Ironically, the award also made it possible for Stallman to vote. Months before the award, a fire in Stallman’s apartment house had consumed his few earthly possessions. By the time of the award, Stallman was listing himself as a “squatter”[12] at 545 Technology Square. “[The registrar of voters] didn’t want to accept that as my address”, Stallman would later recall. “A newspaper article about the MacArthur grant said that and then they let me register”.[13]
Most importantly, the MacArthur money gave Stallman more freedom. Already dedicated to the issue of software freedom, Stallman chose to use the additional freedom to increase his travels in support of the GNU Project mission.
Interestingly, the ultimate success of the GNU Project and the free software movement in general would stem from one of these trips. In 1990, Stallman paid a visit to the Polytechnic University in Helsinki, Finland. Among the audience members was 21-year-old Linus Torvalds, future developer of the Linux kernel-the free software kernel destined to fill the GNU Project’s most sizable gap.
A student at the nearby University of Helsinki at the time, Torvalds regarded Stallman with bemusement. “I saw, for the first time in my life, the stereotypical long-haired, bearded hacker type”, recalls Torvalds in his 2001 autobiography Just for Fun. “We don’t have much of them in Helsinki”.[14]
While not exactly attuned to the “sociopolitical” side of the Stallman agenda, Torvalds nevertheless appreciated the agenda’s underlying logic: no programmer writes error-free code. By sharing software, hackers put a program’s improvement ahead of individual motivations such as greed or ego protection.
Like many programmers of his generation, Torvalds had cut his teeth not on mainframe computers like the IBM 7094, but on a motley assortment of home-built computer systems. As university student, Torvalds had made the step up from C programming to Unix, using the university’s MicroVAX. This ladder-like progression had given Torvalds a different perspective on the barriers to machine access. For Stallman, the chief barriers were bureaucracy and privilege. For Torvalds, the chief barriers were geography and the harsh Helsinki winter. Forced to trek across the University of Helsinki just to log in to his Unix account, Torvalds quickly began looking for a way to log in from the warm confines of his off-campus apartment.
The search led Torvalds to the operating system Minix, a lightweight version of Unix developed for instructional purposes by Dutch university professor Andrew Tanenbaum. The program fit within the memory confines of a 386 PC, the most powerful machine Torvalds could afford, but still lacked a few necessary features. It most notably lacked terminal emulation, the feature that allowed Torvalds’ machine to mimic a university terminal, making it possible to log in to the MicroVAX from home.
During the summer of 1991, Torvalds rewrote Minix from the ground up, adding other features as he did so. By the end of the summer, Torvalds was referring to his evolving work as the “GNU/Emacs of terminal emulation programs”.[15] Feeling confident, he solicited a Minix newsgroup for copies of the POSIX standards, the software blue prints that determined whether a program was Unix compatible. A few weeks later, Torvalds was posting a message eerily reminiscent of Stallman’s original 1983 GNU posting:
Hello everybody out there using minix- I’m doing a (free) operating system (just a hobby, won’t be big and professional like gnu for 386 (486) AT clones). This has been brewing since April, and is starting to get ready. I’d like any feedback on things people like/dislike in minix, as my OS resembles it somewhat (same physical layout of the file-system (due to practical reasons) among other things).[16]
The posting drew a smattering of responses and within a month, Torvalds had posted a 0.01 version of the operating system-i.e., the earliest possible version fit for outside review-on an Internet FTP site. In the course of doing so, Torvalds had to come up with a name for the new system. On his own PC hard drive, Torvalds had saved the program as Linux, a name that paid its respects to the software convention of giving each Unix variant a name that ended with the letter X. Deeming the name too “egotistical”, Torvalds changed it to Freax, only to have the FTP site manager change it back.
Although Torvalds had set out build a full operating system, both he and other developers knew at the time that most of the functional tools needed to do so were already available, thanks to the work of GNU, BSD, and other free software developers. One of the first tools the Linux development team took advantage of was the GNU C Compiler, a tool that made it possible to process programs written in the C programming language.
Integrating GCC improved the performance of Linux. It also raised issues. Although the GPL’s “viral” powers didn’t apply to the Linux kernel, Torvald’s willingness to borrow GCC for the purposes of his own free software operating system indicated a certain obligation to let other users borrow back. As Torvalds would later put it: “I had hoisted myself up on the shoulders of giants”.[17] Not surprisingly, he began to think about what would happen when other people looked to him for similar support. A decade after the decision, Torvalds echoes the Free Software Foundation’s Robert Chassel when he sums up his thoughts at the time:
You put six months of your life into this thing and you want to make it available and you want to get something out of it, but you don’t want people to take advantage of it. I wanted people to be able to see [Linux], and to make changes and improvements to their hearts’ content. But I also wanted to make sure that what I got out of it was to see what they were doing. I wanted to always have access to the sources so that if they made improvements, I could make those improvements myself.[18]
When it was time to release the 0.12 version of Linux, the first to include a fully integrated version of GCC, Torvalds decided to voice his allegiance with the free software movement. He discarded the old kernel license and replaced it with the GPL. The decision triggered a porting spree, as Torvalds and his collaborators looked to other GNU programs to fold into the growing Linux stew. Within three years, Linux developers were offering their first production release, Linux 1.0, including fully modified versions of GCC, GDB, and a host of BSD tools.
By 1994, the amalgamated operating system had earned enough respect in the hacker world to make some observers wonder if Torvalds hadn’t given away the farm by switching to the GPL in the project’s initial months. In the first issue of Linux Journal, publisher Robert Young sat down with Torvalds for an interview. When Young asked the Finnish programmer if he felt regret at giving up private ownership of the Linux source code, Torvalds said no. “Even with 20/20 hindsight”, Torvalds said, he considered the GPL “one of the very best design decisions” made during the early stages of the Linux project.[19]
That the decision had been made with zero appeal or deference to Stallman and the Free Software Foundation speaks to the GPL’s growing portability. Although it would take a few years to be recognized by Stallman, the explosiveness of Linux development conjured flashbacks of Emacs. This time around, however, the innovation triggering the explosion wasn’t a software hack like Control-R but the novelty of running a Unix-like system on the PC architecture. The motives may have been different, but the end result certainly fit the ethical specifications: a fully functional operating system composed entirely of free software.
12.
See Reuven Lerner, “Stallman wins $240,000 MacArthur award”, MIT,
13.
See Michael Gross, “Richard Stallman: High School Misfit, Symbol of Free Software, MacArthur-certified Genius” (1999).
14.
See Linus Torvalds and David Diamond,
15.
See Linus Torvalds and David Diamond,
17.
See Linus Torvalds and David Diamond,
18.
See Linus Torvalds and David Diamond,
19.
See Robert Young, “Interview with Linus, the Author of Linux”,