Back to the egg - Emacs in the 70s

1 Ab Initio

In the beginning, there was TECO1.
And darkness was on the face of the programmers,
And the MIT hackers said “Let there be Emacs”,
And there was a collection of TECO maros".

This is the first in what is planned to be a series of posts on “The people of Emacs” who have influenced me. I’ve found that the use of Emacs is often a marker for “interesting people doing interesting things”. I have absolutely zero interest in tracking weight lifting stats, but using Emacs on Android as a base for “replacing proprietary software with free alternatives”? Now I’m listening.

Since I’ve been noodling around with Emacs since something like 1979 I’m going to do this in installments. This is part of the Emacs Carnival for December 2025. I suspect I will write several installments after the December carnival has ended.

2 Try this at home kids !

Because Emacs people are fundamentally a hands-on crowd, I’ll try to start each installment with some DIY project. Thanks to fine folks at SDF Public Access UNIX System you can log into a TOPS10 or TOPS20 system today and run the original TECO Emacs. You can run your own PDP-10 on a raspberry pi if you want even more fun.

Go to https://sdf.org/ or https://twenex.org/ to get started.

“If you’re computer doesn’t have 36 bits,
you’re not playing with a full DEC2

Your assignment is to create yourself an account, log in, run Emacs, create a file, and save it. Bonus points for running dired. Double bonus points if it’s a finger plan file (the TOPS20 version of “About Me”). Can you edit a file in Emacs c.a. 1980? How much of what you do today in Emacs was there at the beginning? The past is a foreign land.

Emacs was created on multi-user systems. That had implications for the ethos and the culture around it. This may be new to you if you started computing after, say, 1985. When you log in, run “finger” to see who’s logged in and what they’re doing. You will be presented with BB3 (bulletin-board) messages (system local messages that other people post). There was/is “talk” for connecting with other users on the system.

One of the many things that was cool about the TOPS20 exec (shell) was the command line help system. Type “?” at any point (no return) and you will see the list of valid command options at that point. This was mimicked in Cisco IOS and Columbia C-Kermit. Bits of it were re-implemented in tcsh (TENEX C shell) and later in bash. Type “?”. See what happens.

Triple bonus points - log in using telnet. Who needs ssh? Who needs encrypted passwords on the wire?

3 The people

3.1 RMS - sine non qua

I don’t think I need to or can say anything new about Stallman, his vision, his skill as a programmer, his dedication, etc. For my entire career he was a force, a presence, a coding-monster, a moral inspiration, etc. The Emacs story began 6 or 7 years before FSF, but I think the sharing culture that created it was and is the heart of free software and where it most clearly lives to this day.

3.2 Guy Steele and the CMU connection

Guy Steele contributed to the initial development of Emacs. He was a student at MIT at the time. He later moved to Carnegie-Mellon University (CMU) as an assistant professor. He later co-authored one of the standard references on C and helped standardize Java. I’m assuming he brought Emacs with him from MIT. Secondary effects indicate that this is probably true. I know of at least two other implementations of Emacs that happened at CMU around that time: “Gosmacs” by James Gosling (later of Java fame), and FINE (FINE Is Not Emacs) for TOPS10 (more on that later). My future wife was an undergraduate at CMU at the time and knew Steele and his family.

3.3 More CMU connections

The plot thickens. My older brother went to CMU as an undergrad in the fall of 1978. He was running in the computing circles where Emacs versions were proliferating. My first memory of Emacs is a stack of green-bar print-outs of the source code (TECO) for Emacs that sat for a year or so in my younger brother’s room. He and several of his friends from CMU worked as interns during the summers for Steve Wilhite (he of GIF file format fame) at CompuServe. Steve was developing compilers, run-times and programming support tools (like editors). They brought FINE with them from CMU. CompuServe ran (to its dying day) on DEC10s. I later worked for Steve in my first job out of college. FINE was there and waiting for me.

3.4 The Columbian Exchange.

To keep this post mostly in the 70s I have to mention a group in passing that deserves its own history. It turns out I was in the right place at the right time with (in retrospect) an incredible group of friends.

In high school, before the Apple-II, before Star Wars we had a clique that called itself the “Computer Wizards”. In many cases we were sons of professors and the odd Bell Labs/Western electric/Battelle engineer. We made our own light sabers before there was Star Wars merch on the shelves. Members of the group went on to places like CMU, Stanford, MIT, Berkeley, Microsoft, Amazon, and Google. Various of us were responsible for things such as the Document Object Model (DOM), the Java VM at Sun, C++ standards, bits of the Mach kernel (running to this day as the core OS of macs), high speed networking research, tcsh, IETF standards work and managing a James Bond movie shooting site in Puerto Rico that sometimes doubled as a world famous radio telescope. Several migrated to CMU. Several more of us spent time working at Ohio State University Computer and Information Science (osu-cis) in the early days of GNU/Gnu Emacs. For about 10 years there was a constant flow of people between OSU, CMU and, later, silicon valley. Emacs was part of the ethos and the environment.

4 Feedback and submissions

I’d welcome feedback and corrections. This was mostly written from memories, some of which are over 45 years old.

I’d also welcome other people’s stories. If you’re reading this in December 2025 write something up on your blog and drop me a link. I’ll add it to the carnival. Late submissions might even be accepted.

See Emacs Carnival for December 2025 for instructions.


  1. TECO is an editor originally for use with paper tapes that provided the substrate for the initial version of Emacs. It provided the flexibility and power needed to create Emacs. This was apparently appreciated by others as well, as I was peripherally aware of a separate set of TECO macros curated inside CompuServe. I confess to never having used it directly for anything serious, but since every key did something when you typed it the joke was “Type your name and see what happens”. ↩︎

  2. Digital Equipment Corporation (DEC) made the PDP-10 line of computers that had a 36 bit architecture. They were word (not byte) addressable with the addresses being stored in 18 bits of a word. For a while, fans of DEC’s “Big” PDP-10s had an attitude about their 16 bit (PDP11) and 32bit (VAX) cousins. ↩︎

  3. I wrote a “BB” knock-off on Vax/VMS in Fortran-77 using the EDT editor long ago. People tried to port many good ideas form the TOPS10/20/TENEX/ITS systems to Unix and other environments. ↩︎

comments powered by Disqus