Many know Donald Knuth as the author of The Art of Computer Programming, or TAOCP, an epic four volume set. However fewer are aware the steadfast devotion he applied to perfecting the typography of his books. Consider this story as he was about to publish the second edition of Volume 2.
The TAOCP project progressed smoothly for three volumes—published in 1968, 1969, and 1973—but advancements in the subject matter soon overtook the writing. When, in 1978, a second edition of Volume 2 was required, the Monotype was unfortunately dying out, replaced by newfangled “photocomposition.”
Dr. Knuth looked at his photo-typeset proofs in horror. Gone were the elegant text and math displays that exemplified a fine technical publication. In their place were pages full of words and symbols that, except for the use of typographic fonts, may as well have been prepared on a typewriter. This is not how Dr. Knuth felt his work should be presented to the world.
Like most of us who know how to write code, he decided to do something about it. The story continues in a highly relatable fashion.
He decided to take a break from writing and devise a method of harnessing zeros and ones to replicate the quality he knew possible from his experience with Monotype composition. Dr. Knuth guessed it might take six months, or at the outside, a year.
In the end, it took about ten years to create TeX (the composition software), Metafont (a program for creating fonts for use with TeX), and a collection of the fonts he needed to produce TAOCP, as well as a new approach to writing computer code—“literate programming.” 1
I love this story, not only because it highlights that even great minds like Knuth can drastically underestimate the time it takes to create something, but because he really did let his book sit for ten years so he could perfect its printing method. He was thoroughly unsatisfied with solutions offered by the marketplace. Rather than settle and get his book on the shelves faster he set out to solve the problem. Not many people give quality this consideration. Fewer still act and see it through.
Lest you think Knuth was simply being picky, read pages 338–343 of his Mathematical Typography lecture 2. Here he shows some preliminary examples of mathematical typography from the Transactions of the American Mathematical Society journal in order to “educate your eyes”. He notes various periods of “typographic turmoil” that marred the publication.
Knuth certainly has an eye for detail.
Just like Knuth held out printing his book, he has also held out on releasing an electronic edition for similar reasons.
It’s been 22 years since the PDF format was introduced. Seven years have passed since the first Kindle was introduced. And still most electronic books have yet to catch up to, let alone exceed, the typographic beauty and utility that marked pre-computer typographic craftsmanship.
And Knuth is not the only author to see this as a problem. Matthew Butterick, who has pondered & practiced typography more than most, expressed it this way.
…our newest reading technologies—the Kindle, the iPad—are leaving typography behind. This is a mistake. Those who care about the written word should be invested in preserving typography, because as we lose typography, we’ll also start to lose some of the possibilities of the written word. 3
Matthew also seems hesitant to release less-than-ideal ebooks. Listen to how he describes the Kindle edition of his book Practical Typography.
It may be the best-designed Kindle book available. Which would still make it one of the least impressive achievements of my career. 4
But I digress — let’s return to Knuth.
With the recent release of TAOCP in electronic form Knuth again demonstrates his dedication to typographic quality.
For many years I’ve resisted temptations to put out a hasty electronic version of The Art of Computer Programming, because the samples sent to me were not well made.
But now, working together with experts at Mathematical Sciences Publishers, Addison-Wesley and I have launched an electronic edition that meets the highest standards. We’ve put special emphasis into making the search feature work well. Thousands of useful “clickable” cross-references are also provided — from exercises to their answers and back, from the index to the text, from the text to important tables and figures, etc.
Note: However, I have personally approved ONLY the PDF versions of these books. Beware of glitches in the ePUB and Kindle versions, etc., which cannot be faithful to my intentions because of serious deficiencies in those alternative formats. 5
If you’ve ever felt the mental agony of a hastily created ebook you will appreciate the effort Knuth has applied to TAOCP. It is my hope that more authors and publishers will pursue typographic quality, not only reclaiming what has been lost, but taking advantage of the many features available to an electronic edition.