The Internet Changes Everything (Again)

The Evolution of EdTech — Part 4 of 13

By Clarence Stephen | March 2026

2,200 words | 9 min read

In 1993, there were 130 websites in the world. By 1996, over 100,000. By the end of the decade, millions. And somewhere in that exponential explosion, a teacher in rural Iowa could — for the first time in history — show her students the same resources available to a kid at a Manhattan prep school.

That was the theory.

The reality was a 28.8 kbps modem that screamed like a fax machine having a panic attack every time it connected, a phone line that kicked you offline whenever someone picked up the receiver, and a fifteen-minute wait to load a single webpage with images. But the idea — that the sum total of human knowledge could be accessible to anyone with a connection — was genuinely intoxicating. And it hit education like a freight train running on dial-up.

Wiring the Schools

In 1994, only 35 percent of American public schools had internet access. The internet was clearly going to be important — the most important communication technology since the printing press — and two-thirds of schools were locked out. Not a gap. A chasm.

The federal response was, by government standards, both swift and massive. The Telecommunications Act of 1996 included the E-Rate program, which provided discounts of 20 to 90 percent on internet access and networking equipment for schools and libraries. The discount scaled to poverty: the poorest schools got the largest subsidies. Between 1998 and 2005, E-Rate distributed over $20 billion to connect American schools to the internet.

The numbers were impressive. By 1998, 89 percent of schools had some form of internet access. By 2002, 99 percent. The connectivity problem was, if not solved, at least addressed at a scale few government programs have matched.

But connecting a school to the internet and actually using the internet to improve education turned out to be — as with every previous technology in this series — two entirely different problems. The second was vastly harder than the first.

The Computer Lab

If you were there, this will sound familiar.

Your class was scheduled for the computer lab — a windowless room with 25 beige desktops arranged in rows, humming with the particular anxiety of hardware being asked to perform beyond its capabilities. You waited while the teacher figured out the login. You opened Netscape Navigator (later Internet Explorer, to the dismay of anyone with opinions about browsers). You typed a URL from the whiteboard, or went to AltaVista and searched for something related to your assignment.

Then you waited. The page loaded in strips, top to bottom, like a window blind slowly opening. Images materialized pixel by pixel. If someone down the hall started downloading a file, your connection crawled.

Eventually you found something. More likely, you found a GeoCities page with a tiled background, animated GIFs of construction workers, and text in six colors and three fonts, written by an enthusiast whose expertise was unclear and whose citations were nonexistent.

You printed it out (your teacher required a physical source), stapled it to your report, and called it research.

That was the internet in education, circa 1997. Chaotic, unreliable, often absurd — and utterly transformative in ways that would take years to become visible.

What Changed

The internet did not, as breathless magazine articles predicted, "replace the textbook" or "make classrooms obsolete" or "democratize education overnight." What it did was subtler and, in retrospect, more consequential.

It changed the fundamental question of education. Before the internet, the central problem was access to information. Knowledge was locked inside books, locked inside libraries, locked inside institutions. If you wanted to learn about quantum mechanics, you needed a library that owned a physics textbook, or a school that offered a physics course, or a teacher who knew something about it.

The internet made the access problem trivial. Information was everywhere. The new challenge — and it's still the defining challenge — wasn't how to find information but how to evaluate it. How to tell credible sources from unreliable ones. How to synthesize multiple perspectives. How to think critically about what you read.

Schools were not remotely prepared. The entire system was built on a scarcity model: the teacher possesses the knowledge, the textbook contains the facts, the student absorbs and reproduces. Suddenly, every student had access to more information than any teacher could master. The power dynamic shifted under educators' feet, though it took institutions years to notice.

It connected people. Before the internet, a student in rural Alaska and a student in Tokyo had essentially no way to interact. Afterward, they could exchange emails, collaborate on projects, share perspectives in real time. Programs like ePals (founded 1996) connected classrooms across continents. The educational potential of talking to someone fundamentally different from you — different place, different language, different assumptions — is enormous, and the internet made it possible at scale for the first time.

It created the first online learning. The mid-1990s saw the earliest experiments in what would become online education. Universities started offering courses via email and rudimentary web platforms. The Virtual High School, founded in 1996, offered online courses to students in schools that lacked the resources or enrollment to provide them locally. A kid in a small town who wanted to study Japanese or AP Physics could take a course taught by a qualified teacher hundreds of miles away.

These early online courses were primitive — mostly text, assignments submitted via email, discussions on bulletin boards. But they planted the seed for everything that followed: MOOCs, Khan Academy, Coursera, and the massive expansion of remote learning that COVID-19 would force on the world two decades later.

The Digital Divide

The internet also exposed — and for a time actively widened — one of education's deepest structural inequities.

Before E-Rate, the numbers were damning. Schools in affluent districts had internet access in roughly half their classrooms. Schools where 75 percent or more of students qualified for free lunch? About 8 percent of classrooms. Eight percent. The same technology being hailed as the great equalizer was distributed with almost cartoonish inequality.

E-Rate narrowed this gap at school, but did nothing about the gap at home. In 1998, households earning more than $75,000 were six times as likely to have internet access as those earning less than $15,000. Kids from wealthy families could research and practice online in their bedrooms. Kids from poor families couldn't.

The term "digital divide" entered the national vocabulary, and it never left. Two decades later, COVID would demonstrate with brutal clarity that the divide had never been closed. It had just been papered over during school hours.

The Dot-Com Boom and EdTech 1.0

The late 1990s saw the first wave of venture capital flowing into educational technology. I find this period fascinating because I've seen the pattern from the investor side. Dozens of startups emerged with names that sounded generated by a particularly optimistic algorithm: LearnStar, eSylvan, Classroom Connect, Lightspan.

Most failed. The dot-com crash of 2000 wiped out the edtech sector almost entirely. The companies that survived — Blackboard (founded 1997), Pearson's digital division, a handful of others — would dominate the next decade. But the first edtech boom taught a lesson the industry keeps forgetting and relearning: schools are not like other markets.

Schools move slowly. They purchase by committee. They're bound by regulations, standards, and procurement processes that would reduce a startup founder to tears. They're chronically underfunded. And the end users — students and teachers — aren't the buyers. The buyers are administrators and school boards, whose incentives diverge sharply from what classrooms actually need.

Every edtech startup that has failed has failed, at least in part, because it underestimated how hard it is to sell to schools. The ones that succeeded learned to work with the system rather than trying to disrupt it — a word the edtech industry borrowed from Silicon Valley and has been misapplying with enthusiasm ever since.

The Legacy

The internet changed education, but not in the ways 1990s prophets predicted. Teachers weren't replaced. Schools weren't rendered obsolete. Opportunity wasn't automatically equalized.

What the internet did was lay the infrastructure — physical, cultural, psychological — for everything that came after. The fiber optic cables. The Wi-Fi networks. The expectation that information should be instantly available. The understanding, still incompletely absorbed by institutions, that the role of education in an information-rich world isn't to deliver knowledge but to develop the capacity to use it.

By 2005, you couldn't imagine education without the internet. It had become like electricity — invisible, essential, taken entirely for granted.

The question, as the new millennium began, was what institutions would build on top of this infrastructure. The answer turned out to be the Learning Management System — a technology so ubiquitous and so universally loathed that it deserves its own chapter.

Next in the series: "The Rise of the LMS and the Standardization Era" — when education met Blackboard, and nobody was happy about it.

Sources

Clarence Stephen is the founder of Readify, an AI-powered literary platform. Yale B.Sc. Physics. Previously: Tiger Global, Morgan Stanley. Learn more at ireadifybooks.com

What could AI do for your classroom?

Readify uses local AI to adapt reading material to each student's level — generating illustrations, simplifying vocabulary, and tracking comprehension in real time. No student data leaves the building.

Bring Readify to Your School Try the Platform

Liked this article?

Get the full Evolution of EdTech series delivered to your inbox, plus weekly insights on AI in education.

Subscribe to Readify Branches