Here is chapter 13 of our ongoing history of Free and Open Source, The Daemon, the GNU and the Penguin, by Dr. Peter H. Salus.
Here are the earlier chapters:
If you wish to review any of the legal documents he references, our Legal Docs page has a section on BSDI documents, including the long-secret 1994 USL-Regents of California Settlement Agreement.
Judge Dickinson R. Debevoise's 1993 Opinion in USL v. BSDI on the parties' various motions to dismiss/amend and USL's motion for preliminary injunction is here, which found, among other things, the following regarding copyright and trade secrets:
Consequently, I find that Plaintiff has failed to
demonstrate a likelihood that it can successfully defend its
copyright in 32V. Plaintiff's claims of copyright violations are
not a basis for injunctive relief. . . .
After reviewing the affidavits of Plaintiff's and
Defendants, experts, a great deal of uncertainty remains as to what
trade secrets Net2 might contain. One fact does seem clear: the
header files, filenames, and function names used by Defendants are
not trade secrets. Defendants could have printed these off of any
of the thousands of unrestricted copies of Plaintiff's binary
object code. (Kashtan Aff. at 9-11.) Moreover, the nonfunctional
elements of the code, such as comments, cannot be trade secrets
because these elements are minimal and confer no competitive
advantage on Defendants. The copied elements that contain
instructions, such as BREAD and CPIO, might perhaps be trade
secrets, but Defendants' experts have argued persuasively that
these instructions are either in the public domain or otherwise
exempt. As Defendants have repeatedly emphasized, much of 32V
seems to be publicly available. . . .
Thus, even if all of the pieces of the 32V code
had been thoroughly revealed in publicly available literature, the
overall organization of the code might remain a trade secret unless
it too had been disclosed.
On the present record, however, it is impossible to
determine whether the overall organization of Net2 has been
disclosed. The record itself contains little information directly
pertinent to this issue. Moreover, the parties' submissions hint
that some of 32V's organization may already be publicly available.
. . . .
A further consideration is that 32V's overall
organization may not even be protectable in the first place.
Berkeley's license to use 32V protects 32V derivatives only to the
extent that they contain certain proprietary information. If
Berkeley excises the proprietary information (as it attempted to do
with Net2), Berkeley is free to distribute derivatives without
restriction. Berkeley has utilized this freedom in the past to
distribute a number of non-proprietary systems and portions of
systems, all apparently without objections from AT&T. These
distributions, to some degree, must have disclosed the overall
organization of 32V. Thus, Berkeley's activities under the
licensing agreement, and AT&T's acceptance of those activities, are
evidence that Berkeley and AT&T interpreted the agreement to allow
the disclosure of at least some of 32V's organization.
In summary, I find that I am unable to ascertain whether
any aspect of Net2 or BSD/386, be it an individual line of code or
the overall system organization, deserves protection as Plaintiff's
trade secret. Since Plaintiff has failed to provide enough
evidence to establish a "reasonable probability" that Net2 or
BSD/386 contain trade secrets, I find that Plaintiff has failed to
demonstrate a likelihood of success on the merits of its claim for
misappropriation of trade secrets. No preliminary injunction will
The Daemon, the GNU and the Penguin
~ by Peter H. Salus
Chapter 13. USL v The Regents of the University of California
From late 1986 on, Keith Bostic would stand up at each USENIX Conference
and announce the progress of his -- the CSRG's -- project: about 35%
of the code was AT&T license free; about 55%; about 77%... The progress
may have seemed slow at times, but the was always some progress.
And there were always loud cheers and resounding applause.
AT&T's lawyers had started off on the wrong foot in the mid-1970s;
the fee structure made AT&T-license-free UNIX a financial necessity.
Among others, John Gilmore nudged the CSRG to produce a license-free
version. After all, it was clear that AT&T hadn't objected to Minix,
which was a UNIX-like system with no AT&T code.
In November 1988, at the BSD Workshop in Berkeley, Keith, Mike Karels
and Kirk McKusick announce the completion and availability of BSD
Networking Release 1.
NET 1 was a subset of the then-current Berkeley system. It was
quite similar to 4.3-Tahoe, including source code and documentation
for the networking portions of the kernel, the C library and utility
programs. It was available without evidence of any prior license
(AT&T or Berkeley), and was (re)distributed via anonymous FTP. The
source carried a Berkeley copyright notice and a legend that
allowed redistribution with attribution.
(The Berkeley license was, and still is, different from the GPL.
Keith and rms had debated the various aspects of the licenses
repeatedly, without convergence. I will discuss this later.)
In June 1991, at the USENIX Conference in Nashville, BSD
Networking Release 2 was available. NET 2 contained far more
than just networking code and, like NET 1, was available with
no prior license. The new features included a new virtual
memory system (derived from Carnegie-Mellon's Mach system, which
had been ported at the University of Utah) and a port to the
It was nearly the end of the line for the CSRG. Karels left
Berkeley in 1992; Bostic and McKusick, in June 1993. But,
NET 2 was a US-Russia collaboration, with contributions by Bill
Jolitz, Donn Seeley, Trent Hein, Vadim Antonov, Mike Karels,
Igor Belchinsky, Pace Willisson, Jeff Polk, and Tony Sanders. It
was turned into a commercial product, known as BSDI (= Berkeley
Software Design, Inc.) and was complete by the end of 1991
and released to the public on April 10, 1993 as 1.0, the long
delay being the consequence of UNIX Systems Laboratories (USL)
filing suit to prevent BSDI from shipping its product.
USL filed for an injunction barring distribution of "BSD/386,
pending trial, arguing that BSD/386 infringed
USL's copyright in its 32V software and misappropriated
USL trade secrets."1 The Court denied USL's request for a
preliminary injunction on March 3, 1993, ruling that USL had
"failed to show a likelihood of success on either its copyright
claim or its trade secret claim."
On March 30, 1993, Judge Dickinson Debevoise of the US District
Court of New Jersey reaffirmed his denial of USL's motion for a
preliminary injunction against BSDI. The Court found that the 32V
source code had been distributed without a copyright notice. The
Court rejected USL's argument that the publication of 32V was
sufficiently limited to avoid a forfeiture, and thus found that
USL had failed to establish that BSD/386 contained any USL
USL subsequently filed a motion for reconsideration, asking the
District Court to hold a new hearing on whether USL had published
32V without a copyright notice. USL argued that the Court's
prior ruling was based on an incorrect finding as to the number
of copies distributed. (USL's motion for reconsideration did
not challenge the Court's ruling that USL had failed to establish
trade secret misappropriation.)
The Court denied USL's motion for reconsideration. Although the
Court amended its prior factual finding as to the number of copies
distributed, it found that the number was not critical to its ruling
on the issue of publication without notice.
It was just under 20 years since Ken had delivered that first
UNIX paper at SOSP and began receiving requests for the software.
It was 15 years since UNIX NEWS became ;login: and
the "UNIX Users Group" turned into USENIX.
And, through all of this, Western Electric, AT&T, and now USL
had learned nothing about the nature of the user community.
What BSDI (and others, like mt Xinu) were trying to do was ensure
the continued development, growth and use of the UNIX operating
system. The suit by USL was an attempt to protect something of
value. But the value had been discovered too late. Maybe Ritchie and
Thompson had handled the system carelessly in the mid-1970s; maybe
BTL employees inadvertently gave UNIX to the public without any
significant restriction. But the users, like O'Dell and Kolstad,
Coulouris and Lions, saw the value.
BSDI had distributed pre-production releases of BSD/386 (Beta
version). It now began international distribution. Full source
was priced at $1000. (In the January/February 1994 ;login:,
Lou Katz wrote: "It works! It works!").
Because of the way the licenses had been granted to the University
of California, the Regents of the University had been included in
USL's suit. In June 1993, the Regents struck back, filing suit
against USL for not complying with agreements made with the CSRG.
In the meantime, Novell acquired USL. A decade later, it is still
unclear exactly what USL believed it was selling, or what Novell
thought it was buying.
To take but one example, the "PROPRIETARY NOTICE" on the System V
errno.h file reads:
This source code is unpublished proprietary information
constituting, or derived under license from AT&T's UNIX (r)
System V. In addition, portions of such source code were
derived from Berkeley 4.3 BSD under license from the Regents
of the University of California.
USL had filed suit in New Jersey, the home of AT&T and Bell Labs
(and USL), clearly thinking that this would yield an advantage
in their suit. When the USL v BSDI and Regents suit was
dismissed, the Regents filed suit in California.
On Friday, February 4, 1994, Novell and the University of California
agree to drop all relevant suits and countersuits. BSDI immediately
announced the availability of a release based on "4.4BSD-Lite."
"We are delighted to settle with USL so that we can devote our
full efforts to creating products and serving our customers," Rob
Kolstad, president of BSDI, said to me. (I will not go into the
corporate history of BSDI here.)
There is no question that The Open Group "owns" the UNIX trademark.
There are many questions as to whether any one entity "owns" UNIX.
In 1995, Novell executed an "Asset Purchase Agreement" with the
Santa Cruz Operation, though exactly what was purchased is
unclear a decade later.
While both The SCO Group (a successor to the Santa Cruz Operation)
and Novell claim rights to the UNIX
source code, many elements of that code carry copyright notices
from the University of California and other companies and
All of this is important because of the various suits filed by
The SCO Group and the ownership claims made. It is clear to me
that while AT&T owned a good portion of several versions of
System V, as a result of the incorporation of code from the
CSRG and other sources, they never had full rights to all the
source of any of the versions. Exactly what was conveyed to
Novell, from Novell to the Santa Cruz Operation, and from
the latter to The SCO Group, may be adjudicated in one of the
1 The story of 32V is in Chapter 6,
Dr. Salus is the author of "A Quarter Century of UNIX" and several other books, including "HPL: Little Languages and Tools", "Big Book of Ipv6 Addressing Rfcs", "Handbook of Programming Languages (HPL): Imperative Programming Languages", "Casting the Net: From ARPANET to INTERNET and Beyond", and "The Handbook of Programming Languages (HPL): Functional, Concurrent and Logic Programming Languages". There is an interview with him, audio and video,"codebytes: A History of UNIX and UNIX Licences" which was done in 2001 at a USENIX conference. Dr. Salus has served as Executive Director of the USENIX Association.
This work is licensed under the Creative Commons
Attribution-NonCommercial-NoDerivs License. To view
a copy of this license, visit
or send a letter to Creative Commons, 559 Nathan Abbott Way,
Stanford, California 94305, USA.