Unlike most books, Volume 4 will not appear all at once. Instead, 128-page fascicles will be released more or less as Knuth finishes writing them. Though not scheduled for publication until 2000 or later, the fascicles are sure to begin circulating informally before then. (A fascicle describing the MMIX machine is already available on the Internet.) Knuth, now 61, hopes to finish the book around 2003 -- though "that's probably slipped by a year or two," he admits. It could be a decade or two into the next millennium before he completes the set.

Peter Gordon, Knuth's editor at Addison-Wesley, sounds wistful when asked about a due date for Volume 4 (which will actually be published as three "sub-volumes"). "From my 20 years' experience in computer-science publishing, the most frequently asked question by far is, 'Where is Volume 4?'" he says. "Nobody has to say more than that, who the author is, what book they're talking about. Just 'Volume 4.'"

Speaking with Knuth, one gets the impression of a man hard pressed to keep up with his mind's high-speed output of ideas. His writing career dates back to 1957, when, as a 19-year-old freshman at the Case Institute of Technology, he earned $25 for the publication of his "Potrzebie System of Weights and Measures" in Mad magazine. His style has remained delightfully literate, featuring sly plays on the jargon of computing, as well as some that are not so sly: Chapter 2 of "Fundamental Algorithms" opens with "Hamlet's" first-act resolution "Yea, from the table of my memory I'll wipe away all trivial fond records."

Besides demonstrating the techniques of clear, efficient coding, Knuth has sought to bring a deeper sense of aesthetics to the discipline. "You try to consider that the program is an essay, a work of literature," he says. "I'm hoping someday that the Pulitzer Prize committee will agree." Prizes would be handed out for "best-written program," he says, only half-joking. Knuth himself has already collected numerous awards, including the National Medal of Science from then-President Jimmy Carter and Japan's prestigious Kyoto Prize.

And though it may take him another quarter century to complete his magnum opus, Knuth is already dreaming up projects to come -- including the computer-aided composition of an orchestral piece based on the Book of Revelations.

At Stanford, Knuth no longer teaches, though he occasionally lectures on whatever happens to interest him at the moment. (This fall will find him at the Massachusetts Institute of Technology, talking about "Things a Computer Scientist Rarely Talks About.") When he has the time, Knuth reads four-hand piano music with friends on his Bosendorfer grand, or plays the 16-rank organ that stands across from it in the music room of the Palo Alto, Calif., home he shares with his wife, Jill. Otherwise, his days are spent sifting through scientific journals, research papers and pages on pages of notes for his next books. "I'm obsessively detail-oriented," Knuth says. An example: During the 10 years or more in which Knuth was occupied designing the TeX typesetting system and revising Volumes 1 through 3, he accumulated a 270-inch stack of such correspondence. Twenty-two and a half feet of heady research may be daunting, but more revealing is the fact that Knuth actually measured it.

Though the world of programming may have little time these days for Knuth's rigorous analytical style and painstaking attention to low-level detail, his work remains an indispensable contribution to the body of knowledge that is computer science. He will perhaps one day be remembered as programming's Dr. Johnson, but the label would do him a disservice, for Knuth's ideas of elegance can be applied to more disciplines than simply the digital realm. Knuth hesitates at this suggestion, then demurs: "Everyday life is like programming, I guess," he says. "If you love something you can put beauty into it."

The Free Software Project
Read Andrew Leonard's book-in-progress on Linux and open source -- and post your comments.