USL v. BSDi was a lawsuit brought in 1992 by UNIX Systems Laboratories against Berkeley Software Design, Inc and the Regents of the University of California over intellectual property related to UNIX. The case was settled out of court 1993 after the judge expressed doubt in the validity of USL's IP, with USL and BSDi mutually agreeing not to litigate further about the majority of the software that would later be developed into the free BSDs.

A similar case, SCO v. IBM, is currently pending, where SCO claims to be USL's successor in interest.

Background

The suit has its roots at the Computer Sciences Research Group (CSRG) at the University of California, Berkeley, which had a license for the source code of UNIX from AT&T's Bell Labs. Students doing Operating systems research at the CSRG modified and extended UNIX over the course of several years. With AT&T's blessing, the CSRG made the first of several releases of the modified operating system in 1978 -- the Berkeley Software Distribution. Because the system contained copyrighted AT&T source code it was only available to other organizations with a source code license for UNIX.

Later, students and faculty would audit the software code for the TCP/IP stack, removing all the AT&T material, and license it to the public in 1988 as NET-1 under the BSD license. When it became apparent that the Berkeley CSRG would soon close, students and faculty began an effort to remove all the remaining AT&T code from the BSD and replace it with their own, an effort that resulted in the 1991 release of NET-2 to the public under the BSD license. NET-2 contained code for nearly an entire UNIX-like system, which the CSRG believed contained no AT&T copyrighted software.

BSDi obtained a copy of NET-2 and filled in the missing pieces as well as porting it to the Intel i386 computer architecture. BSDi then sold the resulting BSD/386 operating system. This drew the ire of AT&T, whose Unix System Labs subsidiary filed suit against BSDi in New Jersey in April of 1990. The suit was later amended to include The Regents of the University of California.

USL's complaint

In the lawsuit, UNIX System Labs would allege that:

• The Regents of the University of California, by releasing NET-2 "based upon, substantially copied from, or derived from proprietary UNIX":
  • Had violated USL's software license with the UC Berkeley .
  • Had violated USL's copyright on UNIX.
  • Had divulged USL's trade secrets by making public "the methods or concepts used therein".
• BSDi had known the UC Berkeley had no right to release the NET-2 source code (and had in fact induced the University into releasing the code), so distributing the BSD/386 code based on it meant BSDi had knowingly violated USL's copyright.
• The UC Berkeley's claim in the announcement of the release of NET-2 that it "requires no previous license [...] from AT&T" was false, and an illegal deceptive trade practice.
• BSDi's claim in their advertising and software license that BSD/386 and the NET-2 code it was derived from "contained no AT&T licensed code" was false, so BSDi was guilty of false advertising and deceptive trade practices.
• BSDi's 1-800-ITS-UNIX telephone number violated USL's trademark on UNIX.

On these grounds, USL asked the court for a preliminary injunction that would bar BSDi and the UC Berkeley from distributing the NET-2 software until the outcome of the case was known.

Many of the trial documents of this case are sealed or unavailable, including the majority of those submitted by USL. Some of those that are available have had portions removed as a term of the case settlement.

Pretrial

Since the allegedly infringing software had been released to the public by the UC Berkeley, most of the case would hinge on the events there. Because UC Berkeley was not originally a party to the suit, the University made its arguments against an injuction in a series of amicus briefs.

The University submitted the licenses UC Berkeley had from AT&T for UNIX, which specifically stated that the copyright on the Berkeley software built upon the 32V version of UNIX belonged to the University. They went on to claim that AT&T had confirmed this by allowing the free redistribution of NET-1, as well as allowing the distribution of the later BSDs, in which specific source code files were marked as containing no AT&T code and so freely redistributable. Since allowing this distribution was legally considered abandoning the copyright, that code could not have been considered a copyrighted part of UNIX even if AT&T had not explicity said it wasn't.

Assuming this was true, the University would still have to show that NET-2 did not contain any validly copyrighted AT&T UNIX code. One claim the University would make was that USL's copyright in the 32V version of UNIX that NET/2 was based on was invalid. At the time 32V was released, US copyright law did not automatically presume that a released work was copyrighted. In order to claim copyright, it was necessary to include copyright notices in the work -- which 32V did not have -- or to register the work with the government. Because AT&T didn't register the copyright on 32V UNIX until 1992, the grace period for registering an already-released work had expired.

The University also claimed that identical and similar lines of source code (which were presented during discovery) did not infringe on USL's copyright because they had become public domain by AT&T's own actions. AT&T had promoted UNIX as a standard, licensing it to universities and allowing UNIX source code to be published and documented in books. The University submitted briefs from the UC Berkeley students and staff, explaining how they'd audited the code by looking for freely available copies of the source code and methods. When they could find none, they said, they removed the code and rewrote it using publicly known techniques -- and so any remaining similarities existed because USL's had effectively abandoned the copyright to them.

The University would also argue that the source code did not infringe because it was necessary for program compatibility -- that certain code could be written only one way and still be compatible with the standards set by organizations like POSIX and AT&T itself, and so was no longer a "creative" work that could receive copyright protection under the law.

Even if the code were validly copyrighted AT&T UNIX code, the University claimed, that would not be a copyright violation because it made up such a small fraction of the whole of NET-2 that it could not legally be considered a derived work.

Terms of the settlement

Until recently, the details of the settlement have been kept secret between the parties, with the general public consensus being that USL and BSDi had mutually agreed not to litigate further about the majority of the software that would later be developed into the free BSDs, after the judge had expressed doubts about the validity of USL's copyrights. In November 2004, a copy of what appears to be the USL v. BSDi settlement agreement was posted to the Groklaw website. This appears to have been obtained from The Regents of the University of California's Office of the General Counsel under the State of California Public Records Law. If this is confirmed to be genuine, this will be the first time that this crucial link in UNIX legal history will have been made public.

to do: summarize the settlement terms

External links

• Court documents from USL v. BSDi (http://cm.bell-labs.com/cm/cs/who/dmr/bsdi/bsdisuit.htm)
• UNIX History timeline (http://www.levenez.com/unix/)
• Berkeley Licenses with AT&T (http://members.cox.net/docs3wilkerson/index.html)
• What appears to be a copy of the USL v. BSDi settlement agreement (http://www.groklaw.net/pdf/USLsettlement.pdf) (Adobe .pdf format), text version (http://www.groklaw.net/article.php?story=20041126130302760)