Today, it is easy to use the words Internet and Web as if they mean the same thing.

They do not.

The Internet is the network. The World Wide Web is one system that runs on top of that network.

And the Web was not the first system for finding and sharing information across the Internet.

Before the Web, people were already sending email, transferring files, reading discussion groups, connecting to remote computers and browsing information systems. There was ARPANET. There was Usenet. There was FTP. There was Gopher. There was WAIS. There were several early hypertext and browser projects.

The World Wide Web did not appear in an empty room.

It appeared in a crowded room and happened to be the idea that eventually connected almost everything.

Before the Web: the Internet already existed

The story actually begins decades before Tim Berners-Lee wrote his proposal.

In the 1960s, researchers were already thinking about networks that could allow computers and people to share information remotely. Those ideas eventually contributed to ARPANET, the network created through the US Advanced Research Projects Agency.

The first four ARPANET nodes went online in 1969. Over the following decades, researchers developed networking technologies and protocols that eventually made the modern Internet possible.

By the 1980s, TCP/IP was becoming the common language that allowed different networks to communicate.

In 1986, the National Science Foundation launched NSFNET, connecting academic researchers to supercomputing centers in the United States. NSFNET would later become one of the major pieces of infrastructure that allowed the Web to spread rapidly in North America.

The US National Science Foundation's history of the Internet describes ARPANET, TCP/IP and NSFNET as important steps toward the commercial Internet we know today.

But none of this was yet the Web.

So what did people actually do on the Internet?

Quite a lot.

They sent email.

They transferred files with FTP.

They logged into remote machines with Telnet.

They participated in Usenet newsgroups.

They chatted on IRC.

They searched specialized databases.

They browsed information through systems such as Gopher.

In other words, the Internet already had applications for communication and information retrieval.

What it lacked was a simple, general-purpose system for connecting documents to one another across different machines.

Hypertext existed before the Web

This part of the story is often forgotten.

Tim Berners-Lee did not invent the idea of linking pieces of information together.

The idea of hypertext is much older.

In 1945, Vannevar Bush described a hypothetical information machine called the Memex in his essay As We May Think. Bush imagined a system where people could associate and navigate between pieces of information rather than reading everything in a fixed linear order.

Later, Ted Nelson developed the concepts of hypertext and hypermedia, and Douglas Engelbart demonstrated systems that allowed people to interact with linked information.

CERN's own WorldWideWeb timeline places developments such as Nelson's hypertext work, Engelbart's NLS, ARPANET, Usenet, HyperCard and SGML alongside the technologies that eventually led to the Web.

The Web was therefore not invented from nothing.

It was a particularly effective combination of several ideas that already existed.

1989: Tim Berners-Lee has a problem

In 1989, Tim Berners-Lee was working at CERN, the European particle physics laboratory in Switzerland.

CERN was an unusual place. Thousands of scientists from different countries worked there, but they often belonged to different institutions and used different computer systems.

Berners-Lee saw a problem with how information was organized.

People had lots of information, but it was spread across different computers and systems. There was no universal way of connecting it.

In March 1989, he wrote a proposal titled Information Management: A Proposal.

His manager, Mike Sendall, reportedly described the proposal as "vague, but exciting."

It was a remarkably understated description of what was about to happen.

Berners-Lee's idea was to combine three things:

• the Internet's existing network infrastructure,
• hypertext, and
• a simple system for publishing and retrieving documents.

That combination became the World Wide Web.

1990: The Web is built

Berners-Lee continued developing the idea with Belgian engineer Robert Cailliau.

By the end of 1990, he had built the first web server and the first browser.

The computer was a NeXT workstation sitting in his office at CERN.

It even had a handwritten warning attached to it:

This machine is a server. DO NOT POWER IT DOWN!!

The first browser was called WorldWideWeb. It was not merely a browser. It was also an editor.

This is one of the more interesting details about the original Web.

Berners-Lee did not initially imagine a system where people only consumed information. The first browser could also create and edit Web pages.

The Web was intended to be read and written.

CERN's reconstruction of the original browser still demonstrates this concept.

The three technologies that made the Web

The Web is often reduced to HTML, but there were really several pieces involved.

HTML

HyperText Markup Language described the structure of documents and allowed them to contain links to other documents.

HTTP

HyperText Transfer Protocol provided the mechanism for requesting and transferring resources between clients and servers.

URLs

Uniform Resource Locators gave those resources addresses.

Put simply:

HTML described the document, HTTP moved it, and URLs told you where to find it.

That combination was extraordinarily powerful because it was simple.

The original HTTP protocol was tiny compared with modern Web infrastructure. The IETF's HTTP/1.1 specification records that HTTP had been used by the World Wide Web since 1990.

1991: The Web escapes CERN

The first version was not particularly useful if only one person could use it.

The original browser ran on a NeXT computer, and very few people had one.

So CERN developed another browser.

The Line Mode Browser was deliberately simple and could run on a much wider range of systems.

In August 1991, Berners-Lee announced the WWW software on Internet newsgroups.

The first Web server outside Europe appeared at SLAC in California in December 1991.

The Web had officially escaped CERN.

1992-1993: The Web meets its competitors

This is where the history gets much more interesting.

The Web was not immediately the obvious winner.

Several other systems were already available, and some were considerably more popular.

Gopher: the Web's most important early competitor

Gopher was developed at the University of Minnesota in 1991.

It was a system for navigating Internet information through hierarchical menus.

Imagine opening a browser and seeing a menu like this:

1. About this server
2. News
3. Documents
4. Software
5. Search
6. Other servers

You selected an item and moved deeper into another menu.

It was simple, fast and well suited to the relatively slow connections of the early 1990s.

Gopher became extremely popular. The University of Minnesota's history of the protocol describes it as one of the first widely used ways of navigating Internet information.

The problem was that Gopher's structure was relatively rigid.

The Web could link almost anything to almost anything else. A document could contain a link in the middle of a paragraph, which could lead to another document on another server.

Gopher was more like navigating a filesystem.

The Web was more like creating a giant network of documents.

For a while, that distinction did not matter much.

Then Gopher made a crucial mistake

In February 1993, the University of Minnesota announced a licensing policy that reserved the right to charge for commercial use of its Gopher server software.

That announcement caused substantial concern among Internet users and administrators.

At almost exactly the same time, CERN was moving in the opposite direction.

On April 30, 1993, CERN released the World Wide Web software into the public domain.

That difference mattered enormously.

Gopher had a large installed base, but the Web was open and could be freely implemented by anyone.

The W3C's history of the Web identifies the Gopher licensing announcement and CERN's release of the Web software as two major events preceding the Web's rapid growth.

There was also a technical advantage.

Gopher's menu structure was excellent for organizing information, but HTML was much more flexible. As browsers became graphical and Internet connections became faster, the Web could incorporate images, typography and increasingly complex layouts.

Gopher had been designed around the limitations of its time.

Those limitations were disappearing.

WAIS

Another important system was WAIS, or Wide Area Information Servers.

WAIS was designed primarily for searching and retrieving information across distributed databases.

Instead of browsing a hierarchy of menus or following links between documents, you could search for information across servers.

In the early 1990s, WAIS was a serious technology. An IETF document from 1994 explicitly compared WAIS, Gopher and the World Wide Web as competing approaches to networked multimedia information.

Read the contemporary RFC 1614: Network Access to Multimedia Information.

WAIS did not disappear because it was useless.

It simply occupied a narrower role, while the Web combined documents, links, media and navigation into one much more general system.

Other early Web browsers that almost nobody remembers

The browser history is also much more crowded than the usual "WorldWideWeb → Mosaic → Netscape → Internet Explorer" story suggests.

CERN lists several early browsers, including:

• ViolaWWW, created by Pei-Yuan Wei.
• Erwise, developed by students at Helsinki University of Technology.
• MidasWWW, developed at SLAC.
• Lynx, a text-based browser that survived long after many of its early competitors disappeared.

ViolaWWW is particularly interesting because it was more ambitious than a simple document viewer. It supported things such as embedded scripts and interactive elements before JavaScript became the standard way to do this.

Erwise was also significant. It was developed by Finnish students and is often described as the first graphical browser for Unix.

Most of these projects disappeared.

Mosaic did not.

1993: Mosaic changes the game

NCSA Mosaic was released in 1993.

It was not the first Web browser and it was not even the first graphical Web browser.

What it did was make the Web approachable.

Mosaic supported graphical interaction and was released for multiple popular platforms, including Windows and Macintosh.

Suddenly, you did not have to be a Unix researcher to understand the Web.

Mosaic was important not simply because it looked nicer. It ran on popular personal computers and helped bring the Web outside research institutions.

Mosaic also supported Gopher, which is an interesting detail. In a sense, Mosaic helped the Web win partly by incorporating some of the functionality of its competitors.

CERN's history records that the Macintosh and Windows versions released in late 1993 had an immediate impact on the spread of the Web.

By the end of 1993, CERN records more than 500 known Web servers.

By the end of 1994, there were around 10,000 Web servers and an estimated 10 million users.

1994 was already being called the "Year of the Web."

1994: Netscape arrives

Mosaic's developers did not stop there.

Marc Andreessen, one of the people behind Mosaic, left NCSA and co-founded Mosaic Communications with Jim Clark.

The company eventually became Netscape Communications, and its browser became Netscape Navigator.

Netscape helped transform the Web from an academic technology into a commercial platform.

Companies started building websites.

Online stores appeared.

Advertising arrived.

Search engines became important.

Web design became an industry.

And the browser itself became a commercial battleground.

The Web wins, but the Internet does not disappear

By the middle of the 1990s, the World Wide Web was clearly becoming the dominant way ordinary users interacted with Internet information.

But this did not mean that everything else disappeared.

Email remained separate.

FTP remained useful.

IRC remained popular.

Usenet survived for decades.

Gopher never completely disappeared either.

The important change was that the Web became the general-purpose interface through which an enormous amount of Internet content could be accessed.

Why did the Web win?

There was no single reason.

It was a combination of several advantages arriving at almost exactly the right moment.

It was open

CERN's decision to release the Web software without royalties was crucial.

Anyone could implement a Web server or browser.

It was decentralized

There was no central database containing every webpage.

Anyone could run a server and connect it to the larger system.

It was simple

HTML was comparatively easy to understand. HTTP was simple. URLs gave resources recognizable addresses.

It was extensible

Images could be added. Then forms. Then CSS. Then JavaScript. Then video. Then applications.

The Web did not need to be redesigned every time computers became more capable.

It was easy to link

This may have been the most important property of all.

One document could point directly to another document somewhere else in the world.

That sounds normal today because we have lived with hyperlinks for decades.

At the time, it was revolutionary.

The Web becomes the Internet's interface

During the second half of the 1990s, the Web expanded rapidly.

Search engines appeared because there were now too many websites to navigate manually.

Amazon launched in 1994.

Yahoo! became a major Web directory.

Google arrived in 1998.

Wikipedia appeared in 2001.

YouTube arrived in 2005.

Facebook launched in 2004.

The Web stopped being a collection of documents and became a platform for almost everything.

But the story did not end there

The Web became dominant, but people never stopped trying to build alternatives.

Some alternatives tried to fix centralization.

Some tried to improve privacy.

Some tried to simplify publishing.

Some tried to make the Internet more decentralized.

And some simply tried to recreate the older, quieter Internet.

Gemini: a smaller Web

One of the more interesting modern examples is Gemini.

Gemini is a lightweight Internet protocol created around the idea of a deliberately limited alternative to the modern Web.

Instead of supporting everything a modern browser can do, Gemini keeps the protocol and document format deliberately small.

A Gemini document is normally plain text with simple links and lightweight formatting.

The result is much closer to the text-oriented Internet of the early 1990s than to today's JavaScript-heavy websites.

Gemini uses its own URI scheme, gemini://, and normally operates on port 1965.

The protocol is documented in its official Gemini documentation.

Gemini is not trying to replace every website.

It is more like an alternative publishing space for people who prefer a smaller, quieter protocol.

Gopher is still here

And surprisingly, Gopher never completely died.

There are still Gopher servers and modern Gopher clients.

That makes Gopher an unusual piece of Internet history: a technology that lost the mainstream competition but never completely disappeared.

Its simplicity is still attractive to a small community of users.

The original Gopher protocol is documented in RFC 1436.

IPFS and the decentralized Web

Another modern direction is very different.

Instead of making the Web smaller, projects such as IPFS try to change how information is addressed and distributed.

Traditional Web addresses generally tell your browser where something is located.

IPFS uses content addressing. Instead of primarily asking "where is this file?", the system can identify content by what the content actually is.

IPFS describes itself as a set of open protocols for addressing, routing and transferring data using content addressing and peer-to-peer networking.

The official IPFS documentation explains the architecture and what IPFS is and is not.

The modern Internet is not one thing

This is perhaps the most important distinction.

When people say "the Internet" today, they are often really talking about several different layers at once.

There is the underlying network.

There are protocols such as TCP/IP.

There is DNS.

There is HTTP.

There is the World Wide Web.

There are email protocols.

There are messaging systems.

There are peer-to-peer networks.

There are decentralized systems.

There are protocols such as Gemini and Gopher.

And there are countless applications built on top of all of them.

The Web won the competition for general-purpose hyperlinked information, but it did not replace the Internet itself.

What the Web actually invented

The Web's biggest innovation was not the Internet.

It was not hypertext either.

It was the particular combination of:

• globally addressable documents,
• hyperlinks between those documents,
• a simple client-server protocol,
• a simple markup language,
• and an open, decentralized publishing model.

That combination turned a network of computers into a network of information.

And that distinction matters.

ARPANET connected computers.

The Internet connected networks.

The Web connected information.

That is why the Web became so much bigger than its original purpose at CERN.

From CERN to almost everything

In 1989, Berners-Lee was trying to solve an information-management problem at a particle physics laboratory.

In 1991, the first Web server was still sitting in his office.

In 1993, CERN released the technology openly.

In 1994, the Web was spreading beyond research institutions.

By the late 1990s, companies were building businesses around it.

By the 2000s, it was becoming a platform.

By the 2010s, much of the software people used every day had moved into the browser.

And today, in 2026, the Web is so deeply embedded in computing that it is easy to forget how unusual the original idea actually was.

The Web was not inevitable.

Gopher existed.

WAIS existed.

Other browsers existed.

Other hypertext systems existed.

There were many possible futures.

The World Wide Web happened to combine openness, simplicity, hyperlinks and graphical browsers at exactly the right moment.

And once enough people started linking to each other, the network effect became almost impossible to stop.

That may be the real origin story of the Web.

Not one invention.

Not one browser.

Not even one person.

It was a collection of ideas that finally found a simple enough way to connect.

Sources and further reading

• CERN: The birth of the Web
• CERN: A short history of the Web
• CERN: WorldWideWeb Browser History
• CERN: WorldWideWeb Timeline
• W3C: The history of the Web
• US National Science Foundation: Birth of the Commercial Internet
• IETF RFC 1436: The Internet Gopher Protocol
• IETF RFC 1614: Network Access to Multimedia Information
• IETF RFC 2616: HTTP/1.1
• IPFS: What is IPFS?
• Gemini Protocol