Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ames!ucsd!ucsdhub!esosun!seismo!uunet!auspex!guy
From: guy@auspex.UUCP (Guy Harris)
Newsgroups: comp.unix.questions
Subject: Re: UNIX 9th edition ????
Keywords: need Berkeley verification
Message-ID: <973@auspex.UUCP>
Date: 8 Feb 89 09:35:28 GMT
References: <487@maxim.ERBE.SE> <8872@alice.UUCP> <316@dcs.UUCP> <9434@cit-vax.Caltech.Edu>
Reply-To: guy@auspex.UUCP (Guy Harris)
Organization: Auspex Systems, Santa Clara
Lines: 98

Some more stuff, for the incurably trivial:

> +++ 'AT&T' branch +++
>PWB: typesetter

PWB/UNIX 1.0 was basically V6-based, although it had some post-V6,
pre-V7 stuff that also appeared elsewhere:

	"typesetter", or "Phototypesetter, Version 7", which included a
	C-language nroff/troff implementation, a newer C compiler with
	more of the features of modern C ("long"s, casts, I forget what
	else), the standard I/O library, and an "lseek" library routine
	implemented atop V6's "seek" that took a "long" as an argument;
	(actually, there were some tiny differences between what was in
	"Phototypesetter, Version 7" and what was in PWB/UNIX 1.0 - so
	tiny I forget what the were)

	assorted kernel changes that, I think, appeared in a "diff"
	listing Ken Thompson sent out called "50 changes to UNIX" or
	something like that.

> --- offshoot ---
>Unix RT: real time
>MERT: real time
> --- end offshoot ---

Other way around; MERT was the first one on this branch (more-or-less
V6, I think, but implemented as a layer atop a real-time kernel), and
UNIX/RT came from that.

>TS1.0:

As I remember from a UNIX/TS 1.0 manual I saw once, this was sort of an
"almost-V7" merged with some stuff from PWB/UNIX.  (V7 file system, for
example.)

>TS2.0:
>SysIII: which spawned Xenix

I think the first Xenix was V7-based, and subsequent Xenix releases
picked up stuff from S3 and S5.  The big thing about V7 was the binary
licensing schedule that allowed vendors to sell UNIX as the OS on their
boxes - many based on 16-bit micros such as the Z8000 or on 16/32-bit
micros such as the 68000 - without their customers having to fork out
$20K or more for a UNIX source license.

I think there are boatloads of internal AT&T UNIX flavors that got
merged into S3; some AT&T people may be able to fill in the details
there.

>TS4.0:
>SysV:
>SysV Release 2:
>SysV R3:
>SysV R4: also known as SunOS 5.0  [in the works]

Well, SunOS 5.0 - or whatever it's called - will be based on S5R4, but
Sun will probably add stuff of their own, just as everybody else
will....

> +++ 'BSD' branch +++
>1 BSD:

V6-based, as I remember.

>2 BSD:

V7-based.

> --- offshoot ---
>2.8 BSD:
>2.9 BSD:
>2.10 BSD: 1986? [also received code from 4.3 BSD]
> --- end offshoot ---

>3 BSD: has merge from 32V
>4 BSD: 1980
>4.1 BSD: 1981
> --- offshoot ---

>version 8:  [apparently back to the trunk 8-]

Yup, although V8 and company are done at Bell Labs Research, not
Berkeley; I'd be more inclined to call it an AT&T branch - or just "back
to the trunk".

>version 9:
>Plan 9:

(To what degree can Plan 9 be thought of as a descendant of V8/V9?)

>SUN OS 2.0: 1983? NFS, received code from 4.2 BSD

SunOS 1.0, more likely.  It got stuff from 4.1cBSD and 4.2BSD.  (I seem
to remember hearing references to SunOS 0.x, for some value of x, but I
don't know what that was or if it existed.)

SunOS 2.0 was the first one with NFS.