The History of the Internet in a Nutshell

Most people alive today have never known a world without the internet. They’ve never had to drive to a library to look something up, never had to wait three days for a letter to arrive, never had to call a business during business hours because that was literally the only option. The internet is so embedded in how life works now that it’s hard to even think of it as a technology that was invented, debugged, argued over, and slowly stitched together by a relatively small number of people over several decades.

But that’s exactly what happened. The internet didn’t arrive fully formed. It started as a Cold War experiment funded by the US military, built by academics who were mostly trying to solve boring technical problems like how to send a file from one computer to another without the whole system crashing if one node went down. Nobody in that early group was sitting around imagining online shopping or social media or video calls. They were just trying to get packets of data from point A to point B reliably. The vision came later, and even then it arrived in pieces, usually invented by someone who didn’t realize they were changing everything.

The history of the Internet is genuinely one of the stranger stories in modern technology, because so much of it happened by accident, by stubborn engineers ignoring critics, by college students building things in dorm rooms, and by companies making decisions that seemed reasonable at the time and turned out to be catastrophically wrong or accidentally brilliant. It involves governments, universities, venture capitalists, teenagers, and at least one person who invented something fundamental and never got rich from it.

What follows is that story, told in a way that actually makes sense: with real dates, real names, and the kind of context that explains why things happened when they did instead of just listing events in order.

TL;DR

• The internet started in 1969 as ARPANET, a US Defense Department research network connecting four universities.
• Tim Berners-Lee invented the World Wide Web in 1989 at CERN, which is what most people actually mean when they say “the internet.”
• The 1990s dot-com boom saw billions invested into internet businesses, most of which failed spectacularly by 2001.
• Broadband replacing dial-up in the early 2000s changed everything about how people used the web.
• Social media, the smartphone, and cloud computing transformed the internet from a destination into infrastructure that runs underneath daily life.
• The internet is now more centralized than its founders intended, with a handful of companies controlling most of what people see and do online.

The History of the Internet Starts With a Nuclear Fear, Not a Tech Vision

Here’s something most people get wrong about where the internet came from. It wasn’t invented to survive nuclear war. That’s a myth that got attached to ARPANET early and stuck because it sounds good. The actual origin is more mundane and more interesting.

ARPANET and the Problem of Sharing Computing Resources

In 1957, the Soviet Union launched Sputnik. The US panicked, as it tended to do during the Cold War, and in 1958 created ARPA, the Advanced Research Projects Agency, specifically to make sure the Americans didn’t fall behind the Soviets in science and technology. ARPA funded research across universities, and by the mid-1960s a problem had emerged: the computers at different universities couldn’t talk to each other. If you were a researcher at MIT with results you wanted to share with someone at Stanford, you mailed them a tape. Physically. In the post.

J.C.R. Licklider, a psychologist turned computer scientist working at ARPA, wrote a memo in 1963 describing what he called an “Intergalactic Computer Network,” where computers could communicate with each other across distances. He wasn’t building anything yet. He was describing a vision that would take another six years to begin existing.

The actual network came from the work of Larry Roberts at ARPA, who in 1966 was tasked with building a network that connected computers at different research universities. The key technical problem was how to send data reliably across a network where any individual connection might fail. The solution, packet switching, was developed independently by Paul Baran at RAND Corporation and Donald Davies at the UK’s National Physical Laboratory. Packet switching broke messages into small chunks called packets, sent them through whatever path was available, and reassembled them at the other end. If one route went down, packets took another. That’s it. That’s the foundational idea behind everything.

October 29, 1969: The First Message

On October 29, 1969, the first message was sent over ARPANET. The receiving end was at Stanford, the sending end was at UCLA. The message was supposed to be “login.” The system crashed after the first two letters. The first word ever transmitted over what would become the internet was “lo.” Which is, honestly, a better origin story than anyone planned.

By 1971, ARPANET connected 15 nodes. By 1973 it had reached the UK and Norway, making it technically international. The network was small, expensive, and entirely limited to universities and defense contractors. Nobody outside that world knew it existed or had any reason to care.

Email: The Killer App Nobody Planned

In 1971, Ray Tomlinson, an engineer at Bolt Beranek and Newman, sent the first email between two computers that were sitting next to each other in the same room. He used the @ symbol to separate the user name from the host machine, a decision he made in about 30 seconds that turned out to be permanent. Email spread faster than anyone expected. By 1973, email was 75% of all ARPANET traffic. The entire original purpose of the network, sharing computing resources, got swamped almost immediately by people just sending messages to each other.

This pattern would repeat many times in internet history. Engineers build something for one purpose, users find a different purpose, the different purpose becomes the actual product. It happened with email, it happened with the web, it happened with Twitter, it happened with Zoom. The technology points one direction and the humans using it immediately walk the other way.

From Military Network to Public Infrastructure: The 1970s and 1980s

ARPANET in the 1970s was genuinely useful for researchers. But it was still a closed system. You needed to be affiliated with a funded institution to access it, and the software was complicated enough that getting online required technical training. The internet as a public thing, something anyone could access, didn’t exist yet and wouldn’t for another 20 years.

TCP/IP: The Protocol That Made Everything Work Together

The problem by the mid-1970s was that ARPANET wasn’t the only network. Universities and companies were building their own networks with their own protocols, and none of them could talk to each other. You could get from UCLA to MIT on ARPANET, but you couldn’t connect from ARPANET to ALOHANET in Hawaii or to the packet radio networks the military was using in the field.

Vint Cerf and Bob Kahn solved this in 1974 by designing TCP/IP, the Transmission Control Protocol and Internet Protocol. TCP/IP is a set of rules for how data should be packaged, addressed, transmitted, and received. Critically, it was designed to work across different types of networks without requiring them to change their underlying architecture. A network using radio signals and a network using copper wire could both speak TCP/IP and exchange data.

January 1, 1983 is sometimes called the “birthday of the internet” because on that date ARPANET officially switched to TCP/IP. The different networks could now interconnect, and the internet, in the technical sense of a network of networks, started to take shape. But it was still not something a regular person could access.

The Usenet, Bulletin Boards, and Early Online Communities

While ARPANET was growing in universities, a parallel world of online communities was developing outside it. Usenet, created in 1980 by Tom Truscott and Jim Ellis at Duke University, was a distributed discussion system where people could post messages to topic-based groups, called newsgroups, and read replies. It ran over ordinary phone lines rather than ARPANET infrastructure and spread to universities and companies rapidly.

By the mid-1980s there were thousands of newsgroups covering topics from computer programming to science fiction to, inevitably, pornography. Usenet was chaotic, lightly moderated, and genuinely unpredictable in the best and worst ways. It was also where a lot of internet culture developed before social media existed, including the concept of “trolling,” the practice of posting deliberately provocative content, which appeared on Usenet in recognizable form by 1992.

Alongside Usenet, Bulletin Board Systems, or BBSs, were phone-connected servers where people dialed in to post messages, download files, and play text-based games. At their peak in the early 1990s, there were over 60,000 BBSs operating in the United States. They were the internet before most people had the internet, and they produced the first generation of people who thought of computers as communication devices rather than calculation machines.

The World Wide Web: When the History of the Internet Really Accelerated

Here’s where things get confusing for a lot of people. The internet and the World Wide Web are not the same thing. The internet is the underlying network infrastructure, the cables, routers, and protocols that move data around. The World Wide Web is an application that runs on top of that infrastructure. Before the web existed, the internet was already real and already useful. But the web is what made it usable for everyone.

Tim Berners-Lee and the Proposal Nobody Took Seriously

In March 1989, a British physicist named Tim Berners-Lee working at CERN, the particle physics laboratory in Switzerland, submitted a proposal to his manager titled “Information Management: A Proposal.” The proposal described a system for managing documents at CERN using a concept he called hypertext, where documents could link to other documents. His manager, Mike Sendall, wrote “vague but exciting” on the cover and approved a small amount of time for Berners-Lee to continue the work.

That’s it. That’s how the World Wide Web started. A vague proposal and a modest go-ahead from a manager at a physics lab.

By 1990, Berners-Lee had written the three foundational technologies that the web still runs on: HTML, the language for formatting documents; HTTP, the protocol for transferring them; and URLs, the addressing system for finding them. He built the first web browser, also called WorldWideWeb, and the first web server, running on a NeXT computer at CERN. The first website went live on August 6, 1991 at the address info.cern.ch and it explained what the World Wide Web was.

Berners-Lee never patented any of it. He made it all open and free deliberately, because he understood that a web that required licensing would never become universal. That decision is why the web grew the way it did. It’s also why he never became a billionaire from it, which seems like a reasonable trade when you look at how many billions it made for everyone who built on top of it.

Mosaic and Netscape: Making the Web Visual

The early web was text-based. Functional, interesting, but not visually compelling. In 1993, Marc Andreessen and Eric Bina at the National Center for Supercomputing Applications at the University of Illinois released Mosaic, the first web browser to display images inline with text. This sounds minor. It was not minor.

Before Mosaic, the web looked like a terminal. After Mosaic, it looked like a magazine, or at least a very basic one. Web traffic grew by 341,634% in 1993. That number is not a typo. The visual web was immediately more interesting to more people, and the growth rate reflects that.

Andreessen graduated, moved to California, and co-founded Netscape in 1994. Netscape Navigator became the dominant web browser within months of release. By late 1994 it had 75% of the browser market. The company went public in August 1995 in one of the most successful IPOs in history up to that point, with shares jumping from $28 to $75 on the first day, and Netscape had never turned a profit. It was the first clear signal that the stock market was prepared to bet heavily on internet companies based entirely on potential.

The Dot-Com Boom, the Crash, and What Was Left Behind

The late 1990s are hard to explain to anyone who didn’t live through them. The combination of a fast-growing web, a stock market in a sustained bull run, and a genuine sense that everything was about to change produced an atmosphere of investment mania that has rarely been matched in financial history.

Pets.com and the Logic of Irrational Exuberance

Between 1995 and 2000, billions of dollars flowed into internet companies with business models that would have been laughed out of a normal investment meeting. Pets.com raised $82.5 million in its IPO in February 2000 despite losing $61.8 million in the nine months prior and despite the fact that shipping heavy bags of dog food to individual households had obvious unit economics problems. The company went bankrupt nine months later. The sock puppet mascot from their Super Bowl ad was auctioned off.

Pets.com is the easy joke, but it wasn’t unique. Webvan raised $375 million and burned through it trying to build same-day grocery delivery infrastructure in the late 1990s, a concept that only became viable two decades later when Amazon, with vastly more capital and logistics expertise, made it work. Kozmo.com promised one-hour delivery of movies, snacks, and video games in cities and lasted until 2001. These companies weren’t stupid. They were early, which in startup terms is often the same as wrong.

The Nasdaq peaked at 5,048.62 on March 10, 2000. By October 2002 it had fallen to 1,114.11, a drop of 78%. Over $5 trillion in market value evaporated. Companies shut down, thousands of tech workers lost jobs, and the phrase “dot-com” briefly became a punchline.

What Actually Survived the Crash

The crash wiped out bad businesses and good ones alike, which is what crashes do. But look at what was left standing and still growing by 2003. Amazon had survived, barely, by shifting toward selling other companies’ products and building infrastructure. Google had launched in 1998 and was already the best search engine anyone had used, funded conservatively and growing on actual usage rather than hype. eBay was profitable and growing. PayPal had been acquired by eBay in 2002. The internet itself was bigger in 2003 than it had been in 2000. More people online, more pages, more commerce, just less insanity around valuations.

The companies that survived the dot-com crash were mostly the ones solving real problems at defensible scale. That’s not a complicated lesson, but it took $5 trillion in market losses to make it obvious.

Broadband, Social Media, and the Internet Becoming Infrastructure

The dial-up era had a specific texture. You connected when you needed to and disconnected when you were done. You heard the modem negotiate a connection. You could not use the phone and the internet at the same time. The internet was a place you visited.

Broadband Changes the Whole Relationship

Broadband penetration in the US went from 9% of households in 2000 to 42% in 2004 to over 60% by 2007. When the connection is always on, behavior changes completely. You stop thinking about connecting and start thinking about what to do. Music, video, photos, communication, shopping: all of it moved online faster once people didn’t have to wait for a connection or worry about tying up the phone line.

Napster, which launched in 1999, already demonstrated that people would happily download music if the connection was fast enough. Napster hit 80 million users before being shut down by lawsuit in 2001. The music industry won the legal battle and lost the war. iTunes launched in 2003, but the underlying behavior Napster had established, music as digital files rather than physical objects, was permanent.

YouTube launched in 2005, was acquired by Google in 2006 for $1.65 billion, and by 2007 was serving 100 million video views per day. Video on the internet only worked because broadband was now common enough that streaming didn’t mean waiting 20 minutes for a 3-minute clip to buffer.

Social Media: The Internet Becomes About People

Friendster launched in 2002 and quickly attracted 3 million users before collapsing under the weight of its own servers. MySpace launched in 2003 and by 2006 was the most visited website in the United States, with 100 million accounts. Facebook launched in February 2004, initially only for Harvard students, expanded to other universities, then to anyone with an email address by September 2006. By 2008 Facebook had 100 million users. By 2012 it had 1 billion.

Twitter launched in 2006. The first tweet was Jack Dorsey on March 21, 2006 writing “just setting up my twttr.” LinkedIn had launched in 2003 and was quietly building the professional network that would eventually be acquired by Microsoft for $26.2 billion in 2016.

Social media didn’t just add features to the internet. It changed what the internet was for most people. Instead of browsing websites or searching for information, people spent time on platforms, publishing, reading, arguing, and connecting with each other. The web became less about documents and more about people, which was either the best thing or the worst thing, depending on who you ask and on which day.

The iPhone and the Mobile Internet

June 29, 2007. Steve Jobs held up an iPhone on a stage in San Francisco and demonstrated what a smartphone could do. Three things: an iPod, a phone, and an internet communicator. The crowd laughed and cheered. It’s impossible to overstate how different that device was from what people were carrying before it.

By 2010 there were 46 million iPhones in use. Android launched in 2008 and by 2013 was the dominant mobile operating system globally. By 2014, mobile internet usage exceeded desktop usage for the first time. The internet had moved from computers to pockets, and that changed everything about how people accessed information, communicated, shopped, and spent their time.

App stores meant that software distribution no longer required a web browser. Instagram launched in 2010 as a mobile-only app and hit 1 million users in three months. The mobile internet was, in many ways, a different internet than the desktop version, faster to access, more visual, more social, and more personal.

The Modern Internet: Cloud, AI, and Centralization Nobody Voted For

The internet of 2026 is technically the same internet as 1991. Same TCP/IP, same HTML, same HTTP. But the way it’s organized and who controls it has changed substantially from what the people who built it expected or intended.

Cloud Computing and the Infrastructure Shift

Amazon Web Services launched in 2006 and quietly became one of the most important businesses in the history of technology. AWS let any company, from a two-person startup to a Fortune 500 corporation, rent computing infrastructure by the hour instead of buying and managing their own servers. Netflix, Airbnb, Slack, Pinterest, and thousands of other companies you use daily run on AWS. Microsoft Azure launched in 2010. Google Cloud followed.

The cloud moved the physical infrastructure of the internet into a small number of large data centers owned by three companies. That’s a fascinating consolidation that happened mostly without public debate. When AWS had an outage in December 2021 affecting their US-East-1 region, it knocked out Tinder, Disney+, Amazon’s own shopping site, and thousands of other services simultaneously. That’s what concentration looks like in practice.

The Platform Problem: Centralization in Plain Sight

The early internet was genuinely decentralized. Websites were independent, email ran across dozens of providers, and no single company controlled access to more than a small piece of the web. The internet’s founders, including Berners-Lee, designed it this way on purpose.

What happened instead is that network effects concentrated attention and data in a small number of platforms. Google handles over 90% of global web searches. Facebook and Instagram together account for a significant portion of time spent on the mobile internet. Amazon controls roughly 40% of US e-commerce. YouTube is where most people go for video. These platforms aren’t the internet, but for most users they’re indistinguishable from it.

Tim Berners-Lee has said publicly that the web has evolved in ways he didn’t intend and doesn’t like. He’s been working since 2018 on a project called Solid, which is trying to give users control over their own data rather than having it stored in silos owned by platforms. Whether that effort succeeds or not, the observation is accurate: the internet started as a distributed system and gradually became centralized in ways that would have seemed impossible to its original architects.

AI and the Next Shift

ChatGPT launched in November 2022. It hit 1 million users in 5 days. By January 2023 it had 100 million users, the fastest consumer app to reach that number in history. The rapid adoption of AI tools is the latest disruption in a long line of disruptions in the history of the Internet.

AI is changing how people search for information, which is a direct challenge to Google’s core business model. Instead of searching for a list of pages and clicking through them, people are increasingly asking AI assistants questions and getting direct answers. The implications for content publishers, search engines, and the broader web economy are still being worked out. What’s clear is that this shift is real and happening fast.

Conclusion

The history of the Internet is really a story about what happens when you give smart people a problem, let them build something nobody fully understands yet, and then release it to a few billion humans who use it in ways nobody predicted. Every decade since 1969 has added something that changed what the internet was and what it was for. Packet switching, then TCP/IP, then the web, then email at scale, then broadband, then social media, then smartphones, then cloud computing, then AI.

None of it was inevitable. All of it was built by specific people making specific decisions, some brilliant and some terrible, under conditions they couldn’t fully see. That’s worth remembering the next time the internet feels like it’s always been there and always will be exactly as it is now.

Frequently Asked Questions

When was the internet invented?

The internet doesn’t have a single invention date because it was built in stages over decades. ARPANET, the network that became the internet, first connected four universities on October 29, 1969. TCP/IP, the protocol suite that made different networks interconnect, was adopted on January 1, 1983, which some consider the internet’s real birthday. The World Wide Web, which is what most people mean when they say “the internet,” went public on August 6, 1991.

Who invented the internet?

No single person invented the internet. The key contributors include J.C.R. Licklider, who first described the concept of interconnected computing in 1963; Vint Cerf and Bob Kahn, who designed TCP/IP in 1974; and Tim Berners-Lee, who invented the World Wide Web in 1989-1991. ARPANET, the foundation of the internet, was built by teams at universities and at Bolt Beranek and Newman, a Boston technology firm contracted by ARPA.

What was ARPANET?

ARPANET was the Advanced Research Projects Agency Network, a computer network funded by the US Department of Defense and launched in 1969. It initially connected four universities: UCLA, Stanford Research Institute, UC Santa Barbara, and the University of Utah. ARPANET was the first network to use packet switching at scale and served as the technical and organizational foundation for what became the public internet. It was formally decommissioned in 1990.

What is the difference between the internet and the World Wide Web?

The internet is the global infrastructure of interconnected networks, including the physical cables, routers, and protocols that move data between computers. The World Wide Web is one application that runs on top of the internet, using HTTP and HTML to serve linked documents through browsers. Email, FTP, and online gaming also run on the internet but are not the web. Most people use “internet” and “web” interchangeably, but they refer to different things technically.

What caused the dot-com crash?

The dot-com crash was caused by a combination of overinvestment in internet companies with unsustainable business models, a stock market that rewarded user growth over profitability, and a general correction that was probably inevitable given the valuation multiples that had accumulated. The Nasdaq peaked on March 10, 2000 and fell 78% by October 2002. Companies that were burning cash without a path to profit collapsed, including high-profile failures like Pets.com, Webvan, and Kozmo.com. Companies that were solving real problems with sustainable economics mostly survived.

When did most people start using the internet?

Mainstream internet adoption started accelerating in the mid-1990s after the World Wide Web and graphical browsers like Mosaic and Netscape Navigator made it accessible to non-technical users. In 1995, about 16 million people worldwide were online, representing less than 0.5% of the global population. By 2000 that number had grown to 361 million. Broadband adoption in the early 2000s increased usage time significantly, and the launch of smartphones between 2007 and 2010 brought mobile internet access to a much larger share of the global population.

Who controls the internet today?

The internet has no single owner or controller. Physical infrastructure is owned by a mix of telecom companies, governments, and private corporations. Domain names and IP addresses are managed by ICANN, a nonprofit organization. The content people access online is increasingly concentrated on a small number of platforms: Google dominates search, Meta dominates social networking in many markets, Amazon dominates e-commerce and cloud infrastructure. Governments increasingly regulate what their citizens can access, with countries like China operating a heavily filtered version of the internet behind what is commonly called the Great Firewall.

What was the first website?

The first website was created by Tim Berners-Lee at CERN and went live on August 6, 1991. The URL was info.cern.ch and the page described what the World Wide Web was and how to use it. CERN restored the original page in 2013, and it remains accessible at that URL today. The content is simple text with links, which is both appropriate for 1991 and somewhat humbling given what the web became.

How fast has the internet grown?

The growth has been extraordinary by any measure. ARPANET had 4 nodes in 1969 and 15 by 1971. The web had roughly 130 websites in 1993 and over 1 million by 1997. There are now estimated to be over 1.1 billion websites, though only a fraction are actively maintained. Internet users grew from 16 million in 1995 to 5.4 billion in 2023, representing about 67% of the world’s population. Mobile internet users now account for the majority of global internet traffic.

What comes after the internet as we know it?

That depends on who you ask and which technology you’re tracking. Decentralized web projects like Web3 and Tim Berners-Lee’s Solid project are trying to redistribute control away from large platforms and back toward individual users. AI is changing how information is accessed and processed, with implications for search, publishing, and how content is created at scale. Faster connectivity through 5G and eventually 6G networks is enabling applications that weren’t practical on earlier mobile infrastructure. The infrastructure will evolve, but the internet’s basic architecture has proven durable enough that wholesale replacement seems unlikely. Layers get added. The foundation stays.