by Peter H. Salus
<[email protected]>

Peter H. Salus is the author of A Quarter Century of UNIX (1994) and Casting the Net (1995).

20 Years Ago in U[SE]NIX

The January 1979 (Santa Monica) meeting had been a success: 350 attendees. The June meeting was scheduled for Toronto. The planning was underway.

First, there was an announcement of the "UNIX Users Group Conference" (the long arm of the AT&T lawyers hadn't reached to Queens Park or St. Joseph Street yet), to be held at the University of Toronto, "Wednesday, June 20, 1979 through Saturday, June 23, 1979," with "Registration: Tuesday evening, June 19, 1979, 7:00PM - 9:00PM."

Second, there was an announcement of the "Software Tools User Group Meeting," to be held "June 19, the day before the general UNIX users meeting."

I'll write about the meeting(s) in June. But in many ways, 1979 was to be an important and eventful one for UNIX. Toward the end of 1978, an early version of 32V -- the port to the VAX by Charlie Roberts's group in Holmdel -- made its way from New Jersey to California. The cohort at the CSRG immediately began working on turning it into 3BSD. At the same time, the group at Bell Labs were turning V6 and 32V into V7.

By the time of its release, V7 was a truly wonderful system. Less than two years after DEC announced the VAX, here was a 32-bit OS (32V) and a new and updated version of UNIX for the PDP-11. Moreover, it contained Steve Bourne's new shell, as well as grep, uucp, awk, lex, lint, etc. The V7 manual was also the first to be commercially published: I still have my copy of the Holt, Rinehart and Winston printing of 1979.

V7 also served as input to 3BSD. In fact, when it appeared, 3BSD was everything that Bill Joy had said he wanted Berkeley's "product" to be.

Bill had said that he was tired of doing releases that were made up of utilities and decided he wanted to release systems. 32V and V7 (which Berkeley had gotten in January) were the material he worked with.

3BSD was a complete, bootable system. It had a bootblock at the beginning of the tape, so you could roll it onto raw hardware. 3BSD had a virtual-memory-based kernel, and all the utilities had been booted across. Perhaps more importantly, if you wanted to run UNIX with a paging system, you had to run 3BSD.

One of the people who wanted to run UNIX was Brian Harvey. In January 1979, Brian went to Lincoln-Sudbury Regional High School and persuaded the school board to float a bond issue for computer equipment. Brian's 15-year-olds ran UNIX on a PDP-11.

Also in early 1979, Jim Kulp at the IIASAS in Laxenburg, Austria, bought a VAX/780 and ran 3BSD on it.

Commercialization had also begun. In 1977 there was Interactive Systems; in 1978, P.J. Plauger's Whitesmiths compiler and Idris, the first UNIX clone. Now, in 1979, with UNIX looking forward to its 10th birthday, there came XENIX -- a collaboration between Microsoft and the Santa Cruz Operation -- the first UNIX implementation for the Intel 8086.

There was going to be a lot to talk about in Toronto.