This Software Isn't Beer

In 1980, a printer jammed at MIT.

Richard Stallman, a programmer at the Artificial Intelligence Laboratory, wanted to fix it. Not the hardware (that was Xerox's problem). He wanted to modify the software so the printer would notify users when their jobs got stuck. He'd done it before with the old printer. A few lines of code, and people stopped wasting trips down the hall to find their pages sitting in a jammed machine.

But when Stallman tracked down the source code for the new Xerox 9700, he hit a wall. A professor at Carnegie Mellon had worked on the printer software at Xerox PARC. Stallman asked for the code, just as he always had.

The professor refused. He had signed a non-disclosure agreement.

To Stallman, this wasn't just a "no." It was a violation of the hacker ethic that defined his world. Now, a contract stood between a programmer and his tools. The hardware in his lab was no longer his to control. He was powerless.

That refusal lit a fuse. What followed would reshape how software is made, shared, and owned. But to understand why a jammed printer mattered so much, you need to know what was being lost.

Freedom of Software

The Idealist: Richard Stallman and the Free Software Movement

In the 1960s and '70s, software typically came bundled with hardware, and sharing source code was the norm. User groups like SHARE and DECUS facilitated the exchange of programs and technical knowledge. Programmers built on each other's work freely. It was simply how things were done.

Stallman had grown up in this culture. The AI Lab at MIT was its purest expression: a place where hackers shared code like scientists shared research. When the professor refused to hand over the printer source, Stallman saw it as a betrayal of everything the community stood for.

"I was so angry I couldn't think of a way to express it," Stallman later recalled. "So I just turned away and walked out without another word."

It wasn't just about a printer. It was a sign of things to come. Software was being locked away behind corporate walls, and the culture of sharing was dying.

Stallman spent the next few years watching the community he loved dissolve. Colleagues left for lucrative jobs at startups. The AI Lab's culture of open collaboration gave way to proprietary projects and NDAs. By 1983, he'd had enough.

He responded with an act of sheer audacity. He wouldn't just complain about the system; he would replace it. If the Operating System (OS) was proprietary, users would always be at the mercy of whoever controlled it. But if the OS was free, programmers could build an entire ecosystem of free software on top of it. Stallman chose to clone Unix because it was portable, well-documented, and already popular in universities and businesses. A free Unix-compatible system would let users escape proprietary software without abandoning their existing workflows.

In September 1983, he announced the GNU Project (a recursive acronym: "GNU's Not Unix"), an ambitious effort to create an entirely free Unix-like operating system. Two years later, he published the GNU Manifesto in Dr. Dobb's Journal and founded the Free Software Foundation, declaring war on proprietary software.

Stallman's vision was radical: software should be free, not just in cost, but in freedom. Users should be able to inspect it, modify it, and share it. "Free as in freedom, not free as in beer," as he famously put it. This wasn't about money; it was about power and control.

Enforcing Freedom

Stallman wasn't the only one watching the culture shift. The battle lines were being drawn across the industry.

In 1976, a young Bill Gates fired off an angry letter to the hobbyist community. His "Open Letter to Hobbyists" accused them of stealing Microsoft's Altair BASIC. "Who cares if the people who worked on it get paid?" he wrote. The hobbyists fired back. Some pointed out that Gates had developed the software on Harvard's government-funded computers. Others questioned his claimed $40,000 in development costs. The letter drew a line in the sand. On one side: software as product, protected by law. On the other: software as shared knowledge, built by communities. The hobbyists responded by creating Tiny BASIC, free interpreters written and distributed by volunteers, proving that useful software didn't require corporate backing.

The tension between these two worlds would soon catch Stallman in the crossfire. In 1981, James Gosling had written a version of Emacs for Unix and shared it freely, asking only for a letter acknowledging his authorship. Stallman incorporated some of Gosling's display code into an early version of GNU Emacs. But in 1983, Gosling sold his Emacs to a company called UniPress, who began selling it commercially. The code Stallman had used in good faith was now proprietary. UniPress demanded he stop distributing GNU Emacs with any of Gosling's code. Stallman had to rewrite those portions from scratch.

The betrayal stung. UniPress believed "hobbyists and academics could never produce an Emacs that could compete" with their commercial product. They were wrong, but the experience crystallized a hard lesson: in a world of lawyers, good faith was a vulnerability. Code shared without legal protection could be seized and turned against the community that built it.

Stallman realized he needed a weapon.

In 1989, he forged one: the GNU General Public License (GPL). The GPL introduced "copyleft," a concept that performed a kind of legal jujitsu. It used copyright law (normally a tool for restriction) to guarantee freedom. If you built on GPL code and distributed it, you were legally bound to keep your contribution open under the same license. It was a virus of liberty, designed to spread openness wherever it touched.

Meanwhile, researchers at universities like MIT and UC Berkeley took a different path. They developed permissive licenses: the BSD license in 1980, the MIT License around 1987. These let anyone use their code for any purpose, even proprietary products. The tradeoff was intentional. Permissive licenses meant less legal friction, fewer hoops to jump through. That made adoption easier and contributions more likely, especially from companies wary of copyleft's viral nature. Less ideological, more pragmatic. Two philosophies, both pushing back against the corporate tide.

The Pragmatist: Linus Torvalds and the Linux Revolution

While Stallman was fighting legal battles, a very different kind of revolution was brewing in Helsinki. It came not from ideology, but from a 21-year-old who just wanted a better operating system.

On August 25, 1991, Linus Torvalds posted a now-famous message to the comp.os.minix Usenet newsgroup:

I'm doing a (free) operating system (just a hobby, won't be big and professional like gnu) for 386(486) AT clones.

The self-deprecating parentheticals turned out to be spectacularly wrong. That hobby became the Linux kernel, and it arrived at exactly the right moment.

By 1991, Stallman's GNU project had built most of a complete operating system: compilers, editors, utilities, and libraries. But one critical piece was missing: the kernel. GNU's own kernel project, called Hurd, was plagued by delays. It used an ambitious microkernel architecture that was technically elegant but fiendishly complex to implement. Development dragged on for years, and Stallman himself would later admit he was "not very optimistic" about it.

Linux filled that gap. When combined with GNU's tools, it created the first complete free operating system: GNU/Linux. Suddenly, free software wasn't just a philosophy. It booted.

The BSD Lawsuit: A Fork in the Road

Linux also benefited from a legal storm that paralyzed its main competitor. In 1992, AT&T's Unix System Laboratories sued Berkeley Software Design (BSDi) and the University of California over BSD Unix, claiming it contained proprietary AT&T code. The lawsuit cast a legal cloud over the entire BSD ecosystem. FreeBSD and NetBSD couldn't be safely released until the case settled in early 1994.

Linus Torvalds himself acknowledged this twist of fate. He told Meta Magazine in 1993: "If 386BSD had been available when I started on Linux, Linux would probably never had happened." The lawsuit didn't just delay BSD. It handed Linux a critical window to capture developer mindshare.

What made Linux explode was Torvalds' approach: he welcomed feedback and contributions from the community. Early on, that meant patches from strangers, messy debates, and rapid iteration; yet it also created a decentralized development model where people could work in parallel on their own goals. This wasn't driven by ideology; it was driven by results. The approach proved faster, more reliable, and higher quality than traditional closed development.

The Cathedral vs. Bazaar

This development model became so influential that Eric S. Raymond wrote an essay about it. "The Cathedral and the Bazaar", published in 1997, contrasted two approaches to building software. The Cathedral represents centralized, careful development: one team, one vision, released when ready. The Bazaar represents distributed development: many contributors working in parallel, releasing frequently.

Raymond's key insight was counterintuitive: "Given enough eyeballs, all bugs are shallow." The more people who could see the code, the faster problems got found and fixed. What looked like chaos was actually a more efficient debugging machine than any corporate QA department could build.

The essay argued that the Bazaar model could produce software as good as, or better than, the Cathedral. It remains a valuable read today. It would become a catalyst for one of the most significant moments in open source history.

Catching Fire: The Browser Wars and the Open Web

By the mid-1990s, free software had proven itself in server rooms and universities. But it remained invisible to most of the world, a movement of hackers talking to hackers. For free software to go mainstream, it needed a crisis, a business model, and a new name.

Embrace, Extend and Extinguish

In the 1990s, it was still the early days of the internet. Public interest exploded, and Netscape's Navigator browser, released in 1994, quickly became the dominant way to navigate the nascent web. Their IPO in 1995, the largest at the time, signaled that this was a real new technological wave worth paying attention to.

Microsoft saw it as a threat to its operating system business model and moved aggressively. Critics later summarized the playbook as "Embrace, Extend, and Extinguish," a framing that surfaced in and around the U.S. antitrust era. First, they embraced the existing standards, in part by licensing Mosaic code (an ancestor of Navigator) to get a jumpstart. Next, they extended web standards with proprietary features, creating incompatibilities and friction for competitors; some sites only worked on Internet Explorer. Finally, they extinguished competition using established power: bundling Internet Explorer with Windows, pushing OEM and partner deals, and making it painfully expensive to bet against the default.

In the end, it was not a fair battle. In a last-ditch effort, Netscape announced in 1998 that it would open-source its browser code because it couldn't compete with Microsoft with its own resources. Raymond's "Cathedral and the Bazaar" played a direct role. A Netscape employee cited it as external validation when convincing executives that open-sourcing was viable. Raymond was even invited to Netscape's headquarters for a day-long strategy session to help design their source release. Up until this point, open source software was seen by businesses as a countercultural, non-commercial oddity. While the move wasn't able to save Netscape, a phoenix would rise from its ashes.

LAMP Lights Up

While the corporate giants fought a scorched-earth war for the browser window, a quiet insurgency was taking over the server room. The headlines were all about Netscape and Microsoft, but the real revolution wasn't about how we viewed the web. It was about how we built it.

In server rooms and dorm rooms around the world, four technologies (Linux, Apache, MySQL, and PHP) were converging into a stack that would power the early internet. A German magazine editor coined the acronym "LAMP" in 1998 to describe this combination, demonstrating that free software could replace expensive commercial packages.

The Apache story captures how this happened. In mid-1994, the lead developer left the National Center for Supercomputing Applications (NCSA), and development on the most popular web server software stalled overnight. Webmasters who depended on it were left stranded with bugs to fix and features to add, but no one to send patches to.

Brian Behlendorf was one of them. A 21-year-old building websites for Wired magazine's HotWired project, he needed the NCSA server to handle user registration, but it couldn't. So he patched the code himself. Then he discovered he wasn't alone. Other webmasters were doing the same thing, each fixing their own problems in isolation.

Behlendorf set up a mailing list to coordinate. By February 1995, eight developers had pooled their patches and forked the NCSA codebase. They called it Apache ("a patchy server," as the legend goes). Working loosely together across time zones, they eventually rewrote the entire program. By 1999, Apache had grown into the Apache Software Foundation, a non-profit designed to ensure the code and community could outlive any single contributor.

The rest of the stack came together the same way: Linux provided a stable base, MySQL made databases accessible, and PHP turned static web pages into dynamic applications. The LAMP stack mattered because it was accessible. No licensing fees. No vendor lock-in. A kid with a $500 computer could build the same applications as a Fortune 500 company. The giants of Web 2.0 proved its power: Wikipedia, Facebook, and WordPress all started on LAMP. The stack didn't just lower the barrier to entry. It obliterated it.

And if you were that kid, it felt unreal. You'd install Linux off a magazine CD or a friend's burned disc, fight a few drivers, and suddenly you had a real server. Apache would serve pages the way "real" sites did. PHP would take a static page and make it breathe. MySQL would remember things. For the first time, the limiting factor wasn't licensing. It was how stubborn you were.

The Bridge: OSI and Bringing Open Source to Business

The movement had a branding problem. Every time someone heard "free software," they assumed it meant "zero cost." Explaining "free as in freedom, not free as in beer" got old fast, and it wasn't winning over enterprise buyers who associated "free" with "cheap" and "unsupported."

The solution came from an unlikely source. Christine Peterson wasn't a programmer. She ran Foresight Institute, a nanotechnology think tank. But she'd been attending weekly meetings in late 1997 where a small group discussed how to promote free software to a broader audience. The terminology issue kept coming up: "those new to the term 'free software' assume it is referring to the price," Peterson later wrote. "At this point, a discussion on software has turned into one about the price of an alcoholic beverage."

On February 2, 1998, Eric Raymond arrived in town to work with Netscape on their plan to release the browser code. Peterson had been mulling over alternatives, and between meetings that week, she landed on "open source." She ran it by a few people. Some liked it; a friend in marketing thought "open" was overused. But she didn't have a better idea.

The key moment came on February 5th at VA Research. Raymond, Peterson, and a small group of developers gathered to discuss strategy. Peterson felt awkward proposing a term as a non-programmer. "My influence within the free software community was weak," she admitted. So instead of formally suggesting it, one of the developers simply dropped "open source" into a sentence about another topic.

No one reacted at first. Then, a few minutes later, someone else used the term without noticing. Peterson exchanged a glance with him across the room. The term was catching on, spreading through the conversation organically.

By the end of the meeting, the group explicitly discussed terminology. Several alternatives were proposed: "freely distributable," "cooperatively developed," "sourceware." But "open source" won out. Raymond endorsed it. The next day, Linus Torvalds gave his blessing.

It was a strategic coup, and a philosophical rupture. They were effectively deciding to strip the movement of its moral philosophy to make it palatable to the boardroom.

Richard Stallman initially considered adopting the term, then changed his mind. To him, "open source" was a betrayal. "The philosophy of open source considers issues in terms of how to make software 'better,' in a practical sense only," he later wrote. "For the free software movement, however, nonfree software is a social problem." Some supporters of open source saw it as a "marketing campaign for free software." Stallman saw it as selling out.

This philosophical split persists to this day. The Free Software Foundation still insists on "free software," framing the issue as one of ethics and user rights. The business world embraced "open source," focusing on practical benefits. Two movements, sharing most of the same code, divided by what that code meant.

In April 1998, Tim O'Reilly's "Freeware Summit" made the new terminology official. Leaders of major projects gathered and voted to adopt "open source" as the movement's public label. By the time they left, the summit had been renamed the "Open Source Summit."

That same year, Raymond and Bruce Perens founded the Open Source Initiative (OSI) to formalize the movement. They derived the Open Source Definition from Debian's Free Software Guidelines, written by the Debian community in 1997. The rebranding worked. It made the concept palatable to the corporate world.

Commercialization

With the terminology settled, the next question was inevitable: money. The code was free, so how do you build a business?

Red Hat's origin story sounds like a startup cliché: a small businessman met a geek at a tech conference. Marc Ewing was the geek, hacking and debugging his own Linux distribution from his home in Durham, North Carolina. He named it after his grandfather's red Cornell lacrosse cap, which he'd worn in his college computer lab. ("If you need help, look for the guy in the red hat.") Bob Young was the businessman, running a computer supply catalog from his wife's sewing closet in Connecticut.

Young had just been laid off. "I'm unemployed with three kids, a big mortgage and a net worth of something below what it had been when I graduated from college 15 years earlier," he later said. "But for that disaster, I wouldn't have stumbled into the open-source opportunity."

He noticed growing interest in Linux and started buying Ewing's CDs. He sold out so many times that they joined forces in 1995.

Red Hat's bet was simple: corporations needed guarantees. They needed someone to call when things broke. They needed training, documentation, and a stable, tested version they could trust for mission-critical applications. Red Hat offered all of that, not by selling software, but by selling support.

Wall Street noticed. On August 11, 1999, Red Hat went public. The stock opened at $14 and closed at $52, a 270% gain on day one. Within three days, it hit $85. At the height of the dot-com bubble, it was one of the largest first-day gains in Wall Street history.

IBM noticed too. With Microsoft dominating the server market via Windows NT, Linux offered a robust alternative. It fit perfectly into IBM's new strategy: selling expertise rather than licenses. By late 2000, IBM announced a $1 billion investment to make its enterprise servers compatible with Linux. Within two years, they'd recouped the investment. This validation from one of America's tech giants was a major milestone, especially after years of attacks from Microsoft, who had called open source "risky," "unstable," and "legally dangerous."

Aftermath: The Phoenix Rises

Netscape's source code lived on, but the road was brutal. The codebase was a mess: years of corporate cruft, feature bloat, and technical debt. Engineers had to rewrite it almost from scratch. Meanwhile, Microsoft had won the browser war. By 2001, Internet Explorer commanded over 90% of the market. Microsoft disbanded the IE team. Why invest in a product with no competition?

What followed was a digital dark age. Internet Explorer 6 reigned virtually unchallenged, going five years without a major update. Innovation stalled.

If you used the web in 2003, you remember the misery. Pop-up ads spawned faster than you could close them. Spyware hijacked homepages. Standards were ignored. IE6 was dubbed "the least secure software on the planet." Proprietary extensions like ActiveX fragmented web standards, forcing developers to litter their code with conditional hacks. Microsoft had won, and the web was rotting.

But in the wreckage of Netscape, something was stirring.

Blake Ross was 14 years old when Netscape released its source code in 1998. He dove in, submitting bug fixes and patches. By 2001, he'd earned a summer internship at Netscape's Mountain View headquarters. His mom flew across the country and drove him to work every day. He was still in high school.

Ross saw the problem up close. AOL had split Netscape's browser division into competing groups, and the result was a bloated mess. The interface had become impossibly complicated. Features that sounded good in meetings, like automatic form fillers, turned into distractions in practice. "Just getting started using the web in Netscape was an exercise in frustration," Ross later recalled.

He stayed up until 3 or 4 in the morning reading customer feedback. "People were saying that Internet Explorer was so bad and Netscape was so bad," he said. "I knew that there was something out there that people wanted and nobody was giving it to them."

Ross and a senior Netscape engineer started working nights on a back-to-basics browser. They forked the code, stripped out the bloat, and built a new interface from scratch. Ross drew inspiration from Google's founders: "They practically invented the 'Oh, duh' school of usability."

They pitched it to Netscape's executives. The response was dismissive: "This is just the browser we have, but with fewer features." Which was exactly the point. A browser should do one thing: let people surf the web. Netscape eventually spun off the project to the Mozilla Foundation.

On November 9, 2004, Firefox 1.0 launched. It was fast, secure, and simple. It blocked pop-ups. It had tabbed browsing. It respected web standards. Within nine months, it had been downloaded over 60 million times. A teenager and a small team of volunteers had built something that millions of people preferred over the product of the wealthiest corporation on Earth.

The browser wars reignited. Microsoft, forced out of complacency, announced plans to upgrade Internet Explorer for the first time in years. Firefox proved that open source could go head-to-head with tech giants, and win.

The Foundation Was Laid

Two decades earlier, Richard Stallman had asked a colleague for some printer code and been told no. That refusal had seemed like a small thing, a minor frustration in a university lab. But it had crystallized something. The belief that software should be shared. That users deserve control over the tools they depend on. That collaboration beats isolation.

By the mid-2000s, that belief had become infrastructure. Linux ran the servers. Apache handled the requests. Firefox sat on millions of desktops. Red Hat was a billion-dollar company. The revolution wasn't countercultural anymore, it was mainstream.

The ideological battles between "free software" and "open source" continued. Stallman never stopped insisting that freedom mattered more than convenience. But the practical impact was undeniable: a generation of programmers had proven that open collaboration could produce world-class software, and that sharing code wasn't naive idealism. It was a better way to build.

The foundation was laid. What came next would test whether it could hold.

In the end, it wasn't about the beer. It was about who gets to choose.

This is the first post in a series exploring the open source ecosystem. Future posts will dive into how open source became big business: GitHub democratizing contribution, venture capital flooding in, Microsoft's dramatic pivot from calling Linux "a cancer" to becoming one of its largest contributors, the impact of AI, sustainability challenges, and where it's all heading.

This Software Isn't Beer

Freedom of Software

The Idealist: Richard Stallman and the Free Software Movement

Enforcing Freedom

The Pragmatist: Linus Torvalds and the Linux Revolution

The BSD Lawsuit: A Fork in the Road

The Cathedral vs. Bazaar

Catching Fire: The Browser Wars and the Open Web

Embrace, Extend and Extinguish

LAMP Lights Up

The Bridge: OSI and Bringing Open Source to Business

Commercialization

Aftermath: The Phoenix Rises

The Foundation Was Laid

References

The Birth of the Free Software Movement

Enforcing Freedom

The Linux Revolution

Browser Wars & Open Web

LAMP Stack

Commercialization

OSI & Modern Definitions

Aftermath & Firefox