10 Jun
2021
10 Jun
'21
9:58 a.m.
Warner Losh <imp(a)bsdimp.com> wrote:
So I've seen a number of places that talk about
Unix TS 3.0 -> 4.0 -> 5.0
progression and how System III was released and System V was released and
System IV was internal only.
What I've not seen is the "why" part of this. Why was it internal only?
Warner
I did some contract programming at Southern Bell (one of the operating
companies) for a while, starting in the summer of 1982. We used
USG Unix 4.0 on a PDP-11/70. There were also vaxen.
It was no longer being called PWB by then. In fact, I think that PWB
was V6 based and an ancestor of System III.
System III had been released to the world. When I asked why AT&T
hadn't released Unix 4.0, I was told that the policy was to release
one version behind what was being run internally.
After the Bell System break up (1/1/1984), AT&T decided to just
release what was current, thus the jump to System V, with "System IV"
never having gotten outside the Bell System.
When I was doing the contract work, although we were on Unix 4.0,
we used documentation from Unix 3.0; they didn't bother reprinting /
updating the manuals since the primary changes were apparently
performance improvements in the kernel. (For example, they moved
to hash tables for many things instead of simple linked lists.)
I still have the reference manual and the photocopies I made of all
the other documentation. :-)
There was also a screen editor, named 'se' (NOT related to the Georgia
Tech 'se' editor) which they'd managed to shoehorn onto the '11.
It was kinda cool. I used both it and 'ed'. vi was also available but
I found the modal stuff weird and didn't end up learning it until a year
or two later. :-)
Hope this is of interest.
Arnold