Advent of Computing - Episode 13 - Minitel, the French Network Connection

Episode Date: September 22, 2019

Today we are dipping back into the deep and complex history of the proto-internet. We are going to be looking at Minitel, a France-Wide-Web that was built in the 1980s as a way to help the country stay relevant in the digital age. Like the show? Then why not head over and support me on Patreon. Perks include early access to future episodes, and stickers: https://www.patreon.com/adventofcomputing Important dates in this episode: 1980: Minitel Program Networks France

It may be falling a little bit out of fashion nowadays, but if you look at the URL for most websites, then you'll see the prefix www. That abbreviation is, of course, shorthand for World Wide Web. It's so common to just call it the internet or the web that often we don't give much thought to the idea of worldwide in its name. of Worldwide in its name. Even though large-scale networking research had been going on since as far back as the mid-early 50s, there was never a guarantee that network connections would be made internationally. And in a lot of cases, the exact opposite was true. Even ARPANET, the direct predecessor of the internet, was only ever designed to network the US independently. It may be strange to think about it, but in the early era of what would become the web, worldwide was rarely, if ever,

part of the discussion. But despite the small scale of these early, isolated networks, a lot of them were hugely successful. Some of these networks were actually longer lived than our modern incarnation of the internet. Today, I want to tell you the story of Minitel, a French network that started as a modernization project in 1978, but quickly turned into a proto-internet that would continue to operate all the way until 2012. So what was Minitel? How is it connected to the larger context of the internet?

And how on earth did it manage to the larger context of the internet? And how on earth did it manage to survive so far into the modern day? Welcome back to Adrin of Computing. I'm your host, Sean Haas. This is episode 13, Minitel, the French Network Connection. And today, we're dipping back into the deep and complex history of networking. But instead of looking at Cold War-driven struggles for networking like in the US and Russia, we'll be examining a reaction to a different kind of conflict. The 1970s and 1980s saw a radical shift in how computers were viewed and used.

This era is often called the microcomputer revolution, because for the first time, there were computers small enough and cheap enough that they could start making their way out of massive institutions. And as computers moved into the wider world, a lot would start to change. On the small scale, automation via newly computerizable tasks made a lot of workers redundant, and that led to large-scale unemployment. And on a much larger level, a new market was starting to be created, and technology that was talked about as science fiction in the earlier half of the 20th century was starting to become an everyday reality. This all created a scenario

where nations had to come to grips with the changing world. They could either find a way to adapt or get swept off into irrelevancy. Now, different countries had vastly different responses to these technological growing pains. In the US, this change wouldn't be very abrupt. There have been designs on the digital future both from inside and outside the government since at least the 1930s. So in a lot of cases, American industry was in the driver's seat during this transition. The UK is an example of a different approach. In that country, the BBC's computer literacy project worked to train citizens to pivot them into new jobs and opportunities they were opening up with the spread of computers. In France, this adjustment came to

focus on networking. And in the tail end of the 70s, this would give birth to a wide-scale project colloquially called Minitel. So how exactly did the issue of digital modernization in France lead to a nationalized computer network? And how did France go about creating a France-wide web over a decade before the first web browser would ever be written? Perhaps one of the best ways to survive the microcomputer revolution and remain relevant was to keep up with the United States. Now I said at the top that France chose to do this via networking, but that's kind of burying the lead. The goal of networking the nation wasn't an end to itself, but rather it was a way to kickstart the French IT sector and help domestic computer companies become more competitive with foreign juggernauts like IBM.

In a lot of ways, the French plan would end up being somewhat similar to what the English government accomplished with the BBC's computer literacy project. But instead of focusing on education, the French solution would come in the form of more direct economic stimulus. So let me back up and explain a few things before we dive deeper into the details on how France would try to jumpstart their IT sector. And well to start with, maybe jumpstart isn't even the right word to use here. It's more like they built it up from scratch. Now to put it lightly, France was a late comer to the computer game. The earliest French built mainframes had only started coming out to market in the very late 60s. So by the time the 70s came around, there wasn't an established computer industry in the country, at least not a very large one. Today, this wouldn't really be an issue.

Internationalization has gotten to the point where it's hard to tell where a computer even comes from. But that wasn't yet the case in the 70s, and as industry was starting to shift into the digital age, it became increasingly important to the nation's bottom line to have at least some kind of IT sector. However, the need to modernize wasn't purely economic. The French government saw an increasing dependence on foreign-made computers and foreign-written software, and as they noticed that, that brought up a second key concern, sovereignty. As French industry became more and more dependent on computers in general,

it also became more and more dependent on foreign companies. This came into clear focus in 1966. During the 60s, France's Atomic Energy Commission had been working towards developing a hydrogen bomb, and their labs had been primarily using American-made mainframes. But their progress would come to a halt when the US government banned IBM and CDC, two of the larger computer manufacturers in the country, from exporting computers to the Atomic Energy Commission in France. For the states, this was simply a play at stopping nuclear proliferation. But this struck deep at France's national sovereignty. This event made it clear that reliance on American computers and software left France at the mercy of American government and industry. But the fiasco over the H-bomb project wouldn't be the final straw. In fact, it wouldn't even be an outlier.

American industry and government would continue to encroach into the 1970s. And France's own computer industry wasn't able to keep up on its own. The president of France at the time, Giscard d'Estaing, put it bluntly, to quote, For France, American domination of telecommunications and computers is a threat to its independence in the crucially significant if not overriding area of more industry in the country going all the way back to the mid-60s, but none had met with widespread success. This would change in 1976, when the French government decided to take a serious action, and that turned out to actually be commissioning a report.

It doesn't sound that serious, but really a lot of innovation and new policy does come out of these kinds of actions. This report would end up being known as the Computerization of Technology. end up being known as the computerization of technology. The paper itself, written by Simon Nora and Alain Mink, outlines the problem facing France and a few possible solutions. Specifically, the report focuses on how the government could improve both the computing and telecommunications sector in France, with the goal of escaping American domination.

And reading from the source, Nora and Mink don't pull any punches. When addressing the larger issue, they wrote, quote, knowledge will end up being shaped, as it always has been, by the available stock of information, leaving to others, i.e. to American databases, the responsibility of organizing the French collective memory while being content to plummet is to accept a form of cultural alienation. Installing French databases is an imperative of national sovereignty. End quote. And this cuts to the root of fear in that era.

As computerization was very much changing society, whoever had control of that process could control what society would change into. And having American-dependent computers and networks could quickly turn into having an American-dependent society. The core sentiment is something that we still struggle with today. As the internet becomes more and more casually integrated in society,. As the internet becomes more and more casually integrated in society, corporations are able to exact more and more control over the society. Technology today may be different, but the issue remains the same as in 1976. Computerization in society put the deepest fears of France into words. American domination of computing and telecom

wasn't just an economic threat, or even just a matter of sovereignty. In fact, it was a fight over the continued existence of French society. But how could this threat be stopped? In the coming months, a plan would start to form from within the French government, building upon what was written about in Computerization of Society. This would soon be known as the Telematics Plan. And once complete, it essentially boils down to this. Get France Online. Basically, this would entail the French government making something akin to a dot-com bubble, but nearly a decade prior to the internet's conception. The plan was to provide mass government investment

to create an IT sector from scratch. In the end, this would involve modernizing the national telecom grid, creating a nationalized internet-like service accessible to every French citizen, creating a nationalized internet-like service accessible to every French citizen, making a user base from scratch, and carving out a space for private industry on this new network. And as shocking as it may seem, by the end of the decade, all of these goals would be met, and then some. So that's the overview, but how did the specifics of this new French network shake out?

Like I said earlier, Minitel wasn't just a network or a terminal, it was an entire platform. It may be easiest to think about as analogous to the internet, at least partially. You have a hardware layer that's made up of servers, the actual network cabling itself, and the end user's computer system. This hardware layer has to work together with a layer of software that actually manages routing information and eventually rendering things like webpages on the user's terminal. That's the broad strokes at least. The modern internet that we all use was designed and built sometime in the 1990s. The exact date varies a little bit depending on

who you talk to. But the design and development stages of Minitel started as far back as 1978. Now you may be tempted to think that over a decade of difference would just mean that Minitel's a scaled down and less sophisticated version of the internet. But that's not really the case. And it's also not a predecessor to the internet, because both of them would end up living side by side for a number of years. So let's look at how Minitel worked, and how the French government was able to create a network that's similar but distinct from the later internet. Something important in the conception of Minitel was its intent. You see, the network had very different goals

from the internet. While the internet grew out of networks that were used for government and academic tasks, Minitel was designed from the ground up to be used by the average citizen, and as a platform for both private companies and public institutions to offer services. The goal for Minitel was to get as many users and companies on board as possible, as quickly as possible. So a lot of the choices made in designing Minitel were to make it very palatable to the general public. Another key difference that I want to make clear is that Minitel wasn't some always connected service like the internet is today. Instead, Minitel was more of an on-demand experience. Most users didn't spend days jacked into the network,

but would log on to do a specific task and then log off. So with that in mind, I want to go into the specifics and talk about what exactly made Minitel Minitel. The first part I want to cover is also the most visible and probably most recognizable part, and that's the actual Minitel terminal. In the analogy of the modern net, these take the place of a user's computer and web browser, but they weren't really fully-fledged computers. Instead, they're what's known as a dumb terminal. That's advice that just does one thing, connect to another computer. And when it comes to hardware, that was really all a Minitel terminal could do.

It was composed of the bare essentials. When you got a Minitel terminal and opened it up, you'd see that it only has a small CRT screen that can only display black and white text, a keyboard, and a modem. Now, that's not very impressive, and some later models would come with different options, even a color screen, but the general loadout is very bare minimum. Now, modems aren't really something that's used anymore, so I think it bears a little bit of explanation.

Basically, it's an old style of network interface. What's interesting about a modem is that it's not entirely digital, even though it's connected to a digital computer. It sends and receives information as analog signals, specifically as sound. Just as a side note to that, the intro and outro bumper for this podcast is actually a sample from a modem connecting up to a network. So basically a modem allows you to turn digital data to analog and then back into digital. They don't see much use anymore, but in the 80s and 90s they were fairly ubiquitous. They allow computers to send data over a phone line, despite the telephone grid being strictly analog, and incapable of doing any digital data transfer. And in that period that they were active, this was used to connect computers up to networks. So moving on from that modem

tangent and going back to the actual terminal, what I find interesting is you can see the whole ideology behind the Minitel platform made into a physical device in these terminals. They're small, unobtrusive, and built for convenience. The screen on most models of Minitel terminals were only about 8 or 9 inches. The keyboard was connected to the terminal by a hinge, so it could be folded up over the screen when not in use, and it was small enough to stash away until you needed it. To make the Minitel terminal even more convenient, the French government handed out units for free. Of course you could buy higher end models, but most Minitels were given away. This may sound like a huge waste of money on the part of the state, but it actually served two very important purposes.

First of all, it made an instant user base for Minitel. It lowered the barrier for entry to the platform to essentially nothing. And secondly, it gave a direct boost to French computer manufacturers. A big part of the initiative was to make sure that all free Minitel terminals were ordered from French companies. So right off the bat, more money was being pushed into the burgeoning IT sector, and more users were immediately able to get onto the network. The next part that I want to talk about is the network itself, or what actually makes all these free terminals useful. is the network itself, or what actually makes all these free terminals useful.

The, and I'm going to butcher this, Centre National d'Etudes des Télécommunications, or as I'm going to call it if I ever say it again in this episode, CNET, France's National Centre for Telecommunications Studies, had been working on a national network going back as far as 1971. This network was called Transpac, and it was somewhat similar to the American ARPANET. In fact, a lot of the technology used in Transpac was borrowed directly from the Americans. Transpac itself was fully digital, and for the time, it was pretty cutting edge. And by the time plans for Minitel were being drafted, Transpac was a relatively mature network and it reached most of France.

But the problem is, Minitel users didn't connect directly to the digital network. They couldn't. Instead, they would dial into a gateway over existing analog phone lines. Partly, this was due to the fact that most homes just didn't have direct feeds into a new nationalized digital network. And partly it was to give the French government a way to centralize control over Minitel as a platform. These gateways were called point d'accès vidéo text, or pavIs, and were used to mediate and translate traffic from the

analog phone lines into the digital TransPack feed. Once connected to a PAVI, a Minitel user could then link up to any server on the nationwide network. The PAVIs also served as directories of available services on the network. One key part of the design here is that everything I've described, from Transpac to the phone lines to the pavis, is owned and operated by the French government. This meant that the state had full control over how people used the network and were able to track that use. But outside of the actual network,

Minitel would be relatively open to third parties. And this split state and private ecosystem was done by design, and it's something very unique to Minitel. Other networks didn't have this. It was done as a way to create a new IT sector for France. The government would provide the entire network and allow third parties to hook up to it. And what exactly were these third parties doing on the network? Well, that brings us to the next part, the services on the network, or what people could actually connect to once online. Instead of the websites that we visit today,

Minitel had services. These ran the gamut from phone directories to chat rooms to digital banking. And importantly, almost all of these services were offered by private sector businesses. Setting up a new service was also relatively easy. All you had to do was connect your server to Transpac and then register it to be listed on one of the Pavi's directories. And as far as hardware, you could have any computer that could connect to a network. There were mainframes that ran services, but there were also microcomputers like Apple IIs. Overall, it made the barrier to entry for being a service provider very, very low. And that made it insanely easy to go into business.

And that brings us to the last part of the equation, at least as far as the internet allegory goes, and that's the media, or what's actually being sent over the network to users. Keep in mind, this is well over a decade before HTML would hit the scene. And even if HTML had been available, a Minitel terminal just couldn't display it, since they had text-only displays. Instead, Minitel used a technology called VideoText. However, it's not really a one-to-one replacement for HTML. In some ways, it's a lot more sophisticated. First of all, VideoText could only display text,

but it could send that in multiple colors. But importantly, VideoText could do two-way communication. You could receive data from a server to a terminal and you could send data back from a terminal to a server. This meant that video text services were interactive and that's something that you just don't get with plain HTML on its own. Now one part of Minitel that doesn't have a direct corollary to the modern web is a system called Kiosk. This was the billing service that was used with Minitel. You see, using the network wasn't entirely free. There had to be some way for the French government to recoup money, and for service providers to rake in profits.

profits. Services were billed by the number of minutes a user spent connected and could be billed at different tiers ranging from totally free for some government services to just over two francs a minute for premium sites. What made kiosk work so well was that it was integrated with the customer's existing phone service bill. So instead of using some new e-commerce system, you could just log in and let the phone company deal with all the billing. At the end of the month, you get your normal bill, except it has a new line item for Minitel usage. This was all possible thanks to the design of Minitel's network. Since control over the entire network was nationalized, including the phone lines and the pavies, it was simple for the

government to see who was using the services and bill accordingly. This was also a huge boon to companies that ran Minitel services. As a private company, you didn't have to deal with your own e-commerce or billing. Instead, the kiosk system would automatically manage that for you. It was a real win-win. It made Minitel easy to use and feel familiar to consumers, and at the same time, it made business easier to handle for service providers. Minitel services would start hitting the wires as early as 1980, first in Saint-Malo, France, but by 1982, the service would cover the entire country. One of the first services offered on the new network was simple but effective. A digitized and searchable phone directory. And now, that may not

sound impressive, but I think that service alone really encapsulates what Minitel was set to be. It's not a radically new system that's meant to reinvent life. Instead, it's a way to improve what's already possible. And at its height in the mid-1990s, Minitel would have somewhere north of 25,000 different services available, and it would be regularly used by over 25 million people. Just to do some quick math for fun, for the time, that amounts to about 41% of people in France being online with Minitel. Now, the World Wide Web wouldn't see that kind of penetration until sometime around 2000. So going by sheer numbers alone, it's clear to see that Minitel did exactly as planned. It created a massive new market in France from nothing.

So what did these thousands of services look like, and how did they bring in so many people? This is one thing that I find remarkable, but a lot of the services offered by Minitel are the same popular services online today. People have never really changed, it's just the technology gets better. There were services geared towards work, play, and everything in between. This includes practical services like banking. In fact, digital banking was one of the more popular Minitel services. But it wasn't just for managing your bank account at home. Minitel opened up the market for day trading, since you could now easily and quickly access stock markets from anywhere with a phone line without ever having to place an actual call.

There were also services for day-to-day tasks. This included things like ordering tickets for travel or events, or even home delivery services all years before counterparts like Ticketmaster or Amazon ever existed. You could even register for university on Minitel. And in fact, a lot of public universities would require digital registration. Outside of the practical, existed more entertaining services. All the way back in the 80s, Minitel sites were serving multiplayer games and contests. Games on the system could range from simple quizzes to full-blown 3D mazes. But one of the most popular types of services were what were called massageries, or chat rooms. Decades before

anyone would ever even think about anonymity online, many tell users in France were holding digital masquerades. Connected folk would use massageries just to chat with others, make friends, or even for online dating. And since people are how we are, one of the ever-present services on the network was the adult chat room, often dubbed Minitel Rose. And it's just as it sounds. And even better, it was built by the minute, so it became really a profitable type of service. One of the ironic results of Minitel was the sheer amount of print material that was produced for it. Soon after release, printed directories for Minitel services would start to be published, complete with glossy ads for premium services. In fact, France Telecom, the very government agency that was in charge of keeping the Minitel project going, would end up publishing some of these directories themselves.

On top of the directories were the expected books about using the terminal, or tips and tricks for the network. Even more interesting, though, were the ads, both in print and plastered on walls and billboards. And, well, a lot of these ads were for Minitel Rose services. In this era in France, it became common to see large billboards with a scantily clad woman that simply had a bold print number to dial. And it's important to keep in mind that every service I've mentioned was usable on an 8-inch black-and-white text display, I've mentioned was usable on an 8-inch black and white text display. Decades and decades before the first network connection over the internet would ever be made. Alright, I think it's that time again. Let's wrap this episode up. So overall, Minitel did exactly what it set out to do.

It helped France stay competitive during the digital revolution by creating a new market from scratch. And along the way, it connected an entire nation well before the internet ever could. But there's still a lot more to the story. And a big part of that is the preservation of Minitel. The network was finally disconnected in 2012, but users and services had already disappeared. Luckily, there's an active community of hackers and historians that have been working on keeping the terminals alive and tracking down every shred of information. But for systems like Minitel, that can be difficult, since much of the software behind the system was never properly preserved. And beyond that, the user base that made Minitel alive has since moved on. While working on this episode, one of my best resources was a book called Minitel,

Welcome to the Internet, written by Julian Mayland and Kevin Driscoll. The two have been working since before the shutdown of Minitel to document and preserve the platform. I was lucky enough to get the chance to talk to both of them, and even better, the whole thing was recorded. So if all goes to plan, then I should be putting out that interview this coming Monday. To see their amazing work with Minitel, you can go to Minitel.us or, and I highly recommend this option, track down a copy of Minitel Welcome to the Internet. It's really a great read. And one more quick announcement.

Adren of Computing has been online for a little over six months now, so I decided it was about time to get some merchandise out. If you want to show support for the show, or just have the coolest new shirt in your neighborhood, then this is your chance. I have a TeePublic store set up for the podcast. I'll have a link to it in store set up for the podcast. I'll have a link to it in the website and in the show's description. And thanks for listening

to Advent of Computing. Remember to come back next week for my interview with Julian Mayland and Kevin Driscoll. Until then, if you like the show, then why not take a minute to share it with a friend? You can always rate and review on Apple Podcasts podcasts if you have any comments or suggestions for a future show then go ahead and shoot me a tweet i'm at advent of comp on twitter and as always have a great rest of your day