Louis Proyect: The Unrepentant Marxist

June 12, 2017

An uncommon life in Common Business-Oriented Language (COBOL)

Filed under: computers — louisproyect @ 6:25 pm

In many ways, the best thing about the NY Times is the obituary since it amounts to a small-scale biography. If given a choice between a documentary and a narrative film, I generally lean toward the documentary because the real lives of people are far more interesting than what a screenwriter can think up. The same thing is true when it comes to a biography versus a novel. Why would I want to read something written by Jonathan Franzen when my time could be better spent on a biography of Ho Chi Minh or John Brown, just two that are sitting on my bookshelf right now?

On June 4th, the obituary for Jean Sammet appeared. Although I am pretty familiar with the lives of people who took part in the information revolution, I had never heard of her before. She was one of the six people who got together in 1959 to write the Common Business-Oriented Language, more familiarly known as COBOL. This was a language I used from 1970 until 1995 or so when I switched over to a Unix platform at Columbia University developing client-server systems in perl and java.

Jean Sammet, who brought computing into the business mainstream, at the University of Maryland in 1979 to deliver a lecture. Credit: Ben Shneiderman

From 1970 to 1978, I used my COBOL skills to facilitate moves from one city to another during my time as an SWP member. In those years, being a qualified COBOL programmer could usually land you a job within a week after moving to a new city. Furthermore, it enabled you to change jobs every 2 years or so with a 10 percent salary increase. And most importantly, it allowed you to exist in the corporate world without having to become part of the machine. Even after leaving the SWP, my computer skills continued to pay off, even at a place like Goldman-Sachs. The job served my ends just as much as it did my employer. This might have been obvious as indicated by a Newsday article about the Nicaragua solidarity movement in New York:

Lou Proyect works in a Wall Street investment bank, one of 25 “database administrators” who sit in a numbing row of fluorescent-blanched cubicles and stares at computers until the end of the day. It is the latest variation on the kind of job he has held for 19 years. Tacked to the wall of his cubicle is the latest article cut out from PC Week, a personal computer trade magazine: “IBM’s PS/2s aren’t all that revolutionary.” Neither, he says, is Lou Proyect.

It was in another computer magazine that he learned of the shortage of computer programmers in Nicaragua, because so many of the skilled middle class were leaving:

“This neon sign kept on in my head: ‘Nicaragua Needs You.’ This was using skills I had always taken for granted.”

He went to conduct a two-week workshop in computer programing as part of TecNica, a national organization which in the last two years has sent to Nicaragua about 50 New Yorkers, mostly computer programmers, some engineers, and one typesetter, one medical lab technician, one boiler mechanic, one travel agent. “Most are not hot radicals,” Proyect says. “They’re people very much like Ben Linder, taken up with the idea of helping the poor.”

I would go so far as to say that maybe half the people who went on TecNica delegations had more in common with Jean Sammet than they had with me. Born in 1928, she graduated from Mount Holyoke with a mathematics degree. Enrolled as a math grad student at the U. of Illinois in 1949, she ran into her first computer that revolted her. She said, “I thought of a computer as some obscene piece of hardware that I wanted nothing to do with.”

It was only when she ran some punch cards through a computer that she was transformed. “To my utter astonishment. I loved it.”

This exactly how I reacted when I started off as a programmer trainee at Metropolitan Life in 1968 using a COBOL-like language developed in-house called English Language. When I discovered that testing software was like doing puzzles, I couldn’t believe I was going to get paid to have fun. Here in a nutshell is what a COBOL programmer does. The code has been simplified but not by much.

Open input employee_file.
Open output check_file.
01 employee_record.
03 Employee_ID  picture x(10).
03 First_Name  picture x(30).
03 Last_Name  picture x(30).
03 Wage_amount  picture 9(5)v99.
03 Hours_worked picture 999.
01 check_record.
03 Pay_to_first_name  picture x(30).
03 Pay_to_last_name  picture x(30).
03 check_amount  pic zzzzz.zz.

Read employee_file into employee_record at end go to End_job.
Move First_Name to Pay_to_first_name.
Move Last_Name to Pay_to_last_name.
Compute check_amout = Wage_amount * Hours_worked.
Write check_record.
Go to P1.

Close employee_file.
Close check_file.
Now all of this might seem quite mundane. The program reads through a file, calculates the wage and then writes a check. Except the program would not work. Take a minute to see why. Done? It won’t work because I spelled “amout” rather than “amount” under P1. You might understand what I meant but not a computer!

American corporations have been running payroll applications like this since the 1960s but in a place like Nicaragua most businesses, many of which were owned by Somoza before 1979, did not have computers. Someone had to sit down with a calculator and do all of this manually, including the signature on the check. In the TecNica video, one volunteer recounts how the introduction of a modest computer cut the time for a state-owned enterprise dramatically. My time spent in Nicaragua convinced me that automation made socialism possible for the first time in human history, something that cybernetics expert Stafford Beer hoped would help transform Chile. It was only Nixon and Reagan’s intervention that showed how difficult it was to build socialism, even with the best of intentions by the government and having leading-edge technology at its disposal. A counter-revolutionary army supported by the most powerful capitalist nation in history was capable of stopping even the most determined movements for change.

The obit described the cultural environment for professional programmers in the early 50s:

In the early 1950s, the computer industry was in its infancy, with no settled culture or rigid career paths. Lois Haibt, a contemporary of Ms. Sammet’s at IBM, where Ms. Sammet worked for nearly three decades, observed, “They took anyone who seemed to have an aptitude for problem-solving skills — bridge players, chess players, even women.”

While by the time I entered the field it could no longer be described as in its “infancy”, it was nothing like today when most programmers have little interest except in making big money. Goldman-Sachs is now looking for computer science graduates who more likely than not have never read a single novel in their life except “Atlas Shrugged”. In 1970, computer science was barely getting off the ground. Most programmers I worked with back then were people who fell into it like me. With a liberal arts degree, it was very tempting to take a job as a programmer trainee that expected very little from you except to be competent in your trade (it could hardly be called a profession.)

When I went to work for the First National Bank of Boston in 1970, some of my co-workers had been affected by the student movement to some degree or another. There was a guy who had just graduated Dartmouth who had very poor work habits and spent most of the day talking about the Grateful Dead to anybody who listened. I sat next to a guy named Richard who worked there as a consultant. He was very knowledgeable about the arts and politics and someone I spent much of the day wasting time with chatting about socialism, 12-tone music and Godard films. Another consultant was a Harvard graduate who was about 5 years older than me and skeptical about radical politics to say the least. I dragged him once to see Camejo speak and he summed him up as “too febrile”. This was around the time I began to realize that not everybody was open to socialism.

I never met the guy at the bank who probably had more guts than me. There was a whiteboard in the cafeteria that was used for design sessions but someone had the brilliant idea to write something like this on it when nobody was looking: “The capitalist system is destroying the United States while it is killing countless Vietnamese peasants. Now is the time to demonstrate your opposition to such a monstrous system.” And beneath it in capital letters and underlined, you could read: “DO NOT ERASE”. People working for banks are so used to authority that the agitprop stood up until late in the morning when security guards got the okay from upper management to erase it.

From Boston I went down to Houston, Texas in 1973 and went to work for Texas Commerce Bank. I reported to Billy Penrod, a guy who looked and talked like a cowboy and who was a former Texas A&M running back from Gonzalez, Texas. Like most people in Gonzalez, Billy  was a racist. He once described Gonzalez as a sundown town, even though he didn’t use that term. He put it this way, “Colored people understood that they shouldn’t get caught in Gonzalez after dark.” Despite his retrograde views, I learned to admire Billy as the consummate systems analyst. We were developing a personal trust system that kept track of the estates owned by the oil millionaires. Billy was from Jean Sammet’s generation and got started out wiring IBM tab machines, used for accounting systems before there were computers. He got so good at managing them that he went to work as an IBM consultant implementing tab-machine based accounting systems around the country.

In 1975, I moved back to New York to work on automating the SWP headquarters, including the Militant newspaper and Pathfinder. This was the first time I began to suspect that I had joined a cult. The in-grown, zombie-like atmosphere at West Street made me feel ten times more alienated than I ever felt in a bank.

I went to work for Salomon Brothers during the day while doing West Street systems development by night. I didn’t stay long at Salomon but long enough to work with Michael Bloomberg who had me and a business analyst automating SBIL, their branch office in London. I rather liked Bloomberg even though he was an even bigger skunk than Billy. He was a sexist and racist pig who once yelled out “Look at the tits on that broad” when a Latina was delivering coffee on the trading floor.

When I got a mediocre review at Salomon, I went out and found another job in a week with ACI (Automated Concepts, Inc.), one of many “job shops” that hired programmers for very good salaries in the 1970s and 80s before they went to work for themselves as a contract programmer making even bigger money. One of my last jobs was as a self-employed contract programmer making $500 per day. This was in 1989 just when the job market was tightening up irrevocably and when those asshole libertarian smart-ass computer science majors were taking over.

ACI was a fun place to work. I got a big kick out of the CEO Fred Harris who was into EST, a self-improvement cult not nearly so bad as Scientology but pretty bad. Fred didn’t quite know what to make of me but he appreciated the fact that I had a rather “elevated” mind as well as being a crackerjack COBOL programmer. I have no idea whatever happened to ACI and Fred Harris but I used to dream about going up to their offices on 386 Park Avenue South getting my next assignment.

One of my last consulting assignments with ACI introduced me to Gabriel Manfugas, the son of a former Batista soldier who had fled to the USA in the early 60s. Gabriel and I became fast friends even though he had no use for my politics. We used to smoke pot, even during lunch, as we walked around mid-town. I got to know his friends, who were upwardly mobile Latinos from Washington Heights and programmers like him. By this time, programmers—including them—were computer science graduates but nobody could mistake them from the jerks I used to run into at Goldman-Sachs. Mostly, they were looking for angles to make them wealthy like starting their own consulting company. I was 20 years older than all of them, who saw me as a father figure—subversive politics and all. I enjoyed plenty of cocaine binges with them in the 1980s and have fond memories of all of them who are now in senior management positions at various corporations.



  1. The NY Times obit for Jean Sammet is a reminder that in the early days of computing, the programming world was very much a woman’s world. Most of the early programmers were women, like Sammet and Admiral Grace Hopper, under whom Sammet had worked when developing COBOL. That was because during the Second World War, defense labs around the country hire young women who were either math or science graduates to work as computers. In those days, a computer was a human being whose job it was to crank out complex calculations for things like artillery tables. That sort of work required people who were proficient in math and who could put up with the tedium of doing multiple calculations a day. Since most of the male graduates were in the service at that time, young women were hired to do this work. When the first electronic digital computers were built, some of these young women were redeployed to write programs for the newfangled machines. So most of the early programmers were women. Nowadays, the tech world is largely a male dominated world. But that was not always the case.

    Comment by Jim Farmelant — June 12, 2017 @ 6:44 pm

  2. Hey naber süper bi yazı olmuş.

    Comment by Profesyonel tasima ekibi istanbul — June 12, 2017 @ 6:58 pm

  3. Thanks for sharing this, Louis. I got my CS degree in 2002, but I agree the “asshole libertarian smart-ass computer science majors” are still ruining the scene. The industry is still extremely unregulated and rewards a lot of unhealthy work habits and aggressive office culture. Startup culture is used where it’s needed, since it carries social cachet, even when the company is not a startup. Even startups are becoming more corporatized and mechanized by VC firms. For the last decade, a strange trend I’ve seen is programmers imposing project management processes on themselves based on ideas culled from business books. Some of them are based on Toyota manufacturing systems. I guess who needs a Fred Taylor when the workers do it to themselves?

    The wages are still above average, but the job titles are so specific about your experience, it’s hard to switch places often. You definitely lived in the golden years.

    Comment by Aaron — June 13, 2017 @ 2:12 pm

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: