Hacker News Reader

Comment by cafard | Hacker News Reader

Unix was originally created to run on mainframes? Brian Kernighan was one of the creators of Unix?

The Kernighan thing is a kinda and a definitional problem.

No he didn’t work on the kernels or C. But he did work on awk and write a lot on top of roff that made Unix useful to Bell. It really started as a documentation system internally. Also he was the primary test driver of a lot: building tools with tools. I think he’s too humble about his work.

His memoir, mentioned by another commentator, is an amazing read. I highly recommend it.

I’d give Brian Kernighan credit for a lot of the userspace of early UNiX. Besides awk, he wrote and/or contributed to many other tools, wrote lots of documentation, man pages, filmed videos introducing UNIX, played historian and documentarian roles as well. Today, you might call him Technical Marketing Manager or Developer Advocate. He did a lot to make UNIX usable and known to the world outside of AT&T.

Yeah, huge early adopter of lex and yacc. Really drove the potential of pipes. He’d take half baked ideas and show Thompson, Richie, et al the real potential. Kicked off many late nights of their inspired work. Such a supportive human being, too.

I’ve been using awk for 20 years, and when I read his memoir I had a light bulb go off and really got it like I never had before. Such an amazingly powerful tool.

He’s an incredible writer. So many of his books have demystified things that were vague to me. I still read his papers that use real terminals, etc. because I always obtain some kind of mini enlightenment moment. The Unix Programming Environment is so great and the books he supervised like TCP/IP Illustrated by Stevens.

I keep the first volume of the latter on my desk to show “that’s not how that works.” Those diagrams <3.

It was created for the PDP-11, a minicomputer. It was smaller than a mainframe but larger than a microcomputer (also know as the personal computer).

You are correct about Kernighan and if you want to know more about the story behind early UNIX, I recommend this book: "UNIX: A History and a Memoir".

Unix was originally created on a PDP-7, and later ported to an 11/20.

Neither is substantially larger than a rack-mount PC.

You are right, my inflationist memory transformed the 7 into a 11 !

Here's a tale about bringing Unix V0 up on a recovered PDP-7 at the Living Computers Museum + Lab[0]. They logged in as 'dmr'.

[0] https://livingcomputers.org/Blog/Restoring-UNIX-v0-on-a-PDP-...

That looks substantially larger than a rack-mount PC.

> It was smaller than a mainframe but larger than a microcomputer (also know as the personal computer).

Microcomputers didn't exist until R2E released the Micral N years later, and François Gernelle came up with this word. The Personal Computer was created by IBM in 1981, eight years after the birth of the microcomputer.

You're not entirely wrong though, UNIX (think X as cross platform) was technically born with the C language on the PDP-11.

The early "Unics" was written in assembly for the 18 bits PDPs ISA (more precisely a PDP-7, but the ISA was shared with the 1, 4, 7, 9 and 15). Then they came up with a new high level language to port it to the 16 bits PDP-11 ISA, which eventually became C.

Most microcomputers were called "Personal Computers" back even before IBM's PC came out. i.e. the Apple II was a personal computer. The generic term just meant you were not sharing the machine with someone else.

Reply on news.ycombinator.com