Inventing language

Nov 10, 2009 by Larry Hardesty
Institute Professor and Associate Provost for Faculty Equity Barbara Liskov. Photo: Donna Coveney

(PhysOrg.com) -- Last Thursday, the day after the New York Yankees won their first World Series of the 21st century, MIT Institute Professor Barbara Liskov, the 2008 recipient of the Turing Award — frequently called the Nobel Prize for computer science — delivered the first lecture of the 2009 Dertouzos Lecture Series.

John Guttag, the Dugald C. Jackson Professor in the Department for and , introduced Liskov, and he greeted the crowd of more than 150 people wearing a Yankees cap. The cap, he claimed, had a purpose other than pouring salt in the wounds of disappointed Boston Red Sox fans: His PowerPoint presentation was titled “Barbara Liskov: the Derek Jeter of Computer Science” — a reference to the Yankees’ star shortstop that drew hisses from the audience. Guttag’s presentation also featured photos of Liskov dressed in renaissance clothing and putting a tray of cookies in the oven — “moonlighting as a baker to put her son through Harvard,” Guttag said, “which was all the more poignant, since he could have gone to MIT for free.”

Guttag waxed serious long enough to describe Liskov as one of his own most valued mentors, and then Liskov took the podium. She opened with a joke of her own: In large part, her Turing Award honored work she had done in the 1970s, laying down principles for the organization of programming languages that today are almost universally followed. After the award was announced, she said, her husband spent a lot of time on the computer Googling the reaction, and at some point, “he came upon a quote from someone who said, ‘What did she get this award for? Everyone knows this, anyway.’”

After that, Liskov was all business. The thrust of her talk was that, in the 1970s, it was emphatically not the case that “everyone knew this,” and she described in great detail the intellectual environment in which she did her pioneering work. The talk was not for the uninitiated: she began by describing several papers from the early 1970s from which she had drawn inspiration — with titles like “Go To Statement Considered Harmful” and “Information Distribution Aspects of Design Methodology” — but first conceded to the audience that while “many of you have read them, I’m not sure all of you have.”

Liskov explained that, in the fall of 1972, after reviewing the literature in the field, she came up with the idea for what today are called abstract data types. Traditionally, a computer program would be a long list of exhaustively detailed instructions, and anyone reading the code — including the original programmer — could easily get lost. Abstract data types are, effectively, repositories for the computational details of the program, which let the programmer concentrate on the big picture. A complicated program turns into some rather simple interactions between the abstract data types. And indeed, the programmer can later change the details of the data types’ instantiation — how they do their low-level computations — without changing the overall structure of the program.

Liskov explained how, after coming up with the idea of abstract data types, she and some collaborators created a programming language, CLU, which put some of her ideas into practice. The rest of her talk was largely a demonstration that CLU prefigured most of the ideas that are commonplace in today’s programming languages — ideas with names like polymorphism, type hierarchy, and exception handling.

During the question-and-answer session that followed the talk, Liskov was asked the secret of her success. Part of her answer — which must have chagrined some members of the audience — was that “I don’t work that many hours a day.” “I always went home at night, and didn’t work in the evening,” she said. “I always found that downtime to be really useful.” She also, however, emphasized the importance of pursuing research that’s interesting — rather than, say, the research that will generate the most publications. That way, she said, “at the end, if you fail, at least you did something interesting, rather than doing something boring and also failing.” After the laughter died down, she added, “Or doing something boring and then forgetting how to do something interesting.”

Provided by Massachusetts Institute of Technology (news : web)

Explore further: Off-world manufacturing is a go with space printer

add to favorites email to friend print save as pdf

Related Stories

Epilepsy breakthrough on horizon

Aug 31, 2006

Researchers at MIT are developing a device that could detect and prevent epileptic seizures before they become debilitating.

Goodbye to faulty software?

Jul 15, 2008

Will it ever be possible to buy software guaranteed to be free from bugs? A team of European researchers think so. Their work on the mathematical foundations of programming could one day revolutionise the ...

'Last Lecture' prof's program to be updated

Jul 31, 2009

(AP) -- Carnegie Mellon University will release an updated version of the animation-based software developed by late "last lecture" professor Randy Pausch to teach computer programming.

Recommended for you

Off-world manufacturing is a go with space printer

4 hours ago

On Friday, the BBC reported on a NASA email exchange with a space station which involved astronauts on the International Space Station using their 3-D printer to make a wrench from instructions sent up in ...

First drone in Nevada test program crashes in demo

21 hours ago

A drone testing program in Nevada is off to a bumpy start after the first unmanned aircraft authorized to fly without Federal Aviation Administration supervision crashed during a ceremony in Boulder City.

Fully automated: Thousands of blood samples every hour

Dec 19, 2014

Siemens is supplying automation technology for the longest and one of the most cutting-edge sample processing lines in any clinical laboratory. The line, or automation track, 200 meters long, in Marlborough, ...

Explainer: What is 4-D printing?

Dec 19, 2014

Additive manufacturing – or 3D printing – is 30 years old this year. Today, it's found not just in industry but in households, as the price of 3D printers has fallen below US$1,000. Knowing you can p ...

User comments : 0

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.