--- Begin Message ---
|
Subject: |
chapters on Stallman, Wall & Raymond for new book |
|
Date: |
Tue, 27 Nov 2001 00:33:25 -0500 |
Dear Mr. Dachary,
I am the author of Technomanifestos: Visions of the Information
Revolutionaries, to be published in early 2002. Because of your
involvement with GNU and the FSF, I have included a draft (html format)
of the chapters on Richard Stallman and Larry Wall and Eric S. Raymond
for your enjoyment and, if possible, for your review and feedback.
Following incorporation of feedback, our publisher, Texere, may send you
a full manuscript and invite you to provide an endorsement for the
bookÂ’s back cover, if you find it appropriate to do so.
Technomanifestos tells the stories of people who have written seminal
works and shaped the field of information technology, from cybernetics
to the Internet and beyond. The visionaries featured, including Norbert
Wiener, Vannevar Bush, Alan Turing, Ted Nelson, Tim Berners-Lee, Jaron
Lanier and others—all ascribe to humanistic beliefs about the
intersection of technology and society. Technomanifestos attempts to
describe their groundbreaking ideas, their personalities, and their
passion. The book discusses the Information Revolution in a broader
social context and answers the question of what makes it revolutionary;
as well as the unending struggle to ensure that the institutions which
control power—technology, corporations, government—are humanitarian, not
totalitarian.
Please email me at address@hidden if you have any comments or
questions, if you feel I have left out important detail, or would like
to receive any other individual chapters electronically. I intend the
book to be as broad but thorough as possible.
As the manuscript is on an accelerated schedule, the sooner any
response, the better. Again, any response would be appreciated, but I
understand if your schedule makes it impossible to do so.
Yours,
Adam Brate
Chapter Ten
Sharing the Source
Richard M. Stallman and Larry Wall
Electronic freedoms will be at the center of the
whirlwind of the coming years. Either we will fight for
them, or they will be taken from us like candy from a
child.–Ted Nelson
Ideas are strange and magical. We can share an idea with someone else as
easily as handing over a jacket. If two people share the jacket, only one
can wear it at a time unless they rip it in half, rendering it useless.
Unless both are the same size, it will only fit one of them well. When two
people share an idea, however, they both can enjoy full use of it at all
times. The idea shapes itself to fit each person. And it will never wear
out; rather, the more itÂ’s used and shared, the better it gets. But who
would confuse an idea with a jacket? People can learn, teach and collaborate
with ideas. No one can learn a jacket. Ideas inhabit the mental world;
jackets inhabit the physical.
The Information Revolution is bringing the mental and physical worlds
together. Not long ago, just about the only way to store an idea outside the
mind was in a book; now, it can be stored electronically with infinite
variation. Marshall McLuhan recognized how profound the disjunction from the
Gutenberg Age to the Electronic Age would be, just by observing the changes
wrought by the first electronic media; the telegraph, the light bulb, radio,
television. They only presaged the arrival of the networked computer, the
ultimate media of ideas.
With computers, ideas are encoded as software, the code. The code inhabits
the physical world, stored on punch cards, magnetic tape, a memory chip, a
floppy disk, a CD. It could even be stored within the threads of a jacket.
Software is both idea and object, mental and physical. The institutions of
the Industrial Revolution, built around mastery of the physical world, treat
software as if it was no different from a jacket. Individuals have been able
to see the difference; but for their insight to matter to society, all the
institutions must transform.
* * *
By the 1970s, the proliferation of computers had created a market for
software development. People, most famously Bill Gates, realized that they
could demand that people pay for the privilege of using code they wrote, or
at least controlled, and call anyone who just copied the software (like they
had done in the past) thieves and pirates. Bill GatesÂ’s version of a
technomanifesto was his article "An Open Letter to Hobbyists," which led the
charge to criminalize the once open, cross-pollinating programming
community. He wrote the letter in 1975, when personal computer hobbyists
started sharing the BASIC interpreter he had written for profit. He believed
these hobbyists, who had sent back thanks and suggestions for improvements,
but no money, were stealing from him. At the time, the community thought
Gates was a jerk for calling sharing stealing. But the Gates perspective
allowed corporations to make billions of dollars, and it soon prevailed.
Once a few people became millionaire by forming companies to write
proprietary software, thousands followed suit. The major corporations got
into the act. In the early decades of computing, most code was proprietary,
but tied to a particular system. By 1980, proprietary, licensed software was
increasingly packaged separately as a stand-alone product, making
programmers and businesspeople realize there were riches to be made.
There wasnÂ’t any easy way for programmers to share and distribute their
creations. Many now worked for institutions and corporations who didnÂ’t want
them to give away anything. So those who still wanted to live as they had
before, the hackers, surreptitiously worked on pet projects, sneaking them
out the back door. The spirit of sharing cool stuff freely, for personal
pride and communal good, had been forced underground but still existed,
aided by the growth of the ARPANET and then the Internet and Web. The
worldwide network let hackers punch holes through the proprietary shells of
the companies they worked for and sneak programs, like notes under the desk,
to one another.
Two people, Richard M. Stallman and Larry Wall, grew to embrace the
philosophy of openness and collaboration and led the charge to establish a
technological, social, and moral alternative to the cut-throat
commercialism. Stallman wrote "The GNU Manifesto" (1985) and the "GNU
General Public License" (1989) for free software. WallÂ’s manifestos come in
the form of keynotes which summarize his years of developing the programming
language Perl: "First Perl Conference Keynote" (1997), and "The Third State
of the Onion" (1999).
The Information Revolution, they believe, has wrought a "post-scarcity"
world wherein knowledge can be freer and better than ever before. Despite
the encroaching corporate proprietary attitude, these revolutionaries
attempt to hardwire social consciousness into the technologies of the next
century.
Edens
Richard Matthew StallmanÂ’s fascination with computers and programming drove
him in to the famed MIT Artificial Intelligence Lab in 1971. Up until that
point, his freshman year at Harvard, he had precious little access to
computers. Although he eventually earned a magna cum laude physics degree
from Harvard, the computer center there was like most others then and now:
heavily regulated, with a strong hierarchy that determined who was allowed
to use the computing resources. With the hackers at the MIT AI Lab Stallman
discovered a completely different world, one in which students would and
could literally break down the door if a professor dared lock up a computer
terminal.
Stallman had a preternatural interest in computers. As a child growing up in
Manhattan in the 1950s he scrounged up computer manuals and wrote down
programs and algorithms, imagining the day heÂ’d get to run them. At times
that may have seemed like just a fantasy. Although Stallman was extremely
bright and self-aware, devouring science fiction books from an early age, he
had trouble getting along with teachers and other children. After several
years of devastating school experiences, he was sent to a special high
school for troubled students. There he felt branded, trapped in a finite
world. Richard decided to figure out how to survive the system, and improved
his grades dramatically. In 1969, the summer of the ARPANET, he got a summer
job at an IBM laboratory in New York and started programming on a computer
for the first time. In his last year of high school he transferred to the
local public school, which enabled him to get into Harvard.
The AI Lab was Eden for the happy hacker, the clever programmer who coded
for the hell of it. This was the home of Marvin Minsky and Seymour Papert,
and the training grounds for researchers the likes of Len Kleinrock, Larry
Roberts, and Bob Taylor, pioneers of ARPANET. The hacker there neednÂ’t be a
student at MIT. Or even be past adolescence. He (or she, although rare at
the time) just needed to have a passion for programming. While some would
leave to get married and have a family, new hackers invariably took their
place.
To be a hacker there meant that your ethical code was driven by the progress
of the computer code–it was wrong, almost evil, to keep code or computing
resources to yourself. Hackers respected one another because they were good
at what they did, not because they had money or a title. This led to
profound conflicts with other ethical systems, particularly ones that give
supremacy to the individual ownership of ideas. From the perspective of the
hacker, keeping an idea or a new program to yourself was a spit in the eye
of everyone else–you were obstructing progress and increasing waste, helping
entropy instead. From the opposite perspective, hackers were constantly
invading secrets and building systems without necessary security.
Everyone in StallmanÂ’s group celebrated creativity and individuality,
diverse interests, and a sense of play. The hackers controlled their own
time and, unlike the professors and graduate students, ignored the
traditional rules and roles. With almost no official power in the MIT, they
maintained their status by being demonstrably the best computer programmers
around. StallmanÂ’s days of programming code there were leavened by Bulgarian
singing, folk dancing, and playing practical jokes. He kept his hair long
and stringy. Stallman felt safe within the confines of the MIT community,
seemingly beyond the grip of the Vietnam War, which he considered
wrong-headed. Despite a low number, he wasnÂ’t drafted. The ninth floor of
Tech Square remained his home and haven, sometimes literally: There was one
summer when he had no other residence, and slept next to his beloved
computer.
By the mid-1970s Stallman had established himself. Within the computing
community the three initial login handle RMS, Richard M. Stallman, was
well-known and well-respected, in no small part due to the advanced text
editor and programming environment he created, called Emacs.
* * *
Meanwhile, on the opposite coast of the United States, Larry Wall was
receiving an education in a completely different environment. He was raised
in south Los Angeles as an evangelical Protestant; his father was a pastor
and both parents descended from a long line of preachers. In 1961 the Walls
moved to Pasadena, then to Bremerton, Washington in 1963. It was there
where, in 1972, Larry Wall first began programming on a school calculator.
He almost managed to get a game of tic-tac-toe working on the machine, but
couldnÂ’t quite fit the program into the 120 steps that was the calculatorÂ’s
limit.
Wall enrolled at Seattle Pacific University as a chemistry and music major.
He switched to premed, but eventually became more interested in computer
science and began spending a lot of time at SPUÂ’s computer center. In fact,
Wall spent three years working in the computer center, just taking classes
when he felt like it. There he helped a young Bill Gates, who, ever mindful
of his cash flow, would schlep over from the nearby University of Washington
to get cheap computer access. Wall finished his undergraduate degree eight
years after he entered with a self-made major in Natural and Artificial
Languages; in other words, in linguistics and computer science. Like other
information revolutionaries, his interests spanned a wide diversity of
fields.
Larry Wall intended to parlay his training in linguistics with his religious
convictions to become a Bible-translating missionary. He and his new wife,
Gloria, moved back to California for graduate school in linguistics at
Berkeley and UCLA, while also studying at the missionary Summer Institute of
Linguistics. They planned to go to Africa and learn unwritten languages,
devise writing systems, and translate works including the Bible–work
simultaneously spiritual, unusual, and intellectual. But Larry had become
allergic to everything from seafood to wheat, eggs, and tomatoes. Going to
the far reaches of Africa might be detrimental to his health. So he turned
instead back to programming, which, as it turned out, was fortunate for the
computer revolution ahead.
Surfing the Code
Programmers like Richard Stallman and Larry Wall have been on the front
lines of the Information Revolution. They are among the most involved users
of computers. They delve into the system, tweaking or even rebuilding it to
serve their desires. They are creating the environment for everyone else.
Today, most programmers work inside of a preconstructed environment. In John
von Neumann and Alan TuringÂ’s day, everyone who wanted to use a computer
needed to know just how it worked, or at least the numeric codes it could
understand. Soon, however, people designed systems that would automatically
load onto the computer, systems which others could use. These systems hide
as much of the complexity of the digital machinery as possible so that
people can just sit down at a computer and do what they want. TodayÂ’s
operating systems have taken that concept a good distance, establishing a
graphic environment inhabited by virtual objects, represented on monitors by
icons and windows, menus and buttons. The relationship between the virtual
world on the monitor (what Alan Kay calls the user illusion) and the bits
flowing through the computerÂ’s processor is extremely arcane and complex.
But there are other environments spaces where users can get closer to the
workings of the computer: programming environments.
Programming environments make the task of designing complex programs to run
on computers feasible. Computers run only machine code–binary instructions
in the language particular to the processor inside. Each different kind of
computer understands a different language of machine code. The processor
inside a IBM PC uses a different language than that inside a Macintosh. In
the earliest days, programmers had to flip switches, punch up, or type in
series of numbers to load instructions into a computer. Soon people devised
first-level systems, known as assembly languages, which would let people
write their programs using memorizable codes, like "ADD" or "JMP," which the
computer would directly transcribe into the corresponding numerical codes it
could run.
>From the start, programmers began devising computer languages which allow
people to write their programs in something resembling English.[1] In such
languages, one command might be the equivalent of dozens of assembler
commands. Then people can write more complex programs than they could have
if they had to use only the very basic commands the computer directly
understands. Software associated with the language, the compiler, then takes
the source program and does the task of translating it as quickly and
efficiently as possible into the intended numerical code which the computer
can execute. To write the fastest and most efficient programs required the
hand-crafting of assembler code, but that task is near impossible for large
programs. Often programmers combine the two approaches, hand-tuning in
assembly the most time-critical components, while maintaining a
comprehensible framework through compiler code.
Thus a program usually exists in two forms: the understandable "source
code," and the executable "machine code." The source code, which consists of
lines of text, is like a recipe; the machine code is the resultant dish. One
can look at the machine code to try to determine what the computer will do,
but that is usually a near-impossible task, just as itÂ’s often difficult to
exactly recreate a dish. And most programs are as complex as, say, a
Twinkie, with dozens of ingredients and finely tuned chemical additives and
an industrial cooking process. It would be impossible to duplicate a Twinkie
just by analyzing the finished product. The source code, however, is the
recipe that allows you to create and modify your own Twinkies.[2]
Unix
The university computing departments of the 1970s, where Stallman and Wall
learned to program, were defined by the symbiotic growth of ARPANET, the
operating system Unix, and the programming language C. Unix and C
represented the culmination of all the advances of the previous two decades
of computing. Flexible and powerful, Unix had the greatest strength of being
portable, whereas previous operating systems were designed for a particular
model of computer. Because implementations of C s prung up for nearly every
hardware system, programmers could write code for large, complex programs
that would remain comprehensible and flexible on various platforms, but
still compile to be almost as fast as if they had be handcrafted in
assembly. As advances in hardware accelerated and new models proliferated,
Unix became the dominant operating system for the networked computer centers
across the country, and then the world. Unix and C also became the
environment and the benchmark for Stallman and Wall to create their own
work.
In that revolutionary year of 1969, as the ARPANET went up, Ken Thompson of
Bell Labs invented Unix. Following the lessons of the best in theoretical
computer science, Thompson designed Unix to be a toolkit of simple,
complementary programs that could be linked together to perform complex
tasks. The development of Unix was driven by the toolkit philosophy, which
can be broken down into four precepts:
1. Write programs to be single-purpose tools.
2. Make the output of every program usable by other programs.
3. Make a working program as early as possible.
4. If a tool would help your work, use it. If it doesnÂ’t exist, build it.
This philosophy is well grounded in the real world–it’s how a carpenter uses
his tools to build a bookshelf or a house. ItÂ’s also very similar to the
Society of Mind concept being developed by Marvin Minsky. Computer
scientists everywhere were realizing that systems of great complexity and
creativity, from artificial intelligence systems to operating systems, can
best be built by the symbiotic coordination of many simple parts. The only
thing the toolkit philosophy glossed over was that the programmer was
required to handle all of the coordination, an often arduous task,
especially since the Unix tools actually werenÂ’t as simple as they should
have been.
Thompson and the rest of the Bell Labs researchers in his team found that
their scientific approach was paying great dividends. They soon had a usable
Unix system up and running. In 1973 a friend of Thompson from UC Berkeley
saw their talk and was impressed enough to ask for a copy of the system.
Berkeley soon became the second great center for Unix development, led by
the graduate student Bill Joy. In 1975, the same year as the first personal
computer debuted, Ken Thompson came to Berkeley on sabbatical from Bell
Labs. Under his guidance, Bill Joy and another graduate student, Chuck
Haley, began to improve and extend the Berkeley Unix system. By 1977, the
year Steve Jobs founded Apple Computer, Joy put together the a computer tape
with the Berkeley Software Distribution (BSD) of Unix, sending out thirty
copies to those who had the necessary AT&T Unix licenses. Soon enough, Bill
Joy had become a one-man missionary for Unix, whose popularity was steadily
growing. But the explosion was just around the corner, when personal
computers would converge with these professional tools.
A New Propriety
As the 1980s rolled around Stallman and Wall became central players in the
expanding computing universe. Suddenly, computing became part of the real
world, which turned 180 degrees from the excesses and successes of the 1960s
and 1970s. Ronald ReaganÂ’s election as president marked the collapse of
liberalism and the rise of the yuppie, replacing free love, the hippie, and
the civil rights movement with corporate conservatism, the entrepreneur, the
Greed Decade. And thousands of those entrepreneurs were into computers.
Whereas there once had been dozens of computer companies, there were now
hundreds. The hottest companies in the world were computer companies that
hadnÂ’t existed a decade before, like Apple and Atari. And while people had
been making money by selling machines, this was the first decade where
millions, even billions of dollars could be made just by selling software to
run on those machines. It was the decade of the personal computer,
commercial software, and with RMS and Larry Wall, the decade of the birth of
the free software movement.
The development that shaped Stallman and WallÂ’s careers began in 1980, as
the popularity of Unix rose. They worked on the powerful computers that were
converging on some variant of Unix as an operating system; they were outside
of the personal computer boom. The changes in the distribution and licensing
of Unix changed their lives. AT&T realized Unix could be a commercial
product, and stopped contributing the Bell Labs developments to the rest of
the Unix licensees–thereafter, Berkeley became the central focus of
community advances. Meanwhile, DARPA (the renamed ARPA; the D stands for
Defense) wanted to consolidate their ARPANET network on the same operating
system, and chose Unix. Instead of going with AT&TÂ’s version, they supported
the Berkeley development of Unix by Joy.
In 1982, Unix users from Stanford and Berkeley, including Bill Joy, formed
Sun Microsystems. Sun became an overnight success by selling, by the
standards of the time, very inexpensive workstations running the Berkeley
variant of Unix. The combination of the growth of the ARPANET and the
proliferation of Sun workstations brought forth the Unix explosion. In 1984,
the AT&T monopoly was split up; Unix came out from its Bell Labs nest as a
stand-alone commercial product. At the time there were three main variants
of Unix: AT&T Unix, BerkeleyÂ’s BSD, and SunÂ’s Unix, SunOS (later dubbed
Solaris). They coexisted uneasily but peacefully, advancing steadily. All
the versions were governed by AT&TÂ’s license, and for every copy of Unix
sold, money was paid into AT&TÂ’s coffers.
A Symbolic Stand
Meanwhile, the sea change to proprietary software hit the MIT AI Lab hard,
and Stallman the hardest of all. The first glimmerings were in 1977, when
Xerox gave a graphics printer to the AI Lab. The hackers hacked around with
its software, extending and improving its functionality, and squashing bugs.
Then Xerox gave the AI Lab a faster printer, one of the first laser printers
out of Xerox PARC research. But this time the software was run by a
separate, dedicated and inaccessible computer. The Lab lost functionality
without any way of getting it back, since the hackers couldnÂ’t get at the
code. Around the same time, Digital discontinued the PDP-10, the only system
on which StallmanÂ’s system could run. The AI Lab would have to use a new
operating system. And it looked like it would probably be proprietary.
But it was the war between the hackers themselves that destroyed StallmanÂ’s
world once and for all.
Led by the senior hacker Richard Greenblatt, the hackers at MIT had the
dream of designing and selling a stand-alone machine that would run the
operating system LISP. The LISP machine would not only would it be cool and
hackable, but also represent the best of technology. They knew about the
ideas of Alan Kay and others, and incorporated those into the ideals of the
AI Lab, planning to make the ultimate personal computer. But when in 1979,
Greenblatt insisted that LISP Machine Incorporated (LMI) be a hackerish,
small-scale company, most of the other hackers didnÂ’t go along. They instead
decided to form a heavily funded company, Symbolics, with the goal of
getting big, making money fast, and thus having a strong influence on the
world.
In 1981, Symbolics hired away the hackers en masse from the AI Lab. Soon all
the hackers who were involved with either company had to resign from the AI
Lab, leaving just Stallman to hold the torch. When MIT got a contract to use
the Symbolics LISP machines, Stallman saw his opportunity to fight back. The
LISP operating system had once been a joint endeavor between the AI Lab,
Symbolics, and LMI. But Symbolics made their system proprietary. Moreover,
they forced the hackers to choose sides: for or against them.
Stallman stood alone against Symbolics. When new versions of the Symbolics
system would appear at MIT, Stallman figured out how to reproduce the
features, with his own code. Since Stallman didnÂ’t hide his work, LMI could
use it to keep up with Symbolics. Working on his own, Stallman outhacked all
of SymbolicsÂ’ programmers for two years. While some at Symbolics simply saw
Stallman as a wretched weasel, most of them could look past their corporate
loyalty to be awed by one of the greatest hackers of all time. Stallman was
John Henry to the steam locomotive of Symbolics.
Stallman knew he couldnÂ’t fight Symbolics forever, but he did it for long
enough. He gave GreenblattÂ’s company enough time to get a foothold in the
LISP machine market. By 1983 it was time for Stallman to move on.
Non-disclosure agreements were being shoved in his face. The hackers were
gone. The bureaucratic, secret-laden, hierarchical, selfish, institutional
world had overtaken the AI Lab. The computer and the culture to which he
pledged his life–had died. But what to do next?
* * *
Stallman decided to develop a free Unix-compatible operating system, which
he dubbed GNU (GNUÂ’s Not Unix, recursive acronyms being yet another AI Lab
hacker tradition), in 1983.[3] He could see that Unix, while certainly not
his favorite operating system, with plenty of hierarchy and passwords, was
becoming the dominant system in the programming world. Since Stallman wanted
to create a free operating system that everyone would use, he knew that it
would have to be compatible with Unix. He decided that when 1984 began, he
would quit the AI Lab and begin his fight against the Orwellian forces that
ripped apart his community. GNU would be free to all–shared and enjoyed as
the basis for an unfettered computing experience. His first idea was to get
funding from hardware companies, who would benefit from a non-proprietary
operating system. But they wouldnÂ’t make an investment in a long-haired
hackerÂ’s plan, however brilliant he was or however proven his skills. So he
took his case to the community, by writing a technomanifesto that defined a
movement, the Free Software Movement.
The GNU Manifesto
Richard StallmanÂ’s "GNU Manifesto" was originally published in the March
1985 edition of Dr. DobbÂ’s Journal, whose readers were the same type of
people that read Stewart BrandÂ’s The Whole Earth Catalog a decade earlier.
The "GNU Manifesto" is RMSÂ’s manifesto, expressly named, which defined the
GNU philosophy and established the free software movement as a social,
political, and moral cause. RMS had initially written a version of the
manifesto intended to appeal to computer companies. He argued the benefits
of free software, and was looking for a few investments of maybe $50,000
from a few different companies that could reap millions of dollars in
benefits down the road. But every corporation thought it was a crazy idea.
So RMS rewrote the manifesto for a new audience: the community of
programmers, his fellow hackers. He enjoined them to join the fight to
preserve the way of life of the programmer against the encroachments of the
world of proprietary software. Moreover, he argued that it was immoral to
not share code, and he would lead by example.
Why I Must Write GNU: I consider that the golden rule requires that if I
like a program I must share it with other people who like it. Software
sellers want to divide the users and conquer them, making each user agree
not to share with others. I refuse to break solidarity with other users in
this way.–"The GNU Manifesto"
What hurt Stallman the most about software companies was not that they
profited by selling programs, but that by doing so they harmed the society
of computer users. Without proprietary software, programmers have no reason
not to work together, improving one anothersÂ’ lives by improving the code.
They have no need to hoard ideas, because the only benefit comes from having
others use them. Most important, working together creates the bond of
friendship, respect, and love, while retaining the goad of competition;
proprietary software means working against each other, which leads
inevitably to distrust, disrespect, and cynicism.
The fundamental act of friendship among programmers is the sharing of
programs; marketing arrangements now typically used essentially forbid
programmers to treat others as friends. The purchaser of software must
choose between friendship and obeying the law. Naturally, many decide that
friendship is more important. But those who believe in law often do not feel
at ease with either choice. They become cynical and think that programming
is just a way of making money.–"The GNU Manifesto"
Stallman knew that the battle against proprietary software had to start at
the core, the bedrock of computing, the operating system. The operating
system is the environment in which all other software operates; it is the
world the programmer inhabits. In the real world, we all enjoy the benefits
of public space–natural resources are common resources.
Why All Computer Users Will Benefit: Once GNU is written, everyone will be
able to obtain good system software free, just like air.–"The GNU Manifesto"
The idea of free software certainly includes the idea that people are not
obligated to pay money for the use of the GNU system. But money was not
StallmanÂ’s central concern; it was the health of the community, of all of
society. The primary benefit of not having restrictions on breathing the air
is not that we donÂ’t have to pay. The primary benefit is that we can live.
The economic benefit of inexpensive software is a mere distraction, in
StallmanÂ’s mind, to the societal benefits of a healthy, trusting,
collaborative programming community that can share the source code of the
system.
Stallman listed the three primary benefits of free system software: because
the sources will be available, any user can make personal modifications;
schools will have a much more educational computing environment; and the
overhead and perniciousness of regulating use of system software will be
eliminated.
Consider a space station where air must be manufactured at great cost:
Charging each breather per liter of air may be fair, but wearing the metered
gas mask all day and all night is intolerable even if everyone can afford to
pay the air bill. And the TV cameras everywhere to see if you ever take the
mask off are outrageous. It's better to support the air plant with a head
tax and chuck the masks.–"The GNU Manifesto"
By discussing such an economic system, Stallman wasnÂ’t ignoring the monetary
reality that programmers need to be paid. He was simply asserting that as
computers become ever more important to society, they become natural
resources like air, and should be handled as such. System software becomes
part of the societal environment and should have no restrictions or
impediments to its free use. The predictive accuracy of his seemingly
outrageous analogy of TV cameras everywhere to check on gas masks is
chilling–this kind of round-the-clock, privacy-eliminating monitoring is
endemic in todayÂ’s world of proprietary operating systems.
To the programmer, information, code, is his lifeblood. Without the ability
to understand and circulate it, he might as well be dead. On the same
grounds decades later, Stallman had rediscovered Norbert WienerÂ’s
realization that cybernetic feedback of information is crucial to the health
of an information-driven society.
Copying all or parts of a program is as natural to a programmer as
breathing, and as productive. It ought to be as free.–"The GNU Manifesto"
DevilÂ’s Advocate
Just as Turing anticipated objections to artificial intelligence in his
manifesto, "Computing Machinery and Intelligence," so did Stallman
anticipate objections to free software in his GNU Manifesto. Stallman knew
that the prevailing wisdom was that free software couldnÂ’t work.
Objection: "Nobody will use it if it is free, because that means they can't
rely on any support." And "You have to charge for the program to pay for
providing the support."
StallmanÂ’s rebuttal, "If people would rather pay for GNU plus service than
get GNU free without service, a company to provide just service to people
who have obtained GNU free ought to be profitable," has been borne out.
There are a number of thriving companies that make money by providing
service to users of GNU software.
Objection: You need to advertise to reach lots of people, and thereÂ’s no use
in advertising something thatÂ’s free
Stallman argued that objection canÂ’t work both ways: If GNU would benefit
from advertising, then a company could make a profit by advertising its
services in distributing the system. If, on the other hand, the distribution
of GNU is perfect without any outside help, then it wouldnÂ’t need any
advertising to reach lots of people.
Objection: "My company needs a proprietary operating system to get a
competitive edge."
Stallman had a flat response for that objection: GNU will remove operating
system software from the realm of competition. He knew that if GNU was as
good as any proprietary operating system out there, free market economics
would eliminate the proprietary systems. Just as thereÂ’s no competition for
the atmosphere in the breathable air market, there would be no competition
for GNU, as the free software juggernaut took over.
Objection: "Don't people have a right to control how their creativity is
used?"
This has actually become the most lasting objection to StallmanÂ’s arguments.
People who agree with everything else Stallman says disagree with StallmanÂ’s
feelings. RMS made it very clear that he in particular felt the only moral
choice one could make was to not control how his creativity is used. Many
others are not as willing to be so saint-like, and simply resent his moral
posturing. But RMS didnÂ’t insist that others follow his lead. Rather, he
reminded them that at least in the United States, individual control over
their creations isnÂ’t an intrinsic right. Rather, the Constitution gives
control over the arts and sciences to society.
All intellectual property rights[4] are just licenses granted by society
because it was thought, rightly or wrongly, that society as a whole would
benefit by granting them. But in any particular situation, we have to ask:
are we really better off granting such license? What kind of act are we
licensing a person to do?–"The GNU Manifesto"
Objection: Competition makes things get done better.
That objection is true for healthy competition, but Stallman believed that
proprietary software, with its restrictions and lawsuits, made for unhealthy
competition. Healthy competition is like a road race, where the victory goes
to the swiftest. But competition can also include other strategies, such as
attacking the other runners. When that happens, everyone finishes late.
Stallman was personally wounded by what he experienced Symbolics doing,
trying all they could to eliminate competition in the AI Lab or at LMI. The
only good way to compete in software, Stallman believed, was to write better
software, not threaten your competitors, make customers sign restrictive
deals, and hide your work. Those tactics allow software companies to make
money fast, but itÂ’s money at the expense of progress.
Low-paying organizations do poorly in competition with high-paying ones, but
they do not have to do badly if the high-paying ones are banned.–"The GNU
Manifesto"
At first the idea of banning high-paying organizations seems absurd, but
historically itÂ’s the road of society he has chosen. The greatest profits
from the least effort are to be made by organizations that threaten direct
violence, like protection rackets, or abuse a monopoly on a necessary
product or commodity, or sell an addictive product, like drugs and gambling,
or donÂ’t pay taxes. All of those organizations are illegal, because their
profits come at the expense of the prosperity of society.
The Ultimate Objection: I want to get paid for programming!
StallmanÂ’s basic response is that money isnÂ’t everything. He countered the
reasoning that Bill Gates had professed in 1975, that the most important
thing in software development is to prevent society from sharing software so
that the individual developer can make as much money as possible. Plenty of
people were programming, devoting their entire lives to writing code,
because of the pure joy of the task, and for fame and appreciation. The AI
Lab hackers, while paid much less than if they had gone into corporations,
were the most famous coders of the time. They are the ones whose names are
remembered, whose exploits are remembered. Programmers, like artists and
musicians, would find ways to code even if they were paid little or less.
Pay for programmers will not disappear, only become less.–"The GNU
Manifesto"
He was certain that free software wouldnÂ’t eliminate the ability of
programmers to get paid. That, he felt, was an irrational fear. When people
do something useful, thereÂ’s money in it somewhere, no matter the
particulars of the law. Speaking of the law, the legal code is free and
available to all, but lawyers and judges, who use it and create new case
law, are some of the most highly paid professionals in the world. Stallman
could think of numerous ways a good programmer could make money on free
software–and time has proven him right.
The GNU General Public License
The "GNU Manifesto" inspired enough people to join StallmanÂ’s crusade that
the Free Software Foundation (FSF) was born in 1985. While they began
writing and distributing pieces of the GNU system, such as the Emacs text
editor, a C compiler, and a debugger, each of which became favorites in the
Unix world, Stallman also worked on figuring out how to formalize the
manifestoÂ’s intentions with the power of law.
Stallman could simply have released his code into the public domain, without
any copyright at all, where all works go when their copyright license
expires. There are no restrictions whatsoever on what someone can do with a
creative work that is in the public domain. You can chop out half the scenes
of Hamlet, add some juggling and dialogue from Socrates, and still call it
ShakespeareÂ’s Hamlet, without running afoul of the law (though it may be
unwatchable). One could stick the entire text of Hamlet in the middle of a
book such as this one without breaking the law. If GNU was public domain,
software companies could take the source code and add whatever bits of it
they wanted to their proprietary software. Stallman certainly didnÂ’t want to
write a free operating system that would simply enrich the likes of AT&T or
Microsoft. So he wrote the "GNU General Public License" (GNU GPL), which
protects works with the power of copyright.
He knew that just as proprietary software developers used the strength of
copyright to prevent users from sharing code, there must be a way for the
FSF to use the strength of copyright to ensure that users shared code. The
first version of the GNU GPL debuted in 1985.[5] While the "GNU Manifesto"
proclaimed the free software philosophy, it was the GNU GPL that was
StallmanÂ’s sword. The license, which summarizes the GNU philosophy amidst
its legal clauses, establishes what Stallman coined "copyleft."
Stallman learned the term of "copyleft" from hacker Don Hopkins, who
scribbled "Copyleft–all rights reversed," on an envelope he sent to his
friend during the early years of the GNU project. The above phrase was
brilliantly hackerish, an awful double pun on the standard phrase
"Copyright–all rights reserved." Stallman, who loves puns with a childlike
delight, latched on to the phrase to describe his innovation.
When we speak of free software, we are referring to freedom, not price.–"GNU
General Public License"
By 1989, Stallman had learned that people were confused by the term "free
software." They thought all he cared about was some kind of pinko,
no-money-allowed software with which no one could get paid for their work.
But that had nothing to do with what he cared about. He just believed that
nonproprietary software was necessary to the exercise of intrinsic personal
freedoms–free software for a free society.
The GPL protects what Stallman identified as the four freedoms necessary for
a healthy programming community:
1. The freedom to run the program, for any purpose.
2. The freedom to study how the program works, and adapt it to your needs.
Access to the source code is a precondition for this.
3. The freedom to redistribute copies so you can help your neighbor.
4. The freedom to improve the program, and release your improvements to
the public, so that the whole community benefits. Access to the source
code is a precondition for this.
The GPL consists of two parts: requiring the above freedoms, and
guaranteeing the preservation of the above freedoms in modified versions.
The GPL states that if someone modifies or uses GPL-protected code in a new
work, the new work must also be released under the GPL. RMS dubbed this form
of copyright protection "copyleft," as it uses copyright to protect the
ability to copy, not prevent the ability to copy. The protection is not
against copying and redistribution, but against incorporation into nonfree
works. If code is released under the GPL, people will always be able to look
at it and tinker with it.
The license agreements of most software companies try to keep users at the
mercy of those companies.–"GNU General Public License"
The GPL forces people to admit that their work rests upon the shoulders of
others. They have to respect the wishes of other people, the community. And
the free software community values sharing. If copylefting leads to a bad
system, then projects released under the GPL will lose. People will just use
alternatives, and development of the GPL-protected projects will wither
away. But Stallman was sure that wouldnÂ’t happen: he believed that
programmers with GPL-protected freedoms write the best code.
Perl
As Stallman was growing the Free Software Foundation, Larry Wall released
his master creation, Perl, out into the world. The programming language
would be his ultimate act of creation and evolution. Like Stallman, Larry
Wall loved science fiction as a child, but he stopped reading the genre for
years because he couldnÂ’t reconcile his fundamentalist environment with the
scientific and largely atheistic tone of science fiction. Bringing faith and
science together gave him much of his strength. To Wall, artistry is the
crafting of order out of chaos, a beautiful whole out of individual
brushstrokes or notes. He sees evolution as the most random, chaotic process
of all–that God chose that as one of his tools confirms to Wall God’s
ultimate artistry.
The genesis of Perl began in the early 1980s when Wall was working as a Unix
support programmer at the Cold War giant computer corporation Burroughs[6]–a
job which gave him plenty of time to hack around writing his own programs,
intended to make his job easier and more enjoyable–and releasing them for
anyone to use to make their lives easier and more enjoyable. These programs,
the newsreader rn, the programming tools patch and metaconfig, and a
computer game, warp, along with his active participation in Internet
newsgroups, made his name quickly well known. Patch was especially popular,
as it allowed people to easily share updates for programs over the
low-bandwith Usenet. Each of his programs were great bits of code, but he
still had plenty of free time and lots of coding zeal. He was ready to take
on a real challenge, something that would really improve the way people used
their systems.
In 1986, WallÂ’s supervisor gave him a giant task with a tight deadline.
WallÂ’s division in Burroughs/Unisys was assigned to a National Security
Agency project, codename "Blacker," which involved an encrypted, high-speed,
bicoastal network–testing out a kind of top-secret, paranoid alternative to
the ARPANET. Wall was told to create a configuration and monitoring system
for the network, which had both Digital and Sun Microsystems machines on
either coast, in a single month. So, using the Unix toolkit, including some
of the stuff he wrote, he hacked together a solution, transforming the
Usenet news system into a command distribution system. Problem solved.
Then the supervisor told him to assemble reports from the jumble of log
files on the different computers. Wall reached into the Unix toolkit and
discovered that the tool he wanted to use, awk, wasnÂ’t up to the task. Awk
had crucial bits missing and was too slow. Larry knew he had to write a new
tool to do the job, but he was tired of making do with the kludgy jumble of
the Unix toolkit, in which some tools were very good and some were awful. He
wanted duct tape. He wanted a Swiss Army knife. He wanted to develop a tool
that would continue to solve different problems in the future.
As Wall described his realization, he determined the three chief virtues of
a programmer:
Laziness, impatience, and hubris.
This was a prankish play on the Biblical virtues of the excellent wife
(humility, forbearance, love, and diligence) [Philippians 4:2,3].
If Wall worked really hard, he probably could have figured out a way to get
awk to do the job. But he was too lazy. Doing that work would take a long,
boring time, and awk was a slow, slow program. But he was too impatient. And
he believed he could do the job better himself. He had the hubris, the
near-arrogant pride of the hacker.
So Wall wrote a super text-processing tool, a complete programming language,
which he soon dubbed perl. The story of the name well describes Larry WallÂ’s
mindset: His first thought was to name the tool after his wife, Gloria. Then
he came up with pearl, which was both beautiful and was obliquely
Scriptural: JesusÂ’ parable of a merchant "seeking goodly pearls, who, when
he had found one pearl of great price, went and sold all that he had, and
bought it" [Matthew:13:46]. With that in mind, he came up with a phrase to
justify the name as an acronym: Practical Extraction And Report Language.
But Unix commands were rarely five letters long, and he heard of a
process-control language with the same name. So Wall changed the name to
perl, which allowed him to keep the first acronymic expansion and come up
with a second, hacker-humor one: Pathologically Eclectic Rubbish Lister.
Throughout 1987 Wall worked on Perl, releasing the first version on December
18, 1987 to the comp.sources.misc newsgroup on the Internet. Perl embodied
two precepts which Wall called the Perl slogans:
The first Perl slogan:
ThereÂ’s more than one way to do it.
While thatÂ’s certainly true in the real world, it is rarely true for
computers. Programmers design their software to work in a very particular,
logical way–and only one logical way. If your methods of thinking or
problem-solving are at all different, thatÂ’s too bad. You have to learn to
think the way the programmer intended you to think.
The second Perl slogan:
Easy things should be easy, and hard things should be possible.
Wall adapted this maxim from the similar pronouncement Alan Kay made in
"Microelectronics and the Personal Computer" in 1977. Wall understood that
90 per cent of most programs were relatively simple and straightforward
tasks. The remaining 10 percent posed the real challenge. Wall designed Perl
to be the ultimate text-processing language, the 90 per cent that is easy. C
is the 100% possible language, and Wall incorporated the C commands
necessary to do bit-level hacking into Perl. But whereas a C program to
print out "hello world" would take five or six lines of definitions and
configurations, the equivalent Perl program looks like this:
print "hello world"
ThatÂ’s what he means by, "Easy things should be easy."
Together, the two slogans reveal the powerful nature of Perl. By following
the first slogan, Perl is able to follow the second slogan for programmers
of different levels of ability. What is easy for a master hacker wonÂ’t
necessarily be easy for a novice. But PerlÂ’s multiple avenues allow
programmers to learn new tricks and skills, developing new abilities and
knowledge over time. Some of the tools in Perl are very straightforward;
others are complex. If a programmer doesnÂ’t understand the complex tool, he
doesnÂ’t have to use it. But he may use it for a simple task at first, then
steadily learn its nuances. Thus over time that which once was hard to
understand now becomes easy to understand–and then easy to program.
These slogans run counter to much of the history of programming languages.
Computer languages are generally designed for either the novice or the
master programmer, not the lazy, impatient, hubristic programmer, who can be
both. The traditional slogans would be: "ThereÂ’s one right way to do it" and
"Everything should be easy–if it’s not, it’s not possible." Some would argue
that the slogan for some languages would be, "Everything should be hard."
Linux
In 1990, the Free Software Foundation had just about finished the GNU
project. They had a complete, best-of-breed operating system, except for one
detail: the kernel. The kernel, as the name implies, implements the core
functionality of the system. ItÂ’s what sits between a running application
above and the computer hardware below. It was the last of the major
essential components the Free Software Foundation needed to build for a
working GNU system. And the GNU kernel development project was moving
terribly slowly.
The GNU people wanted to build a superadvanced kernel with all of the
greatest theoretical advances made practical. But doing so was terribly
difficult: a kernel has to not only work, it also has to work as fast as
possible. And making something both complex and optimally efficient is a
ferocious task. They worked on the kernel, which they dubbed the GNU Hurd,
for years, struggling to make it work.
Like the lazy, impatient, and hubristic Larry Wall, a Finnish graduate
student named Linus Torvalds thought he could do the job. Torvalds, studying
computer science at Helsinki University, just wanted to make something that
would work. It didnÂ’t need to be perfect or beautiful or work on every
machine ever made. In 1991 he released a kernel for the 386 platform under
the GPL, designed to work as part of a complete GNU operating system. He
called the kernel and the system Linux. When Linus first released Linux,
Stallman was still very gung-ho about the Hurd kernel. A lot of computer
scientists thought that the Linux kernel wasn't going to go anywhere.
Stallman believed the consensus, and didn't really mind that Linus was
calling the entire system "Linux," even though it was a GNU operating
system.
>From the perspective of most computer scientists, Linux was awful, ignoring
all the new ideas and just using old, safe, boring ones. And it worked only
on cheap computers. From the perspective of thousands of others, it worked!
On cheap computers! Linus had followed one of the number one rules that Ken
Thompson had followed in creating Unix: Make something that works as soon as
possible–then start fixing it. The Internet allowed people to contribute
quickly and massively to the project, rapidly improving its reliability and
power.
By 1993, the GNU system with Linux as its kernel was comparable to the
commercial Unixes, and it and the free BSD projects killed off the smaller
commercial Unixes. They were both growing, and growing fast. 1989 had been a
crucial year in the history of BSD: the first freely redistributable release
of BSD code, without any licensing restrictions other than preservation of
the copyright notice. The Berkeley developers had separated their code from
the AT&T code–though it took a few more years to write free versions of the
AT&T code. When that happened, the 1990s saw numerous complete BSD systems
as different developers "forked" the code: NetBSD, FreeBSD, and OpenBSD.
Each fork was associated with a set of goals and personalities; the
competition was by turns healthy and disruptive. In 1992, the
AT&T-controlled Unix System Laboratories sued the commercial arm of the BSD
distribution, Berkeley Software Design, to force the halt of any BSD sales.
There were hundreds of pages of briefs, injunction requests, and
counter-suits. Only by January 1994 did both sides come to a settlement. The
suit and the fractious forks considerably harmed the momentum of BSD. Linux,
as its advocates called the entire system, just kept growing and took an
insurmountable lead, soon to be used by not thousands, but millions.
When corporations started getting involved with Linux, Stallman started to
have a problem with Linux developers and users not giving credit to GNU. It
became obvious that the companies that adopted Linux were deliberately
avoiding mention of Stallman, the FSF, the GPL, and GNU–deliberately
avoiding mention of the principles of free software. In response, Stallman
began to push for the system to be called GNU/Linux, a name that properly
describes the nature of the system and is slowly growing in acceptance as
people realize how much the GNU project started it all.
Perls of Wisdom
Larry Wall shares StallmanÂ’s belief in the freedoms of free software; Perl,
of course, has always been free and nonproprietary. The essential difference
is in how Wall goes about proselytizing these beliefs to others.
Wall, with some humor and humility, sees the similarities between his role
as the leader of the Perl community and the prophets in the Bible. His most
famous sermons have taken on the form of the Perl Conference keynotes, which
he came to call the State of the Onion speeches–combining an obvious pun on
the Presidential State of the Union talks with the slightly more subtle, but
no less painful, reference to pearl onions. In each one he reveals his own
talents and interests, including the paths he could have gone down:
linguist, musician, chemist, missionary. He famously wears Hawaiian shirts,
or, for more formal occasions, neon tuxedos. While Stallman is stout with
Jesus hair, Wall is slim, with unruly, fine brown hair, oversized glasses,
and an infectious grin under a large, well-groomed moustache. Wall looks for
all the world like an enthusiastic science teacher toiling away in a public
high school, sticking with a sense of style developed in the 1970s but with
humor and intelligence.
Like Stallman, Wall takes the stand to tell programmers that:
It is better to give than to receive.–"First State of the Perl Onion"
But he believes that Stallman pushes too hard, that his opinions are too
extreme to be the best kind of leader. RMS, from the beginning of the GNU
project, has antagonized people simply by living the nonprofit life. When
the free software movement began, the only way to make money by programming
was to write proprietary software. There was no infrastructure to support a
commercial model for free software. The only possibilities were to
compromise, like Larry Wall did, and work for proprietary projects for a
salary but write free software whenever possible, or to make very little
money, as Stallman did. And Stallman was an especially brilliant hacker
without a family to support.
Only if we have the fundamental right to own information do we also have the
fundamental right to give information away, freely and without coercion.
Simply because we want to, not because we have to.–"First State of the Perl
Onion"
While Wall believes that the chief virtues of a programmer are laziness,
impatience, and hubris, they are the virtues of passionate individuality.
But the antonymic attributes are the virtues of community: diligence,
patience, and humility.
As a prominent figure in the computing community, Larry Wall found himself
drawn to the complex center, where the worlds of proprietary and free
software battle it out. At one end of the free software world, pulling hard
away from the proprietary world, is Richard Stallman. At the opposite end of
the proprietary software world, pulling just as hard away, is Bill Gates.
Comparing this situation to an acetylene molecule, he says:
Let me be specific. Some folks in this room are extremely leery of Bill
[Gates]. Others are extremely leery of Richard [Stallman] . These people
tend to be leery not only of the opposite hydrogen, but also the opposite
carbon. They are supplying the repulsive forces, because they fear the
opposite extreme.–"Third State of the Perl Onion
[[insert Rocheworld diagram]]
What Wall found is that though he lives in the complexity of the free
software world, he is more comfortable near the center. He felt that his
relationship with Tim OÂ’Reilly, who has created a publishing empire founded
upon proprietary materials for programmers (like WallÂ’s own Perl programming
manual, the "Camel Book") is mutually beneficial and to his liking. He
thinks OÂ’Reilly represents the best of the proprietary world, encouraging
the growth of the free software world, and even hiring Wall in 1996 simply
to work on Perl. Any increase in the usage of Perl translates into increased
sales of the OÂ’Reilly-published Perl manuals. Wall wanted the Perl culture
to include both the fanatics, the people who love Perl and only work with
Perl, and the multiple joiners, the missionaries and merchants, who see Perl
as a useful tool among many.
While maintaining all these virtues seems paradoxical, Wall asserts that
they are not. What is required is an ability to change oneÂ’s perspective.
There are times to ignore oneÂ’s natural impulses, and be humble and patient.
But there are also times to trust oneÂ’s instincts. While this seems like
common sense, most programmers rarely use common sense. They prefer logical
systems that give only one right answer. By promoting a complex, paradoxical
way of looking at things, Wall is challenging technologists to transform
their philosophy of life. With Perl, he is promoting a balanced, flexible,
and organic philosophy in contrast to the rule-driven, inflexible, and
mechanical philosophy that has dominated both computing and theology.
To that end, Wall sees that everything is interconnected and even
intertwingled. He admires messiness; complexity. After all, he designed Perl
to be two things: a text-processing language and a glue language. Glue works
because it is messy–because it is a complex chemical soup, able to interact
and bind with all kinds of surfaces. Perl, with its complex and redundant
design, fills in the cracks like glue, with all of its messiness and
strength. Neither glue nor Perl solves every problem, but they let you bring
things together.
If there's a germ of an important idea in Perl Culture, it's this: that too
much control is just as deadly as too little control. We need control and we
need chaos. We need order, and disorder. Simplicity, and complexity.
Carefulness, and recklessness. Poise, and panic. Science, and art.–"First
State of the Perl Onion"
[[insert the messiness picture]]
The trick is that as long as you can find the patterns within complexity,
the messiness ceases to be a problem. You can try to follow all the air
molecules bouncing around a room or you can understand them usefully in
terms of temperature and pressure. You can try to understand the entire
World Wide Web or follow your interests to get the answers you need. As long
as youÂ’re given the tools to find what you need amidst the messiness, as
long as there are some rules, even if theyÂ’re arbitrary or contradictory,
you can navigate successfully. As Marvin Minsky tells us in Society of Mind,
human minds can handle complexity by dint of our agents that can "unstick"
it. This applies to everything from programming code to reprogramming our
minds.
[Insert the messiness picture with shapes revealed]
Afterward
>From the perspective of the outside world, Richard Stallman and Larry Wall
seem identical. They are both programmers, organizers, and proselytizers of
two of the largest and most popular free software projects in the world: GNU
and Perl. They are popular speakers and prolific expositors of an
alternative to the world of commercial software development. They love their
work.
And yet the differences abound. Stallman thinks people like Tim OÂ’Reilly are
opportunistic corrupters of a good system; Wall thinks people like publisher
Tim OÂ’Reilly are the best thing that could happen to non-proprietary
software. Stallman founded, lives, and breathes the Free Software Movement;
Wall sometimes compares them to communists. What that really means is that
Stallman is perfectly happy to use a carrot-and-stick approach to expanding
the world of free software, while Wall prefers to use the carrot more than
the stick. While both attach the GPL to their work, Wall sees nothing wrong
in supporting the proprietary model of the OÂ’Reilly manuals, as they
complement PerlÂ’s free, online documentation. Because good documentation is
essential to the use of complex programs, Stallman thinks that making such
documentation unshareable is wrong.
Wall believes that the coercive nature of StallmanÂ’s GPL is indicative of
StallmanÂ’s atheism. He believes that a sense of morality and obligation
derives from a belief in God, not from social pressures. He sees the
corporate attempts to control individual behavior and the communist attempts
to form collectives as two sides of the same coin. He has even called it a
"hive mentality," a science-fiction concept in which people lose all sense
of identity and act like ants or bees in a colony, mindlessly fulfilling the
goals of the collective. Wall finds StallmanÂ’s moral attitude to be
impractical and polarizing, promoting conflict instead of peace.
His own moral guidance comes from the Christian belief that every person is
valuable in GodÂ’s eye, but at the same time should strive to be humble.
Larry WallÂ’s measure of success is intelligent cooperation and compassion,
not just for those who agree with him, but for those who disagree. Whereas
Stallman has drawn the battlelines in the sand, Wall strives to achieve
redemptive subversion from the heart outward, whether subverting a bad
person, a bad company, or a bad culture.
Richard Stallman, motivated by the powerful emotions of the downtrodden, the
dismissed, the disdained, has created a counterbalance to the world of
proprietary software in the GNU project and the Free Software Foundation. He
doesnÂ’t believe his efforts support a hive mentality, or that people
shouldnÂ’t be selfish or celebrate individuality. Rather, he is against
giving individuals the power to dominate and divide others. In the ensuing
years this quiet and shy man has slowly come out of his shell, a
near-religious figure, simultaneously loved and reviled.
In the course of standing up to Symbolics, starting the GNU project, and
leading the free software fight, Richard Stallman discovered that he could
do more than program. He has slowly gained the confidence to tackle problems
that arenÂ’t algorithmic. He now actively supports the political causes that
he feels are connected to the ideals behind free software–from abortion
rights to marijuana legalization. He believes the forces that destroyed his
hacker community, the placement of profit over people, are destroying
communities all over the globe. Nike sweatshops in Indonesia, police
brutality and racial profiling, first-world nations preventing poor
countries from manufacturing inexpensive AIDS medication–all abuses of power
by institutions over people. He sees a direct connection between the free
software movement and the global protests against the trade agreements ruled
by corporate interests and dehumanizing economic policies. Stallman has
nothing against entrepeneurialism and business–but he sees little good in
the massive, unchecked power wielded by the gigantic multi-national
conglomerates with interests utterly removed from that of the freedom of
individuals.
To both Stallman and Wall, programming is primarily an artistic endeavor, an
act of intellectual joy and love. They program because it affords them
freedom: freedom to experiment, freedom to create, freedom to share.
Programming is both an individual and collaborative act. Stallman found his
greatest happiness in the community of the AI Lab hackers. Wall has found
his in fostering the Perl community. In fact, both spend as much time now,
if not more, leading, goading, and encouraging the free software community
as they do just coding.[7]
The two master programmers live at the intersection of art and business. To
program a computer is to communicate with it, to learn its language, to
teach it language. As they became leaders of a movement with a strong impact
on society, they have begun to build a culture that can handle the
complexities of technology and the complexities of human relationships. And
the greatest challenge to that culture are the entrenched and powerful
corporations who have built empires on proprietary software and trade
secrets.
Footnotes
1 Fortran, COBOL, LISP and Algol all started out in the 1950s. BASIC and
Pascal were developed in the 60s.
2 Of course, this analogy breaks down in some important ways; namely, any
program, be it source code or machine code, can be duplicated exactly, as
it’s pure information. There’s no replicator machine for food–yet. See
Chapter 12.
3 GNU would be compatible with Unix, but none of the code would come from
the AT&T-controlled Unix. Thus, GNUÂ’s Not Unix.
4 Nowadays Stallman dislikes people talking about intellectual property
rights collectively. People are too apt to confuse the nature and effects of
copyright, patents, and trademarks. They donÂ’t ask that question, "What kind
of act are we licensing a person to do?"
5 The GNU GPL in its initial incarnation was the GNU Emacs General Public
License; then there was the GNU GCC General Public License, as well as GPLÂ’s
for BISON, NetHack, and the other GNU applications. On February 1, 1989 the
Free Software Foundation released the GNU General Public License, not tied
to any particular application.
6 When Wall first began at the company, it was the System Development
Corporation, later bought by Burroughs. In November 1986 Burroughs merged
with Sperry to become Unisys.
7 Larry Wall is currently leading the effort to build Perl 6, a complete
rewrite of Perl that will take years.
References
Richard Stallman, "The GNU Manifesto" 1985
http://www.gnu.org/gnu/manifesto.html.
Richard M. Stallman, "GNU General Public License," 1989.
http://www.gnu.org/licenses/gpl.html.
Stallman, The Free Software Definition, 1996.
http://www.gnu.org/philosophy/free-sw.html.
Larry Wall, Programming Perl (Sebastopol, Calif.: OÂ’Reilly & Associates,
1991) 4.
Larry Wall Programming Perl, Second Edition (Sebastopol, Calif.: OÂ’Reilly &
Associates, 1996) ix. Though this slogan first appeared in the second
edition of the Camel book, the idea was expressed, albeit more diffusely, in
the first edition.
Larry Wall "First Perl Conference keynote", 1997, also known as the "First
State of the Perl Onion." http://www.wall.org/~larry/keynote/keynote.html
Larry Wall, "Second State of the Onion," 1998.
http://www.wall.org/~larry/onion/onion.html
Larry Wall, "Third State of the Perl Onion," 1999.
http://www.perl.com/1999/08/onion/talk.html
Chapter Eleven
The ®evolution and the Law
Eric S. Raymond and Lawrence Lessig
Undoubtedly, it would be very pleasant for us mice if
the predatory cats of the world were to be belled,
but–who is going to do it? Who is going to assure us
that ruthless power will not find its way back into the
hands of those most avid for it?–Norbert Wiener
The philosophy behind Richard Stallman and Larry WallÂ’s free software has
implications far beyond computers. It is a new model for the creation,
distribution, and regulation of knowledge. It incorporates Norbert WienerÂ’s
belief in the circulation of information through regulatory feedback. It
embraces Doug EngelbartÂ’s theory that technology can help people
communicate; that humankind will evolve by sharing problems and solutions.
It supports Seymour Papert and Alan KayÂ’s conviction that people work best
when they are equipped to think globally and experiment freely. It eases the
problems of information overload by providing a structure in which countless
people can bootstrap an open system in an open society. It holds the promise
to be the infrastructure of Marshall McLuhanÂ’s global village, transforming
business, government, and technology alike.
Despite its promise, many forces oppose the nonproprietary development of
programming code. The World Wide Web is one testament to this, as it morphed
from a research center into a sphere of economic activity. The Code of the
Net has changed, symbolically and in actuality, since Tim Berners-Lee
introduced it in the early 1990s. Exclusively Web businesses, the dotcoms,
spontaneously generated into being, offering everything from books to
cut-rate electronics, twenty-four hour junk food delivery, digital design,
pornography, and dirt-cheap airline tickets. The corporations that own
commercial territories like AOL and shopping site can’t–or won’t–operate in
the old way. They can track people through their credit card information.
Access to sites can be barred. Users began to realize that corporate and
governmental interests could threaten their rights or privacy and free
speech by implementing new, proprietary architectures of control.
When Norbert Wiener applied cybernetics to human systems—society—he came to
the conclusion that information is meant to be free—not in the monetary
sense, but in the sense of unhampered exchange. He adopted the Greek word
kubernetes, meaning ‘steersman,’ to imply the way in which a system must be
regulated so that the flow of information (whether nerve cells in a body or
electronic pulses in a machine) can function optimally. Some may see a
contradiction in this: It means that a system needs to be regulated in order
to be "free." Regulation can mean a degree of governmental control. Or it
could mean a sort of free-market self-regulation, depending upon your point
of view.
In the mid-90s a legal scholar, L. Lawrence Lessig III, and a software
activist, Eric S. Raymond, contemplated the laws of cyberspace. Raymond and
Lessig started to wonder: To what extent, then, is the Web really free? In
what direction are we headed? How does control, regulation, affect the
Internet and the Web?
Lessig has aimed for a balance between private and public interests and the
enforcement of the rights of privacy and free speech through good
governmental regulation. Raymond, who also calls himself a "wandering
anthropologist and troublemaking philosopher," has adopted a development
method on efficiency grounds. His belief is that the open Internet can fuel
the open market, which translates into personal freedoms and the
obsolescence of government. Raymond’s and Lessig’s manifestos–The Cathedral
and the Bazaar (1997) and Code and Other Laws of Cyberspace (1999),
respectively–express opposing and overlapping visions concerning the
implications of open code on the Net. Both stand against the dramatic
backdrop of the biggest power struggles of the Information Revolution to
date, involving governmental regulation of information and the power of
major corporations like Microsoft and AOL Time Warner.
The Bazaar
Eric S. Raymond (best known by his handle "ESR") shares a lot in common with
Ted Nelson and Richard Stallman (RMS). All three realized early on that the
way information is structured, delivered, and accessed on a computer can be
intensely political. They believe knowledge should be shared. They detest
the idea of any authority barring access to information. They are extreme
personalities who dabble in eccentric pursuits. Raymond, a martial arts
aficionado, devises epic Viking/Anglo-Saxon poetry for live-action role
playing games. Modeling himself as a character in HeinleinÂ’s science
fiction, he provides a link on his Web page to his gun rights site where he
rhapsodizes about the ethics of gun-wielding–"It all comes down to you . . .
No one elseÂ’s finger is on the trigger but your own. . . ." Stallman, a
fellow science fiction fan, lists his hobbies as flying, folk dancing, and
physics. Nelson has attended Californian New Age workshops. Raymond suffers
from congenital cerebral palsy. Stallman suffers from a leg injury. Nelson
suffers from anxiety and depression. All three are often found in T-shirts.
Raymond, Stallman, and Ted Nelson are all master propagators of the hacker
myth. In the traditon of the hackers in StallmanÂ’s AI Lab they have
fortified ordinary computer users with a heroic mission: Defend your rights!
They have written manifestos that have not only shaped the computer
fanaticÂ’s self-image, but also, in so doing, the worldÂ’s hardware and
software environments. Between them, they have created the legend of the
heroic hacker as a rogue individual who intrepidly fights against massive
and menacing power, usually governmental or corporate.
In the 1987 update to Computer Lib, Ted Nelson asserted: NOW YOU CAN BE
OPPRESSED BY COMPUTERS IN YOUR LIVING ROOM. To Nelson, the enemies were
authoritarian companies like IBM, the government, and all others who had
vested interests in either harvesting private information or forcing the
public into one stagnant paradigm. To Stallman, the evil is the repression
of the hacker community for the sake of proprietary software. To Raymond (in
the 1990s and 2000s), the enemy is the monolithic entity–quintessentially
Microsoft or the government. RaymondÂ’s manifesto, The Cathedral and the
Bazaar, addresses the folly of the hierarchical corporation (the
"Cathedral") and the viability of an open code alternative (the "Bazaar").
I believed that the most important software . . . needed to be built like
cathedrals, carefully crafted by individual wizards or small bands of mages
working in splendid isolation, with no beta to be released before its time.
Linus Torvalds's style of development–release early and often, delegate
everything you can, be open to the point of promiscuity–came as a surprise.
No quiet, reverent cathedral-building here–rather, the Linux community
seemed to resemble a great babbling bazaar of differing agendas and
approaches . . . out of which a coherent and stable system could seemingly
emerge only by a succession of miracles.–The Cathedral and the Bazaar
Eric S. Raymond published the first draft of The Cathedral and the Bazaar in
1997, a year before he came up with an alternative name for free software:
open source ("the bazaar"). Christine Peterson, the executive director of
the Foresight Institute, a nanotechnology think tank, coined the actual
term. A group of like-minded free software advocates, including Tim OÂ’Reilly
and Larry Wall enthusiastically supported RaymondÂ’s initiative.
Although Richard Stallman repeatedly said information wants to be "free" as
in free speech–not free beer–outsiders were turned off by the Free Software
FoundationÂ’s openly political mission to fight proprietary software.
Raymond, although one of the first contributors to the GNU Project, stated,
"In the battle we are fighting now, ideology is just a handicap. We need to
be making arguments based on economics and development processes and
expected return. We do not need to behave like Communards pumping our fists
on the barricades. This is a losing strategy." He and his colleagues
considered open source a more marketable term than "free software," with
fewer anticommercial connotations. The Open Source Definition offers
companies more leeway than StallmanÂ’s GPL, because one may incorporate open
source code into proprietary software. The open source premise of the
voluntary renunciation of copyright is justified by market economics
(although not necessarily driven by economic factors). The two camps are
functionally similar but philosophically distinct. The Free Software
Movement is based on the argument that source code should be available
because it harms society to do otherwise. The Open Source promoters argue
that source code should be available because it is of maximal economic
benefit.
Rules of the Bazaar
The leading open source software of the late 1990s and early 2000s is the
GNU operating system, overwhelmingly in its Linux incarnation, initiated by
its low-key yet charismatic leader, Linus Torvalds.[1] Before Linux, Raymond
thought major software could only be developed in "cathedrals," by
hacker-priests or arcane corporate bureaucracies. But a wild gaggle of
volunteers, the "bazaar," developed Linux–and it worked amazingly well.
Somehow it tapped the collective IQ that Doug Engelbart sought through his
NLS system. So Raymond sought to discover how the bazaar worked, against all
expectations. He tried it with a free software project of his own (the
fetchmail program), and proselytized it in The Cathedral and the Bazaar. His
manifesto features practical maxims:
"Release early. Release often. And listen to your customers"
"Perfection (in design) is achieved not when there is nothing more to add,
but rather when thereÂ’s nothing more to take away."[2]
"Given enough eyeballs, all bugs are shallow."
"The Cathedral and the Bazaar" made open source plausible for any project
hooked into the Net. Raymond disclosed that his open source fetchmail
project alone fetched more than 600 contributors–1200 eyeballs. What
closed-source developer could compete with that? Microsoft couldnÂ’t hire 600
"cathedral builders" to work on just one piece of a system–the equivalent of
one stained glass window.
In the cathedral-builder view of programming, bugs and development problems
are tricky, insidious deep phenomena. It takes months of scrutiny by a
dedicated few to develop confidence that youÂ’ve winkled them all out. . . .
In the bazaar view, on the other hand, you assume that bugs are generally
shallow phenomena–or, at least, that they turn shallow quickly when exposed
to a thousand eager co-developers pounding on every new release.–The
Cathedral and the Bazaar
Bazaar-style development is an organic process executed on an unprecedented
scale; it only works under the proper conditions. The bazaar cannot be
complete anarchy. For one, Raymond noted that an open source project cannot
be initiated in a ground-up grassroots style. A project must originate with
a Stallmanesque or Torvaldian inspiration: one person (or group of people)
with a kernel of a program and a vision for growing it. If he or she builds
it–creates a stable foundation and guides the construction–they will come.
After all, even a populist uprising must have its leaders. GNU wouldnÂ’t
exist without Stallman and those who have been inspired by his ideas and
code. Linux wouldnÂ’t exist if it werenÂ’t for both its humble, encouraging,
creative Finnish leader and his cadre of developers. Nor would fetchmail (or
perhaps the programmer-as-hero myth) exist without the flamboyant Eric S.
Raymond.
On the other hand, the open source bazaar encourages ground-up work in
project execution: testing, debugging, developing and overall code
improvement. As with Alan KayÂ’s Smalltalk, organic modularity is key to the
construction: Programmers can swap and switch organelles of code with
greater ease. A hierarchy of sorts is implicit, at least in the sense that
each project should have a leader/coordinator–a charismatic, modest, wise
individual who can gently guide a community of free-thinking programmers
willing to work for, well, free.
Raymond believes that in open source an internal market ultimately selects
what leaders and projects will be successful. The community chooses its own
leaders by only developing projects headed by the successful coordinators.
This, he asserts, is a new leadership style that is not based on a power
relationship, but on consensus.
"Treating Your Users as Co-Developers is Your Least-Hassle Route to Rapid
Code Improvement and Effective Debugging"
In cathedral-style development, the world is divided into two: developers
and users. Only those inside the cathedral get to see and manipulate the
secret source code–everyone outside must accept what is sent out. In the
bazaar, everyone, from the master hacker to the newbie user, is allowed to
see the source. Any user can be a developer. Everyone can contribute–and if
the project is properly designed, everyone will.
Just as the leader/coordinator of an open source project isnÂ’t a tyrannical
master, the thousands of users-developers building it arenÂ’t drones. Nor are
they completely altruistic, Raymond notes. Richard Stallman exhorts fellow
programmers to write free code from moral incentives–a conviction that
knowledge should be shared, not hoarded. Raymond emphasizes more selfish
principles. His stand is that the pragmatic programmer is likely to not mind
proprietary commercialism as much as Stallman. He or she simply supports
open source to use and develop better tools than ever possible in a
proprietary arrangement.
To many programmers, the fun of the collaborative coding challenge is an
incentive enough. The fun factor, after all, is what Ted Nelson emphasized
when he wrote Computer Lib and Alan Kay when he and Adele Goldberg wrote "A
Dynamic Personal Media." Using computers and writing code should be an
enjoyable, deeply creative and personally rewarding experience, as Papert
found with LOGO. Just as a fair guarantees better turnout than a Sunday
church service, the bazaar atmosphere of an open source project attracts
more people because they have fun. Marshall McLuhan and Abbie HoffmanÂ’s
dream that the electronic age would make drudgery and work obsolete is
coming true.
It may turn out that one of the most important effects of open sourceÂ’s
success will be to teach us that play is the most economically efficient
mode of creative work. –The Cathedral and the Bazaar
* * *
The principle Raymond emphasizes the most is status or "egoboo"–a term he
took from science-fiction fandom, meaning ego boost. A good programmer gets
credit from his or her peers. Raymond, the amateur anthropologist, explains
that the drive to compete for social status is hardwired into human culture.
Programmers have a gift culture, not unlike aboriginal cultures that live in
places where the climate is mild and the food is abundant. As a result,
status is more important than ownership. And status is defined by what one
gives away as a gift. In return for the best gifts, the best code, one gets
egoboo. And greater support and participation for oneÂ’s own projects.
Sometimes "gift-giving" translates into material reward. But egoboo is a
currency unto itself–perhaps a more valuable one than monetary when other
human needs are already fulfilled; most gift-giving programmers are
gainfully employed in some capacity. The theoretical underpinning of this is
straight from the 1960s: The question is no longer how one makes money, but
how to satisfy oneÂ’s psychological needs.
. . . By properly rewarding the egos of many other hackers, a strong
developer/coordinator can use the Internet to capture the benefits of having
lots of co-developers without having a project collapse into a chaotic
mess.–The Cathedral and the Bazaar
The Cathedral and the Bazaar was written so that a person with an MBA could
appreciate the benefits of open source. Through the open source model the
code would benefit from thousands of contributors catching bugs and
investing time, energy and belief in the product. Good open source software
never dies–new contributors keep improving it year after year. You don’t
need management to cajole and interest their underlings–egoboo provides a
stronger motivation than whip-cracking. Once the project becomes a must-have
application, other people would do the work of porting it to new
technologies.
The publication of The Cathedral and the Bazaar online and in book form
(OÂ’Reilly, 1999) resulted in a convincing and resounding triumph for the
bazaar–and ESR himself. His supporters abbreviated it to "CatB." CatB’s
readership extended far and wide online and offline. On January 22, 1998,
about eight months after the paper debuted, Eric Hahn, Executive V.P. and
CTO of Netscape Communications, made an unprecedented announcement: They
would give away the source code for Netscape Communicator. Hahn emailed
Raymond to thank him for the thoughts behind The Cathedral and the Bazaar,
informing him that CatB was an inspirational influence behind the decision.
Netscape executives called him to hire him as a consultant (along with
Torvalds and Stallman). Raymond hung up the phone in a daze. It took him
some time to be able to think, but he soon realized how big this was. The
open source movement just got the proverbial key to the executive bathroom.
But he knew that if Netscape went down the toilet, so could any credibility
for open source.
Either way, it was significant egoboo for ESR.
Netscape
Unlike Tim Berners-Lee, Jim Clark, and Marc Andreessen intended to develop a
commercial browser for the Web. While Berners-LeeÂ’s inspiration was pure
research, Andreessen and Clark focused on development, marketing and
customer service. In 1994 they founded Mosaic Communications Corporation.
Later that year they renamed themselves Netscape and changed their browserÂ’s
name from Mosaic to Navigator. Navigator, downloadable for free, accelerated
the development of the Web by many orders of magnitude. It rapidly became
the dominant Web browser, easily fending off even MicrosoftÂ’s attempt at
competition in 1995, Internet Explorer.
In 1998, Microsoft, initially behind the curve on the Internet, announced
its intention to integrate Explorer 4.0 in its new operating system, Windows
98. It would be part of the desktop on everyoneÂ’s computer. The move, if
successful, could demolish Netscape. Clark and Andreessen hoped that going
open source would gain Netscape a wider market share over Explorer.
Clark and Andreessen were convinced that they could harness the power of
open source and still make a profit by becoming service-based. This
reflected RaymondÂ’s belief that software is inherently a service-based
business that has been viewed and treated as a manufacturing-type business.
The first major service-based business success was Red Hat–a GNU/Linux
distributor and service provider completed its initial public offering in
1999 to great success. Although Red Hat has no proprietary claim whatsoever
on the code its is distributing and supporting–anyone can do the same
thing–the company thrived by building its own reputation and brand. That
reputation allows Red Hat to keep the best Linux developers on its payroll,
so that it wonÂ’t fall behind any other company. With the new open
source-based business model, corporations like Red Hat are turning a profit
and synergistically supporting Linux. IBM pledged $1 million to Linux
development, to revive sales of their mainframes and become the world's
biggest software service provider. Raymond was right–the Netscape deal got
open source into even the most corporate of all boardrooms.
While Microsoft bet on its monopolistic cathedral, Netscape threw open its
gates to the bazaar. Netscape set up the non-profit Mozilla.org, to
coordinate the development process both inside and outside the corporation
and to maintain the source code. NetscapeÂ’s paid engineers spearheaded the
project and the larger Web community contributes more as the process
develops. The only way Netscape could compete with Bill Gates, Raymond
reckoned, was to change the rules.
Just as Norbert Wiener asserted that information cannot be locked in secret
vaults for long, so Raymond insists that NetscapeÂ’s move toward open source
will inevitably happen throughout the industry. In RaymondÂ’s opinion
closed-source Microsoft is doomed because its profit incentive is founded
upon trade secrets, not improved software. A company like Microsoft tries to
spend as little as possible on supporting its software while making the
consumer believe heÂ’s getting his moneyÂ’s worth. Raymond, like Stallman,
believes itÂ’s only a matter of time before consumers wake up and realize
that they are being scammed–and that there is a better alternative.
Lessig and Goliath
The Netscape vs. Microsoft story is quintessentially the struggle between
the piranhic corporation versus the innovative small fish. By 1998
MicrosoftÂ’s plan to integrate (to bundle) Internet Explorer into its Windows
98 operating system jumpstarted an antitrust case with the U.S. Department
of Justice and nineteen states. Microsoft asserted that an integrated
browser worked better with other Microsoft applications. The Justice
Department believed it was a classic use of one monopoly, in operating
systems, to establish another, in Web browsers–a violation of the Sherman
Antitrust Law of 1890.[3]
Could all the promise of open source actually protect Netscape from
extinction more than 80 percent of all desktops were Windows systems and
Explorer becomes part of the Windows code? What incentive would users have
to download Navigator if Explorer were the default browser on their desktop,
removable only by extreme measures? Customers couldnÂ’t really choose for
themselves.
The person driving the case against Microsoft in the court of public opinion
was the young, wiry, bespectacled Harvard professor named Lawrence Lessig.
Like Eric Raymond, Lessig studied philosophy at the University of
Pennsylvania, and later entered a field dedicated to revealing the effects
of technology on society. Raymond decided to enter it through his work as a
programmer. Lessig entered it as a lawyer.
Although he was born in South Dakota, Lessig grew up in the small town of
Williamsport, Pennsylvania, best known as the home of the Little League
World Series. After graduating from Penn with a degree in economics, he
followed Alan Turing, Marshall McLuhan, and Seymour Papert to Cambridge
University, getting a masterÂ’s in philosophy in 1986. He then went on to
Yale Law School and quickly distinguished himself as a top constitutional
lawyer after graduating in 1989. He clerked for Supreme Court Justice
Antonin Scalia as the token liberal surrounded by collegially hostile
conservatives. At the age of thirty, he joined the faculty at the University
of Chicago, later Harvard and finally Stanford Law School. With his
involvement in the Microsoft case and related writings, Lessig became the
definitive legal expert on the Internet and computing.
LessigÂ’s life has informed and reflected his beliefs about the role of
government in society. In 1985, while at Cambridge University, he smuggled a
heart valve to a Jewish dissident behind the Iron Curtain in the Soviet
Union. His wife, Bettina Neuefeind, investigated the genocidal crimes of the
Yugoslavian government in Kosovo. Lessig has learned firsthand the lasting
and personal effects of the Cold War. He saw how political and legal
doctrine is not a vague, impersonal theory but the architecture, lasting for
generations, which can determine whether people live or die. He could see
that the determining architecture of the future lay within the electronic
musings of computers.
* * *
Lessig advised the Microsoft caseÂ’s district court judge, Thomas Penfield
Jackson, on the legal and technological issues of the case. Lessig compared
Microsoft to the Ma Bell monopoly, and declared Microsoft illegal for
annihilating its competition in the browser market. He argued that
MicrosoftÂ’s browser and operating system should be regarded as two separate
and discrete products, despite MicrosoftÂ’s assertions to the contrary.
In 1999 Lessig published the best-selling Code and Other Laws of Cyberspace.
In Code the self-proclaimed constitutionalist argues that government should
intelligently regulate the Internet to prevent evils such as monopolies and
infringements of individual rights like privacy and free speech. Raymond and
other cyberlibertarians believe that the Net is an inherently free and open
space that should be devoid of any sort of regulation. Lessig argues that
the absence of regulation is not freedom. Lessig wanted the Internet to
continue to be a place of freedom as Licklider and Berners-Lee intended–not
a place of oppressive control, ruled by monopolistic corporations like
Microsoft. We need laws for the Net that protect the freedoms of the
people–to the greatest extent possible. We need the law to step in
occasionally.
Liberty in cyberspace will not come from the absence of the state. . . . We
build a world where freedom can flourish not by removing from society any
self-conscious control; we build a world where freedom can flourish by
setting it in a place where a particular kind of self-conscious control
survives. . . . Foundations get laid, they donÂ’t magically appear . . . We
are beginning to see in cyberspace that this building, or laying, is not the
work of an invisible hand. . . . Cyberspace, left to itself, will not
fulfill the promise of freedom. Left to itself, cyberspace will become the
perfect tool of control. –Code and Other Laws of Cyberspace
Lessig thought Microsoft represented every possible threat to a free
cyberspace. He joked in an email message about "selling his soul" by using
Microsoft software. He accused the company of goading and threatening
software rivals and hardware makers like Compaq by manipulating licensing
agreements and access to codes needed to make non-Microsoft programs
compatible with Windows. Microsoft ferociously attacked Lessig, declaring
him biased and improperly appointed as "special master" to the judge. The
judge rejected the charge.
In LessigÂ’s "friend of the court" letter he notes that the courts had been
hesitant to apply centuries of legal doctrine to the software industry
because they maintain the view that "code is different." Lessig believed
that "it is a mistake to fetishize code in this way." Computer technology
might seem like magic, but it isnÂ’t.
On June 7, 2000, with LessigÂ’s counsel, Judge Thomas Penfield Jackson ruled
that Microsoft did indeed violate the antitrust laws. He compared the
corporation to a drug gang and likened Bill Gates to Napoleon. He ordered
the company to be split in two and placed restrictions on its business
practices The breakup would have placed the Windows operating system in one
company and created a second business for everything else, including
software applications like Internet Explorer. Judge Jackson also set a long
list of restrictions on MicrosoftÂ’s conduct to last three years if the
breakup order withstood appeal, and ten years if it didnÂ’t. Microsoft, of
course, immediately appealed.
On June 28, 2001, the appeals court determined that Microsoft had violated
the Sherman Antitrust Act and improperly commingled software code of the
browser and operating system. They reversed the decision, however, to split
Microsoft into two separate companies, and reproved Jackson for opining to
reporters about the case. They sent the case down to a new judge to
determine the remedy for MicrosoftÂ’s multiple violations.
Navigator Loses Its Way
While Microsoft was on the stand for antitrust in the browser war, Netscape
and Mozilla proceeded with its open source strategy. In 1999 Eric Raymond
described Netscape and Mozilla as only a "qualified success" for open
source. Mozilla, Raymond claimed, violated one of the foremost rules of open
source outlined in The Cathedral and the Bazaar: Release early and release
often. Jamie Zawinski, one of the key project leaders, resigned on MozillaÂ’s
one-year anniversary, declaring that Netscape was not a truly
network-collaborative project. Not enough outside programmers participated
in project development, Zawinski complained that Netscape would not go along
with what he wanted to do: Ship end-user software. His famous last words:
"Open source is not magic pixie dust."
What went wrong with open source? Zawinski, who like ESR and RMS went by his
initials, JWZ, feared it was the fact that, unlike other open source
projects, Mozilla was too much a subsidiary of Netscape, a corporate entity.
In his tenure at Mozilla Zawinski tried hard to convince hackers that he and
his fellow Netscape Communications employees were just humble guides. The
real Mozilla project existed in the hands of the community at large.
In November 1998 the unthinkable happened: AOL, AmericaÂ’s most popular
Internet service provider, and a corporate behemoth on the order of
Microsoft, bought out Netscape Communications for $4 billion. The deal was
good for investors, but not for users, who knew that large corporate mergers
are detrimental to personal freedoms on the Net. Zawinski at first hoped
that AOL would be enlightened about Mozilla, and not act as a "shambling
inhuman beast of a corporation." But his hopes were soon dashed. Thousands
of Netscape employees followed the departure of Netscape founder Marc
Andreessen, the charismatic leader of NetscapeÂ’s unconventional culture.
Zawinski, after tendering his resignation, fomented against Netscape for
merging into AOL. The Internet was changing from a place of freedom to a
place of oppression. He wrote:
AOL is about centralization and control of content. . .
. Everything that is good about the Internet, everything
that differentiates it from television, is about
empowerment of the individual. I don't want to be a part
of an effort that could result in the elimination of all
that.
In January 2000 AOL became an even more shambling, inhuman beast with the
acquisition of Time Warner, the worldÂ’s largest media company and number one
cable provider, for about $163 billion in stock. The deal was the biggest
merger ever. By the time MicrosoftÂ’s antitrust case was appealed in 2001,
the browser war seemed over. As Mozilla slowly developed, Microsoft
exploited its power to give Internet Explorer a dominating lead in market
share. Netscape became a minor province in Time WarnerÂ’s media empire, from
magazines and movie studios to cartoons and cable networks.
In the aftermath of the merger, Zawinski voiced that the AOL-Time Warner
acquisition is an atrocious example of centralization and control of all
media. AOL, he feared, would become the primary channel by which the public
receives information. Of this he said:
It leads inevitably to a reduction of choices and a
reduction of viewpoints that can become heard. That kind
of control of the entire communication infrastructure,
from content creation, through marketing, to end-user
delivery, is just a disaster as far as true Democracy
goes.
The Architecture and the Message
In 1948 Norbert Wiener warned us that it is when
The Lords of Things as They Are protect themselves from
hunger by wealth, from public opinion by privacy and
anonymity, from private criticism by the laws of libel
and the possession of the means of communication, that
ruthlessness can reach its most sublime levels. Of all
these . . . the control of the means of communication is
the most effective and the most important.
Abbie Hoffman fought against the Lords of the Things as They Are a
generation later when television became the dominant means of communication.
Can the Net escape the same fate, of becoming under the control of the Lords
of totalitarian multinational corporations and governmental bureaucracies?
How are the Lords making the Net controllable, or in LessigÂ’s terms,
regulable?
The institutions are making the Net regulable, Lessig discovered, under the
umbrella of commerce. Companies built commerce-enabling architectures on top
of and into the free Net of Licklider and Berners-Lee, ignoring or
deliberately causing harm to individual freedoms. One way commerce-enabling
architectures affect the common citizen is through proprietary
identification technologies, such as "cookies." Companies prefer dealing
with people they can identify–so they can tailor their product to consumers’
spending patterns and special interests. The more information they have,
from your favorite color to your birth date, fingerprint, and social
security number, the better. People, in the role of consumers, have reason
to give away their information and lose anonymity–access to sites, secure
transactions, and personalization. By selling products to you, companies get
to own a bit of you.
The more powerful and prevalent the corporation, the more information it can
attain about you. Lessig provides the example of America Online. Users must
log into AOL using a password. AOL can identify who you are and track your
postings and purchases. AOL hires content police to ensure your behavior is
enforced. AOL can block communication between certain members. AOL can
monitor your every move. AOLÂ’s code can not be changed by its members. AOL
limits its crowds to twenty-three people at a time. AOL is not a democracy;
the management controls the space. Not to mention knowing your spending
patterns and special interests and applying them in commercially useful ways
throughout the AOL Time Warner media empire. The software code is the legal
code and it is a tool of control. In return, AOL provides conveniences to
its members like informing them when other members are online and catering
to individual interests.
Life will be easier for those who carry a [digital] ID than for those who
donÂ’t. Servers will make exchanges cheaper, or simpler, if data can be
authenticated. Just as it is easier to accept cookies automatically, so too
will it be easier to authenticate facts about yourself. Life in an
authenticating world will be simpler for those who authenticate.–Code and
Other Laws of Cyberspace
Government, Lessig claims, may join forces with commerce to build this
architecture of regulation on the Net. While the government may not regulate
the Net directly, it can through commercial entities–by regulating them. The
government could regulate gambling sites (by requiring that a digital ID
verifies that the user lives in a state where gambling is legal) which, in
turn, would force the market to regulate users on the Net. But why stop at
gambling sites? All sorts of information pass over the Net which interest
the government–political opinions, classified data, even songs and
mathematical equations can all be regulated by the government.
Any sort of filtering or agent technology can operate hand in hand with
digital IDs to cripple the tenets of democracy. W3CÂ’s PICs software for
content selection, championed by Berners-Lee for parents to censor the web
for their children, is one such filter. It censors the Net just as an
old-fashioned English butler would bar entry to an estate. Lessig realized
the technology enables users to ban anything that doesnÂ’t match their
personal ideologies (for example, only viewing sites approved by the
Christian Right). A government or corporation could make filters part of the
architecture of the Net. Such zoning and censorship could occur at any
level–hardwired into the browser or into the ISP. The Net itself will become
more fragmented, more subservient to special interests, less cohesive.
If the Net is so corrupted, the technologies imagined and developed by
Engelbart, Kay and Nelson would actually work against their ideals. The Net
wouldnÂ’t be a tool for dialogue, collaboration, and democracy. If
precautionary steps arenÂ’t taken, the Net could actually become
anti-democratic.
What happens if everyone can, in effect, have a butler? Would such a world
be consistent with the values of the First Amendment? . . . What values
should we choose? In my view, we should not opt for perfect filtering. We
should not design for the most efficient system of censoring–or, at least,
we should not do this in a way that allows invisible upstream filtering. Nor
should we opt for overfiltering so long as the tendency worldwide is to
overfilter speech. If there is speech the government has an interest in
controlling, then let that control be obvious to the users. Only when
regulation is transparent is a political response possible.–Code and Other
Laws of Cyberspace
Corporations and governments can even regulate the Net geographically by
programming servers to discriminate against users from certain locations.
Countries like China could build a highly regulated Net to block access to
politically charged information to outsiders. Corporations could bar certain
people under certain ages from certain places on the Internet. Whoever has
access to encrypted identification data can discriminate based on age, sex,
medical status, political affiliation, race.
Open Control
Lawrence Lessig, like Eric S. Raymond, is not against private enterprise.
Neither the lawyer nor the open source activist is bound by ethical and
moral imperatives to keep the Net wholly free of commercial interests.
Lessig, as demonstrated by his stand in the Microsoft antitrust trial,
believes in an active government, although he does voice his warnings and
criticisms. Unlike Raymond, he doesnÂ’t believe that the invisible hand of
the free market will guide the Net toward greatest liberty. What we must
aspire to, he reasons, is a fine balance between regulation and freedom.
Just as the framers of the U.S. Constitution aimed to ensure an equilibrium
of power between the federal government and the individual states, industry,
and individuals, so we need checks and balances on the powers-that-be that
control the technological architecture of the Net.
Architecture is a kind of law: It determines what people
can and cannot do. When commercial interests determine
the architecture, they create a kind of privatized law.
I am not against private enterprise; my strong
presumption in most cases is to let the market produce.
But isnÂ’t it absolutely clear that there must be limits
on this presumption? That public values are not
exhausted by the sum of what IBM might desire? That what
is good for America Online is not necessarily good for
America? . . .
The limits to corporate and governmental power lie in the transparency of
intention. Regulation isnÂ’t inherently wrong, Lessig argues; what is
dangerous is invisible regulation. A closed, proprietary architecture hides
commercial and governmental intent. If intent is hidden, the architecture
can encourage any kind of wrong, like covertly favoring one power, as in
Berners-LeeÂ’s example of agent technologies that find only sponsorsÂ’
products. The more code is open and available to the public, the more
government and corporate power is constrained from abuse of power.
Which all leads back to . . . free software and open source. Lessig states
that open code is the crucial check on power on the Net. For this reason he
has praised Eric S. RaymondÂ’s The Cathedral and the Bazaar as the compelling
essay that has defined the movement. "Open code is open control–there is
control, but the user is aware of it." Code developed in the bazaar makes it
yet more transparent to the knowledgeable public. .Each individual part is
comprehensible to an individual who understands the principles of code.[4]
The bazaar regulates code transparently, and thus limits the absolute power
of corporations and governments. Open code is, as Lessig deems it, a "A
Freedom of Information Act for network regulation." He deems it the
foundation to an open society.
We are just beginning to see why the architecture of the space matters–in
particular, why the ownership of that architecture matters. If the code of
cyberspace is owned . . . it can be controlled; if it is not owned, control
is much more difficult. . . . In a way that the American founders would have
instinctively understood, . . . "open code" . . . is itself a check on
arbitrary power. A structural guarantee of constitutionalized liberty, it
functions as a type of separation of powers in the American constitutional
tradition.–Code and Other Laws of Cyberspace
Yet, even if the code is open, how will the law be upheld? Lessig argues
that the courts must determine constitutional values and must make
constitutional choices about the Net. A dialogue must begin–now–about what
values the Net should have. To what extent does the original Constitution
guide us on the Net? What new values must we articulate for democracy and
the rights of citizens? Who will make them lucid and enforceable?
My fear about cyberspace . . . is that the institutions most responsible for
articulating constitutional values will simply stand back while issues of
constitutional import are legislatively determined. The institutions most
responsible for articulating constitutional values today are the courts. My
sense is they will step back because they feel . . . that these are new
questions that cyberspace has raised. Their newness will make them feel
political, and when a question feels political, courts step away from
resolving it.–Code
Ideally, Lessig answers, the values of free speech, privacy, due process and
equality on the Net should be resolved in a responsible, democratic forum.
But, again, he asks: If there is no government or court system that
represents and enforces these values, then who will? If code is law, who are
the lawmakers? What values are being embedded in the code?"
On these questions Lessig, the liberal and Raymond, the libertarian, part
ways.
RaymondÂ’s Retort
In 1999, the society of Computer Professionals for Social Responsibility
presented the Norbert Wiener Award to the open source movement. Eric S.
Raymond, Richard M. Stallman, Larry Wall, and Brian Behlendorf (of Apache)
accepted the award. Raymond wrote the following about liberty and law on the
Net in his acceptance speech:
All too often, people who invoke ‘social responsibility’
are demanding that we give up individual liberty–that we
accept just a bit more taxation, just a bit more
regulation, just a bit more governmental intrusiveness,
all for the supposed good of society. . . .
We cannot leave the defense of our liberty to
politicians . . . It is the duty of every citizen–and of
every socially responsible programmer–not merely to
passively resist the erosion of liberty, but to actively
promote and extend liberty; to enlarge the private
sphere; to take power away from government so that
individuals and voluntary groups may peacefully work out
their destinies.
To Raymond, the libertarian, the Net will regulate itself, provided that the
NetÂ’s users become like him: socially conscious and liberty-driven. In
general, he says, he doesnÂ’t argue from open source to politics, because he
despises people who warp science to advance a political agenda. The
Cathedral and the Bazaar was written without a directed political
motivation. Despite this, his libertarianism allows him to see patterns
between politics and open source.
The libertarian view is that the invisible hand of the marketplace will be
enough to balance the interests of the individuals with that of the
corporations. The reason: Most businesses will find it necesssary to go open
source anyway. They wonÂ’t have a choice. ItÂ’s simply that open source works
best. No closed source company can possibly compete with the quantity of
programmers working on a code, nor can such code be as valuable in the open
market without being open. The Microsofts and AOL-Time Warners will be
ultimately restrained on the Internet by an architecture that will stay open
source by default. No need for DJ intervention or any other governmental
mechanism that may muck up free-market mechanisms.
Unlike Lessig, Raymond believes that free market capitalism is superior to
any form of government intervention–even if the latter, through the letter
of the law, intends to maintain open code on the Net. The libertarian ethic
is that government should have limited power. To Lessig, the struggle is for
regulation that provides the most freedom and opportunity to the people.
Government, through the law and court system, can help. To Raymond, the
battle is the noble, responsible individual versus a power-hungry and
bureaucratic government.
This battle needs to be fought on at least two fronts;
technological (by developing technologies that empower
individuals and put their activities and communications
out of government reach) and political (by using
provisions of the U.S.'s Constitution and analogous
charters and traditions in other countries to head off
government attempts to censor and control content).
A Jeffersonian Balance
Thus far the Net has been a democratic export worldwide in its facility for
free speech, its distribution of secure encryption, and its resistance of
censorship. Lessig invokes the spirit of Licklider and the Intergalactic
Network as example of how revolution and innovation is possible. The early
Internet, after all, became popular only because of its generosity of
spirit–privacy, free speech, free browsers, cheap access. Anyone with access
to a computer–even at a local library–can get an e-mail account and an
"emperorÂ’s library" at his or her fingertips. But Lessig disagrees with
Raymond that the Net (the Internet and Web) will remain this way through
free-market enlightened self-interest. Lessig argues that the Net will be
regulated–controlled and abused–by companies, potentially monopolies, if no
other regulation exists to counter it. It is no more intrinsically free and
open than Russia was after the Soviet Union fell and corruption filled the
vacuum. Markets need certain conditions, certain architectures, before they
can succeed. They need governmental regulation. He thinks even independent
bodies of experts like W3C or the Internet Engineering Task Force (IETF) are
subject to the interest of companies that write the code that shapes the
Internet that shapes our lives.
Furthermore, Lessig states that RaymondÂ’s interpretation of "regulation" as
defined by politicians and bureaucrats and "pseudo-cooperation with
governments" is limited and limiting. ItÂ’s "warmed-over Ayn Rand."[5] Lessig
also believes contract law, regulated property rights, and antitrust laws
are examples of constructive regulation. He insists that free markets wonÂ’t
function without this constructive form of regulation.
Countries must come to an agreement about how law will regulate and about
any norms that they will impose on private ordering. . . . It will require
the nations of the world to come to a common understanding about this space
and to develop a common strategy for dealing with its regulation.–Code and
Other Laws of Cyberspace
In an online debate on the Website of the political journal The American
Prospect, Lessig railed against Raymond:
The techno-anarchist, self-congratulatory,
fuck-the-government, give me code and Coke (-a-cola,
that is) attitude is doing the movement harm. While the
movement turns its energy away from government, while it
cocoons itself with this lullaby about how little from
government we need, others less principled and others
less convinced dominate the debate in Washington. While
the open-source crowd has scorned those who would speak
of regulation, regulations abound. Bad regulation, no
doubt, but bad regulation is the product of a process
where the good was not heard.
* * *
Lessig believes that collective values ought to determine the architecture
of the Internet, and that free market capitalism will not represent such
values. Federal and state law can aspire to do so.
How the Constitution deals with creative work–ideas–is the prime example of
this. Thomas Jefferson wrote in 1813, " . . . Ideas should freely spread
from one to another over the globe, for the moral and mutual instruction of
man . . . [They should be] incapable of confinement or exclusive
appropriation." Constraint of ideas, Jefferson reasoned, harms society.
Monopolists would hoard ideas to strengthen their own fiefdoms. Ideas belong
in the "intellectual commons," the public domain–the same conclusion Richard
Stallman reached 172 years later.
However, if there is no regulation, then the authors of ideas suffer. An
author tries to sell a book, and a giant publishing company can just sell
their own version, without giving the author anything. A musician writes a
beautiful song, and anyone can make their own record of it, again without
paying the idea-maker. If the author, the musician, the artist canÂ’t get
paid, theyÂ’ll stop creating ideas and start selling vegetables. Society
suffers if new ideas arenÂ’t created.
If the law didnÂ’t protect the author at all, there would be fewer authors.
The law has a reason to protect the rights of authors, as least insofar as
doing so gives them an incentive to produce.–Code and Other Laws of
Cyberspace
The Constitution of the United States calls for copyright of creative
work–including books, music, movies, and software–giving the authors of
creative works a monopoly over the publication (the copying) of their works
for a limited time, after which time such work will go into the public
domain. While copyright seems to serve the interests of the authors, its
real purpose is to serve the public. Copyright keeps authors from
starving–but should remain brief to encourage the free flow of ideas amongst
society. Lessig believes that the law of copyright should be upheld–even on
the Internet–emphasizing that the law stipulates for a limited time and with
limits such as the right of fair use (the right to use and reproduce
excerpts of copyrighted material).
Unfair Use
But not all law is good. Lessig cites the 1998 Digital Millennium Copyright
Act (DMCA) as one of the most dangerous forms of regulation yet passed for
the Net. This, he says to Raymond, is the result of "bad regulation" or
"coopted government" that threatens free markets if better regulation is not
put into place instead. Lessig believes that if the open source movement
ignores the government, then the powers-that-be will pass laws that oppose
freedom. The DMCA is a law that Lessig regards an ugly collaboration between
government and business.
The principle of fair use is that copyright law may not restrict the basic
right to free speech. People have the right to use copyrighted works, like
sharing a book with a friend, taping a television show, making a mix-tape of
favorite songs, or presenting movie clips in a film criticism class. The
DMCA restricts or forbids that right, if the book, show, song, or movie is
in digital form. Today only some creative works are mass-produced in digital
form (such as CD’s)–in the near future, nearly all will be. Everything will
be software, and the DMCA defines copyright law for software.
The DMCA criminalizes any behavior that could threaten technological methods
which keep software proprietary. A person doesnÂ’t have to deliberately
attack such "protection" to go to jail–she could simply explain why it’s
flawed, or even put up a website linking to someone elseÂ’s explanation. The
law prevents any exploration of weaknesses or violations of privacy in
security and encryption programs. It fails to distinguish between capacity
and action–it’s like jailing chemists because they could build a bomb, or
gun-owners because they could shoot someone, or car owners because they
could drive someone over.
Raymond responds to acts like the DMCA with a confident shrug reminiscent of
Yippie nonchalance. "We will fight these impositions through political
actions, technical subversion and other means because it is in our nature to
do so. Often, we will win. When we lose we will route around the damage and
carry onÂ….In the long run, technology shapes politics far more than the
reverse."
Yet Lessig is adamant that only the proper application of law can stand up
to big business and bad regulation. When public interests are at stake, then
the collective should choose–through the mechanism of the law. He continues
to rail against the DMCA through articles and arguments published in
magazine and posted on the Net. The future of the DMCA remains to be seen.
Afterward
As certain as Lessig is that courts and legislators must fight for
fundamental rights and free speech on the Net, he acknowledges that the
current system is not working to do so. He detects a weariness and wariness
on the behalf of the American people for governmentÂ’s ability to solve
collective problems. Likewise, the courts have been reluctant to be judged
political. The courts, he urges, must be political if they are to avoid
becoming an enemy of the people. When the president and Congress, corrupted
by a corporatized election process, fail to defend the principles of the
Constitution, the courts must do so. And they must act fast, as ever more
pernicious laws which strip away Constitutional values go into effect.
If we hate government, it is not because the idea of collective values is
anathema. If we hate government, it is because we have grown tired of our
own government. We have grown weary of its betrayals, of its games, of the
interests that control it. We must find a way to get over it. . . . We stand
at the edge of an era that demands we make fundamental choices about what
life in this space, and therefore in real space, will be like . . . The
values of free speech, privacy, due process, and equality define who we are.
If there is no government to insist on these values, then who will?–Code and
Other Laws of Cyberspace
Incidentally, the critical group most resistant to LessigÂ’s manifesto may be
those to whom Raymond refers as "his people"–the motley programmers and
hackers of the world, the very people who build the architecture of the Net.
Raymond has bequeathed a theory, a language, and a mythic tradition to
legions of programmers worldwide. His own libertarian politics infuses this
ethic. In his essay "How to Become a Hacker," Raymond reifies the hacker
community (the bazaar) as a self-organized, independent, and conspicuously
anti-authoritarian collective. Raymond believes that the selfishness of the
libertarian programmer is superior to any claim to collective
interest–better yet, the two end meeting as a form of mutual "enlightened
self-interest." The open source society runs on a free market of egoboo. And
egoboo, in turn, requires the sort of cooperation and "intellectual commons"
that Lessig rallies to forge through law and government.
We may view LinusÂ’s method as a way to create an efficient market in
"egoboo"–as to connect the selfishness of individual hackers as firmly as
possible to difficult ends that can only be achieved through sustained
cooperation. –The Cathedral and the Bazaar
Although egoboo may not be enough to foster the sort of democratic values
that Lessig seeks for the Net, the efforts of the free software and open
source community inspire what Lessig calls the transparency of intention.
Raymond predicts that the infrastructure of the Net will remain all open
source, as it was in the beginning with Tim Berners-Lee. It will be
moderated by consortia like ICANN and W3C and service-based corporations
like Red Hat Linux. Databases and development tools will be a mix of open
and closed code. Applications, he predicts, will be that last to go open
source, although "we can expect that the eventual destiny of any software
technology is to either to die of become part of the open infrastructure
itself."
The future of open source can be as unlimited as its human capital. It may
lead to a reasoned resurgence of the free society fought for by the Yippies,
expanding from software to become a new collaborative and nonproprietary
paradigm for education, science, and the arts. It may fragment even beyond
the current conflict between the Free Software Foundation and the Open
Source Initiative. It may, as Raymond and Lessig hope in their separate
ways, help limit the power of monopolies and oppressive government actions,
problems that have plagued democracy since its inception. This battle in the
Information Revolution, then, may be fought on more even ground.
Footnotes
1 In his writings, Raymond simply calls the system "Linux."
2 This is a direct gibe at Microsoft, whose applications are renowned for
"featuritis," the phenomenon of adding as many features to a single
application as possible, whether or not theyÂ’re necessary or even barely
useful.
3 Sponsored by the brother of the Civil War general, the Sherman Antitrust
Law of 1890 invoked in the Microsoft case was a populist backlash against
the industrial monopolies of the late nineteenth century in oil, steel, and
rail. Theodore Roosevelt and Edward Taft, his successor, invoked it to break
the Standard Oil monopoly in 1911. RockefellerÂ’s company had infuriated the
American people by raising prices to sponsor a war against foreign
competition. Consumers had no alternative and oil prices skyrocketed.
Prosecutors accused Bill Gates of being the new Rockefeller.
4 As society spends more time in cyberspace, within the architecture of
code, it becomes ever more important that every citizen understand the
principles of programming. Seymour Papert, Ted Nelson, and Alan Kay also
explicitly argued that the ability to program computers is crucial to the
preservation of democracy in the information age.
5 Raymond retorts that he is no more a warmed-over Randite than Lessig is a
warmed-over Marxist-Leninist, and that in political and policy debates
"regulation" usually indicates one party’s favor by law–not an equilibrium
of power.
References
Eric S. Raymond The Cathedral and the Bazaar, (California: OÂ’Reilly) 1999
Andrew Leonard, "Let My Software Go!" April 1998
http://www.salon.com/21st/feature/1998/04/cov_14feature.html 6 May 2001.
Lawrence Lessig, Code and Other Laws of Cyberspace, (New York: Perseus,
1999).
Jamie Zawinski "Fear and Loathing on the Merger Trail" 23 November 1998
http://www.mozilla.org/fear.html 6 May 2001.
David Streitfield "An Awkward Anniversary" 17 March 2001.
http://www.washingtonpost.com/wp-srv/WPlate/2000-03/17/119l-031700-idx.html
6 May 2001.
Wiener, 1948
Eric S. Raymond, "On Socially Responsible Computing," 19 October 1999
http://www.tuxedo.org/~esr/writings/cpsr-speech.html 6 May 2001.
Eric S. Raymond, "Defending Network Freedom"
http://www.tuxedo.org/~esr/netfreedom/index.html 6 May 2001.
Lawrence Lessig and Eric S. Raymond TAP Controversy: Should Public Policy
Support Open Source Software?" 27 March 2000
http://www.prospect.org/controversy/open_source/lessig-l-2.html 6 May 2001
--- End Message ---