On Sun, Jul 20, 2025 at 9:52 AM Will Senn <will.senn@gmail.com> wrote:

I remember having discussions about vi vs emacs in the mid 1990's. I'm
curious if those were the first public wars about editors, or if y'all
remember earlier flamewars on the subject? Maybe KEDIT vs EDIT?

Hmm I don't think it was really a new thing, as I think the UNIX and VMS editor wars occurred long after earlier conflicts. Hey, even UNIX had way more than just ex/vi and emacs. There was the Rand Editor and Fred, to name two others, who were popular in the early PDP-11 UNIX days. Also, remember that for the early DEC PDPs 1/6/10 machines, there were Stopgap, SonOfStopgap (a.k.a. SOS - which became DEC's EDIT), Teco, and eventually EMACS [remember UNIX emacs is a clone of the original TOPS-10/ITS EMACS]; and probably others that I am forgetting - those the ones I used in PDP-10 days. In fact, I learned Emacs on the PDP-10, but neither Gosling nor Zimmerman had yet written their C versions. I used Fred as my first video editor on UNIX and did not learn ex(1) until a few years later, when I was at Tektronix, as it was not available on the CMU systems. By the time EMACS arrived (and required a 32-bit system), I was fluent with it and never bothered to go back.

I'll also note that the PDP-8 and PDP-11 families had their own set of editors, and that was just in the DEC world. The IBM world was really not any different. Any interactive system needs an editor. In all cases, the line editors were first because that was what you could use with a Teletype Model 28 and later Model 33 ASR or a Friden (Singer/Link) Flexowriter. With the advent of the "glass tty," the editor wars took on new life as several editors emerged on different systems.. As I said, UNIX had a number before ex/vi. The Rand Editor was the most widely used in the UNIX world early on. [ I don't remember why we used Cornell's Fred over the Rand Editor at CMU in those days, but I know we had both].

To make matters worse, until UNIX came along, the concept of file types and "access methods" was fairly well ingrained in those systems. So features like embedded line numbers (or not) would get into the war. The DEC PDP 1/6/10 Stopgap family added line numbers that compilers and other tools were aware of. IBM ISAM files did the same thing, which is how we sometimes prepared our source code TSS, as you could create card images from ISAM file easily. At the time, I remember there were arguments about needing line numbers or not. A dirty secret about me, I turn on line numbers in vi, because I often find it faster to refer to things that way. That is probably because I learned the TSS editor first on the IBM 360 and then used SOS on TOPS, all before I started using Unix. So when I did, I learned to use ed(1) with its dynamic line numbering and never stopped, even when vi mode showed up in ex(1).

Clem