The Early Internet

The Internet is Not the Web

Before we go any further: the internet and the web are completely different things. This is a distinction almost everyone gets wrong, and it matters for understanding why things cost what they do.

The internet is the infrastructure—the pipes, protocols, and wiring that lets computers talk to each other globally. It's been around since 1969. The web is an application layer that runs on top of the internet. It didn't exist until 1989, and it's just one of many things the internet can do.

You can have email, file transfers, video streaming, and entire corporate intranets on the internet without any web browser ever entering the picture. The confusion between these two terms is foundational to misunderstanding web history and cost.

ARPANET: The Military-Academic Birth

On September 2, 1969, four computers became the first nodes of ARPANET, funded by the U.S. Department of Defense's Advanced Research Projects Agency (ARPA). The nodes were at UCLA, Stanford, UC Santa Barbara, and the University of Utah. The goal was audaciously simple: create a network that could survive partial destruction and still function.

This was the height of Cold War paranoia. The thinking went: if a centralized network like AT&T's telephone system got bombed, you're done. But a distributed network with redundant paths could keep running even with nodes knocked offline. This paranoid requirement accidentally created something beautiful: a network that was inherently resilient and decentralized.

The early ARPANETers weren't thinking about commerce or advertising. They were researchers, physicists, and computer scientists who wanted to share data and collaborate. This ethos—open, non-commercial, peer-to-peer—would shape the internet's DNA for decades.

Email and the TCP/IP Revolution

In 1971, Ray Tomlinson sent the first network email. It was unremarkable by today's standards—just a test message—but it revealed something profound: the internet's most killer app wasn't file transfer or resource sharing. It was asynchronous communication. Email would become the reason millions of people got on the internet, and it remains foundational today.

By the early 1980s, ARPANET was facing a crisis. It was sprawling into multiple networks with different protocols, and they didn't all talk to each other cleanly. Vint Cerf and Bob Kahn designed TCP/IP to solve this: a universal protocol that any network could adopt, allowing them all to communicate. On January 1, 1983, ARPANET switched to TCP/IP. This moment is considered the official birth of the modern internet.

What made TCP/IP genius was its simplicity and universality. It wasn't a closed, proprietary system. It was published openly, free to implement on any hardware. This openness would prove crucial to the internet's dominance over competing network technologies like OSI or IBM's SNA.

DNS: Making Addresses Human-Readable

In the early days, you accessed computers on the network by their IP address, like 192.168.1.1. But with thousands of machines coming online, remembering IP addresses became impossible. The first solution was a simple text file called HOSTS.TXT, maintained by SRI and distributed via FTP. Every computer downloaded a fresh copy regularly.

By 1984, Paul Mockapetris invented the Domain Name System (DNS). Instead of one centralized list, DNS is a distributed database of domain names and their corresponding IP addresses. When you type "example.com" into a browser, your computer queries a DNS server to convert that human-readable name into an IP address. The system is hierarchical, redundant, and infinitely scalable.

DNS is so fundamental that we barely notice it. But without it, the internet would require you to remember IP addresses for every site you visit. The web would never have become mainstream. This single invention—elegant, distributed, resilient—is one of the internet's greatest engineering achievements and a perfect example of why the internet era cost so much less to deploy than the proprietary network era before it.

The Pre-Web Internet: Gopher, FTP, Usenet, and BBS

From 1983 through 1989, the internet grew—but not visibly to the general public. Academics, researchers, and early hobbyists were online, but the tools were arcane and the content was niche.

FTP (File Transfer Protocol) was the way to move files across the network. You'd use command-line FTP to download software, papers, or data. It still works today, though it's been largely replaced by HTTP and cloud storage.

Usenet was a distributed system for discussing topics in hierarchical newsgroups. Before forums, before Reddit, before social media, Usenet was how the internet discussed things. It was chaotic, sometimes brilliant, often terrible, and completely decentralized. Each server would exchange messages with its peers, creating a global conversation without any central authority.

Gopher (1991) was a hierarchical document retrieval system. You'd navigate through menu structures of text and links. It was more user-friendly than FTP but clunkier than what would come next. At its peak, thousands of Gopher servers existed worldwide. After the web won, most Gopher servers vanished completely, though a few hardy souls still maintain Gopher sites.

BBS (Bulletin Board Systems) were dial-up boards where people connected via modem to chat, play games, and share files. BBSes predated the internet but thrived on ARPANET and other networks in the 80s. Running a BBS meant hosting a computer at home, paying phone bills, and managing a community of users. The aesthetic of BBS culture—ANSI art, ASCII graphics, modem speed bragging—became part of hacker folklore.

Early Internet Timeline

The Foundations Remain Unchanged

Here's what should blow your mind: the TCP/IP stack from 1983 is still the internet's foundation. The DNS system from 1984 still resolves domains exactly the way Mockapetris designed it. When you deploy a website today, you're building on top of architecture created by academics in military-funded labs in the 1960s and 70s.

This is why the internet is so cheap compared to proprietary alternatives. The basic infrastructure was built on public funding, released as open standards, and maintained communally. There's no toll booth. There's no licensing fee. There's no vendor lock-in at the protocol level. Everyone can implement TCP/IP. Everyone can run a DNS server. This openness created competition and drove costs to the floor.

Why This History Matters to Your Website Cost

You might be wondering: why does a history lesson matter to the cost of building a website? Because cost is downstream of architecture.

If the internet had remained proprietary—if, say, AT&T or IBM had won the network wars—you'd be paying per-packet fees. You'd need special hardware. You'd be dependent on a single vendor. Instead, the internet's open architecture created a commodity market. Anyone can offer internet service. Anyone can host a website. Supply and competition drive prices down.

When you pay for web hosting today, you're benefiting from 40+ years of accumulated infrastructure, standardization, and competition. The next time you look at a hosting bill, you're seeing the fruits of decisions made in 1973 by Vint Cerf in his lab.

Next: The Web Arrives

The internet had been humming along for two decades—stable, invisible, used only by academics and government researchers. Then in 1989, a physicist at CERN named Tim Berners-Lee had an idea about how to make it useful to humans. The web wasn't a breakthrough in networking. It was a breakthrough in how people could interact with the network. And it would change everything.