High-tech industry considered harmful: a little-heard of Dijkstra paper

(This article also appears in The Farm, but I gave it a more clever title here.)

[ via java.blogs via

Brian’s Waste of

Time ] The late great computer scientist Edsger Dijkstra’s

contributions to the field are many. Among the best known are his

solution to the Shortest

Path Problem, his invention of the Dining

Philosophers Problem and semaphores,

and his classic paper, Go To Statement Considered

Harmful.

As a computer science student, I’ve read many of his papers, but here’s

a little-known one that he wrote in 1995 — by hand,

no less — called Why American Computing Science Seems

Incurable. It’s

available as a PDF file contaning scans of handwritten sheets. Here’s a

sample:

In the essay, Dijkstra argues that the pressures that the high-tech

industry is adversely affecting academic research. He says that

industry pressure is causing the definition of being a good programmer

to change from someone who is “able to design more effective and

trustworthy programs” and who knows “how to do it efficiently” to

somewho who thinks of “‘industrial acceptance’ as quality criterion”

and writes programs such that “its main feature [is] that one could

apply it unthinkingly.” Programming, he says is becoming less a branch

of applied math and more a branch of keeping the high-tech industry

afloat, a problem aggravated by “a total lack of faith in [America’s]

educational system and a deep-rooted mistrust of

intellectuals.”

It’s a fascinating read, and it just might send you running to the

bookstore to buy all those Knuth

volumes and start brushing up on your algorithms.