Microsoft Won the Browser War So Completely the Government Couldn't Save the Loser.
Internet Explorer went from under 20% to over 90% of the market in roughly six years. The weapon wasn't better code - it was Windows itself. And the lawsuit that proved it ended with Microsoft keeping everything.
Comes with a free Standards-War Battle Map template.
In 1996, fewer than one in five people on the web used Internet Explorer. By 2000, more than four in five did, and within a couple of years the number crept past ninety.6 No browser in history climbed that fast on the merits, and IE didn't climb on the merits. It climbed because it came pre-installed on a machine almost everyone already owned. The browser war wasn't won in the code. It was won at the factory, before the customer ever touched the keyboard.
The story everyone tells is a triumph of engineering: Microsoft out-built Netscape, IE got better, and the better product won. Almost none of that is the cause. IE 1.0 wasn't even Microsoft's own code - it was licensed Spyglass Mosaic, taken in late 1994 for roughly $2 million upfront.5 What Microsoft owned was not a better browser. It was the one thing Netscape could never buy at any price: the desktop itself.
The browser that rode in on someone else's monopoly
Here is the mechanism, worked all the way down. A browser in the mid-1990s was a thing you went and got - Netscape's came on a disk or a download, an act of deliberate choice. Microsoft removed the choice. It bundled IE free with every copy of Windows, and Windows sat on the overwhelming majority of the world's PCs.6 Distribution wasn't a marketing channel for IE; it was Windows, and Windows was a monopoly. The court later found exactly this - that Microsoft held monopoly power over PC operating systems and used it to crush threats, Netscape among them.1 The genius and the crime were the same act. Microsoft didn't beat Netscape in the browser market. It moved the fight to a market it already owned and let the geography do the winning.
And there was a tell hidden in the licensing. Spyglass had given Microsoft IE's original code for an upfront fee plus a royalty on each copy sold. By bundling IE free into Windows, Microsoft made the per-copy royalty evaporate - you can't collect a cut of a price that's now zero. Spyglass sued and settled for $8 million in early 1997.5 The same move that starved Netscape of oxygen also starved Microsoft's own supplier. 'Free' was never generosity. It was a weapon, aimed in every direction at once.
| Netscape's path | Microsoft's path | |
|---|---|---|
| How users got the browser | Sought it out, downloaded it | Already installed on Windows |
| Price to the user | Free, but a deliberate choice | Free, and the default |
| The real asset | The browser | The operating system underneath it |
| What had to win | The product | The distribution |
The strategy was so blatant it became the evidence against it
Most monopolies are hard to prosecute because the harm is diffuse and the intent is buried. Microsoft's wasn't. On May 18, 1998, the Department of Justice and the attorneys general of twenty states plus the District of Columbia filed jointly - a coordinated federal-state action, not a lone agency taking a flyer.8 They had a clean theory: monopoly in operating systems, used to tie a browser to it and smother a rival. The same OS leverage that made the strategy so effective made it impossible to disguise. Brilliance at scale leaves fingerprints, and these were everywhere.
The trial produced the most famous line of the era - that Microsoft intended to 'cut off Netscape's air supply.' It's worth being precise about it, because the legend has hardened into something the record doesn't fully support. The phrase came from Intel's Steven McGeady, testifying that a senior Microsoft executive had said it.7 That executive denied ever saying it. The government's own filings noted the awkward seam: at his deposition he'd answered 'it's possible, but I just don't recall,' and at trial it became 'never said.'4 The DOJ argued the shift undermined his credibility - and it's telling that the case didn't need the quote at all. The distribution numbers were the confession. The air supply was being cut off in plain sight, whether or not anyone had narrated it out loud.
“It's possible, but I just don't recall.”4
Microsoft lost the verdict and kept the company
Then came the part that reverses the moral. In November 1999 the court found the monopoly; in April 2000 it ruled Microsoft had violated both Sections 1 and 2 of the Sherman Act.1 In June 2000 it ordered the company split in two - an operating-system company and an applications company, the corporate equivalent of a death sentence.3 For about a year, the conventional story is that Microsoft was broken up. It wasn't. In June 2001 the D.C. Circuit affirmed the core finding - monopoly maintenance, the real crime - but vacated the breakup entirely and even reversed the finding that Microsoft had tried to monopolize the browser market itself.2 The remedy was sent back, and what finally landed was a consent decree about sharing programming interfaces and giving PC makers a little more flexibility. Microsoft kept its monopoly, kept its applications, kept IE on every desktop.
So both halves of the headline are true at once. Microsoft won the browser war - decisively, near-totally - and it lost the antitrust battle on the law. But losing the battle cost it almost nothing, because the appeals court took away the one remedy that would have mattered and left the prize untouched. The verdict was a moral defeat with no operational teeth. The browser war's loser was Netscape. The antitrust case's real loser was the idea that antitrust could still move fast enough to matter.
The durable lesson isn't 'bundling is evil' - it's that distribution can beat product whenever one player controls a chokepoint everyone else has to pass through. Netscape was competing for the browser; Microsoft was competing on the desktop, a layer below, where it had already won. If you have a platform, the cheapest way to win an adjacent fight is to make your entry the default rather than a choice. But carry the second half of the story too: a win this clean attracts enforcement precisely because it's this clean. Microsoft's edge survived the lawsuit only because the remedy collapsed on appeal - which is luck, not strategy. Build the default; don't bet your company on the regulator being too slow to unwind it.
The deepest irony is that Microsoft was found guilty of exactly what it did, and the punishment confirmed there was no punishment. It paid Spyglass $8 million for the code it weaponized, paid years of legal fees for a monopoly it got to keep, and walked away with Internet Explorer sitting on more than nine of every ten machines on earth.6 The browser war proved that the best place to win a market is one layer beneath it. The antitrust case proved something colder: that you can be caught, named, and convicted, and still keep the thing you weren't supposed to have - if the system that catches you can't decide what to do once it has.
Standards-War Battle Map
A one-page canvas for a winner-take-most format war: who's fighting, which platforms and partners back each side, where the installed base sits today, and which forces will tip the market before it locks. Blank to map a standards fight you're entering; filled as the worked example charting the battle the story describes. Use it to spot which side has already won and which is still spending to lose.
The worked example unlocks with a subscription. See plans →
Sources
Where this comes from — the filings, records, and reporting behind it.
- 1Judge Jackson issued Findings of Fact on November 5, 1999, holding that Microsoft's dominance of the x86-based PC operating systems market constituted a monopoly and that Microsoft had taken actions to crush threats to that monopoly including from Netscape, Java, Apple, and others. On April 3, 2000, he issued Conclusions of Law finding Microsoft violated Sections 1 and 2 of the Sherman Act.
- 2The D.C. Circuit Court of Appeals (June 28, 2001) affirmed Microsoft's monopoly maintenance violation under Sherman Act §2 but reversed the District Court's finding of attempted monopolization of the browser market and vacated the breakup remedy, remanding the tying claim under the rule of reason.
- 3The District Court ordered Microsoft split into two companies — one for operating systems, one for applications — in its Final Judgment of June 7, 2000 (97 F. Supp. 2d 59). This breakup was subsequently overturned on appeal.
- 4The DOJ's proposed findings of fact (filed with the court) documented that Paul Maritz at his deposition said 'it's possible, but I just don't recall' when asked about the 'cut off Netscape's air supply' statement, but at trial testified he 'never said' it — an inconsistency the government argued undermined his credibility.
- 5Microsoft licensed Spyglass Mosaic in December 1994/January 1995 for an upfront payment of approximately $2 million as the codebase for Internet Explorer 1.0. After Microsoft bundled IE free with Windows and paid only the minimum quarterly fee, Spyglass sued; the dispute settled for $8 million on January 22, 1997.
- 6Internet Explorer was bundled with every copy of Windows, which held over 95% of the desktop OS market by 2004. IE's adoption went from below 20% in 1996 to roughly 40% in 1998 and over 80% in 2000, achieving a peak share of approximately 94–95% in the early 2000s.
- 7Intel VP Steven McGeady testified at trial (November 1998) that Microsoft Senior VP Paul Maritz stated an intent to 'extinguish' and 'smother' Netscape and to 'cut off Netscape's air supply.' Maritz denied the quote in January 1999 testimony. Microsoft's attorney argued McGeady's own handwritten notes did not contain the word 'extinguish.'
- 8The DOJ and the Attorneys General of 20 states (plus D.C.) jointly filed antitrust complaints against Microsoft on May 18, 1998, charging monopolization of the PC operating systems market, attempted monopolization of the browser market, and illegal tying of IE to Windows under the Sherman Act.