This is part of our article series published as “History of FreeBSD“. Subscribe to our article series to find out more about the beginnings of FreeBSD
In this second part of our series on the history of FreeBSD, we continue to trace the pre-history of FreeBSD and the events that would eventually shape the project and the future of open source software.
386BSD Becomes Popular
The initial release of 386BSD worked but was barely functional. So, Bill and Lynne Jolitz started working on creating a full release. They began by removing and replacing the pieces that had unsatisfactory licenses or no longer worked the way they should. The latter were pieces of code that had been implemented years earlier using designed that were archaic at this point.
They also removed tools that duplicated functionality. According to Bill, “The Unix systems then offered eight or 10 different competing mechanisms to do basically the same thing. The question in 386BSD was, How many of those could we get rid of? We increased the capabilities and reduced the size by a factor of 35.“
When it was released, 386BSD 0.1 was a hit. The Jolitz’s made it available via a public ftp server and only expected to receive a couple of hundred downloads. Instead, they got over 250,000. This publicity was due in part to an article in Dr. Dobb’s Journal, a famous early computer magazine. It even caught the eye of Linux creator Linus Torvalds. Linus told Meta Magazine in 1993, “If 386BSD had been available when I started on Linux, Linux would probably never had happened.“
Time to Commercialise BSD
Around the same time, someone decided it was time to create a company to promote and sell a commercial version of BSD. Thus Berkeley Software Design, Incorporated or BSDi came into being. It was founded by Rick Adams (founder of one of the largest internet service providers of the 90s, UUNET) and members of Berkeley’s Computer Systems Research Group (CSRG). This included Keith Bostic, Kirk McKusick, Mike Karels, Bill Jolitz, and Donn Seeley.
BSDi started selling their commercial version of BSD in January of 1992 after receiving permission from UC Berkeley. It was originally called BSD/386, but later changed to BSD/OS. Anyone could get both the source and the binaries for $995. This was a 99% discount on the price AT&T charged for their System V Unix. They even set up an easy to remember phone number for people who wanted to order copies: 1-800-ITS-Unix.
The AT&T Part of The Story
In the late 1950s, AT&T was forced to accept a consent decree from the US government to end an anti-trust lawsuit. As part of that consent decree, AT&T had to limit its business endeavours to its national telephone system and special projects for the federal government. Once educational institutions became aware of AT&T’s Unix system in the 1970s, they requested access to it for their computer labs. Legally AT&T couldn’t be in the computer business. So, AT&T’s lawyers came up with a workaround. They would license Unix to universities, but they would operate under a “no advertising, no support, no bug fixes, payment in advance” plan.
Since the educational institutions that licensed Unix couldn’t expect support from the creators of the system, many of them shared bug fixes and improvements. Thus the Berkeley Software Distribution (BSD) was born.
In the early 1990s, the US government broke up AT&T and removed the consent decree. Now AT&T could move into other business ventures. UNIX System Laboratories or USL was one of those ventures. This subsidiary was created solely to develop and sell Unix. When USL caught wind of BSDi’s marketing strategy, their lawyers jumped into action. They sent a letter demanding that BSDi do two things: drop the 1-800-ITS-Unix phone number and make it clear in the advertisements that BSDi’s product was not Unix. There are differing accounts as to whether or, not they got rid of the phone number, but they did fulfil the second part of USL’s demand.
USL was still not content with the fact that BSDi was selling a competing product, so they decided to take them to court. The suit filed by USL alleged that BSDi’s product included code and trade secrets that belonged to USL. They also asked for an injunction to prevent BSDi from selling their product until the lawsuit had been resolved because it could damage USL.
At the pre-trial hearing, BSDi pointed out that they were using code from the University of California, Berkeley, except for six kernel files that had been written by Bill Jolitz to replace copyrighted code. They felt that USL couldn’t sue them over freely available code. The judge agreed and told USL “restate their complaint based solely on the six files or he would dismiss it”. USL decided instead to add the University of California at Berkeley to the suit and refile it. They also added an injunction to halt the shipment of Networking Release 2 by the University. Towards the end of 1992, Judge Dickinson R. Debevoise from New Jersey heard the arguments for the injunction. After deliberating 6 weeks, he denied the injunction and threw out all but two complaints.
The University of California filed a counter-suit against USL in California district court. This counter-suit pointed out that USL’s product System V included code from BSD and they had failed to credit the University per the license USL had agreed to. If their complaint was judged to be correct, the University requested that “USL be forced to reprint all their documentation with the appropriate due credit added, to notify all their licensees of their oversight, and to run full-page advertisements in major publications such as *The Wall Street Journal* and *Fortune* magazine notifying the business world of their inadvertent oversight”.
Before the court could make a ruling. USL was purchased from AT&T by Novell. The CEO of Novell said that he would prefer to compete in the marketplace than in court. In 1993, the two sides opted for settlement talks instead. However, progress was slow because neither side wanted to budge. Finally, they reached an agreement in early 1994. The terms of the agreement were secret for many years, but were eventually released under the California Public Records law.
Only three files had to be removed from Networking Release 2 and changes were made to other files. The University of California added USL copyrights to 70 files. Even with the new copyrights, the files would be free to redistribute.
After the Lawsuit
Even though the supporters of a free and open-source BSD were victorious, the UNIX System Laboratories, Inc. v. Berkeley Software Design, Inc. lawsuit caused immense damage. It effectively stalled development on BSD for several years. No one wanted to risk getting pulled into the lawsuit. Many of the people who would have contributed to BSD instead contributed code and bug fixes to Linux.
According to Jordan Huddard (one of the creators of FreeBSD), this episode “effectively ceded the territory to Linux during a critical period in which there was a leadership vacuum and lots of Fear, Uncertainty and Doubt in the BSD community”.
BSDi continued to sell commercial BSD services. In 2000, they were purchased by Walnut Creek (the operators of CDROM.com and parent company for Slackware). The following year, BSDi sold its software business to Wind River Systems, and the remaining company was then renamed to iXsystems.
Like this article? Share it!
Pingback: ● NEWS ● #KlaraSystems #Licensing ☞ History of FreeBSD: Part 2: BSD… | Dr. Roy Schestowitz (罗伊)
Pingback: Valuable News – 2020/10/26 | 𝚟𝚎𝚛𝚖𝚊𝚍𝚎𝚗
Pingback: FreeBSD and ZFS Curated Content for the Holidays | Klara Inc.
Pingback: History of FreeBSD – Net/1 and Net/2 – A Path to Freedom | Klara Inc.
Pingback: The Birth of UNIX | Klara Inc.