Sam Williams

For Want of a Printer

I fear the Greeks. Even when they bring gifts. ---Virgil The Aeneid

The new printer was jammed, again. 

Richard M. Stallman, a staff software programmer at theMassachusetts Institute of Technology's ArtificialIntelligence Laboratory (AI Lab), discovered themalfunction the hard way. An hour after sending off a50-page file to the office laser printer, Stallman, 27, broke off a productive work session to retrieve hisdocuments. Upon arrival, he found only four pages inthe printer's tray. To make matters even morefrustrating, the four pages belonged to another user, meaning that Stallman's print job and the unfinishedportion of somebody else's print job were still trappedsomewhere within the electrical plumbing of the lab'scomputer network. 

Waiting for machines is an occupational hazard whenyou're a software programmer, so Stallman took hisfrustration with a grain of salt. Still, the differencebetween waiting for a machine and waiting on a machineis a sizable one. It wasn't the first time he'd beenforced to stand over the printer, watching pages printout one by one. As a person who spent the bulk of hisdays and nights improving the efficiency of machinesand the software programs that controlled them, Stallman felt a natural urge to open up the machine, look at the guts, and seek out the root of the problem. 

Unfortunately, Stallman's skills as a computerprogrammer did not extend to the mechanical-engineeringrealm. As freshly printed documents poured out of themachine, Stallman had a chance to reflect on other waysto circumvent the printing jam problem. 

How long ago had it been that the staff members at theAI Lab had welcomed the new printer with open arms?Stallman wondered. The machine had been a donation fromthe Xerox Corporation. A cutting edge prototype, it wasa modified version of the popular Xerox photocopier. Only instead of making copies, it relied on softwaredata piped in over a computer network to turn that datainto professional-looking documents. Created byengineers at the world-famous Xerox Palo Alto ResearchFacility, it was, quite simply, an early taste of thedesktop-printing revolution that would seize the restof the computing industry by the end of the decade. 

Driven by an instinctual urge to play with the best newequipment, programmers at the AI Lab promptlyintegrated the new machine into the lab's sophisticatedcomputing infrastructure. The results had beenimmediately pleasing. Unlike the lab's old laserprinter, the new Xerox machine was fast. Pages cameflying out at a rate of one per second, turning a20-minute print job into a 2-minute print job. The newmachine was also more precise. Circles came out lookinglike circles, not ovals. Straight lines came outlooking like straight lines, not low-amplitude sine waves. 

It was, for all intents and purposes, a gift too goodto refuse. 

It wasn't until a few weeks after its arrival that themachine's flaws began to surface. Chief among thedrawbacks was the machine's inherent susceptibility topaper jams. Engineering-minded programmers quicklyunderstood the reason behind the flaw. As aphotocopier, the machine generally required the directoversight of a human operator. Figuring that thesehuman operators would always be on hand to fix a paperjam, if it occurred, Xerox engineers had devoted theirtime and energies to eliminating other pesky problems. In engineering terms, user diligence was built into the system. 

In modifying the machine for printer use, Xeroxengineers had changed the user-machine relationship ina subtle but profound way. Instead of making themachine subservient to an individual human operator, they made it subservient to an entire networkedpopulation of human operators. Instead of standingdirectly over the machine, a human user on one end ofthe network sent his print command through an extendedbucket-brigade of machines, expecting the desiredcontent to arrive at the targeted destination and inproper form. It wasn't until he finally went to checkup on the final output that he realized how little ofthe desired content had made it through. 

Stallman himself had been of the first to identify theproblem and the first to suggest a remedy. Yearsbefore, when the lab was still using its old printer, Stallman had solved a similar problem by opening up thesoftware program that regulated the printer on thelab's PDP-11 machine. Stallman couldn't eliminate paperjams, but he could insert a software command thatordered the PDP-11 to check the printer periodicallyand report back to the PDP-10, the lab's centralcomputer. To ensure that one user's negligence didn'tbog down an entire line of print jobs, Stallman alsoinserted a software command that instructed the PDP-10to notify every user with a waiting print job that theprinter was jammed. The notice was simple, somethingalong the lines of "The printer is jammed, please fixit, " and because it went out to the people with themost pressing need to fix the problem, chances werehigher that the problem got fixed in due time. 

As fixes go, Stallman's was oblique but elegant. Itdidn't fix the mechanical side of the problem, but itdid the next best thing by closing the information loopbetween user and machine. Thanks to a few additionallines of software code, AI Lab employees couldeliminate the 10 or 15 minutes wasted each week inrunning back and forth to check on the printer. Inprogramming terms, Stallman's fix took advantage of theamplified intelligence of the overall network. 

"If you got that message, you couldn't assume somebodyelse would fix it, " says Stallman, recalling the logic. "You had to go to the printer. A minute or two afterthe printer got in trouble, the two or three people whogot messages arrive to fix the machine. Of those two orthree people, one of them, at least, would usually knowhow to fix the problem. "

Such clever fixes were a trademark of the AI Lab andits indigenous population of programmers. Indeed, thebest programmers at the AI Lab disdained the termprogrammer, preferring the more slangy occupationaltitle of hacker instead. The job title covered a hostof activities-everything from creative mirth making tothe improvement of existing software and computersystems. Implicit within the title, however, was theold-fashioned notion of Yankee ingenuity. To be ahacker, one had to accept the philosophy that writing asoftware program was only the beginning. Improving aprogram was the true test of a hacker's skills. For more on the term "hacker, "see Appendix B. 

Such a philosophy was a major reason why companies likeXerox made it a policy to donate their machines andsoftware programs to places where hackers typicallycongregated. If hackers improved the software, companies could borrow back the improvements, incorporating them into update versions for thecommercial marketplace. In corporate terms, hackerswere a leveragable community asset, an auxiliaryresearch-and-development division available at minimal cost. 

It was because of this give-and-take philosophy thatwhen Stallman spotted the print-jam defect in the Xeroxlaser printer, he didn't panic. He simply looked for away to update the old fix or " hack" for the newsystem. In the course of looking up the Xeroxlaser-printer software, however, Stallman made atroubling discovery. The printer didn't have anysoftware, at least nothing Stallman or a fellowprogrammer could read. Until then, most companies hadmade it a form of courtesy to publish source-codefiles-readable text files that documented theindividual software commands that told a machine whatto do. Xerox, in this instance, had provided softwarefiles in precompiled, or binary, form. Programmers werefree to open the files up if they wanted to, but unlessthey were an expert in deciphering an endless stream ofones and zeroes, the resulting text was pure gibberish. 

Although Stallman knew plenty about computers, he wasnot an expert in translating binary files. As a hacker, however, he had other resources at his disposal. Thenotion of information sharing was so central to thehacker culture that Stallman knew it was only a matterof time before some hacker in some university lab orcorporate computer room proffered a version of thelaser-printer source code with the desired source-code files. 

After the first few printer jams, Stallman comfortedhimself with the memory of a similar situation yearsbefore. The lab had needed a cross-network program tohelp the PDP-11 work more efficiently with the PDP-10. The lab's hackers were more than up to the task, butStallman, a Harvard alumnus, recalled a similar programwritten by programmers at the Harvard computer-sciencedepartment. The Harvard computer lab used the samemodel computer, the PDP-10, albeit with a differentoperating system. The Harvard computer lab also had apolicy requiring that all programs installed on thePDP-10 had to come with published source-code files. 

Taking advantage of his access to the Harvard computerlab, Stallman dropped in, made a copy of thecross-network source code, and brought it back to theAI Lab. He then rewrote the source code to make it moresuitable for the AI Lab's operating system. With nomuss and little fuss, the AI Lab shored up a major gapin its software infrastructure. Stallman even added afew features not found in the original Harvard program, making the program even more useful. "We wound up usingit for several years, " Stallman says. 

From the perspective of a 1970s-era programmer, thetransaction was the software equivalent of a neighborstopping by to borrow a power tool or a cup of sugarfrom a neighbor. The only difference was that inborrowing a copy of the software for the AI Lab, Stallman had done nothing to deprive Harvard hackersthe use of their original program. If anything, Harvardhackers gained in the process, because Stallman hadintroduced his own additional features to the program, features that hackers at Harvard were perfectly free toborrow in return. Although nobody at Harvard ever cameover to borrow the program back, Stallman does recall aprogrammer at the private engineering firm, Bolt, Beranek & Newman, borrowing the program and adding afew additional features, which Stallman eventuallyreintegrated into the AI Lab's own source-code archive. 

"A program would develop the way a city develops, " saysStallman, recalling the software infrastructure of theAI Lab. "Parts would get replaced and rebuilt. Newthings would get added on. But you could always look ata certain part and say, `Hmm, by the style, I see thispart was written back in the early 60s and this partwas written in the mid-1970s. '"

Through this simple system of intellectual accretion, hackers at the AI Lab and other places built up robustcreations. On the west coast, computer scientists at UCBerkeley, working in cooperation with a few low-levelengineers at AT&T, had built up an entire operatingsystem using this system. Dubbed Unix, a play on anolder, more academically respectable operating systemcalled Multics, the software system was available toany programmer willing to pay for the cost of copyingthe program onto a new magnetic tape and shipping it. Not every programmer participating in this culturedescribed himself as a hacker, but most shared thesentiments of Richard M. Stallman. If a program orsoftware fix was good enough to solve your problems, itwas good enough to solve somebody else's problems. Whynot share it out of a simple desire for good karma?

The fact that Xerox had been unwilling to share itssource-code files seemed a minor annoyance at first. Intracking down a copy of the source-code files, Stallmansays he didn't even bother contacting Xerox. "They hadalready given us the laser printer, " Stallman says. "Why should I bug them for more?"

When the desired files failed to surface, however, Stallman began to grow suspicious. The year before, Stallman had experienced a blow up with a doctoralstudent at Carnegie Mellon University. The student, Brian Reid, was the author of a useful text-formattingprogram dubbed Scribe. One of the first programs thatgave a user the power to define fonts and type styleswhen sending a document over a computer network, theprogram was an early harbinger of HTML, the linguafranca of the World Wide Web. In 1979, Reid made thedecision to sell Scribe to a Pittsburgh-area softwarecompany called Unilogic. His graduate-student careerending, Reid says he simply was looking for a way tounload the program on a set of developers that wouldtake pains to keep it from slipping into the publicdomain. To sweeten the deal, Reid also agreed to inserta set of time-dependent functions- "time bombs" insoftware-programmer parlance-that deactivated freelycopied versions of the program after a 90-dayexpiration date. To avoid deactivation, users paid thesoftware company, which then issued a code that defusedthe internal time-bomb feature. 

For Reid, the deal was a win-win. Scribe didn't fallinto the public domain, and Unilogic recouped on itsinvestment. For Stallman, it was a betrayal of theprogrammer ethos, pure and simple. Instead of honoringthe notion of share-and-share alike, Reid had inserteda way for companies to compel programmers to pay forinformation access. 

As the weeks passed and his attempts to track downXerox laser-printer source code hit a brick wall, Stallman began to sense a similar money-for-codescenario at work. Before Stallman could do or sayanything about it, however, good news finally trickledin via the programmer grapevine. Word had it that ascientist at the computer-science department atCarnegie Mellon University had just departed a job atthe Xerox Palo Alto Research Center. Not only had thescientist worked on the laser printer in question, butaccording to rumor, he was still working on it as partof his research duties at Carnegie Mellon. 

Casting aside his initial suspicion, Stallman made afirm resolution to seek out the person in questionduring his next visit to the Carnegie Mellon campus. 

He didn't have to wait long. Carnegie Mellon also had alab specializing in artificial-intelligence research, and within a few months, Stallman had abusiness-related reason to visit the Carnegie Melloncampus. During that visit, he made sure to stop by thecomputer-science department. Department employeesdirected him to the office of the faculty memberleading the Xerox project. When Stallman reached theoffice, he found the professor working there. 

In true engineer-to-engineer fashion, the conversationwas cordial but blunt. After briefly introducinghimself as a visitor from MIT, Stallman requested acopy of the laser-printer source code so that he couldport it to the PDP-11. To his surprise, the professorrefused to grant his request. 

"He told me that he had promised not to give me acopy, " Stallman says. 

Memory is a funny thing. Twenty years after the fact, Stallman's mental history tape is notoriously blank inplaces. Not only does he not remember the motivatingreason for the trip or even the time of year duringwhich he took it, he also has no recollection of theprofessor or doctoral student on the other end of theconversation. According to Reid, the person most likelyto have fielded Stallman's request is Robert Sproull, aformer Xerox PARC researcher and current director ofSun Laboratories, a research division of thecomputer-technology conglomerate Sun Microsystems. During the 1970s, Sproull had been the primarydeveloper of the laser-printer software in questionwhile at Xerox PARC. Around 1980, Sproull took afaculty research position at Carnegie Mellon where hecontinued his laser-printer work amid other projects. 

"The code that Stallman was asking for was leading-edgestate-of-the-art code that Sproull had written in theyear or so before going to Carnegie Mellon, " recallsReid. "I suspect that Sproull had been at CarnegieMellon less than a month before this request came in. "

When asked directly about the request, however, Sproulldraws a blank. "I can't make a factual comment, " writesSproull via email. "I have absolutely no recollectionof the incident. "

With both participants in the brief conversationstruggling to recall key details-including whether theconversation even took place-it's hard to gauge thebluntness of Sproull's refusal, at least as recalled byStallman. In talking to audiences, Stallman has maderepeated reference to the incident, noting thatSproull's unwillingness to hand over the source codestemmed from a nondisclosure agreement, a contractualagreement between Sproull and the Xerox Corporationgiving Sproull, or any other signatory, access thesoftware source code in exchange for a promise ofsecrecy. Now a standard item of business in thesoftware industry, the nondisclosure agreement, or NDA, was a novel development at the time, a reflection ofboth the commercial value of the laser printer to Xeroxand the information needed to run it. "Xerox was at thetime trying to make a commercial product out of thelaser printer, " recalls Reid. "They would have beeninsane to give away the source code. "

For Stallman, however, the NDA was something elseentirely. It was a refusal on the part of Xerox andSproull, or whomever the person was that turned downhis source-code request that day, to participate in asystem that, until then, had encouraged softwareprogrammers to regard programs as communal resources. Like a peasant whose centuries-old irrigation ditch hadgrown suddenly dry, Stallman had followed the ditch toits source only to find a brand-spanking-newhydroelectric dam bearing the Xerox logo. 

For Stallman, the realization that Xerox had compelleda fellow programmer to participate in this newfangledsystem of compelled secrecy took a while to sink in. Atfirst, all he could focus on was the personal nature ofthe refusal. As a person who felt awkward and out ofsync in most face-to-face encounters, Stallman'sattempt to drop in on a fellow programmer unannouncedhad been intended as a demonstration of neighborliness. Now that the request had been refused, it felt like amajor blunder. "I was so angry I couldn't think of away to express it. So I just turned away and walked outwithout another word, " Stallman recalls. "I might haveslammed the door. Who knows? All I remember is wantingto get out of there. "

Twenty years after the fact, the anger still lingers, so much so that Stallman has elevated the event into amajor turning point. Within the next few months, aseries of events would befall both Stallman and the AILab hacker community that would make 30 seconds worthof tension in a remote Carnegie Mellon office seemtrivial by comparison. Nevertheless, when it comes timeto sort out the events that would transform Stallmanfrom a lone hacker, instinctively suspicious ofcentralized authority, to a crusading activist applyingtraditional notions of liberty, equality, andfraternity to the world of software development, Stallman singles out the Carnegie Mellon encounter forspecial attention. 

"It encouraged me to think about something that I'dalready been thinking about, " says Stallman. "I alreadyhad an idea that software should be shared, but Iwasn't sure how to think about that. My thoughtsweren't clear and organized to the point where I couldexpress them in a concise fashion to the rest of the world. "

Although previous events had raised Stallman's ire, hesays it wasn't until his Carnegie Mellon encounter thathe realized the events were beginning to intrude on aculture he had long considered sacrosanct. As an eliteprogrammer at one of the world's elite institutions, Stallman had been perfectly willing to ignore thecompromises and bargains of his fellow programmers justso long as they didn't interfere with his own work. Until the arrival of the Xerox laser printer, Stallmanhad been content to look down on the machines andprograms other computer users grimly tolerated. On therare occasion that such a program breached the AI Lab'swalls-when the lab replaced its venerable IncompatibleTime Sharing operating system with a commercialvariant, the TOPS 20, for example-Stallman and hishacker colleagues had been free to rewrite, reshape, and rename the software according to personal taste. 

Now that the laser printer had insinuated itself withinthe AI Lab's network, however, something had changed. The machine worked fine, barring the occasional paperjam, but the ability to modify according to personaltaste had disappeared. From the viewpoint of the entiresoftware industry, the printer was a wake-up call. Software had become such a valuable asset thatcompanies no longer felt the need to publicize sourcecode, especially when publication meant givingpotential competitors a chance to duplicate somethingcheaply. From Stallman's viewpoint, the printer was aTrojan Horse. After a decade of failure, privatelyowned software-future hackers would use the term "proprietary" software-had gained a foothold inside theAI Lab through the sneakiest of methods. It had comedisguised as a gift. 

That Xerox had offered some programmers access toadditional gifts in exchange for secrecy was alsogalling, but Stallman takes pains to note that, ifpresented with such a quid pro quo bargain at a youngerage, he just might have taken the Xerox Corporation upon its offer. The awkwardness of the Carnegie Mellonencounter, however, had a firming effect on Stallman'sown moral lassitude. Not only did it give him thenecessary anger to view all future entreaties withsuspicion, it also forced him to ask the uncomfortablequestion: what if a fellow hacker dropped intoStallman's office someday and it suddenly becameStallman's job to refuse the hacker's request forsource code?

"It was my first encounter with a nondisclosureagreement, and it immediately taught me thatnondisclosure agreements have victims, " says Stallman, firmly. "In this case I was the victim. [My lab and I]were victims. "

It was a lesson Stallman would carry with him throughthe tumultuous years of the 1980s, a decade duringwhich many of his MIT colleagues would depart the AILab and sign nondisclosure agreements of their own. Because most nondisclosure aggreements (NDAs) hadexpiration dates, few hackers who did sign them sawlittle need for personal introspection. Sooner orlater, they reasoned, the software would become publicknowledge. In the meantime, promising to keep thesoftware secret during its earliest development stageswas all a part of the compromise deal that allowedhackers to work on the best projects. For Stallman, however, it was the first step down a slippery slope. 

"When somebody invited me to betray all my colleaguesin that way, I remembered how angry I was when somebodyelse had done that to me and my whole lab, " Stallmansays. "So I said, `Thank you very much for offering methis nice software package, but I can't accept it onthe conditions that you're asking for, so I'm going todo without it. '"

As Stallman would quickly learn, refusing such requestsinvolved more than personal sacrifice. It involvedsegregating himself from fellow hackers who, thoughsharing a similar distaste for secrecy, tended toexpress that distaste in a more morally flexiblefashion. It wasn't long before Stallman, increasinglyan outcast even within the AI Lab, began billinghimself as "the last true hacker, " isolating himselffurther and further from a marketplace dominated byproprietary software. Refusing another's request forsource code, Stallman decided, was not only a betrayalof the scientific mission that had nurtured softwaredevelopment since the end of World War II, it was aviolation of the Golden Rule, the baseline moraldictate to do unto others as you would have them dounto you. 

Hence the importance of the laser printer and theencounter that resulted from it. Without it, Stallmansays, his life might have followed a more ordinarypath, one balancing the riches of a commercialprogrammer with the ultimate frustration of a lifespent writing invisible software code. There would havebeen no sense of clarity, no urgency to address aproblem others weren't addressing. Most importantly, there would have been no righteous anger, an emotionthat, as we soon shall see, has propelled Stallman'scareer as surely as any political ideology or ethical belief. 

"From that day forward, I decided this was something Icould never participate in, " says Stallman, alluding tothe practice of trading personal liberty for the sakeof convenience-Stallman's description of the NDAbargain-as well as the overall culture that encouragedsuch ethically suspect deal-making in the first place. "I decided never to make other people victims just likeI had been a victim. "

2001: A Hacker's Odyssey

The New York University computer-science departmentsits inside Warren Weaver Hall, a fortress-likebuilding located two blocks east of Washington SquarePark. Industrial-strength air-conditioning vents createa surrounding moat of hot air, discouraging loiterersand solicitors alike. Visitors who breach the moatencounter another formidable barrier, a securitycheck-in counter immediately inside the building'ssingle entryway. 

Beyond the security checkpoint, the atmosphere relaxessomewhat. Still, numerous signs scattered throughoutthe first floor preach the dangers of unsecured doorsand propped-open fire exits. Taken as a whole, thesigns offer a reminder: even in the relatively tranquilconfines of pre-September 11, 2001, New York, one cannever be too careful or too suspicious. 

The signs offer an interesting thematic counterpoint tothe growing number of visitors gathering in the hall'sinterior atrium. A few look like NYU students. Mostlook like shaggy-aired concert-goers milling outside amusic hall in anticipation of the main act. For onebrief morning, the masses have taken over Warren WeaverHall, leaving the nearby security attendant withnothing better to do but watch Ricki Lake on TV andshrug her shoulders toward the nearby auditoriumwhenever visitors ask about "the speech. "

Once inside the auditorium, a visitor finds the personwho has forced this temporary shutdown of buildingsecurity procedures. The person is Richard M. Stallman, founder of the GNU Project, original president of theFree Software Foundation, winner of the 1990 MacArthurFellowship, winner of the Association of ComputingMachinery's Grace Murray Hopper Award (also in 1990), corecipient of the Takeda Foundation's 2001 TakedaAward, and former AI Lab hacker. As announced over ahost of hacker-related web sites, including the GNUProject's own http://www. Gnu. Org site, Stallman is inManhattan, his former hometown, to deliver a muchanticipated speech in rebuttal to the MicrosoftCorporation's recent campaign against the GNU GeneralPublic License. 

The subject of Stallman's speech is the history andfuture of the free software movement. The location issignificant. Less than a month before, Microsoft seniorvice president Craig Mundie appeared at the nearby NYUStern School of Business, delivering a speech blastingthe General Public License, or GPL, a legal deviceoriginally conceived by Stallman 16 years before. Builtto counteract the growing wave of software secrecyovertaking the computer industry-a wave first noticedby Stallman during his 1980 troubles with the Xeroxlaser printer-the GPL has evolved into a central toolof the free software community. In simplest terms, theGPL locks software programs into a form of communalownership-what today's legal scholars now call the"digital commons"-through the legal weight ofcopyright. Once locked, programs remain unremovable. Derivative versions must carry the same copyrightprotection-even derivative versions that bear only asmall snippet of the original source code. For thisreason, some within the software industry have taken tocalling the GPL a "viral" license, because it spreadsitself to every software program it touches. Actually, the GPL's powers are notquite that potent. According to section 10 of the GNU General PublicLicense, Version 2 (1991), the viral nature of thelicense depends heavily on the Free SoftwareFoundation's willingness to view a program as aderivative work, not to mention the existing licensethe GPL would replace. 

If you wish to incorporate parts of the Program intoother free programs whose distribution conditions aredifferent, write to the author to ask for permission. For software that is copyrighted by the Free SoftwareFoundation, write to the Free Software Foundation; wesometimes make exceptions for this. Our decision willbe guided by the two goals of preserving the freestatus of all derivatives of our free software and ofpromoting the sharing and reuse of software generally. 

"To compare something to a virus is very harsh, " saysStallman. "A spider plant is a more accuratecomparison; it goes to another place if you activelytake a cutting. "

For more information on the GNU General Public License, visit [http://www. Gnu. Org/copyleft/gpl. Html. ]

In an information economy increasingly dependent onsoftware and increasingly beholden to softwarestandards, the GPL has become the proverbial "bigstick. " Even companies that once laughed it off assoftware socialism have come around to recognize thebenefits. Linux, the Unix-like kernel developed byFinnish college student Linus Torvalds in 1991, islicensed under the GPL, as are many of the world's mostpopular programming tools: GNU Emacs, the GNU Debugger, the GNU C Compiler, etc. Together, these tools form thecomponents of a free software operating systemdeveloped, nurtured, and owned by the worldwide hackercommunity. Instead of viewing this community as athreat, high-tech companies like IBM, Hewlett Packard, and Sun Microsystems have come to rely upon it, sellingsoftware applications and services built to ride atopthe ever-growing free software infrastructure. 

They've also come to rely upon it as a strategic weaponin the hacker community's perennial war againstMicrosoft, the Redmond, Washington-based company that, for better or worse, has dominated the PC-softwaremarketplace since the late 1980s. As owner of thepopular Windows operating system, Microsoft stands tolose the most in an industry-wide shift to the GPLlicense. Almost every line of source code in theWindows colossus is protected by copyrights reaffirmingthe private nature of the underlying source code or, atthe very least, reaffirming Microsoft's legal abilityto treat it as such. From the Microsoft viewpoint, incorporating programs protected by the "viral" GPLinto the Windows colossus would be the softwareequivalent of Superman downing a bottle of Kryptonitepills. Rival companies could suddenly copy, modify, andsell improved versions of Windows, rendering thecompany's indomitable position as the No. 1 provider ofconsumer-oriented software instantly vulnerable. Hencethe company's growing concern over the GPL's rate ofadoption. Hence the recent Mundie speech blasting theGPL and the " open source" approach to softwaredevelopment and sales. And hence Stallman's decision todeliver a public rebuttal to that speech on the samecampus here today. 

20 years is a long time in the software industry. Consider this: in 1980, when Richard Stallman wascursing the AI Lab's Xerox laser printer, Microsoft, the company modern hackers view as the most powerfulforce in the worldwide software industry, was still aprivately held startup. IBM, the company hackers usedto regard as the most powerful force in the worldwidesoftware industry, had yet to to introduce its firstpersonal computer, thereby igniting the currentlow-cost PC market. Many of the technologies we nowtake for granted-the World Wide Web, satellitetelevision, 32-bit video-game consoles-didn't evenexist. The same goes for many of the companies that nowfill the upper echelons of the corporate establishment, companies like AOL, Sun Microsystems, Amazon. Com, Compaq, and Dell. The list goes on and on. 

The fact that the high-technology marketplace has comeso far in such little time is fuel for both sides ofthe GPL debate. GPL-proponents point to the shortlifespan of most computer hardware platforms. Facingthe risk of buying an obsolete product, consumers tendto flock to companies with the best long-term survival. As a result, the software marketplace has become awinner-take-all arena. See Shubha Ghosh, "Revealing the Microsoft WindowsSource Code, " Gigalaw. Com (January, 2000). Http://www. Gigalaw. Com/articles/ghosh-2000-01-p1. Html The current, privately owned software environment, GPL-proponents say, leads to monopoly abuse andstagnation. Strong companies suck all the oxygen out ofthe marketplace for rival competitors and innovative startups. 

GPL-opponents argue just the opposite. Selling softwareis just as risky, if not more risky, than buyingsoftware, they say. Without the legal guaranteesprovided by private software licenses, not to mentionthe economic prospects of a privately owned "killerapp" (i. E. , a breakthrough technology that launches anentirely new market), Killer apps don't have to be proprietary. Witness, ofcourse, the legendary Mosaic browser, a program whosecopyright permits noncommercial derivatives withcertain restrictions. Still, I think the reader getsthe point: the software marketplace is like thelottery. The bigger the potential payoff, the morepeople want to participate. For a good summary of thekiller-app phenomenon, see Philip Ben-David, "WhateverHappened to the `Killer App'?" e-Commerce News(December 7, 2000). Companies lose the incentive to participate. Onceagain, the market stagnates and innovation declines. AsMundie himself noted in his May 3 address on the samecampus, the GPL's "viral" nature "poses a threat" toany company that relies on the uniqueness of itssoftware as a competitive asset. Added Mundie: It alsofundamentally undermines the independent commercialsoftware sector because it effectively makes itimpossible to distribute software on a basis whererecipients pay for the product rather than just thecost of distributionSee Craig Mundie, "The Commercial Software Model, "senior vice president, Microsoft Corp. Excerpted froman online transcript of Mundie's May 3, speech to theNew York University Stern School of Business. 

http://www. Ecommercetimes. Com/perl/story/5893. Html 001, http://www. Microsoft. Com/presspass/exec/craig/05-03sharedsource. Asp The mutual success of GNU/ LinuxThe acronym GNU stands for "GNU's not Unix. "In anotherportion of the May 29, 2001, NYU speech, Stallmansummed up the acronym's origin: We hackers always lookfor a funny or naughty name for a program, becausenaming a program is half the fun of writing theprogram. We also had a tradition of recursive acronyms, to say that the program that you're writing is similarto some existing program . . . I looked for a recursiveacronym for Something Is Not UNIX. And I tried all 26letters and discovered that none of them was a word. Idecided to make it a contraction. That way I could havea three-letter acronym, for Something's Not UNIX. And Itried letters, and I came across the word "GNU. " Thatwas it. Although a fan of puns, Stallman recommendsthat software users pronounce the "g" at the beginningof the acronym (i. E. , "gah-new"). Not only does thisavoid confusion with the word "gnu, " the name of theAfrican antelope, Connochaetes gnou, it also avoidsconfusion with the adjective "new. " "We've been workingon it for 17 years now, so it is not exactly new anymore, " Stallman says. Source: author notes and onlinetranscript of "Free Software: Freedom and Cooperation, "Richard Stallman's May 29, 2001, speech at New York University. 

http://www. Gnu. Org/events/rms-nyu-2001-transcript. Txt, the amalgamated operating system built around theGPL-protected Linux kernel, and Windows over the last10 years reveals the wisdom of both perspectives. Nevertheless, the battle for momentum is an importantone in the software industry. Even powerful vendorssuch as Microsoft rely on the support of third-partysoftware developers whose tools, programs, and computergames make an underlying software platform such asWindows more attractive to the mainstream consumer. Citing the rapid evolution of the technologymarketplace over the last 20 years, not to mention hisown company's admirable track record during thatperiod, Mundie advised listeners to not get too carriedaway by the free software movement's recent momentum:Two decades of experience have shown that an economicmodel that protects intellectual property and abusiness model that recoups research and developmentcosts can create impressive economic benefits anddistribute them very broadly. Such admonitions serve asthe backdrop for Stallman's speech today. Less than amonth after their utterance, Stallman stands with hisback to one of the chalk boards at the front of theroom, edgy to begin. 

If the last two decades have brought dramatic changesto the software marketplace, they have brought evenmore dramatic changes to Stallman himself. Gone is theskinny, clean-shaven hacker who once spent his entiredays communing with his beloved PDP-10. In his placestands a heavy-set middle-aged man with long hair andrabbinical beard, a man who now spends the bulk of histime writing and answering email, haranguing fellowprogrammers, and giving speeches like the one today. Dressed in an aqua-colored T-shirt and brown polyesterpants, Stallman looks like a desert hermit who juststepped out of a Salvation Army dressing room. 

The crowd is filled with visitors who share Stallman'sfashion and grooming tastes. Many come bearing laptopcomputers and cellular modems, all the better to recordand transmit Stallman's words to a waiting Internetaudience. The gender ratio is roughly 15 males to 1female, and 1 of the 7 or 8 females in the room comesin bearing a stuffed penguin, the official Linuxmascot, while another carries a stuffed teddy bear. 

<Graphic file:/home/craigm/books/free_0201. Png>

Richard Stallman, circa 2000. "I decided I woulddevelop a free software operating system or die trying. . Of old age of course. " Photo courtesy ofhttp://www. Stallman. Org. 

Agitated, Stallman leaves his post at the front of theroom and takes a seat in a front-row chair, tapping afew commands into an already-opened laptop. For thenext 10 minutes Stallman is oblivious to the growingnumber of students, professors, and fans circulating infront of him at the foot of the auditorium stage. 

Before the speech can begin, the baroque rituals ofacademic formality must be observed. Stallman'sappearance merits not one but two introductions. MikeUretsky, codirector of the Stern School's Center forAdvanced Technology, provides the first. 

"The role of a university is to foster debate and tohave interesting discussions, " Uretsky says. "Thisparticular presentation, this seminar falls right intothat mold. I find the discussion of open sourceparticularly interesting. "

Before Uretsky can get another sentence out, Stallmanis on his feet waving him down like a stranded motorist. 

"I do free software, " Stallman says to rising laughter. "Open source is a different movement. "

The laughter gives way to applause. The room is stockedwith Stallman partisans, people who know of hisreputation for verbal exactitude, not to mention hismuch publicized 1998 falling out with the open sourcesoftware proponents. Most have come to anticipate suchoutbursts the same way radio fans once waited for JackBenny's trademark, "Now cut that out!" phrase duringeach radio program. 

Uretsky hastily finishes his introduction and cedes thestage to Edmond Schonberg, a professor in the NYUcomputer-science department. As a computer programmerand GNU Project contributor, Schonberg knows whichlinguistic land mines to avoid. He deftly summarizesStallman's career from the perspective of a modern-dayprogrammer. 

"Richard is the perfect example of somebody who, byacting locally, started thinking globally [about]problems concerning the unavailability of source code, "says Schonberg. "He has developed a coherent philosophythat has forced all of us to reexamine our ideas of howsoftware is produced, of what intellectual propertymeans, and of what the software community actually represents. "

Schonberg welcomes Stallman to more applause. Stallmantakes a moment to shut off his laptop, rises out of hischair, and takes the stage. 

At first, Stallman's address seems more Catskillscomedy routine than political speech. "I'd like tothank Microsoft for providing me the opportunity to beon this platform, " Stallman wisecracks. "For the pastfew weeks, I have felt like an author whose book wasfortuitously banned somewhere. "

For the uninitiated, Stallman dives into a quick freesoftware warm-up analogy. He likens a software programto a cooking recipe. Both provide useful step-by-stepinstructions on how to complete a desired task and canbe easily modified if a user has special desires orcircumstances. "You don't have to follow a recipeexactly, " Stallman notes. "You can leave out someingredients. Add some mushrooms, 'cause you likemushrooms. Put in less salt because your doctor saidyou should cut down on salt-whatever. "

Most importantly, Stallman says, software programs andrecipes are both easy to share. In giving a recipe to adinner guest, a cook loses little more than time andthe cost of the paper the recipe was written on. Software programs require even less, usually a fewmouse-clicks and a modicum of electricity. In bothinstances, however, the person giving the informationgains two things: increased friendship and the abilityto borrow interesting recipes in return. 

"Imagine what it would be like if recipes were packagedinside black boxes, " Stallman says, shifting gears. "You couldn't see what ingredients they're using, letalone change them, and imagine if you made a copy for afriend. They would call you a pirate and try to put youin prison for years. That world would create tremendousoutrage from all the people who are used to sharingrecipes. But that is exactly what the world ofproprietary software is like. A world in which commondecency towards other people is prohibited or prevented. "

With this introductory analogy out of the way, Stallmanlaunches into a retelling of the Xerox laser-printerepisode. Like the recipe analogy, the laser-printerstory is a useful rhetorical device. With itsparable-like structure, it dramatizes just how quicklythings can change in the software world. Drawinglisteners back to an era before Amazon. Com one-clickshopping, Microsoft Windows, and Oracle databases, itasks the listener to examine the notion of softwareownership free of its current corporate logos. 

Stallman delivers the story with all the polish andpractice of a local district attorney conducting aclosing argument. When he gets to the part about theCarnegie Mellon professor refusing to lend him a copyof the printer source code, Stallman pauses. 

"He had betrayed us, " Stallman says. "But he didn'tjust do it to us. Chances are he did it to you. "

On the word "you, " Stallman points his index fingeraccusingly at an unsuspecting member of the audience. The targeted audience member's eyebrows flinchslightly, but Stallman's own eyes have moved on. Slowlyand deliberately, Stallman picks out a second listenerto nervous titters from the crowd. "And I think, mostlylikely, he did it to you, too, " he says, pointing at anaudience member three rows behind the first. 

By the time Stallman has a third audience member pickedout, the titters have given away to general laughter. The gesture seems a bit staged, because it is. Still, when it comes time to wrap up the Xerox laser-printerstory, Stallman does so with a showman's flourish. "Heprobably did it to most of the people here in thisroom-except a few, maybe, who weren't born yet in1980, " Stallman says, drawing more laughs. "[That's]because he had promised to refuse to cooperate withjust about the entire population of the planet Earth. "

Stallman lets the comment sink in for a half-beat. "Hehad signed a nondisclosure agreement, " Stallman adds. 

Richard Matthew Stallman's rise from frustratedacademic to political leader over the last 20 yearsspeaks to many things. It speaks to Stallman's stubbornnature and prodigious will. It speaks to the clearlyarticulated vision and values of the free softwaremovement Stallman helped build. It speaks to thehigh-quality software programs Stallman has built, programs that have cemented Stallman's reputation as aprogramming legend. It speaks to the growing momentumof the GPL, a legal innovation that many Stallmanobservers see as his most momentous accomplishment. 

Most importantly, it speaks to the changing nature ofpolitical power in a world increasingly beholden tocomputer technology and the software programs thatpower that technology. 

Maybe that's why, even at a time when mosthigh-technology stars are on the wane, Stallman's starhas grown. Since launching the GNU Project in 1984, 5Stallman has been at turns ignored, satirized, vilified, and attacked-both from within and without thefree software movement. Through it all, the GNU Projecthas managed to meet its milestones, albeit with a fewnotorious delays, and stay relevant in a softwaremarketplace several orders of magnitude more complexthan the one it entered 18 years ago. So too has thefree software ideology, an ideology meticulouslygroomed by Stallman himself. 

To understand the reasons behind this currency, ithelps to examine Richard Stallman both in his own wordsand in the words of the people who have collaboratedand battled with him along the way. The RichardStallman character sketch is not a complicated one. Ifany person exemplifies the old adage "what you see iswhat you get, " it's Stallman. 

"I think if you want to understand Richard Stallman thehuman being, you really need to see all of the parts asa consistent whole, " advises Eben Moglen, legal counselto the Free Software Foundation and professor of law atColumbia University Law School. "All those personaleccentricities that lots of people see as obstacles togetting to know Stallman really are Stallman: Richard'sstrong sense of personal frustration, his enormoussense of principled ethical commitment, his inabilityto compromise, especially on issues he considersfundamental. These are all the very reasons Richard didwhat he did when he did. "

Explaining how a journey that started with a laserprinter would eventually lead to a sparring match withthe world's richest corporation is no easy task. Itrequires a thoughtful examination of the forces thathave made software ownership so important in today'ssociety. It also requires a thoughtful examination of aman who, like many political leaders before him, understands the malleability of human memory. Itrequires an ability to interpret the myths andpolitically laden code words that have built up aroundStallman over time. Finally, it requires anunderstanding of Stallman's genius as a programmer andhis failures and successes in translating that geniusto other pursuits. 

When it comes to offering his own summary of thejourney, Stallman acknowledges the fusion ofpersonality and principle observed by Moglen. "Stubbornness is my strong suit, " he says. "Most peoplewho attempt to do anything of any great difficultyeventually get discouraged and give up. I never gave up. "

He also credits blind chance. Had it not been for thatrun-in over the Xerox laser printer, had it not beenfor the personal and political conflicts that closedout his career as an MIT employee, had it not been fora half dozen other timely factors, Stallman finds itvery easy to picture his life following a differentcareer path. That being said, Stallman gives thanks tothe forces and circumstances that put him in theposition to make a difference. 

"I had just the right skills, " says Stallman, summingup his decision for launching the GNU Project to theaudience. "Nobody was there but me, so I felt like, `I'm elected. I have to work on this. If not me, who?'" Endnotes

1. Actually, the GPL's powers are not quite thatpotent. According to section 10 of the GNU GeneralPublic License, Version 2 (1991), the viral nature ofthe license depends heavily on the Free SoftwareFoundation's willingness to view a program as aderivative work, not to mention the existing licensethe GPL would replace. 

If you wish to incorporate parts of the Program intoother free programs whose distribution conditions aredifferent, write to the author to ask for permission. For software that is copyrighted by the Free SoftwareFoundation, write to the Free Software Foundation; wesometimes make exceptions for this. Our decision willbe guided by the two goals of preserving the freestatus of all derivatives of our free software and ofpromoting the sharing and reuse of software generally. 

"To compare something to a virus is very harsh, " saysStallman. "A spider plant is a more accuratecomparison; it goes to another place if you activelytake a cutting. "

For more information on the GNU General Public License, visit

[http://www. Gnu. Org/copyleft/gpl. Html. ]

A Portrait of the Hacker as a Young Man

Richard Stallman's mother, Alice Lippman, stillremembers the moment she realized her son had a special gift. 

"I think it was when he was eight, " Lippman recalls. 

The year was 1961, and Lippman, a recently divorcedsingle mother, was wiling away a weekend afternoonwithin the family's tiny one-bedroom apartment onManhattan's Upper West Side. Leafing through a copy ofScientific American, Lippman came upon her favoritesection, the Martin Gardner-authored column titled"Mathematical Games. " A substitute art teacher, Lippmanalways enjoyed Gardner's column for the brain-teasersit provided. With her son already ensconced in a bookon the nearby sofa, Lippman decided to take a crack atsolving the week's feature puzzle. 

"I wasn't the best person when it came to solving thepuzzles, " she admits. "But as an artist, I found theyreally helped me work through conceptual barriers. "

Lippman says her attempt to solve the puzzle met animmediate brick wall. About to throw the magazine downin disgust, Lippman was surprised by a gentle tug onher shirt sleeve. 

"It was Richard, " she recalls, "He wanted to know if Ineeded any help. "

Looking back and forth, between the puzzle and her son, Lippman says she initially regarded the offer withskepticism. "I asked Richard if he'd read themagazine, " she says. "He told me that, yes, he had andwhat's more he'd already solved the puzzle. The nextthing I know, he starts explaining to me how to solve it. "

Hearing the logic of her son's approach, Lippman'sskepticism quickly gave way to incredulity. "I mean, Ialways knew he was a bright boy, " she says, "but thiswas the first time I'd seen anything that suggested howadvanced he really was. "

Thirty years after the fact, Lippman punctuates thememory with a laugh. "To tell you the truth, I don'tthink I ever figured out how to solve that puzzle, " shesays. "All I remember is being amazed he knew the answer. "

Seated at the dining-room table of her second Manhattanapartment-the same spacious three-bedroom complex sheand her son moved to following her 1967 marriage toMaurice Lippman, now deceased-Alice Lippman exudes aJewish mother's mixture of pride and bemusement whenrecalling her son's early years. The nearby dining-roomcredenza offers an eight-by-ten photo of Stallmanglowering in full beard and doctoral robes. The imagedwarfs accompanying photos of Lippman's nieces andnephews, but before a visitor can make too much of it, Lippman makes sure to balance its prominent placementwith an ironic wisecrack. 

"Richard insisted I have it after he received hishonorary doctorate at the University of Glasgow, " saysLippman. "He said to me, `Guess what, mom? It's thefirst graduation I ever attended. '"1

Such comments reflect the sense of humor that comeswith raising a child prodigy. Make no mistake, forevery story Lippman hears and reads about her son'sstubbornness and unusual behavior, she can deliver atleast a dozen in return. 

"He used to be so conservative, " she says, throwing upher hands in mock exasperation. "We used to have theworst arguments right here at this table. I was part ofthe first group of public city school teachers thatstruck to form a union, and Richard was very angry withme. He saw unions as corrupt. He was also very opposedto social security. He thought people could make muchmore money investing it on their own. Who knew thatwithin 10 years he would become so idealistic? All Iremember is his stepsister coming to me and saying, `What is he going to be when he grows up? A fascist?'"

As a single parent for nearly a decade-she andRichard's father, Daniel Stallman, were married in1948, divorced in 1958, and split custody of their sonafterwards-Lippman can attest to her son's aversion toauthority. She can also attest to her son's lust forknowledge. It was during the times when the two forcesintertwined, Lippman says, that she and her sonexperienced their biggest battles. 

"It was like he never wanted to eat, " says Lippman, recalling the behavior pattern that set in around ageeight and didn't let up until her son's high-schoolgraduation in 1970. "I'd call him for dinner, and he'dnever hear me. I'd have to call him 9 or 10 times justto get his attention. He was totally immersed. "

Stallman, for his part, remembers things in a similarfashion, albeit with a political twist. 

"I enjoyed reading, " he says. "If I wanted to read, andmy mother told me to go to the kitchen and eat or go tosleep, I wasn't going to listen. I saw no reason why Icouldn't read. No reason why she should be able to tellme what to do, period. Essentially, what I had readabout, ideas such as democracy and individual freedom, I applied to myself. I didn't see any reason to excludechildren from these principles. "

The belief in individual freedom over arbitraryauthority extended to school as well. Two years aheadof his classmates by age 11, Stallman endured all theusual frustrations of a gifted public-school student. It wasn't long after the puzzle incident that hismother attended the first in what would become a longstring of parent-teacher conferences. 

"He absolutely refused to write papers, " says Lippman, recalling an early controversy. "I think the last paperhe wrote before his senior year in high school was anessay on the history of the number system in the westfor a fourth-grade teacher. "

Gifted in anything that required analytical thinking, Stallman gravitated toward math and science at theexpense of his other studies. What some teachers saw assingle-mindedness, however, Lippman saw as impatience. Math and science offered simply too much opportunity tolearn, especially in comparison to subjects andpursuits for which her son seemed less naturallyinclined. Around age 10 or 11, when the boys inStallman's class began playing a regular game of touchfootball, she remembers her son coming home in a rage. "He wanted to play so badly, but he just didn't havethe coordination skills, " Lippman recalls. "It made himso angry. "

The anger eventually drove her son to focus on math andscience all the more. Even in the realm of science, however, her son's impatience could be problematic. Poring through calculus textbooks by age seven, Stallman saw little need to dumb down his discourse foradults. Sometime, during his middle-school years, Lippman hired a student from nearby Columbia Universityto play big brother to her son. The student left thefamily's apartment after the first session and nevercame back. "I think what Richard was talking about wentover his head, " Lippman speculates. 

Another favorite maternal anecdote dates back to theearly 1960s, shortly after the puzzle incident. Aroundage seven, two years after the divorce and relocationfrom Queens, Richard took up the hobby of launchingmodel rockets in nearby Riverside Drive Park. Whatstarted as aimless fun soon took on an earnest edge asher son began recording the data from each launch. Likethe interest in mathematical games, the pursuit drewlittle attention until one day, just before a majorNASA launch, Lippman checked in on her son to see if hewanted to watch. 

"He was fuming, " Lippman says. "All he could say to mewas, `But I'm not published yet. ' Apparently he hadsomething that he really wanted to show NASA. "

Such anecdotes offer early evidence of the intensitythat would become Stallman's chief trademark throughoutlife. When other kids came to the table, Stallmanstayed in his room and read. When other kids playedJohnny Unitas, Stallman played Werner von Braun. "I wasweird, " Stallman says, summing up his early yearssuccinctly in a 1999 interview. "After a certain age, the only friends I had were teachers. "See Michael Gross, "Richard Stallman:High SchoolMisfit, Symbol of Free Software, MacArthur-certifiedGenius" (1999). This interview is one of the mostcandid Stallman interviews on the record. I recommendit highly. 

http://www. Mgross. Com/interviews/stallman1. Html

Although it meant courting more run-ins at school, Lippman decided to indulge her son's passion. By age12, Richard was attending science camps during thesummer and private school during the school year. Whena teacher recommended her son enroll in the ColumbiaScience Honors Program, a post-Sputnik program designedfor gifted middle- and high-school students in New YorkCity, Stallman added to his extracurriculars and wassoon commuting uptown to the Columbia University campuson Saturdays. 

Dan Chess, a fellow classmate in the Columbia ScienceHonors Program, recalls Richard Stallman seeming a bitweird even among the students who shared a similar lustfor math and science. "We were all geeks and nerds, buthe was unusually poorly adjusted, " recalls Chess, now amathematics professor at Hunter College. "He was alsosmart as shit. I've known a lot of smart people, but Ithink he was the smartest person I've ever known. "

Seth Breidbart, a fellow Columbia Science HonorsProgram alumnus, offers bolstering testimony. Acomputer programmer who has kept in touch with Stallmanthanks to a shared passion for science fiction andscience-fiction conventions, he recalls the15-year-old, buzz-cut-wearing Stallman as "scary, "especially to a fellow 15-year-old. 

"It's hard to describe, " Breidbart says. "It wasn'tlike he was unapproachable. He was just very intense. [He was] very knowledgeable but also very hardheaded insome ways. "

Such descriptions give rise to speculation: arejudgment-laden adjectives like "intense" and"hardheaded" simply a way to describe traits that todaymight be categorized under juvenile behavioraldisorder? A December, 2001, Wired magazine articletitled "The Geek Syndrome" paints the portrait ofseveral scientifically gifted children diagnosed withhigh-functioning autism or Asperger Syndrome. In manyways, the parental recollections recorded in the Wiredarticle are eerily similar to the ones offered byLippman. Even Stallman has indulged in psychiatricrevisionism from time to time. During a 2000 profilefor the Toronto Star, Stallman described himself to aninterviewer as "borderline autistic, "See Judy Steed, Toronto Star, BUSINESS, (October 9, 2000): C03. His vision of free software and socialcooperation stands in stark contrast to the isolatednature of his private life. A Glenn Gould-likeeccentric, the Canadian pianist was similarlybrilliant, articulate, and lonely. Stallman considershimself afflicted, to some degree, by autism: acondition that, he says, makes it difficult for him tointeract with people. A description that goes a long way toward explaining alifelong tendency toward social and emotional isolationand the equally lifelong effort to overcome it. 

Such speculation benefits from the fast and loosenature of most so-called " behavioral disorders"nowadays, of course. As Steve Silberman, author of "The Geek Syndrome, " notes, American psychiatrists haveonly recently come to accept Asperger Syndrome as avalid umbrella term covering a wide set of behavioraltraits. The traits range from poor motor skills andpoor socialization to high intelligence and an almostobsessive affinity for numbers, computers, and ordered systems. See SteveSilberman, "The Geek Syndrome, " Wired(December, 2001). Reflecting on the broad nature of this umbrella, Stallman says its possible that, if born 40 yearslater, he might have merited just such a diagnosis. Then again, so would many of his computer-world colleagues. 

"It's possible I could have had something like that, "he says. "On the other hand, one of the aspects of thatsyndrome is difficulty following rhythms. I can dance. In fact, I love following the most complicated rhythms. It's not clear cut enough to know. "

Chess, for one, rejects such attempts atback-diagnosis. "I never thought of him [as] havingthat sort of thing, " he says. "He was just veryunsocialized, but then, we all were. "

Lippman, on the other hand, entertains the possibility. She recalls a few stories from her son's infancy, however, that provide fodder for speculation. Aprominent symptom of autism is an oversensitivity tonoises and colors, and Lippman recalls two anecdotesthat stand out in this regard. "When Richard was aninfant, we'd take him to the beach, " she says. "Hewould start screaming two or three blocks before wereached the surf. It wasn't until the third time thatwe figured out what was going on: the sound of the surfwas hurting his ears. " She also recalls a similarscreaming reaction in relation to color: "My mother hadbright red hair, and every time she'd stoop down topick him up, he'd let out a wail. "

In recent years, Lippman says she has taken to readingbooks about autism and believes that such episodes weremore than coincidental. "I do feel that Richard hadsome of the qualities of an autistic child, " she says. "I regret that so little was known about autism back then. "

Over time, however, Lippman says her son learned toadjust. By age seven, she says, her son had become fondof standing at the front window of subway trains, mapping out and memorizing the labyrinthian system ofrailroad tracks underneath the city. It was a hobbythat relied on an ability to accommodate the loudnoises that accompanied each train ride. "Only theinitial noise seemed to bother him, " says Lippman. "Itwas as if he got shocked by the sound but his nerveslearned how to make the adjustment. "

For the most part, Lippman recalls her son exhibitingthe excitement, energy, and social skills of any normalboy. It wasn't until after a series of traumatic eventsbattered the Stallman household, she says, that her sonbecame introverted and emotionally distant. 

The first traumatic event was the divorce of Alice andDaniel Stallman, Richard's father. Although Lippmansays both she and her ex-husband tried to prepare theirson for the blow, she says the blow was devastatingnonetheless. "He sort of didn't pay attention when wefirst told him what was happening, " Lippman recalls. "But the reality smacked him in the face when he and Imoved into a new apartment. The first thing he saidwas, `Where's Dad's furniture?'"

For the next decade, Stallman would spend his weekdaysat his mother's apartment in Manhattan and his weekendsat his father's home in Queens. The shuttling back andforth gave him a chance to study a pair of contrastingparenting styles that, to this day, leaves Stallmanfirmly opposed to the idea of raising children himself. Speaking about his father, a World War II vet whopassed away in early 2001, Stallman balances respectwith anger. On one hand, there is the man whose moralcommitment led him to learn French just so he could bemore helpful to Allies when they'd finally come. On theother hand, there was the parent who always knew how tocraft a put-down for cruel effect. Regrettably, I did not get a chance tointerview DanielStallman for this book. During the early research forthis book, Stallman informed me that his fathersuffered from Alzheimer's. When I resumed research inlate 2001, I learned, sadly, that Daniel Stallman haddied earlier in the year. 

"My father had a horrible temper, " Stallman says. "Henever screamed, but he always found a way to criticizeyou in a cold, designed-to-crush way. "

As for life in his mother's apartment, Stallman is lessequivocal. "That was war, " he says. "I used to say inmy misery, `I want to go home, ' meaning to thenonexistent place that I'll never have. "

For the first few years after the divorce, Stallmanfound the tranquility that eluded him in the home ofhis paternal grandparents. Then, around age 10 hisgrandparents passed away in short succession. ForStallman, the loss was devastating. "I used to go andvisit and feel I was in a loving, gentle environment, "Stallman recalls. "It was the only place I ever foundone, until I went away to college. "

Lippman lists the death of Richard's paternalgrandparents as the second traumatic event. "It reallyupset him, " she says. He was very close to both hisgrandparents. Before they died, he was very outgoing, almost a leader-of-the-pack type with the other kids. After they died, he became much more emotionally withdrawn. "

From Stallman's perspective, the emotional withdrawalwas merely an attempt to deal with the agony ofadolescence. Labeling his teenage years a "purehorror, " Stallman says he often felt like a deaf personamid a crowd of chattering music listeners. 

"I often had the feeling that I couldn't understandwhat other people were saying, " says Stallman, recalling the emotional bubble that insulated him fromthe rest of the adolescent and adult world. "I couldunderstand the words, but something was going onunderneath the conversations that I didn't understand. I couldn't understand why people were interested in thethings other people said. "

For all the agony it produced, adolescence would have aencouraging effect on Stallman's sense ofindividuality. At a time when most of his classmateswere growing their hair out, Stallman preferred to keephis short. At a time when the whole teenage world waslistening to rock and roll, Stallman preferredclassical music. A devoted fan of science fiction, Madmagazine, and late-night TV, Stallman cultivated adistinctly off-the-wall personality that fed off theincomprehension of parents and peers alike. 

"Oh, the puns, " says Lippman, still exasperated by thememory of her son's teenage personality. "There wasn'ta thing you could say at the dinner table that hecouldn't throw back at you as a pun. "

Outside the home, Stallman saved the jokes for theadults who tended to indulge his gifted nature. One ofthe first was a summer-camp counselor who handedStallman a print-out manual for the IBM 7094 computerduring his 12th year. To a preteenager fascinated withnumbers and science, the gift was a godsend. Stallman, an atheist, wouldprobably quibble with thisdescription. Suffice it to say, it was somethingStallman welcomed. See previous note 1: "As soon as Iheard about computers, I wanted to see one and playwith one. " By the end of summer, Stallman was writing out paperprograms according to the 7094's internalspecifications, anxiously anticipating getting a chanceto try them out on a real machine. 

With the first personal computer still a decade away, Stallman would be forced to wait a few years beforegetting access to his first computer. His first chancefinally came during his junior year of high school. Hired on at the IBM New York Scientific Center, anow-defunct research facility in downtown Manhattan, Stallman spent the summer after high-school graduationwriting his first program, a pre-processor for the 7094written in the programming language PL/I. "I firstwrote it in PL/I, then started over in assemblerlanguage when the PL/I program was too big to fit inthe computer, " he recalls. 

After that job at the IBM Scientific Center, Stallmanhad held a laboratory-assistant position in the biologydepartment at Rockefeller University. Although he wasalready moving toward a career in math or physics, Stallman's analytical mind impressed the lab directorenough that a few years after Stallman departed forcollege, Lippman received an unexpected phone call. "Itwas the professor at Rockefeller, " Lippman says. "Hewanted to know how Richard was doing. He was surprisedto learn that he was working in computers. He'd alwaysthought Richard had a great future ahead of him as a biologist. "

Stallman's analytical skills impressed faculty membersat Columbia as well, even when Stallman himself becamea target of their ire. "Typically once or twice an hour[Stallman] would catch some mistake in the lecture, "says Breidbart. "And he was not shy about letting theprofessors know it immediately. It got him a lot ofrespect but not much popularity. "

Hearing Breidbart's anecdote retold elicits a wry smilefrom Stallman. "I may have been a bit of a jerksometimes, " he admits. "But I found kindred spiritsamong the teachers, because they, too, liked to learn. Kids, for the most part, didn't. At least not in thesame way. "

Hanging out with the advanced kids on Saturdaynevertheless encouraged Stallman to think more aboutthe merits of increased socialization. With collegefast approaching, Stallman, like many in his ColumbiaScience Honors Program, had narrowed his list ofdesired schools down to two choices: Harvard and MIT. Hearing of her son's desire to move on to the IvyLeague, Lippman became concerned. As a 15-year-oldhigh-school junior, Stallman was still having run-inswith teachers and administrators. Only the year before, he had pulled straight A's in American History, Chemistry, French, and Algebra, but a glaring F inEnglish reflected the ongoing boycott of writingassignments. Such miscues might draw a knowing chuckleat MIT, but at Harvard, they were a red flag. 

During her son's junior year, Lippman says shescheduled an appointment with a therapist. Thetherapist expressed instant concern over Stallman'sunwillingness to write papers and his run-ins withteachers. Her son certainly had the intellectualwherewithal to succeed at Harvard, but did he have thepatience to sit through college classes that required aterm paper? The therapist suggested a trial run. IfStallman could make it through a full year in New YorkCity public schools, including an English class thatrequired term papers, he could probably make it atHarvard. Following the completion of his junior year, Stallman promptly enrolled in summer school at Louis D. Brandeis High School, a public school located on 84thStreet, and began making up the mandatory art classeshe had shunned earlier in his high-school career. 

By fall, Stallman was back within the mainstreampopulation of New York City high-school students. Itwasn't easy sitting through classes that seemedremedial in comparison with his Saturday studies atColumbia, but Lippman recalls proudly her son's abilityto toe the line. 

"He was forced to kowtow to a certain degree, but hedid it, " Lippman says. "I only got called in once, which was a bit of a miracle. It was the calculusteacher complaining that Richard was interrupting hislesson. I asked how he was interrupting. He saidRichard was always accusing the teacher of using afalse proof. I said, `Well, is he right?' The teachersaid, `Yeah, but I can't tell that to the class. Theywouldn't understand. '"

By the end of his first semester at Brandeis, thingswere falling into place. A 96 in English wiped awaymuch of the stigma of the 60 earned 2 years before. Forgood measure, Stallman backed it up with top marks inAmerican History, Advanced Placement Calculus, andMicrobiology. The crowning touch was a perfect 100 inPhysics. Though still a social outcast, Stallmanfinished his 11 months at Brandeis as the fourth-rankedstudent in a class of 789. 

<Graphic file:/home/craigm/books/free_0306. Png>

Stallman's senior-year transcript at Louis D. BrandeisH. S. , November, 1969. Note turnaround in English classperformance. "He was forced to kowtow to a certaindegree, " says his mother, "but he did it. "

Outside the classroom, Stallman pursued his studieswith even more diligence, rushing off to fulfill hislaboratory-assistant duties at Rockefeller Universityduring the week and dodging the Vietnam protesters onhis way to Saturday school at Columbia. It was there, while the rest of the Science Honors Program studentssat around discussing their college choices, thatStallman finally took a moment to participate in thepreclass bull session. 

Recalls Breidbart, "Most of the students were going toHarvard and MIT, of course, but you had a few going toother Ivy League schools. As the conversation circledthe room, it became apparent that Richard hadn't saidanything yet. I don't know who it was, but somebody gotup the courage to ask him what he planned to do. "

Thirty years later, Breidbart remembers the momentclearly. As soon as Stallman broke the news that he, too, would be attending Harvard University in the fall, an awkward silence filled the room. Almost as if oncue, the corners of Stallman's mouth slowly turnedupward into a self-satisfied smile. 

Says Breidbart, "It was his silent way of saying, `That's right. You haven't got rid of me yet. '"

Impeach God

Although their relationship was fraught with tension, Richard Stallman would inherit one noteworthy traitfrom his mother: a passion for progressive politics. 

It was an inherited trait that would take severaldecades to emerge, however. For the first few years ofhis life, Stallman lived in what he now admits was a"political vacuum. "See Michael Gross, "Richard Stallman: High SchoolMisfit, Symbol of Free Software, MacArthur-certifiedGenius" (1999). Like most Americans during the Eisenhower age, theStallman family spent the 50s trying to recapture thenormalcy lost during the wartime years of the 1940s. 

"Richard's father and I were Democrats but happy enoughto leave it at that, " says Lippman, recalling thefamily's years in Queens. "We didn't get involved muchin local or national politics. "

That all began to change, however, in the late 1950swhen Alice divorced Daniel Stallman. The move back toManhattan represented more than a change of address; itrepresented a new, independent identity and a jarringloss of tranquility. 

"I think my first taste of political activism came whenI went to the Queens public library and discoveredthere was only a single book on divorce in the wholelibrary, " recalls Lippman. "It was very controlled bythe Catholic church, at least in Elmhurst, where welived. I think that was the first inkling I had of theforces that quietly control our lives. "

Returning to her childhood neighborhood, Manhattan'sUpper West Side, Lippman was shocked by the changesthat had taken place since her departure to HunterCollege a decade and a half before. The skyrocketingdemand for postwar housing had turned the neighborhoodinto a political battleground. On one side stood thepro-development city-hall politicians and businessmenhoping to rebuild many of the neighborhood's blocks toaccommodate the growing number of white-collar workersmoving into the city. On the other side stood the poorIrish and Puerto Rican tenants who had found anaffordable haven in the neighborhood. 

At first, Lippman didn't know which side to choose. Asa new resident, she felt the need for new housing. As asingle mother with minimal income, however, she sharedthe poorer tenants' concern over the growing number ofdevelopment projects catering mainly to wealthyresidents. Indignant, Lippman began looking for ways tocombat the political machine that was attempting toturn her neighborhood into a clone of the Upper East Side. 

Lippman says her first visit to the local Democraticparty headquarters came in 1958. Looking for a day-carecenter to take care of her son while she worked, shehad been appalled by the conditions encountered at oneof the city-owned centers that catered to low-incomeresidents. "All I remember is the stench of rottenmilk, the dark hallways, the paucity of supplies. I hadbeen a teacher in private nursery schools. The contrastwas so great. We took one look at that room and left. That stirred me up. "

The visit to the party headquarters proveddisappointing, however. Describing it as "theproverbial smoke-filled room, " Lippman says she becameaware for the first time that corruption within theparty might actually be the reason behind the city'sthinly disguised hostility toward poor residents. Instead of going back to the headquarters, Lippmandecided to join up with one of the many clubs aimed atreforming the Democratic party and ousting the lastvestiges of the Tammany Hall machine. Dubbed theWoodrow Wilson/FDR Reform Democratic Club, Lippman andher club began showing up at planning and city-councilmeetings, demanding a greater say. 

"Our primary goal was to fight Tammany Hall, CarmineDeSapio and his henchman, "Carmine DeSapio holds the dubious distinction ofbeingthe first Italian-American boss of Tammany Hall, theNew York City political machine. For more informationon DeSapio and the politics of post-war New York, seeJohn Davenport, "Skinning the Tiger: Carmine DeSapioand the End of the Tammany Era, " New York Affairs(1975): 3:1. Says Lippman. "I was the representative to the citycouncil and was very much involved in creating a viableurban-renewal plan that went beyond simply adding moreluxury housing to the neighborhood. "

Such involvement would blossom into greater politicalactivity during the 1960s. By 1965, Lippman had becomean "outspoken" supporter for political candidates likeWilliam Fitts Ryan, a Democratic elected to Congresswith the help of reform clubs and one of the first U. S. Representatives to speak out against the Vietnam War. 

It wasn't long before Lippman, too, was an outspokenopponent of U. S. Involvement in Indochina. "I wasagainst the Vietnam war from the time Kennedy senttroops, " she says. "I had read the stories by reportersand journalists sent to cover the early stages of theconflict. I really believed their forecast that itwould become a quagmire. "

Such opposition permeated the Stallman-Lippmanhousehold. In 1967, Lippman remarried. Her new husband, Maurice Lippman, a major in the Air National Guard, resigned his commission to demonstrate his oppositionto the war. Lippman's stepson, Andrew Lippman, was atMIT and temporarily eligible for a student deferment. Still, the threat of induction should that defermentdisappear, as it eventually did, made the risk of U. S. Escalation all the more immediate. Finally, there wasRichard who, though younger, faced the prospect ofchoosing between Vietnam or Canada when the war lastedinto the 1970s. 

"Vietnam was a major issue in our household, " saysLippman. "We talked about it constantly: what would wedo if the war continued, what steps Richard or hisstepbrother would take if they got drafted. We were allopposed to the war and the draft. We really thought itwas immoral. "

For Stallman, the Vietnam War elicited a complexmixture of emotions: confusion, horror, and, ultimately, a profound sense of political impotence. Asa kid who could barely cope in the mild authoritarianuniverse of private school, Stallman experienced ashiver whenever the thought of Army boot camp presented itself. 

"I was devastated by the fear, but I couldn't imaginewhat to do and didn't have the guts to go demonstrate, "recalls Stallman, whose March 18th birthday earned hima dreaded low number in the draft lottery when thefederal government finally eliminated collegedeferments in 1971. "I couldn't envision moving toCanada or Sweden. The idea of getting up by myself andmoving somewhere. How could I do that? I didn't knowhow to live by myself. I wasn't the kind of person whofelt confident in approaching things like that. "

Stallman says he was both impressed and shamed by thefamily members who did speak out. Recalling a bumpersticker on his father's car likening the My Laimassacre to similar Nazi atrocities in World War II, hesays he was "excited" by his father's gesture ofoutrage. "I admired him for doing it, " Stallman says. "But I didn't imagine that I could do anything. I wasafraid that the juggernaut of the draft was going todestroy me. "

Although descriptions of his own unwillingness to speakout carry a tinge of nostalgic regret, Stallman says hewas ultimately turned off by the tone and direction ofthe anti-war movement. Like other members of theScience Honors Program, he saw the weekenddemonstrations at Columbia as little more than adistracting spectacle. Chess, another Columbia Science Honors Program alum, describes the protests as "background noise. " "We wereall political, " he says, "but the SHP was imporant. Wewould never have skipped it for a demonstration. " Ultimately, Stallman says, the irrational forcesdriving the anti-war movement became indistinguishablefrom the irrational forces driving the rest of youthculture. Instead of worshiping the Beatles, girls inStallman's age group were suddenly worshipingfirebrands like Abbie Hoffman and Jerry Rubin. To a kidalready struggling to comprehend his teenage peers, escapist slogans like "make love not war" had ataunting quality. Not only was it a reminder thatStallman, the short-haired outsider who hated rock 'n'roll, detested drugs, and didn't participate in campusdemonstrations, wasn't getting it politically; hewasn't "getting it" sexually either. 

"I didn't like the counter culture much, " Stallmanadmits. "I didn't like the music. I didn't like thedrugs. I was scared of the drugs. I especially didn'tlike the anti-intellectualism, and I didn't like theprejudice against technology. After all, I loved acomputer. And I didn't like the mindlessanti-Americanism that I often encountered. There werepeople whose thinking was so simplistic that if theydisapproved of the conduct of the U. S. In the VietnamWar, they had to support the North Vietnamese. Theycouldn't imagine a more complicated position, I guess. "

Such comments alleviate feelings of timidity. They alsounderline a trait that would become the key toStallman's own political maturation. For Stallman, political confidence was directly proportionate topersonal confidence. By 1970, Stallman had becomeconfident in few things outside the realm of math andscience. Nevertheless, confidence in math gave himenough of a foundation to examine the anti-war movementin purely logical terms. In the process of doing so, Stallman had found the logic wanting. Although opposedto the war in Vietnam, Stallman saw no reason todisavow war as a means for defending liberty orcorrecting injustice. Rather than widen the breachbetween himself and his peers, however, Stallmanelected to keep the analysis to himself. 

In 1970, Stallman left behind the nightly dinnertimeconversations about politics and the Vietnam War as hedeparted for Harvard. Looking back, Stallman describesthe transition from his mother's Manhattan apartment tolife in a Cambridge dorm as an "escape. " Peers whowatched Stallman make the transition, however, sawlittle to suggest a liberating experience. 

"He seemed pretty miserable for the first while atHarvard, " recalls Dan Chess, a classmate in the ScienceHonors Program who also matriculated at Harvard. "Youcould tell that human interaction was really difficultfor him, and there was no way of avoiding it atHarvard. Harvard was an intensely social kind of place. "

To ease the transition, Stallman fell back on hisstrengths: math and science. Like most members of theScience Honors Program, Stallman breezed through thequalifying exam for Math 55, the legendary "boot camp"class for freshman mathematics "concentrators" atHarvard. Within the class, members of the ScienceHonors Program formed a durable unit. "We were the mathmafia, " says Chess with a laugh. "Harvard was nothing, at least compared with the SHP. "

To earn the right to boast, however, Stallman, Chess, and the other SHP alumni had to get through Math 55. Promising four years worth of math in two semesters, the course favored only the truly devout. "It was anamazing class, " says David Harbater, a former "mathmafia" member and now a professor of mathematics at theUniversity of Pennsylvania. "It's probably safe to saythere has never been a class for beginning collegestudents that was that intense and that advanced. Thephrase I say to people just to get it across is that, among other things, by the second semester we werediscussing the differential geometry of Banachmanifolds. That's usually when their eyes bug out, because most people don't start talking about Banachmanifolds until their second year of graduate school. "

Starting with 75 students, the class quickly melteddown to 20 by the end of the second semester. Of that20, says Harbater, "only 10 really knew what they weredoing. " Of that 10, 8 would go on to become futuremathematics professors, 1 would go on to teach physics. 

"The other one, " emphasizes Harbater, "was Richard Stallman. "

Seth Breidbart, a fellow Math 55 classmate, remembersStallman distinguishing himself from his peers even then. 

"He was a stickler in some very strange ways, " saysBreidbart. There is a standard technique in math whicheverybody does wrong. It's an abuse of notation whereyou have to define a function for something and whatyou do is you define a function and then you prove thatit's well defined. Except the first time he did andpresented it, he defined a relation and proved thatit's a function. It's the exact same proof, but he usedthe correct terminology, which no one else did. That'sjust the way he was. "

It was in Math 55 that Richard Stallman began tocultivate a reputation for brilliance. Breidbartagrees, but Chess, whose competitive streak refused toyield, says the realization that Stallman might be thebest mathematician in the class didn't set in until thenext year. "It was during a class on Real Analysis, which I took with Richard the next year, " says Chess, now a math professor at Hunter College. "I actuallyremember in a proof about complex valued measures thatRichard came up with an idea that was basically ametaphor from the calculus of variations. It was thefirst time I ever saw somebody solve a problem in abrilliantly original way. "

Chess makes no bones about it: watching Stallman'ssolution unfold on the chalkboard was a devastatingblow. As a kid who'd always taken pride in being thesmartest mathematician the room, it was like catching aglimpse of his own mortality. Years later, as Chessslowly came to accept the professional rank of agood-but-not-great mathematician, he had Stallman'ssophomore-year proof to look back on as a tauntingearly indicator. 

"That's the thing about mathematics, " says Chess. "Youdon't have to be a first-rank mathematician torecognize first-rate mathematical talent. I could tellI was up there, but I could also tell I wasn't at thefirst rank. If Richard had chosen to be amathematician, he would have been a first-rank mathematician. "

For Stallman, success in the classroom was balanced bythe same lack of success in the social arena. Even asother members of the math mafia gathered to take on theMath 55 problem sets, Stallman preferred to work alone. The same went for living arrangements. On the housingapplication for Harvard, Stallman clearly spelled outhis preferences. "I said I preferred an invisible, inaudible, intangible roommate, " he says. In a rarestroke of bureaucratic foresight, Harvard's housingoffice accepted the request, giving Stallman a one-roomsingle for his freshman year. 

Breidbart, the only math-mafia member to share a dormwith Stallman that freshman year, says Stallman slowlybut surely learned how to interact with other students. He recalls how other dorm mates, impressed byStallman's logical acumen, began welcoming his inputwhenever an intellectual debate broke out in the diningclub or dorm commons. 

"We had the usual bull sessions about solving theworld's problems or what would be the result ofsomething, " recalls Breidbart. "Say somebody discoversan immortality serum. What do you do? What are thepolitical results? If you give it to everybody, theworld gets overcrowded and everybody dies. If you limitit, if you say everyone who's alive now can have it buttheir children can't, then you end up with anunderclass of people without it. Richard was justbetter able than most to see the unforeseencircumstances of any decision. "

Stallman remembers the discussions vividly. "I wasalways in favor of immortality, " he says. "I wasshocked that most people regarded immortality as a badthing. How else would we be able to see what the worldis like 200 years from now?"

Although a first-rank mathematician and first-ratedebater, Stallman shied away from clear-cut competitiveevents that might have sealed his brilliant reputation. Near the end of freshman year at Harvard, Breidbartrecalls how Stallman conspicuously ducked the Putnamexam, a prestigious test open to math studentsthroughout the U. S. And Canada. In addition to givingstudents a chance to measure their knowledge inrelation to their peers, the Putnam served as a chiefrecruiting tool for academic math departments. According to campus legend, the top scorerautomatically qualified for a graduate fellowship atany school of his choice, including Harvard. 

Like Math 55, the Putnam was a brutal test of merit. Asix-hour exam in two parts, it seemed explicitlydesigned to separate the wheat from the chaff. Breidbart, a veteran of both the Science Honors Programand Math 55, describes it as easily the most difficulttest he ever took. "Just to give you an idea of howdifficult it was, " says Breidbart, "the top score was a120, and my score the first year was in the 30s. Thatscore was still good enough to place me 101st in the country. "

Surprised that Stallman, the best student in the class, had passed on the test, Breidbart says he and a fellowclassmate cornered him in the dining common anddemanded an explanation. "He said he was afraid of notdoing well, " Breidbart recalls. 

Breidbart and the friend quickly wrote down a fewproblems from memory and gave them to Stallman. "Hesolved all of them, " Breidbart says, "leading me toconclude that by not doing well, he either meant comingin second or getting something wrong. "

Stallman remembers the episode a bit differently. "Iremember that they did bring me the questions and it'spossible that I solved one of them, but I'm pretty sureI didn't solve them all, " he says. Nevertheless, Stallman agrees with Breidbart's recollection that fearwas the primary reason for not taking the test. Despitea demonstrated willingness to point out theintellectual weaknesses of his peers and professors inthe classroom, Stallman hated the notion ofhead-to-head competition. 

"It's the same reason I never liked chess, " saysStallman. "Whenever I'd play, I would become soconsumed by the fear of making a single mistake that Iwould start making stupid mistakes very early in thegame. The fear became a self-fulfilling prophecy. "

Whether such fears ultimately prompted Stallman to shyaway from a mathematical career is a moot issue. By theend of his freshman year at Harvard, Stallman had otherinterests pulling him away from the field. Computerprogramming, a latent fascination throughout Stallman'shigh-school years, was becoming a full-fledged passion. Where other math students sought occasional refuge inart and history classes, Stallman sought it in thecomputer-science laboratory. 

For Stallman, the first taste of real computerprogramming at the IBM New York Scientific Center hadtriggered a desire to learn more. "Toward the end of myfirst year at Harvard school, I started to have enoughcourage to go visit computer labs and see what theyhad. I'd ask them if they had extra copies of anymanuals that I could read. "

Taking the manuals home, Stallman would examine machinespecifications, compare them with other machines healready knew, and concoct a trial program, which hewould then bring back to the lab along with theborrowed manual. Although some labs balked at thenotion of a strange kid coming off the street andworking on the lab machinery, most recognizedcompetence when they saw it and let Stallman run theprograms he had created. 

One day, near the end of freshman year, Stallman heardabout a special laboratory near MIT. The laboratory waslocated on the ninth floor an off-campus building inTech Square, the newly built facility dedicated toadvanced research. According to the rumors, the labitself was dedicated to the cutting-edge science ofartificial intelligence and boasted the cutting-edgemachines and software programs to match. 

Intrigued, Stallman decided to pay a visit. 

The trip was short, about 2 miles on foot, 10 minutesby train, but as Stallman would soon find out, MIT andHarvard can feel like opposite poles of the sameplanet. With its maze-like tangle of interconnectedoffice buildings, the Institute's campus offered anaesthetic yin to Harvard's spacious colonial-villageyang. The same could be said for the student body, ageeky collection of ex-high school misfits known morefor its predilection for pranks than its politicallypowerful alumni. 

The yin-yang relationship extended to the AI Lab aswell. Unlike Harvard computer labs, there was nograd-student gatekeeper, no clipboard waiting list forterminal access, no explicit atmosphere of "look butdon't touch. " Instead, Stallman found only a collectionof open terminals and robotic arms, presumably theartifacts of some A. I. Experiment. 

Although the rumors said anybody could sit down at theterminals, Stallman decided to stick with the originalplan. When he encountered a lab employee, he asked ifthe lab had any spare manuals it could loan to aninquisitive student. "They had some, but a lot ofthings weren't documented, " Stallman recalls. "Theywere hackers after all. "

Stallman left with something even better than a manual:a job. Although he doesn't remember what the firstproject was, he does remember coming back to the AI Labthe next week, grabbing an open terminal and writingsoftware code. 

Looking back, Stallman sees nothing unusual in the AILab's willingness to accept an unproven outsider atfirst glance. "That's the way it was back then, " hesays. "That's the way it still is now. I'll hiresomebody when I meet him if I see he's good. Why wait?Stuffy people who insist on putting bureaucracy intoeverything really miss the point. If a person is good, he shouldn't have to go through a long, detailed hiringprocess; he should be sitting at a computer writing code. "

To get a taste of "bureaucratic and stuffy, " Stallmanneed only visit the computer labs at Harvard. There, access to the terminals was doled out according toacademic rank. As an undergrad, Stallman usually had tosign up or wait until midnight, about the time mostprofessors and grad students finished their daily workassignments. The waiting wasn't difficult, but it wasfrustrating. Waiting for a public terminal, knowing allthe while that a half dozen equally usable machineswere sitting idle inside professors' locked offices, seemed the height of illogic. Although Stallman paidthe occasional visit to the Harvard computer labs, hepreferred the more egalitarian policies of the AI Lab. "It was a breath of fresh air, " he says. "At the AILab, people seemed more concerned about work than status. "

Stallman quickly learned that the AI Lab's first-come, first-served policy owed much to the efforts of avigilant few. Many were holdovers from the days ofProject MAC, the Department of Defense-funded researchprogram that had given birth to the first time-shareoperating systems. A few were already legends in thecomputing world. There was Richard Greenblatt, thelab's in-house Lisp expert and author of MacHack, thecomputer chess program that had once humbled A. I. Critic Hubert Dreyfus. There was Gerald Sussman, original author of the robotic block-stacking programHACKER. And there was Bill Gosper, the in-house mathwhiz already in the midst of an 18-month hacking bendertriggered by the philosophical implications of thecomputer game LIFE. See Steven Levy, Hackers (Penguin USA [paperback], 1984): 144. Levy devotes about five pages to describingGosper's fascination with LIFE, a math-based softwaregame first created by British mathematician JohnConway. I heartily recommend this book as a supplement, perhaps even a prerequisite, to this one. 

Members of the tight-knit group called themselves "hackers. " Over time, they extended the "hacker"description to Stallman as well. In the process ofdoing so, they inculcated Stallman in the ethicaltraditions of the "hacker ethic . " To be a hacker meantmore than just writing programs, Stallman learned. Itmeant writing the best possible programs. It meantsitting at a terminal for 36 hours straight if that'swhat it took to write the best possible programs. Mostimportantly, it meant having access to the bestpossible machines and the most useful information atall times. Hackers spoke openly about changing theworld through software, and Stallman learned theinstinctual hacker disdain for any obstacle thatprevented a hacker from fulfilling this noble cause. Chief among these obstacles were poor software, academic bureaucracy, and selfish behavior. 

Stallman also learned the lore, stories of how hackers, when presented with an obstacle, had circumvented it increative ways. Stallman learned about " lock hacking, "the art of breaking into professors' offices to"liberate" sequestered terminals. Unlike their pamperedHarvard counterparts, MIT faculty members knew betterthan to treat the AI Lab's terminal as privateproperty. If a faculty member made the mistake oflocking away a terminal for the night, hackers werequick to correct the error. Hackers were equally quickto send a message if the mistake repeated itself. "Iwas actually shown a cart with a heavy cylinder ofmetal on it that had been used to break down the doorof one professor's office, "Gerald Sussman, an MIT faculty member and hackerwhosework at the AI Lab predates Stallman's, disputes thismemory. According to Sussman, the hackers never brokeany doors to retrieve terminals. Stallman says. 

Such methods, while lacking in subtlety, served apurpose. Although professors and administratorsoutnumbered hackers two-to-one inside the AI Lab, thehacker ethic prevailed. Indeed, by the time ofStallman's arrival at the AI Lab, hackers and the AILab administration had coevolved into something of asymbiotic relationship. In exchange for fixing themachines and keeping the software up and running, hackers earned the right to work on favorite petprojects. Often, the pet projects revolved aroundimproving the machines and software programs evenfurther. Like teenage hot-rodders, most hackers viewedtinkering with machines as its own form of entertainment. 

Nowhere was this tinkering impulse better reflectedthan in the operating system that powered the lab'scentral PDP-6 mini-computer. Dubbed ITS, short for theIncompatible Time Sharing system, the operating systemincorporated the hacking ethic into its very design. Hackers had built it as a protest to Project MAC'soriginal operating system, the Compatible Time SharingSystem, CTSS, and named it accordingly. At the time, hackers felt the CTSS design too restrictive, limitingprogrammers' power to modify and improve the program'sown internal architecture if needed. According to onelegend passed down by hackers, the decision to buildITS had political overtones as well. Unlike CTSS, whichhad been designed for the IBM 7094, ITS was builtspecifically for the PDP-6. In letting hackers writethe systems themselves, AI Lab administratorsguaranteed that only hackers would feel comfortableusing the PDP-6. In the feudal world of academicresearch, the gambit worked. Although the PDP-6 wasco-owned in conjunction with other departments, A. I. Researchers soon had it to themselves. 

ITS boasted features most commercial operating systemswouldn't offer for years, features such asmultitasking, debugging, and full-screen editingcapability. Using it and the PDP-6 as a foundation, theLab had been able to declare independence from ProjectMAC shortly before Stallman's arrival. I apologize for the whirlwind summary ofITS' genesis, an operating system many hackers still regard as theepitome of the hacker ethos. For more information onthe program's political significance, see SimsonGarfinkel, Architects of the Information Society:Thirty-Five Years of the Laboratory for ComputerScience at MIT (MIT Press, 1999). 

As an apprentice hacker, Stallman quickly becameenamored with ITS. Although forbidding to mostnewcomers, the program contained many built-in featuresthat provided a lesson in software development tohacker apprentices such as himself. 

"ITS had a very elegant internal mechanism for oneprogram to examine another, " says Stallman, recallingthe program. "You could examine all sorts of statusabout another program in a very clean, well-specified way. "

Using this feature, Stallman was able to watch howprograms written by hackers processed instructions asthey ran. Another favorite feature would allow themonitoring program to freeze the monitored program'sjob between instructions. In other operating systems, such a command would have resulted in half-computedgibberish or an automatic systems crash. In ITS, itprovided yet another way to monitor the step-by-step performance. 

"If you said, `Stop the job, ' it would always bestopped in user mode. It would be stopped between twouser-mode instructions, and everything about the jobwould be consistent for that point, " Stallman says. "Ifyou said, `Resume the job, ' it would continue properly. Not only that, but if you were to change the status ofthe job and then change it back, everything would beconsistent. There was no hidden status anywhere. "

By the end of 1970, hacking at the AI Lab had become aregular part of Stallman's weekly schedule. From Mondayto Thursday, Stallman devoted his waking hours to hisHarvard classes. As soon as Friday afternoon arrived, however, he was on the T, heading down to MIT for theweekend. Stallman usually timed his arrival to coincidewith the ritual food run. Joining five or six otherhackers in their nightly quest for Chinese food, hewould jump inside a beat-up car and head across theHarvard Bridge into nearby Boston. For the next twohours, he and his hacker colleagues would discusseverything from ITS to the internal logic of theChinese language and pictograph system. Followingdinner, the group would return to MIT and hack codeuntil dawn. 

For the geeky outcast who rarely associated with hishigh-school peers, it was a heady experience, suddenlyhanging out with people who shared the samepredilection for computers, science fiction, andChinese food. "I remember many sunrises seen from a carcoming back from Chinatown, " Stallman would recallnostalgically, 15 years after the fact in a speech atthe Swedish Royal Technical Institute. "It was actuallya very beautiful thing to see a sunrise, 'cause that'ssuch a calm time of day. It's a wonderful time of dayto get ready to go to bed. It's so nice to walk homewith the light just brightening and the birds startingto chirp; you can get a real feeling of gentlesatisfaction, of tranquility about the work that youhave done that night. "See Richard Stallman, "RMS lecture at KTH (Sweden), "(October 30, 1986). 

http://www. Gnu. Org/philosophy/stallman-kth. Html

The more Stallman hung out with the hackers, the morehe adopted the hacker worldview. Already committed tothe notion of personal liberty, Stallman began toinfuse his actions with a sense of communalresponsibility. When others violated the communal code, Stallman was quick to speak out. Within a year of hisfirst visit, Stallman was the one breaking into lockedoffices, trying to recover the sequestered terminalsthat belonged to the lab community as a whole. In truehacker fashion, Stallman also sought to make his ownpersonal contribution to the art of lock hacking. Oneof the most artful door-opening tricks, commonlyattributed to Greenblatt, involved bending a stiff wireinto a cane and attaching a loop of tape to the longend. Sliding the wire under the door, a hacker couldtwist and rotate the wire so that the long end touchedthe door knob. Provided the adhesive on the tape held, a hacker could open the doorknob with a few sharp twists. 

When Stallman tried the trick, he found it good butwanting in a few places. Getting the tape to stickwasn't always easy, and twisting the wire in a way thatturned the doorknob was similarly difficult. Stallmanremembered that the hallway ceiling possessed tilesthat could be slid away. Some hackers, in fact, hadused the false ceiling as a way to get around lockeddoors, an approach that generally covered theperpetrator in fiberglass but got the job done. 

Stallman considered an alternative approach. What if, instead of slipping a wire under the door, a hackerslid away one of the panels and stood over the door jamb?

Stallman took it upon himself to try it out. Instead ofusing a wire, Stallman draped out a long U-shaped loopof magnetic tape, fastening a loop of adhesive tape atthe base of the U. Standing over the door jamb, hedangled the tape until it looped under the doorknob. Lifting the tape until the adhesive fastened, he thenpulled on the left end of the tape, twisting thedoorknob counter-clockwise. Sure enough, the dooropened. Stallman had added a new twist to the art oflock hacking. 

"Sometimes you had to kick the door after you turnedthe door knob, " says Stallman, recalling the lingeringbugginess of the new method. "It took a little bit ofbalance to pull it off. "

Such activities reflected a growing willingness onStallman's part to speak and act out in defense ofpolitical beliefs. The AI Lab's spirit of direct actionhad proved inspirational enough for Stallman to breakout of the timid impotence of his teenage years. Breaking into an office to free a terminal wasn't thesame as taking part in a protest march, but it waseffective in ways that most protests weren't. It solvedthe problem at hand. 

By the time of his last years at Harvard, Stallman wasbeginning to apply the whimsical and irreverent lessonsof the AI Lab back at school. 

"Did he tell you about the snake?" his mother asks atone point during an interview. "He and his dorm matesput a snake up for student election. Apparently it gota considerable number of votes. "

Stallman verifies the snake candidacy with a fewcaveats. The snake was a candidate for election withinCurrier House, Stallman's dorm, not the campus-widestudent council. Stallman does remember the snakeattracting a fairly significant number of votes, thanksin large part to the fact that both the snake and itsowner both shared the same last name. "People may havevoted for it, because they thought they were voting forthe owner, " Stallman says. "Campaign posters said thatthe snake was `slithering for' the office. We also saidit was an `at large' candidate, since it had climbedinto the wall through the ventilating unit a few weeksbefore and nobody knew where it was. "

Running a snake for dorm council was just one ofseveral election-related pranks. In a later election, Stallman and his dorm mates nominated the housemaster's son. "His platform was mandatory retirement atage seven, " Stallman recalls. Such pranks paled incomparison to the fake-candidate pranks on the MITcampus, however. One of the most successfulfake-candidate pranks was a cat named Woodstock, whichactually managed to outdraw most of the humancandidates in a campus-wide election. "They neverannounced how many votes Woodstock got, and theytreated those votes as spoiled ballots, " Stallmanrecalls. "But the large number of spoiled ballots inthat election suggested that Woodstock had actuallywon. A couple of years later, Woodstock wassuspiciously run over by a car. Nobody knows if thedriver was working for the MIT administration. "Stallman says he had nothing to do with Woodstock'scandidacy, "but I admired it. "In an email shortly after this book went into itsfinaledit cycle, Stallman says he drew political inspirationfrom the Harvard campus as well. "In my first year ofHarvard, in a Chinese History class, I read the storyof the first revolt against the Chin dynasty, " he says. "The story is not reliable history, but it was verymoving. "

At the AI Lab, Stallman's political activities had asharper-edged tone. During the 1970s, hackers faced theconstant challenge of faculty members andadministrators pulling an end-run around ITS and itshacker-friendly design. One of the first attempts camein the mid-1970s, as more and more faculty membersbegan calling for a file security system to protectresearch data. Most other computer labs had installedsuch systems during late 1960s, but the AI Lab, throughthe insistence of Stallman and other hackers, remaineda security-free zone. 

For Stallman, the opposition to security was bothethical and practical. On the ethical side, Stallmanpointed out that the entire art of hacking relied onintellectual openness and trust. On the practical side, he pointed to the internal structure of ITS being builtto foster this spirit of openness, and any attempt toreverse that design required a major overhaul. 

"The hackers who wrote the Incompatible TimesharingSystem decided that file protection was usually used bya self-styled system manager to get power over everyoneelse, " Stallman would later explain. "They didn't wantanyone to be able to get power over them that way, sothey didn't implement that kind of a feature. Theresult was, that whenever something in the system wasbroken, you could always fix it. "See Richard Stallman (1986). 

Through such vigilance, hackers managed to keep the AILab's machines security-free. Over at the nearby MITLaboratory for Computer Sciences, however, security-minded faculty members won the day. The LCSinstalled its first password-based system in 1977. Onceagain, Stallman took it upon himself to correct what hesaw as ethical laxity. Gaining access to the softwarecode that controlled the password system, Stallmanimplanted a software command that sent out a message toany LCS user who attempted to choose a unique password. If a user entered "starfish, " for example, the messagecame back something like: I see you chose the password"starfish. " I suggest that you switch to the password"carriage return. " It's much easier to type, and alsoit stands up to the principle that there should be no passwords. See StevenLevy, Hackers (Penguin USA [paperback], 1984): 417. I have modified this quote, which Levy alsouses as an excerpt, to illustrate more directly how theprogram might reveal the false security of the system. Levy uses the placeholder "[such and such]. " Users who did enter "carriage return"-that is, userswho simply pressed the Enter or Return button, enteringa blank string instead of a unique password-left theiraccounts accessible to the world at large. As scary asthat might have been for some users, it reinforced thehacker notion that Institute computers, and evenInstitute computer files, belonged to the public, notprivate individuals. Stallman, speaking in an interviewfor the 1984 book Hackers, proudly noted that one-fifthof the LCS staff accepted this argument and employedthe blank-string password. See Steven Levy, Hackers (Penguin USA [paperback], 1984): 417. 

Stallman's null-string crusade would prove ultimatelyfutile. By the early 1980s, even the AI Lab's machineswere sporting password-based security systems. Even so, it represents a major milestone in terms of Stallman'spersonal and political maturation. To the objectiveobserver familiar with Stallman's later career, itoffers a convenient inflection point between the timidteenager afraid to speak out even on issues oflife-threatening importance and the adult activist whowould soon turn needling and cajoling into a full-timeoccupation. 

In voicing his opposition to computer security, Stallman drew on many of the forces that had shaped hisearly life: hunger for knowledge, distaste forauthority, and frustration over hidden procedures andrules that rendered some people clueless outcasts. Hewould also draw on the ethical concepts that wouldshape his adult life: communal responsibility, trust, and the hacker spirit of direct action. Expressed insoftware-computing terms, the null string representsthe 1. 0 version of the Richard Stallman politicalworldview-incomplete in a few places but, for the mostpart, fully mature. 

Looking back, Stallman hesitates to impart too muchsignificance to an event so early in his hackingcareer. "In that early stage there were a lot of peoplewho shared my feelings, " he says. "The large number ofpeople who adopted the null string as their passwordwas a sign that many people agreed that it was theproper thing to do. I was simply inclined to be anactivist about it. "

Stallman does credit the AI Lab for awakening thatactivist spirit, however. As a teenager, Stallman hadobserved political events with little idea as to how asingle individual could do or say anything ofimportance. As a young adult, Stallman was speaking outon matters in which he felt supremely confident, matters such as software design, communalresponsibility, and individual freedom. "I joined thiscommunity which had a way of life which involvedrespecting each other's freedom, " he says. "It didn'ttake me long to figure out that that was a good thing. It took me longer to come to the conclusion that thiswas a moral issue. "

Hacking at the AI Lab wasn't the only activity helpingto boost Stallman's esteem. During the middle of hissophomore year at Harvard, Stallman had joined up witha dance troupe that specialized in folk dances . Whatbegan as a simple attempt to meet women and expand hissocial horizons soon expanded into yet another passionalongside hacking. Dancing in front of audiencesdressed in the native garb of a Balkan peasant, Stallman no longer felt like the awkward, uncoordinated10-year-old whose attempts to play football had endedin frustration. He felt confident, agile, and alive. For a brief moment, he even felt a hint of emotionalconnection. He soon found being in front of an audiencefun, and it wasn't long thereafter that he begancraving the performance side of dancing almost as muchas the social side. 

Although the dancing and hacking did little to improveStallman's social standing, they helped him overcomethe feelings of weirdness that had clouded hispre-Harvard life. Instead of lamenting his weirdnature, Stallman found ways to celebrate it. In 1977, while attending a science-fiction convention, he cameacross a woman selling custom-made buttons. Excited, Stallman ordered a button with the words "Impeach God"emblazoned on it. 

For Stallman, the "Impeach God" message worked on manylevels. An atheist since early childhood, Stallmanfirst saw it as an attempt to set a "second front" inthe ongoing debate on religion. "Back then everybodywas arguing about God being dead or alive, " Stallmanrecalls. "`Impeach God' approached the subject of Godfrom a completely different viewpoint. If God was sopowerful as to create the world and yet do nothing tocorrect the problems in it, why would we ever want toworship such a God? Wouldn't it be better to put him on trial?"

At the same time, "Impeach God" was a satirical take onAmerica and the American political system. TheWatergate scandal of the 1970s affected Stallmandeeply. As a child, Stallman had grown up mistrustingauthority. Now, as an adult, his mistrust had beensolidified by the culture of the AI Lab hackercommunity. To the hackers, Watergate was merely aShakespearean rendition of the daily power strugglesthat made life such a hassle for those withoutprivilege. It was an outsized parable for what happenedwhen people traded liberty and openness for securityand convenience. 

Buoyed by growing confidence, Stallman wore the buttonproudly. People curious enough to ask him about itreceived the same well-prepared spiel. "My name isJehovah, " Stallman would say. "I have a special plan tosave the universe, but because of heavenly securityreasons I can't tell you what that plan is. You're justgoing to have to put your faith in me, because I seethe picture and you don't. You know I'm good because Itold you so. If you don't believe me, I'll throw you onmy enemies list and throw you in a pit where InfernalRevenue Service will audit your taxes for eternity. "

Those who interpreted the spiel as a word-for-wordparody of the Watergate hearings only got half themessage. For Stallman, the other half of the messagewas something only his fellow hackers seemed to behearing. One hundred years after Lord Acton warnedabout absolute power corrupting absolutely, Americansseemed to have forgotten the first part of Acton'struism: power, itself, corrupts. Rather than point outthe numerous examples of petty corruption, Stallmanfelt content voicing his outrage toward an entiresystem that trusted power in the first place. 

"I figured why stop with the small fry, " says Stallman, recalling the button and its message. "If we went afterNixon, why not going after Mr. Big. The way I see it, any being that has power and abuses it deserves to havethat power taken away. "

Small Puddle of Freedom

Ask anyone who's spent more than a minute in RichardStallman's presence, and you'll get the samerecollection: forget the long hair. Forget the quirkydemeanor. The first thing you notice is the gaze. Onelook into Stallman's green eyes, and you know you're inthe presence of a true believer. 

To call the Stallman gaze intense is an understatement. Stallman's eyes don't just look at you; they lookthrough you. Even when your own eyes momentarily shiftaway out of simple primate politeness, Stallman's eyesremain locked-in, sizzling away at the side of yourhead like twin photon beams. 

Maybe that's why most writers, when describingStallman, tend to go for the religious angle. In a 1998Salon. Com article titled "The Saint of Free Software, "Andrew Leonard describes Stallman's green eyes as"radiating the power of an Old Testament prophet. "See Andrew Leonard, "TheSaint of Free Software, "Salon. Com (August 1998). 

http://www. Salon. Com/21st/feature/1998/08/cov_31feature. Html A 1999 Wired magazine article describes the Stallmanbeard as "Rasputin-like, "See Leander Kahney, "Linux's Forgotten Man, " WiredNews(March 5, 1999). Http://www. Wired. Com/news/print/0, 1294, 18291, 00. Html while a London Guardian profile describes the Stallmansmile as the smile of "a disciple seeing Jesus. "See "Programmer on moral highground; Free software isa moral issue for Richard Stallman believes in freedomand free software. " London Guardian (November 6, 1999). These are just a small sampling of the religiouscomparisons. To date, the most extreme comparison hasto go to Linus Torvalds, who, in his autobiography-seeLinus Torvalds and David Diamond, Just For Fun: TheStory of an Accidentaly Revolutionary (HarperCollinsPublishers, Inc. , 2001): 58-writes "Richard Stallman isthe God of Free Software. " Honorable mention goes toLarry Lessig, who, in a footnote description ofStallman in his book-see Larry Lessig, The Future ofIdeas (Random House, 2001): 270-likens Stallman toMoses: . . . As with Moses, it was another leader, Linus Torvalds, who finally carried the movement intothe promised land by facilitating the development ofthe final part of the OS puzzle. Like Moses, too, Stallman is both respected and reviled by allies withinthe movement. He is [an] unforgiving, and hence formany inspiring, leader of a critically important aspectof modern culture. I have deep respect for theprinciple and commitment of this extraordinaryindividual, though I also have great respect for thosewho are courageous enough to question his thinking andthen sustain his wrath. In a final interview withStallman, I asked him his thoughts about the religiouscomparisons. "Some people do compare me with an OldTestament prophent, and the reason is Old Testamentprophets said certain social practices were wrong. Theywouldn't compromise on moral issues. They couldn't bebought off, and they were usually treated with contempt. "

Such analogies serve a purpose, but they ultimatelyfall short. That's because they fail to take intoaccount the vulnerable side of the Stallman persona. Watch the Stallman gaze for an extended period of time, and you will begin to notice a subtle change. Whatappears at first to be an attempt to intimidate orhypnotize reveals itself upon second and third viewingas a frustrated attempt to build and maintain contact. If, as Stallman himself has suspected from time totime, his personality is the product of autism orAsperger Syndrome, his eyes certainly confirm thediagnosis. Even at their most high-beam level ofintensity, they have a tendency to grow cloudy anddistant, like the eyes of a wounded animal preparing togive up the ghost. 

My own first encounter with the legendary Stallman gazedates back to the March, 1999, LinuxWorld Conventionand Expo in San Jose, California. Billed as a "comingout party" for the Linux software community, theconvention also stands out as the event thatreintroduced Stallman to the technology media. Determined to push for his proper share of credit, Stallman used the event to instruct spectators andreporters alike on the history of the GNU Project andthe project's overt political objectives. 

As a reporter sent to cover the event, I received myown Stallman tutorial during a press conferenceannouncing the release of GNOME 1. 0, a free softwaregraphic user interface. Unwittingly, I push an entirebank of hot buttons when I throw out my very firstquestion to Stallman himself: do you think GNOME'smaturity will affect the commercial popularity of theLinux operating system?

"I ask that you please stop calling the operatingsystem Linux, " Stallman responds, eyes immediatelyzeroing in on mine. "The Linux kernel is just a smallpart of the operating system. Many of the softwareprograms that make up the operating system you callLinux were not developed by Linus Torvalds at all. Theywere created by GNU Project volunteers, putting intheir own personal time so that users might have a freeoperating system like the one we have today. To notacknowledge the contribution of those programmers isboth impolite and a misrepresentation of history. That's why I ask that when you refer to the operatingsystem, please call it by its proper name, GNU/Linux. "

Taking the words down in my reporter's notebook, Inotice an eerie silence in the crowded room. When Ifinally look up, I find Stallman's unblinking eyeswaiting for me. Timidly, a second reporter throws out aquestion, making sure to use the term " GNU/Linux"instead of Linux. Miguel de Icaza, leader of the GNOMEproject, fields the question. It isn't until halfwaythrough de Icaza's answer, however, that Stallman'seyes finally unlock from mine. As soon as they do, amild shiver rolls down my back. When Stallman startslecturing another reporter over a perceived error indiction, I feel a guilty tinge of relief. At least heisn't looking at me, I tell myself. 

For Stallman, such face-to-face moments would servetheir purpose. By the end of the first LinuxWorld show, most reporters know better than to use the term "Linux"in his presence, and wired. Com is running a storycomparing Stallman to a pre-Stalinist revolutionaryerased from the history books by hackers andentrepreneurs eager to downplay the GNU Project'soverly political objectives. 2 Other articles follow, and while few reporters call the operating systemGNU/Linux in print, most are quick to credit Stallmanfor launching the drive to build a free softwareoperating system 15 years before. 

I won't meet Stallman again for another 17 months. During the interim, Stallman will revisit SiliconValley once more for the August, 1999 LinuxWorld show. Although not invited to speak, Stallman does managed todeliver the event's best line. Accepting the show'sLinus Torvalds Award for Community Service-an awardnamed after Linux creator Linus Torvalds-on behalf ofthe Free Software Foundation, Stallman wisecracks, "Giving the Linus Torvalds Award to the Free SoftwareFoundation is a bit like giving the Han Solo Award tothe Rebel Alliance. "

This time around, however, the comments fail to makemuch of a media dent. Midway through the week, Red Hat, Inc. , a prominent GNU/Linux vendor, goes public. Thenews merely confirms what many reporters such as myselfalready suspect: "Linux" has become a Wall Streetbuzzword, much like "e-commerce" and "dot-com" beforeit. With the stock market approaching the Y2K rolloverlike a hyperbola approaching its vertical asymptote, all talk of free software or open source as a politicalphenomenon falls by the wayside. 

Maybe that's why, when LinuxWorld follows up its firsttwo shows with a third LinuxWorld show in August, 2000, Stallman is conspicuously absent. 

My second encounter with Stallman and his trademarkgaze comes shortly after that third LinuxWorld show. Hearing that Stallman is going to be in Silicon Valley, I set up a lunch interview in Palo Alto, California. The meeting place seems ironic, not only because of therecent no-show but also because of the overallbackdrop. Outside of Redmond, Washington, few citiesoffer a more direct testament to the economic value ofproprietary software. Curious to see how Stallman, aman who has spent the better part of his life railingagainst our culture's predilection toward greed andselfishness, is coping in a city where evengarage-sized bungalows run in the half-million-dollarprice range, I make the drive down from Oakland. 

I follow the directions Stallman has given me, until Ireach the headquarters of Art. Net, a nonprofit "virtualartists collective. " Located in a hedge-shrouded housein the northern corner of the city, the Art. Netheadquarters are refreshingly run-down. Suddenly, theidea of Stallman lurking in the heart of Silicon Valleydoesn't seem so strange after all. 

I find Stallman sitting in a darkened room, tappingaway on his gray laptop computer. He looks up as soonas I enter the room, giving me a full blast of his200-watt gaze. When he offers a soothing "Hello, " Ioffer a return greeting. Before the words come out, however, his eyes have already shifted back to thelaptop screen. 

"I'm just finishing an article on the spirit ofhacking, " Stallman says, fingers still tapping. "Take a look. "

I take a look. The room is dimly lit, and the textappears as greenish-white letters on a blackbackground, a reversal of the color scheme used by mostdesktop word-processing programs, so it takes my eyes amoment to adjust. When they do, I find myself readingStallman's account of a recent meal at a Koreanrestaurant. Before the meal, Stallman makes aninteresting discovery: the person setting the table hasleft six chopsticks instead of the usual two in frontof Stallman's place setting. Where most restaurantgoers would have ignored the redundant pairs, Stallmantakes it as challenge: find a way to use all sixchopsticks at once. Like many software hacks, thesuccessful solution is both clever and silly at thesame time. Hence Stallman's decision to use it as anillustration. 

As I read the story, I feel Stallman watching meintently. I look over to notice a proud but child-likehalf smile on his face. When I praise the essay, mycomment barely merits a raised eyebrow. 

"I'll be ready to go in a moment, " he says. 

Stallman goes back to tapping away at his laptop. Thelaptop is gray and boxy, not like the sleek, modernlaptops that seemed to be a programmer favorite at therecent LinuxWorld show. Above the keyboard rides asmaller, lighter keyboard, a testament to Stallman'saging hands. During the late 1980s, when Stallman wasputting in 70- and 80-hour work weeks writing the firstfree software tools and programs for the GNU Project, the pain in Stallman's hands became so unbearable thathe had to hire a typist. Today, Stallman relies on akeyboard whose keys require less pressure than atypical computer keyboard. 

Stallman has a tendency to block out all externalstimuli while working. Watching his eyes lock onto thescreen and his fingers dance, one quickly gets thesense of two old friends locked in deep conversation. 

The session ends with a few loud keystrokes and theslow disassembly of the laptop. 

"Ready for lunch?" Stallman asks. 

We walk to my car. Pleading a sore ankle, Stallmanlimps along slowly. Stallman blames the injury on atendon in his left foot. The injury is three years oldand has gotten so bad that Stallman, a huge fan of folkdancing, has been forced to give up all dancingactivities. "I love folk dancing inherently, " Stallmanlaments. "Not being able to dance has been a tragedyfor me. "

Stallman's body bears witness to the tragedy. Lack ofexercise has left Stallman with swollen cheeks and apot belly that was much less visible the year before. You can tell the weight gain has been dramatic, becausewhen Stallman walks, he arches his back like a pregnantwoman trying to accommodate an unfamiliar load. 

The walk is further slowed by Stallman's willingness tostop and smell the roses, literally. Spotting aparticularly beautiful blossom, he tickles theinnermost petals with his prodigious nose, takes a deepsniff and steps back with a contented sigh. 

"Mmm, rhinophytophilia, "At the time, I thought Stallman was referring to theflower's scientific name. Months later, I would learnthat rhinophytophilia was in fact a humorous referenceto the activity, i. E. , Stallman sticking his nose intoa flower and enjoying the moment. For another humorousStallman flower incident, visit:http://www. Stallman. Org/texas. Html he says, rubbing his back. 

The drive to the restaurant takes less than threeminutes. Upon recommendation from Tim Ney, formerexecutive director of the Free Software Foundation, Ihave let Stallman choose the restaurant. While somereporters zero in on Stallman's monk-like lifestyle, the truth is, Stallman is a committed epicure when itcomes to food. One of the fringe benefits of being atraveling missionary for the free software cause is theability to sample delicious food from around the world. "Visit almost any major city in the world, and chancesare Richard knows the best restaurant in town, " saysNey. "Richard also takes great pride in knowing what'son the menu and ordering for the entire table. "

For today's meal, Stallman has chosen a Cantonese-styledim sum restaurant two blocks off University Avenue, Palo Alto's main drag. The choice is partially inspiredby Stallman's recent visit to China, including alecture stop in Guangdong province, in addition toStallman's personal aversion to spicier Hunanese andSzechuan cuisine. "I'm not a big fan of spicy, "Stallman admits. 

We arrive a few minutes after 11 a. M. And findourselves already subject to a 20-minute wait. Giventhe hacker aversion to lost time, I hold my breathmomentarily, fearing an outburst. Stallman, contrary toexpectations, takes the news in stride. 

"It's too bad we couldn't have found somebody else tojoin us, " he tells me. "It's always more fun to eatwith a group of people. "

During the wait, Stallman practices a few dance steps. His moves are tentative but skilled. We discuss currentevents. Stallman says his only regret about notattending LinuxWorld was missing out on a pressconference announcing the launch of the GNOMEFoundation. Backed by Sun Microsystems and IBM, thefoundation is in many ways a vindication for Stallman, who has long championed that free software andfree-market economics need not be mutually exclusive. Nevertheless, Stallman remains dissatisfied by themessage that came out. 

"The way it was presented, the companies were talkingabout Linux with no mention of the GNU Project at all, "Stallman says. 

Such disappointments merely contrast the warm responsecoming from overseas, especially Asia, Stallman notes. A quick glance at the Stallman 2000 travel itinerarybespeaks the growing popularity of the free softwaremessage. Between recent visits to India, China, andBrazil, Stallman has spent 12 of the last 115 days onUnited States soil. His travels have given him anopportunity to see how the free software concepttranslates into different languages of cultures. 

"In India many people are interested in free software, because they see it as a way to build their computinginfrastructure without spending a lot of money, "Stallman says. "In China, the concept has been muchslower to catch on. Comparing free software to freespeech is harder to do when you don't have any freespeech. Still, the level of interest in free softwareduring my last visit was profound. "

The conversation shifts to Napster, the San Mateo, California software company, which has become somethingof a media cause cÈlËbre in recent months. The companymarkets a controversial software tool that lets musicfans browse and copy the music files of other musicfans. Thanks to the magnifying powers of the Internet, this so-called "peer-to-peer" program has evolved intoa de facto online juke box, giving ordinary music fansa way to listen to MP3 music files over the computerwithout paying a royalty or fee, much to recordcompanies' chagrin. 

Although based on proprietary software, the Napstersystem draws inspiration from the long-held Stallmancontention that once a work enters the digital realm-inother words, once making a copy is less a matter ofduplicating sounds or duplicating atoms and more amatter of duplicating information-the natural humanimpulse to share a work becomes harder to restrict. Rather than impose additional restrictions, Napsterexecs have decided to take advantage of the impulse. Giving music listeners a central place to trade musicfiles, the company has gambled on its ability to steerthe resulting user traffic toward other commercial opportunities. 

The sudden success of the Napster model has put thefear in traditional record companies, with good reason. Just days before my Palo Alto meeting with Stallman, U. S. District Court Judge Marilyn Patel granted arequest filed by the Recording Industry Association ofAmerica for an injunction against the file-sharingservice. The injunction was subsequently suspended bythe U. S. Ninth District Court of Appeals, but by early2001, the Court of Appeals, too, would find the SanMateo-based company in breach of copyright law, 5 adecision RIAA spokesperson Hillary Rosen would laterproclaim proclaim a "clear victory for the creativecontent community and the legitimate online marketplace. "See "A Clear Victoryfor Recording Industry in NapsterCase, " RIAA press release (February 12, 2001). Http://www. Riaa. Com/PR_story. Cfm?id=372

For hackers such as Stallman, the Napster businessmodel is scary in different ways. The company'seagerness to appropriate time-worn hacker principlessuch as file sharing and communal informationownership, while at the same time selling a servicebased on proprietary software, sends a distressingmixed message. As a person who already has a hardenough time getting his own carefully articulatedmessage into the media stream, Stallman isunderstandably reticent when it comes to speaking outabout the company. Still, Stallman does admit tolearning a thing or two from the social side of theNapster phenomenon. 

"Before Napster, I thought it might be OK for people toprivately redistribute works of entertainment, "Stallman says. "The number of people who find Napsteruseful, however, tells me that the right toredistribute copies not only on a neighbor-to-neighborbasis, but to the public at large, is essential andtherefore may not be taken away. "

No sooner does Stallman say this than the door to therestaurant swings open and we are invited back insideby the host. Within a few seconds, we are seated in aside corner of the restaurant next to a large mirrored wall. 

The restaurant's menu doubles as an order form, andStallman is quickly checking off boxes before the hosthas even brought water to the table. "Deep-fried shrimproll wrapped in bean-curd skin, " Stallman reads. "Bean-curd skin. It offers such an interesting texture. I think we should get it. "

This comment leads to an impromptu discussion ofChinese food and Stallman's recent visit to China. "Thefood in China is utterly exquisite, " Stallman says, hisvoice gaining an edge of emotion for the first timethis morning. "So many different things that I've neverseen in the U. S. , local things made from localmushrooms and local vegetables. It got to the pointwhere I started keeping a journal just to keep track ofevery wonderful meal. "

The conversation segues into a discussion of Koreancuisine. During the same June, 2000, Asian tour, Stallman paid a visit to South Korea. His arrivalignited a mini-firestorm in the local media thanks to aKorean software conference attended by Microsoftfounder and chairman Bill Gates that same week. Next togetting his photo above Gates's photo on the front pageof the top Seoul newspaper, Stallman says the bestthing about the trip was the food. "I had a bowl ofnaeng myun, which is cold noodles, " says Stallman. "These were a very interesting feeling noodle. Mostplaces don't use quite the same kind of noodles foryour naeng myun, so I can say with complete certaintythat this was the most exquisite naeng myun I ever had. "

The term "exquisite" is high praise coming fromStallman. I know this, because a few moments afterlistening to Stallman rhapsodize about naeng myun, Ifeel his laser-beam eyes singeing the top of my right shoulder. 

"There is the most exquisite woman sitting just behindyou, " Stallman says. 

I turn to look, catching a glimpse of a woman's back. The woman is young, somewhere in her mid-20s, and iswearing a white sequinned dress. She and her male lunchcompanion are in the final stages of paying the check. When both get up from the table to leave therestaurant, I can tell without looking, becauseStallman's eyes suddenly dim in intensity. 

"Oh, no, " he says. "They're gone. And to think, I'llprobably never even get to see her again. "

After a brief sigh, Stallman recovers. The moment givesme a chance to discuss Stallman's reputation vis-ý-visthe fairer sex. The reputation is a bit contradictoryat times. A number of hackers report Stallman'spredilection for greeting females with a kiss on theback of the hand. See Mae Ling Mak, "Mae Ling's Story" (December 17, 1998). 

http://www. Crackmonkey. Org/pipermail/crackmonkey/1998q4/003006. HtmSo far, Mak is the only person I've found willing tospeak on the record in regard to this practice, although I've heard this from a few other femalesources. Mak, despite expressing initial revulsion atit, later managed to put aside her misgivings and dancewith Stallman at a 1999 LinuxWorld show. Http://www. Linux. Com/interact/potd. Phtml?potd_id=44 A May 26, 2000 Salon. Com article, meanwhile, portraysStallman as a bit of a hacker lothario. Documenting thefree software-free love connection, reporter AnnaleeNewitz presents Stallman as rejecting traditionalfamily values, telling her, "I believe in love, but notmonogamy. "See Annalee Newitz, "If Code is Free Why Not Me?"Salon. Com (May 26, 2000). 

Stallman lets his menu drop a little when I bring thisup. "Well, most men seem to want sex and seem to have arather contemptuous attitude towards women, " he says. "Even women they're involved with. I can't understandit at all. "

I mention a passage from the 1999 book Open Sources inwhich Stallman confesses to wanting to name theill-fated GNU kernel after a girlfriend at the time. The girlfriend's name was Alix, a name that fitperfectly with the Unix developer convention of puttingan "x" at the end of any new kernel name-e. G. , "Linux. "Because the woman was a Unix system administrator, Stallman says it would have been an even more touchingtribute. Unfortunately, Stallman notes, the kernelproject's eventual main developer renamed the kernel HURD. See Richard Stallman, "The GNU Operating System and theFree Software Movement, " Open Sources (O'Reilly &Associates, Inc. , 1999): 65. Although Stallman and the girlfriend later broke up, the story triggers an automatic question: for all themedia imagery depicting him as a wild-eyed fanatic, isRichard Stallman really just a hopeless romantic, awandering Quixote tilting at corporate windmills in aneffort to impress some as-yet-unidentified Dulcinea?

"I wasn't really trying to be romantic, " Stallman says, recalling the Alix story. "It was more of a teasingthing. I mean, it was romantic, but it was alsoteasing, you know? It would have been a delightful surprise. "

For the first time all morning, Stallman smiles. Ibring up the hand kissing. "Yes, I do do that, "Stallman says. "I've found it's a way of offering someaffection that a lot of women will enjoy. It's a chanceto give some affection and to be appreciated for it. "

Affection is a thread that runs clear through RichardStallman's life, and he is painfully candid about itwhen questions arise. "There really hasn't been muchaffection in my life, except in my mind, " he says. Still, the discussion quickly grows awkward. After afew one-word replies, Stallman finally lifts up hismenu, cutting off the inquiry. 

"Would you like some shimai?" he asks. 

When the food comes out, the conversation slalomsbetween the arriving courses. We discuss the oft-notedhacker affection for Chinese food, the weekly dinnerruns into Boston's Chinatown district during Stallman'sdays as a staff programmer at the AI Lab, and theunderlying logic of the Chinese language and itsassociated writing system. Each thrust on my partelicits a well-informed parry on Stallman's part. 

"I heard some people speaking Shanghainese the lasttime I was in China, " Stallman says. "It wasinteresting to hear. It sounded quite different [fromMandarin]. I had them tell me some cognate words inMandarin and Shanghainese. In some cases you can seethe resemblance, but one question I was wondering aboutwas whether tones would be similar. They're not. That'sinteresting to me, because there's a theory that thetones evolved from additional syllables that got lostand replaced. Their effect survives in the tone. Ifthat's true, and I've seen claims that that happenedwithin historic times, the dialects must have divergedbefore the loss of these final syllables. "

The first dish, a plate of pan-fried turnip cakes, hasarrived. Both Stallman and I take a moment to carve upthe large rectangular cakes, which smell like boiledcabbage but taste like potato latkes fried in bacon. 

I decide to bring up the outcast issue again, wonderingif Stallman's teenage years conditioned him to takeunpopular stands, most notably his uphill battle since1994 to get computer users and the media to replace thepopular term "Linux" with "GNU/Linux. "

"I believe it did help me, " Stallman says, chewing on adumpling. "I have never understood what peer pressuredoes to other people. I think the reason is that I wasso hopelessly rejected that for me, there wasn'tanything to gain by trying to follow any of the fads. It wouldn't have made any difference. I'd still be justas rejected, so I didn't try. "

Stallman points to his taste in music as a key exampleof his contrarian tendencies. As a teenager, when mostof his high school classmates were listening to Motownand acid rock, Stallman preferred classical music. Thememory leads to a rare humorous episode from Stallman'smiddle-school years. Following the Beatles' 1964appearance on the Ed Sullivan Show, most of Stallman'sclassmates rushed out to purchase the latest Beatlesalbums and singles. Right then and there, Stallmansays, he made a decision to boycott the Fab Four. 

"I liked some of the pre-Beatles popular music, "Stallman says. "But I didn't like the Beatles. Iespecially disliked the wild way people reacted tothem. It was like: who was going to have a Beatlesassembly to adulate the Beatles the most?"

When his Beatles boycott failed to take hold, Stallmanlooked for other ways to point out the herd-mentalityof his peers. Stallman says he briefly consideredputting together a rock band himself dedicated tosatirizing the Liverpool group. 

"I wanted to call it Tokyo Rose and the Japanese Beetles. "

Given his current love for international folk music, Iask Stallman if he had a similar affinity for Bob Dylanand the other folk musicians of the early 1960s. Stallman shakes his head. "I did like Peter, Paul andMary, " he says. "That reminds me of a great filk. "

When I ask for a definition of "filk, " Stallmanexplains the concept. A filk, he says, is a popularsong whose lyrics have been replaced with parodylyrics. The process of writing a filk is calledfilking, and it is a popular activity among hackers andscience-fiction aficionados. Classic filks include "OnTop of Spaghetti, " a rewrite of "On Top of Old Smokey, "and "Yoda, " filk-master "Weird" Al Yankovic's StarWars-oriented rendition of the Kinks tune, "Lola. "

Stallman asks me if I would be interested in hearingthe folk filk. As soon as I say yes, Stallman's voicebegins singing in an unexpectedly clear tone: How muchwood could a woodchuck chuck, If a woodchuck could chuckwood?How many poles could a polak lock, If a polak couldlock poles?How many knees could a negro grow, If a negrocould grow knees?The answer, my dear, is stick it inyour ear. The answer is to stick it in your ear. Thesinging ends, and Stallman's lips curl into anotherchild-like half smile. I glance around at the nearbytables. The Asian families enjoying their Sunday lunchpay little attention to the bearded alto in their midst. For more Stallmanfilks, visithttp://www. Stallman. Org/doggerel. Html. To hear Stallmansinging "The Free Software Song, " visithttp://www. Gnu. Org/music/free-software-song. Html. After a few moments of hesitation, I finally smile too. 

"Do you want that last cornball?" Stallman asks, eyestwinkling. Before I can screw up the punch line, Stallman grabs the corn-encrusted dumpling with his twochopsticks and lifts it proudly. "Maybe I'm the one whoshould get the cornball, " he says. 

The food gone, our conversation assumes the dynamics ofa normal interview. Stallman reclines in his chair andcradles a cup of tea in his hands. We resume talkingabout Napster and its relation to the free softwaremovement. Should the principles of free software beextended to similar arenas such as music publishing? I ask. 

"It's a mistake to transfer answers from one thing toanother, " says Stallman, contrasting songs withsoftware programs. "The right approach is to look ateach type of work and see what conclusion you get. "

When it comes to copyrighted works, Stallman says hedivides the world into three categories. The firstcategory involves "functional" works-e. G. , softwareprograms, dictionaries, and textbooks. The secondcategory involves works that might best be described as"testimonial"-e. G. , scientific papers and historicaldocuments. Such works serve a purpose that would beundermined if subsequent readers or authors were freeto modify the work at will. The final category involvesworks of personal expression-e. G. , diaries, journals, and autobiographies. To modify such documents would beto alter a person's recollections or point ofview-action Stallman considers ethically unjustifiable. 

Of the three categories, the first should give usersthe unlimited right to make modified versions, whilethe second and third should regulate that rightaccording to the will of the original author. Regardless of category, however, the freedom to copyand redistribute noncommercially should remainunabridged at all times, Stallman insists. If thatmeans giving Internet users the right to generate ahundred copies of an article, image, song, or book andthen email the copies to a hundred strangers, so be it. "It's clear that private occasional redistribution mustbe permitted, because only a police state can stopthat, " Stallman says. "It's antisocial to come betweenpeople and their friends. Napster has convinced me thatwe also need to permit, must permit, even noncommercialredistribution to the public for the fun of it. Becauseso many people want to do that and find it so useful. "

When I ask whether the courts would accept such apermissive outlook, Stallman cuts me off. 

"That's the wrong question, " he says. "I mean nowyou've changed the subject entirely from one of ethicsto one of interpreting laws. And those are two totallydifferent questions in the same field. It's useless tojump from one to the other. How the courts wouldinterpret the existing laws is mainly in a harsh way, because that's the way these laws have been bought bypublishers. "

The comment provides an insight into Stallman'spolitical philosophy: just because the legal systemcurrently backs up businesses' ability to treatcopyright as the software equivalent of land titledoesn't mean computer users have to play the gameaccording to those rules. Freedom is an ethical issue, not a legal issue. "I'm looking beyond what theexisting laws are to what they should be, " Stallmansays. "I'm not trying to draft legislation. I'mthinking about what should the law do? I consider thelaw prohibiting the sharing of copies with your friendthe moral equivalent of Jim Crow. It does not deserve respect. "

The invocation of Jim Crow prompts another question. How much influence or inspiration does Stallman drawfrom past political leaders? Like the civil-rightsmovement of the 1950s and 1960s, his attempt to drivesocial change is based on an appeal to timeless values:freedom, justice, and fair play. 

Stallman divides his attention between my analogy and aparticularly tangled strand of hair. When I stretch theanalogy to the point where I'm comparing Stallman withDr. Martin Luther King, Jr. , Stallman, after breakingoff a split end and popping it into his mouth, cuts me off. 

"I'm not in his league, but I do play the same game, "he says, chewing. 

I suggest Malcolm X as another point of comparison. Like the former Nation of Islam spokesperson, Stallmanhas built up a reputation for courting controversy, alienating potential allies, and preaching a messagefavoring self-sufficiency over cultural integration. 

Chewing on another split end, Stallman rejects thecomparison. "My message is closer to King's message, "he says. "It's a universal message. It's a message offirm condemnation of certain practices that mistreatothers. It's not a message of hatred for anyone. Andit's not aimed at a narrow group of people. I inviteanyone to value freedom and to have freedom. "

Even so, a suspicious attitude toward politicalalliances remains a fundamental Stallman charactertrait. In the case of his well-publicized distaste forthe term "open source, " the unwillingness toparticipate in recent coalition-building projects seemsunderstandable. As a man who has spent the last twodecades stumping on the behalf of free software, Stallman's political capital is deeply invested in theterm. Still, comments such as the "Han Solo" wisecrackat the 1999 LinuxWorld have only reinforced theStallman's reputation in the software industry as adisgrunted mossback unwilling to roll with political ormarketing trends. 

"I admire and respect Richard for all the work he'sdone, " says Red Hat president Robert Young, summing upStallman's paradoxical political nature. "My onlycritique is that sometimes Richard treats his friendsworse than his enemies. "

Stallman's unwillingness to seek alliances seemsequally perplexing when you consider his politicalinterests outside of the free software movement. VisitStallman's offices at MIT, and you instantly find aclearinghouse of left-leaning news articles coveringcivil-rights abuses around the globe. Visit his website, and you'll find diatribes on the DigitalMillennium Copyright Act, the War on Drugs, and theWorld Trade Organization. 

Given his activist tendencies, I ask, why hasn'tStallman sought a larger voice? Why hasn't he used hisvisibility in the hacker world as a platform to boostrather than reduce his political voice. 

Stallman lets his tangled hair drop and contemplatesthe question for a moment. 

"I hesitate to exaggerate the importance of this littlepuddle of freedom, " he says. "Because the morewell-known and conventional areas of working forfreedom and a better society are tremendouslyimportant. I wouldn't say that free software is asimportant as they are. It's the responsibility Iundertook, because it dropped in my lap and I saw a wayI could do something about it. But, for example, to endpolice brutality, to end the war on drugs, to end thekinds of racism we still have, to help everyone have acomfortable life, to protect the rights of people whodo abortions, to protect us from theocracy, these aretremendously important issues, far more important thanwhat I do. I just wish I knew how to do something about them. "

Once again, Stallman presents his political activity asa function of personal confidence. Given the amount oftime it has taken him to develop and hone the freesoftware movement's core tenets, Stallman is hesitantto jump aboard any issues or trends that mighttransport him into uncharted territory. 

"I wish I knew I how to make a major difference onthose bigger issues, because I would be tremendouslyproud if I could, but they're very hard and lots ofpeople who are probably better than I am have beenworking on them and have gotten only so far, " he says. "But as I see it, while other people were defendingagainst these big visible threats, I saw another threatthat was unguarded. And so I went to defend againstthat threat. It may not be as big a threat, but I wasthe only one there. "

Chewing a final split end, Stallman suggests paying thecheck. Before the waiter can take it away, however, Stallman pulls out a white-colored dollar bill andthrows it on the pile. The bill looks so clearlycounterfeit, I can't help but pick it up and read it. Sure enough, it is counterfeit. Instead of bearing theimage of a George Washington or Abe Lincoln, the bill'sfront side bears the image of a cartoon pig. Instead ofthe United States of America, the banner above the pigreads "United Swines of Avarice. " The bill is for zerodollars, and when the waiter picks up the money, Stallman makes sure to tug on his sleeve. 

"I added an extra zero to your tip, " Stallman says, yetanother half smile creeping across his lips. 

The waiter, uncomprehending or fooled by the look ofthe bill, smiles and scurries away. 

"I think that means we're free to go, " Stallman says. 

The Emacs Commune

The AI Lab of the 1970s was by all accounts a specialplace. Cutting-edge projects and top-flight researchersgave it an esteemed position in the world of computerscience. The internal hacker culture and its anarchicpolicies lent a rebellious mystique as well. Onlylater, when many of the lab's scientists and softwaresuperstars had departed, would hackers fully realizethe unique and ephemeral world they had once inhabited. 

"It was a bit like the Garden of Eden, " says Stallman, summing up the lab and its software-sharing ethos in a1998 Forbes article. "It hadn't occurred to us not to cooperate. "See JoshMcHugh, "For the Love of Hacking, " Forbes(August 10, 1998). Http://www. Forbes. Com/forbes/1998/0810/6203094a. Html

Such mythological descriptions, while extreme, underline an important fact. The ninth floor of 545Tech Square was more than a workplace for many. Forhackers such as Stallman, it was home. 

The word "home" is a weighted term in the Stallmanlexicon. In a pointed swipe at his parents, Stallman, to this day, refuses to acknowledge any home beforeCurrier House, the dorm he lived in during his days atHarvard. He has also been known to describe leavingthat home in tragicomic terms. Once, while describinghis years at Harvard, Stallman said his only regret wasgetting kicked out. It wasn't until I asked Stallmanwhat precipitated his ouster, that I realized I hadwalked into a classic Stallman setup line. 

"At Harvard they have this policy where if you pass toomany classes they ask you to leave, " Stallman says. 

With no dorm and no desire to return to New York, Stallman followed a path blazed by Greenblatt, Gosper, Sussman, and the many other hackers before him. Enrolling at MIT as a grad student, Stallman rented anapartment in nearby Cambridge but soon viewed the AILab itself as his de facto home. In a 1986 speech, Stallman recalled his memories of the AI Lab duringthis period: I may have done a little bit more livingat the lab than most people, because every year or twofor some reason or other I'd have no apartment and Iwould spend a few months living at the lab. And I'vealways found it very comfortable, as well as nice andcool in the summer. But it was not at all uncommon tofind people falling asleep at the lab, again because oftheir enthusiasm; you stay up as long as you possiblycan hacking, because you just don't want to stop. Andthen when you're completely exhausted, you climb overto the nearest soft horizontal surface. A very informalatmosphere. See Stallman (1986). The lab's home-like atmosphere could be aproblem at times. What some saw as a dorm, othersviewed as an electronic opium den. In the 1976 bookComputer Power and Human Reason, MIT researcher JosephWeizenbaum offered a withering critique of the "computer bum, " Weizenbaum's term for the hackers whopopulated computer rooms such as the AI Lab. "Theirrumpled clothes, their unwashed hair and unshavedfaces, and their uncombed hair all testify that theyare oblivious to their bodies and to the world in whichthey move, " Weizenbaum wrote. "[Computer bums] exist, at least when so engaged, only through and for the computers. "See JosephWeizenbaum, Computer Power and Human Reason:From Judgment to Calculation (W. H. Freeman, 1976): 116. 

Almost a quarter century after its publication, Stallman still bristles when hearing Weizenbaum's"computer bum" description, discussing it in thepresent tense as if Weizenbaum himself was still in theroom. "He wants people to be just professionals, doingit for the money and wanting to get away from it andforget about it as soon as possible, " Stallman says. "What he sees as a normal state of affairs, I see as a tragedy. "

Hacker life, however, was not without tragedy. Stallmancharacterizes his transition from weekend hacker tofull-time AI Lab denizen as a series of painfulmisfortunes that could only be eased through theeuphoria of hacking. As Stallman himself has said, thefirst misfortune was his graduation from Harvard. Eagerto continue his studies in physics, Stallman enrolledas a graduate student at MIT. The choice of schools wasa natural one. Not only did it give Stallman the chanceto follow the footsteps of great MIT alumni: WilliamShockley ('36), Richard P. Feynman ('39), and MurrayGell-Mann ('51), it also put him two miles closer tothe AI Lab and its new PDP-10 computer. "My attentionwas going toward programming, but I still thought, well, maybe I can do both, " Stallman says. 

Toiling in the fields of graduate-level science by dayand programming in the monastic confines of the AI Labby night, Stallman tried to achieve a perfect balance. The fulcrum of this geek teeter-totter was his weeklyouting with the folk-dance troupe, his one socialoutlet that guaranteed at least a modicum ofinteraction with the opposite sex. Near the end of thatfirst year at MIT, however, disaster struck. A kneeinjury forced Stallman to drop out of the troupe. Atfirst, Stallman viewed the injury as a temporaryproblem, devoting the spare time he would have spentdancing to working at the AI Lab even more. By the endof the summer, when the knee still ached and classesreconvened, Stallman began to worry. "My knee wasn'tgetting any better, " Stallman recalls, "which meant Ihad to stop dancing completely. I was heartbroken. "

With no dorm and no dancing, Stallman's social universeimploded. Like an astronaut experiencing theaftereffects of zero-gravity, Stallman found that hisability to interact with nonhackers, especially femalenonhackers, had atrophied significantly. After 16 weeksin the AI Lab, the self confidence he'd been quietlyaccumulating during his 4 years at Harvard wasvirtually gone. 

"I felt basically that I'd lost all my energy, "Stallman recalls. "I'd lost my energy to do anythingbut what was most immediately tempting. The energy todo something else was gone. I was in total despair. "

Stallman retreated from the world even further, focusing entirely on his work at the AI Lab. ByOctober, 1975, he dropped out of MIT, never to go back. Software hacking, once a hobby, had become his calling. 

Looking back on that period, Stallman sees thetransition from full-time student to full-time hackeras inevitable. Sooner or later, he believes, thesiren's call of computer hacking would have overpoweredhis interest in other professional pursuits. "Withphysics and math, I could never figure out a way tocontribute, " says Stallman, recalling his strugglesprior to the knee injury. "I would have been proud toadvance either one of those fields, but I could neversee a way to do that. I didn't know where to start. With software, I saw right away how to write thingsthat would run and be useful. The pleasure of thatknowledge led me to want to do it more. "

Stallman wasn't the first to equate hacking withpleasure. Many of the hackers who staffed the AI Labboasted similar, incomplete academic rÈsumÈs. Most hadcome in pursuing degrees in math or electricalengineering only to surrender their academic careersand professional ambitions to the sheer exhilarationthat came with solving problems never before addressed. Like St. Thomas Aquinas, the scholastic known forworking so long on his theological summae that hesometimes achieved spiritual visions, hackers reachedtranscendent internal states through sheer mental focusand physical exhaustion. Although Stallman shunneddrugs, like most hackers, he enjoyed the "high" thatcame near the end of a 20-hour coding bender. 

Perhaps the most enjoyable emotion, however, was thesense of personal fulfillment. When it came to hacking, Stallman was a natural. A childhood's worth oflate-night study sessions gave him the ability to worklong hours with little sleep. As a social outcast sinceage 10, he had little difficulty working alone. And asa mathematician with built-in gift for logic andforesight, Stallman possessed the ability to circumventdesign barriers that left most hackers spinning their wheels. 

"He was special, " recalls Gerald Sussman, an MITfaculty member and former AI Lab researcher. DescribingStallman as a "clear thinker and a clear designer, "Sussman employed Stallman as a research-projectassistant beginning in 1975. The project was complex, involving the creation of an AI program that couldanalyze circuit diagrams. Not only did it involve anexpert's command of Lisp, a programming language builtspecifically for AI applications, but it also requiredan understanding of how a human might approach the same task. 

When he wasn't working on official projects such asSussman's automated circuit-analysis program, Stallmandevoted his time to pet projects. It was in a hacker'sbest interest to improve the lab's softwareinfrastructure, and one of Stallman's biggest petprojects during this period was the lab's editorprogram TECO. 

The story of Stallman's work on TECO during the 1970sis inextricably linked with Stallman's later leadershipof the free software movement. It is also a significantstage in the history of computer evolution, so much sothat a brief recapitulation of that evolution isnecessary. During the 1950s and 1960s, when computerswere first appearing at universities, computerprogramming was an incredibly abstract pursuit. Tocommunicate with the machine, programmers created aseries of punch cards, with each card representing anindividual software command. Programmers would thenhand the cards over to a central system administratorwho would then insert them, one by one, into themachine, waiting for the machine to spit out a new setof punch cards, which the programmer would thendecipher as output. This process, known as " batchprocessing, " was cumbersome and time consuming. It wasalso prone to abuses of authority. One of themotivating factors behind hackers' inbred aversion tocentralization was the power held by early systemoperators in dictating which jobs held top priority. 

In 1962, computer scientists and hackers involved inMIT's Project MAC, an early forerunner of the AI Lab, took steps to alleviate this frustration. Time-sharing, originally known as "time stealing, " made it possiblefor multiple programs to take advantage of a machine'soperational capabilities. Teletype interfaces also madeit possible to communicate with a machine not through aseries of punched holes but through actual text. Aprogrammer typed in commands and read the line-by-lineoutput generated by the machine. 

During the late 1960s, interface design made additionalleaps. In a famous 1968 lecture, Doug Engelbart, ascientist then working at the Stanford ResearchInstitute, unveiled a prototype of the modern graphicalinterface. Rigging up a television set to the computerand adding a pointer device which Engelbart dubbed a "mouse, " the scientist created a system even moreinteractive than the time-sharing system developed aMIT. Treating the video display like a high-speedprinter, Engelbart's system gave a user the ability tomove the cursor around the screen and see the cursorposition updated by the computer in real time. The usersuddenly had the ability to position text anywhere onthe screen. 

Such innovations would take another two decades to maketheir way into the commercial marketplace. Still, bythe 1970s, video screens had started to replaceteletypes as display terminals, creating the potentialfor full-screen-as opposed to line-by-line-editing capabilities. 

One of the first programs to take advantage of thisfull-screen capability was the MIT AI Lab's TECO. Shortfor Text Editor and COrrector, the program had beenupgraded by hackers from an old teletype line editorfor the lab's PDP-6 machine. Ccording to the Jargon File, TECO's name originallystood for Tape Editor and Corrector. 

TECO was a substantial improvement over old editors, but it still had its drawbacks. To create and edit adocument, a programmer had to enter a series ofsoftware commands specifying each edit. It was anabstract process. Unlike modern word processors, whichupdate text with each keystroke, TECO demanded that theuser enter an extended series of editing instructionsfollowed by an "end of command" sequence just to changethe text. Over time, a hacker grew proficient enough towrite entire documents in edit mode, but as Stallmanhimself would later point out, the process required "amental skill like that of blindfold chess. "See Richard Stallman, "EMACS: TheExtensible, Customizable, Display Editor, " AI Lab Memo (1979). Anupdated HTML version of this memo, from which I amquoting, is available athttp://www. Gnu. Org/software/emacs/emacs-paper. Html. 

To facilitate the process, AI Lab hackers had built asystem that displayed both the "source" and "display"modes on a split screen. Despite this innovative hack, switching from mode to mode was still a nuisance. 

TECO wasn't the only full-screen editor floating aroundthe computer world at this time. During a visit to theStanford Artificial Intelligence Lab in 1976, Stallmanencountered an edit program named E. The programcontained an internal feature, which allowed a user toupdate display text after each command keystroke. Inthe language of 1970s programming, E was one of thefirst rudimentary WYSIWYG editors. Short for "what yousee is what you get, " WYSIWYG meant that a user couldmanipulate the file by moving through the displayedtext, as opposed to working through a back-end editor program. "See RichardStallman, "Emacs the Full Screen Editor"(1987). Http://www. Lysator. Liu. Se/history/garb/txt/87-1-emacs. Txt

Impressed by the hack, Stallman looked for ways toexpand TECO's functionality in similar fashion upon hisreturn to MIT. He found a TECO feature calledControl-R, written by Carl Mikkelson and named afterthe two-key combination that triggered it. Mikkelson'shack switched TECO from its usual abstractcommand-execution mode to a more intuitivekeystroke-by-keystroke mode. Stallman revised thefeature in a subtle but significant way. He made itpossible to trigger other TECO command strings, or "macros, " using other, two-key combinations. Where usershad once entered command strings and discarded themafter entering then, Stallman's hack made it possibleto save macro tricks on file and call them up at will. Mikkelson's hack had raised TECO to the level of aWYSIWYG editor. Stallman's hack had raised it to thelevel of a user-programmable WYSIWYG editor. "That wasthe real breakthrough, " says Guy Steele, a fellow AILab hacker at the time. 

By Stallman's own recollection, the macro hack touchedoff an explosion of further innovation. "Everybody andhis brother was writing his own collection of redefinedscreen-editor commands, a command for everything hetypically liked to do, " Stallman would later recall. "People would pass them around and improve them, makingthem more powerful and more general. The collections ofredefinitions gradually became system programs in theirown right. "

So many people found the macro innovations useful andhad incorporated it into their own TECO programs thatthe TECO editor had become secondary to the macro maniait inspired. "We started to categorize it mentally as aprogramming language rather than as an editor, "Stallman says. Users were experiencing their ownpleasure tweaking the software and trading new ideas. 

Two years after the explosion, the rate of innovationbegan to exhibit dangerous side effects. The explosivegrowth had provided an exciting validation of thecollaborative hacker approach, but it had also led toover-complexity. "We had a Tower of Babel effect, " saysGuy Steele. 

The effect threatened to kill the spirit that hadcreated it, Steele says. Hackers had designed ITS tofacilitate programmers' ability to share knowledge andimprove each other's work. That meant being able to sitdown at another programmer's desk, open up aprogrammer's work and make comments and modificationsdirectly within the software. "Sometimes the easiestway to show somebody how to program or debug somethingwas simply to sit down at the terminal and do it forthem, " explains Steele. 

The macro feature, after its second year, began to foilthis capability. In their eagerness to embrace the newfull-screen capabilities, hackers had customized theirversions of TECO to the point where a hacker sittingdown at another hacker's terminal usually had to spendthe first hour just figuring out what macro commandsdid what. 

Frustrated, Steele took it upon himself to the solvethe problem. He gathered together the four differentmacro packages and began assembling a chart documentingthe most useful macro commands. In the course ofimplementing the design specified by the chart, Steelesays he attracted Stallman's attention. 

"He started looking over my shoulder, asking me what Iwas doing, " recalls Steele. 

For Steele, a soft-spoken hacker who interacted withStallman infrequently, the memory still sticks out. Looking over another hacker's shoulder while he workedwas a common activity at the AI Lab. Stallman, the TECOmaintainer at the lab, deemed Steele's work"interesting" and quickly set off to complete it. 

"As I like to say, I did the first 0. 001 percent of theimplementation, and Stallman did the rest, " says Steelewith a laugh. 

The project's new name, Emacs, came courtesy ofStallman. Short for "editing macros, " it signified theevolutionary transcendence that had taken place duringthe macros explosion two years before. It also tookadvantage of a gap in the software programming lexicon. Noting a lack of programs on ITS starting with theletter "E, " Stallman chose Emacs, making it possible toreference the program with a single letter. Once again, the hacker lust for efficiency had left its mark. 

In the course of developing a standard system of macrocommands, Stallman and Steele had to traverse apolitical tightrope. In creating a standard program, Stallman was in clear violation of the fundamentalhacker tenet-"promote decentralization. " He was alsothreatening to hobble the very flexibility that hadfueled TECO's explosive innovation in the first place. 

"On the one hand, we were trying to make a uniformcommand set again; on the other hand, we wanted to keepit open ended, because the programmability wasimportant, " recalls Steele. 

To solve the problem, Stallman, Steele, and fellowhackers David Moon and Dan Weinreib limited theirstandardization effort to the WYSIWYG commands thatcontrolled how text appeared on-screen. The rest of theEmacs effort would be devoted to retaining theprogram's Tinker Toy-style extensibility. 

Stallman now faced another conundrum: if users madechanges but didn't communicate those changes back tothe rest of the community, the Tower of Babel effectwould simply emerge in other places. Falling back onthe hacker doctrine of sharing innovation, Stallmanembedded a statement within the source code that setthe terms of use. Users were free to modify andredistribute the code on the condition that they gaveback all the extensions they made. Stallman dubbed itthe " Emacs Commune. " Just as TECO had become more thana simple editor, Emacs had become more than a simplesoftware program. To Stallman, it was a socialcontract. In an early memo documenting the project, Stallman spelled out the contract terms. "EMACS, " hewrote, "was distributed on a basis of communal sharing, which means that all improvements must be given back tome to be incorporated and distributed. "See Stallman (1979): #SEC34. 

Not everybody accepted the contract. The explosiveinnovation continued throughout the decade, resultingin a host of Emacs-like programs with varying degreesof cross-compatibility. A few cited their relation toStallman's original Emacs with humorously recursivenames: Sine (Sine is not Emacs), Eine (Eine is notEmacs), and Zwei (Zwei was Eine initially). As adevoted exponent of the hacker ethic, Stallman saw noreason to halt this innovation through legalharassment. Still, the fact that some people would soeagerly take software from the community chest, alterit, and slap a new name on the resulting softwaredisplayed a stunning lack of courtesy. 

Such rude behavior was reflected against other, unsettling developments in the hacker community. BrianReid's 1979 decision to embed "time bombs" in Scribe, making it possible for Unilogic to limit unpaid useraccess to the software, was a dark omen to Stallman. "He considered it the most Nazi thing he ever saw inhis life, " recalls Reid. Despite going on to laterInternet fame as the cocreator of the Usenet altheirarchy, Reid says he still has yet to live down that1979 decision, at least in Stallman's eyes. "He saidthat all software should be free and the prospect ofcharging money for software was a crime against humanity. "In a 1996 interviewwith online magazine MEME, Stallman cited Scribe's sale as irksome, but hesitatedto mention Reid by name. "The problem was nobodycensured or punished this student for what he did, "Stallman said. "The result was other people got temptedto follow his example. " See MEME 2. 04. Http://memex. Org/meme2-04. Html

Although Stallman had been powerless to head off Reid'ssale, he did possess the ability to curtail other formsof behavior deemed contrary to the hacker ethos. Ascentral source-code maintainer for the Emacs "commune, "Stallman began to wield his power for political effect. During his final stages of conflict with theadministrators at the Laboratory for Computer Scienceover password systems, Stallman initiated a software " strike, "See Steven Levy, Hackers (Penguin USA [paperback], 1984): 419. Refusing to send lab members the latest version ofEmacs until they rejected the security system on thelab's computers. The move did little to improveStallman's growing reputation as an extremist, but itgot the point across: commune members were expected tospeak up for basic hacker values. 

"A lot of people were angry with me, saying I wastrying to hold them hostage or blackmail them, which ina sense I was, " Stallman would later tell author StevenLevy. "I was engaging in violence against them becauseI thought they were engaging in violence to everyone at large. "

Over time, Emacs became a sales tool for the hackerethic. The flexibility Stallman and built into thesoftware not only encouraged collaboration, it demandedit. Users who didn't keep abreast of the latestdevelopments in Emacs evolution or didn't contributetheir contributions back to Stallman ran the risk ofmissing out on the latest breakthroughs. And thebreakthroughs were many. Twenty years later, users hadmodified Emacs for so many different uses-using it as aspreadsheet, calculator, database, and web browser-thatlater Emacs developers adopted an overflowing sink torepresent its versatile functionality. "That's the ideathat we wanted to convey, " says Stallman. "The amountof stuff it has contained within it is both wonderfuland awful at the same time. "

Stallman's AI Lab contemporaries are more charitable. Hal Abelson, an MIT grad student who worked withStallman during the 1970s and would later assistStallman as a charter boardmember of the Free SoftwareFoundation, describes Emacs as "an absolutely brilliantcreation. " In giving programmers a way to add newsoftware libraries and features without messing up thesystem, Abelson says, Stallman paved the way for futurelarge-scale collaborative software projects. "Itsstructure was robust enough that you'd have people allover the world who were loosely collaborating [and]contributing to it, " Abelson says. "I don't know ifthat had been done before. "In writing this chapter, I've elected to focus moreonthe social significance of Emacs than the softwaresignificance. To read more about the software side, Irecommend Stallman's 1979 memo. I particularlyrecommend the section titled "Research ThroughDevelopment of Installed Tools" (#SEC27). Not only isit accessible to the nontechnical reader, it also shedslight on how closely intertwined Stallman's politicalphilosophies are with his software-design philosophies. A sample excerpt follows: EMACS could not have beenreached by a process of careful design, because suchprocesses arrive only at goals which are visible at theoutset, and whose desirability is established on thebottom line at the outset. Neither I nor anyone elsevisualized an extensible editor until I had made one, nor appreciated its value until he had experienced it. EMACS exists because I felt free to make individuallyuseful small improvements on a path whose end was notin sight. 

Guy Steele expresses similar admiration. Currently aresearch scientist for Sun Microsystems, he remembersStallman primarily as a "brilliant programmer with theability to generate large quantities of relativelybug-free code. " Although their personalities didn'texactly mesh, Steele and Stallman collaborated longenough for Steele to get a glimpse of Stallman'sintense coding style. He recalls a notable episode inthe late 1970s when the two programmers banded togetherto write the editor's "pretty print" feature. Originally conceived by Steele, pretty print wasanother keystroke-triggerd feature that reformattedEmacs' source code so that it was both more readableand took up less space, further bolstering theprogram's WYSIWIG qualities. The feature was strategicenough to attract Stallman's active interest, and itwasn't long before Steele wrote that he and Stallmanwere planning an improved version. 

"We sat down one morning, " recalls Steele. "I was atthe keyboard, and he was at my elbow, " says Steele. "Hewas perfectly willing to let me type, but he was alsotelling me what to type. 

The programming session lasted 10 hours. Throughoutthat entire time, Steele says, neither he nor Stallmantook a break or made any small talk. By the end of thesession, they had managed to hack the pretty printsource code to just under 100 lines. "My fingers wereon the keyboard the whole time, " Steele recalls, "butit felt like both of our ideas were flowing onto thescreen. He told me what to type, and I typed it. "

The length of the session revealed itself when Steelefinally left the AI Lab. Standing outside the buildingat 545 Tech Square, he was surprised to find himselfsurrounded by nighttime darkness. As a programmer, Steele was used to marathon coding sessions. Still, something about this session was different. Workingwith Stallman had forced Steele to block out allexternal stimuli and focus his entire mental energieson the task at hand. Looking back, Steele says he foundthe Stallman mind-meld both exhilarating and scary atthe same time. "My first thought afterward was: it wasa great experience, very intense, and that I neverwanted to do it again in my life. "

A Stark Moral Choice

On September 27, 1983, computer programmers logging onto the Usenet newsgroup net. Unix-wizards encountered anunusual message. Posted in the small hours of themorning, 12:30 a. M. To be exact, and signed byrms@mit-oz, the message's subject line was terse butattention-grabbing. "New UNIX implementation, " it read. Instead of introducing a newly released version ofUnix, however, the message's opening paragraph issued acall to arms: Starting this Thanksgiving I am going towrite a complete Unix-compatible software system calledGNU (for Gnu's Not Unix), and give it away free toeveryone who can use it. Contributions of time, money, programs and equipment are greatly needed. 1 To anexperienced Unix developer, the message was a mixtureof idealism and hubris. Not only did the author pledgeto rebuild the already mature Unix operating systemfrom the ground up, he also proposed to improve it inplaces. The new GNU system, the author predicted, wouldcarry all the usual components-a text editor, a shellprogram to run Unix-compatible applications, acompiler, "and a few other things. "See Richard Stallman, "Initial GNUAnnouncement"(September 1983). Http://www. Gnu. Ai. Mit. Edu/gnu/initial-announcement. Html It would also contain many enticing features thatother Unix systems didn't yet offer: a graphic userinterface based on the Lisp programming language, acrash-proof file system, and networking protocols builtaccording to MIT's internal networking system. 

"GNU will be able to run Unix programs, but will not beidentical to Unix, " the author wrote. "We will make allimprovements that are convenient, based on ourexperience with other operating systems. "

Anticipating a skeptical response on some readers'part, the author made sure to follow up hisoperating-system outline with a brief biographicalsketch titled, "Who am I?": I am Richard Stallman, inventor of the original much-imitated EMACS editor, now at the Artificial Intelligence Lab at MIT. I haveworked extensively on compilers, editors, debuggers, command interpreters, the Incompatible TimesharingSystem and the Lisp Machine operating system. Ipioneered terminal-independent display support in ITS. In addition I have implemented one crashproof filesystem and two window systems for Lisp machines. Asfate would have it, Stallman's fanciful GNU Projectmissed its Thanksgiving launch date. By January, 1984, however, Stallman made good on his promise and fullyimmersed himself in the world of Unix softwaredevelopment. For a software architect raised on ITS, itwas like designing suburban shopping malls instead ofMoorish palaces. Even so, building a Unix-likeoperating system had its hidden advantages. ITS hadbeen powerful, but it also possessed an Achilles' heel:MIT hackers had designed it to take specific advantageof the DEC-built PDP line. When AI Lab administratorselected to phase out the lab's powerful PDP-10 machinein the early 1980s, the operating system that hackersonce likened to a vibrant city became an instant ghosttown. Unix, on the other hand, was designed formobility and long-term survival. Originally developedby junior scientists at AT&T, the program had slippedout under corporate-management radar, finding a happyhome in the cash-strapped world of academic computersystems. With fewer resources than their MIT brethren, Unix developers had customized the software to rideatop a motley assortment of hardware systems:everything from the 16-bit PDP-11-a machine consideredfit for only small tasks by most AI Lab hackers-to32-bit mainframes such as the VAX 11/780. By 1983, afew companies, most notably Sun Microsystems, were evengoing so far as to develop a new generation ofmicrocomputers, dubbed "workstations, " to takeadvantage of the increasingly ubiquitous operating system. 

To facilitate this process, the developers in charge ofdesigning the dominant Unix strains made sure to keepan extra layer of abstraction between the software andthe machine. Instead of tailoring the operating systemto take advantage of a specific machine's resources-asthe AI Lab hackers had done with ITS and thePDP-10-Unix developers favored a more generic, off-the-rack approach. Focusing more on theinterlocking standards and specifications that held theoperating system's many subcomponents together, ratherthan the actual components themselves, they created asystem that could be quickly modified to suit thetastes of any machine. If a user quibbled with acertain portion, the standards made it possible to pullout an individual subcomponent and either fix it orreplace it with something better. Simply put, what theUnix approach lacked in terms of style or aesthetics, it more than made up for in terms of flexibility andeconomy, hence its rapid adoption. See Marshall Kirk McKusick, "Twenty Years ofBerkeleyUnix, " Open Sources (O'Reilly & Associates, Inc. , 1999): 38. 

Stallman's decision to start developing the GNU systemwas triggered by the end of the ITS system that the AILab hackers had nurtured for so long. The demise of ITShad been a traumatic blow to Stallman. Coming on theheels of the Xerox laser printer episode, it offeredfurther evidence that the AI Lab hacker culture waslosing its immunity to business practices in theoutside world. 

Like the software code that composed it, the roots ofITS' demise stretched way back. Defense spending, longa major font for computer-science research, had driedup during the post-Vietnam years. In a desperate questfor new funds, laboratories and universities turned tothe private sector. In the case of the AI Lab, winningover private investors was an easy sell. Home to someof the most ambitious computer-science projects of thepost-war era, the lab became a quick incubator oftechnology. Indeed, by 1980, most of the lab's staff, including many hackers, were dividing its time betweenInstitute and commercial projects. 

What at first seemed like a win-win deal-hackers got towork on the best projects, giving the lab first look atmany of the newest computer technologies coming downthe pike-soon revealed itself as a Faustian bargain. The more time hackers devoted to cutting-edgecommercial projects, the less time they had to devoteto general maintenance on the lab's baroque softwareinfrastructure. Soon, companies began hiring awayhackers outright in an attempt to monopolize their timeand attention. With fewer hackers to mind the shop, programs and machines took longer to fix. Even worse, Stallman says, the lab began to undergo a "demographicchange. " The hackers who had once formed a vocalminority within the AI Lab were losing membership while"the professors and the students who didn't really lovethe [PDP-10] were just as numerous as before. "See Richard Stallman (1986). 

The breaking point came in 1982. That was the year thelab's administration decided to upgrade its maincomputer, the PDP-10. Digital, the corporation thatmanufactured the PDP-10, had discontinued the line. Although the company still offered a high-poweredmainframe, dubbed the KL-10, the new machine required adrastic rewrite or "port" of ITS if hackers wanted tocontinue running the same operating system. Fearfulthat the lab had lost its critical mass of in-houseprogramming talent, AI Lab faculty members pressed forTwenex, a commercial operating system developed byDigital. Outnumbered, the hackers had no choice but to comply. 

"Without hackers to maintain the system, [facultymembers] said, `We're going to have a disaster; we musthave commercial software, '" Stallman would recall a fewyears later. "They said, `We can expect the company tomaintain it. ' It proved that they were utterly wrong, but that's what they did. "

At first, hackers viewed the Twenex system as yetanother authoritarian symbol begging to be subverted. The system's name itself was a protest. Officiallydubbed TOPS-20 by DEC, it was a successor to TOPS-10, acommercial operating system DEC marketed for thePDP-10. Bolt Beranek Newman had deveoped an improvedversion, dubbed Tenex, which TOPS-20 drew upon. Multiple sources: see RichardStallman interview, Gerald Sussman email, and Jargon File 3. 0. 0. Http://www. Clueless. Com/jargon3. 0. 0/TWENEX. Html Stallman, the hacker who coined the Twenex term, sayshe came up with the name as a way to avoid using theTOPS-20 name. "The system was far from tops, so therewas no way I was going to call it that, " Stallmanrecalls. "So I decided to insert a `w' in the Tenexname and call it Twenex. "

The machine that ran the Twenex/TOPS-20 system had itsown derisive nickname: Oz. According to one hackerlegend, the machine got its nickname because itrequired a smaller PDP-11 machine to power itsterminal. One hacker, upon viewing the KL-10-PDP-11setup for the first time, likened it to the wizard'sbombastic onscreen introduction in the Wizard of Oz. "Iam the great and powerful Oz, " the hacker intoned. "Payno attention to the PDP-11 behind that console. "Seehttp://www. As. Cmu. Edu/~geek/humor/See_Figure_1. Txt

If hackers laughed when they first encountered theKL-10, their laughter quickly died when theyencountered Twenex. Not only did Twenex boast built-insecurity, but the system's software engineers haddesigned the tools and applications with the securitysystem in mind. What once had been a cat-and-mouse gameover passwords in the case of the Laboratory forComputer Science's security system, now became anout-and-out battle over system management. Systemadministrators argued that without security, the Ozsystem was more prone to accidental crashes. Hackersargued that crashes could be better prevented byoverhauling the source code. Unfortunately, the numberof hackers with the time and inclination to performthis sort of overhaul had dwindled to the point thatthe system-administrator argument prevailed. 

Cadging passwords and deliberately crashing the systemin order to glean evidence from the resulting wreckage, Stallman successfully foiled the system administrators'attempt to assert control. After one foiled "coupd'etat, " Stallman issued an alert to the entire AI staff. 

"There has been another attempt to seize power, "Stallman wrote. "So far, the aristocratic forces havebeen defeated. " To protect his identity, Stallmansigned the message "Radio Free OZ. "

The disguise was a thin one at best. By 1982, Stallman's aversion to passwords and secrecy had becomeso well known that users outside the AI Laboratory wereusing his account as a stepping stone to the ARPAnet, the research-funded computer network that would serveas a foundation for today's Internet. One such"tourist" during the early 1980s was Don Hopkins, aCalifornia programmer who learned through the hackinggrapevine that all an outsider needed to do to gainaccess to MIT's vaunted ITS system was to log in underthe initials RMS and enter the same three-lettermonogram when the system requested a password. 

"I'm eternally grateful that MIT let me and many otherpeople use their computers for free, " says Hopkins. "Itmeant a lot to many people. "

This so-called "tourist" policy, which had been openlytolerated by MIT management during the ITS years, See "MIT AI Lab TouristPolicy. "http://catalog. Com/hopkins/text/tourist-policy. Html fell by the wayside when Oz became the lab's primarylink to the ARPAnet. At first, Stallman continued hispolicy of repeating his login ID as a password sooutside users could follow in his footsteps. Over time, however, the Oz's fragility prompted administrators tobar outsiders who, through sheer accident or maliciousintent, might bring down the system. When those sameadministrators eventually demanded that Stallman stoppublishing his password, Stallman, citing personalethics, refused to do so and ceased using the Oz systemaltogether. 3

"[When] passwords first appeared at the MIT AI Lab I[decided] to follow my belief that there should be nopasswords, " Stallman would later say. "Because I don'tbelieve that it's really desirable to have security ona computer, I shouldn't be willing to help uphold thesecurity regime. "

Stallman's refusal to bow before the great and powerfulOz symbolized the growing tension between hackers andAI Lab management during the early 1980s. This tensionpaled in comparison to the conflict that raged withinthe hacker community itself. By the time the KL-10arrived, the hacker community had already divided intotwo camps. The first centered around a software companycalled Symbolics, Inc. The second centered aroundSymbolics chief rival, Lisp Machines, Inc. (LMI). Bothcompanies were in a race to market the Lisp Machine, adevice built to take full advantage of the Lispprogramming language. 

Created by artificial-intelligence research pioneerJohn McCarthy, a MIT artificial-intelligence researcherduring the late 1950s, Lisp is an elegant languagewell-suited for programs charged with heavy-dutysorting and processing. The language's name is ashortened version of LISt Processing. FollowingMcCarthy's departure to the Stanford ArtificialIntelligence Laboratory, MIT hackers refined thelanguage into a local dialect dubbed MACLISP. The "MAC"stood for Project MAC, the DARPA-funded researchproject that gave birth to the AI Lab and theLaboratory for Computer Science. Led by AI Labarch-hacker Richard Greenblatt, AI Lab programmersduring the 1970s built up an entire Lisp-basedoperating system, dubbed the Lisp Machine operatingsystem. By 1980, the Lisp Machine project had generatedtwo commercial spin-offs. Symbolics was headed byRussell Noftsker, a former AI Lab administrator, andLisp Machines, Inc. , was headed by Greenblatt. 

The Lisp Machine software was hacker-built, meaning itwas owned by MIT but available for anyone to copy asper hacker custom. Such a system limited the marketingadvantage of any company hoping to license the softwarefrom MIT and market it as unique. To secure anadvantage, and to bolster the aspects of the operatingsystem that customers might consider attractive, thecompanies recruited various AI Lab hackers and set themworking on various components of the Lisp Machineoperating system outside the auspices of the AI Lab. 

The most aggressive in this strategy was Symbolics. Bythe end of 1980, the company had hired 14 AI Labstaffers as part-time consultants to develop itsversion of the Lisp Machine. Apart from Stallman, therest signed on to help LMI. See H. P. Newquist, The Brain Makers: Genius, Ego, andGreed in the Quest for Machines that Think (SamsPublishing, 1994): 172. 

At first, Stallman accepted both companies' attempt tocommercialize the Lisp machine, even though it meantmore work for him. Both licensed the Lisp Machine OSsource code from MIT, and it was Stallman's job toupdate the lab's own Lisp Machine to keep pace with thelatest innovations. Although Symbolics' license withMIT gave Stallman the right to review, but not copy, Symbolics' source code, Stallman says a "gentleman'sagreement" between Symbolics management and the AI Labmade it possible to borrow attractive snippets intraditional hacker fashion. 

On March 16, 1982, a date Stallman remembers wellbecause it was his birthday, Symbolics executivesdecided to end this gentlemen's agreement. The move waslargely strategic. LMI, the primary competition in theLisp Machine marketplace, was essentially using a copyof the AI Lab Lisp Machine. Rather than subsidize thedevelopment of a market rival, Symbolics executiveselected to enforce the letter of the license. If the AILab wanted its operating system to stay current withthe Symbolics operating system, the lab would have toswitch over to a Symbolics machine and sever itsconnection to LMI. 

As the person responsible for keeping up the lab's LispMachine, Stallman was incensed. Viewing thisannouncement as an "ultimatum, " he retaliated bydisconnecting Symbolics' microwave communications linkto the laboratory. He then vowed never to work on aSymbolics machine and pledged his immediate allegianceto LMI. "The way I saw it, the AI Lab was a neutralcountry, like Belgium in World War I, " Stallman says. "If Germany invades Belgium, Belgium declares war onGermany and sides with Britain and France. "

The circumstances of the so-called "Symbolics War" of1982-1983 depend heavily on the source doing thetelling. When Symbolics executives noticed that theirlatest features were still appearing in the AI Lab LispMachine and, by extension, the LMI Lisp machine, theyinstalled a "spy" program on Stallman's computerterminal. Stallman says he was rewriting the featuresfrom scratch, taking advantage of the license's reviewclause but also taking pains to make the source code asdifferent as possible. Symbolics executives arguedotherwise and took their case to MIT administration. According to 1994 book, The Brain Makers: Genius, Ego, and Greed, and the Quest for Machines That Think, written by Harvey Newquist, the administrationresponded with a warning to Stallman to "stay away"from the Lisp Machine project. Ibid. : 196. According to Stallman, MIT administrators backed Stallman up. "I was neverthreatened, " he says. "I did make changes in mypractices, though. Just to be ultra safe, I no longerread their source code. I used only the documentationand wrote the code from that. "

Whatever the outcome, the bickering solidifiedStallman's resolve. With no source code to review, Stallman filled in the software gaps according to hisown tastes and enlisted members of the AI Lab toprovide a continuous stream of bug reports. He alsomade sure LMI programmers had direct access to thechanges. "I was going to punish Symbolics if it was thelast thing I did, " Stallman says. 

Such statements are revealing. Not only do they shedlight on Stallman's nonpacifist nature, they alsoreflect the intense level of emotion triggered by theconflict. According to another Newquist-related story, Stallman became so irate at one point that he issued anemail threatening to "wrap myself in dynamite and walkinto Symbolics' offices. "Ibid. Newquist, who says this anecdote was confirmedbyseveral Symbolics executives, writes, "The messagecaused a brief flurry of excitement and speculation onthe part of Symbolics' employees, but ultimately, noone took Stallman's outburst that seriously. " Although Stallman would deny any memory of the emailand still describes its existence as a "vicious rumor, "he acknowledges that such thoughts did enter his head. "I definitely did have fantasies of killing myself anddestroying their building in the process, " Stallmansays. "I thought my life was over. "

The level of despair owed much to what Stallman viewedas the "destruction" of his "home"-i. E. , the demise ofthe AI Lab's close-knit hacker subculture. In a lateremail interview with Levy, Stallman would liken himselfto the historical figure Ishi, the last survivingmember of the Yahi, a Pacific Northwest tribe wiped outduring the Indian wars of the 1860s and 1870s. Theanalogy casts Stallman's survival in epic, almostmythical, terms. In reality, however, it glosses overthe tension between Stallman and his fellow AI Labhackers prior to the Symbolics-LMI schism. Instead ofseeing Symbolics as an exterminating force, many ofStallman's colleagues saw it as a belated bid forrelevance. In commercializing the Lisp Machine, thecompany pushed hacker principles of engineer-drivensoftware design out of the ivory-tower confines of theAI Lab and into the corporate marketplace wheremanager-driven design principles held sway. Rather thanviewing Stallman as a holdout, many hackers saw him asa troubling anachronism. 

Stallman does not dispute this alternate view ofhistorical events. In fact, he says it was yet anotherreason for the hostility triggered by the Symbolics"ultimatum. " Even before Symbolics hired away most ofthe AI Lab's hacker staff, Stallman says many of thehackers who later joined Symbolics were shunning him. "I was no longer getting invited to go to Chinatown, "Stallman recalls. "The custom started by Greenblatt wasthat if you went out to dinner, you went around or senta message asking anybody at the lab if they also wantedto go. Sometime around 1980-1981, I stopped gettingasked. They were not only not inviting me, but oneperson later confessed that he had been pressured tolie to me to keep their going away to dinner without mea secret. "

Although Stallman felt anger toward the hackers whoorchestrated this petty form of ostracism, theSymbolics controversy dredged up a new kind of anger, the anger of a person about to lose his home. WhenSymbolics stopped sending over its source-code changes, Stallman responded by holing up in his MIT offices andrewriting each new software feature and tool fromscratch. Frustrating as it may have been, it guaranteedthat future Lisp Machine users had unfettered access tothe same features as Symbolics users. 

It also guaranteed Stallman's legendary status withinthe hacker community. Already renowned for his workwith Emacs, Stallman's ability to match the output ofan entire team of Symbolics programmers-a team thatincluded more than a few legendary hackers itself-stillstands has one of the major human accomplishments ofthe Information Age, or of any age for that matter. Dubbing it a "master hack" and Stallman himself a"virtual John Henry of computer code, " author StevenLevy notes that many of his Symbolics-employed rivalshad no choice but to pay their idealistic formercomrade grudging respect. Levy quotes Bill Gosper, ahacker who eventually went to work for Symbolics in thecompany's Palo Alto office, expressing amazement overStallman's output during this period: I can seesomething Stallman wrote, and I might decide it was bad(probably not, but somebody could convince me it wasbad), and I would still say, "But wait aminute-Stallman doesn't have anybody to argue with allnight over there. He's working alone! It's incredibleanyone could do this alone!"See Steven Levy, Hackers (Penguin USA [paperback], 1984): 426. For Stallman, the months spent playing catch up withSymbolics evoke a mixture of pride and profoundsadness. As a dyed-in-the-wool liberal whose father hadserved in World War II, Stallman is no pacifist. Inmany ways, the Symbolics war offered the rite ofpassage toward which Stallman had been careening eversince joining the AI Lab staff a decade before. At thesame time, however, it coincided with the traumaticdestruction of the AI Lab hacker culture that hadnurtured Stallman since his teenage years. One day, while taking a break from writing code, Stallmanexperienced a traumatic moment passing through thelab's equipment room. There, Stallman encountered thehulking, unused frame of the PDP-10 machine. Startledby the dormant lights, lights that once activelyblinked out a silent code indicating the status of theinternal program, Stallman says the emotional impactwas not unlike coming across a beloved family member'swell-preserved corpse. 

"I started crying right there in the equipment room, "he says. "Seeing the machine there, dead, with nobodyleft to fix it, it all drove home how completely mycommunity had been destroyed. "

Stallman would have little opportunity to mourn. TheLisp Machine, despite all the furor it invoked and allthe labor that had gone into making it, was merely asideshow to the large battles in the technologymarketplace. The relentless pace of computerminiaturization was bringing in newer, more powerfulmicroprocessors that would soon incorporate themachine's hardware and software capabilities like amodern metropolis swallowing up an ancient desert village. 

Riding atop this microprocessor wave werehundreds-thousands-of commercial software programs, each protected by a patchwork of user licenses andnondisclosure agreements that made it impossible forhackers to review or share source code. The licenseswere crude and ill-fitting, but by 1983 they had becomestrong enough to satisfy the courts and scare awaywould-be interlopers. Software, once a form of garnishmost hardware companies gave away to make theirexpensive computer systems more flavorful, was quicklybecoming the main dish. In their increasing hunger fornew games and features, users were putting aside thetraditional demand to review the recipe after every meal. 

Nowhere was this state of affairs more evident than inthe realm of personal computer systems. Companies suchas Apple Computer and Commodore were minting freshmillionaires selling machines with built-in operatingsystems. Unaware of the hacker culture and its distastefor binary-only software, many of these users sawlittle need to protest when these companies failed toattach the accompanying source-code files. A fewanarchic adherents of the hacker ethic helped propelthat ethic into this new marketplace, but for the mostpart, the marketplace rewarded the programmers speedyenough to write new programs and savvy enough tocopyright them as legally protected works. 

One of the most notorious of these programmers was BillGates, a Harvard dropout two years Stallman's junior. Although Stallman didn't know it at the time, sevenyears before sending out his message to the net. Unix-wizards newsgroup, Gates, a buddingentrepreneur and general partner with theAlbuquerque-based software firm Micro-Soft, laterspelled as Microsoft, had sent out his own open letterto the software-developer community. Written inresponse to the PC users copying Micro-Soft's softwareprograms, Gates' " Open Letter to Hobbyists" hadexcoriated the notion of communal software development. 

"Who can afford to do professional work for nothing?"asked Gates. "What hobbyist can put three man-yearsinto programming, finding all bugs, documenting hisproduct, and distributing it for free?"See Bill Gates, "An Open Letter toHobbyists" (February3, 1976). To view an online copy of this letter, go to

http://www. Blinkenlights. Com/classiccmp/gateswhine. Html. 

Although few hackers at the AI Lab saw the missive, Gates' 1976 letter nevertheless represented thechanging attitude toward software both among commercialsoftware companies and commercial software developers. Why treat software as a zero-cost commodity when themarket said otherwise? As the 1970s gave way to the1980s, selling software became more than a way torecoup costs; it became a political statement. At atime when the Reagan Administration was rushing todismantle many of the federal regulations and spendingprograms that had been built up during the half centuryfollowing the Great Depression, more than a fewsoftware programmers saw the hacker ethic asanticompetitive and, by extension, un-American. Atbest, it was a throwback to the anticorporate attitudesof the late 1960s and early 1970s. Like a Wall Streetbanker discovering an old tie-dyed shirt hiding betweenFrench-cuffed shirts and double-breasted suits, manycomputer programmers treated the hacker ethic as anembarrassing reminder of an idealistic age. 

For a man who had spent the entire 1960s as anembarrassing throwback to the 1950s, Stallman didn'tmind living out of step with his peers. As a programmerused to working with the best machines and the bestsoftware, however, Stallman faced what he could onlydescribe as a "stark moral choice": either get over hisethical objection for " proprietary" software-the termStallman and his fellow hackers used to describe anyprogram that carried private copyright or end-userlicense that restricted copying and modification-ordedicate his life to building an alternate, nonproprietary system of software programs. Given hisrecent months-long ordeal with Symbolics, Stallman feltmore comfortable with the latter option. "I suppose Icould have stopped working on computers altogether, "Stallman says. "I had no special skills, but I'm sure Icould have become a waiter. Not at a fancy restaurant, probably, but I could've been a waiter somewhere. "

Being a waiter-i. E. , dropping out of programmingaltogether-would have meant completely giving up anactivity, computer programming, that had given him somuch pleasure. Looking back on his life since moving toCambridge, Stallman finds it easy to identify lengthyperiods when software programming provided the onlypleasure. Rather than drop out, Stallman decided tostick it out. 

An atheist, Stallman rejects notions such as fate, dharma, or a divine calling in life. Nevertheless, hedoes feel that the decision to shun proprietarysoftware and build an operating system to help othersdo the same was a natural one. After all, it wasStallman's own personal combination of stubbornness, foresight, and coding virtuosity that led him toconsider a fork in the road most others didn't knowexisted. In describing the decision in a chapter forthe 1999 book, Open Sources, Stallman cites the spiritencapsulated in the words of the Jewish sage Hillel: IfI am not for myself, who will be for me?If I am onlyfor myself, what am I?If not now, when?See Richard Stallman, Open Sources(O'Reilly &Associates, Inc. , 1999): 56. Stallman adds his ownfootnote to this statement, writing, "As an atheist, Idon't follow any religious leaders, but I sometimesfind I admire something one of them has said. " Speaking to audiences, Stallman avoids the religiousroute and expresses the decision in pragmatic terms. "Iasked myself: what could I, an operating-systemdeveloper, do to improve the situation? It wasn't untilI examined the question for a while that I realized anoperating-system developer was exactly what was neededto solve the problem. "

Once he reached that decision, Stallman says, everything else "fell into place. " He would abstainfrom using software programs that forced him tocompromise his ethical beliefs, while at the same timedevoting his life to the creation of software thatwould make it easier for others to follow the samepath. Pledging to build a free software operatingsystem "or die trying-of old age, of course, " Stallmanquips, he resigned from the MIT staff in January, 1984, to build GNU. 

The resignation distanced Stallman's work from thelegal auspices of MIT. Still, Stallman had enoughfriends and allies within the AI Lab to retainrent-free access to his MIT office. He also had theability to secure outside consulting gigs to underwritethe early stages of the GNU Project. In resigning fromMIT, however, Stallman negated any debate aboutconflict of interest or Institute ownership of thesoftware. The man whose early adulthood fear of socialisolation had driven him deeper and deeper into the AILab's embrace was now building a legal firewall betweenhimself and that environment. 

For the first few months, Stallman operated inisolation from the Unix community as well. Although hisannouncement to the net. Unix-wizards group hadattracted sympathetic responses, few volunteers signedon to join the crusade in its early stages. 

"The community reaction was pretty much uniform, "recalls Rich Morin, leader of a Unix user group at thetime. "People said, `Oh, that's a great idea. Show usyour code. Show us it can be done. '"

In true hacker fashion, Stallman began looking forexisting programs and tools that could be convertedinto GNU programs and tools. One of the first was acompiler named VUCK, which converted programs writtenin the popular C programming language intomachine-readable code. Translated from the Dutch, theprogram's acronym stood for the Free UniversityCompiler Kit. Optimistic, Stallman asked the program'sauthor if the program was free. When the authorinformed him that the words "Free University" were areference to the Vrije Universiteit in Amsterdam, Stallman was chagrined. 

"He responded derisively, stating that the universitywas free but the compiler was not, " recalls Stallman. "I therefore decided that my first program for the GNUProject would be a multi-language, multi-platform compiler. "

Eventually Stallman found a Pastel language compilerwritten by programmers at Lawrence Livermore NationalLab. According to Stallman's knowledge at the time, thecompiler was free to copy and modify. Unfortunately, the program possessed a sizable design flaw: it savedeach program into core memory, tying up precious spacefor other software activities. On mainframe systemsthis design flaw had been forgivable. On Unix systemsit was a crippling barrier, since the machines that ranUnix were too small to handle the large filesgenerated. Stallman made substantial progress at first, building a C-compatible frontend to the compiler. Bysummer, however, he had come to the conclusion that hewould have to build a totally new compiler from scratch. 

In September of 1984, Stallman shelved compilerdevelopment for the near term and began searching forlower-lying fruit. He began development of a GNUversion of Emacs, the program he himself had beensupervising for a decade. The decision was strategic. Within the Unix community, the two native editorprograms were vi, written by Sun Microsystems cofounderBill Joy, and ed, written by Bell Labs scientist (andUnix cocreator) Ken Thompson. Both were useful andpopular, but neither offered the endlessly expandablenature of Emacs. In rewriting Emacs for the Unixaudience, Stallman stood a better chance of showing offhis skills. It also stood to reason that Emacs usersmight be more attuned to the Stallman mentality. 

Looking back, Stallman says he didn't view the decisionin strategic terms. "I wanted an Emacs, and I had agood opportunity to develop one. "

Once again, the notion of reinventing the wheel gratedon Stallman's efficient hacker sensibilities. Inwriting a Unix version of Emacs, Stallman was soonfollowing the footsteps of Carnegie Mellon graduatestudent James Gosling, author of a C-based versiondubbed Gosling Emacs or GOSMACS. Gosling's version ofEmacs included an interpreter that exploited asimplified offshoot of the Lisp language calledMOCKLISP. Determined to build GNU Emacs on a similarLisp foundation, Stallman borrowed copiously fromGosling's innovations. Although Gosling had put GOSMACSunder copyright and had sold the rights to UniPress, aprivately held software company, Stallman cited theassurances of a fellow developer who had participatedin the early MOCKLISP interpreter. According to thedeveloper, Gosling, while a Ph. D. Student at CarnegieMellon, had assured early collaborators that their workwould remain accessible. When UniPress caught wind ofStallman's project, however, the company threatened toenforce the copyright. Once again, Stallman faced theprospect of building from the ground up. 

In the course of reverse-engineering Gosling'sinterpreter, Stallman would create a fully functionalLisp interpreter, rendering the need for Gosling'soriginal interpreter moot. Nevertheless, the notion ofdevelopers selling off software rights-indeed, the verynotion of developers having software rights to sell inthe first place-rankled Stallman. In a 1986 speech atthe Swedish Royal Technical Institute, Stallman citedthe UniPress incident as yet another example of thedangers associated with proprietary software. 

"Sometimes I think that perhaps one of the best thingsI could do with my life is find a gigantic pile ofproprietary software that was a trade secret, and starthanding out copies on a street corner so it wouldn't bea trade secret any more, " said Stallman. "Perhaps thatwould be a much more efficient way for me to givepeople new free software than actually writing itmyself; but everyone is too cowardly to even take it. "

Despite the stress it generated, the dispute overGosling's innovations would assist both Stallman andthe free software movement in the long term. It wouldforce Stallman to address the weaknesses of the EmacsCommune and the informal trust system that had allowedproblematic offshoots to emerge. It would also forceStallman to sharpen the free software movement'spolitical objectives. Following the release of GNUEmacs in 1985, Stallman issued " The GNU Manifesto, " anexpansion of the original announcement posted inSeptember, 1983. Stallman included within the documenta lengthy section devoted to the many arguments used bycommercial and academic programmers to justify theproliferation of proprietary software programs. Oneargument, "Don't programmers deserve a reward for theircreativity, " earned a response encapsulating Stallman'sanger over the recent Gosling Emacs episode:

"If anything deserves a reward, it is socialcontribution, " Stallman wrote. "Creativity can be asocial contribution, but only in so far [sic] associety is free to use the results. If programmersdeserve to be rewarded for creating innovativeprograms, by the same token they deserve to be punishedif they restrict the use of these programs. "See Richard Stallman, "The GNUManifesto" (1985). Http://www. Gnu. Org/manifesto. Html

With the release of GNU Emacs, the GNU Project finallyhad code to show. It also had the burdens of anysoftware-based enterprise. As more and more Unixdevelopers began playing with the software, money, gifts, and requests for tapes began to pour in. Toaddress the business side of the GNU Project, Stallmandrafted a few of his colleagues and formed the FreeSoftware Foundation (FSF), a nonprofit organizationdedicated to speeding the GNU Project towards its goal. With Stallman as president and various hacker allies asboard members, the FSF helped provide a corporate facefor the GNU Project. 

Robert Chassell, a programmer then working at LispMachines, Inc. , became one of five charter boardmembers at the Free Software Foundation following adinner conversation with Stallman. Chassell also servedas the organization's treasurer, a role that startedsmall but quickly grew. 

"I think in '85 our total expenses and revenue weresomething in the order of $23, 000, give or take, "Chassell recalls. "Richard had his office, and weborrowed space. I put all the stuff, especially thetapes, under my desk. It wasn't until sometime laterLMI loaned us some space where we could store tapes andthings of that sort. "

In addition to providing a face, the Free SoftwareFoundation provided a center of gravity for otherdisenchanted programmers. The Unix market that hadseemed so collegial even at the time of Stallman'sinitial GNU announcement was becoming increasinglycompetitive. In an attempt to tighten their hold oncustomers, companies were starting to close off accessto Unix source code, a trend that only speeded thenumber of inquiries into ongoing GNU software projects. The Unix wizards who once regarded Stallman as a noisykook were now beginning to see him as a software Cassandra. 

"A lot of people don't realize, until they've had ithappen to them, how frustrating it can be to spend afew years working on a software program only to have ittaken away, " says Chassell, summarizing the feelingsand opinions of the correspondents writing in to theFSF during the early years. "After that happens acouple of times, you start to say to yourself, `Hey, wait a minute. '"

For Chassell, the decision to participate in the FreeSoftware Foundation came down to his own personalfeelings of loss. Prior to LMI, Chassell had beenworking for hire, writing an introductory book on Unixfor Cadmus, Inc. , a Cambridge-area software company. When Cadmus folded, taking the rights to the book downwith it, Chassell says he attempted to buy the rightsback with no success. 

"As far as I know, that book is still sitting on shelfsomewhere, unusable, uncopyable, just taken out of thesystem, " Chassell says. "It was quite a goodintroduction if I may say so myself. It would havetaken maybe three or four months to convert [the book]into a perfectly usable introduction to GNU/Linuxtoday. The whole experience, aside from what I have inmy memory, was lost. "

Forced to watch his work sink into the mire while hiserstwhile employer struggled through bankruptcy, Chassell says he felt a hint of the anger that droveStallman to fits of apoplexy. "The main clarity, forme, was the sense that if you want to have a decentlife, you don't want to have bits of it closed off, "Chassell says. "This whole idea of having the freedomto go in and to fix something and modify it, whateverit may be, it really makes a difference. It makes onethink happily that after you've lived a few years thatwhat you've done is worthwhile. Because otherwise itjust gets taken away and thrown out or abandoned or, atthe very least, you no longer have any relation to it. It's like losing a bit of your life. "

St. Ignucius

The Maui High Performance Computing Center is locatedin a single-story building in the dusty red hills justabove the town of Kihei. Framed by million-dollar viewsand the multimillion dollar real estate of theSilversword Golf Course, the center seems like theultimate scientific boondoggle. Far from the boxy, sterile confines of Tech Square or even the sprawlingresearch metropolises of Argonne, Illinois and LosAlamos, New Mexico, the MHPCC seems like the kind ofplace where scientists spend more time on their tansthan their post-doctoral research projects. 

The image is only half true. Although researchers atthe MHPCC do take advantage of the local recreationalopportunities, they also take their work seriously. According to Top500. Org, a web site that tracks themost powerful supercomputers in the world, the IBM SPPower3 supercomputer housed within the MHPCC clocks inat 837 billion floating-point operations per second, making it one of 25 most powerful computers in theworld. Co-owned and operated by the University ofHawaii and the U. S. Air Force, the machine divides itscomputer cycles between the number crunching tasksassociated with military logistics and high-temperaturephysics research. 

Simply put, the MHPCC is a unique place, a place wherethe brainy culture of science and engineering and thelaid-back culture of the Hawaiian islands coexist inpeaceful equilibrium. A slogan on the lab's 2000 website sums it up: "Computing in paradise. "

It's not exactly the kind of place you'd expect to findRichard Stallman, a man who, when taking in thebeautiful view of the nearby Maui Channel through thepicture windows of a staffer's office, mutters a tersecritique: "Too much sun. " Still, as an emissary fromone computing paradise to another, Stallman has amessage to deliver, even if it means subjecting hispale hacker skin to the hazards of tropical exposure. 

The conference room is already full by the time Iarrive to catch Stallman's speech. The gender breakdownis a little better than at the New York speech, 85%male, 15% female, but not by much. About half of theaudience members wear khaki pants and logo-encrustedgolf shirts. The other half seems to have gone native. Dressed in the gaudy flower-print shirts so popular inthis corner of the world, their faces are a deep shadeof ochre. The only residual indication of geek statusare the gadgets: Nokia cell phones, Palm Pilots, andSony VAIO laptops. 

Needless to say, Stallman, who stands in front of theroom dressed in plain blue T-shirt, brown polyesterslacks, and white socks, sticks out like a sore thumb. The fluorescent lights of the conference room helpbring out the unhealthy color of his sun-starved skin. His beard and hair are enough to trigger beads of sweaton even the coolest Hawaiian neck. Short of having thewords "mainlander" tattooed on his forehead, Stallmancouldn't look more alien if he tried. 

As Stallman putters around the front of the room, a fewaudience members wearing T-shirts with the logo of theMaui FreeBSD Users Group (MFUG) race to set up cameraand audio equipment. FreeBSD, a free software offshootof the Berkeley Software Distribution, the venerable1970s academic version of Unix, is technically acompetitor to the GNU/Linux operating system. Still, inthe hacking world, Stallman speeches are documentedwith a fervor reminiscent of the Grateful Dead and itslegendary army of amateur archivists. As the local freesoftware heads, it's up to the MFUG members to makesure fellow programmers in Hamburg, Mumbai, andNovosibirsk don't miss out on the latest pearls of RMS wisdom. 

The analogy to the Grateful Dead is apt. Often, whendescribing the business opportunities inherent withinthe free software model, Stallman has held up theGrateful Dead as an example. In refusing to restrictfans' ability to record live concerts, the GratefulDead became more than a rock group. They became thecenter of a tribal community dedicated to Grateful Deadmusic. Over time, that tribal community became so largeand so devoted that the band shunned record contractsand supported itself solely through musical tours andlive appearances. In 1994, the band's last year as atouring act, the Grateful Dead drew $52 million in gatereceipts alone. See "Grateful Dead Time Capsule: 1985-1995 NorthAmerican Tour Grosses. " http://www. Accessplace. Com/gdtc/1197. Htm

While few software companies have been able to matchthat success, the tribal aspect of the free softwarecommunity is one reason many in the latter half of the1990s started to accept the notion that publishingsoftware source code might be a good thing. Hoping tobuild their own loyal followings, companies such asIBM, Sun Microsystems, and Hewlett Packard have come toaccept the letter, if not the spirit, of the Stallmanfree software message. Describing the GPL as theinformation-technology industry's "Magna Carta, " ZDNetsoftware columnist Evan Leibovitch sees the growingaffection for all things GNU as more than just a trend. "This societal shift is letting users take back controlof their futures, " Leibovitch writes. "Just as theMagna Carta gave rights to British subjects, the GPLenforces consumer rights and freedoms on behalf of theusers of computer software. "See Evan Leibovitch, "Who's Afraid of Big BadWolves, "ZDNet Tech Update (December 15, 2000). Http://techupdate. Zdnet. Com/techupdate/stories/main/0Y/A

The tribal aspect of the free software community alsohelps explain why 40-odd programmers, who mightotherwise be working on physics projects or surfing theWeb for windsurfing buoy reports, have packed into aconference room to hear Stallman speak. 

Unlike the New York speech, Stallman gets nointroduction. He also offers no self-introduction. Whenthe FreeBSD people finally get their equipment up andrunning, Stallman simply steps forward, startsspeaking, and steamrolls over every other voice in the room. 

"Most of the time when people consider the question ofwhat rules society should have for using software, thepeople considering it are from software companies, andthey consider the question from a self-servingperspective, " says Stallman, opening his speech. "Whatrules can we impose on everybody else so they have topay us lots of money? I had the good fortune in the1970s to be part of a community of programmers whoshared software. And because of this I always like tolook at the same issue from a different direction toask: what kind of rules make possible a good societythat is good for the people who are in it? Andtherefore I reach completely different answers. "

Once again, Stallman quickly segues into the parable ofthe Xerox laser printer, taking a moment to deliver thesame dramatic finger-pointing gestures to the crowd. Healso devotes a minute or two to the GNU/Linux name. 

"Some people say to me, `Why make such a fuss aboutgetting credit for this system? After all, theimportant thing is the job is done, not whether you getrecognition for it. ' Well, this would be wise advice ifit were true. But the job wasn't to build an operatingsystem; the job is to spread freedom to the users ofcomputers. And to do that we have to make it possibleto do everything with computers in freedom. "For narrative purposes, I havehesitated to go in-depthwhen describing Stallman's full definition of software"freedom. " The GNU Project web site lists fourfundamental components: The freedom to run a program, for any purpose (freedom 0). The freedom to study how aprogram works, and adapt it to your needs (freedom 1). The freedom to redistribute copies of a program so youcan help your neighbor (freedom 2). The freedom toimprove the program, and release your improvements tothe public, so that the whole community benefits(freedom 3). For more information, please visit "TheFree Software Definition" athttp://www. Gnu. Org/philosophy/free-sw. Html. 

Adds Stallman, "There's a lot more work to do. "

For some in the audience, this is old material. Forothers, it's a little arcane. When a member of thegolf-shirt contingent starts dozing off, Stallman stopsthe speech and asks somebody to wake the person up. 

"Somebody once said my voice was so soothing, he askedif I was some kind of healer, " says Stallman, drawing aquick laugh from the crowd. "I guess that probablymeans I can help you drift gently into a blissful, relaxing sleep. And some of you might need that. Iguess I shouldn't object if you do. If you need tosleep, by all means do. "

The speech ends with a brief discussion of softwarepatents, a growing issue of concern both within thesoftware industry and within the free softwarecommunity. Like Napster, software patents reflect theawkward nature of applying laws and concepts writtenfor the physical world to the frictionless universe ofinformation technology. The difference betweenprotecting a program under copyright and protecting aprogram under software patents is subtle butsignificant. In the case of copyright, a softwarecreator can restrict duplication of the source code butnot duplication of the idea or functionality that thesource code addresses. In other words, if a developerchooses not to use a software program under theoriginal developer's terms, that second developer isstill free to reverse-engineer the program-i. E. , duplicate the software program's functionality byrewriting the source code from scratch. Suchduplication of ideas is common within the commercialsoftware industry, where companies often isolatereverse-engineering teams to head off accusations ofcorporate espionage or developer hanky-panky. In thejargon of modern software development, companies referto this technique as "clean room" engineering. 

Software patents work differently. According to theU. S. Patent Office, companies and individuals maysecure patents for innovative algorithms provided theysubmit their claims to a public review. In theory, thisallows the patent-holder to trade off disclosure oftheir invention for a limited monopoly of a minimum of20 years after the patent filing. In practice, thedisclosure is of limited value, since the operation ofthe program is often self-evident. Unlike copyright, apatent gives its holder the ability to head off theindependent development of software programs with thesame or similar functionality. 

In the software industry, where 20 years can cover theentire life cycle of a marketplace, patents take on astrategic weight. Where companies such as Microsoft andApple once battled over copyright and the "look andfeel" of various technologies, today's Internetcompanies use patents as a way to stake out individualapplications and business models, the most notoriousexample being Amazon. Com's 2000 attempt to patent thecompany's "one-click" online shopping process. For mostcompanies, however, software patents have become adefensive tool, with cross-licensing deals balancingone set of corporate patents against another in a tenseform of corporate detente. Still, in a few notablecases of computer encryption and graphic imagingalgorithms, software vendors have successfully stifledrival technologies. 

For Stallman, the software-patent issue dramatizes theneed for eternal hacker vigilance. It also underlinesthe importance of stressing the political benefits offree software programs over the competitive benefits. Pointing to software patents' ability to createsheltered regions in the marketplace, Stallman sayscompetitive performance and price, two areas where freesoftware operating systems such as GNU/Linux andFreeBSD already hold a distinct advantage over theirproprietary counterparts, are red herrings compared tothe large issues of user and developer freedom. 

"It's not because we don't have the talent to makebetter software, " says Stallman. "It's because we don'thave the right. Somebody has prohibited us from servingthe public. So what's going to happen when usersencounter these gaps in free software? Well, if theyhave been persuaded by the open source movement thatthese freedoms are good because they lead tomore-powerful reliable software, they're likely to say, `You didn't deliver what you promised. This software'snot more powerful. It's missing this feature. You liedto me. ' But if they have come to agree with the freesoftware movement, that the freedom is important initself, then they will say, `How dare those people stopme from having this feature and my freedom too. ' Andwith that kind of response, we may survive the hitsthat we're going to take as these patents explode. "

Such comments involve a hefty dose of spin, of course. Most open source advocates are equally, if not more, vociferous as Stallman when it comes to opposingsoftware patents. Still, the underlying logic ofStallman's argument-that open source advocatesemphasize the utilitarian advantages of free softwareover the political advantages-remains uncontested. Rather than stress the political significance of freesoftware programs, open source advocates have chosen tostress the engineering integrity of the hackerdevelopment model. Citing the power of peer review, theopen source argument paints programs such as GNU/Linuxor FreeBSD as better built, better inspected and, byextension, more trushworthy to the average user. 

That's not to say the term "open source" doesn't haveits political implications. For open source advocates, the term open source serves two purposes. First, iteliminates the confusion associated with the word"free, " a word many businesses interpret as meaning"zero cost. " Second, it allows companies to examine thefree software phenomenon on a technological, ratherthan ethical, basis. Eric Raymond, cofounder of theOpen Source Initiative and one of the leading hackersto endorse the term, effectively summed up thefrustration of following Stallman down the politicalpath in a 1999 essay, titled " Shut Up and Show Themthe Code": RMS's rhetoric is very seductive to the kindof people we are. We hackers are thinkers and idealistswho readily resonate with appeals to "principle" and"freedom" and "rights. " Even when we disagree with bitsof his program, we want RMS's rhetorical style to work;we think it ought to work; we tend to be puzzled anddisbelieving when it fails on the 95% of people whoaren't wired like we are. 4 Included among that 95%, Raymond writes, are the bulk of business managers, investors, and nonhacker computer users who, throughsheer weight of numbers, tend to decide the overalldirection of the commercial software marketplace. Without a way to win these people over, Raymond argues, programmers are doomed to pursue their ideology on theperiphery of society: When RMS insists that we talkabout "computer users' rights, " he's issuing adangerously attractive invitation to us to repeat oldfailures. It's one we should reject-not because hisprinciples are wrong, but because that kind oflanguage, applied to software, simply does not persuadeanybody but us. In fact, it confuses and repels mostpeople outside our culture. 4 Watching Stallman deliverhis political message in person, it is hard to seeanything confusing or repellent. Stallman's appearancemay seem off-putting, but his message is logical. Whenan audience member asks if, in shunning proprietarysoftware, free software proponents lose the ability tokeep up with the latest technological advancements, Stallman answers the question in terms of his ownpersonal beliefs. "I think that freedom is moreimportant than mere technical advance, " he says. "Iwould always choose a less advanced free program ratherthan a more advanced nonfree program, because I won'tgive up my freedom for something like that. My rule is, if I can't share it with you, I won't take it. "

Such answers, however, reinforce the quasi-religiousnature of the Stallman message. Like a Jew keepingkosher or a Mormon refusing to drink alcohol, Stallmanpaints his decision to use free software in the placeof proprietary in the color of tradition and personalbelief. As software evangelists go, Stallman avoidsforcing those beliefs down listeners' throats. Thenagain, a listener rarely leaves a Stallman speech notknowing where the true path to software righteousness lies. 

As if to drive home this message, Stallman punctuateshis speech with an unusual ritual. Pulling a black robeout of a plastic grocery bag, Stallman puts it on. Outof a second bag, he pulls a reflective yellow computerdisk and places it on his head. The crowd lets out astartled laugh. 

"I am St. Ignucius of the Church of Emacs, " saysStallman, raising his right hand in mock-blessing. "Ibless your computer, my child. "

<Graphic file:books/free_0801. Png>

Stallman dressed as St. Ignucius. Photo by Wouter vanOortmerssen. 

The laughter turns into full-blown applause after a fewseconds. As audience members clap, the computer disk onStallman's head catches the glare of an overhead light, eliciting a perfect halo effect. In the blink of aneye, Stallman goes from awkward haole to Russianreligious icon. 

" Emacs was initially a text editor, " says Stallman, explaining the getup. "Eventually it became a way oflife for many and a religion for some. We call thisreligion the Church of Emacs. "

The skit is a lighthearted moment of self-pardoy, ahumorous return-jab at the many people who might seeStallman's form of software asceticism as religiousfanaticism in disguise. It is also the sound of theother shoe dropping-loudly. It's as if, in donning hisrobe and halo, Stallman is finally letting listeners ofthe hook, saying, "It's OK to laugh. I know I'm weird. "

Discussing the St. Ignucius persona afterward, Stallmansays he first came up with it in 1996, long after thecreation of Emacs but well before the emergence of the"open source" term and the struggle forhacker-community leadership that precipitated it. Atthe time, Stallman says, he wanted a way to "poke funat himself, " to remind listeners that, though stubborn, Stallman was not the fanatic some made him out to be. It was only later, Stallman adds, that others seizedthe persona as a convenient way to play up hisreputation as software ideologue, as Eric Raymond didin an 1999 interview with the linux. Com web site: WhenI say RMS calibrates what he does, I'm not belittlingor accusing him of insincerity. I'm saying that likeall good communicators he's got a theatrical streak. Sometimes it's conscious-have you ever seen him in hisSt. Ignucius drag, blessing software with a diskplatter on his head? Mostly it's unconscious; he's justlearned the degree of irritating stimulus that works, that holds attention without (usually) freaking people out. See "GuestInterview: Eric S. Raymond, " Linux. Com (May18, 1999). Http://www. Linux. Com/interviews/19990518/8/ Stallman takes issue with the Raymond analysis. "It'ssimply my way of making fun of myself, " he says. "Thefact that others see it as anything more than that is areflection of their agenda, not mine. "

That said, Stallman does admit to being a ham. "Are youkidding?" he says at one point. "I love being thecenter of attention. " To facilitate that process, Stallman says he once enrolled in Toastmasters, anorganization that helps members bolster theirpublic-speaking skills and one Stallman recommendshighly to others. He possesses a stage presence thatwould be the envy of most theatrical performers andfeels a link to vaudevillians of years past. A few daysafter the Maui High Performance Computing Centerspeech, I allude to the 1999 LinuxWorld performace andask Stallman if he has a Groucho Marx complex-i. E. , theunwillingness to belong to any club that would have himas a member. Stallman's response is immediate: "No, butI admire Groucho Marx in a lot of ways and certainlyhave been in some things I say inspired by him. Butthen I've also been inspired in some ways by Harpo. "

The Groucho Marx influence is certainly evident inStallman's lifelong fondness for punning. Then again, punning and wordplay are common hacker traits. Perhapsthe most Groucho-like aspect of Stallman's personality, however, is the deadpan manner in which the puns aredelivered. Most come so stealthily-without even thehint of a raised eyebrow or upturned smile-you almosthave to wonder if Stallman's laughing at his audiencemore than the audience is laughing at him. 

Watching members of the Maui High Performance ComputerCenter laugh at the St. Ignucius parody, such concernsevaporate. While not exactly a standup act, Stallmancertainly possesses the chops to keep a roomful ofengineers in stitches. "To be a saint in the Church ofEmacs does not require celibacy, but it does requiremaking a commitment to living a life of moral purity, "he tells the Maui audience. "You must exorcise the evilproprietary operating system from all your computer andthen install a wholly [holy] free operating system. Andthen you must install only free software on top ofthat. If you make this commitment and live by it, thenyou too will be a saint in the Church of Emacs, and youtoo may have a halo. "

The St. Ignucius skit ends with a brief inside joke. Onmost Unix systems and Unix-related offshoots, theprimary competitor program to Emacs is vi, atext-editing program developed by former UC Berkeleystudent and current Sun Microsystems chief scientist, Bill Joy. Before doffing his "halo, " Stallman pokes funat the rival program. "People sometimes ask me if it isa sin in the Church of Emacs to use vi, " he says. "Using a free version of vi is not a sin; it is apenance. So happy hacking. "

After a brief question-and-answer session, audiencemembers gather around Stallman. A few ask forautographs. "I'll sign this, " says Stallman, holding upone woman's print out of the GNU General PublicLicense, "but only if you promise me to use the termGNU/Linux instead of Linux and tell all your friends todo likewise. "

The comment merely confirms a private observation. Unlike other stage performers and political figures, Stallman has no "off" mode. Aside from the St. Ignuciuscharacter, the ideologue you see onstage is theideologue you meet backstage. Later that evening, during a dinner conversation in which a programmermentions his affinity for "open source" programs, Stallman, between bites, upbraids his tablemate: "Youmean free software. That's the proper way to refer to it. "

During the question-and-answer session, Stallman admitsto playing the pedagogue at times. "There are manypeople who say, `Well, first let's invite people tojoin the community, and then let's teach them aboutfreedom. ' And that could be a reasonable strategy, butwhat we have is almost everybody's inviting people tojoin the community, and hardly anybody's teaching themabout freedom once they come in. "

The result, Stallman says, is something akin to athird-world city. People move in, hoping to strike itrich or at the very least to take part in a vibrant, open culture, and yet those who hold the true powerkeep evolving new tricks and strategies-i. E. , softwarepatents-to keep the masses out. "You have millions ofpeople moving in and building shantytowns, but nobody'sworking on step two: getting them out of thoseshantytowns. If you think talking about softwarefreedom is a good strategy, please join in doing steptwo. There are plenty working on step one. We need morepeople working on step two. "

Working on "step two" means driving home the issue thatfreedom, not acceptance, is the root issue of the freesoftware movement. Those who hope to reform theproprietary software industry from the inside are on afool's errand. "Change from the inside is risky, "Stallman stays. "Unless you're working at the level ofa Gorbachev, you're going to be neutralized. "

Hands pop up. Stallman points to a member of the golfshirt-wearing contingent. "Without patents, how wouldyou suggest dealing with commercial espionage?"

"Well, those two questions have nothing to do with eachother, really, " says Stallman. 

"But I mean if someone wants to steal another company'spiece of software. "

Stallman's recoils as if hit by a poisonous spray. "Wait a second, " Stallman says. "Steal? I'm sorry, there's so much prejudice in that statement that theonly thing I can say is that I reject that prejudice. Companies that develop nonfree software and otherthings keep lots and lots of trade secrets, and sothat's not really likely to change. In the olddays-even in the 1980s-for the most part programmerswere not aware that there were even software patentsand were paying no attention to them. What happened wasthat people published the interesting ideas, and ifthey were not in the free software movement, they keptsecret the little details. And now they patent thosebroad ideas and keep secret the little details. So asfar as what you're describing, patents really make nodifference to it one way or another. "

"But if it doesn't affect their publication, " a newaudience member jumps in, his voice trailing off almostas soon as he starts speaking. 

"But it does, " Stallman says. "Their publication istelling you that this is an idea that's off limits tothe rest of the community for 20 years. And what thehell good is that? Besides, they've written it in sucha hard way to read, both to obfuscate the idea and tomake the patent as broad as possible, that it'sbasically useless looking at the published informationto learn anything anyway. The only reason to look atpatents is to see the bad news of what you can't do. "

The audience falls silent. The speech, which began at3:15, is now nearing the 5:00 whistle, and mostlisteners are already squirming in their seats, antsyto get a jump start on the weekend. Sensing thefatigue, Stallman glances around the room and hastilyshuts things down. "So it looks like we're done, " hesays, following the observation with an auctioneer's"going, going, gone" to flush out any last-minutequestioners. When nobody throws their hand up, Stallmansigns off with a traditional exit line. 

"Happy hacking, " he says. Endnotes

1. See "Grateful Dead Time Capsule: 1985-1995 NorthAmerican Tour Grosses. "http://www. Accessplace. Com/gdtc/1197. Htm 2. See EvanLeibovitch, "Who's Afraid of Big Bad Wolves, " ZDNetTech Update (December 15, 2000). 

http://techupdate. Zdnet. Com/techupdate/stories/main/0Y/A>3. For narrative purposes, I have hesitated to goin-depth when describing Stallman's full definition ofsoftware "freedom. " The GNU Project web site lists fourfundamental components: The freedom to run a program, for any purpose (freedom 0). The freedom to study how aprogram works, and adapt it to your needs (freedom 1). The freedom to redistribute copies of a program so youcan help your neighbor (freedom 2). The freedom toimprove the program, and release your improvements tothe public, so that the whole community benefits(freedom 3). For more information, please visit "TheFree Software Definition" athttp://www. Gnu. Org/philosophy/free-sw. Html. 4. See EricRaymond, "Shut Up and Show Them the Code, " onlineessay, (June 28, 1999). 5. See "Guest Interview: EricS. Raymond, " Linux. Com (May 18, 1999). Http://www. Linux. Com/interviews/19990518/8/

The GNU General Public License

By the spring of 1985, Richard Stallman had settled onthe GNU Project's first milestone-a Lisp-based freesoftware version of Emacs. To meet this goal, however, he faced two challenges. First, he had to rebuild Emacsin a way that made it platform independent. Second, hehad to rebuild the Emacs Commune in a similar fashion. 

The dispute with UniPress had highlighted a flaw in theEmacs Commune social contract. Where users relied onStallman's expert insight, the Commune's rules held. Inareas where Stallman no longer held the position ofalpha hacker-pre-1984 Unix systems, forexample-individuals and companies were free to maketheir own rules. 

The tension between the freedom to modify and thefreedom to exert authorial privilege had been buildingbefore GOSMACS. The Copyright Act of 1976 hadoverhauled U. S. Copyright law, extending the legalprotection of copyright to software programs. Accordingto Section 102(b) of the Act, individuals and companiesnow possessed the ability to copyright the "expression"of a software program but not the "actual processes ormethods embodied in the program. "See Hal Abelson, Mike Fischer, and JoanneCostello, "Software and Copyright Law, " updated version (1998). Translated, programmers and companies had the abilityto treat software programs like a story or song. Otherprogrammers could take inspiration from the work, butto make a direct copy or nonsatirical derivative, theyfirst had to secure permission from the originalcreator. Although the new law guaranteed that evenprograms without copyright notices carried copyrightprotection, programmers quickly asserted their rights, attaching coypright notices to their software programs. 

At first, Stallman viewed these notices with alarm. Rare was the software program that didn't borrow sourcecode from past programs, and yet, with a single strokeof the president's pen, Congress had given programmersand companies the power to assert individual authorshipover communally built programs. It also injected a doseof formality into what had otherwise been an informalsystem. Even if hackers could demonstrate how a givenprogram's source-code bloodlines stretched back years, if not decades, the resources and money that went intobattling each copyright notice were beyond mosthackers' means. Simply put, disputes that had once beensettled hacker-to-hacker were now settledlawyer-to-lawyer. In such a system, companies, nothackers, held the automatic advantage. 

Proponents of software copyright had theircounter-arguments: without copyright, works mightotherwise slip into the public domain. Putting acopyright notice on a work also served as a statementof quality. Programmers or companies who attached theirname to the copyright attached their reputations aswell. Finally, it was a contract, as well as astatement of ownership. Using copyright as a flexibleform of license, an author could give away certainrights in exchange for certain forms of behavior on thepart of the user. For example, an author could giveaway the right to suppress unauthorized copies just solong as the end user agreed not to create a commercial offshoot. 

It was this last argument that eventually softenedStallman's resistance to software copyright notices. Looking back on the years leading up to the GNUProject, Stallman says he began to sense the beneficialnature of copyright sometime around the release ofEmacs 15. 0, the last significant pre-GNU Projectupgrade of Emacs. "I had seen email messages withcopyright notices plus simple `verbatim copyingpermitted' licenses, " Stallman recalls. "Thosedefinitely were [an] inspiration. "

For Emacs 15, Stallman drafted a copyright that gaveusers the right to make and distribute copies. It alsogave users the right to make modified versions, but notthe right to claim sole ownership of those modifiedversions, as in the case of GOSMACS. 

Although helpful in codifying the social contract ofthe Emacs Commune, the Emacs 15 license remained too"informal" for the purposes of the GNU Project, Stallman says. Soon after starting work on a GNUversion of Emacs, Stallman began consulting with theother members of the Free Software Foundation on how toshore up the license's language. He also consulted withthe attorneys who had helped him set up the FreeSoftware Foundation. 

Mark Fischer, a Boston attorney specializing inintellectual-property law, recalls discussing thelicense with Stallman during this period. "Richard hadvery strong views about how it should work, " Fischersays, "He had two principles. The first was to make thesoftware absolutely as open as possible. The second wasto encourage others to adopt the same licensing practices. "

Encouraging others to adopt the same licensingpractices meant closing off the escape hatch that hadallowed privately owned versions of Emacs to emerge. Toclose that escape hatch, Stallman and his free softwarecolleagues came up with a solution: users would be freeto modify GNU Emacs just so long as they publishedtheir modifications. In addition, the resulting"derivative" works would also have carry the same GNUEmacs License. 

The revolutionary nature of this final condition wouldtake a while to sink in. At the time, Fischer says, hesimply viewed the GNU Emacs License as a simplecontract. It put a price tag on GNU Emacs' use. Insteadof money, Stallman was charging users access to theirown later modifications. That said, Fischer doesremember the contract terms as unique. 

"I think asking other people to accept the price was, if not unique, highly unusual at that time, " he says. 

The GNU Emacs License made its debut when Stallmanfinally released GNU Emacs in 1985. Following therelease, Stallman welcomed input from the generalhacker community on how to improve the license'slanguage. One hacker to take up the offer was futuresoftware activist John Gilmore, then working as aconsultant to Sun Microsystems. As part of hisconsulting work, Gilmore had ported Emacs over toSunOS, the company's in-house version of Unix. In theprocess of doing so, Gilmore had published the changesas per the demands of the GNU Emacs License. Instead ofviewing the license as a liability, Gilmore saw it asclear and concise expression of the hacker ethos. "Upuntil then, most licenses were very informal, " Gilmore recalls. 

As an example of this informality, Gilmore cites acopyright notice for trn, a Unix utility. Written byLarry Wall, future creator of the Perl programminglanguage, patch made it simple for Unix programmers toinsert source-code fixes-" patches" in hackerjargon-into any large program. Recognizing the utilityof this feature, Wall put the following copyrightnotice in the program's accompanying README file:

Copyright (c) 1985, Larry Wall You may copy the trn kitin whole or in part as long as you don't try to makemoney off it, or pretend that you wrote it. See Trn Kit README. Http://www. Za. Debian. Org/doc/trn/trn-readme

Such statements, while reflective of the hacker ethic, also reflected the difficulty of translating the loose, informal nature of that ethic into the rigid, legallanguage of copyright. In writing the GNU EmacsLicense, Stallman had done more than close up theescape hatch that permitted proprietary offshoots. Hehad expressed the hacker ethic in a mannerunderstandable to both lawyer and hacker alike. 

It wasn't long, Gilmore says, before other hackersbegan discussing ways to "port" the GNU Emacs Licenseover to their own programs. Prompted by a conversationon Usenet, Gilmore sent an email to Stallman inNovember, 1986, suggesting modification: You shouldprobably remove "EMACS" from the license and replace itwith "SOFTWARE" or something. Soon, we hope, Emacs willnot be the biggest part of the GNU system, and thelicense applies to all of it. See John Gilmore, quoted from email to author. Gilmore wasn't the onlyperson suggesting a more general approach. By the endof 1986, Stallman himself was at work with GNUProject's next major milestone, a source-code debugger, and was looking for ways to revamp the Emacs license sothat it might apply to both programs. Stallman'ssolution: remove all specific references to Emacs andconvert the license into a generic copyright umbrellafor GNU Project software. The GNU General PublicLicense, GPL for short, was born. 

In fashioning the GPL, Stallman followed the softwareconvention of using decimal numbers to indicateprototype versions and whole numbers to indicate matureversions. Stallman published Version 1. 0 of the GPL in1989 (a project Stallman was developing in 1985), almost a full year after the release of the GNUDebugger, Stallman's second major foray into the realmof Unix programming. The license contained a preamblespelling out its political intentions:

The General Public License is designed to make surethat you have the freedom to give away or sell copiesof free software, that you receive source code or canget it if you want it, that you can change the softwareor use pieces of it in new free programs; and that youknow you can do these things. 

To protect your rights, we need to make restrictionsthat forbid anyone to deny you these rights or to askyou to surrender the rights. These restrictionstranslate to certain responsibilities for you if youdistribute copies of the software, or if you modify it. See Richard Stallman, etal. , "GNU General PublicLicense: Version 1, " (February, 1989). Http://www. Gnu. Org/copyleft/copying-1. 0. Html

In fashioning the GPL, Stallman had been forced to makean additional adjustment to the informal tenets of theold Emacs Commune. Where he had once demanded thatCommune members publish any and all changes, Stallmannow demanded publication only in instances whenprogrammers circulated their derivative versions in thesame public manner as Stallman. In other words, programmers who simply modified Emacs for private useno longer needed to send the source-code changes backto Stallman. In what would become a rare compromise offree software doctrine, Stallman slashed the price tagfor free software. Users could innovate withoutStallman looking over their shoulders just so long asthey didn't bar Stallman and the rest of the hackercommunity from future exchanges of the same program. 

Looking back, Stallman says the GPL compromise wasfueled by his own dissatisfaction with the Big Brotheraspect of the original Emacs Commune social contract. As much as he liked peering into other hackers'systems, the knowledge that some future source-codemaintainer might use that power to ill effect forcedhim to temper the GPL. 

"It was wrong to require people to publish allchanges, " says Stallman. "It was wrong to require themto be sent to one privileged developer. That kind ofcentralization and privilege for one was not consistentwith a society in which all had equal rights. "

As hacks go, the GPL stands as one of Stallman's best. It created a system of communal ownership within thenormally proprietary confines of copyright law. Moreimportantly, it demonstrated the intellectualsimilarity between legal code and software code. Implicit within the GPL's preamble was a profoundmessage: instead of viewing copyright law withsuspicion, hackers should view it as yet another systembegging to be hacked. 

"The GPL developed much like any piece of free softwarewith a large community discussing its structure, itsrespect or the opposite in their observation, needs fortweaking and even to compromise it mildly for greateracceptance, " says Jerry Cohen, another attorney whohelped Stallman with the creation of the license. "Theprocess worked very well and GPL in its severalversions has gone from widespread skeptical and attimes hostile response to widespread acceptance. "

In a 1986 interview with Byte magazine, Stallman summedup the GPL in colorful terms. In addition toproclaiming hacker values, Stallman said, readersshould also "see it as a form of intellectual jujitsu, using the legal system that software hoarders have setup against them. "See David Betz and Jon Edwards, "Richard Stallmandiscusses his public-domain [sic] Unix-compatiblesoftware system with BYTE editors, " BYTE (July, 1996). (Reprinted on the GNU Project web site:http://www. Gnu. Org/gnu/byte-interview. Html. ) Thisinterview offers an interesting, not to mention candid, glimpse at Stallman's political attitudes during theearliest days of the GNU Project. It is also helpful intracing the evolution of Stallman's rhetoric. Describing the purpose of the GPL, Stallman says, "I'mtrying to change the way people approach knowledge andinformation in general. I think that to try to ownknowledge, to try to control whether people are allowedto use it, or to try to stop other people from sharingit, is sabotage. " Contrast this with a statement to theauthor in August 2000: "I urge you not to use the term`intellectual property' in your thinking. It will leadyou to misunderstand things, because that termgeneralizes about copyrights, patents, and trademarks. And those things are so different in their effects thatit is entirely foolish to try to talk about them atonce. If you hear somebody saying something aboutintellectual property, without quotes, then he's notthinking very clearly and you shouldn't join. "Years later, Stallman would describe the GPL's creationin less hostile terms. "I was thinking about issuesthat were in a sense ethical and in a sense politicaland in a sense legal, " he says. "I had to try to dowhat could be sustained by the legal system that we'rein. In spirit the job was that of legislating the basisfor a new society, but since I wasn't a government, Icouldn't actually change any laws. I had to try to dothis by building on top of the existing legal system, which had not been designed for anything like this. "

About the time Stallman was pondering the ethical, political, and legal issues associated with freesoftware, a California hacker named Don Hopkins mailedhim a manual for the 68000 microprocessor. Hopkins, aUnix hacker and fellow science-fiction buff, hadborrowed the manual from Stallman a while earlier. As adisplay of gratitude, Hopkins decorated the returnenvelope with a number of stickers obtained at a localscience-fiction convention. One sticker in particularcaught Stallman's eye. It read, "Copyleft (L), AllRights Reversed. " Following the release of the firstversion of GPL, Stallman paid tribute to the sticker, nicknaming the free software license "Copyleft. " Overtime, the nickname and its shorthand symbol, abackwards "C, " would become an official Free SoftwareFoundation synonym for the GPL. 

The German sociologist Max Weber once proposed that allgreat religions are built upon the "routinization" or"institutionalization" of charisma. Every successfulreligion, Weber argued, converts the charisma ormessage of the original religious leader into a social, political, and ethical apparatus more easilytranslatable across cultures and time. 

While not religious per se, the GNU GPL certainlyqualifies as an interesting example of this"routinization" process at work in the modern, decentralized world of software development. Since itsunveiling, programmers and companies who have otherwiseexpressed little loyalty or allegiance to Stallman havewillingly accepted the GPL bargain at face value. A fewhave even accepted the GPL as a preemptive protectivemechanism for their own software programs. Even thosewho reject the GPL contract as too compulsory, stillcredit it as influential. 

One hacker falling into this latter group was KeithBostic, a University of California employee at the timeof the GPL 1. 0 release. Bostic's department, theComputer Systems Research Group (SRG), had beeninvolved in Unix development since the late 1970s andwas responsible for many key parts of Unix, includingthe TCP/IP networking protocol, the cornerstone ofmodern Internet communications. By the late 1980s, AT&T, the original owner of the Unix brand name, beganto focus on commercializing Unix and began looking tothe Berkeley Software Distribution, or BSD, theacademic version of Unix developed by Bostic and hisBerkeley peers, as a key source of commercial technology. 

Although the Berkeley BSD source code was shared amongresearchers and commercial programmers with asource-code license, this commercialization presented aproblem. The Berkeley code was intermixed withproprietary AT&T code. As a result, Berkeleydistributions were available only to institutions thatalready had a Unix source license from AT&T. As AT&Traised its license fees, this arrangement, which had atfirst seemed innocuous, became increasingly burdensome. 

Hired in 1986, Bostic had taken on the personal projectof porting BSD over to the Digital EquipmentCorporation's PDP-11 computer. It was during thisperiod, Bostic says, that he came into closeinteraction with Stallman during Stallman's occasionalforays out to the west coast. "I remember vividlyarguing copyright with Stallman while he sat atborrowed workstations at CSRG, " says Bostic. "We'd goto dinner afterward and continue arguing aboutcopyright over dinner. "

The arguments eventually took hold, although not in theway Stallman would have liked. In June, 1989, Berkeleyseparated its networking code from the rest of theAT&T-owned operating system and distributed it under aUniversity of California license. The contract termswere liberal. All a licensee had to do was give creditto the university in advertisements touting derivative programs. The Universityof California's "obnoxious advertisingclause" would later prove to be a problem. Looking fora less restrictive alternative to the GPL, some hackersused the University of California, replacing"University of California" with the name of their owninstution. The result: free software programs thatborrowed from dozens of other programs would have tocite dozens of institutions in advertisements. In 1999, after a decade of lobbying on Stallman's part, theUniversity of California agreed to drop this clause. In contrast to the GPL, proprietary offshoots werepermissible. Only one problem hampered the license'srapid adoption: the BSD Networking release wasn't acomplete operating system. People could study the code, but it could only be run in conjunction with otherproprietary-licensed code. 

Over the next few years, Bostic and other University ofCalifornia employees worked to replace the missingcomponents and turn BSD into a complete, freelyredistributable operating system. Although delayed by alegal challenge from Unix Systems Laboratories-the AT&Tspin-off that retained ownership of the Unix brandname-the effort would finally bear fruit in the early1990s. Even before then, however, many of the Berkeleyutilities would make their way into Stallman's GNU Project. 

"I think it's highly unlikely that we ever would havegone as strongly as we did without the GNU influence, "says Bostic, looking back. "It was clearly somethingwhere they were pushing hard and we liked the idea. "

By the end of the 1980s, the GPL was beginning to exerta gravitational effect on the free software community. A program didn't have to carry the GPL to qualify asfree software-witness the case of the BSD utilities-butputting a program under the GPL sent a definitemessage. "I think the very existence of the GPLinspired people to think through whether they weremaking free software, and how they would license it, "says Bruce Perens, creator of Electric Fence, a popularUnix utility, and future leader of the Debian GNU/Linuxdevelopment team. A few years after the release of theGPL, Perens says he decided to discard Electric Fence'shomegrown license in favor of Stallman's lawyer-vettedcopyright. "It was actually pretty easy to do, " Perens recalls. 

Rich Morin, the programmer who had viewed Stallman'sinitial GNU announcement with a degree of skepticism, recalls being impressed by the software that began togather under the GPL umbrella. As the leader of a SunOSuser group, one of Morin's primary duties during the1980s had been to send out distribution tapescontaining the best freeware or free softwareutilities. The job often mandated calling up originalprogram authors to verify whether their programs werecopyright protected or whether they had been consignedto the public domain. Around 1989, Morin says, he beganto notice that the best software programs typicallyfell under the GPL license. "As a software distributor, as soon as I saw the word GPL, I knew I was home free, "recalls Morin. 

To compensate for the prior hassles that went intocompiling distribution tapes to the Sun User Group, Morin had charged recipients a convenience fee. Now, with programs moving over to the GPL, Morin wassuddenly getting his tapes put together in half thetime, turning a tidy profit in the process. Sensing acommercial opportunity, Morin rechristened his hobby asa business: Prime Time Freeware. 

Such commercial exploitation was completely within theconfines of the free software agenda. "When we speak offree software, we are referring to freedom, not price, "advised Stallman in the GPL's preamble. By the late1980s, Stallman had refined it to a more simplemnemonic: "Don't think free as in free beer; think freeas in free speech. "

For the most part, businesses ignored Stallman'sentreaties. Still, for a few entrepreneurs, the freedomassociated with free software was the same freedomassociated with free markets. Take software ownershipout of the commercial equation, and you had a situationwhere even the smallest software company was free tocompete against the IBMs and DECs of the world. 

One of the first entrepreneurs to grasp this conceptwas Michael Tiemann, a software programmer and graduatestudent at Stanford University. During the 1980s, Tiemann had followed the GNU Project like an aspiringjazz musician following a favorite artist. It wasn'tuntil the release of the GNU C Compiler in 1987, however, that he began to grasp the full potential offree software. Dubbing GCC a "bombshell, " Tiemann saysthe program's own existence underlined Stallman'sdetermination as a programmer. 

"Just as every writer dreams of writing the greatAmerican novel, every programmer back in the 1980stalked about writing the great American compiler, "Tiemman recalls. "Suddenly Stallman had done it. It wasvery humbling. "

"You talk about single points of failure, GCC was it, "echoes Bostic. "Nobody had a compiler back then, untilGCC came along. "

Rather than compete with Stallman, Tiemann decided tobuild on top of his work. The original version of GCCweighed in at 110, 000 lines of code, but Tiemannrecalls the program as surprisingly easy to understand. So easy in fact that Tiemann says it took less thanfive days to master and another week to port thesoftware to a new hardware platform, NationalSemiconductor's 32032 microchip. Over the next year, Tiemann began playing around with the source code, creating a native compiler for the C+ programminglanguage. One day, while delivering a lecture on theprogram at Bell Labs, Tiemann ran into some AT&Tdevelopers struggling to pull off the same thing. 

"There were about 40 or 50 people in the room, and Iasked how many people were working on the native codecompiler, " Tiemann recalls. "My host said theinformation was confidential but added that if I took alook around the room I might get a good general idea. "

It wasn't long after, Tiemann says, that the light bulbwent off in his head. "I had been working on thatproject for six months, " Tiemann says. I just thoughtto myself, whether it's me or the code this is a levelof efficiency that the free market should be ready to reward. "

Tiemann found added inspiration in the GNU Manifesto, which, while excoriating the greed of some softwarevendors, encourages other vendors to consider theadvantages of free software from a consumer point ofview. By removing the power of monopoly from thecommerical software question, the GPL makes it possiblefor the smartest vendors to compete on the basis ofservice and consulting, the two most profit-richcorners of the software marketplace. 

In a 1999 essay, Tiemann recalls the impact ofStallman's Manifesto. "It read like a socialistpolemic, but I saw something different. I saw abusiness plan in disguise. "7. See Michael Tiemann, "Future of Cygnus Solutions:AnEntrepreneur's Account, " Open Sources (O'Reilly &Associates, Inc. , 1999): 139. 

Teaming up with John Gilmore, another GNU Project fan, Tiemann launched a software consulting servicededicated to customizing GNU programs. Dubbed CygnusSupport, the company signed its first developmentcontract in February, 1990. By the end of the year, thecompany had $725, 000 worth of support and development contracts. 

GNU Emacs, GDB, and GCC were the "big three" ofdeveloper-oriented tools, but they weren't the onlyones developed by Stallman during the GNU Project'sfirst half decade. By 1990, Stallman had also generatedGNU versions of the Bourne Shell (rechristened theBourne Again Shell, or BASH), YACC (rechristenedBison), and awk (rechristened gawk). Like GCC, everyGNU program had to be designed to run on multiplesystems, not just a single vendor's platform. In theprocess of making programs more flexible, Stallman andhis collaborators often made them more useful as well. 

Recalling the GNU universalist approach, Prime TimeFreeware's Morin points to a critical, albeit mundane, software package called hello. "It's the hello worldprogram which is five lines of C, packaged up as if itwere a GNU distribution, " Morin says. "And so it's gotthe Texinfo stuff and the configure stuff. It's got allthe other software engineering goo that the GNU Projecthas come up with to allow packages to port to all thesedifferent environments smoothly. That's tremendouslyimportant work, and it affects not only all of[Stallman's] software, but also all of the other GNUProject software. "

According to Stallman, improving software programs wassecondary to building them in the first place. "Witheach piece I may or may not find a way to improve it, "said Stallman to Byte. "To some extent I am getting thebenefit of reimplementation, which makes many systemsmuch better. To some extent it's because I have been inthe field a long time and worked on many other systems. I therefore have many ideas to bring to bear. "See Richard Stallman, BYTE(1986). 

Nevertheless, as GNU tools made their mark in the late1980s, Stallman's AI Lab-honed reputation for designfastidiousness soon became legendary throughout theentire software-development community. 

Jeremy Allison, a Sun user during the late 1980s andprogrammer destined to run his own free softwareproject, Samba, in the 1990s, recalls that reputationwith a laugh. During the late 1980s, Allison beganusing Emacs. Inspired by the program'scommunity-development model, Allison says he sent in asnippet of source code only to have it rejected by Stallman. 

"It was like the Onion headline, " Allison says. "`Child's prayers to God answered: No. '"

Stallman's growing stature as a software programmer, however, was balanced by his struggles as a projectmanager. Although the GNU Project moved from success tosuccess in creation of developer-oriented tools, itsinability to generate a working kernel-the central"traffic cop" program in all Unix systems thatdetermines which devices and applications get access tothe microprocessor and when-was starting to elicitgrumbles as the 1980s came to a close. As with most GNUProject efforts, Stallman had started kerneldevelopment by looking for an existing program tomodify. According to a January 1987 "Gnusletter, "Stallman was already working to overhaul TRIX, a Unixkernel developed at MIT. 

A review of GNU Project "GNUsletters" of the late 1980sreflects the management tension. In January, 1987, Stallman announced to the world that the GNU Projectwas working to overhaul TRIX, a Unix kernel developedat MIT. A year later, in February of 1988, the GNUProject announced that it had shifted its attentions toMach, a lightweight "micro-kernel" developed atCarnegie Mellon. All told, however, official GNUProject kernel development wouldn't commence until 1990. See "HURD History. "http://www. Gnu. Org/software/hurd/history. Html

The delays in kernel development were just one of manyconcerns weighing on Stallman during this period. In1989, Lotus Development Corporation filed suit againstrival software company, Paperback SoftwareInternational, for copying menu commands in Lotus'popular 1-2-3 Spreadsheet program. Lotus' suit, coupledwith the Apple -Microsoft "look and feel" battle, provided a troublesome backdrop for the GNU Project. Although both suits fell outside the scope of the GNUProject, both revolved around operating systems andsoftware applications developed for the personalcomputer, not Unix-compatible hardware systems-theythreatened to impose a chilling effect on the entireculture of software development. Determined to dosomething, Stallman recruited a few programmer friendsand composed a magazine ad blasting the lawsuits. Hethen followed up the ad by helping to organize a groupto protest the corporations filing the suit. Callingitself the League of Programming Freedom, the groupheld protests outside the offices of Lotus, Inc. Andthe Boston courtroom hosting the Lotus trial. 

The protests were notable. According to a League of Programming Freedom Press, theprotests were notable for featuring the firsthexadecimal protest chant: 1-2-3-4, toss the lawyersout the door; 5-6-7-8, innovate don't litigate;9-A-B-C, 1-2-3 is not for me; D-E-F-O, look and feelhave got to go

http://lpf. Ai. Mit. Edu/Links/prep. Ai. Mit. Edu/demo. Final. Release They document the evolving nature of softwareindustry. Applications had quietly replaced operatingsystems as the primary corporate battleground. In itsunfulfilled quest to build a free software operatingsystem, the GNU Project seemed hopelessly behind thetimes. Indeed, the very fact that Stallman had felt itnecessary to put together an entirely new groupdedicated to battling the "look and feel" lawsuitsreinforced that obsolescence in the eyes of some observers. 

In 1990, the John D. And Catherine T. MacArthurFoundation cerified Stallman's genius status when itgranted Stallman a MacArthur fellowship, thereforemaking him a recipient for the organization's so-called"genius grant. " The grant, a $240, 000 reward forlaunching the GNU Project and giving voice to the freesoftware philosophy, relieved a number of short-termconcerns. First and foremost, it gave Stallman, anonsalaried employee of the FSF who had been supportinghimself through consulting contracts, the ability todevote more time to writing GNU code. I use the term "writing" here loosely. About the timeof the MacArthur award, Stallman began sufferingchronic pain in his hands and was dictating his work toFSF-employed typists. Although some have speculatedthat the hand pain was the result of repetitive stressinjury, or RSI, an injury common among softwareprogrammers, Stallman is not 100% sure. "It was NOTcarpal tunnel syndrome, " he writes. "My hand problemwas in the hands themselves, not in the wrists. "Stallman has since learned to work without typistsafter switching to a keyboard with a lighter touch. 

Ironically, the award also made it possible forStallman to vote. Months before the award, a fire inStallman's apartment house had consumed his few earthlypossessions. By the time of the award, Stallman waslisting himself as a "squatter"See Reuven Lerner, "Stallman wins $240, 000MacArthuraward, " MIT, The Tech (July 18, 1990). Http://the-tech. Mit. Edu/V110/N30/rms. 30n. Html at 545 Technology Square. "[The registrar of voters]didn't want to accept that as my address, " Stallmanwould later recall. "A newspaper article about theMacArthur grant said that and then they let me register. "See Michael Gross, "Richard Stallman: High SchoolMisfit, Symbol of Free Software, MacArthur-certifiedGenius" (1999). 

Most importantly, the MacArthur money gave Stallmanmore freedom. Already dedicated to the issue ofsoftware freedom, Stallman chose to use the additionalfreedom to increase his travels in support of the GNUProject mission. 

Interestingly, the ultimate success of the GNU Projectand the free software movement in general would stemfrom one of these trips. In 1990, Stallman paid a visitto the Polytechnic University in Helsinki, Finland. Among the audience members was 21-year-old LinusTorvalds, future developer of the Linux kernel-the freesoftware kernel destined to fill the GNU Project's mostsizable gap. 

A student at the nearby University of Helsinki at thetime, Torvalds regarded Stallman with bemusement. "Isaw, for the first time in my life, the stereotypicallong-haired, bearded hacker type, " recalls Torvalds inhis 2001 autobiography Just for Fun. "We don't havemuch of them in Helsinki. "See Linus Torvalds and David Diamond, Just For Fun:TheStory of an Accidentaly Revolutionary (HarperCollinsPublishers, Inc. , 2001): 58-59. 

While not exactly attuned to the "sociopolitical" sideof the Stallman agenda, Torvalds neverthelessappreciated the agenda's underlying logic: noprogrammer writes error-free code. By sharing software, hackers put a program's improvement ahead of individualmotivations such as greed or ego protection. 

Like many programmers of his generation, Torvalds hadcut his teeth not on mainframe computers like the IBM7094, but on a motley assortment of home-built computersystems. As university student, Torvalds had made thestep up from C programming to Unix, using theuniversity's MicroVAX. This ladder-like progression hadgiven Torvalds a different perspective on the barriersto machine access. For Stallman, the chief barrierswere bureaucracy and privilege. For Torvalds, the chiefbarriers were geography and the harsh Helsinki winter. Forced to trek across the University of Helsinki justto log in to his Unix account, Torvalds quickly beganlooking for a way to log in from the warm confines ofhis off-campus apartment. 

The search led Torvalds to the operating system Minix, a lightweight version of Unix developed forinstructional purposes by Dutch university professorAndrew Tanenbaum. The program fit within the memoryconfines of a 386 PC, the most powerful machineTorvalds could afford, but still lacked a few necessaryfeatures. It most notably lacked terminal emulation, the feature that allowed Torvalds' machine to mimic auniversity terminal, making it possible to log in tothe MicroVAX from home. 

During the summer of 1991, Torvalds rewrote Minix fromthe ground up, adding other features as he did so. Bythe end of the summer, Torvalds was referring to hisevolving work as the "GNU/Emacs of terminal emulation programs. "See LinusTorvalds and David Diamond, Just For Fun: TheStory of an Accidentaly Revolutionary (HarperCollinsPublishers, Inc. , 2001): 78. Feeling confident, he solicited a Minix newsgroup forcopies of the POSIX standards, the software blue printsthat determined whether a program was Unix compatible. A few weeks later, Torvalds was posting a messageeerily 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) ATclones). This has been brewing since April, and isstarting to get ready. I'd like any feedback on thingspeople like/dislike in minix, as my OS resembles itsomewhat (same physical layout of the file-system (dueto practical reasons) among other things). See "Linux 10th Anniversary. "http://www. Linux10. Org/history/

The posting drew a smattering of responses and within amonth, Torvalds had posted a 0. 01 version of theoperating system-i. E. , the earliest possible versionfit for outside review-on an Internet FTP site. In thecourse of doing so, Torvalds had to come up with a namefor the new system. On his own PC hard drive, Torvaldshad saved the program as Linux, a name that paid itsrespects to the software convention of giving each Unixvariant a name that ended with the letter X. Deemingthe name too "egotistical, " Torvalds changed it toFreax, only to have the FTP site manager change it back. 

Although Torvalds had set out build a full operatingsystem, both he and other developers knew at the timethat most of the functional tools needed to do so werealready available, thanks to the work of GNU, BSD, andother free software developers. One of the first toolsthe Linux development team took advantage of was theGNU C Compiler, a tool that made it possible to processprograms written in the C programming language. 

Integrating GCC improved the performance of Linux. Italso raised issues. Although the GPL's "viral" powersdidn't apply to the Linux kernel, Torvald's willingnessto borrow GCC for the purposes of his own free softwareoperating system indicated a certain obligation to letother users borrow back. As Torvalds would later putit: "I had hoisted myself up on the shoulders of giants. "See Linus Torvalds andDavid Diamond, Just For Fun: TheStory of an Accidentaly Revolutionary (HarperCollinsPublishers, Inc. , 2001): 96-97. Not surprisingly, he began to think about what wouldhappen when other people looked to him for similarsupport. A decade after the decision, Torvalds echoesthe Free Software Foundation's Robert Chassel when hesums up his thoughts at the time: You put six months ofyour life into this thing and you want to make itavailable and you want to get something out of it, butyou don't want people to take advantage of it. I wantedpeople to be able to see [Linux], and to make changesand improvements to their hearts' content. But I alsowanted to make sure that what I got out of it was tosee what they were doing. I wanted to always haveaccess to the sources so that if they madeimprovements, I could make those improvements myself. See Linus Torvalds andDavid Diamond, Just For Fun: TheStory of an Accidentaly Revolutionary (HarperCollinsPublishers, Inc. , 2001): 94-95. 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 freesoftware movement. He discarded the old kernel licenseand replaced it with the GPL. The decision triggered aporting spree, as Torvalds and his collaborators lookedto other GNU programs to fold into the growing Linuxstew. Within three years, Linux developers wereoffering their first production release, Linux 1. 0, including fully modified versions of GCC, GDB, and ahost of BSD tools. 

By 1994, the amalgamated operating system had earnedenough respect in the hacker world to make someobservers wonder if Torvalds hadn't given away the farmby switching to the GPL in the project's initialmonths. In the first issue of Linux Journal, publisherRobert Young sat down with Torvalds for an interview. When Young asked the Finnish programmer if he feltregret at giving up private ownership of the Linuxsource code, Torvalds said no. "Even with 20/20hindsight, " Torvalds said, he considered the GPL "oneof the very best design decisions" made during theearly stages of the Linux project. See Robert Young, "Interview with Linus, theAuthor ofLinux, " Linux Journal (March 1, 1994). Http://www. Linuxjournal. Com/article. Php?sid=2736

That the decision had been made with zero appeal ordeference to Stallman and the Free Software Foundationspeaks to the GPL's growing portability. Although itwould take a few years to be recognized by Stallman, the explosiveness of Linux development conjuredflashbacks of Emacs. This time around, however, theinnovation triggering the explosion wasn't a softwarehack like Control-R but the novelty of running aUnix-like system on the PC architecture. The motivesmay have been different, but the end result certainlyfit the ethical specifications: a fully functionaloperating system composed entirely of free software. 

As his initial email message to the comp. Os. Minixnewsgroup indicates, it would take a few months beforeTorvalds saw Linux as anything less than a holdoveruntil the GNU developers delivered on the HURD kernel. This initial unwillingness to see Linux in politicalterms would represent a major blow to the Free SoftwareFoundation. 

As far as Torvalds was concerned, he was simply thelatest in a long line of kids taking apart andreassembling things just for fun. Nevertheless, whensumming up the runaway success of a project that couldhave just as easily spent the rest of its days on anabandoned computer hard drive, Torvalds credits hisyounger self for having the wisdom to give up controland accept the GPL bargain. 

"I may not have seen the light, " writes Torvalds, reflecting on Stallman's 1991 Polytechnic Universityspeech and his subsequent decision to switch to theGPL. "But I guess something from his speech sunk in . "See Linus Torvalds andDavid Diamond, Just For Fun: TheStory of an Accidentaly Revolutionary (HarperCollinsPublishers, Inc. , 2001): 59. Interview offers an interesting, not to mentioncandid, glimpse at Stallman's political attitudesduring the earliest days of the GNU Project. It is alsohelpful in tracing the evolution of Stallman'srhetoric. Describing the purpose of the GPL, Stallmansays, "I'm trying to change the way people approachknowledge and information in general. I think that totry to own knowledge, to try to control whether peopleare allowed to use it, or to try to stop other peoplefrom sharing it, is sabotage. " Contrast this with astatement to the author in August 2000: "I urge you notto use the term `intellectual property' in yourthinking. It will lead you to misunderstand things, because that term generalizes about copyrights, patents, and trademarks. And those things are sodifferent in their effects that it is entirely foolishto try to talk about them at once. If you hear somebodysaying something about intellectual property, withoutquotes, then he's not thinking very clearly and youshouldn't join. "

GNU/Linux

By 1993, the free software movement was at acrossroads. To the optimistically inclined, all signspointed toward success for the hacker cultur. Wiredmagazine, a funky, new publication offering stories ondata encryption, Usenet, and software freedom, wasflying off magazine racks. The Internet, once a slangterm used only by hackers and research scientists, hadfound its way into mainstream lexicon. Even PresidentClinton was using it. The personal computer, once ahobbyist's toy, had grown to full-scale respectability, giving a whole new generation of computer users accessto hacker-built software. And while the GNU Project hadnot yet reached its goal of a fully intact, freesoftware operating system, curious users could stilltry Linux in the interim. 

Any way you sliced it, the news was good, or so itseemed. After a decade of struggle, hackers and hackervalues were finally gaining acceptance in mainstreamsociety. People were getting it. 

Or were they? To the pessimistically inclined, eachsign of acceptance carried its own troublingcountersign. Sure, being a hacker was suddenly cool, but was cool good for a community that thrived onalienation? Sure, the White House was saying all theright things about the Internet, even going so far asto register its own domain name, whitehouse. Gov, but itwas also meeting with the companies, censorshipadvocates, and law-enforcement officials looking totame the Internet's Wild West culture. Sure, PCs weremore powerful, but in commoditizing the PC marketplacewith its chips, Intel had created a situation in whichproprietary software vendors now held the power. Forevery new user won over to the free software cause viaLinux, hundreds, perhaps thousands, were booting upMicrosoft Windows for the first time. 

Finally, there was the curious nature of Linux itself. Unrestricted by design bugs (like GNU) and legaldisputes (like BSD), Linux' high-speed evolution hadbeen so unplanned, its success so accidental, thatprogrammers closest to the software code itself didn'tknow what to make of it. More compilation album thanoperating system, it was comprised of a hacker medleyof greatest hits: everything from GCC, GDB, and glibc(the GNU Project's newly developed C Library) to X (aUnix-based graphic user interface developed by MIT'sLaboratory for Computer Science) to BSD-developed toolssuch as BIND (the Berkeley Internet Naming Daemon, which lets users substitute easy-to-remember Internetdomain names for numeric IP addresses) and TCP/IP. Thearch's capstone, of course, was the Linux kernel-itselfa bored-out, super-charged version of Minix. Ratherthan building their operating system from scratch, Torvalds and his rapidly expanding Linux developmentteam had followed the old Picasso adage, "good artistsborrow; great artists steal. " Or as Torvalds himselfwould later translate it when describing the secret ofhis success: "I'm basically a very lazy person wholikes to take credit for things other people actually do. "Torvalds has offeredthis quote in many differentsettings. To date, however, the quote's most notableappearance is in the Eric Raymond essay, "The Cathedraland the Bazaar" (May, 1997). 

http://www. Tuxedo. Org/~esr/writings/cathedral-bazaar/cathedral-bazaar/index. Html

Such laziness, while admirable from an efficiencyperspective, was troubling from a politicalperspective. For one thing, it underlined the lack ofan ideological agenda on Torvalds' part. Unlike the GNUdevelopers, Torvalds hadn't built an operating systemout of a desire to give his fellow hackers something towork with; he'd built it to have something he himselfcould play with. Like Tom Sawyer whitewashing a fence, Torvalds' genius lay less in the overall vision andmore in his ability to recruit other hackers to speedthe process. 

That Torvalds and his recruits had succeeded whereothers had not raised its own troubling question: what, exactly, was Linux? Was it a manifestation of the freesoftware philosophy first articulated by Stallman inthe GNU Manifesto? Or was it simply an amalgamation ofnifty software tools that any user, similarlymotivated, could assemble on his own home system?

By late 1993, a growing number of Linux users had begunto lean toward the latter definition and began brewingprivate variations on the Linux theme. They even becamebold enough to bottle and sell their variations-or"distributions"-to fellow Unix aficionados. The resultswere spotty at best. 

"This was back before Red Hat and the other commercialdistributions, " remembers Ian Murdock, then a computerscience student at Purdue University. "You'd flipthrough Unix magazines and find all these businesscard-sized ads proclaiming `Linux. ' Most of thecompanies were fly-by-night operations that saw nothingwrong with slipping a little of their own source codeinto the mix. "

Murdock, a Unix programmer, remembers being "sweptaway" by Linux when he first downloaded and installedit on his home PC system. "It was just a lot of fun, "he says. "It made me want to get involved. " Theexplosion of poorly built distributions began to dampenhis early enthusiasm, however. Deciding that the bestway to get involved was to build a version of Linuxfree of additives, Murdock set about putting a list ofthe best free software tools available with theintention of folding them into his own distribution. "Iwanted something that would live up to the Linux name, "Murdock says. 

In a bid to "stir up some interest, " Murdock posted hisintentions on the Internet, including Usenet'scomp. Os. Linux newsgroup. One of the first respondingemail messages was from rms@ai. Mit. Edu . As a hacker, Murdock instantly recognized the address. It wasRichard M. Stallman, founder of the GNU Project and aman Murdock knew even back then as "the hacker ofhackers. " Seeing the address in his mail queue, Murdockwas puzzled. Why on Earth would Stallman, a personleading his own operating-system project, care aboutMurdock's gripes over Linux?

Murdock opened the message. 

"He said the Free Software Foundation was starting tolook closely at Linux and that the FSF was interestedin possibly doing a Linux system, too. Basically, itlooked to Stallman like our goals were in line withtheir philosophy. "

The message represented a dramatic about-face onStallman's part. Until 1993, Stallman had been contentto keep his nose out of the Linux community's affairs. In fact, he had all but shunned the renegade operatingsystem when it first appeared on the Unix programminglandscape in 1991. After receiving the firstnotification of a Unix-like operating system that ranon PCs, Stallman says he delegated the task ofexamining the new operating system to a friend. RecallsStallman, "He reported back that the software wasmodeled after System V, which was the inferior versionof Unix. He also told me it wasn't portable. "

The friend's report was correct. Built to run on386-based machines, Linux was firmly rooted to itslow-cost hardware platform. What the friend failed toreport, however, was the sizable advantage Linuxenjoyed as the only freely modifiable operating systemin the marketplace. In other words, while Stallmanspent the next three years listening to bug reportsfrom his HURD team, Torvalds was winning over theprogrammers who would later uproot and replant theoperating system onto new platforms. 

By 1993, the GNU Project's inability to deliver aworking kernel was leading to problems both within theGNU Project and within the free software movement atlarge. A March, 1993, a Wired magazine article bySimson Garfinkel described the GNU Project as "boggeddown" despite the success of the project's many tools. See Simson Garfinkel, "IsStallman Stalled?" Wired(March, 1993). Those within the project and its nonprofit adjunct, the Free Software Foundation, remember the mood asbeing even worse than Garfinkel's article let on. "Itwas very clear, at least to me at the time, that therewas a window of opportunity to introduce a newoperating system, " says Chassell. "And once that windowwas closed, people would become less interested. Whichis in fact exactly what happened. "Chassel's concern about there being a36-month "window"for a new operating system is not unique to the GNUProject. During the early 1990s, free software versionsof the Berkeley Software Distribution were held up byUnix System Laboratories' lawsuit restricting therelease of BSD-derived software. While many usersconsider BSD offshoots such as FreeBSD and OpenBSD tobe demonstrably superior to GNU/Linux both in terms ofperformance and security, the number of FreeBSD andOpenBSD users remains a fraction of the total GNU/Linuxuser population. To view a sample analysis of therelative success of GNU/Linux in relation to other freesoftware operating systems, see the essay by NewZealand hacker, Liam Greenwood, "Why is LinuxSuccessful" (1999). 

Much has been made about the GNU Project's strugglesduring the 1990-1993 period. While some place the blameon Stallman for those struggles, Eric Raymond, an earlymember of the GNU Emacs team and later Stallman critic, says the problem was largely institutional. "The FSFgot arrogant, " Raymond says. "They moved away from thegoal of doing a production-ready operating system todoing operating-system research. " Even worse, "Theythought nothing outside the FSF could affect them. "

Murdock, a person less privy to the inner dealings ofthe GNU Project, adopts a more charitable view. "Ithink part of the problem is they were a little tooambitious and they threw good money after bad, " hesays. "Micro-kernels in the late 80s and early 90s werea hot topic. Unfortunately, that was about the timethat the GNU Project started to design their kernel. They ended up with alot of baggage and it would havetaken a lot of backpedaling to lose it. "

Stallman cites a number of issues when explaining thedelay. The Lotus and Apple lawsuits had providedpolitical distractions, which, coupled with Stallman'sinability to type, made it difficult for Stallman tolend a helping hand to the HURD team. Stallman alsocites poor communication between various portions ofthe GNU Project. "We had to do a lot of work to get thedebugging environment to work, " he recalls. "And thepeople maintaining GDB at the time were not thatcooperative. " Mostly, however, Stallman says he and theother members of the GNU Project team underestimatedthe difficulty of expanding the Mach microkernal into afull-fledged Unix kernel. 

"I figured, OK, the [Mach] part that has to talk to themachine has already been debugged, " Stallman says, recalling the HURD team's troubles in a 2000 speech. "With that head start, we should be able to get it donefaster. But instead, it turned out that debugging theseasynchronous multithreaded programs was really hard. There were timing books that would clobber the files, and that's no fun. The end result was that it tookmany, many years to produce a test version. "See Maui High Performance ComputingCenter Speech. 

Whatever the excuse, or excuses, the concurrent successof the Linux-kernel team created a tense situation. Sure, the Linux kernel had been licensed under the GPL, but as Murdock himself had noted, the desire to treatLinux as a purely free software operating system wasfar from uniform. By late 1993, the total Linux userpopulation had grown from a dozen or so Minixenthusiasts to somewhere between 20, 000 and 100, 000. GNU/Linux user-populationnumbers are sketchy at best, which is why I've provided such a broad range. The100, 000 total comes from the Red Hat "Milestones" site, http://www. Redhat. Com/about/corporate/milestones. Html. What had once been a hobby was now a marketplace ripefor exploitation. Like Winston Churchill watchingSoviet troops sweep into Berlin, Stallman felt anunderstandable set of mixed emotions when it came timeto celebrate the Linux "victory. "I wrote this Winston Churchill analogy beforeStallmanhimself sent me his own unsolicited comment onChurchill: World War II and the determination needed towin it was a very strong memory as I was growing up. Statements such as Churchill's, "We will fight them inthe landing zones, we will fight them on the beaches .. . We will never surrender, " have always resonated forme. 

Although late to the party, Stallman still had clout. As soon as the FSF announced that it would lend itsmoney and moral support to Murdock's software project, other offers of support began rolling in. Murdockdubbed the new project Debian-a compression of his andhis wife, Deborah's, names-and within a few weeks wasrolling out the first distribution. "[Richard'ssupport] catapulted Debian almost overnight from thisinteresting little project to something people withinthe community had to pay attention to, " Murdock says. 

In January of 1994, Murdock issued the " DebianManifesto. " Written in the spirit of Stallman's "GNUManifesto" from a decade before, it explained theimportance of working closely with the Free SoftwareFoundation. Murdock wrote: The Free Software Foundationplays an extremely important role in the future ofDebian. By the simple fact that they will bedistributing it, a message is sent to the world thatLinux is not a commercial product and that it nevershould be, but that this does not mean that Linux willnever be able to compete commercially. For those of youwho disagree, I challenge you to rationalize thesuccess of GNU Emacs and GCC, which are not commercialsoftware but which have had quite an impact on thecommercial market regardless of that fact. 

The time has come to concentrate on the future of Linuxrather than on the destructive goal of enrichingoneself at the expense of the entire Linux communityand its future. The development and distribution ofDebian may not be the answer to the problems that Ihave outlined in the Manifesto, but I hope that it willat least attract enough attention to these problems toallow them to be solved. Shortly after the Manifesto'srelease, the Free Software Foundation made its firstmajor request. Stallman wanted Murdock to call itsdistribution "GNU/Linux. " At first, Murdock says, Stallman had wanted to use the term " Lignux"-"as inLinux with GNU at the heart of it"-but a sample testingof the term on Usenet and in various impromptu hackerfocus groups had merited enough catcalls to convinceStallman to go with the less awkward GNU/Linux. 

Although some would dismiss Stallman's attempt to addthe "GNU" prefix as a belated quest for credit, Murdocksaw it differently. Looking back, Murdock saw it as anattempt to counteract the growing tension between GNUProject and Linux-kernel developers. "There was a splitemerging, " Murdock recalls. "Richard was concerned. "

The deepest split, Murdock says, was over glibc. Shortfor GNU C Library, glibc is the package that letsprogrammers make "system calls" directed at the kernel. Over the course of 1993-1994, glibc emerged as atroublesome bottleneck in Linux development. Because somany new users were adding new functions to the Linuxkernel, the GNU Project's glibc maintainers were soonoverwhelmed with suggested changes. Frustrated bydelays and the GNU Project's growing reputation forfoot-dragging, some Linux developers suggested creatinga " fork"-i. E. , a Linux-specific C Library parallel to glibc. 

In the hacker world, forks are an interestingphenomenon. Although the hacker ethic permits aprogrammer to do anything he wants with a givenprogram's source code, most hackers prefer to pourtheir innovations into a central source-code file or "tree" to ensure compatibility with other people'sprograms. To fork glibc this early in the developmentof Linux would have meant losing the potential input ofhundreds, even thousands, of Linux developers. It wouldalso mean growing incompatibility between Linux and theGNU system that Stallman and the GNU team still hopedto develop. 

As leader of the GNU Project, Stallman had alreadyexperienced the negative effects of a software fork in1991. A group of Emacs developers working for asoftware company named Lucid had a falling out overStallman's unwillingness to fold changes back into theGNU Emacs code base. The fork had given birth to aparallel version, Lucid Emacs, and hard feelings all around. Jamie Zawinski, aformer Lucid programmer who would goon to head the Mozilla development team, has a web sitethat documents the Lucid/GNU Emacs fork, titled, "TheLemacs/FSFmacs Schism. " http://www. Jwz. Org/doc/lemacs. Html

Murdock says Debian was mounting work on a similar forkin glibc source code that motivated Stallman to insiston adding the GNU prefix when Debian rolled out itssoftware distribution. "The fork has since converged. Still, at the time, there was a concern that if theLinux community saw itself as a different thing as theGNU community, it might be a force for disunity. "

Stallman seconds Murdock's recollection. In fact, hesays there were nascent forks appearing in relation toevery major GNU component. At first, Stallman says heconsidered the forks to be a product of sour grapes. Incontrast to the fast and informal dynamics of theLinux-kernel team, GNU source-code maintainers tendedto be slower and more circumspect in making changesthat might affect a program's long-term viability. Theyalso were unafraid of harshly critiquing other people'scode. Over time, however, Stallman began to sense thatthere was an underlying lack of awareness of the GNUProject and its objectives when reading Linuxdevelopers' emails. 

"We discovered that the people who consideredthemselves Linux users didn't care about the GNUProject, " Stallman says. "They said, `Why should Ibother doing these things? I don't care about the GNUProject. It's working for me. It's working for us Linuxusers, and nothing else matters to us. ' And that wasquite surprising given that people were essentiallyusing a variant of the GNU system, and they cared solittle. They cared less than anybody else about GNU. "

While some viewed descriptions of Linux as a "variant"of the GNU Project as politically grasping, Murdock, already sympathetic to the free software cause, sawStallman's request to call Debian's version GNU/Linuxas reasonable. "It was more for unity than for credit, "he says. 

Requests of a more technical nature quickly followed. Although Murdock had been accommodating on politicalissues, he struck a firmer pose when it came to thedesign and development model of the actual software. What had begun as a show of solidarity soon became ofmodel of other GNU projects. 

"I can tell you that I've had my share of disagreementswith him, " says Murdock with a laugh. "In all honestyRichard can be a fairly difficult person to work with. "

In 1996, Murdock, following his graduation from Purdue, decided to hand over the reins of the growing Debianproject. He had already been ceding management dutiesto Bruce Perens, the hacker best known for his work onElectric Fence, a Unix utility released under the GPL. Perens, like Murdock, was a Unix programmer who hadbecome enamored of GNU/Linux as soon as the program'sUnix-like abilities became manifest. Like Murdock, Perens sympathized with the political agenda ofStallman and the Free Software Foundation, albeit from afar. 

"I remember after Stallman had already come out withthe GNU Manifesto, GNU Emacs, and GCC, I read anarticle that said he was working as a consultant forIntel, " says Perens, recalling his first brush withStallman in the late 1980s. "I wrote him asking how hecould be advocating free software on the one hand andworking for Intel on the other. He wrote back saying, `I work as a consultant to produce free software. ' Hewas perfectly polite about it, and I thought his answermade perfect sense. "

As a prominent Debian developer, however, Perensregarded Murdock's design battles with Stallman withdismay. Upon assuming leadership of the developmentteam, Perens says he made the command decision todistance Debian from the Free Software Foundation. "Idecided we did not want Richard's style ofmicro-management, " he says. 

According to Perens, Stallman was taken aback by thedecision but had the wisdom to roll with it. "He gaveit some time to cool off and sent a message that wereally needed a relationship. He requested that we callit GNU/Linux and left it at that. I decided that wasfine. I made the decision unilaterally. Everybodybreathed a sigh of relief. "

Over time, Debian would develop a reputation as thehacker's version of Linux, alongside Slackware, anotherpopular distribution founded during the same 1993-1994period. Outside the realm of hacker-oriented systems, however, Linux was picking up steam in the commercialUnix marketplace. In North Carolina, a Unix companybilling itself as Red Hat was revamping its business tofocus on Linux. The chief executive officer was RobertYoung, the former Linux Journal editor who in 1994 hadput the question to Linus Torvalds, asking whether hehad any regrets about putting the kernel under the GPL. To Young, Torvalds' response had a "profound" impact onhis own view toward Linux. Instead of looking for a wayto corner the GNU/Linux market via traditional softwaretactics, Young began to consider what might happen if acompany adopted the same approach as Debian-i. E. , building an operating system completely out of freesoftware parts. Cygnus Solutions, the company foundedby Michael Tiemann and John Gilmore in 1990, wasalready demonstrating the ability to sell free softwarebased on quality and customizability. What if Red Hattook the same approach with GNU/Linux?

"In the western scientific tradition we stand on theshoulders of giants, " says Young, echoing both Torvaldsand Sir Isaac Newton before him. "In business, thistranslates to not having to reinvent wheels as we goalong. The beauty of [the GPL] model is you put yourcode into the public domain. Young uses the term "public domain" incorrectlyhere. Public domain means not protected by copyright. GPL-protected programs are by definition protected bycopyright. If you're an independent software vendor and you'retrying to build some application and you need amodem-dialer, well, why reinvent modem dialers? You canjust steal PPP off of Red Hat Linux and use that as thecore of your modem-dialing tool. If you need a graphictool set, you don't have to write your own graphiclibrary. Just download GTK. Suddenly you have theability to reuse the best of what went before. Andsuddenly your focus as an application vendor is less onsoftware management and more on writing theapplications specific to your customer's needs. "

Young wasn't the only software executive intrigued bythe business efficiencies of free software. By late1996, most Unix companies were starting to wake up andsmell the brewing source code. The Linux sector wasstill a good year or two away from full commercialbreakout mode, but those close enough to the hackercommunity could feel it: something big was happening. The Intel 386 chip, the Internet, and the World WideWeb had hit the marketplace like a set of monsterwaves, and Linux-and the host of software programs thatechoed it in terms of source-code accessibility andpermissive licensing-seemed like the largest wave yet. 

For Ian Murdock, the programmer courted by Stallman andthen later turned off by Stallman's micromanagementstyle, the wave seemed both a fitting tribute and afitting punishment for the man who had spent so muchtime giving the free software movement an identity. Like many Linux aficionados, Murdock had seen theoriginal postings. He'd seen Torvalds's originaladmonition that Linux was "just a hobby. " He'd alsoseen Torvalds's admission to Minix creator AndrewTanenbaum: "If the GNU kernel had been ready lastspring, I'd not have bothered to even start my project. "This quote is takenfrom the much-publicizedTorvalds-Tanenbaum "flame war" following the initialrelease of Linux. In the process of defending hischoice of a nonportable monolithic kernel design, Torvalds says he started working on Linux as a way tolearn more about his new 386 PC. "If the GNU kernel hadbeen ready last spring, I'd not have bothered to evenstart my project. " See Chris DiBona et al. , OpenSources (O'Reilly & Associates, Inc. , 1999): 224. Like many, Murdock knew the opportunities that hadbeen squandered. He also knew the excitement ofwatching new opportunities come seeping out of the veryfabric of the Internet. 

"Being involved with Linux in those early days wasfun, " recalls Murdock. "At the same time, it wassomething to do, something to pass the time. If you goback and read those old [comp. Os. Minix] exchanges, you'll see the sentiment: this is something we can playwith until the HURD is ready. People were anxious. It'sfunny, but in a lot of ways, I suspect that Linux wouldnever have happened if the HURD had come along more quickly. "

By the end of 1996, however, such "what if" questionswere already moot. Call it Linux, call it GNU/Linux;the users had spoken. The 36-month window had closed, meaning that even if the GNU Project had rolled out itsHURD kernel, chances were slim anybody outside thehard-core hacker community would have noticed. Thefirst Unix-like free software operating system washere, and it had momentum. All hackers had left to dowas sit back and wait for the next major wave to comecrashing down on their heads. Even the shaggy-hairedhead of one Richard M. Stallman. 

Ready or not. 

Open Source

In November, 1995, Peter Salus, a member of the FreeSoftware Foundation and author of the 1994 book, AQuarter Century of Unix, issued a call for papers tomembers of the GNU Project's "system-discuss" mailinglist. Salus, the conference's scheduled chairman, wanted to tip off fellow hackers about the upcomingConference on Freely Redistributable Software inCambridge, Massachusetts. Slated for February, 1996 andsponsored by the Free Software Foundation, the eventpromised to be the first engineering conference solelydedicated to free software and, in a show of unity withother free software programmers, welcomed papers on"any aspect of GNU, Linux, NetBSD, 386BSD, FreeBSD, Perl, Tcl/tk, and other tools for which the code isaccessible and redistributable. " Salus wrote: Over thepast 15 years, free and low-cost software has becomeubiquitous. This conference will bring togetherimplementers of several different types of freelyredistributable software and publishers of suchsoftware (on various media). There will be tutorialsand refereed papers, as well as keynotes by LinusTorvalds and Richard Stallman. See Peter Salus, "FYI-Conference on FreelyRedistributable Software, 2/2, Cambridge" (1995)(archived by Terry Winograd). 

http://hci. Stanford. Edu/pcd-archives/pcd-fyi/1995/0078. Html One of the first people to receive Salus' email wasconference committee member Eric S. Raymond. Althoughnot the leader of a project or company like the variousother members of the list, Raymond had built a tidyreputation within the hacker community as a majorcontributor to GNU Emacs and as editor of The NewHacker Dictionary, a book version of the hackingcommunity's decade-old Jargon File. 

For Raymond, the 1996 conference was a welcome event. Active in the GNU Project during the 1980s, Raymond haddistanced himself from the project in 1992, citing, like many others before him, Stallman's"micro-management" style. "Richard kicked up a fussabout my making unauthorized modifications when I wascleaning up the Emacs LISP libraries, " Raymond recalls. "It frustrated me so much that I decided I didn't wantto work with him anymore. "

Despite the falling out, Raymond remained active in thefree software community. So much so that when Salussuggested a conference pairing Stallman and Torvalds askeynote speakers, Raymond eagerly seconded the idea. With Stallman representing the older, wiser contingentof ITS/Unix hackers and Torvalds representing theyounger, more energetic crop of Linux hackers, thepairing indicated a symbolic show of unity that couldonly be beneficial, especially to ambitious younger(i. E. , below 40) hackers such as Raymond. "I sort ofhad a foot in both camps, " Raymond says. 

By the time of the conference, the tension betweenthose two camps had become palpable. Both groups hadone thing in common, though: the conference was theirfirst chance to meet the Finnish wunderkind in theflesh. Surprisingly, Torvalds proved himself to be acharming, affable speaker. Possessing only a slightSwedish accent, Torvalds surprised audience memberswith his quick, self-effacing wit. Although Linus Torvalds is Finnish, hismother tongueis Swedish. "The Rampantly Unofficial Linus FAQ" offersa brief explanation: Finland has a significant (about6%) Swedish-speaking minority population. They callthemselves "finlandssvensk" or "finlandssvenskar" andconsider themselves Finns; many of their families havelived in Finland for centuries. Swedish is one ofFinland's two official languages. Http://tuxedo. Org/~esr/faqs/linus/ Even more surprising, says Raymond, was Torvalds'equal willingness to take potshots at other prominenthackers, including the most prominent hacker of all, Richard Stallman. By the end of the conference, Torvalds' half-hacker, half-slacker manner was winningover older and younger conference-goers alike. 

"It was a pivotal moment, " recalls Raymond. "Before1996, Richard was the only credible claimant to beingthe ideological leader of the entire culture. Peoplewho dissented didn't do so in public. The person whobroke that taboo was Torvalds. "

The ultimate breach of taboo would come near the end ofthe show. During a discussion on the growing marketdominance of Microsoft Windows or some similar topic, Torvalds admitted to being a fan of Microsoft'sPowerPoint slideshow software program. From theperspective of old-line software purists, it was like aMormon bragging in church about his fondness ofwhiskey. From the perspective of Torvalds and hisgrowing band of followers, it was simply common sense. Why shun worthy proprietary software programs just tomake a point? Being a hacker wasn't about suffering, itwas about getting the job done. 

"That was a pretty shocking thing to say, " Raymondremembers. "Then again, he was able to do that, becauseby 1995 and 1996, he was rapidly acquiring clout. "

Stallman, for his part, doesn't remember any tension atthe 1996 conference, but he does remember later feelingthe sting of Torvalds' celebrated cheekiness. "Therewas a thing in the Linux documentation which says printout the GNU coding standards and then tear them up, "says Stallman, recalling one example. "OK, so hedisagrees with some of our conventions. That's fine, but he picked a singularly nasty way of saying so. Hecould have just said `Here's the way I think you shouldindent your code. ' Fine. There should be no hostility there. "

For Raymond, the warm reception other hackers gave toTorvalds' comments merely confirmed his suspicions. Thedividing line separating Linux developers fromGNU/Linux developers was largely generational. ManyLinux hackers, like Torvalds, had grown up in a worldof proprietary software. Unless a program was clearlyinferior, most saw little reason to rail against aprogram on licensing issues alone. Somewhere in theuniverse of free software systems lurked a program thathackers might someday turn into a free softwarealternative to PowerPoint. Until then, why begrudgeMicrosoft the initiative of developing the program andreserving the rights to it?

As a former GNU Project member, Raymond sensed an addeddynamic to the tension between Stallman and Torvalds. In the decade since launching the GNU Project, Stallmanhad built up a fearsome reputation as a programmer. Hehad also built up a reputation for intransigence bothin terms of software design and people management. Shortly before the 1996 conference, the Free SoftwareFoundation would experience a full-scale staffdefection, blamed in large part on Stallman. BrianYoumans, a current FSF staffer hired by Salus in thewake of the resignations, recalls the scene: "At onepoint, Peter [Salus] was the only staff member workingin the office. "

For Raymond, the defection merely confirmed a growingsuspicion: recent delays such as the HURD and recenttroubles such as the Lucid-Emacs schism reflectedproblems normally associated with software projectmanagement, not software code development. Shortlyafter the Freely Redistributable Software Conference, Raymond began working on his own pet software project, a popmail utility called " fetchmail. " Taking a cuefrom Torvalds, Raymond issued his program with atacked-on promise to update the source code as earlyand as often as possible. When users began sending inbug reports and feature suggestions, Raymond, at firstanticipating a tangled mess, found the resultingsoftware surprisingly sturdy. Analyzing the success ofthe Torvalds approach, Raymond issued a quick analysis:using the Internet as his "petri dish" and the harshscrutiny of the hacker community as a form of naturalselection, Torvalds had created an evolutionary modelfree of central planning. 

What's more, Raymond decided, Torvalds had found a wayaround Brooks' Law. First articulated by Fred P. Brooks, manager of IBM's OS/360 project and author ofthe 1975 book, The Mythical Man-Month, Brooks' Lawheld that adding developers to a project only resultedin further project delays. Believing as most hackersthat software, like soup, benefits from a limitednumber of cooks, Raymond sensed something revolutionaryat work. In inviting more and more cooks into thekitchen, Torvalds had actually found away to make theresulting software better. Brooks' Law is the shorthand summary of the followingquote taken from Brooks' book: Since softwareconstruction is inherently a systems effort-an exercisein complex interrelationships-communication effort isgreat, and it quickly dominates the decrease inindividual task time brought about by partitioning. Adding more men then lengthens, not shortens, theschedule. See Fred P. Brooks, The Mythical Man-Month(Addison Wesley Publishing, 1995)

Raymond put his observations on paper. He crafted theminto a speech, which he promptly delivered before agroup of friends and neighbors in Chester County, Pennsylvania. Dubbed " The Cathedral and the Bazaar, "the speech contrasted the management styles of the GNUProject with the management style of Torvalds and thekernel hackers. Raymond says the response wasenthusiastic, but not nearly as enthusiastic as the onehe received during the 1997 Linux Kongress, a gatheringof Linux users in Germany the next spring. 

"At the Kongress, they gave me a standing ovation atthe end of the speech, " Raymond recalls. "I took thatas significant for two reasons. For one thing, it meantthey were excited by what they were hearing. Foranother thing, it meant they were excited even afterhearing the speech delivered through a language barrier. "

Eventually, Raymond would convert the speech into apaper, also titled "The Cathedral and the Bazaar. " Thepaper drew its name from Raymond's central analogy. GNUprograms were "cathedrals, " impressive, centrallyplanned monuments to the hacker ethic, built to standthe test of time. Linux, on the other hand, was morelike "a great babbling bazaar, " a software programdeveloped through the loose decentralizing dynamics ofthe Internet. 

Implicit within each analogy was a comparison ofStallman and Torvalds. Where Stallman served as theclassic model of the cathedral architect-i. E. , aprogramming "wizard" who could disappear for 18 monthsand return with something like the GNU CCompiler-Torvalds was more like a genial dinner-partyhost. In letting others lead the Linux designdiscussion and stepping in only when the entire tableneeded a referee, Torvalds had created a developmentmodel very much reflective of his own laid-backpersonality. From the Torvalds' perspective, the mostimportant managerial task was not imposing control butkeeping the ideas flowing. 

Summarized Raymond, "I think Linus's cleverest and mostconsequential hack was not the construction of theLinux kernel itself, but rather his invention of theLinux development model. "See Eric Raymond, "The Cathredral and the Bazaar"(1997). 

In summarizing the secrets of Torvalds' managerialsuccess, Raymond himself had pulled off a coup. One ofthe audience members at the Linux Kongress was TimO'Reilly, publisher of O'Reilly & Associates, a companyspecializing in software manuals and software-relatedbooks (and the publisher of this book). After hearingRaymond's Kongress speech, O'Reilly promptly invitedRaymond to deliver it again at the company's inauguralPerl Conference later that year in Monterey, California. 

Although the conference was supposed to focus on Perl, a scripting language created by Unix hacker Larry Wall, O'Reilly assured Raymond that the conference wouldaddress other free software technologies. Given thegrowing commercial interest in Linux and Apache, apopular free software web server, O'Reilly hoped to usethe event to publicize the role of free software increating the entire infrastructure of the Internet. From web-friendly languages such as Perl and Python toback-room programs such as BIND (the Berkeley InternetNaming Daemon), a software tool that lets users replacearcane IP numbers with the easy-to-remember domain-nameaddresses (e. G. , amazon. Com), and sendmail, the mostpopular mail program on the Internet, free software hadbecome an emergent phenomenon. Like a colony of antscreating a beautiful nest one grain of sand at a time, the only thing missing was the communal self-awareness. O'Reilly saw Raymond's speech as a good way to inspirethat self-awareness, to drive home the point that freesoftware development didn't start and end with the GNUProject. Programming languages, such as Perl andPython, and Internet software, such as BIND, sendmail, and Apache, demonstrated that free software was alreadyubiquitous and influential. He also assured Raymond aneven warmer reception than the one at Linux Kongress. 

O'Reilly was right. "This time, I got the standingovation before the speech, " says Raymond, laughing. 

As predicted, the audience was stocked not only withhackers, but with other people interested in thegrowing power of the free software movement. Onecontingent included a group from Netscape, the MountainView, California startup then nearing the end game ofits three-year battle with Microsoft for control of theweb-browser market. 

Intrigued by Raymond's speech and anxious to win backlost market share, Netscape executives took the messageback to corporate headquarters. A few months later, inJanuary, 1998, the company announced its plan topublish the source code of its flagship Navigator webbrowser in the hopes of enlisting hacker support infuture development. 

When Netscape CEO Jim Barksdale cited Raymond's"Cathedral and the Bazaar" essay as a major influenceupon the company's decision, the company instantlyelevated Raymond to the level of hacker celebrity. Determined not to squander the opportunity, Raymondtraveled west to deliver interviews, advise Netscapeexecutives, and take part in the eventual partycelebrating the publication of Netscape Navigator'ssource code. The code name for Navigator's source codewas "Mozilla": a reference both to the program'sgargantuan size-30 million lines of code-and to itsheritage. Developed as a proprietary offshoot ofMosaic, the web browser created by Marc Andreessen atthe University of Illinois, Mozilla was proof, yetagain, that when it came to building new programs, mostprogrammers preferred to borrow on older, modifiable programs. 

While in California, Raymond also managed to squeeze ina visit to VA Research, a Santa Clara-based companyselling workstations with the GNU/Linux operatingsystem preinstalled. Convened by Raymond, the meetingwas small. The invite list included VA founder LarryAugustin, a few VA employees, and Christine Peterson, president of the Foresight Institute, a Silicon Valleythink tank specializing in nanotechnology. 

"The meeting's agenda boiled down to one item: how totake advantage of Netscape's decision so that othercompanies might follow suit?" Raymond doesn't recallthe conversation that took place, but he does rememberthe first complaint addressed. Despite the best effortsof Stallman and other hackers to remind people that theword "free" in free software stood for freedom and notprice, the message still wasn't getting through. Mostbusiness executives, upon hearing the term for thefirst time, interpreted the word as synonymous with"zero cost, " tuning out any follow up messages in shortorder. Until hackers found a way to get past thiscognitive dissonance, the free software movement facedan uphill climb, even after Netscape. 

Peterson, whose organization had taken an activeinterest in advancing the free software cause, offeredan alternative: open source. 

Looking back, Peterson says she came up with the opensource term while discussing Netscape's decision with afriend in the public relations industry. She doesn'tremember where she came upon the term or if sheborrowed it from another field, but she does rememberher friend disliking the term. 5

At the meeting, Peterson says, the response wasdramatically different. "I was hesitant aboutsuggesting it, " Peterson recalls. "I had no standingwith the group, so started using it casually, nothighlighting it as a new term. " To Peterson's surprise, the term caught on. By the end of the meeting, most ofthe attendees, including Raymond, seemed pleased by it. 

Raymond says he didn't publicly use the term "opensource" as a substitute for free software until a dayor two after the Mozilla launch party, when O'Reillyhad scheduled a meeting to talk about free software. Calling his meeting "the Freeware Summit, " O'Reillysays he wanted to direct media and community attentionto the other deserving projects that had alsoencouraged Netscape to release Mozilla. "All these guyshad so much in common, and I was surprised they didn'tall know each other, " says O'Reilly. "I also wanted tolet the world know just how great an impact the freesoftware culture had already made. People were missingout on a large part of the free software tradition. "

In putting together the invite list, however, O'Reillymade a decision that would have long-term politicalconsequences. He decided to limit the list towest-coast developers such as Wall, Eric Allman, creator of sendmail, and Paul Vixie, creator of BIND. There were exceptions, of course: Pennsylvania-residentRaymond, who was already in town thanks to the Mozillalaunch, earned a quick invite. So did Virginia-residentGuido van Rossum, creator of Python. "Frank Willison, my editor in chief and champion of Python within thecompany, invited him without first checking in withme, " O'Reilly recalls. "I was happy to have him there, but when I started, it really was just a local gathering. "

For some observers, the unwillingness to includeStallman's name on the list qualified as a snub. "Idecided not to go to the event because of it, " saysPerens, remembering the summit. Raymond, who did go, says he argued for Stallman's inclusion to no avail. The snub rumor gained additional strength from the factthat O'Reilly, the event's host, had feuded publiclywith Stallman over the issue of software-manualcopyrights. Prior to the meeting, Stallman had arguedthat free software manuals should be as freely copyableand modifiable as free software programs. O'Reilly, meanwhile, argued that a value-added market for nonfreebooks increased the utility of free software by makingit more accessible to a wider community. The two hadalso disputed the title of the event, with Stallmaninsisting on "Free Software" over the less politicallyladen "Freeware. "

Looking back, O'Reilly doesn't see the decision toleave Stallman's name off the invite list as a snub. "At that time, I had never met Richard in person, butin our email interactions, he'd been inflexible andunwilling to engage in dialogue. I wanted to make surethe GNU tradition was represented at the meeting, so Iinvited John Gilmore and Michael Tiemann, whom I knewpersonally, and whom I knew were passionate about thevalue of the GPL but seemed more willing to engage in afrank back-and-forth about the strengths and weaknessesof the various free software projects and traditions. Given all the later brouhaha, I do wish I'd invitedRichard as well, but I certainly don't think that myfailure to do so should be interpreted as a lack ofrespect for the GNU Project or for Richard personally. "

Snub or no snub, both O'Reilly and Raymond say the term"open source" won over just enough summit-goers toqualify as a success. The attendees shared ideas andexperiences and brainstormed on how to improve freesoftware's image. Of key concern was how to point outthe successes of free software, particularly in therealm of Internet infrastructure, as opposed to playingup the GNU/Linux challenge to Microsoft Windows. Butlike the earlier meeting at VA, the discussion soonturned to the problems associated with the term "freesoftware. " O'Reilly, the summit host, remembers aparticularly insightful comment from Torvalds, a summit attendee. 

"Linus had just moved to Silicon Valley at that point, and he explained how only recently that he had learnedthat the word `free' had two meanings-free as in`libre' and free as in `gratis'-in English. "

Michael Tiemann, founder of Cygnus, proposed analternative to the troublesome "free software" term:sourceware. "Nobody got too excited about it, " O'Reillyrecalls. "That's when Eric threw out the term `open source. '"

Although the term appealed to some, support for achange in official terminology was far from unanimous. At the end of the one-day conference, attendees put thethree terms-free software, open source, orsourceware-to a vote. According to O'Reilly, 9 out ofthe 15 attendees voted for "open source. " Although somestill quibbled with the term, all attendees agreed touse it in future discussions with the press. "We wantedto go out with a solidarity message, " O'Reilly says. 

The term didn't take long to enter the nationallexicon. Shortly after the summit, O'Reilly shepherdedsummit attendees to a press conference attended byreporters from the New York Times, the Wall StreetJournal, and other prominent publications. Within a fewmonths, Torvalds' face was appearing on the cover ofForbes magazine, with the faces of Stallman, Perlcreator Larry Wall, and Apache team leader BrianBehlendorf featured in the interior spread. Open sourcewas open for business. 

For summit attendees such as Tiemann, the solidaritymessage was the most important thing. Although hiscompany had achieved a fair amount of success sellingfree software tools and services, he sensed thedifficulty other programmers and entrepreneurs faced. 

"There's no question that the use of the word free wasconfusing in a lot of situations, " Tiemann says. "Opensource positioned itself as being business friendly andbusiness sensible. Free software positioned itself asmorally righteous. For better or worse we figured itwas more advantageous to align with the open source crowd. 

For Stallman, the response to the new "open source"term was slow in coming. Raymond says Stallman brieflyconsidered adopting the term, only to discard it. "Iknow because I had direct personal conversations aboutit, " Raymond says. 

By the end of 1998, Stallman had formulated a position:open source, while helpful in communicating thetechnical advantages of free software, also encouragedspeakers to soft-pedal the issue of software freedom. Given this drawback, Stallman would stick with the termfree software. 

Summing up his position at the 1999 LinuxWorldConvention and Expo, an event billed by Torvaldshimself as a "coming out party" for the Linuxcommunity, Stallman implored his fellow hackers toresist the lure of easy compromise. 

"Because we've shown how much we can do, we don't haveto be desperate to work with companies or compromiseour goals, " Stallman said during a panel discussion. "Let them offer and we'll accept. We don't have tochange what we're doing to get them to help us. You cantake a single step towards a goal, then another andthen more and more and you'll actually reach your goal. Or, you can take a half measure that means you don'tever take another step and you'll never get there. "

Even before the LinuxWorld show, however, Stallman wasshowing an increased willingness to alienate his moreconciliatory peers. A few months after the FreewareSummit, O'Reilly hosted its second annual PerlConference. This time around, Stallman was inattendance. During a panel discussion lauding IBM'sdecision to employ the free software Apache web serverin its commercial offerings, Stallman, taking advantageof an audience microphone, disrupted the proceedingswith a tirade against panelist John Ousterhout, creatorof the Tcl scripting language. Stallman brandedOusterhout a "parasite" on the free software communityfor marketing a proprietary version of Tcl viaOusterhout's startup company, Scriptics. "I don't thinkScriptics is necessary for the continued existence ofTcl, " Stallman said to hisses from the fellow audience members. See MalcolmMaclachlan, "Profit Motive Splits OpenSource Movement, " TechWeb News (August 26, 1998). Http://content. Techweb. Com/wire/story/TWB19980824S0012

"It was a pretty ugly scene, " recalls Prime TimeFreeware's Rich Morin. "John's done some prettyrespectable things: Tcl, Tk, Sprite. He's a real contributor. "

Despite his sympathies for Stallman and Stallman'sposition, Morin felt empathy for those troubled byStallman's discordant behavior. 

Stallman's Perl Conference outburst would momentarilychase off another potential sympathizer, Bruce Perens. In 1998, Eric Raymond proposed launching the OpenSource Initiative, or OSI, an organization that wouldpolice the use of the term "open source" and provide adefinition for companies interested in making their ownprograms. Raymond recruited Perens to draft the definition. See Bruce Perens etal. , "The Open Source Definition, "The Open Source Initiative (1998). Http://www. Opensource. Org/docs/definition. Html

Perens would later resign from the OSI, expressingregret that the organization had set itself up inopposition to Stallman and the FSF. Still, looking backon the need for a free software definition outside theFree Software Foundation's auspices, Perens understandswhy other hackers might still feel the need fordistance. "I really like and admire Richard, " saysPerens. "I do think Richard would do his job better ifRichard had more balance. That includes going away fromfree software for a couple of months. "

Stallman's monomaniacal energies would do little tocounteract the public-relations momentum of open sourceproponents. In August of 1998, when chip-maker Intelpurchased a stake in GNU/Linux vendor Red Hat, anaccompanying New York Times article described thecompany as the product of a movement "knownalternatively as free software and open source. "See Amy Harmon, "For Sale: FreeOperating System, " NewYork Times (September 28, 1998). 

http://www. Nytimes. Com/library/tech/98/09/biztech/articles/28linux. Html Six months later, a John Markoff article on AppleComputer was proclaiming the company's adoption of the"open source" Apache server in the article headline. See John Markoff, "AppleAdopts `Open Source' for itsServer Computers, " New York Times (March 17, 1999). 

http://www. Nytimes. Com/library/tech/99/03/biztech/articles/17apple. Html

Such momentum would coincide with the growing momentumof companies that actively embraced the "open source"term. By August of 1999, Red Hat, a company that noweagerly billed itself as "open source, " was sellingshares on Nasdaq. In December, VA Linux-formerly VAResearch-was floating its own IPO to historical effect. Opening at $30 per share, the company's stock priceexploded past the $300 mark in initial trading only tosettle back down to the $239 level. Shareholders luckyenough to get in at the bottom and stay until the endexperienced a 698% increase in paper wealth, a Nasdaq record. 

Among those lucky shareholders was Eric Raymond, who, as a company board member since the Mozilla launch, hadreceived 150, 000 shares of VA Linux stock. Stunned bythe realization that his essay contrasting theStallman-Torvalds managerial styles had netted him $36million in potential wealth, Raymond penned a follow-upessay. In it, Raymond mused on the relationship betweenthe hacker ethic and monetary wealth: Reporters oftenask me these days if I think the open-source communitywill be corrupted by the influx of big money. I tellthem what I believe, which is this: commercial demandfor programmers has been so intense for so long thatanyone who can be seriously distracted by money isalready gone. Our community has been self-selected forcaring about other things-accomplishment, pride, artistic passion, and each other. See Eric Raymond, "Surprised by Wealth, " LinuxToday(December 10, 1999). 

http://linuxtoday. Com/news_story. Php3?ltsn=1999-12-10-001-05-NW-LF Whether or not such comments allayed suspicions thatRaymond and other open source proponents had simplybeen in it for the money, they drove home the opensource community's ultimate message: all you needed tosell the free software concept is a friendly face and asensible message. Instead of fighting the marketplacehead-on as Stallman had done, Raymond, Torvalds, andother new leaders of the hacker community had adopted amore relaxed approach-ignoring the marketplace in someareas, leveraging it in others. Instead of playing therole of high-school outcasts, they had played the gameof celebrity, magnifying their power in the process. 

"On his worst days Richard believes that Linus Torvaldsand I conspired to hijack his revolution, " Raymondsays. "Richard's rejection of the term open source andhis deliberate creation of an ideological fissure in myview comes from an odd mix of idealism andterritoriality. There are people out there who thinkit's all Richard's personal ego. I don't believe that. It's more that he so personally associates himself withthe free software idea that he sees any threat to thatas a threat to himself. "

Ironically, the success of open source and open sourceadvocates such as Raymond would not diminish Stallman'srole as a leader. If anything, it gave Stallman newfollowers to convert. Still, the Raymond territorialitycharge is a damning one. There are numerous instancesof Stallman sticking to his guns more out of habit thanout of principle: his initial dismissal of the Linuxkernel, for example, and his current unwillingness as apolitical figure to venture outside the realm ofsoftware issues. 

Then again, as the recent debate over open source alsoshows, in instances when Stallman has stuck to hisguns, he's usually found a way to gain ground becauseof it. "One of Stallman's primary character traits isthe fact he doesn't budge, " says Ian Murdock. "He'llwait up to a decade for people to come around to hispoint of view if that's what it takes. "

Murdock, for one, finds that unbudgeable nature bothrefreshing and valuable. Stallman may no longer be thesolitary leader of the free software movement, but heis still the polestar of the free software community. "You always know that he's going to be consistent inhis views, " Murdock says. "Most people aren't likethat. Whether you agree with him or not, you reallyhave to respect that. "

A Brief Journey Through Hacker Hell

Richard Stallman stares, unblinking, through thewindshield of a rental car, waiting for the light tochange as we make our way through downtown Kihei. 

The two of us are headed to the nearby town of Pa'ia, where we are scheduled to meet up with some softwareprogrammers and their wives for dinner in about an houror so. 

It's about two hours after Stallman's speech at theMaui High Performance Center, and Kihei, a town thatseemed so inviting before the speech, now seemsprofoundly uncooperative. Like most beach cities, Kiheiis a one-dimensional exercise in suburban sprawl. Driving down its main drag, with its endless successionof burger stands, realty agencies, and bikini shops, it's hard not to feel like a steel-coated morselpassing through the alimentary canal of a giantcommercial tapeworm. The feeling is exacerbated by thelack of side roads. With nowhere to go but forward, traffic moves in spring-like lurches. 200 yards ahead, a light turns green. By the time we are moving, thelight is yellow again. 

For Stallman, a lifetime resident of the east coast, the prospect of spending the better part of a sunnyHawaiian afternoon trapped in slow traffic is enough totrigger an embolism. Even worse is the knowledge that, with just a few quick right turns a quarter mile back, this whole situation easily could have been avoided. Unfortunately, we are at the mercy of the driver aheadof us, a programmer from the lab who knows the way andwho has decided to take us to Pa'ia via the scenicroute instead of via the nearby Pilani Highway. 

"This is terrible, " says Stallman between frustratedsighs. "Why didn't we take the other route?"

Again, the light a quarter mile ahead of us turnsgreen. Again, we creep forward a few more car lengths. This process continues for another 10 minutes, until wefinally reach a major crossroad promising access to theadjacent highway. 

The driver ahead of us ignores it and continues throughthe intersection. 

"Why isn't he turning?" moans Stallman, throwing up hishands in frustration. "Can you believe this?"

I decide not to answer either. I find the fact that Iam sitting in a car with Stallman in the driver seat, in Maui no less, unbelievable enough. Until two hoursago, I didn't even know Stallman knew how to drive. Now, listening to Yo-Yo Ma's cello playing the mournfulbass notes of "Appalachian Journey" on the car stereoand watching the sunset pass by on our left, I do mybest to fade into the upholstery. 

When the next opportunity to turn finally comes up, Stallman hits his right turn signal in an attempt tocue the driver ahead of us. No such luck. Once again, we creep slowly through the intersection, coming to astop a good 200 yards before the next light. By now, Stallman is livid. 

"It's like he's deliberately ignoring us, " he says, gesturing and pantomiming like an air craft carrierlanding-signals officer in a futile attempt to catchour guide's eye. The guide appears unfazed, and for thenext five minutes all we see is a small portion of hishead in the rearview mirror. 

I look out Stallman's window. Nearby Kahoolawe andLanai Islands provide an ideal frame for the settingsun. It's a breathtaking view, the kind that makesmoments like this a bit more bearable if you're aHawaiian native, I suppose. I try to direct Stallman'sattention to it, but Stallman, by now obsessed by theinattentiveness of the driver ahead of us, blows me off. 

When the driver passes through another green light, completely ignoring a "Pilani Highway Next Right, " Igrit my teeth. I remember an early warning relayed tome by BSD programmer Keith Bostic. "Stallman does notsuffer fools gladly, " Bostic warned me. "If somebodysays or does something stupid, he'll look them in theeye and say, `That's stupid. '"

Looking at the oblivious driver ahead of us, I realizethat it's the stupidity, not the inconvenience, that'skilling Stallman right now. 

"It's as if he picked this route with absolutely nothought on how to get there efficiently, " Stallman says. 

The word "efficiently" hangs in the air like a badodor. Few things irritate the hacker mind more thaninefficiency. It was the inefficiency of checking theXerox laser printer two or three times a day thattriggered Stallman's initial inquiry into the printersource code. It was the inefficiency of rewritingsoftware tools hijacked by commercial software vendorsthat led Stallman to battle Symbolics and to launch theGNU Project. If, as Jean Paul Sartre once opined, hellis other people, hacker hell is duplicating otherpeople's stupid mistakes, and it's no exaggeration tosay that Stallman's entire life has been an attempt tosave mankind from these fiery depths. 

This hell metaphor becomes all the more apparent as wetake in the slowly passing scenery. With its multitudeof shops, parking lots, and poorly timed street lights, Kihei seems less like a city and more like a poorlydesigned software program writ large. Instead ofrerouting traffic and distributing vehicles throughside streets and expressways, city planners haveelected to run everything through a single main drag. From a hacker perspective, sitting in a car amidst allthis mess is like listening to a CD rendition of nailson a chalkboard at full volume. 

"Imperfect systems infuriate hackers, " observes StevenLevy, another warning I should have listened to beforeclimbing into the car with Stallman. "This is onereason why hackers generally hate driving cars-thesystem of randomly programmed red lights and oddly laidout one-way streets causes delays which are so goddamnunnecessary [Levy's emphasis] that the impulse is torearrange signs, open up traffic-light control boxes .. . Redesign the entire system. "See Steven Levy, Hackers (Penguin USA[paperback], 1984): 40. 

More frustrating, however, is the duplicity of ourtrusted guide. Instead of searching out a clevershortcut-as any true hacker would do on instinct-thedriver ahead of us has instead chosen to play alongwith the city planners' game. Like Virgil in Dante'sInferno, our guide is determined to give us the fullguided tour of this hacker hell whether we want it or not. 

Before I can make this observation to Stallman, thedriver finally hits his right turn signal. Stallman'shunched shoulders relax slightly, and for a moment theair of tension within the car dissipates. The tensioncomes back, however, as the driver in front of us slowsdown. "Construction Ahead" signs line both sides of thestreet, and even though the Pilani Highway lies lessthan a quarter mile off in the distance, the two-laneroad between us and the highway is blocked by a dormantbulldozer and two large mounds of dirt. 

It takes Stallman a few seconds to register what'sgoing on as our guide begins executing a clumsyfive-point U-turn in front of us. When he catches aglimpse of the bulldozer and the "No Through Access"signs just beyond, Stallman finally boils over. 

"Why, why, why?" he whines, throwing his head back. "You should have known the road was blocked. You shouldhave known this way wouldn't work. You did this deliberately. "

The driver finishes the turn and passes us on the wayback toward the main drag. As he does so, he shakes hishead and gives us an apologetic shrug. Coupled with atoothy grin, the driver's gesture reveals a touch ofmainlander frustration but is tempered with aprotective dose of islander fatalism. Coming throughthe sealed windows of our rental car, it spells out asuccinct message: "Hey, it's Maui; what are you gonna do?"

Stallman can take it no longer. 

"Don't you fucking smile!" he shouts, fogging up theglass as he does so. "It's your fucking fault. This allcould have been so much easier if we had just done itmy way. "

Stallman accents the words "my way" by gripping thesteering wheel and pulling himself towards it twice. The image of Stallman's lurching frame is like that ofa child throwing a temper tantrum in a car seat, animage further underlined by the tone of Stallman'svoice. Halfway between anger and anguish, Stallmanseems to be on the verge of tears. 

Fortunately, the tears do not arrive. Like a summercloudburst, the tantrum ends almost as soon as itbegins. After a few whiny gasps, Stallman shifts thecar into reverse and begins executing his own U-turn. By the time we are back on the main drag, his face isas impassive as it was when we left the hotel 30minutes earlier. 

It takes less than five minutes to reach the nextcross-street. This one offers easy highway access, andwithin seconds, we are soon speeding off toward Pa'iaat a relaxing rate of speed. The sun that once loomedbright and yellow over Stallman's left shoulder is nowburning a cool orange-red in our rearview mirror. Itlends its color to the gauntlet wili wili trees flyingpast us on both sides of the highway. 

For the next 20 minutes, the only sound in our vehicle, aside from the ambient hum of the car's engine andtires, is the sound of a cello and a violin trioplaying the mournful strains of an Appalachian folktune. Endnote

Continuing the Fight

For Richard Stallman, time may not heal all wounds, butit does provide a convenient ally. 

Four years after " The Cathedral and the Bazaar, "Stallman still chafes over the Raymond critique. Healso grumbles over Linus Torvalds' elevation to therole of world's most famous hacker. He recalls apopular T-shirt that began showing at Linux tradeshowsaround 1999. Designed to mimic the original promotionalposter for Star Wars, the shirt depicted Torvaldsbrandishing a lightsaber like Luke Skywalker, whileStallman's face rides atop R2D2. The shirt still grateson Stallmans nerves not only because it depicts him asa Torvalds' sidekick, but also because it elevatesTorvalds to the leadership role in the freesoftware/open source community, a role even Torvaldshimself is loath to accept. "It's ironic, " saysStallman mournfully. "Picking up that sword is exactlywhat Linus refuses to do. He gets everybody focusing onhim as the symbol of the movement, and then he won'tfight. What good is it?"

Then again, it is that same unwillingness to "pick upthe sword, " on Torvalds part, that has left the dooropen for Stallman to bolster his reputation as thehacker community's ethical arbiter. Despite hisgrievances, Stallman has to admit that the last fewyears have been quite good, both to himself and to hisorganization. Relegated to the periphery by theunforeseen success of GNU/Linux, Stallman hasnonetheless successfully recaptured the initiative. Hisspeaking schedule between January 2000 and December2001 included stops on six continents and visits tocountries where the notion of software freedom carriesheavy overtones-China and India, for example. 

Outside the bully pulpit, Stallman has also learned howto leverage his power as costeward of the GNU GeneralPublic License (GPL). During the summer of 2000, whilethe air was rapidly leaking out of the 1999 Linux IPObubble, Stallman and the Free Software Foundationscored two major victories. In July, 2000, Troll Tech, a Norwegian software company and developer of Qt, avaluable suite of graphics tools for the GNU/Linuxoperating system, announced it was licensing itssoftware under the GPL. A few weeks later, SunMicrosystems, a company that, until then, had beenwarily trying to ride the open source bandwagon withoutgiving up total control of its software properties, finally relented and announced that it, too, was duallicensing its new OpenOffice application suite underthe Lesser GNU Public License (LGPL) and the SunIndustry Standards Source License (SISSL). 

Underlining each victory was the fact that Stallman haddone little to fight for them. In the case of TrollTech, Stallman had simply played the role of freesoftware pontiff. In 1999, the company had come up witha license that met the conditions laid out by the FreeSoftware Foundation, but in examining the licensefurther, Stallman detected legal incompatibles thatwould make it impossible to bundle Qt withGPL-protected software programs. Tired of battlingStallman, Troll Tech management finally decided tosplit the Qt into two versions, one GPL-protected andone QPL-protected, giving developers a way around thecompatibility issues cited by Stallman. 

In the case of Sun, they desired to play according tothe Free Software Foundation's conditions. At the 1999O'Reilly Open Source Conference, Sun Microsystemscofounder and chief scientist Bill Joy defended hiscompany's "community source" license, essentially awatered-down compromise letting users copy and modifySun-owned software but not charge a fee for saidsoftware without negotiating a royalty agreement withSun. A year after Joy's speech, Sun Microsystems vicepresident Marco Boerries was appearing on the samestage spelling out the company's new licensingcompromise in the case of OpenOffice, anoffice-application suite designed specifically for theGNU/Linux operating system. 

"I can spell it out in three letters, " said Boerries. "GPL. "

At the time, Boerries said his company's decision hadlittle to do with Stallman and more to do with themomentum of GPL-protected programs. "What basicallyhappened was the recognition that different productsattracted different communities, and the license youuse depends on what type of community you want toattract, " said Boerries. "With [OpenOffice], it wasclear we had the highest correlation with the GPL community. "See MarcoBoerries, interview with author (July, 2000). 

Such comments point out the under-recognized strengthof the GPL and, indirectly, the political genius of manwho played the largest role in creating it. "Thereisn't a lawyer on earth who would have drafted the GPLthe way it is, " says Eben Moglen, Columbia Universitylaw professor and Free Software Foundation generalcounsel. "But it works. And it works because ofRichard's philosophy of design. "

A former professional programmer, Moglen traces his probono work with Stallman back to 1990 when Stallmanrequested Moglen's legal assistance on a privateaffair. Moglen, then working with encryption expertPhillip Zimmerman during Zimmerman's legal battles withthe National Security Administration, says he washonored by the request. "I told him I used Emacs everyday of my life, and it would take an awful lot oflawyering on my part to pay off the debt. "

Since then, Moglen, perhaps more than any otherindividual, has had the best chance to observe thecrossover of Stallman's hacker philosophies into thelegal realm. Moglen says the difference betweenStallman's approach to legal code and software code arelargely the same. "I have to say, as a lawyer, the ideathat what you should do with a legal document is totake out all the bugs doesn't make much sense, " Moglensays. "There is uncertainty in every legal process, andwhat most lawyers want to do is to capture the benefitsof uncertainty for their client. Richard's goal is thecomplete opposite. His goal is to remove uncertainty, which is inherently impossible. It is inherentlyimpossible to draft one license to control allcircumstances in all legal systems all over the world. But if you were to go at it, you would have to go at ithis way. And the resulting elegance, the resultingsimplicity in design almost achieves what it has toachieve. And from there a little lawyering will carryyou quite far. "

As the person charged with pushing the Stallman agenda, Moglen understands the frustration of would-be allies. "Richard is a man who does not want to compromise overmatters that he thinks of as fundamental, " Moglen says, "and he does not take easily the twisting of words oreven just the seeking of artful ambiguity, which humansociety often requires from a lot of people. "

Because of the Free Software Foundation's unwillingnessto weigh in on issues outside the purview of GNUdevelopment and GPL enforcement, Moglen has taken todevoting his excess energies to assisting theElectronic Frontier Foundation, the organizationproviding legal aid to recent copyright defendants suchas Dmitri Skylarov. In 2000, Moglen also served asdirect counsel to a collection of hackers that werejoined together from circulating the DVD decryptionprogram deCSS. Despite the silence of his main clientin both cases, Moglen has learned to appreciate thevalue of Stallman's stubbornness. "There have beentimes over the years where I've gone to Richard andsaid, `We have to do this. We have to do that. Here'sthe strategic situation. Here's the next move. Here'swhat he have to do. ' And Richard's response has alwaysbeen, `We don't have to do anything. ' Just wait. Whatneeds doing will get done. "

"And you know what?" Moglen adds. "Generally, he's been right. "

Such comments disavow Stallman's own self-assessment:"I'm not good at playing games, " Stallman says, addressing the many unseen critics who see him as ashrewd strategist. "I'm not good at looking ahead andanticipating what somebody else might do. My approachhas always been to focus on the foundation, to say`Let's make the foundation as strong as we can make it. '"

The GPL's expanding popularity and continuinggravitational strength are the best tributes to thefoundation laid by Stallman and his GNU colleagues. While no longer capable of billing himself as the "lasttrue hacker, " Stallman nevertheless can take solecredit for building the free software movement'sethical framework. Whether or not other modernprogrammers feel comfortable working inside thatframework is immaterial. The fact that they even have achoice at all is Stallman's greatest legacy. 

Discussing Stallman's legacy at this point seems a bitpremature. Stallman, 48 at the time of this writing, still has a few years left to add to or subtract fromthat legacy. Still, the autopilot nature of the freesoftware movement makes it tempting to examineStallman's life outside the day-to-day battles of thesoftware industry and within a more august, historical setting. 

To his credit, Stallman refuses all opportunities tospeculate. "I've never been able to work out detailedplans of what the future was going to be like, " saysStallman, offering his own premature epitaph. "I justsaid `I'm going to fight. Who knows where I'll get?'"

There's no question that in picking his fights, Stallman has alienated the very people who mightotherwise have been his greatest champions. It is alsoa testament to his forthright, ethical nature that manyof Stallman's erstwhile political opponents stillmanage to put in a few good words for him when pressed. The tension between Stallman the ideologue and Stallmanthe hacker genius, however, leads a biographer towonder: how will people view Stallman when Stallman'sown personality is no longer there to get in the way?

In early drafts of this book, I dubbed this questionthe "100 year" question. Hoping to stimulate anobjective view of Stallman and his work, I askedvarious software-industry luminaries to take themselvesout of the current timeframe and put themselves in aposition of a historian looking back on the freesoftware movement 100 years in the future. From thecurrent vantage point, it is easy to see similaritiesbetween Stallman and past Americans who, while somewhatmarginal during their lifetime, have attainedheightened historical importance in relation to theirage. Easy comparisons include Henry David Thoreau, transcendentalist philosopher and author of On CivilDisobedience, and John Muir, founder of the Sierra Cluband progenitor of the modern environmental movement. Itis also easy to see similarities in men like WilliamJennings Bryan, a. K. A. "The Great Commoner, " leader ofthe populist movement, enemy of monopolies, and a manwho, though powerful, seems to have faded intohistorical insignificance. 

Although not the first person to view software aspublic property, Stallman is guaranteed a footnote infuture history books thanks to the GPL. Given thatfact, it seems worthwhile to step back and examineRichard Stallman's legacy outside the current timeframe. Will the GPL still be something softwareprogrammers use in the year 2102, or will it have longsince fallen by the wayside? Will the term "freesoftware" seem as politically quaint as "free silver"does today, or will it seem eerily prescient in lightof later political events?

Predicting the future is risky sport, but most people, when presented with the question, seemed eager to bite. "One hundred years from now, Richard and a couple ofother people are going to deserve more than afootnote, " says Moglen. "They're going to be viewed asthe main line of the story. "

The "couple other people" Moglen nominates for futuretextbook chapters include John Gilmore, Stallman's GPLadvisor and future founder of the Electronic FrontierFoundation, and Theodor Holm Nelson, a. K. A. Ted Nelson, author of the 1982 book, Literary Machines . Moglensays Stallman, Nelson, and Gilmore each stand out inhistorically significant, nonoverlapping ways. Hecredits Nelson, commonly considered to have coined theterm "hypertext, " for identifying the predicament ofinformation ownership in the digital age. Gilmore andStallman, meanwhile, earn notable credit foridentifying the negative political effects ofinformation control and building organizations-theElectronic Frontier Foundation in the case of Gilmoreand the Free Software Foundation in the case ofStallman-to counteract those effects. Of the two, however, Moglen sees Stallman's activities as morepersonal and less political in nature. 

"Richard was unique in that the ethical implications ofunfree software were particularly clear to him at anearly moment, " says Moglen. "This has a lot to do withRichard's personality, which lots of people will, whenwriting about him, try to depict as epiphenomenal oreven a drawback in Richard Stallman's own life work. "

Gilmore, who describes his inclusion between theerratic Nelson and the irascible Stallman as somethingof a "mixed honor, " nevertheless seconds the Moglenargument. Writes Gilmore: My guess is that Stallman'swritings will stand up as well as Thomas Jefferson'shave; he's a pretty clear writer and also clear on hisprinciples . . . Whether Richard will be as influentialas Jefferson will depend on whether the abstractions wecall "civil rights" end up more important a hundredyears from now than the abstractions that we call"software" or "technically imposed restrictions. "Another element of the Stallman legacy not to beoverlooked, Gilmore writes, is the collaborativesoftware-development model pioneered by the GNUProject. Although flawed at times, the model hasnevertheless evolved into a standard within thesoftware-development industry. All told, Gilmore says, this collaborative software-development model may endup being even more influential than the GNU Project, the GPL License, or any particular software programdeveloped by Stallman: Before the Internet, it wasquite hard to collaborate over distance on software, even among teams that know and trust each other. Richard pioneered collaborative development ofsoftware, particularly by disorganized volunteers whoseldom meet each other. Richard didn't build any of thebasic tools for doing this (the TCP protocol, emaillists, diff and patch, tar files, RCS or CVS orremote-CVS), but he used the ones that were availableto form social groups of programmers who couldeffectively collaborate. Lawrence Lessig, Stanford lawprofessor and author of the 2001 book, The Future ofIdeas, is similarly bullish. Like many legal scholars, Lessig sees the GPL as a major bulwark of the currentso-called "digital commons, " the vast agglomeration ofcommunity-owned software programs, network andtelecommunication standards that have triggered theInternet's exponential growth over the last threedecades. Rather than connect Stallman with otherInternet pioneers, men such as Vannevar Bush, VintonCerf, and J. C. R. Licklider who convinced others tosee computer technology on a wider scale, Lessig seesStallman's impact as more personal, introspective, and, ultimately, unique: [Stallman] changed the debate fromis to ought. He made people see how much was at stake, and he built a device to carry these ideals forward . .. That said, I don't quite know how to place him in thecontext of Cerf or Licklider. The innovation isdifferent. It is not just about a certain kind of code, or enabling the Internet. [It's] much more aboutgetting people to see the value in a certain kind ofInternet. I don't think there is anyone else in thatclass, before or after. Not everybody sees the Stallmanlegacy as set in stone, of course. Eric Raymond, theopen source proponent who feels that Stallman'sleadership role has diminished significantly since1996, sees mixed signals when looking into the 2102crystal ball: I think Stallman's artifacts (GPL, Emacs, GCC) will be seen as revolutionary works, asfoundation-stones of the information world. I thinkhistory will be less kind to some of the theories fromwhich RMS operated, and not kind at all to his personaltendency towards territorial, cult-leader behavior. Asfor Stallman himself, he, too, sees mixed signals: Whathistory says about the GNU Project, twenty years fromnow, will depend on who wins the battle of freedom touse public knowledge. If we lose, we will be just afootnote. If we win, it is uncertain whether peoplewill know the role of the GNU operating system-if theythink the system is "Linux, " they will build a falsepicture of what happened and why. 

But even if we win, what history people learn a hundredyears from now is likely to depend on who dominatespolitically. Searching for his own 19th-centuryhistorical analogy, Stallman summons the figure of JohnBrown, the militant abolitionist regarded as a hero onone side of the Mason Dixon line and a madman on the other. 

John Brown's slave revolt never got going, but duringhis subsequent trial he effectively roused nationaldemand for abolition. During the Civil War, John Brownwas a hero; 100 years after, and for much of the 1900s, history textbooks taught that he was crazy. During theera of legal segregation, while bigotry was shameless, the US partly accepted the story that the South wantedto tell about itself, and history textbooks said manyuntrue things about the Civil War and related events. 

Such comparisons document both the self-perceivedperipheral nature of Stallman's current work and thebinary nature of his current reputation. Although it'shard to see Stallman's reputation falling to the levelof infamy as Brown's did during the post-Reconstructionperiod-Stallman, despite his occasional war-likeanalogies, has done little to inspire violence-it'seasy to envision a future in which Stallman's ideaswind up on the ash-heap. In fashioning the freesoftware cause not as a mass movement but as acollection of private battles against the forces ofproprietary temptation, Stallman seems to have createda unwinnable situation, especially for the manyacolytes with the same stubborn will. 

Then again, it is that very will that may someday proveto be Stallman's greatest lasting legacy. Moglen, aclose observer over the last decade, warns those whomistake the Stallman personality as counter-productiveor epiphenomenal to the "artifacts" of Stalllman'slife. Without that personality, Moglen says, therewould be precious few artifiacts to discuss. SaysMoglen, a former Supreme Court clerk: Look, thegreatest man I ever worked for was Thurgood Marshall. Iknew what made him a great man. I knew why he had beenable to change the world in his possible way. I wouldbe going out on a limb a little bit if I were to make acomparison, because they could not be more different. Thurgood Marshall was a man in society, representing anoutcast society to the society that enclosed it, butstill a man in society. His skill was social skills. But he was all of a piece, too. Different as they werein every other respect, that the person I most nowcompare him to in that sense, all of a piece, compact, made of the substance that makes stars, all the waythrough, is Stallman. In an effort to drive that imagehome, Moglen reflects on a shared moment in the springof 2000. The success of the VA Linux IPO was stillresonating in the business media, and a half dozen freesoftware-related issues were swimming through the news. Surrounded by a swirling hurricane of issues andstories each begging for comment, Moglen recallssitting down for lunch with Stallman and feeling like acastaway dropped into the eye of the storm. For thenext hour, he says, the conversation calmly revolvedaround a single topic: strengthening the GPL. 

"We were sitting there talking about what we were goingto do about some problems in Eastern Europe and what wewere going to do when the problem of the ownership ofcontent began to threaten free software, " Moglenrecalls. "As we were talking, I briefly thought abouthow we must have looked to people passing by. Here weare, these two little bearded anarchists, plotting andplanning the next steps. And, of course, Richard isplucking the knots from his hair and dropping them inthe soup and behaving in his usual way. Anybodylistening in on our conversation would have thought wewere crazy, but I knew: I knew the revolution's righthere at this table. This is what's making it happen. And this man is the person making it happen. "

Moglen says that moment, more than any other, drovehome the elemental simplicity of the Stallman style. 

"It was funny, " recalls Moglen. "I said to him, `Richard, you know, you and I are the two guys whodidn't make any money out of this revolution. ' And thenI paid for the lunch, because I knew he didn't have themoney to pay for it . '" Endnote

Epilogue:

Crushing Loneliness Writing the biography of a livingperson is a bit like producing a play. The drama infront of the curtain often pales in comparison to thedrama backstage. 

In The Autobiography of Malcolm X, Alex Haley givesreaders a rare glimpse of that backstage drama. Stepping out of the ghostwriter role, Haley deliversthe book's epilogue in his own voice. The epilogueexplains how a freelance reporter originally dismissedas a "tool" and "spy" by the Nation of Islamspokesperson managed to work through personal andpolitical barriers to get Malcolm X's life story on paper. 

While I hesitate to compare this book with TheAutobiography of Malcolm X, I do owe a debt ofgratitude to Haley for his candid epilogue. Over thelast 12 months, it has served as a sort of instructionmanual on how to deal with a biographical subject whohas built an entire career on being disagreeable. Fromthe outset, I envisioned closing this biography with asimilar epilogue, both as an homage to Haley and as away to let readers know how this book came to be. 

The story behind this story starts in an Oaklandapartment, winding its way through the various localesmentioned in the book-Silicon Valley, Maui, Boston, andCambridge. Ultimately, however, it is a tale of twocities: New York, New York, the book-publishing capitalof the world, and Sebastopol, California, thebook-publishing capital of Sonoma County. 

The story starts in April, 2000. At the time, I waswriting stories for the ill-fated BeOpen web site(http://www. Beopen. Com/). One of my first assignmentswas a phone interview with Richard M. Stallman. Theinterview went well, so well that Slashdot(http://www. Slashdot. Org/), the popular "news fornerds" site owned by VA Software, Inc. (formerly VALinux Systems and before that, VA Research), gave it alink in its daily list of feature stories. Withinhours, the web servers at BeOpen were heating up asreaders clicked over to the site. 

For all intents and purposes, the story should haveended there. Three months after the interview, whileattending the O'Reilly Open Source Conference inMonterey, California, I received the following emailmessage from Tracy Pattison, foreign-rights manager ata large New York publishing house:

To: sam@BeOpen. Com Subject: RMS InterviewDate: Mon, 10Jul 2000 15:56:37 -0400Dear Mr. Williams, I read your interview with Richard Stallman on BeOpenwith great interest. I've been intrigued by RMS and hiswork for some time now and was delighted to find yourpiece which I really think you did a great job ofcapturing some of the spirit of what Stallman is tryingto do with GNU-Linux and the Free Software Foundation. 

What I'd love to do, however, is read more - and Idon't think I'm alone. Do you think there is moreinformation and/or sources out there to expand andupdate your interview and adapt it into more of aprofile of Stallman? Perhaps including some moreanecdotal information about his personality andbackground that might really interest and enlightenreaders outside the more hardcore programming scene?

The email asked that I give Tracy a call to discuss theidea further. I did just that. Tracy told me hercompany was launching a new electronic book line, andit wanted stories that appealed to an early-adopteraudience. The e-book format was 30, 000 words, about 100pages, and she had pitched her bosses on the idea ofprofiling a major figure in the hacker community. Herbosses liked the idea, and in the process of searchingfor interesting people to profile, she had come acrossmy BeOpen interview with Stallman. Hence her email to me. 

That's when Tracy asked me: would I be willing toexpand the interview into a full-length feature profile?

My answer was instant: yes. Before accepting it, Tracysuggested I put together a story proposal she couldshow her superiors. Two days later, I sent her apolished proposal. A week later, Tracy sent me a followup email. Her bosses had given it the green light. 

I have to admit, getting Stallman to participate in ane-book project was an afterthought on my part. As areporter who covered the open source beat, I knewStallman was a stickler. I'd already received a halfdozen emails at that point upbraiding me for the use of"Linux" instead of "GNU/Linux. "

Then again, I also knew Stallman was looking for waysto get his message out to the general public. Perhapsif I presented the project to him that way, he would bemore receptive. If not, I could always rely upon thecopious amounts of documents, interviews, and recordedonline conversations Stallman had left lying around theInternet and do an unauthorized biography. 

During my research, I came across an essay titled"Freedom-Or Copyright?" Written by Stallman andpublished in the June, 2000, edition of the MITTechnology Review, the essay blasted e-books for anassortment of software sins. Not only did readers haveto use proprietary software programs to read them, Stallman lamented, but the methods used to preventunauthorized copying were overly harsh. Instead ofdownloading a transferable HTML or PDF file, readersdownloaded an encrypted file. In essence, purchasing ane-book meant purchasing a nontransferable key tounscramble the encrypted content. Any attempt to open abook's content without an authorized key constituted acriminal violation of the Digital Millennium CopyrightAct, the 1998 law designed to bolster copyrightenforcement on the Internet. Similar penalties held forreaders who converted a book's content into an openfile format, even if their only intention was to readthe book on a different computer in their home. Unlikea normal book, the reader no longer held the right tolend, copy, or resell an e-book. They only had theright to read it on an authorized machine, warnedStallman: We still have the same old freedoms in usingpaper books. But if e-books replace printed books, thatexception will do little good. With "electronic ink, "which makes it possible to download new text onto anapparently printed piece of paper, even newspaperscould become ephemeral. Imagine: no more used bookstores; no more lending a book to your friend; no moreborrowing one from the public library-no more "leaks"that might give someone a chance to read withoutpaying. (And judging from the ads for Microsoft Reader, no more anonymous purchasing of books either. ) This isthe world publishers have in mind for us. See "Safari Tech Books Online;Subscriber Agreement:Terms of Service. "

http://safari. Oreilly. Com/mainhlp. Asp?help=serviceNeedless to say, the essay caused some concern. NeitherTracy nor I had discussed the software her companywould use nor had we discussed the type of copyrightthat would govern the e-book's usage. I mentioned theTechnology Review article and asked if she could giveme information on her company's e-book policies. Tracypromised to get back to me. 

Eager to get started, I decided to call Stallman anywayand mention the book idea to him. When I did, heexpressed immediate interest and immediate concern. "Did you read my essay on e-books?" he asked. 

When I told him, yes, I had read the essay and waswaiting to hear back from the publisher, Stallman laidout two conditions: he didn't want to lend support toan e-book licensing mechanism he fundamentally opposed, and he didn't want to come off as lending support. "Idon't want to participate in anything that makes melook like a hypocrite, " he said. 

For Stallman, the software issue was secondary to thecopyright issue. He said he was willing to ignorewhatever software the publisher or its third-partyvendors employed just so long as the company specifiedwithin the copyright that readers were free to make anddistribute verbatim copies of the e-book's content. Stallman pointed to Stephen King's The Plant as apossible model. In June, 2000, King announced on hisofficial web site that he was self-publishing The Plantin serial form. According to the announcement, thebook's total cost would be $13, spread out over aseries of $1 installments. As long as at least 75% ofthe readers paid for each chapter, King promised tocontinue releasing new installments. By August, theplan seemed to be working, as King had published thefirst two chapters with a third on the way. 

"I'd be willing to accept something like that, "Stallman said. "As long as it also permitted verbatim copying. "

I forwarded the information to Tracy. Feeling confidentthat she and I might be able to work out an equitablearrangement, I called up Stallman and set up the firstinterview for the book. Stallman agreed to theinterview without making a second inquiry into thestatus issue. Shortly after the first interview, Iraced to set up a second interview (this one in Kihei), squeezing it in before Stallman headed off on a 14-dayvacation to Tahiti. 

It was during Stallman's vacation that the bad newscame from Tracy. Her company's legal department didn'twant to adjust its copyright notice on the e-books. Readers who wanted to make their books transferablewould either have to crack the encryption code orconvert the book to an open format such as HTML. Eitherway, the would be breaking the law and facing criminal penalties. 

With two fresh interviews under my belt, I didn't seeany way to write the book without resorting to the newmaterial. I quickly set up a trip to New York to meetwith my agent and with Tracy to see if there was acompromise solution. 

When I flew to New York, I met my agent, HenningGuttman. It was our first face-to-face meeting, andHenning seemed pessimistic about our chances of forcinga compromise, at least on the publisher's end. Thelarge, established publishing houses already viewed thee-book format with enough suspicion and weren't in themood to experiment with copyright language that made iteasier for readers to avoid payment. As an agent whospecialized in technology books, however, Henning wasintrigued by the novel nature of my predicament. I toldhim about the two interviews I'd already gathered andthe promise not to publish the book in a way that madeStallman "look like a hypocrite. " Agreeing that I wasin an ethical bind, Henning suggested we make that ournegotiating point. 

Barring that, Henning said, we could always take thecarrot-and-stick approach. The carrot would be thepublicity that came with publishing an e-book thathonored the hacker community's internal ethics. Thestick would be the risks associated with publishing ane-book that didn't. Nine months before Dmitri Skylarovbecame an Internet cause celebre, we knew it was only amatter of time before an enterprising programmerrevealed how to hack e-books. We also knew that a majorpublishing house releasing an encryption-protectede-book on Richard M. Stallman was the softwareequivalent of putting "Steal This E-Book" on the cover. 

After my meeting with Henning, I put a call intoStallman. Hoping to make the carrot more enticing, Idiscussed a number of potential compromises. What ifthe publisher released the book's content under a splitlicense, something similar to what Sun Microsystems haddone with Open Office, the free software desktopapplications suite? The publisher could then releasecommercial versions of the e-book under a normalformat, taking advantage of all the bells and whistlesthat went with the e-book software, while releasing thecopyable version under a less aesthetically pleasingHTML format. 

Stallman told me he didn't mind the split-license idea, but he did dislike the idea of making the freelycopyable version inferior to the restricted version. Besides, he said, the idea was too cumbersome. Splitlicenses worked in the case of Sun's Open Office onlybecause he had no control over the decision making. Inthis case, Stallman said, he did have a way to controlthe outcome. He could refuse to cooperate. 

I made a few more suggestions with little effect. Aboutthe only thing I could get out of Stallman was aconcession that the e-book's copyright restrict allforms of file sharing to "noncommercial redistribution. "

Before I signed off, Stallman suggested I tell thepublisher that I'd promised Stallman that the workwould be free. I told Stallman I couldn't agree to thatstatement but that I did view the book as unfinishablewithout his cooperation. Seemingly satisfied, Stallmanhung up with his usual sign-off line: "Happy hacking. "

Henning and I met with Tracy the next day. Tracy saidher company was willing to publish copyable excerpts ina unencrypted format but would limit the excerpts to500 words. Henning informed her that this wouldn't beenough for me to get around my ethical obligation toStallman. Tracy mentioned her own company's contractualobligation to online vendors such as Amazon. Com. Evenif the company decided to open up its e-book contentthis one time, it faced the risk of its partnerscalling it a breach of contract. Barring a change ofheart in the executive suite or on the part ofStallman, the decision was up to me. I could use theinterviews and go against my earlier agreement withStallman, or I could plead journalistic ethics and backout of the verbal agreement to do the book. 

Following the meeting, my agent and I relocated to apub on Third Ave. I used his cell phone to callStallman, leaving a message when nobody answered. Henning left for a moment, giving me time to collect mythoughts. When he returned, he was holding up the cell phone. 

"It's Stallman, " Henning said. 

The conversation got off badly from the start. Irelayed Tracy's comment about the publisher'scontractual obligations. 

"So, " Stallman said bluntly. "Why should I give a damnabout their contractual obligations?"

Because asking a major publishing house to risk a legalbattle with its vendors over a 30, 000 word e-book is atall order, I suggested. 

"Don't you see?" Stallman said. "That's exactly why I'mdoing this. I want a signal victory. I want them tomake a choice between freedom and business as usual. "

As the words "signal victory" echoed in my head, I feltmy attention wander momentarily to the passing foottraffic on the sidewalk. Coming into the bar, I hadbeen pleased to notice that the location was less thanhalf a block away from the street corner memorializedin the 1976 Ramones song, "53rd and 3rd, " a song Ialways enjoyed playing in my days as a musician. Likethe perpetually frustrated street hustler depicted inthat song, I could feel things falling apart as quicklyas they had come together. The irony was palpable. After weeks of gleefully recording other people'slaments, I found myself in the position of trying topull off the rarest of feats: a Richard Stallman compromise. 

When I continued hemming and hawing, pleading thepublisher's position and revealing my growing sympathyfor it, Stallman, like an animal smelling blood, attacked. 

"So that's it? You're just going to screw me? You'rejust going to bend to their will?"

I brought up the issue of a dual-copyright again. 

"You mean license, " Stallman said curtly. 

"Yeah, license. Copyright. Whatever, " I said, feelingsuddenly like a wounded tuna trailing a rich plume ofplasma in the water. 

"Aw, why didn't you just fucking do what I told you todo!" he shouted. 

I must have been arguing on behalf of the publisher tothe very end, because in my notes I managed to save afinal Stallman chestnut: "I don't care. What they'redoing is evil. I can't support evil. Good-bye. "

As soon as I put the phone down, my agent slid afreshly poured Guinness to me. "I figured you mightneed this, " he said with a laugh. "I could see youshaking there towards the end. "

I was indeed shaking. The shaking wouldn't stop untilthe Guinness was more than halfway gone. It felt weird, hearing myself characterized as an emissary of "evil. "It felt weirder still, knowing that three monthsbefore, I was sitting in an Oakland apartment trying tocome up with my next story idea. Now, I was sitting ina part of the world I'd only known through rock songs, taking meetings with publishing executives and drinkingbeer with an agent I'd never even laid eyes on untilthe day before. It was all too surreal, like watchingmy life reflected back as a movie montage. 

About that time, my internal absurdity meter kicked in. The initial shaking gave way to convulsions oflaughter. To my agent, I must have looked like aanother fragile author undergoing an untimely emotionalbreakdown. To me, I was just starting to appreciate thecynical beauty of my situation. Deal or no deal, Ialready had the makings of a pretty good story. It wasonly a matter of finding a place to tell it. When mylaughing convulsions finally subsided, I held up mydrink in a toast. 

"Welcome to the front lines, my friend, " I said, clinking pints with my agent. "Might as well enjoy it. "

If this story really were a play, here's where it wouldtake a momentary, romantic interlude. Disheartened bythe tense nature of our meeting, Tracy invited Henningand I to go out for drinks with her and some of hercoworkers. We left the bar on Third Ave. , headed downto the East Village, and caught up with Tracy and her friends. 

Once there, I spoke with Tracy, careful to avoid shoptalk. Our conversation was pleasant, relaxed. Beforeparting, we agreed to meet the next night. Once again, the conversation was pleasant, so pleasant that theStallman e-book became almost a distant memory. 

When I got back to Oakland, I called around to variousjournalist friends and acquaintances. I recounted mypredicament. Most upbraided me for giving up too muchground to Stallman in the preinterview negotiation. Aformer j-school professor suggested I ignore Stallman's"hypocrite" comment and just write the story. Reporterswho knew of Stallman's media-savviness expressedsympathy but uniformly offered the same response: it'syour call. 

I decided to put the book on the back burner. Even withthe interviews, I wasn't making much progress. Besides, it gave me a chance to speak with Tracy without runningthings past Henning first. By Christmas we had tradedvisits: she flying out to the west coast once, meflying out to New York a second time. The day beforeNew Year's Eve, I proposed. Deciding which coast tolive on, I picked New York. By February, I packed up mylaptop computer and all my research notes related tothe Stallman biography, and we winged our way to JFKAirport. Tracy and I were married on May 11. So muchfor failed book deals. 

During the summer, I began to contemplate turning myinterview notes into a magazine article. Ethically, Ifelt in the clear doing so, since the originalinterview terms said nothing about traditional printmedia. To be honest, I also felt a bit more comfortablewriting about Stallman after eight months of radiosilence. Since our telephone conversation in September, I'd only received two emails from Stallman. Bothchastised me for using "Linux" instead of "GNU/Linux"in a pair of articles for the web magazine UpsideToday. Aside from that, I had enjoyed the silence. InJune, about a week after the New York Universityspeech, I took a crack at writing a 5, 000-wordmagazine-length story about Stallman. This time, thewords flowed. The distance had helped restore my lostsense of emotional perspective, I suppose. 

In July, a full year after the original email fromTracy, I got a call from Henning. He told me thatO'Reilly & Associates, a publishing house out ofSebastopol, California, was interested in the runningthe Stallman story as a biography. The news pleased me. Of all the publishing houses in the world, O'Reilly, the same company that had published Eric Raymond's TheCathedral and the Bazaar, seemed the most sensitive tothe issues that had killed the earlier e-book. As areporter, I had relied heavily on the O'Reilly bookOpen Sources as a historical reference. I also knewthat various chapters of the book, including a chapterwritten by Stallman, had been published with copyrightnotices that permitted redistribution. Such knowledgewould come in handy if the issue of electronicpublication ever came up again. 

Sure enough, the issue did come up. I learned throughHenning that O'Reilly intended to publish the biographyboth as a book and as part of its new Safari Tech BooksOnline subscription service. The Safari user licensewould involve special restrictions, 1 Henning warned, but O'Reilly was willing to allow for a copyright thatpermitted users to copy and share and the book's textregardless of medium. Basically, as author, I had thechoice between two licenses: the Open PublicationLicense or the GNU Free Documentation License. 

I checked out the contents and background of eachlicense. The Open Publication License (OPL)See "The Open Publication License:Draft v1. 0" (June 8, 1999). 

http://opencontent. Org/openpub/ gives readers the right to reproduce and distribute awork, in whole or in part, in any medium "physical orelectronic, " provided the copied work retains the OpenPublication License. It also permits modification of awork, provided certain conditions are met. Finally, theOpen Publication License includes a number of options, which, if selected by the author, can limit thecreation of "substantively modified" versions orbook-form derivatives without prior author approval. 

The GNU Free Documentation License (GFDL), See "The GNU Free DocumentationLicense: Version 1. 1"(March, 2000). 

http://www. Gnu. Org/copyleft/fdl. Html meanwhile, permits the copying and distribution of adocument in any medium, provided the resulting workcarries the same license. It also permits themodification of a document provided certain conditions. Unlike the OPL, however, it does not give authors theoption to restrict certain modifications. It also doesnot give authors the right to reject modifications thatmight result in a competitive book product. It doesrequire certain forms of front- and back-coverinformation if a party other than the copyright holderwishes to publish more than 100 copies of a protectedwork, however. 

In the course of researching the licenses, I also madesure to visit the GNU Project web page titled "VariousLicenses and Comments About Them. "Seehttp://www. Gnu. Org/philosophy/license-list. Html On that page, Ifound a Stallman critique of the Open PublicationLicense. Stallman's critique related to the creation ofmodified works and the ability of an author to selecteither one of the OPL's options to restrictmodification. If an author didn't want to select eitheroption, it was better to use the GFDL instead, Stallmannoted, since it minimized the risk of the nonselectedoptions popping up in modified versions of a document. 

The importance of modification in both licenses was areflection of their original purpose-namely, to givesoftware-manual owners a chance to improve theirmanuals and publicize those improvements to the rest ofthe community. Since my book wasn't a manual, I hadlittle concern about the modification clause in eitherlicense. My only concern was giving users the freedomto exchange copies of the book or make copies of thecontent, the same freedom they would have enjoyed ifthey purchased a hardcover book. Deeming either licensesuitable for this purpose, I signed the O'Reillycontract when it came to me. 

Still, the notion of unrestricted modificationintrigued me. In my early negotiations with Tracy, Ihad pitched the merits of a GPL-style license for thee-book's content. At worst, I said, the license wouldguarantee a lot of positive publicity for the e-book. At best, it would encourage readers to participate inthe book-writing process. As an author, I was willingto let other people amend my work just so long as myname always got top billing. Besides, it might even beinteresting to watch the book evolve. I pictured latereditions looking much like online versions of theTalmud, my original text in a central column surroundedby illuminating, third-party commentary in the margins. 

My idea drew inspiration from Project Xanadu(http://www. Xanadu. Com/), the legendary softwareconcept originally conceived by Ted Nelson in 1960. During the O'Reilly Open Source Conference in 1999, Ihad seen the first demonstration of the project's opensource offshoot Udanax and had been wowed by theresult. In one demonstration sequence, Udanax displayeda parent document and a derivative work in a similartwo-column, plain-text format. With a click of thebutton, the program introduced lines linking eachsentence in the parent to its conceptual offshoot inthe derivative. An e-book biography of Richard M. Stallman didn't have to be Udanax-enabled, but givensuch technological possibilities, why not give users achance to play around?Anybody willing to "port" this book over to Udanax, thefree software version of Xanadu, will receiveenthusiastic support from me. To find out more aboutthis intriguing technology, visit

http://www. Udanax. Com/. 

When Laurie Petrycki, my editor at O'Reilly, gave me achoice between the OPL or the GFDL, I indulged thefantasy once again. By September of 2001, the month Isigned the contract, e-books had become almost a deadtopic. Many publishing houses, Tracy's included, wereshutting down their e-book imprints for lack ofinterest. I had to wonder. If these companies hadtreated e-books not as a form of publication but as aform of community building, would those imprints have survived?

After I signed the contract, I notified Stallman thatthe book project was back on. I mentioned the choiceO'Reilly was giving me between the Open PublicationLicense and the GNU Free Documentation License. I toldhim I was leaning toward the OPL, if only for the factI saw no reason to give O'Reilly's competitors a chanceto print the same book under a different cover. Stallman wrote back, arguing in favor of the GFDL, noting that O'Reilly had already used it several timesin the past. Despite the events of the past year, Isuggested a deal. I would choose the GFDL if it gave methe possibility to do more interviews and if Stallmanagreed to help O'Reilly publicize the book. Stallmanagreed to participate in more interviews but said thathis participation in publicity-related events woulddepend on the content of the book. Viewing this as onlyfair, I set up an interview for December 17, 2001 in Cambridge. 

I set up the interview to coincide with a business tripmy wife Tracy was taking to Boston. Two days beforeleaving, Tracy suggested I invite Stallman out to dinner. 

"After all, " she said, "he is the one who brought us together. "

I sent an email to Stallman, who promptly sent a returnemail accepting the offer. When I drove up to Bostonthe next day, I met Tracy at her hotel and hopped the Tto head over to MIT. When we got to Tech Square, Ifound Stallman in the middle of a conversation just aswe knocked on the door. 

"I hope you don't mind, " he said, pulling the door openfar enough so that Tracy and I could just barely hearStallman's conversational counterpart. It was ayoungish woman, mid-20s I'd say, named Sarah. 

"I took the liberty of inviting somebody else to havedinner with us, " Stallman said, matter-of-factly, giving me the same cat-like smile he gave me back inthat Palo Alto restaurant. 

To be honest, I wasn't too surprised. The news thatStallman had a new female friend had reached me a fewweeks before, courtesy of Stallman's mother. "In fact, they both went to Japan last month when Richard wentover to accept the Takeda Award, " Lippman told me atthe time. Alas, I didn't find out about the Takeda Foundation'sdecision to award Stallman, along with Linus Torvaldsand Ken Sakamura, with its first-ever award for"Techno-Entrepreneurial Achievement for Social/EconomicWell-Being" until after Stallman had made the trip toJapan to accept the award. For more information aboutthe award and its accompanying $1 million prize, visitthe Takeda site, http://www. Takeda-foundation. Jp/. 

On the way over to the restaurant, I learned thecircumstances of Sarah and Richard's first meeting. Interestingly, the circumstances were very familiar. Working on her own fictional book, Sarah said she heardabout Stallman and what an interesting character hewas. She promptly decided to create a character in herbook on Stallman and, in the interests of researchingthe character, set up an interview with Stallman. Things quickly went from there. The two had been datingsince the beginning of 2001, she said. 

"I really admired the way Richard built up an entirepolitical movement to address an issue of profoundpersonal concern, " Sarah said, explaining herattraction to Stallman. 

My wife immediately threw back the question: "What wasthe issue?"

"Crushing loneliness. "

During dinner, I let the women do the talking and spentmost of the time trying to detect clues as to whetherthe last 12 months had softened Stallman in anysignificant way. I didn't see anything to suggest theyhad. Although more flirtatious than I remembered-aflirtatiousness spoiled somewhat by the number of timesStallman's eyes seemed to fixate on my wife'schest-Stallman retained the same general level ofprickliness. At one point, my wife uttered an emphatic"God forbid" only to receive a typical Stallman rebuke. 

"I hate to break it to you, but there is no God, "Stallman said. 

Afterwards, when the dinner was complete and Sarah haddeparted, Stallman seemed to let his guard down alittle. As we walked to a nearby bookstore, he admittedthat the last 12 months had dramatically changed hisoutlook on life. "I thought I was going to be aloneforever, " he said. "I'm glad I was wrong. "

Before parting, Stallman handed me his "pleasure card, "a business card listing Stallman's address, phonenumber, and favorite pastimes ("sharing good books, good food and exotic music and dance") so that I mightset up a final interview. 

<Graphic file:/home/craigm/books/free_ep10. Png>

Stallman's "pleasure" card, handed to me the night ofour dinner. 

The next day, over another meal of dim sum, Stallmanseemed even more lovestruck than the night before. Recalling his debates with Currier House dorm matersover the benefits and drawbacks of an immortalityserum, Stallman expressed hope that scientists mightsome day come up with the key to immortality. "Now thatI'm finally starting to have happiness in my life, Iwant to have more, " he said. 

When I mentioned Sarah's "crushing loneliness" comment, Stallman failed to see a connection between lonelinesson a physical or spiritual level and loneliness on ahacker level. "The impulse to share code is aboutfriendship but friendship at a much lower level, " hesaid. Later, however, when the subject came up again, Stallman did admit that loneliness, or the fear ofperpetual loneliness, had played a major role infueling his determination during the earliest days ofthe GNU Project. 

"My fascination with computers was not a consequence ofanything else, " he said. "I wouldn't have been lessfascinated with computers if I had been popular and allthe women flocked to me. However, it's certainly truethe experience of feeling I didn't have a home, findingone and losing it, finding another and having itdestroyed, affected me deeply. The one I lost was thedorm. The one that was destroyed was the AI Lab. Theprecariousness of not having any kind of home orcommunity was very powerful. It made me want to fightto get it back. "

After the interview, I couldn't help but feel a certainsense of emotional symmetry. Hearing Sarah describewhat attracted her to Stallman and hearing Stallmanhimself describe the emotions that prompted him to takeup the free software cause, I was reminded of my ownreasons for writing this book. Since July, 2000, I havelearned to appreciate both the seductive and therepellent sides of the Richard Stallman persona. LikeEben Moglen before me, I feel that dismissing thatpersona as epiphenomenal or distracting in relation tothe overall free software movement would be a grievousmistake. In many ways the two are so mutually definingas to be indistinguishable. 

While I'm sure not every reader feels the same level ofaffinity for Stallman-indeed, after reading this book, some might feel zero affinity-I'm sure most will agree. Few individuals offer as singular a human portrait asRichard M. Stallman. It is my sincere hope that, withthis initial portrait complete and with the help of theGFDL, others will feel a similar urge to add their ownperspective to that portrait. 

Appendix A : Terminology

For the most part, I have chosen to use the termGNU/Linux in reference to the free software operatingsystem and Linux when referring specifically to thekernel that drives the operating system. The mostnotable exception to this rule comes in Chapter 9 . Inthe final part of that chapter, I describe the earlyevolution of Linux as an offshoot of Minix. It is safeto say that during the first two years of the project'sdevelopment, the operating system Torvalds and hiscolleagues were working on bore little similarity tothe GNU system envisioned by Stallman, even though itgradually began to share key components, such as theGNU C Compiler and the GNU Debugger. 

This decision further benefits from the fact that, prior to 1993, Stallman saw little need to insist on credit. 

Some might view the decision to use GNU/Linux for laterversions of the same operating system as arbitrary. Iwould like to point out that it was in no way aprerequisite for gaining Stallman's cooperation in themaking of this book. I came to it of my own accord, partly because of the operating system's modular natureand the community surrounding it, and partly because ofthe apolitical nature of the Linux name. Given thatthis is a biography of Richard Stallman, it seemedinappropriate to define the operating system inapolitical terms. 

In the final phases of the book, when it became clearthat O'Reilly & Associates would be the book'spublisher, Stallman did make it a condition that I use"GNU/Linux" instead of Linux if O'Reilly expected himto provide promotional support for the book afterpublication. When informed of this, I relayed myearlier decision and left it up to Stallman to judgewhether the resulting book met this condition or not. At the time of this writing, I have no idea whatStallman's judgment will be. 

A similar situation surrounds the terms "free software"and "open source. " Again, I have opted for the morepolitically laden "free software" term when describingsoftware programs that come with freely copyable andfreely modifiable source code. Although more popular, Ihave chosen to use the term "open source" only whenreferring to groups and businesses that have championedits usage. But for a few instances, the terms arecompletely interchangeable, and in making this decisionI have followed the advice of Christine Peterson, theperson generally credited with coining the term. "The`free software' term should still be used incircumstances where it works better, " Peterson writes. "[`Open source'] caught on mainly because a new termwas greatly needed, not because it's ideal. "

Appendix B Hack, Hackers, and Hacking

To understand the full meaning of the word " hacker, "it helps to examine the word's etymology over the years. 

The New Hacker Dictionary, an online compendium ofsoftware-programmer jargon, officially lists ninedifferent connotations of the word "hack" and a similarnumber for "hacker. " Then again, the same publicationalso includes an accompanying essay that quotes PhilAgre, an MIT hacker who warns readers not to be fooledby the word's perceived flexibility. "Hack has only onemeaning, " argues Agre. "An extremely subtle andprofound one which defies articulation. "

Regardless of the width or narrowness of thedefinition, most modern hackers trace the word back toMIT, where the term bubbled up as popular item ofstudent jargon in the early 1950s. In 1990 the MITMuseum put together a journal documenting the hackingphenomenon. According to the journal, students whoattended the institute during the fifties used the word"hack" the way a modern student might use the word"goof. " Hanging a jalopy out a dormitory window was a"hack, " but anything harsh or malicious-e. G. , egging arival dorm's windows or defacing a campus statue-felloutside the bounds. Implicit within the definition of"hack" was a spirit of harmless, creative fun. 

This spirit would inspire the word's gerund form:"hacking. " A 1950s student who spent the better part ofthe afternoon talking on the phone or dismantling aradio might describe the activity as "hacking. " Again, a modern speaker would substitute the verb form of"goof"-"goofing" or "goofing off"-to describe the same activity. 

As the 1950s progressed, the word "hack" acquired asharper, more rebellious edge. The MIT of the 1950s wasoverly competitive, and hacking emerged as both areaction to and extension of that competitive culture. Goofs and pranks suddenly became a way to blow offsteam, thumb one's nose at campus administration, andindulge creative thinking and behavior stifled by theInstitute's rigorous undergraduate curriculum. With itsmyriad hallways and underground steam tunnels, theInstitute offered plenty of exploration opportunitiesfor the student undaunted by locked doors and "NoTrespassing" signs. Students began to refer to theiroff-limits explorations as "tunnel hacking. " Aboveground, the campus phone system offered similaropportunities. Through casual experimentation and duediligence, students learned how to perform humoroustricks. Drawing inspiration from the more traditionalpursuit of tunnel hacking, students quickly dubbed thisnew activity "phone hacking. "

The combined emphasis on creative play andrestriction-free exploration would serve as the basisfor the future mutations of the hacking term. The firstself-described computer hackers of the 1960s MIT campusoriginated from a late 1950s student group called theTech Model Railroad Club. A tight clique within theclub was the Signals and Power (S&P) Committee-thegroup behind the railroad club's electrical circuitrysystem. The system was a sophisticated assortment ofrelays and switches similar to the kind that controlledthe local campus phone system. To control it, a memberof the group simply dialed in commands via a connectedphone and watched the trains do his bidding. 

The nascent electrical engineers responsible forbuilding and maintaining this system saw their activityas similar in spirit to phone hacking. Adopting thehacking term, they began refining it even further. Fromthe S&P hacker point of view, using one less relay tooperate a particular stretch of track meant having onemore relay for future play. Hacking subtly shifted froma synonym for idle play to a synonym for idle play thatimproved the overall performance or efficiency of theclub's railroad system at the same time. Soon S&Pcommittee members proudly referred to the entireactivity of improving and reshaping the track'sunderlying circuitry as "hacking" and to the people whodid it as "hackers. "

Given their affinity for sophisticated electronics-notto mention the traditional MIT-student disregard forclosed doors and "No Trespassing" signs-it didn't takelong before the hackers caught wind of a new machine oncampus. Dubbed the TX-0, the machine was one of thefirst commercially marketed computers. By the end ofthe 1950s, the entire S&P clique had migrated en masseover to the TX-0 control room, bringing the spirit ofcreative play with them. The wide-open realm ofcomputer programming would encourage yet anothermutation in etymology. "To hack" no longer meantsoldering unusual looking circuits, but cobblingtogether software programs with little regard to"official" methods or software-writing procedures. Italso meant improving the efficiency and speed ofalready-existing programs that tended to hog up machineresources. True to the word's roots, it also meantwriting programs that served no other purpose than toamuse or entertain. 

A classic example of this expanded hacking definitionis the game Spacewar, the first interactive video game. Developed by MIT hackers in the early 1960s, Spacewarhad all the traditional hacking definitions: it wasgoofy and random, serving little useful purpose otherthan providing a nightly distraction for the dozen orso hackers who delighted in playing it. From a softwareperspective, however, it was a monumental testament toinnovation of programming skill. It was also completelyfree. Because hackers had built it for fun, they saw noreason to guard their creation, sharing it extensivelywith other programmers. By the end of the 1960s, Spacewar had become a favorite diversion for mainframeprogrammers around the world. 

This notion of collective innovation and communalsoftware ownership distanced the act of computerhacking in the 1960s from the tunnel hacking and phonehacking of the 1950s. The latter pursuits tended to besolo or small-group activities. Tunnel and phonehackers relied heavily on campus lore, but theoff-limits nature of their activity discouraged theopen circulation of new discoveries. Computer hackers, on the other hand, did their work amid a scientificfield biased toward collaboration and the rewarding ofinnovation. Hackers and "official" computer scientistsweren't always the best of allies, but in the rapidevolution of the field, the two species of computerprogrammer evolved a cooperative-some might saysymbiotic-relationship. 

It is a testament to the original computer hackers'prodigious skill that later programmers, includingRichard M. Stallman, aspired to wear the same hackermantle. By the mid to late 1970s, the term "hacker" hadacquired elite connotations. In a general sense, acomputer hacker was any person who wrote software codefor the sake of writing software code. In theparticular sense, however, it was a testament toprogramming skill. Like the term "artist, " the meaningcarried tribal overtones. To describe a fellowprogrammer as hacker was a sign of respect. To describeoneself as a hacker was a sign of immense personalconfidence. Either way, the original looseness of thecomputer-hacker appellation diminished as computersbecame more common. 

As the definition tightened, "computer" hackingacquired additional semantic overtones. To be a hacker, a person had to do more than write interestingsoftware; a person had to belong to the hacker"culture" and honor its traditions the same way amedieval wine maker might pledge membership to avintners' guild. The social structure wasn't as rigidlyoutlined as that of a guild, but hackers at eliteinstitutions such as MIT, Stanford, and Carnegie Mellonbegan to speak openly of a "hacker ethic": theyet-unwritten rules that governed a hacker's day-to-daybehavior. In the 1984 book Hackers, author Steven Levy, after much research and consultation, codified thehacker ethic as five core hacker tenets. 

In many ways, the core tenets listed by Levy continueto define the culture of computer hacking. Still, theguild-like image of the hacker community was underminedby the overwhelmingly populist bias of the softwareindustry. By the early 1980s, computers were popping upeverywhere, and programmers who once would have had totravel to top-rank institutions or businesses just togain access to a machine suddenly had the ability torub elbows with major-league hackers via the ARPAnet. The more these programmers rubbed elbows, the more theybegan to appropriate the anarchic philosophies of thehacker culture in places like MIT. Lost within thecultural transfer, however, was the native MIT culturaltaboo against malicious behavior. As youngerprogrammers began employing their computer skills toharmful ends-creating and disseminating computerviruses, breaking into military computer systems, deliberately causing machines such as MIT Oz, a popularARPAnet gateway, to crash-the term "hacker" acquired apunk, nihilistic edge. When police and businesses begantracing computer-related crimes back to a few renegadeprogrammers who cited convenient portions of thehacking ethic in defense of their activities, the word"hacker" began appearing in newspapers and magazinestories in a negative light. Although books likeHackers did much to document the original spirit ofexploration that gave rise to the hacking culture, formost news reporters, "computer hacker" became a synonymfor "electronic burglar. "

Although hackers have railed against this perceivedmisusage for nearly two decades, the term's rebelliousconnotations dating back to the 1950s make it hard todiscern the 15-year-old writing software programs thatcircumvent modern encryption programs from the 1960scollege student, picking locks and battering down doorsto gain access to the lone, office computer terminal. One person's creative subversion of authority isanother person's security headache, after all. Even so, the central taboo against malicious or deliberatelyharmful behavior remains strong enough that mosthackers prefer to use the term " cracker"-i. E. , aperson who deliberately cracks a computer securitysystem to steal or vandalize data-to describe thesubset of hackers who apply their computing skills maliciously. 

This central taboo against maliciousness remains theprimary cultural link between the notion of hacking inthe early 21st century and hacking in the 1950s. It isimportant to note that, as the idea of computer hackinghas evolved over the last four decades, the originalnotion of hacking-i. E. , performing pranks or exploringunderground tunnels-remains intact. In the fall of2000, the MIT Museum paid tradition to the Institute'sage-old hacking tradition with a dedicated exhibit, theHall of Hacks. The exhibit includes a number ofphotographs dating back to the 1920s, including oneinvolving a mock police cruiser. In 1993, students paidhomage to the original MIT notion of hacking by placingthe same police cruiser, lights flashing, atop theInstitute's main dome. The cruiser's vanity licenseplate read IHTFP, a popular MIT acronym with manymeanings. The most noteworthy version, itself datingback to the pressure-filled world of MIT student lifein the 1950s, is "I hate this fucking place. " In 1990, however, the Museum used the acronym as a basis for ajournal on the history of hacks. Titled, The Institutefor Hacks Tomfoolery and Pranks, the journal offers anadept summary of the hacking. 

"In the culture of hacking, an elegant, simple creationis as highly valued as it is in pure science, " writesBoston Globe reporter Randolph Ryan in a 1993 articleattached to the police car exhibit. "A Hack differsfrom the ordinary college prank in that the eventusually requires careful planning, engineering andfinesse, and has an underlying wit and inventiveness, "Ryan writes. "The unwritten rule holds that a hackshould be good-natured, non-destructive and safe. Infact, hackers sometimes assist in dismantling their ownhandiwork. "

The urge to confine the culture of computer hackingwithin the same ethical boundaries is well-meaning butimpossible. Although most software hacks aspire to thesame spirit of elegance and simplicity, the softwaremedium offers less chance for reversibility. Dismantling a police cruiser is easy compared withdismantling an idea, especially an idea whose time hascome. Hence the growing distinction between "black hat"and "white hat"-i. E. , hackers who turn new ideas towarddestructive, malicious ends versus hackers who turn newideas toward positive or, at the very least, informative ends. 

Once a vague item of obscure student jargon, the word"hacker" has become a linguistic billiard ball, subjectto political spin and ethical nuances. Perhaps this iswhy so many hackers and journalists enjoy using it. Where that ball bounces next, however, is anybody's guess. 

Appendix C GNU Free Documentation License (GFDL)

GNU Free Documentation License Version 1. 1, March 2000Copyright (C) 2000 Free Software Foundation, Inc. 59Temple Place, Suite 330, Boston, MA 02111-1307 USAEveryone is permitted to copy and distribute verbatimcopies of this license document, but changing it is notallowed. PREAMBLE The purpose of this License is tomake a manual, textbook, or other written document"free" in the sense of freedom: to assure everyone theeffective freedom to copy and redistribute it, with orwithout modifying it, either commercially ornoncommercially. Secondarily, this License preservesfor the author and publisher a way to get credit fortheir work, while not being considered responsible formodifications made by others. 

This License is a kind of "copyleft, " which means thatderivative works of the document must themselves befree in the same sense. It complements the GNU GeneralPublic License, which is a copyleft license designedfor free software. 

We have designed this License in order to use it formanuals for free software, because free software needsfree documentation: a free program should come withmanuals providing the same freedoms that the softwaredoes. But this License is not limited to softwaremanuals; it can be used for any textual work, regardless of subject matter or whether it is publishedas a printed book. We recommend this Licenseprincipally for works whose purpose is instruction orreference. APPLICABILITY AND DEFINITIONS This Licenseapplies to any manual or other work that contains anotice placed by the copyright holder saying it can bedistributed under the terms of this License. The"Document", below, refers to any such manual or work. Any member of the public is a licensee, and isaddressed as "you. "

A "Modified Version" of the Document means any workcontaining the Document or a portion of it, eithercopied verbatim, or with modifications and/ortranslated into another language. 

A "Secondary Section" is a named appendix or afront-matter section of the Document that dealsexclusively with the relationship of the publishers orauthors of the Document to the Document's overallsubject (or to related matters) and contains nothingthat could fall directly within that overall subject. (For example, if the Document is in part a textbook ofmathematics, a Secondary Section may not explain anymathematics. ) The relationship could be a matter ofhistorical connection with the subject or with relatedmatters, or of legal, commercial, philosophical, ethical or political position regarding them. 

The "Invariant Sections" are certain Secondary Sectionswhose titles are designated, as being those ofInvariant Sections, in the notice that says that theDocument is released under this License. 

The "Cover Texts" are certain short passages of textthat are listed, as Front-Cover Texts or Back-CoverTexts, in the notice that says that the Document isreleased under this License. 

A "Transparent" copy of the Document means amachine-readable copy, represented in a format whosespecification is available to the general public, whosecontents can be viewed and edited directly andstraightforwardly with generic text editors or (forimages composed of pixels) generic paint programs or(for drawings) some widely available drawing editor, and that is suitable for input to text formatters orfor automatic translation to a variety of formatssuitable for input to text formatters. A copy made inan otherwise Transparent file format whose markup hasbeen designed to thwart or discourage subsequentmodification by readers is not Transparent. A copy thatis not "Transparent" is called "Opaque. "

Examples of suitable formats for Transparent copiesinclude plain ASCII without markup, Texinfo inputformat, LaTeX input format, SGML or XML using apublicly available DTD, and standard-conforming simpleHTML designed for human modification. Opaque formatsinclude PostScript, PDF, proprietary formats that canbe read and edited only by proprietary word processors, SGML or XML for which the DTD and/or processing toolsare not generally available, and the machine-generatedHTML produced by some word processors for outputpurposes only. 

The "Title Page" means, for a printed book, the titlepage itself, plus such following pages as are needed tohold, legibly, the material this License requires toappear in the title page. For works in formats which donot have any title page as such, "Title Page" means thetext near the most prominent appearance of the work'stitle, preceding the beginning of the body of the text. VERBATIM COPYING You may copy and distribute theDocument in any medium, either commercially ornoncommercially, provided that this License, thecopyright notices, and the license notice saying thisLicense applies to the Document are reproduced in allcopies, and that you add no other conditions whatsoeverto those of this License. You may not use technicalmeasures to obstruct or control the reading or furthercopying of the copies you make or distribute. However, you may accept compensation in exchange for copies. Ifyou distribute a large enough number of copies you mustalso follow the conditions in section 3. 

You may also lend copies, under the same conditionsstated above, and you may publicly display copies. COPYING IN QUANTITY If you publish printed copies ofthe Document numbering more than 100, and theDocument's license notice requires Cover Texts, youmust enclose the copies in covers that carry, clearlyand legibly, all these Cover Texts: Front-Cover Textson the front cover, and Back-Cover Texts on the backcover. Both covers must also clearly and legiblyidentify you as the publisher of these copies. Thefront cover must present the full title with all wordsof the title equally prominent and visible. You may addother material on the covers in addition. Copying withchanges limited to the covers, as long as they preservethe title of the Document and satisfy these conditions, can be treated as verbatim copying in other respects. 

If the required texts for either cover are toovoluminous to fit legibly, you should put the firstones listed (as many as fit reasonably) on the actualcover, and continue the rest onto adjacent pages. 

If you publish or distribute Opaque copies of theDocument numbering more than 100, you must eitherinclude a machine-readable Transparent copy along witheach Opaque copy, or state in or with each Opaque copya publicly-accessible computer-network locationcontaining a complete Transparent copy of the Document, free of added material, which the general network-usingpublic has access to download anonymously at no chargeusing public-standard network protocols. If you use thelatter option, you must take reasonably prudent steps, when you begin distribution of Opaque copies inquantity, to ensure that this Transparent copy willremain thus accessible at the stated location until atleast one year after the last time you distribute anOpaque copy (directly or through your agents orretailers) of that edition to the public. 

It is requested, but not required, that you contact theauthors of the Document well before redistributing anylarge number of copies, to give them a chance toprovide you with an updated version of the Document. MODIFICATIONS You may copy and distribute a ModifiedVersion of the Document under the conditions ofsections 2 and 3 above, provided that you release theModified Version under precisely this License, with theModified Version filling the role of the Document, thuslicensing distribution and modification of the ModifiedVersion to whoever possesses a copy of it. In addition, you must do these things in the Modified Version:

1. Use in the Title Page (and on the covers, if any) atitle distinct from that of the Document, and fromthose of previous versions (which should, if there wereany, be listed in the History section of the Document). You may use the same title as a previous version if theoriginal publisher of that version gives permission. 

2. List on the Title Page, as authors, one or morepersons or entities responsible for authorship of themodifications in the Modified Version, together with atleast five of the principal authors of the Document(all of its principal authors, if it has less than five). 

3. State on the Title page the name of the publisher ofthe Modified Version, as the publisher. 

4. Preserve all the copyright notices of the Document. 

5. Add an appropriate copyright notice for yourmodifications adjacent to the other copyright notices. 

6. Include, immediately after the copyright notices, alicense notice giving the public permission to use theModified Version under the terms of this License, inthe form shown in the Addendum below. 

7. Preserve in that license notice the full lists ofInvariant Sections and required Cover Texts given inthe Document's license notice. 

8. Include an unaltered copy of this License. 

9. Preserve the section entitled "History, " and itstitle, and add to it an item stating at least thetitle, year, new authors, and publisher of the ModifiedVersion as given on the Title Page. If there is nosection entitled "History" in the Document, create onestating the title, year, authors, and publisher of theDocument as given on its Title Page, then add an itemdescribing the Modified Version as stated in theprevious sentence. 

10. Preserve the network location, if any, given in theDocument for public access to a Transparent copy of theDocument, and likewise the network locations given inthe Document for previous versions it was based on. These may be placed in the "History" section. You mayomit a network location for a work that was publishedat least four years before the Document itself, or ifthe original publisher of the version it refers togives permission. 

11. In any section entitled "Acknowledgements" or"Dedications, " preserve the section's title, andpreserve in the section all the substance and tone ofeach of the contributor acknowledgements and/ordedications given therein. 

12. Preserve all the Invariant Sections of theDocument, unaltered in their text and in their titles. Section numbers or the equivalent are not consideredpart of the section titles. 

13. Delete any section entitled "Endorsements. " Such asection may not be included in the Modified Version. 

14. Do not retitle any existing section as"Endorsements" or to conflict in title with anyInvariant Section. 

If the Modified Version includes new front-mattersections or appendices that qualify as SecondarySections and contain no material copied from theDocument, you may at your option designate some or allof these sections as invariant. To do this, add theirtitles to the list of Invariant Sections in theModified Version's license notice. These titles must bedistinct from any other section titles. 

You may add a section entitled "Endorsements, " providedit contains nothing but endorsements of your ModifiedVersion by various parties-for example, statements ofpeer review or that the text has been approved by anorganization as the authoritative definition of a standard. 

You may add a passage of up to five words as aFront-Cover Text, and a passage of up to 25 words as aBack-Cover Text, to the end of the list of Cover Textsin the Modified Version. Only one passage ofFront-Cover Text and one of Back-Cover Text may beadded by (or through arrangements made by) any oneentity. If the Document already includes a cover textfor the same cover, previously added by you or byarrangement made by the same entity you are acting onbehalf of, you may not add another; but you may replacethe old one, on explicit permission from the previouspublisher that added the old one. 

The author(s) and publisher(s) of the Document do notby this License give permission to use their names forpublicity for or to assert or imply endorsement of anyModified Version. COMBINING DOCUMENTS You may combinethe Document with other documents released under thisLicense, under the terms defined in section 4 above formodified versions, provided that you include in thecombination all of the Invariant Sections of all of theoriginal documents, unmodified, and list them all asInvariant Sections of your combined work in its license notice. 

The combined work need only contain one copy of thisLicense, and multiple identical Invariant Sections maybe replaced with a single copy. If there are multipleInvariant Sections with the same name but differentcontents, make the title of each such section unique byadding at the end of it, in parentheses, the name ofthe original author or publisher of that section ifknown, or else a unique number. Make the sameadjustment to the section titles in the list ofInvariant Sections in the license notice of thecombined work. 

In the combination, you must combine any sectionsentitled "History" in the various original documents, forming one section entitled "History"; likewisecombine any sections entitled "Acknowledgements, " andany sections entitled "Dedications. " You must deleteall sections entitled "Endorsements. " COLLECTIONS OFDOCUMENTS You may make a collection consisting of theDocument and other documents released under thisLicense, and replace the individual copies of thisLicense in the various documents with a single copythat is included in the collection, provided that youfollow the rules of this License for verbatim copyingof each of the documents in all other respects. 

You may extract a single document from such acollection, and distribute it individually under thisLicense, provided you insert a copy of this Licenseinto the extracted document, and follow this License inall other respects regarding verbatim copying of thatdocument. AGGREGATION WITH INDEPENDENT WORKS Acompilation of the Document or its derivatives withother separate and independent documents or works, inor on a volume of a storage or distribution medium, does not as a whole count as a Modified Version of theDocument, provided no compilation copyright is claimedfor the compilation. Such a compilation is called an"aggregate, " and this License does not apply to theother self-contained works thus compiled with theDocument, on account of their being thus compiled, ifthey are not themselves derivative works of the Document. 

If the Cover Text requirement of section 3 isapplicable to these copies of the Document, then if theDocument is less than one quarter of the entireaggregate, the Document's Cover Texts may be placed oncovers that surround only the Document within theaggregate. Otherwise they must appear on covers aroundthe whole aggregate. TRANSLATION Translation isconsidered a kind of modification, so you maydistribute translations of the Document under the termsof section 4. Replacing Invariant Sections withtranslations requires special permission from theircopyright holders, but you may include translations ofsome or all Invariant Sections in addition to theoriginal versions of these Invariant Sections. You mayinclude a translation of this License provided that youalso include the original English version of thisLicense. In case of a disagreement between thetranslation and the original English version of thisLicense, the original English version will prevail. TERMINATION You may not copy, modify, sublicense, ordistribute the Document except as expressly providedfor under this License. Any other attempt to copy, modify, sublicense or distribute the Document is void, and will automatically terminate your rights under thisLicense. However, parties who have received copies, orrights, from you under this License will not have theirlicenses terminated so long as such parties remain infull compliance. FUTURE REVISIONS OF THIS LICENSE TheFree Software Foundation may publish new, revisedversions of the GNU Free Documentation License fromtime to time. Such new versions will be similar inspirit to the present version, but may differ in detailto address new problems or concerns. Seehttp://www. Gnu. Org/copyleft/. 

Each version of the License is given a distinguishingversion number. If the Document specifies that aparticular numbered version of this License "or anylater version" applies to it, you have the option offollowing the terms and conditions either of thatspecified version or of any later version that has beenpublished (not as a draft) by the Free SoftwareFoundation. If the Document does not specify a versionnumber of this License, you may choose any version everpublished (not as a draft) by the Free SoftwareFoundation. ADDENDUM: How to Use This License for YourDocuments To use this License in a document you havewritten, include a copy of the License in the documentand put the following copyright and license noticesjust after the title page: Copyright (C) YEAR YOURNAME. Permission is granted to copy, distribute and/ormodify this document under the terms of the GNU FreeDocumentation License, Version 1. 1 or any later versionpublished by the Free Software Foundation; with theInvariant Sections being LIST THEIR TITLES, with theFront-Cover Texts being LIST, and with the Back-CoverTexts being LIST. A copy of the license is included inthe section entitled "GNU Free Documentation License". If you have no Invariant Sections, write "with noInvariant Sections" instead of saying which ones areinvariant. If you have no Front-Cover Texts, write "noFront-Cover Texts" instead of "Front-Cover Texts beingLIST"; likewise for Back-Cover Texts. 

If your document contains nontrivial examples ofprogram code, we recommend releasing these examples inparallel under your choice of free software license, such as the GNU General Public License, to permit theiruse in free software.