Unix History

[Kenneth Almquist]

> > This raises the burning question:  whatever happened to
> > Systems I and II, and, especially, IV? ...

Back in 1978, I think it was, a grand new numbering scheme for UNIX releases
put out by USG was devised.  Release numbers were set back to 1.0 and the
operating system was renamed UNIX/TS.  (The "TS" stands for time sharing.)
At the same time, MERT, a different operating system whose primary use was
to run a UNIX emulator, was renamed UNIX/RT (the "RT" stood for real time)
and also had its release numbers set back to 1.0.  The idea was that com-
patable releases of UNIX/TS and UNIX/RT would have the same release numbers.
UNIX/RT proceeded to bite the dust and the name of UNIX/TS gradually shifted
back to plain "UNIX".

USG started producing top level releases once a year.  Second level releases
were used to make features available as they were written.  Third level
releases were limited to bug fixes.  For example, UNIX 3.0 was followed by
UNIX 3.0.1, which was basicly UNIX 3.0 with a few bug fixes.  (It was also
followed by UNIX 3.1, which may or may not have come out before UNIX 3.0.1.)

The first two versions of USG UNIX that were release externally to AT&T
were UNIX 3.0.1 and UNIX 5.0.1.  They were called System III and System V
externally.  There were no systems named System I, II, and IV.

> System I was, approximately, PWB, which was released.

PWB UNIX was not developed by USG.  One of the goals enunciated in the
switch to the new release naming scheme was to eliminate the various special
versions of UNIX floating around.  SCCS was included in UNIX/RT 1.0.  The
PWB code didn't make it into USG UNIX until release 2.0.

I don't know much about the version of PWB UNIX that was released externally,
but I think that predated UNIX/TS 1.0 significantly, so there were probably
UNIX/TS 1.0 features not in the external version of PWB UNIX, as well as vice
versa.

> System II was an improved PWB, incorporating some useful
> things like a souped-up shell.

The Borne shell (which was in UNIX/TS 1.0) did not undergo any major changes
between UNIX/TS 1.0 and UNIX 5.0.

> Since "System V" has now
> become a magic marketing buzzword, the top-level numbering is absolutely
> frozen for external purposes, and all future releases will be V.something
> rather than VI, VII, etc.

The real horror is that the internal numbering system has been abandoned,
so that even those of us inside AT&T have to suffer with names like
"System V Release 2 Issue 2".  Does driving people up the wall with a
chaotic numbering scheme really encourage them to buy from AT&T?

> If you want to get a look at what System IV was like, check out the PDP11
> distribution of System V.  PDP11 SysV is really SysIV.

Not really.

> Although AT&T won't admit it in so many words, they effectively
> abandoned work on PDP11 Unix a long time ago.

Most new features made it to the PDP11.  The major exception is the news
object module format, which makes SDB possible.  SDB is a UNIX 3.0 feature
which did not make it to the PDP11.  The UNIX 4.0 and UNIX 5.0 releases
for the PDP11 were pretty complete.

> For example, although some of the SysV performance work wouldn't fit
> on the 11, *some* of it would.  None of it was applied to PDP11 SysV.

The PDP-11 uses hashing to implement the sleep/wakeup facility in the kernel
under System V.  I don't know of any other System V performance work that
was applied to the PDP11 (but then if it had been it would probably would not
have helped a machine the size of a PDP-11 much.)

>  The PDP11 SysV shared-memory stuff is also different from and
> incompatible with the regular SysV version.

Right, but it is present, which is a difference from UNIX 4.0.  The IPC
stuff was first released in UNIX 4.2.
				Kenneth Almquist
				ihnp4!houxm!hropus!ka	(official name)
				ihnp4!opus!ka		(shorter path)