Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!spool.mu.edu!howland.erols.net!news.sprintlink.net!news-peer.sprintlink.net!newsfeed.internetmci.com!in3.uu.net!news.artisoft.com!usenet
From: Terry Lambert <terry@lambert.org>
Newsgroups: comp.unix.bsd.misc
Subject: Re: Old BSD Source Code
Date: Sun, 29 Sep 1996 16:21:05 -0700
Organization: Me
Lines: 113
Message-ID: <324F0461.79D91FDA@lambert.org>
References: <52129f$ogf@Venus.mcs.com> <19960922121710134506@pppx231.berlin.snafu.de> <AWB.96Sep25081534@margo.cstr.ed.ac.uk> <52kjah$r4r@herald.concentric.net>
NNTP-Posting-Host: hecate.artisoft.com
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 2.01 (X11; I; Linux 1.1.76 i486)
Daniel Ts'o wrote:
]
] In article <AWB.96Sep25081534@margo.cstr.ed.ac.uk>,
] awb@cstr.ed.ac.uk (Alan W Black) wrote:
] :Technically the released source before 4.4-lite has not been shown to
] :contain proprietary code as it has never been tested in court. In
] :fact most people would go further and say that there wasn't any
] :proprietary code in the released versions, and that may be the reason
] :USL (who owned Unix at the time) never actually went to court.
]
] I'm not sure exactly what you are claiming here.
He is claiming that there were no USL trade secrets embodied
in the BSD code, since they had been previously disclosed and
were therfore not secret.
] Why is there any question whether USL-owned UNIX code is
] "propriety"?
There isn't the question is whether the USL-owned code in
the Net/1 and Net/2 releases was held "proprietary" -- ie: if
it was released in violation of the Western Electric license.
] You aren't contesting that AT&T licensed the source to UCB,
] limiting its distribution, are you ?"
If he isn't, I am. The Western Electric license to UCB was
never updated by UCB *precisely* because the newer license
would have imposed distribution restrictions which would have
prevented joint projects with other licensees.
If you will remember your history, the telco was specifically
enjoined from making profit on software as part of the
first consent decree as a regulated monopoly under the
Sherman antitrust acts. When this restriction was eased,
they quickly changed the license, but UCB did not relicense
because of the additional restrictions, and was therfore
not bound by the new license. Inconvenient for Western
Electric, but all perfectly legal.
A similar license was in force at UNSW (University of New
South Wales) at the time of their publication of the V7
source code in book form by the University, as part of the
course materials for a class. Western Electric got UNSW to
"reup" their license, and subsequnt publication carried the
Western Electric copyright, and the restrictions that you
mistakenly believe were in force in the Western Electric/UCB
agreement, or at the time of initial publication.
] The Unix source license was actually not based on
] copyright nor patent, but trade secret.
During the USL/UCB lawsuit, Dennis Ritchie offered to witness
for the defense that the code did not embody any trade secrets.
I have his posting on tape in case it ever comes down to an
argument in court again.
For what it's worth, I was a kernel hacker at Novell before
and after the Novell/USL purchase, and later worked in Novell
USG (UNIX Systems Group), "the former USL". I made the same
offer, as did a number of other Novell USG employees, who felt
that the lawsuit was without any technical merit. Trade secrecy
is a technically-based distinction.
I categorically deny that the Net/1 and Net/2 code releases
embody any trade secrets which were not already publically
disclosed by prior publication and/or legal for UCB/UNSW
to disclose under the terms of the license in force at the
time of the disclosure.
The withdrawl of the Net/1 and Net/2 sources by UCB must, by
definition of the license and the lack of termination conditions,
be considered as a voluntary act of appeasement to end the legal
battle, not as something which they were legally required to do.
That is, in fact, why the Net/1 and Net/2 code is still available
from many places -- as is FreeBSD 1.x and 386BSD 0.1, both Net/2
derived source bases. The distinguishing feature of the sites
that carry it is their ability to withstand legal clubbing long
enough to club back, and so they are immune to intimidation
tactics (MIT was prepared to defend the UCB code base if
necessary -- MIT has more money than most small countries because
of royalties, etc. [billions, in fact]).
To complicate matters, when it became clear that USL might extend
their suit against BSDI to include the free OS implementations,
the source for those implementations was moved "off shore" --
to non-Berne signatory countries, and countries without copyright
or patent treaties with the US. USL had failed before they
began extending their cloud; there is no way to enforce a border
check for trade secrets. You can see the legacy of this in the
fact that all of the crypto code is imported from South Africa
rather than exported from the US. Everyone who wants crypto
code anywhere in the world where it is legal for an individual
to posess it, can get it without violating US ITAR restrictions.
The ITAR restrictions are useless because the code did not
originate in the US.
In closing, Western Electric *purchased* a UNIX following with
those trade secrets as coin. USL was not free to attempt to
reclaim the Western Electric coin simply because it had suddenly
increased in value. Once spent, the coin was spent.
Regards,
Terry Lambert
terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.