[TUHS] advent of "modern" Unix OS

[Larry McVoy via TUHS]

On Sat, Apr 25, 2026 at 04:32:35PM +0000, Andrew Lynch via TUHS wrote:
> HiI've been watching the conversation on the various Unix and Unix-like operating systems and began to wonder.?? Is there such thing as an era of "modern" Unix OS?????
> There seems to me a vast difference between the early Unix variants and what came later suggesting that at some point a "modern" era began.
> If so, what constitutes a modern Unix??? When did it begin?
> I'd propose the modern era began with Unix or variants for 32-bit architectures *with* hardware memory protection (MMU) and "large" RAM (16MB?).?? I'd further say machines of the class became common place somewhere in the 1988 to 1990 time frame.?? Possibly some examples before then like VAX or other mini-computers or high end workstations like HP, Sun, or Apollo.
> I'd say the difference is the advent of full 32-bit architectures with MMU like 68020 with its PMMU vs. 68000/68010 or 80386 vs. 80286.?? Yes, you could push earlier architectures with special hardware but at some point there was a "quantum leap" into a new era of hardware that carried Unix and variants forward.
> Examples of what I'd consider modern Unix or variants would be AT&T System V, 386BSD & later *BSDs, Linux/GNU, etc.
> So is the concept of a modern era of Unix vs. prior Unix technologies real or an illusion?

It's very real and I'd say it started with 4.x BSD.  BSD added networking
and VM that worked.  BSD added stuff like talk and finger that made the
system much more community friendly.  When I ported networking to SCO it
was pretty miserable, it was a very minimal system.  I get that we all
are proud of what the original Unix managed to do on a PDP-11, it's
remarkable how much stuff worked.  But that said, go back to that era
and tell me you don't prefer what 4BSD gave you.  I absolutely do.

And if you have had the pleasure of running on SunOS 4.x, that was a much
more polished version of BSD.  Mojo did the VM system that gave us the
first version of mmap() (that I'm aware of, maybe Multics had one?).
I think it was Rob Gingell & crew that gave us shared libraries.
There was a lot of good systems work done in that source base.
-- 
---
Larry McVoy           Retired to fishing          http://www.mcvoy.com/lm/boat