Coding Freedom: The Ethics and Aesthetics of Hacking

code on an Internet newsgroup, comp.os.minix , with the hope of coaxing feedback from others and to allow other programmers to “play” with it. His first posting to the minix newsgroup on August 25, 1991, when he announced his project, reflects his initially humble intentions (which he contrasted with the FSF GNU project): “I’m doing a (free) operating system (just a hobby, won’t be big and professional like gnu). 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.” At the end of this message, he predicts, incorrectly, that the OS “probably never will support anything other than AT-harddisks, as that’s all I have :).” 15 Given just how common Linux is today, Torvalds’s statement is famous among F/OSS developers for its historical irony.
    Yet a kernel, on its own, is a far cry from a functional OS. Since Stallman’s GNU project had already made many components required for an OS, but had not yet developed a fully operational kernel, Torvalds decided to integrate GNU’s copylefted software applications and components with his kernel. This decision would prove crucial; it required Torvalds to license Linux under the GNU GPL, and assured that the Linux source code would remain accessible throughout its many future modifications and versions. In a 1994 interview, Torvalds remarked that choosing the GPL license was “one of the very best design decisions I ever did, along with accepting code that was copyrighted by other holders (under the same copyright conditions, of course).” 16 Pairing the GNU project with Linux was also a marriage between the purely technical motivations of Torvalds and the philosophical, political motivations of Stallman—a marriage that would come to see some tense moments in the future.
    Indeed, Stallman was not known for his deft leadership skills, while Torvalds’s was to become well known for his open and effective style of leadership. Stallman was foremost a political crusader, attempting to salvage what he saw as the withering away of a culture; Torvalds was a fierce technical pragmatist, embodying a no-frills sensibility commonly championed by many hackers. In marked contrast to Stallman, who tightly controlled the development of FSF software, Torvalds was keen to receive any feedbackfrom peers through newsgroups, where programmers could contribute bug fixes and improvements that, if deemed worthwhile by Torvalds (who became known as the project’s “benevolent dictator”), would be incorporated into new versions of the Linux kernel. Unlike earlier generations of hackers, Torvalds could now do a significant amount of work from the comforts of home (thanks to the personal computer and an Internet connection), and in the process of developing the new kernel, he became a skillful leader, coordinating the contributions of geographically dispersed developers over the Internet.
    The Linux kernel development project helped usher in a new era of networked hacking, in which project leadership validates its status as much through its ability to evaluate and coordinate contributions from others as through the leaders’ own technical prowess. This mature form of networked hacking differed in at least three respects from previous instances of hacker collaboration: production was not affiliated solely with a single institution; production occurred largely independent of market pressures and conditions; and contributions, from previously unknown third parties, were encouraged and, if deemed technically helpful, accepted. Through this experimentation, hackers would ultimately produce software applications robust enough to compete with proprietary software in the market, although few knew this at the time.
    Before the advent of Linux, the idea that complicated software systems could be produced by geographically dispersed hackers was largely disparaged (Raymond 1999). While it would be an exaggeration to claim that long-distance collaboration between programmers was nonexistent prior to the Linux project, its pace was slow, its scale was contained, and its effects were often piecemeal, especially since such collective laboring had required sending tapes over postal mail. 17 In this period, the FSF had already released a number of widely used and technically respected software tools and applications, based on the integrated work of many programmers.
    Linux was, even if not