· Menlo Park, CA
Haute Vulgarisation
Richard Stevens, Silvanus Thompson and Isaac Asimov wrote about hard things in plain English without dumbing them down. What made those books work, and why we get so few of them.
#books #writing #learning #programming
It was hard to learn how to write networked programs until I stumbled upon Richard Stevens’ books: I didn’t know their genre then, but I enjoyed them very much.
Other authors have written about complex topics the same way. Silvanus Thompson did it for calculus in “Calculus Made Easy”, and Isaac Asimov did it for physics in “Understanding Physics”.
The connective tissue is the language: plain English prose that exposes a hard topic without dumbing it down. The genre has a name, haute vulgarisation, which is French for high-brow popularization. It is a peculiar name given that we’re referring to books whose vocabulary lets a normal person understand what’s in them. I like these books because they have a high ratio of understanding gained to time spent studying: you absorb a topic smoothly, without the book feeling like a paper brick, and without vocabulary indigestion. What makes them so good?
First, the author’s mastery of the topic. It is rare to find someone who knows a complex topic well enough to talk about it. Writing about it is even harder. But then writing about it simply enough that a highly motivated kid can follow is a step beyond what most authors are capable of.
Let’s take Stevens as the example.
He had mastered the topic so completely that the field felt finite. Most books apologize in advance by putting “Introduction to” in the title, with the author feeling sorry for the poor reader. It felt minor until I realized how bold it would be to write a book entitled “The Drumming Book” after playing drums for twenty years.
He introduced concepts and vocabulary slowly, and used a new term only when he needed it. I grew accustomed to the sound of the new words, and a third of the way through, the once foreign terms had become my friends. Compare this to introductory math books whose first sentence begins with “Let n be…”. Those feel like waiting at a bus stop with strangers. His felt like sitting in a friend’s living room drinking coffee.
He used examples. A lot of examples. The Polish edition I used came in two volumes of about a thousand pages each, and more than half of those pages were small programs that acted as companions to the words. Each abstract idea was immediately put in motion by something practical that I could run, tinker with, and tweak to see a different result.
He also illustrated his books very well. “TCP/IP Illustrated” (the Polish edition was titled “TCP/IP Bible”) has hundreds of diagrams and hundreds of code listings, and the rest is pure English prose. It is as if he knew that some part of the population are visual thinkers, and he catered to both readers and watchers.
So why don’t we have more books like this?
To start with, there aren’t many people who want to be authors. Most people who learn network programming or physics in 2026 do it to solve a problem at work and move on, sometimes without understanding the problem at all. They aren’t planning to give up their jobs to explain the topic to others.
Perhaps it’s because understanding the complex topic requires exceptional effort, and then to explain it to a general audience requires divine skill. You need to escape your head, forget all the learnt terms, meanings and connections and put yourself in the reader’s mind, just to then see the topic again, but with their fresh eyes.
We also have a chicken and egg problem: if we have no good books, how does the future author learn the topic before they are ready to explain it? Stevens must have learnt network programming from the RFCs and the BSD source code, which had to be a royal pain in the neck. His books are like startups: he was the only man crazy enough to go through all this work and then write a book.
What I found works is following the trail from one good book to the next. I found Brian Kernighan’s books partly because Stevens mentioned them, but more because Stevens’ books looked beautiful. He typeset them himself, in troff, and said so in the prefaces. When I went looking into how, I found Kernighan, who wrote much of the troff toolchain, and whose books turned out to have the same quality. Good authors leave a trail to other good authors. When you want to learn a topic, find the one book that works for you, and it will point you to the rest.
It’s possible that the books I like to read are the books that the author had fun writing. Asimov stopped teaching at Boston University in 1958 to write full time, and then released dozens of books over dozens of years with remarkable consistency.
If you know any good books of this type in your field, email me.