The female roots of all computer science, vol 17: Barbara Liskov

“The Architect of Modern Algorithms” (Quanta) is a recently popular link among some computer nerds on Facebook (all of the sharers, when I last checked, identified as older white males):

Barbara Liskov pioneered the modern approach to writing code.

But by the late 1960s, advances in computing power had outpaced the abilities of programmers. Many computer scientists created programs without thought for design. They wrote long, incoherent algorithms riddled with “goto” statements — instructions for the machine to leap to a new part of the program if a certain condition is satisfied. Early coders relied on these statements to fix unforeseen consequences of their code, but they made programs hard to read, unpredictable and even dangerous.

When she was still a young professor at the Massachusetts Institute of Technology, she led the team that created the first programming language that did not rely on goto statements. The language, CLU (short for “cluster”), relied on an approach she invented — data abstraction — that organized code into modules. Every important programming language used today, including Java, C++ and C#, is a descendant of CLU.

Note that in the discredited male-authored history of computer nerdism, the modern programming language dates back at least to ALGOL 60, developed when Professor Liskov was 21 years old. The public war on goto was waged not by Liskov, but by the developers of ALGOL and Edsger W. Dijkstra, a Dutch curmudgeon, who wrote “Go To Statement Considered Harmful” in 1968, pointing out that “the remark about the undesirability of the go to statement is far from new” and describes some of the history since at least 1959 (criticism by by Heinz Zemanek). Note that Dijkstra is also known for saying “The use of COBOL cripples the mind; its teaching should, therefore, be regarded as a criminal offense.” Liskov, for her part, was known at MIT primarily for developing and teaching the standard software engineering class, 6.170, in which the CLU language was used by students. She was a usually modest and always hard-working person who believed that properly engineered software could function perfectly: “If you find a bug in your code, you should be as embarrassed as if you found a cockroach in your kitchen,” she memorably noted (we had a lot of cockroaches in our East Campus dorm and they were regularly visible during visits to restaurants in Central Square at the time!).

[The article also notes that Liskov is concerned about the impact of the Internet:

I’m worried about the divorced couple in which the husband publishes slander about the wife, including information about where she lives. There is terrible stuff going on.

Yet if one of these two sued the other, the most common precursor to their divorced status, the lawsuit, and anything said by a party during it, as well as the mailing address where the plaintiff wants the checks sent, was already public information, available to anyone who wanted to go down to the courthouse, decades before women developed microprocessors and TCP/IP. (see this study on Massachusetts court records, though records of litigation following out-of-wedlock sex are sealed) Reporters were covering divorce litigation in newspaper stories prior to the computer age, e.g., a November 11, 1939 piece in the NYT describing an allegation of “cruelty”, and one from December 2, 1934, “a charge of extreme cruelty won a divorce here today for Mrs. Edith Crocker Sanger from Prentice Sanger of New York City.” Divorce was apparently a good business even in the Depression. From September 24, 1931: “More than $1,000,000 was handed to Mrs. Eunice Essig Brach of Winnetka today with a divorce from her husband, Frank V. Brach, president of a candy company.” Certainly someone launching a divorce lawsuit and obtaining a profitable judgment in 2019 gets a lot less publicity than he or she would have prior to the Internet.]

Readers: What will the next edition in the “female roots of all computer science” saga be? What other fundamental technologies can be plausibly attributed to a person who identified as a “woman”? My vote: find a woman to replace William Shockley as developer of the semiconductor transistor and Silicon Valley. How can it be done? Here’s a National Public Radio story that credits Hedy Lamarr with having invented frequency hopping. Wikipedia contradicts this story to some extent and the actual patent to Ms. Lamarr and George Anthell reveals that they narrowly claimed a specific piano roll-style mechanism for controlling frequency hopping, not the broad invention of frequency hopping. So we need to find an early patent on a specific application of semiconductor transistors in which one of the inventors has a female-sounding name. Then we can discover the female roots of the modern transistor and rely on the fact that reporters won’t read the patent claims to see that they narrowly cover an application of transistors, not the transistor itself.

Also, will this article on Barbara Liskov and the promotion of the article by “allies” have the desired effect of getting more people who identify as “women” into computer nerdism? The article reveals that Barbara Liskov, despite having invented essentially all of practical programming technology, was not popularly recognized until she reached the age of 80. Moreover, she describes having to struggle as a result of her identification as a “woman” (see also a 2008 interview, in which she notes that “there were a large percentage of women” at her first programming job at MITRE in the early 1960s, at which she learned FORTRAN (several years after inventing ALGOL?) and then got a PhD working with John McCarthy, credited for now at least with the development of Lisp, and then met Dijkstra in 1969 (giving him the idea to write his 1968 screed against goto?)). Compare to Britney Spears, a top-of-the-charts success at age 17 who has never described being a cisgender female as a career handicap in her industry. Why wouldn’t a cisgender female aware of both Liskov and Spears conclude that computer science should be a last resort?


9 thoughts on “The female roots of all computer science, vol 17: Barbara Liskov

  1. Hedy Lamarr’s frequency hopping idea relied on a primitive electromechanical piano roll type mechanism to change the frequencies and was completely impractical (if the piano rolls on the receiving and sending end got out of synch as was inevitable then you’d lose communications) and was never used by the Navy. She is the inventor of modern spread spectrum radio systems in the same way that Icarus is the inventor of the airplane.

  2. > What will the next edition in the “female roots of all computer science” saga be?

    Titania McGrath.

  3. That “our tribe invented everything” saga is not at all different from Soviet propaganda which (for example) claimed that steam engine and incadescent light bulb were invented by Russians. Jaded Soviet citizenry did take that crap with a lot of derision and sarcastic humor.* Americans didn’t develop immunity to Bolshevik propaganda yet.

    * For example: At one May Day parade some students dragged a humongous paper effigy of a microchip stenciled with “Soviet VLSIs are the biggest VLSIs in the world!”

  4. The article is obviously written by someone who does not have a very firm grip on CS or it’s history. Supposing that they (Susan D’Agostino) are an arbiter of what CS history is not only childish, it is also silly. As for Liskov, your tone notwithstanding, she is very much a pioneer of our field. Or do you think that her Turing award was also given through misattribution?

  5. Anonymous: Certainly I would not want to argue that any of the Turing Award winners are undeserving. However, shows that quite a few have been recognized in the area of programming language design and/or algorithms, e.g., Alan Perlis, Edsger W. Dijkstra, Donald E. Knuth, John Backus, Robert W. Floyd, Kenneth E. Iverson (who can live without APL?), Tony Hoare, Dennis M. Ritchie, Niklaus Wirth, Robin Milner (ML! Something that actually could reduce bugs), Ole-Johan Dahl, Kristen Nygaard (sounds female, so maybe a legend could be built around Professor Nygaard), Alan Kay (Smalltalk from Mr. Bigtalk), and Peter Naur (ALGOL again). MIT students forced to use CLU in 6.171 certainly did not say “CLU is so much better than any of the languages we could get paid to use. Professor Liskov towers over all of these other folks and should be recognized as the true pioneer.”

    As a thought experiment, imagine that Barbara Liskov, Russ Atkinson, Alan Snyder, and Craig Schaffert had all gone to medical school and become radiologists. Busy diagnosing patients, therefore, they never built CLU. How would today’s popular programming languages, e.g., Java, Visual Basic, PHP, Python, and C++, be different?

  6. You missed an obvious “Trans Sister” pun that would’ve allowed you to ride another of your hobbyhorses. Not sure Lynn Conway can claim credit for an invention as fundamental as that of Shockley et al., though.
    A while back, I became obsessed with the BBC dramedy about the British microcomputer revolution, Micro Men (2009). I couldn’t work out why they portrayed the original BBC Micro designers as very stereotypical white male nerds, when one of their number was the 20-year-old Sophie Wilson. (Think of all the movies with implausibly young, beautiful women playing world experts on nuclear weapons, etc.) Then I discovered that in the early 80s, Sophie Wilson presented as Roger.

Comments are closed.