Command Line Heroes - The World of the World Wide Web

Episode Date: June 15, 2021

1995 laid the groundwork for a truly global World Wide Web. But not every country took the same path to connecting to the internet. Some resisted, wanting to create their own version. Others had to fight for access, not wanting to be left behind. And while we made huge strides in connecting the world in those early years, we still have a long way to go. Julien Mailland recounts the rollout of France’s Minitel service—how it was years ahead of the internet, but eventually lost its lead. Steve Goldstein explains what was involved in building the infrastructure to expand the NSFNET beyond the United States. Gianluigi Negro shares how China pushed for its connection, and how different it would be compared to the typical U.S. connection. And Christian O'Flaherty covers how costs weighed heavily on Argentina’s attempts to join the growing international network. Clip of Madam Hu courtesy of Asia Internet History Project. Clip from 'A Glimpse of the Future' courtesy of Richard Seltzer. If you want to read up on some of our research on the global internet rollout, you can check out all our bonus material over at redhat.com/commandlineheroes. The page is built in the style of 1995—check it out.Follow along with the episode transcript.

From Georgia to Palo Alto, from Oslo to Singapore, a vast array of information is being made available in attractive, easy-to-use form and for free over the Internet. A global electronic mall is under construction. People congregate here, interact here, and find the information they want. In 1995, the world was coming online. But that Global Electronic Mall wasn't being built everywhere in the same way. There's an old quote from the author William Gibson that you might know. The future is already here, it's just not evenly distributed. And that was never more true than the mid-90s, when countries around the planet

were racing to join the digital revolution. By that point, 9% of Americans were already online, but only 2% of people living in Eastern Europe and Asia had access to the internet. In some histories, the internet is this American invention that gets packed up and exported to everybody else. It's a cartoonish idea of a global village where everybody comes to hang out. But things were a lot more complicated than that. Every nation has its own economic, cultural, and political dynamics that fundamentally changed the way they came online. I'm Saranya Dbarik, and this is Command Line Heroes, an original podcast from Red Hat. We've devoted all season to that pivotal year, 1995, when earlier networks like the ARPANET and the NSFnet gave way to the World Wide Web. But when we talk about the shift to online life, we often take an American point of view. Yet different countries have very different

on-ramps to the internet and the World Wide Web. So for this episode, we're breaking it up into three mini-episodes. We'll visit three different nations to learn how they each built their own online futures back in 1995. Let's start our time-traveling journey with France. In 1995, the Internet was arriving in France. But France Telecom, the giant telecommunications company now known as Orange, was not so happy about the Internet's arrival. One senior executive was even telling journalists the Internet would be banned.

And why? Because France already had an internet of its own, thank you very much. I grew up with this system called Minitel. Julian Maland is an associate professor of telecommunications at Indiana University. But he grew up in France in the 70s and 80s, and he wrote a book on the network that thrived during that time. The book was called Minitel, and that was also the name of France's precursor to the internet. So what was Minitel? Minitel was a computer system that was rolled out into large-scale production by the French government through the post and telephone ministry in 1983.

You'd head down to your local post office and they'd give you a small terminal called a Minitel. Imagine a tiny desktop computer complete with a tiny keyboard and modem, all totally free. You'd bring your terminal home, plug it into your phone line, and voila, access to a wide variety of Minitel sites. Starting very early in the mid-80s, basically everyone in France had access to things that wouldn't reach the rest of the world until the late 90s. Anything from online chat,

online commerce, any sort of information you can get online. Now, before you start tweeting a correction at me, yes, the U.S. had bulletin board services. And yes, other countries had systems too. There was Presto in the U.K. and BTX in Germany. But in the 1980s, nobody's network had anything like the reach that Minitel had in France. Millions of Minitel terminals were installed, and it's estimated that, eventually, half the French population was using them. Very much everyone was online and had access to all these services that only became mainstream in the rest of the world after 1995. How does this happen? Necessity was the mother of invention. In the 1970s, France's old analog telephone network was in shambles.

New lines were being installed by a nationalized company called PTT, and they could take years to install a new line. It was basically a disaster. It was not at par to what you think of one of the big world powers. At the same time, the U.S. grew more and more dominant in the field of telecommunications.

France, I think, has always been very concerned about U.S. cultural domination. After the war, people called it coca colonization. So France actually had two problems. They had to update their crumbling phone system, for starters, and they also had to fend off the Americans. The solution to both problems

lay in digitizing the telephone network. There was a very important report in 1978 called The Computerization of Society, and it basically said we can fix both our problems by creating a computer system that will be hooked up to the telephone network. And that was Minitel. By handing out Minitel systems to everyone in France, free of charge, you could jumpstart the marketplace. You're creating a huge need

for French content and French tech to elbow aside the American companies who'd otherwise dominate. And at the same time, France's phone network is getting digitized. Minitel went national in 1982, and for a time, it thrived. French citizens were enjoying a kind of web experience that Americans wouldn't have for years. There were chat rooms, there was e-commerce, online banking.

And you also see at that time the emergence of the connected home, the smart house, so you can connect your thermostat at home, your VCR, your garden sprinklers, two minutes out. I mean, honestly, France was sort of killing it. By publicly funding all this, giving a free computer to anybody who asked, the state had overnight created an enormous platform where private enterprise could come and innovate. They even had an app store called the Minitel Kiosk that wasn't so different from the one Apple has today. So by 1995, you might think this incredible Minitel network was going to spread to the four corners of the earth. But...

The export effort was a disaster. I think by 1995, France had given up on exporting Minitel. Other nations were trying to develop their own systems, and those other nations weren't necessarily willing to give millions of terminals away for free the way France had. Without such a huge investment on behalf of the public, the Minitel system wouldn't make sense. One of the reasons Minitel really succeeded was because of this state intervention, but you couldn't really replicate all of that in other countries. And I think that that is one of the main reasons why Minitel did not export well. There was an effort to bring Minitel to America, by the way. In 1991, a company called 101 Online arrived in San Francisco. It was a disaster. They were trying to push this 1983 terminals onto the Bay Area in 1991.

That didn't work out too well for obvious reasons. What's fascinating to me is Malin's reply when we asked about the eventual failure of Minitel. I was not expecting this answer. I don't think Minitel. I was not expecting this answer. I don't think Minitel failed at all. It failed at export, but it did not fail on a domestic level.

It was very successful on the domestic level. When you think about the fact that you had a country that was completely online 20 years before most Americans had ever heard of the Internet, I would call that a success. We have this idea today of what the Internet is supposed to be. The World Wide Web is supposed to cover the planet in, yeah, a World Wide Web. But the story of Minitel and France's futuristic adventure reminds us that there's more than one possibility

for a networked nation. Ironically, the infrastructure advances that Minitel made possible ended up creating a smoother rollout for the American-style internet when it came to France. And Minitel itself was shut down in 2012 because it couldn't evolve into the kind of global force that the web could. But maybe it was never meant to. Maybe its beauty lies in the way it revitalized a nation right at the moment when a change was needed. And I'm with Maland. I'd call that a success.

Okay, where are we flying to next? Let's head over to the other side of the world, to China. What did their 1995 look like? China had the CERNET, which began connecting their academic institutions. But they were eager to make connections with researchers around the globe. To do that, they'd need to patch into an American network backbone that could route them to all the other countries. For the first time at the start of 1995, 64,000 lines connected Beijing and Shanghai to the U.S. Internet access was offered to the public via the telephone network, but getting to that point had been a long road. Chinese authorities had been asking for years to connect to the Internet, but the Americans had refused because of political tensions between the countries.

The pressure to connect was mounting, though. A project from the National Science Foundation, the International Connections Management Project, or ICM, began to offer a little hope. Steve Goldstein was the program's director. and work with the academic networks of other countries to help connect them to what was at that point, at least in the early 90s, pretty much the core of the internet, which was the NSFnet, National Science Foundation Network. Not a small job. Goldstein describes how the actual connections were made.

The physical connection can be satellite, it can be copper or later on it could be fiber. It could be acoustic or it could even be smoke signals. It doesn't matter. It's going to be some way of transmitting that information physically. But once that comes in, it's translated to another way or protocol that the routers recognize, and then the routers route or route that signal through the network. They came into a point of presence usually at one of the NSFnet regional networks, which were then connected to the NSFnet backbone, and thereby to all the other regional networks. They were especially eager to build connections

to research centers around the globe. Goldstein himself lobbied for the creation of the ICM. In 1991, a partnership with Sprint made it possible. Sprint at that time was just getting into internet business and they saw this as an opportunity and actually they zero bid the solicitation, meaning that they said, OK, we'll do it. No charge. Sprint turned out to be a marvelous partner for the five years of the ICM project. They always bent over backwards to make things work. Goldstein reminded us these international connections weren't blazing fast. When they set one up for Poland, for example, it was 19.6 kilobits per second. I was in one of their universities and they let me sit down to read my mail.

And I gave the instruction include, which meant download all my mail. And that was a stupid thing because there were like 200 messages and the whole country was only at 19.6 kilobits per second. And I just clogged up their entire line. Today, Poland's connected at tens of gigabits per second. But that gives you an idea where these links to the larger world began. The ICM started arranging connections to the NSFnet for dozens of countries, even France, despite any hopes that Minitel would take over the world.

But for years, China wasn't allowed to connect. And remember, China didn't yet have the ability to build a global network of their own. In fact, every country on the planet needed that NSFnet backbone to reach outside their own borders with a TCP IP connection. Most of those regional networks were academic, and they were eager to access a series of other academic networks. For example, there were 280 universities in America alone that were tied to the NSF-Net. Meanwhile, access to foreign academic networks was a huge incentive for the National Science Foundation, too. Despite those political hurdles, American academics were very interested in connecting with their Chinese counterparts. There were some key institutions in that country that wanted to collaborate, sometimes in almost real time, with sister institutions in the United States.

So, for example, there was the Institute of High Energy Physics near Beijing, wanted to work on the high energy physics programs, which were international, involve the United States, involve Europe, and so forth. So it was a matter of collaboration. So that connection was warmly greeted from that point of view. Championing the idea of China's connection to the global network was computer scientist Shi Hanhu, speaking here with the Asia Internet History Project. Chinese science and technology people, there is a keen desire for the cheaper communication

and the transmission of the big amount of scientific data during their scientific cooperation with their overseas cooperators. So that is the major reason why China at that time has such a motivation to acquire the internet. In 1994, Hu traveled to Washington, D.C. for a committee meeting and made a special side trip to see the chair of the NSF, Neil Lane. She made the case that China had to be connected to the NSFnet. And it happened. On April 20, 1994, China became connected to the World Wide Web. Goldstein remembers Hu being a strong advocate for Chinese sovereignty, despite their need to connect on a shared network.

At one point, she had asked, well, OK, now can we connect? I think I answered and I said, yes, but will you agree that any academic person in China can use the link, not just members of the Communist Party? And her answer was, well, do you have such restrictions in the United States? And she caught me. I had nothing to say. I said, OK, you win. There was a lot at stake for China, and arguably it wasn't just about scientific collaboration.

Gianluigi Negro, a professor at the University of Siena, researches the history of Siena, researches the history of the Chinese internet. He told us that connecting with the NSF-Net was part of China's larger movement toward openness. At the time, China started to literally be more connected from not only the economic point of view, but also the political one. And this is very important because between the end of the 80s and the beginning of the 90s,

there were a lot of projects that supported telecommunication industry development in China. Opening up economically and opening up academically went hand in hand. The World Bank began making enormous loans to China, helping them develop their networks. 1994 saw decades worth of change suddenly happen once, that day when Madam Hu's request was granted and the NSF connected China to the internet.

For the first time, China was connected to the TCP-EP internet connection. So this is very important from an infrastructure point of view. The Chinese internet was never going to be the same as the American internet, though. The so-called Great Firewall regulates what Chinese users can access. Sites like Baidu and Weibo dominate instead of Google and Instagram. But Professor Necro reminded us that despite those differences, the Chinese internet parallels the American one in other ways. Think about the mentality or about the Chinese application or the Chinese services, the management that they have.

Think about Alibaba. Although they have a very strong connection with the Chinese government, business mentality is very, very close to the U.S. ones. This is a reality we see around the globe. While countries came online in their own way, and while every country's internet still navigates distinct politics and economic realities,

they're also universals. The long work of creating a global networking infrastructure is about maintaining difference while forging those common bonds. The internet we have today operates in that careful balance between universals and variations. We've got time for one more stop. Let's head over to the Americas now, South America. Our final destination, Argentina. By the spring of 1995, they were selling their first commercial internet connections. Email and local networks were established by the early 90s, but connecting to the internet itself was a much larger challenge.

We asked Christian O'Flaherty, Regional Vice President for Latin America and the Caribbean at the Internet Society, to talk us through those early days. International data links were not easy to get, very expensive. The bandwidth was low. There were a lot of impediments from the at-the-time incumbents or monopolies for international data services.

In the early networking days that O'Flaherty is describing, the early 90s, there were a few nodes with international connections, but they were super expensive satellite connections. The general public didn't have access at all. Academics in Argentina were pushing for a more genuine connection to networks beyond their borders. It was mostly the pressure from those professors and researchers

that were working in Europe and the US and were using the internet in those countries. So when they came back to Argentina and other countries, they started looking for ways to get similar services. You remember that monopoly OflarFlaherty mentioned? That was an international company called Talental. And O'Flaherty told us that, in the 90s, Talental was dragging its feet, delaying the arrival of an affordable internet.

They didn't want to provide that service. A truly online Argentina was never going to be possible until somebody got them connected in an affordable way. And there didn't seem to be a local solution. But physicist Emma Perez Ferreira was looking beyond their borders. She's considered the person most responsible for bringing the internet to the people of Argentina. She was a national pioneer and also a well-respected pioneer in the region. Lima was able to insist on many different areas of the government and elsewhere, and she was able to get this authorization to connect to the internet, getting service from whoever was able to provide it. If Telentel wasn't going to step up, then she would.

Since the 1980s, Ferreira had been the director of a network called Retina that connected academics at a regional level. But she saw the potential for her mission to grow in the 90s when America began providing the NSFnet's backbone to the world. Suddenly, through that ICM project of Steve Goldstein's, it was possible for Argentina to patch into that backbone and join a global internet. She received authorization to create a direct link via Florida.

O'Flalarity describes the collaboration with American partners. There was a lot of support and interest from different parties to help developing countries, and in particular Latin American countries, to connect to the internet. Starting in 1992, the connection to the NSFnet ran through the city of Homestead in Florida using the Pan AmSat satellite. The router was managed by Sprint, and the National Science Foundation paid the management port fees. Many Latin American countries received support for equipment, and the port was provided free

of charge. They facilitated the connection through trainings and help on getting our own IP addresses and things like that. But getting Argentina's institutions to step up wasn't as easy as you might think. Because unlike the situation we heard about in France, most authorities in Argentina didn't yet understand what the internet could provide. It was difficult to demonstrate that it was worth paying for those services for something that nobody knew about.

So nobody used the internet when she was asking for funds and convincing the universities and the research institutions that it was worth spending the money that way. Ultimately, leaders like Ferreira were able to tie Argentina's networks to America's NSFnet. Up until that time, less than 0.1% of people in Argentina were internet users. But those numbers started exploding once the NSFnet connection was made. And today, three quarters of Argentinians are online. We could keep on going. There are dozens of other stories we can't tell in a single episode. Every country had its own heroes, its own unique issues

when connecting to the rest of the world. Nobody's 1995 was the same. Our 1995 season has been telling a story about how the modern online experience arrived. But we have to remember, history doesn't really happen that way because these other stories and timelines were always running in parallel. And yes, 1995 was a dramatic shift point when the World Wide Web started racing around the planet, bringing with it a certain sameness to the internet.

But when we dig into our history, we remember that the networks we share actually live inside distinct cultures with their own distinct histories and definitely their own futures. 41% of the world is still not online. And over the next couple of decades, as billions more people join the internet, it can't be just about exporting some American tech. And it can't be about folding the world into one tidy version of events. It's going to be about the Internet meeting every country's own ambitions and hopes. You can learn more about how the internet came to France, China, Argentina, and other countries by checking out our bonus material at redhat.com slash command line heroes.

Next time in our season finale, the dot-com bubble that began in 1995 is finally going to burst. We'll see what lessons were learned when the year 2000 turned the tech world on its head. Until then, I'm Saranya Dbarik, and this is Command Line Heroes, an original podcast from Red Hat. Keep on coding. Hi, I'm Jeff Ligon. I'm the Director of Engineering for Edge and Automotive at Red Hat. The number one hesitation that I think folks have about Edge is that they assume they're not ready for it. They think that it'll mean starting over from scratch,

or that every time their business needs to change, they'll be re-engineering solutions and re-evaluating vendors. And look, Edge is complex, and it does mean a different approach, particularly for devices at the far edge. But with Red Hat, it doesn't mean starting over with new platforms. We've taken the same enterprise solutions that simplify and accelerate work across the hybrid cloud and optimized them for use at the edge. So you can manage your edge environments just like you do the others.

Come find out how at redhat.com slash edge.