Marc Andreessen and Netscape

Zusammenfassung

In early 1993, a twenty-one-year-old undergraduate at the University of Illinois wrote the first graphical web browser accessible to ordinary people. Mosaic brought the web out of academic obscurity and onto millions of desktops. Two years later, the company Marc Andreessen co-founded to commercialize that insight staged the most dramatic stock market debut of its era — widely credited with igniting the dot-com boom. Netscape lost the Browser Wars to Microsoft, was absorbed by AOL, and formally shut down in 2008. But what Mosaic started — the commercial, graphical, visually navigable web — never stopped. Andreessen went on to build one of the most influential venture capital firms in Silicon Valley and argue, presciently, that software was eating the world.

The NCSA Lab and the Line-Mode Browser

Marc Lowell Andreessen was born on July 9, 1971, in Cedar Falls, Iowa, and grew up in New Lisbon, Wisconsin. His childhood passion was computers, not sports; he taught himself BASIC from a library book and wrote simple programs for a TRS-80. He enrolled at the University of Illinois at Urbana-Champaign to study computer science, and in 1992, as an undergraduate needing part-time income, took a job at the National Center for Supercomputing Applications — NCSA — for $6.85 per hour.

NCSA operated some of the best computing infrastructure at any American university: high-bandwidth internet connections, powerful workstations, a staff of engineers who had been following the early internet since ARPANET. It was there that Andreessen encountered the World Wide Web, newly released by Tim Berners-Lee at CERN. The web’s promise was clear: hyperlinked documents on any computer anywhere, navigable from a single interface. The web’s execution was not. The standard browser was the line-mode browser — it displayed text in a terminal window. Images required a separate viewer application launched in another window. Links were indicated by numbers in brackets. Navigation required typing commands.

“This is too hard,” Andreessen said. Or words to that effect.

He approached NCSA staff scientist Eric Bina with a proposal: build a graphical browser that displayed images inline with text, worked with a mouse, and ran on the Windows personal computers that most non-technical users owned. Bina was receptive. They began working after hours, during the winter of 1992 to 1993, in extended coding sprints that Bina recalled as sustained and intense. Bina wrote much of the core code; Andreessen shaped the design decisions and pushed the pace. Their collaboration produced something neither could have done alone.

Mosaic

Mosaic 1.0 was released on January 23, 1993, initially for Unix/X11 systems, with Windows and Macintosh versions following in the fall of the same year. Its defining feature — the one that changed everything — was inline images. For the first time, a web browser displayed photographs, diagrams, and graphics directly in the flow of a document, between paragraphs of text, without requiring a separate window or viewer. It seems obvious in retrospect. At the time, it was a revelation: the web was suddenly a visual medium.

The inline image required a new HTML tag. In February 1993, Andreessen proposed to the www-talk mailing list — the informal forum where Berners-Lee and other early web developers discussed the protocol’s evolution — that HTML should include an <IMG> tag allowing browsers to embed images specified by URL directly in pages. The proposal was contested. Berners-Lee and others were committed to a principle of separating document structure from visual presentation; embedding display-specific elements in HTML seemed architecturally wrong.

Info

Andreessen shipped the <IMG> tag in Mosaic anyway, without waiting for consensus. Within months, every other browser had implemented it, because web authors were using it and browsers that didn’t support it rendered those pages broken. The fait accompli became the standard. The episode established a precedent that would recur throughout web history: pragmatic implementation by a dominant browser could overrule architectural principle. The web’s evolution has been shaped as much by what browser vendors shipped unilaterally as by what standards bodies agreed.

Within months of the Windows release, Mosaic had two million users. The number of websites, which had been doubling every few months throughout 1992, began doubling faster. NCSA distributed Mosaic free of charge. The decision maximized adoption at the cost of revenue — a trade that would prove formative for Andreessen’s subsequent thinking about internet business models.

Andreessen graduated in December 1993 and moved to California, taking a job at a small software company in Silicon Valley. He was twenty-two years old and had already created the application that was changing the internet.

Jim Clark and the Netscape Founding

In early 1994, Jim Clark contacted Andreessen through a mutual introduction. Clark had founded Silicon Graphics — the workstation company that made the 3D graphics hardware used in Hollywood film production — and had recently resigned from its board following a dispute over strategic direction. He had money, energy, and no immediate plan.

Clark asked Andreessen what they should build together. Andreessen’s answer, essentially, was: a better Mosaic. Not a modified version of the NCSA code, which belonged to the University of Illinois, but a new browser written entirely from scratch, with a professional engineering team, commercial-grade performance, and a business model.

There were legal constraints to navigate. Andreessen had been an NCSA employee when he wrote Mosaic. Several of his former NCSA colleagues were recruited for the new company, which created further friction with Illinois. The company was founded in April 1994 as Mosaic Communications Corporation, then renamed Netscape Communications Corporation in November 1994 after the University of Illinois objected to the use of the “Mosaic” name. The browser was renamed Netscape Navigator.

Clark provided the initial capital and the business experience. Andreessen was twenty-two, the technical visionary, the person who had demonstrated that the web was a mass-market phenomenon before anyone had used those words.

Navigator and the Ninety Percent

Netscape Navigator 1.0 shipped on December 15, 1994, and was made available for free download for non-commercial use. It was meaningfully better than any competing browser: faster, more stable, with better HTML support and a more coherent user interface. Web developers began testing their sites in Netscape, which meant sites were optimized for Netscape, which meant users chose Netscape because sites looked right in it. The positive feedback loop was self-reinforcing.

By early 1995, Netscape held approximately seventy-five to eighty percent of the browser market. By mid-1995, it had reached roughly ninety percent. The web was, functionally, Netscape’s web. What Netscape rendered, the web looked like. What Netscape’s engineers decided to implement in HTML or JavaScript, web developers used. Netscape had become, without intending to, the de facto standard-setter for the web platform.

The company’s financial position was peculiar. Navigator was nominally free for non-commercial use, but commercial users paid license fees. Netscape’s real revenue came from server software — the Netscape Enterprise Server that companies used to host their websites and run commerce applications. The browser was the distribution mechanism for an audience that was being sold server software.

The IPO That Started an Era

On August 9, 1995, Netscape Communications went public. The company was sixteen months old and had not yet made a profit. Its S-1 filing contained no projections of when it would. The stock was priced at twenty-eight dollars per share after the underwriters doubled the price from the originally planned fourteen. At market open, demand was so intense that trading was delayed for nearly two hours while the underwriters found equilibrium. When it finally opened, the stock immediately reached seventy-one dollars. It closed at fifty-eight dollars and twenty-five cents, giving a company that had existed for less than a year and a half a market capitalization of approximately $2.7 billion.

Andreessen, twenty-four years old, appeared on the cover of Time magazine — photographed barefoot, seated on a throne — with the headline “The Golden Geeks.” It was the first indication to mainstream America that the internet was a financial phenomenon, not merely a technical one.

The Netscape IPO is widely credited as the igniting event of the dot-com boom. It established that internet companies could achieve enormous valuations based on user growth and strategic position rather than current earnings. Within months, investment banks were lining up dozens of internet companies to take public on the same logic. The period and its aftermath are covered in The Dot-com Bubble.

Microsoft’s Response and the Browser Wars

Microsoft had watched the internet’s growth with increasing alarm. In May 1995, Bill Gates circulated an internal memorandum — “The Internet Tidal Wave” — declaring the internet the most important development since the IBM PC and ordering every Microsoft product team to orient its work toward it.

The strategic threat Microsoft identified was specifically Netscape’s. Navigator, running on top of Windows, was becoming a platform within a platform. If developers wrote applications to run inside the browser — which Netscape was actively encouraging, having shipped JavaScript in Navigator 2.0 — the underlying operating system would become irrelevant. Windows would be reduced to a utility layer, its value commoditized. The applications relationship between developer and platform would belong to Netscape, not Microsoft.

Microsoft’s response was Industrial. It licensed Mosaic source code (through Spyglass, Inc., which held the master license from the University of Illinois), built Internet Explorer, and bundled it with Windows 95 at no charge. Then with Windows 98. It required PC manufacturers to display IE prominently on the Windows desktop as a condition of their Windows license, and prohibited them from prominently featuring competing browsers. Microsoft employed over one thousand engineers on Internet Explorer at the peak of the browser wars. The development resources were essentially unlimited.

Internet Explorer was technically behind Navigator in 1995. By 1997, Microsoft had largely caught up. By 1999, IE’s market share had surpassed Netscape’s. By 2000, it was at roughly sixty percent and climbing. Netscape’s revenue had collapsed as IE — free, integrated, and pre-installed on every Windows computer — replaced Navigator for most users.

Dead End: The Platform That Was Right Too Early

Warnung

Netscape’s core strategic argument — that the browser would become the platform, making the underlying operating system irrelevant — was correct as a long-term prediction and fatal as a short-term business plan. The browser did eventually become the platform: by the 2010s, web applications had substantially displaced desktop software for most users. But the transition required JavaScript engines more than ten times faster than those available in 1995, APIs (XMLHttpRequest, WebSockets, WebGL, WebAssembly) that Netscape hadn’t shipped and couldn’t ship with 1990s hardware, and broadband connections that less than five percent of the population had in 1999. Netscape was right about the destination. It could not survive long enough to see the infrastructure catch up with the vision. The lesson is not that Netscape was wrong but that being right about the future by ten years is commercially indistinguishable from being wrong.

Netscape was acquired by AOL for $4.2 billion in stock in November 1998. The acquisition was a lifeline that also marked the end: AOL was a company building a walled-garden internet experience, and the open-web browser Netscape had championed was ideologically at odds with everything AOL represented. Netscape 6, a complete rewrite using the nascent Mozilla platform, shipped in November 2000 and was widely regarded as slow, buggy, and too late. Netscape formally ceased development in 2008.

The open-sourced Netscape code, however, did not die. Released as Mozilla in 1998, it became Firefox under the leadership of Mitchell Baker — the browser that eventually broke Internet Explorer’s monopoly and established that a community-built open-source browser could compete with the world’s largest software company.

After Netscape: Loudcloud, Opsware, and a16z

Andreessen co-founded Loudcloud in 1999 with Ben Horowitz. The company provided hosted internet infrastructure — essentially what would later be called cloud computing — to the dot-com companies that were proliferating at speed. When the bubble burst in 2000 and 2001, most of Loudcloud’s customers collapsed or dramatically reduced their spending. The company pivoted to enterprise software under the name Opsware, survived by selling its cloud division to EDS, and was eventually acquired by HP for $1.6 billion in 2007.

In 2009, Andreessen and Horowitz co-founded Andreessen Horowitz — known by its shorthand, a16z, which represents the sixteen letters between the “a” and the “z” in their combined name. The firm grew rapidly into one of Silicon Valley’s most influential venture capital organizations. Its model distinguished itself from traditional VC by providing portfolio companies with an extensive services infrastructure — recruiters, executive coaches, marketing advisors, legal specialists — rather than capital and board seats alone. Early investments included Facebook, Twitter, Airbnb, GitHub, Skype, and Lyft.

In August 2011, Andreessen published “Why Software Is Eating the World” in the Wall Street Journal. The essay argued that every industry — logistics, retail, financial services, healthcare, education — would be disrupted by software companies building on the internet’s infrastructure. It became one of the most cited pieces of technology writing of the decade, the investment thesis of a generation made explicit.

In 2023, he published “The Techno-Optimist Manifesto” — a forceful, deliberately provocative argument for accelerationist technology development, explicitly opposing what he characterized as an emerging anti-progress consensus among regulatory agencies, academic institutions, and parts of the technology industry itself. The manifesto attracted controversy for its tone and for the positions it took on effective altruism, AI safety, and degrowth economics.

The undergraduate who wrote a browser in an after-hours coding sprint in 1992 had become one of the most powerful and divisive voices in the technology industry. Whether that trajectory represents a fulfillment or a betrayal of what the early web promised depends, in large part, on what you thought the early web was for.