How the Internet Happened

by Brian McCullough

  It helped that the NCSA was relatively flush with cash and resources in the early 1990s. It had gotten a large amount of funding thanks to the recently passed High Performance Computing Act of 1991, more commonly referred to as the “Gore Bill.”* All the wired infrastructure, all the superfast computing machines and the small army of undergraduate and graduate students the NCSA employed to assist with research projects, were paid for and paid by, in part, the government.

  “NCSA was heaven,” remembers one of the students working there in the early nineties, Aleks Totic. “They had all the toys, from Thinking Machines to Crays to Macs to beautiful networks. It was awesome.”3

  Another student programmer, Jon Mittelhauser, would remember, “We were all just kids hanging out in the basement of what was called the software development group.”4 The professors who ran the research programs that were the NCSA’s bread and butter assigned the projects, and the pool of “kids” in the basement coded away to the profs’ specifications.

  In 1992, one of those kids was a twenty-one-year-old by the name of Marc Andreessen. Born in Cedar Falls, Iowa, on July 9, 1971, Andreessen grew up in New Lisbon, Wisconsin (pop. 1,450),5 where his father was a feed salesman and his mother was a shipping clerk at Lands’ End. Computers fascinated Andreessen when he was a child, and he taught himself how to program at an early age. But he was no prodigy. Built large, at six feet two, with a loud, excitable personality, he was not exactly a wallflower, and it set him apart. Another NCSA student programmer named Rob McCool remembers of Andreessen, “All of the [computer science students] I’d come across were all quiet, kind of nerdy types. And here’s this gigantic Scandinavian guy with a purple computer and he’s wild-eyed and telling me about all this stuff that’s gonna be great.”6

  Andreessen was voluble and enthusiastic, but he also had an antiauthoritarian, independent streak that his peers came to appreciate. When a research team Andreessen and McCool were part of got hung up on a coding problem relating to an assigned project, Andreessen simply junked the existing framework and hacked together his own solution. “And I was like, ‘Dude, really? Can you do that?’ ” McCool remembers. “And he was like, ‘Yeah, well, my boss hasn’t noticed yet.’ ”7

  Andreessen had joined the NCSA as a part-time student programmer, doing menial coding work for $6.85 an hour. The researcher who hired Andreessen was Ping Fu, who had had a hand in the groundbreaking “morphing” computer graphics featured in the recent feature film Terminator 2: Judgment Day. Andreessen’s main task at the NCSA was coding Fu’s visualization projects. But what really caught Andreessen’s imagination during those hours in the NCSA basement, the computing technology that he was telling McCool and others was “gonna be great,” was that latest and greatest thing on the Internet: the World Wide Web.

  With the NCSA’s fast computers and even faster Internet connections, Andreessen and the other kids in the basement were perfectly positioned to catch the wave of the Web when it took off. In fact, the NCSA was just the sort of academic research organization that Tim Berners-Lee was fervently hoping would adopt his invention. At this point in the web’s development, Berners-Lee had just recently open-sourced his project to the world, in the hopes that he could “let a thousand flowers bloom” by inviting others to contribute to the project’s development. At the time, there were maybe a couple hundred software developers in the entire world experimenting with the web, and they all hung out and exchanged ideas with Berners-Lee on a Usenet newsgroup called WWW-Talk.

  In November 1992, there were only a few dozen WWW servers in the world. By the end of that same month, one of them happened to be at the NCSA, courtesy of Marc Andreessen.8 On November 16, 1992, Andreessen showed up in the WWW-Talk message group for the first time, joining the various conversations about HTML, web servers and web design and generally volunteering to pitch in on the grand project of moving the web forward.9

  Moving forward meant a better web browser. A browser is a software application that allows a user to both navigate and view the web. Berners-Lee himself had coded the first browser back when he had invented the web. But, as a part of his new crowdsourcing efforts, he had thrown the door open to anyone who wanted to try their hand at coding a better one. Dozens of developers around the world accepted the invitation, and several of them turned out to be students around the same age as Andreessen. At the University of Kansas, several students created the text-based Lynx browser. Pei-Yuan Wei developed the ViolaWWW browser while pursuing a degree at UC Berkeley. If you wanted to make a splash in the early web community, the way to do it was to code and release a better browser, and Marc Andreessen wanted to make a splash.

  Andreessen himself would later describe the early web this way:

  PC Windows had penetrated all the desktops, the Mac was a huge success, and point-and-click interfaces had become part of everyday life. But to use the Net you still had to understand Unix. . . . And the current users had little interest in making it easier. In fact, there was a definite element of not wanting to make it easier, of actually wanting to keep the riffraff out.10

  Andreessen’s big idea in the winter of 1992–93 was to let the riffraff in. He wanted to release a simpler, more user-friendly browser. He wanted it to be point-and-click and windowed. He wanted to make the web look familiar to someone who was comfortable using a personal computer, as opposed to the Unix workstations most of the researchers on the web were used to. And, crucially, he wanted the web to look as sexy as it felt to people like him who were enthusiastic converts. He wanted to add pictures. Says Aleks Totic: “[Andreessen] was like, ‘Oh, there could be newspapers on the Net and all this information can be out there for everyone. How phenomenal could that be?’ ”11 In short, Andreessen had a vision for the web in which someday everything would be possible: graphics, news, commerce, even cat videos.

  So, Andreessen turned his special brand of infectious enthusiasm on his fellow NCSA coders. The first person he targeted was his colleague Eric Bina. Bina was older than Andreessen (almost thirty) and a full-time, salaried NCSA employee. Bina was also a much better programmer than Andreessen was. Bina initially begged off the project, but Andreessen’s enthusiasm and persistence eventually won him over. The “browser project” that Andreessen and Bina undertook began sometime in December 1992. Bina wrote the majority of the original code, but the features were also what made their browser such a leap forward, and it was Andreessen who was coming up with the features.

  In a little over a month of nearly round-the-clock coding, they had their browser ready. It was called X Mosaic. On Saturday, January 23, 1993, the official “0.5” version of the browser was posted to the Internet on the NCSA’s servers. The accompanying release note from Andreessen himself said:

  By the power vested in me by nobody in particular, alpha/beta version 0.5 of . . . X Mosaic is hereby released.

  The last line of the message was the FTP address telling others where they could go to download and install the browser themselves. Within days, no less a web authority than Tim Berners-Lee forwarded and endorsed Andreessen’s announcement:

  An exciting new World-Wide Web browser has come out, written by Marc Andreessen of NCSA.

  This browser was called “X Mosaic” because it was designed to work with X Window, a graphical user interface popular with users of Unix machines. It was designed for the computers that researchers and academics used. In other words, it was preaching to the already-converted web choir. And that was not what Andreessen was after, of course. Using X Mosaic as a proof of concept, he turned his enthusiasm on others in the NCSA basement to get them to write versions of his browser for the computers that the riffraff used.

  NCSA’s young programmers signed on to program these versions, each according to his own platform of choice. Jon Mittelhauser and Chris Wilson developed the PC version. Aleks “Mac Daddy” Totic and Mike McCool wrote the Macintosh port. And since X Mosaic handled only the consumption end of the web experience, the growing team thought it would be
a good idea to tackle the delivery end as well. Thus, McCool’s twin brother Rob wrote Mosaic web server and publishing software that would eventually be released alongside the browsers.

  The kids in the basement did their thing and then released it out into the world. That was how the web worked in 1993; that was what Tim Berners-Lee had hoped would happen. If someone had a better way of doing things, they coded it up and made it available for other people to try. If others liked it, they downloaded it. If they didn’t, well, they didn’t. And if these users had problems, found bugs, had ideas for improvements or wanted to contribute new features, then they got in touch with the creators over email or the Usenet message boards and bitched about it. The kids at the NCSA, surrounded by empty pizza boxes and soda cans, released updated versions—and then maybe a week later they released another updated version based on user feedback.12 The process was very communal and very real-time.

  ■

  WITHIN EIGHTEEN MONTHS, Mosaic was the biggest thing on the web, and probably the biggest thing on the Internet at large. In January of 1993, shortly after Mosaic launched, the number of websites in existence was in the hundreds. By the end of 1994, the number of websites in the world had passed tens of thousands.13 In a similar time frame, the number of web hosts had risen tenfold.14 In a way, one could argue that Mosaic helped make the web, and vice versa. As the first browser designed for the common computer user, Mosaic had a symbiotic sort of chicken-and-egg relationship with the web. For millions of PC and Mac users, Mosaic was their first glimpse of the web. Once they saw what the web could do, they wanted to go off and code their own websites.

  Within those first eighteen months of launch, Mosaic probably delivered 3 million browsers into users’ hands.15 That may seem like a small number, but then, there probably weren’t many more than 3 million people on the web before that point. Toward the tail end of 1994, Mosaic was adding as many as 600,000 new users every month. It is safe to say that by that point the vast majority of people surfing the web did so via a Mosaic web browser.

  The key innovation of the Mosaic browser was Andreessen’s insight that in order to make the web sexier, he simply had to release a browser that enabled the sexiness he imagined. On February 25, 1993, mere weeks after Mosaic’s initial beta launch, Andreessen was on the WWW-Talk message boards making a proposed addition to HTML of an “inline” image tag that would allow for images to be coded directly into web pages. Prior browsers opened images—and really any non-HTML file type—as a separate window. Inline images would make web page design more akin to the page layout of a magazine or newspaper.

  Adding color and sexiness to the web was part of what made Mosaic take off, and part of what made the web take off at exactly the same time. But even the web’s creator was among those who felt that Andreessen’s penchant for multimedia was a little much. Andreessen later admitted, “Tim [Berners-Lee] bawled me out in the summer of ’93 for adding images to the thing.”16

  ■

  “HE ONLY WANTED TEXT,” Andreessen has said of Berners-Lee’s objections when they finally met face-to-face at the World Wide Web Wizards Workshop in Cambridge, Massachusetts, the first true developer conference. “He specifically didn’t want magazines. He had a very pure vision. He basically wanted it [the web] used for scientific papers. His view was that images are the first step on the road to hell. And the road to hell is multimedia content and magazines, garishness and games and consumer stuff. I’m a Midwestern tinkerer type. If people want images, they get images. Bring it on.”17

  For his part, Berners-Lee has denied that images discomfited him. “Of course we did approve of images, in fact we had images on the Web before anybody else,” he has said. But then he adds, “Like diagrams in talks for example.”18

  Years later, even Mosaic cocreator Eric Bina would admit that he had reservations about adding images and multimedia to the web. At the time, he was mainly concerned about bandwidth issues (this was the era of dial-up modems; images could take entire minutes to load onscreen) but he was also worried that he and Andreessen were opening the floodgates to frivolity and junk. “And I was right! People abused it horribly,” Bina said later. “But Marc was also right. As a result of the glitz and glitter, thousands of people wasted time to put in pretty pictures and valuable information on the Web, and millions of people use it.”19

  ■

  THE MILLIONS OF DOWNLOADS to users around the world meant that Mosaic was probably the most successful software product ever designed for—or released on—the Internet up to that point. By the end of 1994, it was clear that the World Wide Web was rapidly taking over the Internet at large. For the millions of Mosaic users, the web almost was the Internet. But then, those millions of users were not exclusively the academics and researchers the web had been designed for. Increasingly, they were also home computer users; business computer users; the uninitiated; the uninvited; the riffraff. Mosaic had become the most successful project in computer science by leaving the computer scientists behind and appealing to the mainstream. Fortune magazine named the Mosaic browser one of its products of the year (alongside the Wonderbra and Mighty Morphin Power Rangers), writing, “This software is transforming the Internet into a workable web . . . instead of an intimidating domain of nerds.”20

  As these things tend to go, the popularity of Mosaic, and the nerd celebrity of the Mosaic team especially, started to create friction within the NCSA. The higher-ups at the center had originally thought of Mosaic as just another software project, very much fitting into their larger purview as a computing research institution. But, over the course of 1993, the status of the Mosaic browser project changed; it became a major NCSA priority. At first the NCSA bigwigs didn’t seem to “have any clue who we were, and we liked it that way,” said Jon Mittelhauser, but once Mosaic took off, “We suddenly found ourselves in meetings with forty people planning our next features, as opposed to the five of us making plans at 2 AM over pizzas and Cokes. Aleks, who had basically done the Mac version [on his own], suddenly found out that there were three or four other people working on it with him, according to the NCSA. And they were like his bosses, telling him what to do.”21

  Chris Wilson was another of the NCSA student programmers who would later go on to work at Microsoft and develop the first Internet Explorer browsers. “I think that Marc and some of the other guys there really wanted to see NCSA just drop everything else like a hot rock and go totally support the web and scale up to do it,” Wilson says. “If you were in a startup and you saw one of your products getting so much attention and having so much potential, absolutely you’d figure out how to do that, right? You’d go mortgage your house.”22

  In fact, the Mosaic team already was functioning like a software startup in all but name, while the NCSA was still thinking of the browser as a glorified research project. Increasingly, this conflict in vision extended to the very structure of the core team of programmers. The part-time student coders were muscled out as the higher-ups assigned seasoned, full-time employees to the project. It was suggested to Andreessen especially that, for the good of the project, he should step aside and let more experienced hands take over. “Don’t you think it’s time to give someone else a chance to share the glory?”23 he was asked.

  In December of 1993, Mosaic and the web made the front page of the New York Times. NCSA director Larry Smarr was pictured and quoted: “Mosaic is the first window into cyberspace,” he said.24 Neither Marc Andreessen nor anyone else on the Mosaic team was even mentioned.

  “[Andreessen] had to lead at NCSA,” says Aleks Totic. “And if he couldn’t lead, he had to leave.”25

  Andreessen was due to graduate that same December. He didn’t even bother to pick up his diploma. By the end of 1993, just a year after launching the Mosaic browser, Marc Andreessen was in Silicon Valley looking for work.

  ■

  THE SILICON VALLEY that Marc Andreessen found himself in by early 1994 was actually at a historical low ebb, considering what was in store. Th
e short but sharp recession of 1990–91 hit the technology industry hard. PC shipments fell by 8% in 1991, the first such drop in recorded industry history.26

  “I thought I had missed the whole thing,” Andreessen would later say of his arrival in California. “The overwhelming mood in the Valley when I arrived was that it was done. The PC was done, and by the way, the Valley was probably done because there was nothing else to do.”27

  Forget the Valley, in 1994, what was the something else that Marc Andreessen could do? To us now, the answer is obvious: form a startup; get venture capital backing; release a product; gain millions of users; go public; become a billionaire. This is only the obvious path to modern minds because of the “something else” that Marc Andreessen would do in 1994: cofound Net­scape, the first true Internet company, the first real “dot-com.” At the time, there was no template for Marc Andreessen to do a web startup, because Marc Andreessen hadn’t created that template yet.

  “I had some idea that I wanted to be part of a new company,” Andreessen says, “but I didn’t even know what a VC [venture capitalist] was.”28

  ■

  JIM CLARK IS FAMOUS in Silicon Valley history for having founded three different billion-dollar companies. By the beginning of 1994, Clark was just departing billion-dollar company number one: Silicon Graphics (SGI). Jim Clark’s tenure at Silicon Graphics was not ending on a happy note. Despite being the founder, despite being largely responsible for the development of modern computer-aided design and computer graphics (those dinosaurs in Jurassic Park? You can thank Silicon Graphics for those), despite turning SGI into a multibillion-dollar publicly traded enterprise, Clark found himself edged out of his own company.