Good day everyone, I just wanted to share that I've put up a bit of info as well as some book covers concerning UNIX standards that were published from the 80s til now:
https://wiki.tuhs.org/doku.php?id=publications:standards
I did my best to put down a bit of information about the /usr/group, POSIX, SVID, and SUS/Open Group standards, although there's certainly more to each story than what I put down there. Still, hopefully it serves to lay out a bit of the history of the actual standards produced over time.
I'm kicking myself because one of the things I could've produced a picture of but didn't save at the time is the cover of IEEE 1003.2, a copy of this popped up on eBay some time in the past year and for reasons I can't recall I didn't order it, nor did I save the picture from the auction at the time. In any case, if anyone has any published standards that are not visually represented in this article, I'm happy to add any photos or scans you can provide to the page.
Also pardon if the bit on spec 1170/SUS may be shorter than the others. Admittedly even having most of this on the desk in front of me right now, I'm fuzzy on the lines between POSIX, the Single UNIX Specification, the "Open Group Specification", spec 1170, etc. or if these are all names that ultimately just refer to different generations of the same thing. Part of getting this information put down is hoping someone will be along to correct inaccuracies :)
Anywho, that's all for now. Feel free to suggest any corrections or additions!
- Matt G.
FYI, this just got passed by Vint Cerf. Very sad news.
---------- Forwarded message ---------
From: vinton cerf via Internet-history <internet-history(a)elists.isoc.org>
Date: Tue, Jun 4, 2024 at 3:18 PM
Subject: [ih] Mike Karels has died
To: internet-history <internet-history(a)elists.isoc.org>
Mike Karels died on Sunday. I don’t have any details other than:
https://www.facebook.com/groups/BSDCan/permalink/10159552565206372/https://www.gearty-delmore.com/obituaries/michael-mike-karels
Mike was deeply involved in the Berkeley BSD releases as I recall, after he
inherited the TCP/IP implementation for Unix from Bill Joy (am I
remembering that correctly?).
RIP
v
--
Internet-history mailing list
Internet-history(a)elists.isoc.org
https://elists.isoc.org/mailman/listinfo/internet-history
Today after trying to decipher the online help for vim and neovim, I
decided I'd had enough and I opted for nvi - the bug for bug vi
compatible that I've used for so long on FreeBSD. It handles cursor
keys, these days (my biggest gripe back when, now I'm not so sure it's
an improvement). It's in-app help pages are about 300 lines long, the
docs are just four of the 4.4 docs: An Introduction to Display Editing
with VI, Edit: A tutorial, EX Reference Manual, and VI-EX Reference
Manual - all very well written and understandable. It does everything I
really need it to do without the million and one extensions and
"enhancements" the others offer.
In doing the docs research, I found many, many references to a "/Vi
Quick Reference card"/ in the various manpages and docs. I googled and
googled some more and of course got thousands of hits (really many
thousands), but I can't seem to find the actual card referenced. I'm
pretty sure what I want to find is a scanned image or pdf of the card
for 4.4bsd.
Do y'all happen to know of where I might find the golden quick ref card
for vi from back in the 4.4bsd days or did it even really exist?
Will
I keep Lomuto and Lomuto, "A Unix Primer", Prentice-Hall (1983) on my
shelf, not as a reference, but because I like to savor the presentation.
The Lomutos manage to impart the Unix ethos while maintaining focus on the
title in a friendly style that is nevertheless succinct and accurate.
Doug
A few years ago, someone -- and I've forgotten who, forgive me -- kindly gave me a copy of the source code for a UNIX for the AT&T PC6300 called IN/ix, developed by INTERACTIVE Systems. I have found precious little about this system online. Apparently the PC/ix UNIX for the IBM PC XT is fairly well preserved, but I can't find much about IN/ix.
For what it's worth, the login herald in the source code reads:
"IN/ix Office System (c) Copyright INTERACTIVE Systems Corp. 1983, 1988"
Presumably this was PC/ix, but targeting the AT&T 6300? Does anyone have any more knowledge of IN/ix?
If you're interested in digging into it yourself, I've dropped the source here:
https://archives.loomcom.com/pc6300/
(N.B.: All the files inside the zip are compressed, that's just how I got it)
-Seth
--
Seth Morabito * Poulsbo, WA * https://loomcom.com/
> Does anyone here have any source material they can point me to
> documenting the existence of a port of BSD curses to Unix Version 7?
Curses appears in the v8 manual but not v7. Of course a
conclusion that it was not ported to v7 turns on dates. Does
v7 refer to a point in time or an interval that extended until we
undertook to prepare the v8 manual? Obviously curses was
ported during or before that interval. If curses was available
when the v7 manual was prepared, I (who edited both editions)
evidently was unaware of any dependence on it then.
Doug
So I've been doing a bit of reading on 1A and 4ESS technologies lately, getting
a feel for the state of things just prior to 3B and 5ESS popping onto the scene,
and came across some BSTJ references to the programming environments involved
in the 4ESS and TSPS No. 1 systems.
The general assembly system targeting the 1A machine language was known as
SPC-SWAP (SWitching Assembly Program)[1](p. 206) and ran under OS/360/370, with
editing typically performed in QED. This then gave way to the EPL (ESS
Programming Language) and ultimately EPLX (EPL eXtra)[2](p. 1)[3](p. 8)
languages which, among other things, were used for later 4ESS work with cross-
compilers for at least TSS/360 by the sounds of it.
Are there any recollections of attempts by the Bell System to rebase any of
these 1A-targeting environments into UNIX, or by the time UNIX was being
considered more broadly for Bell System projects, was 3B/5ESS technology well on
the way, rendering attempting to move entrenched IBM-based environments for the
older switching computation systems moot?
For the record, in addition to the evolution of ESS to the 5ESS generation, a
revision of TSPS, 1B, was also introduced which was rebased on the 3B20D
processor and utilized the same 3B cross-compilation SGS under UNIX as other 3B-
targeted applications[4]. Interestingly, the paper on software development
in [4](p. 109) still makes reference to Programmer's Workbench as of 1982,
implying that nomenclature may have still been the norm at some Bell Labs sites
such as Naperville, Illinois, although I can't tell if they're referring to
PWB as in the branch of UNIX or the environment of make, sccs, etc.
Additionally, is anyone aware of surviving accessible specimens of SWAP
assembly, EPL, or EPLX code or literature beyond the BSTJ references and paper
referenced in the IEEE library below? Thanks for any insights!
- Matt G.
[1] - https://bitsavers.org/magazines/Bell_System_Technical_Journal/BSTJ_V58N06_1…
[2] - https://ieeexplore.ieee.org/document/810323
[3] - https://bitsavers.org/magazines/Bell_System_Technical_Journal/BSTJ_V60N06_1…
[4] - https://bitsavers.org/magazines/Bell_System_Technical_Journal/BSTJ_V62N03_1…
> Doug McIlroy was generating random regular expressions
Actually not. I exhaustively (within limits) tested an RE recognizer
without knowingly generating any RE either mechanically or by hand.
The trick: From recursive equations (easily derived from the grammar of
REs), I counted how many REs exist up to various limits on token counts,
Then I generated all strings that satisfied those limits, turned the
recognizer loose on them and counted how many it accepted. Any disagreement
of counts revealed the existence (but not any symptom) of bugs.
Unlike most diagnostic techniques, this scheme produces a certificate of
(very high odds on) correctness over a representative subdomain. The scheme
also agnostically checks behavior on bad inputs as well as good. It does
not, however, provide a stress test of a recognizer's capacity limits. And
its exponential nature limits its applicability to rather small domains.
(REs have only 5 distinct kinds of token.)
Doug
