Advent of Computing - Episode 20 - PLATO Part 2: An Online Revolution

Episode Date: December 30, 2019

In the conclusion to our discussion of PLATO we look at the final incarnation of the system: PLATO IV. How did an educational machine turn into one of the earliest online communities? What was it like to use PLATO at it's height? Along the way we will look at the software, hardware, and video games that made PLATO so special. 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: 1964: Plasma Display Patented 1972: PLATO IV Launches at University of Illinois 1973: Empire, First MMO, Developed for PLATO IV

Legend tells of a room buried deep in the University of Illinois campus. Some call this place the zoo. It's said that this was a magical location, one where many a student would find themselves trapped. The zoo has been described as the kind of place where you could spend your whole life and not even notice until it was far too late. Inside the zoo, it was possible to voyage to distant worlds, fight against aliens to protect the earth, delve into deep dungeons to hunt for long-lost treasure,

or even warp into the distant future. In the zoo, it was as easy to talk to someone hundreds of miles away as if they were sitting right next to you. You could even do something as mundane as reading the daily news. There really wasn't a reason to ever go out of the zoo. Well, that is except for one thing. And that's when classes started again in the morning. You see, in addition to being a digital wonderland, the zoo was just an operating classroom. In reality, it was room 165A and B, a joint computer lab used for teaching classes at the University of Illinois. The room itself was nothing more than around 80 Play-Doh terminals neatly arranged on desks.

But during the night, and some stolen hours between classes, the zoo came back to life. PLATO 4, the ultimate iteration of the PLATO project, first came into use in 1972. It was a total redesign, addressing all the issues of the project's earlier iterations. It was posed to change how the world learned, but it had another side to it. This new machine also created a hotbed for hackers, gamers, and tech heads. So how did an educational machine end up living a double life? And how did it spawn one of the earliest Welcome back to Adjunct of Computing. I'm your host, Sean Haas, and this is episode 20, Play-Doh Part 2, An Online Revolution.

This is going to be the conclusion to our dive into the Play-Doh project. So if you haven't already, then I'd recommend going back and listening to part one of this series. That's episode 19. It provides a lot of the background that we'll be building on today. In this episode, we're going to look at the hardware, software, and the community that made up Play-Doh 4. This was the final iteration of the system, and like I mentioned last time, this is where everything is going to change. The early days of the project, Play-Doh 1, 2, and 3 I mentioned last time, this is where everything is going to change. The early days of the project, Plato 1, 2, and 3, were essentially a dry run. Plato 3 would be used in classrooms at the University of Illinois in the latter half of the 1960s, but it had some hard limits on its usability. Plato 4 would expand beyond its home campus, but to do so, the platform was totally revamped.

By its height, PLATO 4 would have multiple networked supercomputers installed around the world. Each central computer would support over 4,000 terminals. And in doing so, it pushed past its original goals, changing from a purely educational platform to something completely different. So how did Plato transcend its early roots? How did it overcome the limitations imposed by its very own design? And how did it build an early online community entirely unconnected from the ARPANET? I think it's fair to say that Plato3 was truly a triumph, but only in a limited scope. The system, backed up by a CDC-1604 mainframe, could only run up to 20 terminals at once. So, that means it could only ever teach up to 20 students at once.

Part of that limitation came down to raw computing power. The CDC mainframe being used was relatively powerful for the time, but timesharing, the method that Plato used to share resources amongst multiple terminals, was still in its very early years. There just wasn't an established way to break up computing resources like this, and it really starts to show in the design of Plato 3. It seems like one of the big issues was, as always, memory constraints. This comes down to how TVs or really any graphics display works. Something like a TV doesn't actually store an image, it just displays a signal. Even to keep something as simple as a static image on a screen, you need to constantly be sending that image to the screen. For analog

devices, such as the setup used by a TV news station in this time period, that was a relatively simple issue. A camera will keep sending out information frame after frame until you turn it off. But when we get into the digital realm, we run into a couple more complicating factors. To keep something as simple as even a welcome message on a screen, you need somewhere to store that. The obvious solution is in a computer's memory. That's what we do today. Every pixel of a display is kept stored in the memory on the computer that it's attached to. Then that data is just fed out to the monitor. That kind of image data starts to add up pretty quickly.

For the sake of argument, let's say that we wanted to display a black and white image, 512 pixels by 512, and we wanted to put that onto a screen. Now, that's a pretty low resolution by modern standards, but it would be considered respectable in the 60s. Each pixel would need to have a single bit, either an on for a white pixel or off for a black pixel. So the entire display is roughly 262 kilobytes worth of data. If you wanted to drive some 20 independent screens, as is the case with Play-Doh 3, that comes up to just over 5 megabytes of storage.

Now today, that wouldn't be a problem. Storage is cheap, memory is cheaper. But at the time, that was a nearly insurmountable task. The CDC-1604 computer that was in use had 192 kilobytes of system memory. That's not even enough to store a single screen worth of data. The PLATO team found a way to get around the issue by using a combination of image slides on a computer-controlled projector and storage tubes. Thus, it would shift a lot of the memory load off the central computer. But as we've seen with PLAToh 3, that workaround leads to its own problems. The main issue being the communication lines involved. In Play-Doh's versions 1, 2, and 3, a composite image from the slide selector and a storage tube had to be sent as a CCTV signal to a modified TV screen.

That would serve as the terminal. This worked, but was complicated, expensive, and not entirely reliable. That takes us up to around 1963 or so. While PLATO-3 was finishing development and entering use in some of the first computerized classrooms, the team at the Coordinated Science Lab, aka CSL, aka the lab where PLATO lived, were already starting to work on the system's successor. And there were some clear goals as to what the still theoretical Plato 4 had to be able to do. Goal one was to find a way to get rid of the storage tubes. They were just too cumbersome at this point in time. And goal two was to somehow get more terminals attached.

20 students at a time just isn't enough. As we'll start to see, these two goals are actually very interconnected. Anyway, the storage tubes were of prime concern. In the early stages of Play-Doh, they made everything possible, but they'd outlived their usefulness. of Plato, they made everything possible, but they'd outlive their usefulness. Bitzer had been searching for a solution as far back as 1961, but the problem was that there really wasn't any technology that could directly replace them. One of the few leads would come from a very unexpected source, and it makes for a relatively interesting cameo appearance in this whole story. it makes for a relatively interesting cameo appearance in this whole story.

It turns out that there was a set of patents registered in the late 50s for a device that may be able to replace the storage tubes. Out of the handful of patents that are concerned with this device, the ones of interest were for a quote-unquote luminous display device and a memory device that used excited gas to both display and store data. Who was the inventor of these patents, you might ask? It was none other than Doug Engelbart, the future inventor of the mouse and developer of the first graphical user interface. Prior to his more well-known works, it turns out that he had been looking at a similar problem to the one that Plato was looking at. He had devised a solution, but it's unclear if his plans ever went beyond the test bench. In 1968, when he gave his famous demo, he was using a variation on the storage tube to CCTV setup, so more than likely the display research met a dead end.

The Engelbart patents laid out an interesting possibility for the team back at Illinois. The principle of operation for this so-called plasma display is intriguing in its own right. The entire apparatus is something like a much more complex and computerized neon light. Gas trapped in an evacuated glass container when you pump it with just enough energy will change states into a plasma, which has a few interesting properties. First of all, and most excitingly, it glows. But it has another property that's a little less flashy. Given the right conditions, it will stay as a glowing plasma even after you reduce the power that you're putting into the system. The reason behind this is actually super cool. At least it seems super

cool if you're a nerd like I am. Let's think back to chemistry class for a second. Plasma is one state that matter can exist in, and you get matter to that state by putting more and more energy into something that's already in a gaseous state. Once you hit a certain level of energy, the gas transitions into a plasma. To get a plasma back to being a gas, you just lower the amount of energy. To get a plasma back to being a gas, you just lower the amount of energy. But, and here's what makes this so interesting, for certain gases or gas mixes, that energy level where it transitions from a plasma back to a gas is considerably lower than the energy level where it would transition from a gas to a plasma. It's subtle, but by keeping a tube full of gas at just the right energy level by putting electricity into it, you can quickly flip it into a plasma by giving it just a little extra kick of power. Then, when you drop back down, the tube will keep glowing. To turn it off,

you just have to dip the power below your baseline for a moment. Given a little more wiring, you can even read if it's on or off. And from the outside, that, my friends, looks and functions exactly like a single bit of computer memory, even if on the inside it's just a super fancy expensive neon tube. Still with me? Well, that's just part one of the two patents in play here. That's the memory device. Doing all that complicated work only gives you a single bit of memory, and that's not really enough to do much with. Engelbart's later patent, the luminous display device, is where the payoff comes.

If you have a single bit of memory, then why not just make a bunch of those and put them all together? The display device is actually just as simple as that. It's a matrix of these smaller plasma elements. A little smart wiring makes it so that any of the bits of the grid can be flipped on or off independently. You can imagine that when Bitzer and his cohort first ran across this, they were pretty excited. In practice, it acts like a storage tube, so it could be a drop-in replacement. But it had some key advantages. It's smaller than a storage tube.

The whole assembly for a plasma display would be much less than an inch thick. It wouldn't lose an image like a storage tube does over time. As long as there's power to the display, the plasma can remember where it was last stored as. If produced correctly, it also had the possibility to be much cheaper and much more reliable. Altogether, this could be the device that really was needed to change Play-Doh up. The team at CSL could ditch both the storage tubes and the TV signals. It could be possible to put a plasma display in each terminal. Then, instead of sending an analog CCTV signal, an image could just be sent to each terminal as digital data over something like a modem.

The implications were very clear. A plasma display would make it possible to have more remote PLATO terminals, and just more terminals in general. With the older systems, a dedicated storage tube had to be kept in the server room for each terminal you wanted to connect, which made adding a new terminal a real headache. But with plasma displays and a little extra terminal hardware, you could make it so that all you had to do was plug in a new terminal and then just log on. But things wouldn't be fixed overnight. The problem was that the plasma display was only an idea, and it was only on paper. I can't stress this enough, but at this point in the story, 1961 or 62, it's only an idea. No one had gotten this to work in any meaningful way. Engelbart's patents were floating around, so there was a trail to follow, but a lot of the details would still

need to be worked out. But this kind of uncertainty hadn't stopped the team at CSL before. In 1962, Robert Wilson, a grad student who was already researching at the lab, would take on the new project. Wilson started producing single-pixel prototypes, but problems were constantly mounting with his prototypes. Working with Bitzer, Eugene Slotow, another professor and Plato engineer, and a few other researchers, it would take a couple of years to get just a single pixel to reliably light up, remember its state, and turn off. But once they had a working pixel, it was simple enough to scale up. The team constructed larger and larger screens in the range of a few dozen pixels until they were satisfied that their design would work on larger scales. Then, in order to get

the prototype actually to a usable size, CSL contracted out to a nearby glass company to manufacture the final screens. These finished plasma displays, or rather, the very first plasma displays, were markedly similar to the design in Engelbart's patents. It was a 9-inch, 512x512 array of glass plasma memory elements, each connected up in a matrix. But it diverged on the specifics. This new display was monochrome, only displaying one color. And thanks to the gas mixture, that color was a warm orange that would become Plato's signature for decades to come. In 1964, the plasma display would be patented, and sometime around 1965 or so, the first production screens would start coming off

the nearby assembly line. But that was only one part of the equation. The plasma display was a revolutionary technology, but for Bitzer, it was only one piece in a larger puzzle. For him, it came down to the fact that Plato 4 would need to have a better display and storage device. It just happened that along the way, he and his colleagues would create the first plasma screens. That's not to say that it was an easy invention. It took years of intricate work. But it wasn't an end unto itself. But it wasn't an end unto itself. Play-Doh 4 would be built and designed, while Play-Doh 3 was still in active use at the University of Illinois campus. Now, that makes it easy to think that 4 was just an evolution of 3.

Not only has the version number increased, but so too a lot of the new system's features are just plain bigger. But that misses a lot, and the fact is that PLATO4 was a shift for the project. We have already met the new display that was so core to PLATO4, but that wasn't the only change. That fancy new screen meant that PLATO4 could totally ditch the old way of sending CCTV signals and key presses separately. This new terminal used a modem, a device which allowed data to be sent two ways over a phone line. Thus, communications were made vastly simpler and more cheap. Play-Doh 4 also added a new form of input to augment the humble and now very well-used keyboard. That would be a touchscreen. That's

right, in the 1960s there was a computer with both a flat panel display and a touchscreen. Well, kinda. This is another place where we see the dichotomy of Plato's design. The touch detection part of the system was actually very low-tech, especially compared to the fresh-off-the-line miracle of the plasma display. A set of infrared LEDs and infrared light detectors were used to create an invisible 16x16 grid just in front of that new display, which makes 256 points that could register a separate touch. A touch would be detected by waiting for a finger to break one of those invisible lines of light. While crude, it was an effective way of creating a touchscreen using what limited technology was available. The other big challenge was on the back end of things. Even with the plasma display fixing some of the memory issues, the aging CDC-1604 would have to be replaced.

Bitzer was planning for the PLATO-4 system to be able to handle up to 496 terminals at once. That's a tall order for any mainframe to handle in those days. The lab eventually settled on using a more powerful CDC machine, the 6400. This new machine had a ton of memory, which meant programmers on Play-Doh would have a lot more room to breathe. But there's more to it than just that. For the time it was released, the CDC 6000 series was well known for being screaming fast computers. In fact, the 6600, a very close relative of the system initially used by PLATO 4, is known today as one of the first supercomputers. Compared to that, the 6400 was a little smaller, but it's clear to see that PLATO had moved on from its early days of ILLIAC. It was now playing in a whole nother league.

Despite the new supercomputer, and the new screen, and all the new internals to the terminals, the system wasn't totally new. Play-Doh 4 still had a slide selector, but it was radically changed. Each terminal now had a built-in microfilm projector, and some lessons still made use of slides in this new format. The plasma display came in handy here too. Since the screen was actually transparent, the microfilm image was projected just behind the terminal's display. It was a seamless and easy way to use slides, and a lot more simple than the massively shared slide selector of earlier iterations. Also stuffed inside each Play-Doh 4 terminal was a sound synthesizer, a voice synthesizer, and a device for playing back small audio samples.

The entire sound aspect of Play-Doh is something that I'm planning to cover at a later date, so right now, let's just leave it at the fact that these new models could do a lot, especially when it came to audio. In 1972, Play-DO 4 would officially launch. There had been prototypes in use for testing and demonstrations before that, but 72 is when installations would start to pop up in classrooms. This is around the time that the zoo that I mentioned at the top started to come together. There were a few large labs full of PLATO terminals on the University of Illinois campus, but that was just part of the new model. Remote terminals started to be installed almost immediately. All you needed was access to a phone line to connect up to the U of I supercomputer, so adding a terminal as far away as Hawaii was now possible and relatively practical.

A PLATO network would start to form across the country, completely separate from ARPANET. Using existing telephone infrastructure made expansion across America relatively easy. In the first decade of PLATO 4, new terminals would be installed as far away as Hawaii to Delaware. For a time, it was the largest networked system in the country. But what was it like to use a Plato terminal? The newest iteration of this project took a lot of cues from Plato 3, at least on the software side. It was totally self-hosting. A teacher or programmer could develop an entire lesson from a terminal and then students could use that lesson on the same hardware. But outside of things like text editors, lessons, and the more mundane is where things start to get interesting. Here's something to keep in mind. Despite being designed for teaching,

Plato 4 was essentially just a huge network of terminals connected up to a shared supercomputer. connected up to a shared supercomputer. In 72, nothing on that scale had ever existed. The ARPANET, an early ancestor to the internet, was a contemporary, but there were some major differences. First off, ARPANET was smaller than PLATO up until sometime in the early 1980s. There were also large structural differences. ARPANET was a distributed network of many, many mainframes, whereas PLATO was centralized around one powerful computer. This meant that for early internet users, the world was a set of interconnected islands. But for PLATO, it was more like a single shared continent. And then there's the matter of access. Plato terminals were targeted at schools, including primary, high school, and college level. So it had a pretty large and diverse spread.

On the flip side, ARPANET was used by some universities, the US government, and the military. Not quite as open as Plato. All this is to point out that despite having some similarities, Plato is a different beast than the early internet. I mean, it's entirely different than the modern internet also. It had different goals and a different type of end user. And in a lot of ways, Plato was a better incubator for innovation than the early days of the ARPANET. Even outside of the wires and bits, it had a different culture. Play-Doh is widely cited as one of the first online communities, and that was possible thanks to the software that users would write for it. One of the upsides to having such a large and

distributed user base was that a lot of new software was developed relatively quickly. One of the innovations that would tie together that community was a program called Notes. It was originally developed in 1973, first named Pad, and later incorporated into the core Plato system as Note Files. It filled a relatively simple space, one that we're all familiar with today, talking with other users on the network. There had been some earlier attempts at this on the system. One was a smaller program called Disqus. Another was just using empty files to pass notes around. Pad, and later notes,

was a much more sophisticated solution. Plato admins would maintain a list of topics, the epitomous note files, that a user could page through. Once inside a note file, you could create a new post, and then users could read your post and reply to it. Internet forums and eventually social media sites would come to replace this functionality, but Plato did it first, as far back as 73. Another development that shaped the community was an early instant messaging client called Tacomatic. Like notes, this is another program that seemed dead simple in retrospect, but for the time, it was a game changer. Some early Play-Doh 4 applications, primarily some video games, had a throwaway feature where users could talk to other

players with short messages. It was more of a novelty than actually useful, mainly used for trash talk. Eventually, Play-Doh users started to pick up on the possible usefulness of this feature, and Talk-O-Matic was written. Originally, the program could only handle one-on-one chats, but it soon expanded to running chat rooms with many multiple users. chats, but it soon expanded to running chat rooms with many multiple users. One quirk of Takamatic was that instead of sending a message once the inner key is hit like most modern chat systems, it would send out text as soon as a letter was typed. It made it feel a lot more conversational that way. The program became a hit almost immediately, and like with note files before it,

the programmers that maintained Plato itself noticed the demand for the program. Talk-o-matic was integrated into Plato as term talk, making IM chat a mainstay of the system. The centralized nature of Plato made programs like Notes and Talk-o-matic relatively easy to implement. And, in a sort of feedback kind of mechanism, these programs and other interactive content helped further centralize the Play-Doh community. But outside of lessons and this proto-social media, there is another broad category of programs running on these systems. Video games. Software and hardware preservation is something very near and dear to my heart. I usually save talking about that until the very end of an episode, but I think in this case, it's appropriate to start early. One of the big problems with

preserving systems like Plato or ARPANET or really any early network systems is the fact that you can't preserve that user base. A large portion of what made Plato, Plato was the community. And that's something you really can't emulate. Or at least we haven't figured out how to emulate people like that yet. But we do have the next best thing. CyberOne is a project that's working to keep Play-Doh software alive. And from what I've seen, they're doing a fantastic job of it. They maintain an emulated Play-Doh mainframe and a terminal emulator that you can download. All you have to do is sign up for a login, and then you can experience a lot of Play-Doh for yourself.

Of course, there aren't as many users online as there were during the system's height, but it's not entirely empty. Last time I logged in, there were something like 60 users online. Anyway, thanks to CyberOne, it's possible to still play a lot of early Play-Doh video games. And that's where we get into some very interesting territory. We've already seen the tendency for Play-Doh users to write networked software, and the games that they spun out are no exception. A lot of the games written for the system are networked multiplayer games, some supporting up to 32 players at once.

That's right. multiplayer games, some supporting up to 32 players at once. That's right, before the Atari 2600 was released, a lucky few were playing what amounted to MMOs. So, what were these games like to play? Now, it should come as no surprise that there were a whole lot of Dungeons and Dragons inspired games on Play-Doh. D&D had came out relatively close to the same time as Play-Doh 4, play-doh dnd had came out relatively close to the same time as play-doh 4 and a lot of game nerds are also a lot of computer nerds one of the earliest examples is simply called dnd and it's a pretty complete role-playing game written in 1975 dnd is one of a long line of dungeon crawling rpgs you control a character from an overhead view, you traverse a deep dungeon, fight monsters, loot treasure, and rack up points.

Something that I find particularly interesting about D&D is that there's actually an end goal. A lot of early video games didn't have a way to win, per se. Not so for D&D. The final challenge is to find an artifact known as the Orb. To do so, you have to fight your way down 15 brutally difficult levels and eventually defeat a golden dragon, the game's boss, and possibly the first boss fight in any video game. D&D is a fun game, but there are more impressive examples of that genre on Play-Doh. Also developed in 75 was a game called Moria. This is another dungeon adventure, but with a twist. The first difference you'll notice is that Moria is actually played in 3D.

Now, it's pretty primitive, this is 75, and the actual 3D portion of the game is rendered in a pretty small image, but I think we can forgive the authors for that. Like other early dungeon crawlers, Moria is essentially a huge maze. It's made all the more confusing since you see everything from the first person's perspective. You get used to it after a while, and that's when you start noticing what's just below the surface. Moria is still very fun and compelling. It's both a large and a deep game. The game world is broken up into two main sections. You have the city and the wilderness. Inside the city, you're safe. There's no monsters, and you can even find shops to buy and sell goods. The bartering system is pretty interesting on

its own. Instead of just selecting from a list of items, you get to haggle back and forth with a virtual shopkeeper, have a little conversation, and eventually you settle on a fair enough price. The city is also where you run into the next big difference in Moria. It's multiplayer. You can run into other players, form a party, and then travel out into the wilderness together. Once outside the city, you get into the danger zone, so to speak. The wilderness is populated with random monster encounters. When you run into them, you can either fight them or find some way to escape. Killing monsters gains you loot, and you can spend that back at the city. Unlike D&D, however, there's no end goal.

Instead, you mainly wander around a huge and confusing maze fighting monsters and buying more gear. Over the course of the system, maybe a dozen or more RPGs were made in a relatively similar mold to D&D and Moria. One other interesting game in that genre to note is called Future War. One other interesting game in that genre to note is called Future War. The difference here is in setting. Instead of high fantasy, Future War takes place in the far-flung year of 2020, in the aftermath of a devastating war between humans, aliens, and cyborgs. You control a human, alien, or robot, and traverse another large 3D maze with a team of other players. Now, I've heard a few comparisons of Future War to the later and much more popular Doom,

and there are marked similarities. The earlier game is far simpler, but a lot of the theme and setting is relatively similar. There are environmental hazards like rubble and radioactive waste, and Future War is actually a full-on first-person shooter. Your character is armed with a gun that stays in view as you travel through the maze, and you fight enemies in what amounts to real-time combat. But don't get the impression that Play-Doh only had RPGs. There was a widespread of games across the system's lifespan. Clones of D&D were just one of

them. There were all the usual suspects too. You got chess, you got checkers, you even have mahjong and a selection of card games all rendered in orange and black. There's even a flight simulator called Spazum, complete with 32 player combat. But outside of all that fun that could be had in those games was a whole other type of digital playground. The game was called Empire, and on release in 1973, it was a totally new type of experience. On the surface, Empire is deceptively simple. It's a multiplayer space combat game. From a top-down view, you pilot a spaceship around the galaxy, fight other ships, find planets, invade planets, and conquer. Empire supports up to 32 players on four different teams. You have the Federation, Romulans, Klingons, and the Orions. That's right,

Empire was made by a bunch of Star Trek nerds. The gameplay boils down to taking over the galaxy for your team, which is a lot easier said than done. If you've watched any Star Trek, then you probably know the drill. You have to manage your ship's weapons, shields, engines, and energy reserves. If you take damage, you have to send a party to make repairs lest your ship explode. It's the prototype of more modern arena shooters, complete with team-based combat. On its face, Empire is fun and engaging, but there's more to it than that just under the surface. You see, the entire game world in Empire is persistent, meaning that even after you log

off, the game keeps being played. You may finish a run with a quadrant under your control, turn off your computer, and go to bed only to wake up to find an enemy raid took over all your hard-earned gains. Since first showing up on terminals, Empire became a perennial favorite. Games like Moria were impressive due to their sheer scale, but Empire was something different. Both in and out of game, it was an addiction. There are a lot of stories floating around of teams meeting up to play late into the night in the zoo, only leaving in the early hours of the morning to get breakfast. Thanks to the persistence of the game,

any planets that were wondering of those all-nighters would remain in the victors' hands. At least for a while. Empire would have an early community of enthusiasts around it, and that would continue on for years. As I've been saying, a lot of these games are still very fun today, but there's some big caveats to that. One problem is the simple fact that Play-Doh never had a mouse, and as near as I can tell, its touchscreen was primarily used in actual lessons and not as much in games. This means a lot of keyboard mashing, which normally wouldn't be an issue. But here's the thing. Play-Doh used a very custom keyboard layout with now non-standard buttons labeled things like Next or Help, Data, Lab. Needless to say, a keyboard today doesn't have that. Beyond the archaic layout,

you still need to use keys and combinations of keys to do anything in a game. And since screen space was at a premium, a lot of games don't tell you what keys do what. In my experience, playing some Play-Doh games requires a lot of trial and error or consulting help files. That being said, if you take the time to log into CyberOne, you're in for a real treat. A taste of gaming history. Alright, that brings us to the end of our discussion of Play-Doh. How an ambitious educational platform evolved into a shining example of ingenuity. PLATO 4 and derivatives of the system would remain in service until as recently as 2004.

The technology eventually changed hands from the University of Illinois to CDC, the manufacturer of the mainframes used by PLATO. Over its considerable lifespan, uncounted users came into contact with the orange glow of the little machine. Most were students. Even with its thriving community of hackers and gamers, Play-Doh was still primarily a teaching tool. But the hackers and gamers of Play-Doh really helped push the system to the very limit of what was possible in that era. Many of the innovations that appear on Play-Doh in the 70s wouldn't be replicated elsewhere for years, sometimes decades. It truly created one of the first vibrant online communities, a feat that would be hard to match until the modern day.

Once again, I'd highly recommend tracking down a copy of The Friendly Orange Glow by Brian Deer. It's an amazing account of the history of Plato, and it's the most complete source that I've found in my research. And if you want to see what the machine was like for yourself, then check out cyber1.org. You can apply for your very own Plato account there. And thanks for listening to Advent of Computing. I'll be back in two weeks time with a new episode. Until then, if you like the show, please take a minute to share it with your friends. I also appreciate any ratings and reviews. You can post those over at Apple Podcasts. If you have any comments or suggestions for a future

topic, 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.