Hank: Do you have a smartphone?  You probably have access to the internet almost everywhere you go, and being able to access what is basically the sum total of human knowledge whenever you want, that's a new thing. Pretty cool. Last year, the United Nations actually declared uncensored internet access a human right that deserves protection, but it actually took a long time for the internet to become publicly accessible at all, even in the country where it started.

The Internet grew from four computers in the western US in 1969 to a global network of networks connecting more than 20,000 computers by the end of 1987, but it was still restricted to specific universities and corporations which used it for specific types of collaboration and research. So how did we get from 20,000 computers on guarded networks in 1987 to the fundamental human right to internet access just 30 years later. In other words, when did the regular Joes start logging on?

Some of the first tastes of the future came in the 1970s when a few companies started selling access to the networks. They were standalone networks that weren't connected to the main central network that was the internet but their users could do everything from sending e-mails to checking updated weather reports to playing games and chatting on some of the first ever instant messaging programs. And on some networks like Compuserve's Micronet, users could even read stories from major newspapers right on their home computers.  Like, imagine, a newspaper on your computer.  But as great as these networks were, they still had their limits. Some of them, like Micronet, were only available nights and weekends when businesses that normally used Compuserve's networks were closed, and they were all pretty much isolated islands, since they weren't directly connected to each other or to the broader internet, because throughout the 80s, no matter how big these private networks got, they weren't allowed on the internet. Arpanet, and the other networks that formed the basis of the internet, were government run and government funded, so the businesses and universities using them weren't allowed to have any commercial traffic on the network.

It was fine to use the internet to download some data or email your colleagues a copy of your latest paper, but you weren't supposed to advertise a new product on there, and you definitely weren't supposed to charge the general public to come online. The internet was supposed to be for research, not for making money, which is pretty weird to think about considering how many of us use the internet these days to make a living.

Around the same time, Arpanet's administrators were looking to hand off responsibility for maintaining the internet. Arpanet had long since accomplished its goals and the Department of Defense wanted to move on. The National Science Foundation's huge network, NSFNET seemed to be the best candidate. It started in 1986 and grew so quickly after connecting to Arpanet that less than a year later, it already needed major upgrades to handle all the new traffic. In 1990, NSFNET officially replaced Arpanet as the backbone of the internet and its more than half a million users, but even before Arpanet was completely out of the pictures, a couple enterprising companies were connecting regular people to the internet, though there is some debate about which company was there first and when.

NSFNET also had a policy about banning commercial traffic on the network, but in 1988, they decided to try connecting a couple of those private networks' email servers to NSFNET and a year later, users of Compuserve and an email service called MCI Mail could send the first commercial emails across the internet.

The next year, 1989, we got the first internet service providers, or ISPs. There were companies that usually didn't have any network of their own, they just connected people to a local network and to the internet. You'll hear a lot of people say that the first commercial ISP was a company called The World, which opened near Boston in December 1989, but others say the first one was down in Australia instead, and then there are people who tell you that there weren't any true internet service providers until Congress passed a law in 1992 allowing for commercial traffic on NSFNET.

The truth is, of course, that lots of different companies opened their doors or, like, their wires between 1989 and 199s. But all of the offered slightly different services. Some were just email, some were their own networks that partly connected to the internet, and some offered internet access without an online community of their own. So in this situation, different companies can all kind of claim that they were the first.

More ISPs were created in next few years and in 1995 NSFnet shut down for good and handed everything over to the ISPs. But even in the early 1990s the Internet was nothing like it is today. For one thing, of course, it was slow, so slow, like downloading a picture of Captain Picard took like half the day. Not that I know that from experience. To connect to a network, your computer placed a phone call to the network through a modem, which translated between the digital signals used in computers and the analog signals used in land line phones. Then, your computer would talk to the other computers on the network using the phone lines between them.

This way of getting online became known as dial-up and if you've ever used a dial-up internet connection you know exactly how frustratingly slow it could be. Though to be clear, when we were first doing it, it did not seem slow, it seemed like magic. And since phone lines were already crisscrossed the United States when Arpanet was invented, it just made sense to use them instead of trying to invent something completely new.

Part of why dial-up slow was that there is a limit to how quickly you can cram information down a phone line. To transmit a lot of information, you need a signal with a really high frequency, meaning that it can change very quickly. And for a signal sent down phone lines, that would mean you'd need a really high-pitched sound. But phone lines weren't designed for that sort of signal. They were designed for phone calls. And even though new parents might disagree, humans just don't make sounds that are that high pitched.

So the phone lines weren't designed to transmit signals with very high frequencies that you would need to transmit a lot of information all at once. The first thing a modem did when it connected to a network was to check the highest frequency signal that the wires could handle. And then it slowed the fire hose of ones and zeros coming out of your computer down to that speed.

There's another reason why the internet was different back when the first ISPs were popping up. Back in 1989, there wasn't a single website, because back in 1989 there was no such thing as the Web. Today, a lot of people use the terms "internet" and "web" pretty much interchangeably, but they're actually different things. The internet is came more than a decade before the web.

"Internet" is a shortened form of the word "internetworking," and it was coined in the 70s to refer to the physical cables and computers all connecting together. Today, it usually refers more specifically to all the computers running programs that let them communicate with each other across the giant internet network. The internet began with networks that were designed as a way of remotely sharing programs, films, and access to computers. Then programs like email were added over time.

But these programs tended to organize the information a bit like a tree or even like a whole forest. To get to a specific branch of a program, you had to climb all the way up the trunk and then down a whole bunch of branches. And then to get somewhere else you had to go all the way back to the trunk and over to another set of branches. When the internet was small, it never took too long to get back where you started so that you could get somewhere else.

But as the internet grew, it got harder and harder to navigate unless you knew exactly where you were going. Even smaller individual networks like the one at CERN in Europe were getting out of hand. It was a pain to find everything you needed on the network, especially if you didn't know exactly where to find the thing you were looking for, so you wasted a bunch of time navigating all those different trees.

So in 1989, a scientist named Tim Berners-Lee with some help from his colleague Robert Caillou started working on a better way to arrange all that information. Berners-Lee is often credited with inventing the web and for good reason. He had this big idea to sort of flatten the trees structure. Instead of each file being like it's own little isolated branch, any document or file could direct people to other related files, so that you could easily go from one to the next. 

And he knew exactly the perfect tool for the task: hypertext. Hypertext was invented back in the 60s as a way of going directly from one part of a document to another, like skipping straight from the table of contents to chapter 10 without scrolling all the way there. Then people started using it to link between different documents and it was incorporated into all kinds of programs throughout the 80s. Berners-lee made the hypertext concept the primary way of navigating what he started calling the World-Wide Web, with hyperlinks connecting different web pages. The Web was a hit at CERN and when it was made public in 1993, the final major piece of today's internet was in place. 

Tim Berners-Lee is the reason most websites you go to have addresses that start with http://www. "HTTP" stands for Hypertext Transfer Protcol, the set of programs used to read documents with hyperlinks. Sometimes you'll see "HTTPS," which is just a more secure version of those programs. The "://" is a way of introducing what comes next and the "www" says that the page is part of the World-Wide Web. 

Soon, a whole bunch of different programs popped up for accessing these new hyperlinked web pages, each interpreting and presenting them slightly differently depending on what their users wanted. These were the early web browsers. The ancestors of whatever you're using today to watch this video. Like Netscape Navigator. Remember Netscape? It was first released in 1994 and the code for it was eventually incorporated into Firefox.

With the web and the internet finally paired up and publicly accessible, the number of computers using the internet exploded in the 90s. And billions of dollars were poured into start-up companies on the internet throughout the decade. Wealthy investors piled cash on like eToys.com and GovWorks and WorldCom. Most of these start-ups went bankrupt around the turn of the millennium when the Dot-Com bubble burst and giant corporations like Google and Facebook rose from the ashes. 

So we'll have to wait until next episode of this series to talk about why and how the internet went from a graveyard of failed start-ups to the place it is today. Which is, you know, more than just a graveyard of failed start-ups.

Thanks for watching this episode of SciShow, which is was brought to you by our patrons at Patreon. If you want to help support this show, you can go to patreon.com/scishow and don't forget to go to youtube.com/scishow and subscribe.