What Drove the Rise of Open Source? A Macro‑Historical Analysis

The series will be divided into five articles, covering macro, meso, and micro perspectives:

Analyze the reasons and motivations behind the emergence of open source software at three levels.

Examine the impact of open source software on industry at the meso level.

Study the organizational characteristics and production methods of open source communities at the micro level.

This article focuses on the macro level, analyzing the reasons and motivations for the emergence of open source software.

From the perspective of human civilization, the 18th‑century Industrial Revolution inaugurated the era of industrial civilization; the technological‑social form of industrial civilization is the industrial society. After entering the industrial society, human development can be seen as a history of innovative progress, where technological innovation gradually becomes the main endogenous driver of growth alongside environment, institutions, and culture.

The term "technology" can be split into technology and science; science is the upstream of technology, applied technology builds on underlying technology, which in turn rests on basic science.

Basic science is the foundation of the entire technology and economy edifice. It performs the 0‑to‑1 work, opening new territory, while applied technology adds bricks on that territory, turning scientists' blueprints into reality. Accumulation of basic scientific innovation from quantitative to qualitative change can spark scientific revolutions; similarly, accumulation of underlying‑technology innovation can spark technological revolutions.

Technological revolutions bring a leap in human survival and development means, manifested as cluster inventions and innovations, often accompanied by industrial transformation.

Industrial transformation generally refers to a revolution or change in the structure of the material production sector, the result of widespread application of technological revolutions in production. When the social‑economic operation mode and production methods change, it signifies industrial transformation.

Overall, there is a basic linear relationship among scientific revolution, technological revolution, and industrial revolution. Scientific revolution provides the theoretical premise for technological revolution but does not directly trigger industrial change. Technological revolution directly affects productivity and production methods, becoming the direct driving force for industrial revolution.

Industrial revolution drives comprehensive social change, permeating technology, economy, and all aspects of social life—including ideas, institutions, law, and culture—ultimately advancing social civilization.

This regularity forms the modern civilization's "technology‑economy" development paradigm.

In the past 200+ years we have witnessed three industrial revolutions and five technological revolutions.

The formation mechanism of the technology‑economy paradigm is realized through changes in "key production factors". A key production factor is a specific input or set of inputs in each new techno‑economic paradigm, which may be an important natural resource or industrial product. To qualify, it must have decreasing production cost, unlimited supply capacity, and broad applicability. Thus, the formation and transition of techno‑economic paradigms are linked to the replacement of key production factors, an inevitable result of technological‑economic revolutions.

Software is one of the key production factors of the current information‑technology revolution.

At the birth of computers, the United States developed computers to calculate ballistic trajectories for the military during World War II. In fluid dynamics, calculations became too large for manual methods, creating demand for general‑purpose computers. IBM's second president recognized the future importance, invested heavily, and invited John von Neumann as a consultant.

Subsequent IBM investment produced many classic mainframe series in the 1960s and 1970s, but the equipment cost millions of dollars, affordable only to governments, large banks, and multinational corporations.

The business model at that time bundled hardware, software, and services; to use an IBM system you had to buy IBM hardware and pay high annual service fees, while software was not sold separately.

During this period, the representative software was the UNIX operating system, which experienced the first open‑source versus closed‑source dispute.

UNIX was first created in 1969 at AT&T Bell Labs by Ken Thompson and Dennis Ritchie on a DEC PDP‑7. Over the following years they refined UNIX, rewrote it in C, added multitasking, and continued development within Bell Labs. At that time AT&T held a telephone monopoly agreement that prohibited it from selling software, so UNIX could not be marketed as a product.

In July 1974, Thompson and Ritchie published "The UNIX Time‑Sharing System" in Communications of the ACM, marking UNIX's first exposure to the outside world. In the 1975‑76 academic year, Thompson became a visiting professor at the University of California, Berkeley, bringing graduate students to continue UNIX development. At Berkeley many new features were added. Starting with version V6 in 1976, AT&T licensed UNIX to a few U.S. universities and research institutions (for a small fee or free), providing source code and allowing modifications, which promoted UNIX's spread.

However, this was not open source as understood today; it was limited shared source, not open to anyone. Nevertheless it marked the beginning of the open‑source movement, and the GNU project’s name—"GNU's Not UNIX"—originated from this era.

From UNIX V6 a fork occurred:

Commercial branch: the original AT&T and subsequent commercial versions.

Academic branch: Berkeley's BSD line – 1BSD appeared in 1977.

Later AT&T recognized UNIX's value but could not re‑centralize the source, so it adopted an external licensing model: free for research institutions, fee‑based for enterprises, akin to selling gold as copper. A senior AT&T executive once said, "UNIX is the second most important invention after the transistor," yet AT&T missed the optimal moment to commercialize UNIX.

AT&T then formed the UNIX Support Group (USG) and commercialized UNIX. In 1982 AT&T developed UNIX System III based on V7, a commercial version sold to customers. To resolve the chaotic UNIX version landscape, AT&T integrated various UNIX contributions to create UNIX System V Release 1, a commercial release without source code. Berkeley continued developing BSD as an alternative, leading to System IV and 4.x BSD as two main streams.

In 1986 IEEE defined POSIX as an open operating‑system standard, and AT&T's UNIX secured the trademark. BSD supporters saw themselves as rebels against the AT&T empire. In 1989, Networking Release 1 (Net/1) was created, completely independent of AT&T, allowing BSD‑licensed distribution. In June 1991, Net/2 emerged as a new OS that could be freely released.

That same year, a group of BSD developers left Berkeley to found BSDI (Berkeley Software Design, Inc.), offering a full‑featured commercial UNIX on inexpensive Intel platforms.

AT&T sued BSDI in 1992 for alleged source‑code infringement, halting Net/2's release until the code was cleared. The lawsuit lasted until AT&T sold its UNIX System Laboratories to Novell, which adopted a more open stance, allowing BSDI to release its BSD UNIX after removing AT&T code.

In June 1994, 4.4BSD was released in two forms, one of which—4.4BSD‑Lite—contained no AT&T code and became the foundation of modern BSD systems.

Between 1992 and 1994, the legal disputes delayed BSD development and the broader free‑software movement, allowing the Linux kernel to gain substantial support. Linux and 386BSD (a Net/2 fork) started almost simultaneously; Linus Torvalds noted that if a free Intel‑386 UNIX‑like OS had existed, he might not have created the Linux kernel.

During these disputes, a new "marketplace" production model emerged, contrasting with the traditional "cathedral" model of centralized, closed, and secretive development. This decentralized, open, peer‑reviewed marketplace model is the core idea of later open source.

From the first open‑source versus closed‑source conflict we can see:

Dominant companies lack incentive to open source for commercial gain; open source originates from a few hacker‑spirit researchers and weaker companies.

Compared with Windows and Linux, UNIX had technical advantages but fragmented licensing prevented a universal version, leading to compatibility issues and hindering network effects.

In summary, during the fifth technological revolution of the information age, society's demand for core hardware and software elements is broad and persistent; these core elements become cheaper and more accessible, while intense economic and social conflicts cause a split between open‑source and proprietary software.

Technical‑Economic Paradigm Perspective on Open‑Source Evolution (Part 1)