Skip to main content
Exploring what’s next in tech – Insights, information, and ideas for today’s IT leaders

Unix: Love at first byte

Brian Kernighan’s new book answers the question, “How did a two-person experiment literally grow into something that changed the world?” Anyone who loves Unix will come to know what it felt like to be part of this marvelous adventure.
Brian Kernighan, the “K” in “awk,” has written a wonderful book— UNIX: A History and a Memoir—about the birth and evolution of Unix. It provides not just a record of events and major participants in the effort but a sense of what it was like to be deeply involved in the development of an operating system that started out modestly but has come to dramatically alter much of the technology that underpins our everyday lives today. It answers this question posed in the preface: “How did a two-person experiment literally grow into something that changed the world?” In fact, I couldn’t get past the preface without reflecting on how even I got picked up by that same wave and carried along for the rest of my career.

Accidental intro to Unix 

Early in the book, Kernighan refers to his initial internship at Bell Labs as an “ultimate piece of good luck,” and I have to agree. Bell Labs’ habit (at least back then) of allowing its researchers to find their own focus and expecting them to find others they wanted to work with was a wonderful way to encourage the growth of promising technology. I can imagine them excitedly chatting in offices and hallways, feeding off one another’s ideas. The level of innovation was exceedingly high, especially for an OS that quickly became a top project. That OS wasn’t Unix during its first round but Multics, and only after some setbacks did Unix take its place.

I didn’t start with Unix until it was already a teenager (13 years old). While not quite the piece of good luck that Kernighan enjoyed, my introduction to Unix was still a bit of luck that I practically tripped over. My second job after finishing my computer science studies at the University of Maryland involved programming in a language called SAIL (Stanford Artificial Intelligence Language) on a relatively small computer running Unix. I’d known nothing about Unix when I started the assignment. My team at the time was automating the criminal docket (case records) for the U.S. district courts. I was busy building a model of criminal procedure (Title 18 of the U.S. code) while one of my office mates was focused on the Speedy Trial Act and managing judges’ calendars so that defendants wouldn’t get off scot-free just because their trials didn’t start within 30 days (do I remember this correctly?) of the defendant’s first appearance or indictment.

Over 1M people read enterprise.nxt. Are you one of them?

The command line

The idea was to automate previously paper-only case files and make caseloads easier to manage. Computers In the courthouses were going to be a definite win over court personnel having to communicate with systems in Washington, D.C., over Tymnet, an international data communications network that was heavily used in those days. It didn’t take more than my first few minutes on that first Unix system for me to appreciate how much I could get done on the command line. Commands like grep and pipes were godsends. I’ll never forget how excited I was to discover how easily I could get serious work done by stringing a few commands together.  If I recall correctly, we used a Zilog system about the size of a dorm-room refrigerator.

Unix had by then acquired an impressive set of tools: man, ls, cat, mv, cp, wc, awk, sort, uniq, cmp, diff, od, dd, head, tail, tr, who, and of course, the Bourne shell. For a system that began as a command-line system, Unix had to accommodate users’ needs right there where the cursor waited. It did this so well that I can still remember being amazed that, with a single command, I could do what previously would have required some serious programming to do. I loved it.

I moved into systems administration around 1983. For about 10 years, I was a sysadmin on SunOS and heavily involved in the community, serving on the board of the Sun User Group for six years and helping to organize annual conferences that drew people from around the world. I chaired a SUG conference in Dec 1991 in San Jose, California. I also began to write for print and then online magazines and even pushed out a few books. After that, SunOS (BSD-based) turned into Solaris (Sys V-based). Linux, with to its focus on open source code and free distributions, was gaining attention by 1992.

Unix: A dominant force

Soon afterward, my own focus switched to Linux, and I spent many years administering Red Hat (RHEL) systems while playing with Ubuntu, Fedora, and other distributions on the side. With a growing family of distributions, Linux now dominates cell phones, supercomputers, and a wide array of other devices that power our lives.

Yet, I can still imagine what it might have been like to walk down the hallways of Bell Labs with Kernighan, spending time talking with other engineers about their work, sharing and expanding on ideas. I had some chances to innovate given the array of tools I have had at my disposal on the various Unix and Linux systems I’ve used over the years and have loved nearly every minute of it.

In nine chapters, "UNIX: A History and a Memoir" takes the reader on a journey that starts with people looking for ways to efficiently solve common mathematical problems and then moves through the evolution of tools and ideas that have led to Unix being the dominant force it is in today’s technology.

It may have taken 50 years to get to where we are today, but I’m still taking a deep breath and saying “Wow!”

Useful links:

UNIX: A History and a Memoir

Book review: Unix, the origin story

The Internet is 50! What’s next?

The death of the command line

Sysadmin survival guide

This article/content was written by the individual writer identified and does not necessarily reflect the view of Hewlett Packard Enterprise Company.