deadlines for a smooth transition to manufacture. They were both named Hans, one being thin and wiry, the other being tall, robust, and outgoing. It was an interesting group, especially to someone who had limited horizons.
The first sterling machine was designed and built. After being tested, it was to be shipped to Lloyd's of London, amid much British publicity. But, alas, the machine (at the factory on North St. with the "normal" 1401's) started falling behind schedule. Panic was beginning, and international pressure was brought to bear. The English designer had returned to England, and the two Scots and two Germans were doing the debugging, but not very quickly.
They needed help, and the big-wigs, Chuck Branscom and Jim Ingram decided that that help was named Pokoski. I was surprised and flattered when they told me, but I also felt the pressure immediately. I couldn't fail to feel it; they put it right on my back. I immediately went to North St., and it took me about an hour to find out what the problem was.
I didn't ask; I observed. The little Scottish technicians had been in the British infantry during the war. They told me how they liked to tease the German engineers whom they were supposed to be working with. In particular, they would sing an old WWII ditty, to the tune of Colonel Bogey's March:
Hitler, has only got one ball; Guehring, has two, but they're too small. Himmler, is very similar; But Goebbels, he has none at all!
They basically thought of the two Germans as stuffy, pig-headed losers.
When I spoke with the Germans, it became clear that they detested the puny little Scots technicians, and looked down on them. In fact, being engineers, they refused to do any of the dirty work like probing with the oscilloscope, changing circuit boards, or making wiring changes. That was beneath them. They limited themselves to thinking and making big decisions.
I decided that there was only one way for the machine to meet its deadline. I would have to make all the decisions and pitch in with the dirty work. I would also work fifteen hour days. Actually, I thrived on this stuff. Moreover, I liked both the Scots and the Germans, and we got along well. I just kept between them. They all cooperated well with me. Of course, none of them worked beyond their scheduled times, and the Scots always broke for tea. They all seemed to be oblivious to the deadline.
We chugged along, gradually progressing. One night at about ten o'clock, I was crawling around under the machine when I heard a strange voice say, "Can I help in any way?" It was Chuck Branscomb, the lord high guru, in a sport shirt, checking things out. He said, "If I can hand you tools, probe, or get you a coffee, just let me know. I don't understand the machine, but I can be an extra pair of hands."
In that instant a young man of limited background learned what made America great. There were no pretensions, no class struggles, no aristocracy, no shift times. Just doing what it takes to get the job done. Of course my ego skyrocketed. I'm sure that was the main purpose of Chuck's visit. To let me know that he knew that my job was important!
When I became chairman of the Electrical and Computer Engineering department at UNH thirty years later I remembered Chuck's visit. I made it a point to visit those faculty who were working their butts off, especially after hours, encourage them, and offer to help.
We got the machine back on schedule. It met the deadline and was announced in London to a lot of hoopla. I felt satisfied, and learned a lot about people and pressure. And I made some good friends. The Hanses made it very clear that if I was ever in the vicinity of Sindlfingen, I should look them up. Fifteen years I traveled near there with my young family and thought about it. But we just didn't have the time. I'm sure the Hanses would have treated me like a king. Not only was I a good friend, but they were smart enough to know that I saved their asses, even if they would never admit it.
Have Oscilloscope, Will Travel ;-))
I had noticed that most of the engineers in my department hated traveling. They were just about all family men, and a bit older than me, so I could understand why. But they often needed to go to the hinterlands to fix some 1401 bug which had buffaloed the local IBM customer engineer. I told my boss Bill Murray, that I had noticed that, and that I liked to travel, so maybe he could use me if the need arose.
I had stuck my neck out a bit, since I was not as experienced as the others, and these trips were pressure packed. But Bill's face lit up, and he thanked me, saying that he just hadn't thought of me.
Not long after, I was sent down to Worcester, Massachusetts to fix an unusual problem on the card reader-punch. This was definitely not my strength, and I told Bill. He said that he understood, but Don MacLeod, who, along with Bill, was the expert in this area, was on vacation. Bill wanted to show that he was helping by sending me, and told me that if I got stuck, to call for help.
This trip changed my view of engineering design. Designers generally are reluctant to concern themselves with the servicing of the machine in the field. They are used to working on it in a lab with all kinds of resources and expertise available. When I arrived at the facility in Worcester, the customer engineer explained the problem and breathed a sigh of relief. The problem was all mine now.
I was introduced to the data processing manager of the company, who was a nice guy, and he made me feel at home. He told me where the restaurants were. He said that if any of the female key-punch operators appealed to me, he could arrange a date. There were about twenty. This offer really threw me for a loop and I couldn't tell if he would simply be an intermediary, or put the heat on with job pressure. Anyway I wasn't interested in taking advantage of it. Especially when I found out that the payroll checks were due to be run off in three days and the 1401 was needed for the job.
I went to work immediately and was disheartened to find that the bug was not only intermittent, but was in an optional feature which was very complex and tricky to work with. I worked and worked the next day, noticing the top manager periodically raising hell with the IBM salesman for the lemon he had sold him. The salesman seemed to handle the pressure well, and buffered me from it as best he could. I was getting nowhere despite repeated calls back to Endicott to get advice from Bill, and from Don, who had returned from vacation.
I had no success that day or that evening. After a fitful night, I reluctantly asked Don to come out. He was sympathetic, and flew in late that afternoon. I prayed that he would fix it, but not too quickly. My prayers were answered. It took him several hours to find a design bug which caused a close timing problem and the intermittent failure. Whew, I wasn't too embarrassed, especially after Don and Bill said that I shouldn't be. The funny thing was that Don told me that if he had known about the problem right away, he could have fixed it with little inconvenience. He was a New Hampshire native (in fact a mechanical engineering grad from UNH) and had been on vacation there. At the time I arrived in Worcester, he was driving through on his way back to Endicott!
Field Service is a different world -
That trip significantly changed my attitude toward design. Previously, I thought that the only critical thing was to keep hardware costs down, even at the expense of serviceability. After all, the systems I had worked with could be debugged in the lab with the original designers and the best test equipment at arm's reach. The lab was clean and convenient.
Real life locations weren't like that. Dirt, mice, bugs (The first "bug" was actually a dead moth shorting some wires. Grace Hopper, an early computerist, named it. She was later a navy admiral and one of my heroes because of her penchant for bending the rules to do the right thing.), pressure from customers, and lack of knowledge and equipment were the reality.
I now felt simpatico with the customer engineers, paid more respect to the reliability and serviceability people, and kept this experience in mind while designing products. Years later, I also emphasized this aspect while teaching design to students.
Early that fall, I was asked to go to Chicago to cover a 1401 which was being displayed at a Banking Show there, at the convention center. Tom Schappe also went to cover possible input-output system problems. I think that this was more a "reward" trip for us than anything else. However, it was a new machine, and they didn't want anything to go wrong at the show.
We were to simply be resources that could be called on if the local customer engineers got stuck on a problem. I liked Chicago from my previous visit at Motorola, and expected to have a good time. I did. The computer was set up at the show without problems, and no hitches occurred throughout. We simply hung around during the daytime show hours, and had the nights free. The IBMer running the exhibit told us to keep out of the limelight and not to talk to any potential customers. I guess he didn't want any dumb engineers telling the truth.
As I wondered around the show, looking at the various displays, I noticed that many of them had beautiful female models to catch your attention. They sure caught mine, but I noticed that when they saw my IBM badge, they would walk away, looking for a real customer. So, as soon as I left the IBM area, I removed my badge and had a lot more fun-until the girls, noticing no badge, asked what bank I was with. When I finally confessed, they would walk away.
There was a middle-aged IBMer from marketing at the show for reasons I don't recall. However, he was by himself, and after a while, Tom and I got to know him and we decided to go out to dinner together. Some of the locals tipped us off about a great seafood place called "Ireland's Oyster House". It was just the kind of place I love. It was old. An old mahogony bar. Old tablecloths. Old menus. Old, professional male waiters who knew exactly how to treat you, knew everything on the menu, were invisible unless you needed them, and who never made a mistake. And the seafood was great. Over the next twenty years, until it closed, I made a point of eating there every time I was in the Windy City.
While at dinner, our new friend mentioned that he had a key for the Playboy Club, but had never gone there, and suggested that we try it. For some strange reason, I thought that was a pretty good idea. We called a cab, and the middle aged cabbie struck up a conversation. Pretty soon, he was suggesting that instead of the Playboy Club, we go with him to visit some friends of his.
He said his friends were female, attractive, and friendly. We politely but firmly declined, but he wouldn't take no for an answer. Finally, he said that he would be hanging around outside the Club when we left to see if we were interested then. I wonder why he thought we might change our minds. We ignored him and forgot about it.
At that time, the Chicago Playboy Club was the first and only one in existence. It had been open less than a year. It was heavily publicized in the magazine (which was one of my favorites), and in the media. I still didn't quite know what to expect. We walked up to the door and rang the bell. When the Doorbunny answered, I almost passed out on the spot. I had never seen such a beautiful, sexy, scantily clad woman in my life. In fact, I immediately recognized her as a playmate of the month from a few months back.
As she ushered us in, I saw dozens of them. The Hatcheckbunny, the Poolbunny, and the many Waitressbunnies! We were ushered to a table where we could watch the show (as well as the bunnies). All drinks were $1.50, which was quite reasonable, considering that there was no cover charge. We had a couple of drinks and I had a ball ogling the girls and eavesdropping. I couldn't believe that these waitresses were the same girls whom I had admired as centerfolds for the past few years. I recognized almost all of them.
I seem to have had them memorized, for some reason. Unbelievably, as I listened to them talk to each other, I discovered that they were actually human. They weren't some mystical, magical fantasy. They discussed the tips, and what they were going to do that weekend. Just like real people! It was weird to see the bunnies walking past life-size illuminated color images of their nude centerfold portraits, yet, sound like, as Hefner always said, "the girl next door".
We left after about two hours, in a stupor. We were awakened from it by a cabbie honking his horn. Oh no! Our "buddy" pulled up and opened his door. Our older friend whispered to Tom and me to humor him, but we wouldn't buy what he was selling. At first he was selling a house full of beautiful young nurses. As he drove on, it became one average looking nurse in her thirties.
We could hardly hold our laughter as he made his hard sell. Finally, he pulled up to a house and flashed his lights. The gal came out and stuck her head in the window as the cabbie introduced her. She politely tried to entice us inside, either individually or as a group. We declined, and our buddy gave her five bucks for her time. The cabbie took us to our hotel, reprimanding that we don't know what we missed. I suspect it was his wife.
Office Christmas Party
That December, the Department had another Christmas party. Don MacLeod was a member of the Lions or some other service organization, and they had put on a faux fashion show at one of their parties. It was a takeoff on the weekly fashion show on local TV, the one that Barbara Kollar occasionally modeled for.
The joke was that the models were all cross-dressed men, and Don, as M.C., read a very humorous script. Don had access to the script and costumes, and suggested that we run a similar show at our party. He recruited several of us to act as models, and one night we had a dress rehearsal at his house. Despite Don's encouragement, none of us knew if we would screw up the courage to make our appearance in front of our colleagues and bosses.
I took Sylvia to the party. She was dressed in a bright red party dress, looking very striking. After dinner, Jud and I took a few stiff drinks, attempting to screw up our courage. We barely managed. But it was worth it. Everyone in the audience split their sides laughing. I modeled a tennis outfit, bouncing down the runway with my racket. I was also a very demure bride.
Jud modeled a maternity outfit and brought down the house as he kept eyeing Jim Ingram and glancing at his own pillowed belly. Finally, he sat in Jim's (his boss) lap and hugged him. I recall one of the funniest descriptive lines for a dress was that it was dyed "horse gray". Don played the elderly Jewish owner of the clothing store and M.C. and some other crossdresser played the old maid Jewish aunt who displayed her boutiques.
It was professionally done. I didn't know what the reaction would be when I returned to our table but it was all laughs. Also, just as I figured, all of the old married guys were flocking around Sylvia, talking to her and getting her drinks. I loved it.
Helping start 1401 manufacturing in San Jose
In the spring of 1961, IBM's new facility in San Jose, California was beginning to manufacture 1401's for western U.S. delivery. Of course, it was difficult for them to get up to speed, since they had absolutely no experience with the machine. In fact, their factory people were relatively inexperienced in general, since the plant was only a few years old.
The main product from San Jose had been the 305 RAMAC system. This was essentially the first system which used magnetic disk for storage. I don't recall exact numbers, but the disk drives were about as large as a desk, and could hold only a few megabytes. Nowadays, thousands of megabytes fit in a drive the size of your fist. However it was a big breakthrough at the time, since a disk didn't have to be wound or unwound to find the appropriate data as did a tape drive.
Naturally they ran into problems building and testing the 1401's and called for help. The Endicott bosses wondered who to send, and Bill Murray, remembering my inclination to travel, suggested me. He was reluctant to ask me since it implied a month at San Jose. However, I jumped at the chance. I had never been West at all, and I was aching to visit California.
I noted the timing of it and asked, in passing, if I could stop in St. Louis on the way back. That would be Easter weekend, and I hadn't seen mom and grandpa at Christmas. Bill said, "Sure!" and we began making arrangements. I also pumped all the local guys who had visited San Jose, to find out the lay of the land. I was edgy about the responsibility that I was taking on, but I figured that I could handle it. If not, I would have to swallow my pride and "caw for helrp".
My flight landed in San Francisco, I picked up my rental car for the month, and headed south, down the peninsula. My head was constantly spinning as I admired the scenery. Silicon Valley did not exist at the time. In fact IBM was the start of it, at its southern end. Thus, most of what is now concrete and office buildings was orchards and ranches. It was beautiful. I found my motel and checked in for a month.
The next morning, I found my way to my contact in the design office. It was Hal Eden, the young genius who was responsible for developing the magnetic disk drive. I was thrilled. This was to be my home base, and I was introduced around.
It is especially important to get to know the secretaries in a situation like this, as they know how to get things done. I was then taken to meet the manager of 1401 manufacturing. Boy was he relieved to see me. He simply took the world off his shoulders and placed it on mine. He spent an hour summarizing the problems he had been having, and then asked me what he should do. He said that he would do whatever I said.
It was obvious that I was the twenty-three year old guru. I said that I wanted to spend the rest of the shift wandering around the 1401 line, and I would get back to him tomorrow. He introduced me to his top technicians but I left them to their work and introduced myself to the testers as I wandered from machine to machine. The grapevine had already told them that I was the savior.
Several things quickly became obvious. Firstly, these guys were really inexperienced and scared. I knew that I could teach them a lot about troubleshooting this machine: especially key signals to synchronize the oscilloscope on to get a good look at the bug of the moment. This wasn't in their manuals or classes. I didn't know if this would be difficult or not. It depended on their brainpower.
The second problem was easier to fix. I noticed that some of the bugs that they were working on were design problems which had been solved with engineering changes back in Endicott. However, the changes hadn't yet been installed in San Jose. They were sitting in a pile, because the line was already behind schedule. Whoops! That evening, I thought long and hard about the problems and came to some decisions.
The next morning I told the manager that the best thing that I could do was to make myself a buddy to his testers and get down and dirty with as many of them as often as possible. That is, I would work on the line with them, not hide in his office. They would pick up tips as they looked over my shoulder.
Moreover, since they were running two shifts, I would split my time between them. I would come in at noon, and work until eight, giving about four hours to each. He thought this was a great idea, and thanked me for being willing to give up my evenings. Of course, I didn't mention that I was single, and these hours would give me a chance to cat around at night and sleep in the morning. In actuality, my hours were more flexible than that.
I worked much more than forty hours per week, and I simply made sure that I touched base with both shifts. If there was something that I really wanted to do, I could make some time for it.
The other suggestion I made was to immediately install all engineering changes and keep a chart of the current status of each machine. I was amazed that by the end of the day, a huge chart was set up in front of the manager's office to show each machine's status each day. It was updated each day to show progress. These two ideas seemed to help out a lot.
Another time I saw in the paper that a local airline ran cheap flights from San Jose to Reno. The planes left early Saturday morning, you got a free limo to the casinos, and the plane returned late Saturday night. I believe it cost $40. I had never seen real gambling, so naturally I bought a ticket for that Saturday.
Soon after, the line manager approached me with a serious problem they were having. At IBM, each time a new machine went into production from any plant, the first machine off the line had to go through a "C" test. The first San Jose built machine was currently in C test, and was having an unusual problem. Occasionally, for some unknown reason, all the tape drives would simultaneously begin to rewind. It happened very seldom, maybe once a day. They had been working on the problem for a week, and were stymied.
I looked at all the appropriate signals and they looked good. I replaced some circuit boards which were likely suspects. The technicians said they had already tried this and the failure reappeared. I could never see it fail. I would watch the key signal for hours at a time, hoping to see some strange variation, but it wouldn't fail. Then I would go back to the main line to help out there, when I would hear that the system failed again. This went on for several days. I was spending twelve hours a day on this machine alone, and getting nowhere. The locals knew that it was a tough problem, and now it had been put in my hands.
I called Bill Murray and the other experts in Endicott, but they had never seen such a problem. They offered suggestions, most of which I had already tried. For example, I was suspicious that the wire carrying the key signal was picking up crosstalk from a nearby wire, so I rewired it into another position. No dice! Friday came, and I wondered if I should go to Reno the next day.
Finally, I decided that I would have to cancel my trip and work on it all weekend. I was really disappointed. Plus, I forfeited $40. I couldn't fix it that weekend, and was beginning to run out of ideas. The rest of the unit had been pretty well tested, and soon this test would be holding up shipments. The pressure was tremendous, and I figured that I would "caw forl helrp" on Tuesday if I couldn't fix it sooner.
Monday morning, a casual friend of mine from Endicott showed up. His name was John Campbell, and he was a field engineer in his forties, who had been assigned to the 1401. I had gotten to know and like him at Glendale. He was in San Jose on some other business and stopped by to see the 1401 line. He asked me how things were going. I explained my problem, and said that any suggestions at all would be appreciated. He offered to look over the system. As we were watching, the system failed. I had been watching for a week, and this was the first time that I had seen it fail. My first thought was, "Maybe it will start to fail more often, so I can track it."
But just as that thought entered my mind, John cried out, "What was that sound?" I said that I didn't notice any sound. He then ran to the technicians running the tests, and said, "What was that snapping sound. It sounded like it came from above the machine." They said that it was the solenoid picking that controlled the temperature in the room. He asked if a temperature test was currently being run. They responded in the negative, so he said to turn off the temperature control. The tapes never intermittently rewound again. John had associated the noise with the failure. When he found out what had caused the noise, he figured that the electromagnetic field from the big relay picking somehow caused the circuit which caused rewind to fire. I guess he was right.
He had solved in an hour what I had worked on for a week, and others had worked on a week before that. I never would have found it. I learned a couple of lessons. Keep an open mind! Experience is valuable! (John had seen similar problems in the field.) Use all your senses! I hadn't even heard the sound. I was concentrating so hard on the problem that some of my senses were tuned out. I often used this as an example in my classes on problem solving in later years. I didn't see Reno until forty years later, when I visited Joe and Teryl at Beale AFB.
Late in my tenure at San Jose, another strange problem arose. Once in a while, while punching cards, extra holes would be punched in a card. It happened very seldom, maybe once every few hours, but that was not meeting spec. My heart sank. My weaknesses were the card punch and the printer. Fortunately, they seldom caused trouble. Intermittent bugs were the hardest to find, and this one was very intermittent. They had been working on it for quite a while.
It appeared that the extra punches followed legitimate punches. This indicated that the circuit which activated the punch mechanism wasn't being turned off all the time. I replaced that circuit board, to no avail. It was difficult to keep watching the signals on the oscilloscope for several hours at a time, because one blink might mean you had missed the failure. However, I was suspicious of the electromechanical circuit breaker which reset the punch drivers. On occasion, the signal looked flaky. I replaced the circuit breaker and the system stopped failing. I was ecstatic. However, I could see no physical flaw in the breaker. Several weeks after I returned to Glendale, I heard that the problem had turned up again, and the same "fix" had corrected it.
Finally they figured that the problem might be that the drier, dustier climate in San Jose was causing the circuit breaker contacts to be intermittent. They cleaned them periodically, and failures ceased. The final solution was to put two circuit breakers in parallel. One of them would "always" make good contact. The dust seemed to burn itself off the contacts periodically. I said "always" because, in theory, extra punches would occur if both failed simultaneously. The odds of that occurring are extremely small.
Eventually my month was up. I was a little disappointed. There was still more to see in California, and I liked it there. The facility was modern and beautiful. Moreover the work I was doing was interesting and I was a big wheel, calling my own shots. I had no big projects awaiting me in Glendale. The 1401 line in San Jose was moving pretty smoothly now, and the workers were more proficient. However, I could tell that they and the manager wanted me to stay longer. That would give them a cushion. However, I went to my secretary in the development lab and asked her to order my plane tickets, leaving on Friday for St. Louis, and then departing for Endicott two days later. She looked at me and said, "Are you leaving already?" I said, "Will you miss me?" She turned a deep crimson. I guess I could have had company on my San Francisco expedition and my other forays.
I visited mom and grandpa. I felt better after missing them at Christmas. A few weeks after arriving back at Endicott, I heard through the grapevine that the San Jose people put up a big fight to keep me there longer. However, Bill Murray was determined that I be able to spend Easter in St. Louis with my family, as he had promised, and fought them off. Bill was a wonderful guy. I didn't have the heart to tell him that I would have loved staying in California for a few more weeks and visiting St. Louis after Easter.
Bill was later transferred to Boca Raton, Florida along with many other Endicotters. Eventually Bill retired, spending most of his time on the golf course. He is dead now. He was a wonderful man and a good boss. Straight, honest, hardworking, and of strong character. Everyone liked and respected him.
Minneapolis Data Center
A few days later, my boss set up an assignment for me to go to Minneapolis to the IBM center there, to install a slew of engineering changes and to make sure the machine was in good running shape. This was a key machine, since it was the primary demo system for the whole upper midwest. It had been having some problems, and customers weren't able to test their programs.
There would be three technicians from the plant also going to handle the bulk of the dirty work of wire changes and to help with troubleshooting. It was to be about a three day job, and butted up to the Christmas season. I was pleased to go, and I set it up so that I would fly from Minneapolis to St. Louis for Christmas. I would be able to see mom and grandpa for a couple of days.
As usual, I was apprehensive about the trip. I would be 1500 miles from Endicott with the pressure of the machine on my shoulders. I did know the factory guys who were coming, and they were all competent and pleasant. That helped. I had to take all the engineering change documentation with me, and it was fairly substantial. However, I managed to pack it in my suitcase with my clothes.
I also thought it might be helpful if I had some support material about the 1401 along, in case I got stuck. I figured this stuff wouldn't be available in Minneynoplace. It was all manuals and paperwork and kind of heavy to pack in another suitcase. Besides, my second suitcase would have been too large for it.
I decided to pack the stuff in a cardboard box, and tag it with my name, checking it with my suitcase. I figured that if it got lost, it wouldn't be a catastrophe. So I dropped by a store to get a cardboard box. Unfortunately, there weren't any plain ones of the size I needed. There was a box with red Campbellís Soup cans and printing on its sides, however. I thought about it, and decided, "To hell with how it looks!" and took it. The stuff fit perfectly, but I did feel strange carrying it into the airport, and checking it with my baggage.
I felt like a hobo, except that I did check a legitimate bag, also. My Mohawk flight went from Broome County Airport to Buffalo, where I had to transfer to a Northwest flight to Minneapolis. The Mohawk flight was sparsely populated. (This was not unusual in those days. One time, on a visit to St. Louis, I was flying a leg from Cincinnati to St. Louis on a TWA Constellation with six passengers. The plane had four prop engines, a triple tail, and was probably the largest commercial plane at that time. It held about seventy people, I guess. There were three stewardesses and I played cards with them during the whole flight.)
Anyway as our Mohawk flight headed northwest, across the Finger Lakes, one of the stewardesses, who was about eighteen years old, and a knockout, wearing a little dress about two sizes too small for her, asked me if she could look out my window. She was from Ithaca, and wanted to spot her house. For some reason, I complied and she proceeded to kneel in the aisle seat with her hands on my window and everything else hanging in my face. She pointed out her house and the major points of interest about Cayuga Lake. Then she sat down to talk. She was based in Newark, New Jersey, where she shared an apartment with a couple of other stews.
I perked up and began to envision a fun weekend in the City sometime. Then she proceeded to tell me about her M.D. boyfriend in Buffalo. I quickly became downcast, and proceeded to hate this lecherous sugar daddy. After a while, she left to do her chores, and we landed in Buffalo. As I deplaned and passed her to say goodbye, she blurted out that she had broken up with the Buffalo physician. Before this sank in, I was off the plane, and had to look for my next flight. That gave me something to think about. For forty-five years!
Since the planes in those days only flew a couple of hundred miles per hour, I didn't arrive in Minneynoplace until after dark. At that time, you deplaned down a set of outdoor stairs which were rolled up to the plane's exit. I stepped out the door to a blast of frigid air. It must have been ten below zero, with forty mph winds.
I was blown off balance, and grabbed the metal handrail. My hand immediately froze to it. I was smart enough to restrain from pulling away, and rupturing my skin. I stood for a few seconds, allowing my body heat to do a thaw job. Finally, I gently wiggled my hand free.
I then followed the passengers across the tarmac to the terminal. The wind ripped and drove the hard snow pellets into my frostbitten face. Now I knew what God's country was like. It was like Siberia. I walked to the baggage claim and watched the luggage shoot out. Here came my Campbell's soup! But where was my suitcase? Oh no!
All the other passengers had departed, and all the luggage had been picked up when I finally decided to report my loss to Northwest. They assured me that they would get the bag to me the next day. Not only were all my clothes in the bag, but so was all the paperwork necessary for my job. I forlornly trudged to the taxi-stand with my soup box under my arm. The cabbie gave me a strange look as I said, "Radisson Hotel." We proceeded through the blizzard until we reached downtown.
The Radisson was not exactly a Motel 6. It was more like a Waldorf-Astoria. It was the place where Hubert Humphrey, and later Fritz Mondale, had their headquarters and parties as they appeared on TV on election eve after their presidential losses. I looked at my soup box, wondering how I could hide it. But I had to show some luggage as I checked in. The clerk looked down his nose at me as I told him of my reservation. He started to call a bellhop, and then reconsidered. "Do you have any other luggage?" I told him my story, but I'm not sure he believed me. I went to my room and removed my clothes, seedy from a long day's travel. Oh, well, I would have my suitcase in the morning. After all, this is God's country!
The next morning, I peeked out my door, but no suitcase. I phoned the desk, but no word. I called Northwest, and they assured me that they would have it by noon. I wasn't sure what to do, but the simplest solution was to shower, not brush, not shave, put on my seedy clothes, get breakfast, and go to the IBM Center. The boss there was obliging, but a little concerned. The machine was going to be busy that day, and we could not really start until that night or tomorrow anyway. The system would be all ours for two days after that, and they had a customer waiting to use it when we turned it back. The manager asked if he should call Northwest to apply some IBM pressure. I suggested that he wait until noon. The other fellows showed up, and we set our plans. We would run two twelve hour shifts, from 6 to 6, with two of us on each shift. They urged me to take the day shift since that was when the company interactions would occur. Noon came and no luggage. What a revoltiní development that was. I told the manager, and the guys and I went to lunch. We returned after lunch, and at about 1:30 PM, my suitcase arrived.
The manager smiled and said he had to put on some corporate pressure. I resented that he had to do that. What if I had been some poor little old lady? Why should she be treated any differently? Anyway, we were all relieved, and the timing was such that we were able to sort out the work and plan our strategy. I would wait and get them started, and then my partner and I would go to dinner, and get some rest.
(AUTHORíS NOTE: I have deleted (censored) some interesting but personal events of the next two days.)
When I arrived at the Foshay the next morning, the changes had all been installed, and the diagnostic programs were being run. This took a couple of hours. Everything ran great. The manager had shown up, and I turned the machine over to him, ahead of schedule. Mission accomplished! By early afternoon, I had checked out of the Radisson (with a suitcase, as well as a soup box), eaten lunch, and departed for the airport. I flew to Detroit, from whence I caught a plane to St. Louis. This was either Christmas Eve or the eve of Christmas Eve.
Winter and spring of 1962
Marriage - Introducing Jane
(AUTHORíS NOTE: During the winter and spring of 1962 I went through a very crazy courtship and began a long and happy marriage. That summer I was transferred to another project. However one last event that spring may exemplify the camaraderie of the 1401 gang.)
Jane was certainly much more laid back than I was. She didn't mind if I went hunting or out with the boys. When one of the managers, Al Miller, celebrated his 25th anniversary with IBM, there was a party for him at which everyone got drunk-especially Al. I didn't know Al very well (he was Tom's boss). I had heard that he was a funny guy, but I never saw it.
Somehow, about ten of us ended up at our railroad apartment at about 1 AM. I guess it was because I had beer in the refrigerator, and I invited them. I know the group included Jud, Tom, Paul, Bill Murray, and Al. When I am stone sober, and I see a bunch of drunks, I am usually turned off, especially if some are relatives (like the old family poker games).
Not Jane. We mustn't have been too quiet in this three room mansion, because she came out of the bedroom in her nitie, saw us, and started laughing. She quickly assimilated herself into the group, passing out beer and chips. Someone saw my gym bag on the floor and rummaged through it, pulling out my jock strap. Somehow, it got placed on Al's head as a crown symbolizing his IBM tenure. He wore it proudly the rest of the evening. Eventually, everyone got sick and went home. There were a lot of quiet headaches in the morning. Jane still laughs about it and remembers Al as "the guy wearing the jock strap".
THOSE WERE THE DAYS, MY FRIEND; I THOUGHT THEYíD NEVER END!
Excerpt of e-mail, tale about O M Scott from John, Dec 23, 2018
The first (and probably only) time I saw O M Scott was in late 1959 in the Glendale Lab in Endicott. I was sitting at my desk when Andy Kisacky passed me and mentioned that O M Scott, the division head, was walking down the hallway, apparently going to a meeting. I glanced up to see him and noticed an employee standing by the hall window admiring the first snowfall of the year, occurring outside. About ten minutes later Scott returned in the opposite direction and the guy was still looking out the window. Scott said in passing, "If I come by here again and you're still looking at that snow, you will soon be out in it".
The outside art in San Jose of the metal three hinged shaft which flopped in the breeze reminded me of the common lament of San Jose engineers who had been "called on the carpet". "I just got the shaft up to the second (or first or third) joint." (Was that still said during your tenure there Robert?)