Every day we type addresses that start with three letters almost nobody thinks about: www. They have become as ordinary as the light switch on the wall.
And yet none of it existed a little over 35 years ago – and the idea behind it was born not in a tech company, but in a physics lab.
A problem anyone who has ever worked in an office knows
The story begins at CERN, the large particle physics laboratory near Geneva. In the late 1980s a British computer scientist named Tim Berners-Lee worked there. The problem on his mind was surprisingly mundane and deeply human: thousands of scientists did research at CERN, they came and went constantly, and they left behind countless documents on all kinds of different computers. Nobody could find anything anymore.
Anyone looking for a particular piece of information had to know which machine it was on, how to operate that machine and what format the document was in. A note here, a report there, a contact list on a third system – all separated from each other, all locked away on their own little islands.
Berners-Lee had a simple idea: what if you no longer kept information on islands, but connected it through references instead? What if one document could point to another – with a single click, no matter which computer the other document was on? In 1989 he wrote a proposal for it. His supervisor scribbled in the margin: “Vague, but exciting.”
The most important sentence in this article
Before we go on, we need to clear up a mix-up that almost everyone falls for – and that changes how you understand the whole story:
The World Wide Web is not the internet. And Tim Berners-Lee did not invent the internet.
That sounds surprising, but it is central. The internet – the worldwide networking of computers that can exchange data with one another – had existed for a long time already. Its roots go all the way back to 1969, to the so-called ARPANET, a research network in the United States. Through the 1970s and 1980s it gradually grew into the internet as we still know it at its core today. So when Berners-Lee wrote his proposal in 1989, that network was already two decades old.
What he invented was something else: a method for linking documents together over this existing internet. The World Wide Web is not the network itself, but what runs on top of it.
An image makes it clear: the internet is the road network – the cables, the lines, the connections between computers. The World Wide Web is the traffic of linked pages traveling on those roads. So the roads were already there. Berners-Lee invented the vehicles, the traffic rules and the signposts that made traveling on them usable for everyone in the first place. Email, for example, was already driving on those same roads – the web came along as one more, and especially successful, kind of traffic.
Three building blocks everyone uses today – without knowing it
For this web to work, three things were needed. Berners-Lee designed all three of them between 1989 and 1991. And the remarkable thing is: to this day they are the foundation the entire web stands on.
First: an address for every document. Every page on the web needs a unique address so it can be found again. That is the web address you see at the top of your browser in the address bar. Experts call it a URL – but at heart it is nothing more than the mailing address of a document.
Second: a language for laying out and linking pages. For a computer to know what a heading is, what a paragraph is and – most importantly – where a link to another page sits, a shared language was needed. That language is called HTML. Every website you visit today is written in it.
Third: a method for computers to exchange these pages. When your computer requests a page and the distant computer sends it, both have to use the same way of communicating. That is HTTP – the “http” that used to sit at the start of every web address and that many older readers still laboriously typed out themselves. Today the browser usually adds it on its own, which is why it hardly stands out anymore.
Address, language, exchange method: you are using these three building blocks this very second, while you read this text. Most people have never heard of them – and that is perhaps the best sign of just how good the invention was.
A computer with a handwritten note on it
The early years of the web have something touchingly modest about them. The very first web server in the world was not a vast data center, but a single computer on Berners-Lee’s desk – a so-called NeXT, built by the company Steve Jobs had founded after being pushed out of Apple.
Stuck to that machine was a handwritten note that read: “This machine is a server, DO NOT POWER IT DOWN.” Because if someone had accidentally cut the power, a good portion of the web of the day would simply have vanished.
The very first website went online on August 6, 1991. It was as unspectacular as you can possibly imagine: a plain page explaining what the World Wide Web even was and how you could create pages of your own. No picture, no color, no frills – just text and links. That page still exists today in reconstructed form and is still available online, at its original address at CERN.
The moment the web became interesting for ordinary people came in 1993. That year saw the arrival of Mosaic, the first widely used graphical browser. It could display images and text together on one page – something that seems obvious to us today, but back then felt like a small revelation. Suddenly the web was no longer just something for physicists and computer experts, but something you actually wanted to look at.
The decision that changed everything
And now to the second point that surprises most people – and that is perhaps the most important of the whole story.
The web was given away.
On April 30, 1993, CERN made the web technology irrevocably and free of charge available to everyone. No patent, no license fees, no conditions. Anyone who wanted to run a web server or build a browser was free to do so – without asking anyone for permission and without paying a cent.
That was by no means a given. In the early 1990s there were definitely competing systems for making information accessible over the internet. Some of them were technically more polished than the web. And some charged fees – which scared off many potential users.
Had CERN or Berners-Lee charged license fees for the web, history would probably have unfolded completely differently. The web would have remained one of several competing, mutually incompatible systems – and possibly not even the most successful one. The very reason the whole world settled on it was that anyone could use it, develop it further and build on it for free and without permission. The openness was not a nice side effect. It was the precondition for everything that came after.
From a document problem to the whole world
It is worth pausing for a moment to take in the scale of it. A solution to an internal filing problem at a physics lab – “nobody here can find their documents” – became the infrastructure on which a large part of life now takes place. Shopping, banking, news, dealing with government offices, reconnecting with old friends, looking up a phone number: all of it runs on those three building blocks designed in Geneva between 1989 and 1991.
What is also remarkable is what did not happen. Tim Berners-Lee did not become a billionaire through his invention. He probably could have, had he turned the web into money. Instead he is a university professor today and has spent decades championing an open, free web – the very idea that made the web big in the first place.
The web was never the finished product of a company. It was an idea that was given away, because its inventor understood that it would only be worth something if it belonged to everyone.
Do you still remember?
Many of our readers lived through the early days of the web in the 1990s themselves – and this is exactly where the story gets personal. Do you still remember the first website you ever visited? When did you first go “online” – and how did it feel? That scratchy, screeching sound of the dial-up modem is something many of us still hear in our heads.
What did the web look like for you back then, and what surprised you most when you first saw it? We’d love to hear about it in the comments – we’re curious about your memories.
Comments
It's called Lynx, and has no, repeat no, graphical capability whatsoever - it is text and text only.
And given that I was using a slow dial-up connection, that was an advantage. Text only was slow enough already, downloading images would have taken forever.
I think it's still available. If you want a feel for the earliest days of the Web, give it a try.
And perhaps also try Links.
Hypercard organized information in "stacks" of "cards" that could contain text, graphics/pictures and "hyperlinks" that would do something on the same card (like play a video or animation), link to another card in the stack or even open another stack. I think in later versions it could even open stacks on another computer on the network. Whole games were originally invented as Hypercard stacks (I remember Uninvited and Myst) and only later got their own engine. Though Hypercard was free to use at first with every Mac, it was still proprietary and only ran on Macs (or maybe there was a NeXT version?).
I also remember having known Gopher (text pages hyperlinked over multiple servers over the internet) before the WWW. But I just learned that it was invented later, but became popular before the WWW was widespread. I used Gopher mostly to search through catalogs of libraries all over the country. And although Gopher was developed at the University of Minnesota, even running a server costs licensing fees, at least at later versions.
My first modem was a 2400er, which I used to connect to my university's mainframe (IBM360) and to various BBS, with a Commodore 128 as a terminal (80 characters!). I then used E-Mail, ICQ (chat) and telnet a lot these days. I bought my first Mac when they became affordable with the LC in 1990.
Saw my first web page with an NCSA Mosatic on a Sun SPARC station at the university. I still remember the office and the friend who showed it to me, but not the page - probably an internal one of the university to start with. AltaVista was just a few clicks later.
At home I soldered a cable to connect my then new 14.4k modem to the serial port of the Mac. The university had only
Bravo for your articles
Please keep that in mind.
I can remember surfing through the bowels of CompuServe (formerly 72447, 3424, starting in 1983, and yes, I still remember my first password) to find information from NIH when my sister became ill in 1986. It's a lot easier today.
In 1984, when the radio station I was working for did a remote broadcast from Zihuatanejo, Mexico, we had to send information over a 1200-baud modem via SmartCom's terminal mode. It was interesting to watch what the static from the phone lines added to the screen.
During the pandemic, I anchored nine of hours of news each night from my living room via VOIP -- listeners could not tell that I wasn't in the newsroom until the neighbor's dog decided to bark while my mic was open.
Thank you, Tim Berners-Lee!
I think that html has many similarities to gml and sgml ... that ugly markup language we used to write reports on the mainframe. Anyone remember using dot prompts to format documents? .pa for paragraph etc. etc.
Thanks for the walk down memory lane.
Thanks for the memories.
The shock of using a browser - Google? 1998? - to obtain data from a domain in Australia blew me away. The world's information at my fingertips!
I'm still awed by what browsers can provide.