Why is csh `restricted'?

From owner-pups@minnie.cs.adfa.edu.au Tue Jan  5 15:43:05 1999
Received: (from major@localhost)
	by minnie.cs.adfa.edu.au (8.9.1/8.9.1) id PAA08076
	for pups-liszt; Tue, 5 Jan 1999 15:41:35 +1100 (EST)
Received: from henry.cs.adfa.edu.au (henry.cs.adfa.oz.au [131.236.21.158])
	by minnie.cs.adfa.edu.au (8.9.1/8.9.1) with ESMTP id PAA08071
	for < pups@minnie.cs.adfa.oz.au>; Tue, 5 Jan 1999 15:41:28 +1100 (EST)
Received: (from wkt@localhost)
	by henry.cs.adfa.edu.au (8.9.1/8.9.1) id PAA10444
	for pups@minnie.cs.adfa.oz.au; Tue, 5 Jan 1999 15:42:15 +1100 (EST)
	(envelope-from wkt)
From: Warren Toomey < wkt@henry.cs.adfa.edu.au>
Message-Id: <199901050442.PAA10444@henry.cs.adfa.edu.au>
Subject: Why is csh `restricted'?
To: pups@minnie.cs.adfa.oz.au (Unix Heritage Society)
Date: Tue, 5 Jan 1999 15:42:15 +1100 (EST)
Reply-To: wkt@cs.adfa.oz.au
X-Mailer: ELM [version 2.4ME+ PL43 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-pups@minnie.cs.adfa.edu.au
Precedence: bulk

I've got a questions that's been niggling me, and perhaps someone might
be able to answer it.

The csh was first released in 2bsd, and came with the copyright notice:

/* Copyright (c) 1979 Regents of the University of California */
/*
 * C Shell
 *
 * Bill Joy, UC Berkeley
 * October, 1978
 */

But my memory tells me that, back in the late 80s, people were saying
that the sources to csh were not freely available. And in the tcsh FAQ
(taken from tcsh version 6.00), I see:

	4.  Where can I get csh sources? 

	Csh sources are not public domain. If you do not have an AT&T V3.2
	source licence or better, you are stuck.

So, can anybody tell me if, when and how did the sources to csh become
restricted, or if not, how this urban legend arose??

Many thanks in advance!

	Warren

Date: Mon, 4 Jan 1999 21:26:35 -0800 (PST)
From: "Steven M. Schultz" < sms@moe.2bsd.com>
Message-Id: <199901050526.VAA19409@moe.2bsd.com>
To: wkt@cs.adfa.oz.au
Subject: Re: Why is csh `restricted'?
Cc: pups@minnie.cs.adfa.oz.au
Sender: owner-pups@minnie.cs.adfa.edu.au
Precedence: bulk

Warren -

> The csh was first released in 2bsd, and came with the copyright notice:

> /* Copyright (c) 1979 Regents of the University of California */
> /*
>  * C Shell
>  *
>  * Bill Joy, UC Berkeley
>  * October, 1978
>  */
> 	Csh sources are not public domain. If you do not have an AT&T V3.2
> 	source licence or better, you are stuck.
> 
> So, can anybody tell me if, when and how did the sources to csh become
> restricted, or if not, how this urban legend arose??

	It is not that they "became" restricted.  They always "were" restricted
	because they were derived from the original Bell Labs (later AT&T)
	sources (code borrowed from /bin/sh).  All UNIX sources were, up until 
	you negotiated the deal with SCO, restricted.

	For a long time you either had a multi-kilodollar source license
	or you didn't run UNIX at all.  The binary distributions came a bit
	later.  Initially when 'csh' was being written you had to have a
	source license.  Typically you'd pay (if memory serves) $25k or so
	(quite a chunk of cash in 1979) for a WesternElectric license, park
	the tapes in a rack and send a copy of the license and a check for a 
	few hundred dollars off to UCB to get the software you really intended 
	to run ;)

	You'll note that the copyright lacks the "may be redistributed ..."
	clauses that we typically associate with UCB software.  The famous
	UCB style of copyright ("copyrighted but redistributable") came
	later.

	Steven

From: Warren Toomey < wkt@henry.cs.adfa.edu.au>
Message-Id: <199901050530.QAA11231@henry.cs.adfa.edu.au>
Subject: Re: Why is csh `restricted'?
In-Reply-To: <199901050526.VAA19409@moe.2bsd.com> from "Steven M. Schultz" at "Jan 4, 1999  9:26:35 pm"
To: sms@moe.2bsd.com (Steven M. Schultz)
Date: Tue, 5 Jan 1999 16:30:25 +1100 (EST)
Cc: pups@minnie.cs.adfa.oz.au (Unix Heritage Society)
Reply-To: wkt@cs.adfa.oz.au
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-pups@minnie.cs.adfa.edu.au
Precedence: bulk

In article by Steven M. Schultz:
> Warren -
> 
> > The csh was first released in 2bsd, and came with the copyright notice:
> > /* Copyright (c) 1979 Regents of the University of California */
> > So, can anybody tell me if, when and how did the sources to csh become
> > restricted, or if not, how this urban legend arose??
> 
> 	It is not that they "became" restricted.  They always "were" restricted
> 	because they were derived from the original Bell Labs (later AT&T)
> 	sources (code borrowed from /bin/sh).  All UNIX sources were, up until 
> 	you negotiated the deal with SCO, restricted.
> 
> 	Steven

I didn't know that any of the sources in 1979 2bsd were contaminated with
AT&T sources. I'll go and do a line comparison between V6 sh, V7 sh and
the 2bsd csh, and see if I can find any signs of contamination.

What else in the original 2bsd is contaminated?

Thanks!
	Warren

From: Warren Toomey < wkt@henry.cs.adfa.edu.au>
Message-Id: <199901050550.QAA11350@henry.cs.adfa.edu.au>
Subject: Re: Why is csh `restricted'?
In-Reply-To: <199901050530.QAA11231@henry.cs.adfa.edu.au> from Warren Toomey at "Jan 5, 1999  4:30:25 pm"
To: pups@minnie.cs.adfa.oz.au (Unix Heritage Society)
Date: Tue, 5 Jan 1999 16:50:06 +1100 (EST)
Reply-To: wkt@cs.adfa.oz.au
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-pups@minnie.cs.adfa.edu.au
Precedence: bulk

In article by Warren Toomey:
> In article by Steven M. Schultz:
> > > So, can anybody tell me if, when and how did the sources to csh become
> > > restricted, or if not, how this urban legend arose??
> > 
> > 	It is not that they "became" restricted.  They always "were" restricted
> > 	because they were derived from the original Bell Labs (later AT&T)
> > 	sources (code borrowed from /bin/sh).  All UNIX sources were, up until 
> > 	you negotiated the deal with SCO, restricted.
> > 
> > 	Steven

Steven is right. An investigation into the csh from 2bsd shows that it
is derived from the Mashey shell in 6th Edition UNIX, but not from the
Bourne shell in 7th Edition.

Hmm, I'll have to go and update my UNIX family tree now.

	Warren

Date: Mon, 4 Jan 1999 21:57:01 -0800 (PST)
From: "Steven M. Schultz" < sms@moe.2bsd.com>
Message-Id: <199901050557.VAA19580@moe.2bsd.com>
To: sms@moe.2bsd.com, wkt@cs.adfa.oz.au
Subject: Re: Why is csh `restricted'?
Cc: pups@minnie.cs.adfa.oz.au
Sender: owner-pups@minnie.cs.adfa.edu.au
Precedence: bulk

Warren -

> From: Warren Toomey < wkt@henry.cs.adfa.edu.au>
> I didn't know that any of the sources in 1979 2bsd were contaminated with
> AT&T sources. I'll go and do a line comparison between V6 sh, V7 sh and

	Indeed they were.  ALL sources were considered "contaminated" or
	restricted - that's why for years and years the only 2.x (and 4.x) BSD
	sites were universities or other companies that had source licenses.

> the 2bsd csh, and see if I can find any signs of contamination.
> 
> What else in the original 2bsd is contaminated?

	Anything that I (or other contributors) didn't write ourselves.

	A good case can be made that stuff ported from 4.4-Lite is not
	contaminated (because 4.4-Lite had the legal blessings of AT&T)
	but I was told at one time anything based on the Net-2 stuff could be
	(is?) contaminated.  Alas by the time 4.4-Lite came out the software
	had bloated so much that very little of it can be ported over.  I 
	grabbed a few ideas and pieces out of the kernel - that's where the 
	"sysctl" stuff in 2.11 came from for example.  But the mainline
	applications are GNU based (megabytes and megabytes of memory assumed).
	I'd like to see someone getting GCC to run natively on a PDP-11! 

	That's why the SCO "Ancient Unix" license is such a milestone event and
	is so important (perhaps more so than some folks realize).

	Up until this point you had to have a US$100K budget to gain access
	to the software we can legally obtain for $100 (no 'K') now.

	Steven

From: Warren Toomey < wkt@henry.cs.adfa.edu.au>
Message-Id: <199901050601.RAA11467@henry.cs.adfa.edu.au>
Subject: Contaminated srcs
In-Reply-To: <199901050557.VAA19580@moe.2bsd.com> from "Steven M. Schultz" at "Jan 4, 1999  9:57: 1 pm"
To: sms@moe.2bsd.com (Steven M. Schultz)
Date: Tue, 5 Jan 1999 17:01:49 +1100 (EST)
Cc: pups@minnie.cs.adfa.oz.au (Unix Heritage Society)
Reply-To: wkt@cs.adfa.oz.au
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-pups@minnie.cs.adfa.edu.au
Precedence: bulk

In article by Steven M. Schultz:
> 	A good case can be made that stuff ported from 4.4-Lite is not
> 	contaminated (because 4.4-Lite had the legal blessings of AT&T)
> 	but I was told at one time anything based on the Net-2 stuff could be
> 	(is?) contaminated.  Alas by the time 4.4-Lite came out the software
> 	had bloated so much that very little of it can be ported over.  I 
> 	grabbed a few ideas and pieces out of the kernel - that's where the 
> 	"sysctl" stuff in 2.11 came from for example.  But the mainline
> 	applications are GNU based (megabytes and megabytes of memory assumed).
> 	I'd like to see someone getting GCC to run natively on a PDP-11! 
> 
> 	Steven

Just a thought: much of the stuff in 16-bit Minix was written by people
on Usenet and donated to Minix. The core stuff of course is owned by
Prentice-Hall, but there are some freely-available programs.

	Warren

Date: Tue, 5 Jan 1999 16:42:41 +1030
From: Greg Lehey < grog@lemis.com>
To: "Steven M. Schultz" < sms@moe.2bsd.com>
Cc: wkt@cs.adfa.oz.au, pups@minnie.cs.adfa.oz.au
Subject: Re: Why is csh `restricted'?
Message-ID: <19990105164241.E78349@freebie.lemis.com>
References: <199901050557.VAA19580@moe.2bsd.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
In-Reply-To: <199901050557.VAA19580@moe.2bsd.com>; from Steven M. Schultz on Mon, Jan 04, 1999 at 09:57:01PM -0800
WWW-Home-Page: http://www.lemis.com/~grog
Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia
Phone: +61-8-8388-8286
Fax: +61-8-8388-8725
Mobile: +61-41-739-7062
Sender: owner-pups@minnie.cs.adfa.edu.au
Precedence: bulk

On Monday,  4 January 1999 at 21:57:01 -0800, Steven M. Schultz wrote:
>> From: Warren Toomey < wkt@henry.cs.adfa.edu.au>
>> What else in the original 2bsd is contaminated?
>
> 	Anything that I (or other contributors) didn't write ourselves.
>
> 	A good case can be made that stuff ported from 4.4-Lite is not
> 	contaminated (because 4.4-Lite had the legal blessings of AT&T)
> 	but I was told at one time anything based on the Net-2 stuff could be
> 	(is?) contaminated.

There has been a lot of confusion on this point.  Well, maybe
``disagreement'' would be a better word.  Obviously Net-2 contained
almost only stuff written by contributors, though there was, indeed,
some code which had obviously grown out of Seventh Edition code.  I
think somebody mentioned something like 13 files in the context of the
lawsuit.  I took a look at one (kern_clock.c?), and confirmed that
yes, it looked as if it was derived rather than written from scratch.
On the other hand, there was nothing which AT&T (or the opponent of
the week) could claim to be trade secrets.  And IMO none of this could
have been construed to mean that people couldn't use the sources which
were indisputably completely written by UCB and its contributors.

Greg
--
See complete headers for address, home page and phone numbers
finger grog@lemis.com for PGP public key

Date: Mon, 4 Jan 1999 22:45:35 -0800 (PST)
From: "Steven M. Schultz" < sms@moe.2bsd.com>
Message-Id: <199901050645.WAA19977@moe.2bsd.com>
To: pups@minnie.cs.adfa.oz.au
Subject: Re: Why is csh `restricted'?
Sender: owner-pups@minnie.cs.adfa.edu.au
Precedence: bulk

Greg -

> From: Greg Lehey < grog@lemis.com>
> 
> There has been a lot of confusion on this point.  Well, maybe
> ``disagreement'' would be a better word.  Obviously Net-2 contained

	Hmmm, I think `confusion' is a better fit.  Of course said confusion
	does lead to disagreement eventually ;)

> think somebody mentioned something like 13 files in the context of the

	I'd heard it was 7 files at one time, then 11.  It's a fairly
	small number _but_ the exact list was never disclosed (part of the
	settlement I understand).  Without a list of files the fear (at the
	time) was that "the enemy" could come after you claiming derivation
	of some work from the forbidden files.  Since you didn't know what
	files those were it was hard (impossible) to know what you could or
	couldn't use.

> the week) could claim to be trade secrets.  And IMO none of this could
> have been construed to mean that people couldn't use the sources which
> were indisputably completely written by UCB and its contributors.

	I'm not a lawyer (and don't even play one on the Net;))...  That's
	how you and I (nonlawyer types) think.  The sentiment at the time
	was that up until 4.4-Lite was declared "uncontaminated" there was
	a danger of being legally targeted for using Net-1 and Net-2.

	The point is moot now today because all manner of alternatives
	(FreeBSD for example) exist.  That ready availability may have been 
	a big factor in SCO's allowing inexpensive access to the "original" 
	sources (albeit under 'license' rather than "freely available").

	Steven

Date: Tue, 5 Jan 1999 17:46:39 +1030
From: Greg Lehey < grog@lemis.com>
To: "Steven M. Schultz" < sms@moe.2bsd.com>
Cc: pups@minnie.cs.adfa.oz.au
Subject: Re: Why is csh `restricted'?
Message-ID: <19990105174639.G78349@freebie.lemis.com>
References: <199901050645.WAA19977@moe.2bsd.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
In-Reply-To: <199901050645.WAA19977@moe.2bsd.com>; from Steven M. Schultz on Mon, Jan 04, 1999 at 10:45:35PM -0800
WWW-Home-Page: http://www.lemis.com/~grog
Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia
Phone: +61-8-8388-8286
Fax: +61-8-8388-8725
Mobile: +61-41-739-7062
Sender: owner-pups@minnie.cs.adfa.edu.au
Precedence: bulk

On Monday,  4 January 1999 at 22:45:35 -0800, Steven M. Schultz wrote:
> Greg -
>
>> From: Greg Lehey < grog@lemis.com>
>>
>> There has been a lot of confusion on this point.  Well, maybe
>> ``disagreement'' would be a better word.  Obviously Net-2 contained
>
> 	Hmmm, I think `confusion' is a better fit.  Of course said confusion
> 	does lead to disagreement eventually ;)

We can agree (or is that defuse?) about that.

>> think somebody mentioned something like 13 files in the context of the
>
> 	I'd heard it was 7 files at one time, then 11.  It's a fairly
> 	small number _but_ the exact list was never disclosed (part of the
> 	settlement I understand).  Without a list of files the fear (at the
> 	time) was that "the enemy" could come after you claiming derivation
> 	of some work from the forbidden files.  Since you didn't know what
> 	files those were it was hard (impossible) to know what you could or
> 	couldn't use.

Well, here's an extract from BSDI's announcement dated 8 Feb 1994:

> This broadcast message addresses many of the questions that have arrived
> in my mailbox in the last few days.
>
> Q:  After this lawsuit resolution, is BSDI still in business?
> A:  You bet.  And we're shipping 1.1 early next week.
>
> Q:  The press release was unclear, do I get to keep my current copy
>    of BSD/386?
> A:  The answer is yes!  BSDI is not recalling prior versions.
>    Any USA domestic customer whose support was valid through December,
>    1993 will be shipped the new V1.1 release.  I will be mailing a paper
>    letter to each USA domestic customer detailing their service contract
>    status and verifying the V1.1 shipping address.
>
> Q:  What's all this about `binary-only files'?  Will BSDI continue to
>    ship source code?
> A:  For Version 1.1 only, BSDI will ship the following kernel files
>    in binary format:
>
>	kern/init_main.c	kern/subr_rmap.c	ufs/ufs_bmap.c
>	kern/kern_clock.c	kern/sys_generic.c	ufs/ufs_disksubr.c
>	kern/kern_exit.c	kern/sys_process.c	ufs/ufs_inode.c
>	kern/kern_physio.c	kern/tty.c		ufs/ufs_vnops.c
>	kern/kern_sig.c		kern/tty_subr.c
>	kern/kern_synch.c	kern/vfs_syscalls.c

OK, so it was 16, not 13.  And yes, they didn't say that these were
the ones, but I did look at one and saw the similarities.

> Q:  I noticed your signature changed.  Did you get promoted?
> A:  Yes, we now have a full-time president.  Me!
>
>					Rob Kolstad
>					President, BSDI

Well, some things just keep changing.

Greg
--
See complete headers for address, home page and phone numbers
finger grog@lemis.com for PGP public key