Advent of Computing - Episode 33 - Becoming Portable

Episode Date: June 28, 2020

Portable computing is now totally ubiquitous. There's a good chance you are listening to this episode on a tiny portable computer right now. But where did it all come from? As it turns out the first portable computer was designed all the way back in 1972. This machine, the DynaBook, only ever existed on paper. Despite that handicap, in the coming years it would inspire a huge shift in both personal and portable computing. 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: 1972: DynaBook designed by Alan Kay 1976: NoteTaker project starts 1982: GRiD Compass released

A hallmark of modern technology is that it's everywhere, and it can go anywhere. Nowadays, it's just second nature to walk around with at least one computer. For instance, let's take a look at our intrepid host as he leaves the house for work in the morning. I usually pack a laptop in a messenger bag. That's one computer right there. I'll always have a smartphone in my pocket to stay connected. It's smaller and less powerful, but that is a second computer. And I personally never leave the house without my smartwatch, which is really just a tiny wrist-mounted computer. Most of you listening probably have a similar experience. You won't go far without a computer,

at least nearby. We all look at computers in a very certain way. They are part of our everyday life, at work, at home, and wherever else we may be. The technology has become ubiquitous. It's part of our lives in a deep-rooted and very personal way. Part of what's made that possible, I'd argue a pretty large part, is the fact that there's now so many ways to bring a computer with you. A big part of making computers personal was making computers that could stay on your person at all times. In other words, making computers more portable. Now, you'd probably expect that the journey to portable computers has been a slow one, and you'd be right. It's taken decades and countless new developments to shrink all those parts down to size.

It's plain to see the steady progression even without looking very deep into the matter. However, that's just a cover story that hides more interesting forces at play. The fact is that plans for a small, portable, and highly personal computer were first laid out in the beginning of the 1970s. Since then, these plans have been brewing, changing hands, and spreading amongst influential people in the industry. Parts of these designs have made their way into classrooms, research institutions, the space shuttle, and hey, maybe even your pocket. So let's take a look at the Dynabook,

a theoretical computer designed in the 1970s, and see how the idea of portable computing started and how it started to spread. Welcome back to Advent of Computing. I'm your host, Sean Haas, and this is episode 33, Becoming Portable. Today, we're looking at what I think is a pretty strange story. That's the lineage of portable computers. Now, let me try to explain my thoughts and hopefully you'll come out the other side agreeing with me. So the IBM PC is often used as the mile marker for when the idea of a personal computer was fully realized. And

while I think IBM does deserve a lot of credit, the PC was ultimately just the winner out of a pretty large field of contenders. Just a few years later, the Apple Macintosh showed up, another computer that's been heralded as setting a new standard for personal computing. But once again, the Mac was just one computer out of many that were all aiming at that same idea. Each contender had its own take on what a personal computer would look like, and really how a personal computer would be used. For IBM, a personal computer ended up meaning a cheap and capable machine built with readily available and serviceable parts. parts. Whereas for Apple, a personal computer meant something more along the lines of, let's say,

a personable computer, a computer that anyone could turn on and use without any additional help. Those are just two computers, and we already have some radically different approaches to the same problem. So how do you best make a personal computer? Well, that really depends on who you ask. Portability would end up being another angle of attack. The idea being that a personal computer should be something you can use all the time, more than just at a desk. What makes this so strange is that the development of portable computers runs parallel to other attempts at personal computing. A lot of early portable systems look totally different,

and they're intended to be used in totally different ways than their contemporaries. But once you look under the hood, you can have almost the same hardware driving just a radically different device. And to me, it gets pretty weird to see a slate of familiar-sounding devices packed up in little-known portable systems. So today, we're going to look at these strange early portable computers and the ideas behind them. We're going to start off with the earliest writings about portable computing and see just how modern some of those ideas really were. From there, we'll move towards the first attempts at making portable computers a reality. The first major player in our story is Alan Kay, and I gotta say, I've really started to like Alan Kay. He's someone that isn't that well known out of computing circles, but he should be. He has a

rare mix of technical chops and the ability to explain his work elegantly and simply. And when it comes to early personal computing, Kay had some radical views, but his path to those views was a pretty unique one. Born in 1940, Alan Kay grew up on a healthy diet of science fiction. He often cites his fascination with sci-fi as a stepping stone into the sciences. Now, it's really a small leap from reading stories of space travel to wanting to learn about rockets. from reading stories of space travel to wanting to learn about rockets. Or in Kay's case, reading about futuristic machines led him to learning more about early computers. This would end up exposing Kay to some key ideas early on. And one of the most important may have been a paper called As We May Think by Vannevar Bush. This paper describes a theoretical machine called the Mimex,

a library in a desk that holds books on ribbons of microfilm. It displayed data on flat glass panels and lets the user edit texts and build up associations between pages at will. Just as an aside, if you want to learn more about the Mimex, I did a deep dive into the topic in episode 26. The Mimex would become a fascination for Kay in a number of ways. It was a device that seemed to be pulled out of the pages of a futuristic fantasy. But Bush proposed that the entire device could be built using 1940s technology. As we may think, it did a lot more than just set out a neat idea. It identified how a device like the Mimix could be built, and how it could change the world. It's the kind of article that really makes you

think, and it would help set Kay to thinking more and more about computing. It seemed like the most reasonable path forward would be to get a degree, probably in some computer-adjacent science. to get a degree, probably in some computer-adjacent science. But that wouldn't be nearly as interesting. By the early 1960s, Kay found himself leaving college, sans degree, and joining on with the US Air Force. But instead of flying jets, Kay actually ended up becoming a programmer this way. This may seem kind of strange, but the fact is that the government and the military have always been early adopters of new technology. And in the early 60s, the military was particularly interested in computing. During this period, the Air Force was one of the few institutions that could even afford computers.

So after enlisting in the Air Force, it was pretty quickly realized that Kay had an aptitude for these new technologies. So, he was shuttled off to a government lab to program. For the next few years, Kay would work with early IBM mainframes and eventually some of the first supercomputers. He'd come out the other side of his tour as a very experienced programmer. In 66, he would finally get back to a more ordinary path, enrolling in the University of Colorado Boulder for undergraduate, before leaving to the University of Utah to pursue a PhD in computer science. It was during his graduate studies that Kay started to think more and more about personal computing. He had worked on computers personally for years at this point,

and he knew just how world-changing the technology could be. He also recognized the problems that computers of the era had. But he had a slightly different take on the state of things than a lot of his colleagues. One recurrent refrain from computer scientists is that machines need to become faster and more powerful. But Kay didn't fully agree. Instead, he thought that computers needed to be made more accessible and, eventually, more personal. In his mind, for computers to reach their full potential, they needed to be usable by everyone and anyone. It was while working on his PhD thesis that Kay would come into contact with another infectious idea. We don't have an exact date, but probably sometime in 1967 or 68, Doug Engelbart came to the University of Utah to show off his research into computer interfaces.

At the time, Engelbart was working on the first graphical user interfaces. His system, called the online system, has become legendary for pioneering computer graphics, hypertext, network collaboration, and much, much more. Even the mouse was invented in Engelbart's lab. On his trip to Utah, he gave a talk and showed off some video of NLS running. Kay was one of the attendees to that talk, and one of the converts. Engelbart and his colleagues were taking the first big steps towards personal computing, and Kay was ready to take the steps of his own. His first stab at tackling this problem would take place while he was still in grad school,

and it was called Flex. While less well-known than Kay's later work, I think Flex offers a good example of the out-of-the-box approach he took when solving problems. As Kay explained in his thesis, quote, the Flex system consists of merged hardware and software that is optimized towards handling The idea was to combine specialized hardware with a specialized programming language to create a more user-friendly experience. Here, the man-machine dialogue is key. Kay was trying to find a way to make using a computer feel more effortless, more natural. One side of that was pushing Flex past the standard button and keyboard interfaces common in the 1960s. Kay's machine was capable of displaying simple graphics and handling input from a stylus on a sensor pad. And the entire machine was designed a stylus on a sensor pad, and the entire machine was

designed to be placed on a desktop. On the other side, K worked to develop a higher-level programming language, one that would be easy to learn and easy to use on a day-to-day basis. This would end up being an early object-oriented language, one which describes data and functions in terms of classes and sets. You can even catch a whiff of mimics and NLS in Kay's description of Flex. You were able to create documents and data, complete with hand-drawn images. You could save those documents, link them to other documents, and traverse trees of information with ease. Flex displayed everything on a bitmapped screen, and was controlled with a pointer and a keyboard combo. K had every intention of carrying on Flex beyond

just a thesis project, but that part wouldn't pan out. Once the prototype was up and running, K started to show off his work to non-computer literate folk. Much to his dismay, they just didn't get it. More than that, they didn't like trying to use it. Flex was technically impressive, that's true. But only if you knew what you were looking at. There was just something missing in Kay's calculus. After earning his doctorate, Alan would go into another period of migration. He left Utah for Silicon Valley, spending time around Stanford's AI lab. His initial plan was to go back into academics, but plans changed. In 1970, he was hired to work at the fledgling Xerox Palo Alto Research Center, better known as PARC.

For someone like Kay, who had a profound interest in pushing the boundaries of computing, PARC was probably one of the better places to be. And yes, this is the same Xerox that makes copy machines. On the outside, Xerox doesn't seem like a particularly vibrant company when it comes to innovation. But PARC was a little bit different. It started in 1970 as a subsidiary under Xerox with an express focus on research. And while some of that research was related to printing and copying, the vast majority wasn't. More generally, Park's research was concerned with all aspects of the office, and specifically personal computing, would become a big part of the vision of a modern office. Thanks to the backing of Xerox, Park would have a wide berth to experiment with just what a computer could do and what it could be made to do. It was a nurturing environment for wild ideas, a cross between a functioning company and a think tank.

a cross between a functioning company and a think tank. It was the perfect place for Kay to grow his ideas on personal computing and the perfect place to deconstruct what went wrong with Flex. Over the next two years, Kay would compile his ideas into a new design, which he soon called the Dynabook. The idea had been brewing since before he joined Park, but the full description was codified in 1972. Kay's breakthrough was that a personal computer had to be personal for anyone, not just experienced computer users. That was what had been missing from Flex. Being caught up in the advanced

technical aspects of his work made it easy to miss that most people could just never use the system. If computing was going to make its way out to the untrained masses, it had to come in a form that didn't require training to use. Or at least not very much. But at the same time, that computer would also need to be useful for the more experienced operator. How Kay came to this realization is pretty interesting in its own right. During his migratory period between college and park, he had the chance to visit some of the first labs where school children were being taught how to program.

At that time, the cutting edge in the field was a platform called Logo. It was a stripped-down programming language that was easy enough for kids to pick up. By writing out a Log logo program, a student could program a small robot to move around a classroom and draw lines and shapes with a pen. Kay hadn't seen anything like this before. The idea of creating a system for teaching children about computers was totally new. But it was one of those creations that makes total sense. Computers would become increasingly important in the future, so why not get kids equipped with these new skills as early as possible? Kay really liked the overall premise of early computer education. But he brought his own unique

twist to the concept as always. He realized that Logo had a very similar problem to his own Flex. Well, they actually had sort of opposite problems. While Flex was great for experienced users but useless for the untrained, Logo was good for the untrained while being useless to an experienced computer user. K's vision would start to shift more towards a unified system that someone could grow with. On one end, a child should be able to learn how to program with the system, and then, as they become more proficient, keep using the same computer. Once at PARC, these ideas and musings became much more solid. Given the right environment and resources,

Kay was able to make some solid plans. And in 1972, he published those plans in a paper called A Personal Computer for Children of All Ages. If you go in unprepared, then this is a really, really weird paper. It's the first description of the Dynabook, and one of the first descriptions of a mobile computer. However, don't be fooled. This is not a technical document. Kay opens the paper with this. Quote,

This is not a technical document. Kay opens the paper with this, quote, This note speculates about the emergence of personal, portable information manipulators and their effects when used by both children and adults. Although it should be read as science fiction, current trends in miniaturization and price reduction almost guarantee that many of the notions described will actually happen in the near future. Instead of a dry treatise on how to make a computer more user-friendly, Kay presents speculative fiction. He spends a page or so describing the current problems facing computers

in education, and the use of computers in society at large. The first part ends with Kay offering a solution, a new device which can change the way the world works and how we look at computers, something described as a quote-unquote active book. In most papers, once you're past the introduction and some airing of grievances, you'd expect a full detailed explanation of your solution. But in reading and listening to Alan Kay as much as I have been in preparing for this episode, I can tell you one thing. The expected really isn't his style. Quoting again from his paper. Zap! With a beautiful flash and appropriate noise, Jimmy's spaceship disintegrated. Beth had won space war again. The nine-year-olds were lying on the grass of a park near their home, their Dynabooks hooked together to allow each of them a viewscreen

into the space world where Beth's ship was now floating triumphantly alone. End quote. Yeah, the next page and a half is a short story composed of two vignettes. The first one is about two school children playing on their Dynabook and learning how to program together. The second covers a businessman on a flight. He's getting some work done using his Dynabook, and then eventually he accesses a kiosk to download a book to his computer. There is absolutely no explanation of what a Dynabook actually is until the last half of this paper. So, yeah, a personal computer for children of all ages is often cited as a watershed document for the development of the personal computer. And the middle section is just a science fiction short story. This is not normal at all. And the first time I read it, I had to go back and make sure I had the right file pulled up.

It's not until we get close to the end of the document that Kay actually spills all the spicy details on the Dynabook. So, why did Kay write like this? I think it comes down to his influences. Specifically, Bush's article called As We May Think. In that text, bush laid out the current issues of the time gave a fictionalized account of someone searching through a library from the comfort of their mimics and then he went back to describe what the mimics actually was k is paying homage to bush's work albeit in a slightly more playful manner. With that aside, what actually is

the Dynabook? Well, Kay describes it in shockingly modern terms. It's a flat slate. It's about 9 by 12 inches, less than an inch thick. It has a large rectangular plasma display located directly above a small keyboard. Inside the Dynabook's Slim case is a microprocessor, a good amount of RAM, some type of long-term storage, he's vague on the details, a stylus, and some sort of hardware to manage data I.O. Each of these components has a section dedicated to it, giving some options and explanations for his choices. For instance, Kay explains that a physical keyboard would work fine, but that it would be better to ditch the keyboard altogether so that the screen could be made

bigger. By making the screen touch-sensitive, a keyboard could easily be displayed on the screen itself. He also spends little time discussing the software that his new machine will run, but Kay's a little sparse on those details. The most that he says is that it has to be easy to use and powerful, with the ability to handle graphics and sound as well as text. Overall, the Dynabook sounds an awful lot like a modern tablet computer, but it's described in this strange sci-fi paper from 1972. It sounds like really modern hardware, as described being used in very modern ways, but it's just a description. Ultimately, the Dynabook would be restricted to life on paper.

One of Kay's bold claims was that the Dynabook could be built using 1972 technology, or at least technology currently in development in 1972. at least technology currently in development in 1972. What's more, he says that the Dynabook could be mass-produced and sell for as little as $500. This idea that Dynabook is soon to be possible and at such a cheap price is an important aspect of Kay's vision for personal computing. However, in 72, this type of computer would remain a fantasy. Just for some context, the first microprocessor would hit shelves in 1971. This first chip, the Intel 4004, was an extremely limited processor. By 1972, there were a handful of other options,

but none were particularly powerful or particularly cheap. And there was no way any of them could handle graphics or a sound for that matter. Even something like the Dynabook screen would prove to be a non-starter. Flat plasma displays did exist, but they were about an inch thick, relatively low resolution, and they required a good deal of power to run. While practical for desktop computers, it wasn't quite portable yet. The rest of K's laundry list of components also had similar issues, mainly on the side of cost and power consumption. It just wasn't practically possible to build a real Dynabook, at least not as Kay described it. Over the next few years, progress at Xerox PARC would continue, and although Kay's Dynabook concept was striking

and exciting, it just wasn't something feasible. And when you get down to it, the Dynabook just wasn't possible yet. Kay may have gone a little too far into the realm of science fiction a little too quickly. That being said, a lot of the ideas from a portable computer for children of all ages would stick around in the air at PARC, showing up in other projects. The largest development at PARC in this period was the Alto, a fully graphical computer. The machine would be a huge milestone. Most modern user interfaces are based, in part, on the Alto's work. Windows, icons, a mouse pointer, drop-down menus. Alto would pioneer or greatly improve all of these in this time period. K himself would be instrumental in this project, contributing greatly to the software

side of things. But he didn't really like the Alto. Sure, it was a big step towards a more personal computing experience, but a Dynabook it simply was not. The Alto was big, it was expensive, and even with its graphical interface, most users still needed a good bit of training to use the thing. It definitely wasn't a machine made for children of any age. As the period drug on, K was starting to see a little bit of flex showing up at Xerox.

To him, it looked like the Alto team was set to make a great computer, but not a personal computer. So, he decided the best thing to do would be to jolt his co-workers out of this funk. To quote Kay, professional needs. In January of 1976, I took the whole group to the Payaro Dunes for a three-day off-site to bring up the issue and try to reset the compass. It was called, Let's Burn Our Diskpacks. I used the old aphorism that no biological organism can live in its own waste product to plead for a really fresh start." This strategy had worked for Kay in the past. When he was stuck with Flex, taking a break and restarting had helped him greatly. And evidently, it did some good for the rest of Kay's team. As it turned out, his enthusiasm was contagious as ever.

From these three days outside the office, a new plan would form, soon named the NoteTaker. The new design was a sort of compromise. You could look at it as the Dynabook Revision 2, almost. Gone was the tinge of futurism of the original Dynabook design. The first drawings for the NoteTaker look much more like a typewriter with a small screen mounted above it. look much more like a typewriter with a small screen mounted above it. Plans at this stage were pretty rough, but the core concept was to take what the team had learned from Alto and create a scaled-down and portable version. It would be less powerful, but it would be able to fit on your lap. The note-taker would have a similar software design as that of Flex. K had already created an

object-oriented language environment for the Alto, called Smalltalk. This new laptop would run a refined version of that environment, the idea being that with some refinement, it could be made easier to use while retaining a reasonable degree of power and flexibility. The first draft of the Noteetaker was a much more pragmatic design. Unlike Dynabook, it would lead to a real-world computer. In 76, the project was made official, and a small group of researchers at Xerox set to work turning early drafts into early prototypes. Even from the start, it was clear that Notetaker would be very ambitious. And why shouldn't it be?

Park had already proven itself capable of making gigantic leaps, so Notetaker would just have to be another. Adele Goldberg, a talented programmer just coming off the Alto, would handle most of the Notetaker's environment. That included a new version of Smalltalk for the smaller system, as well as all the graphics. The rest of the team, Larry Tesler and Doug Fairbairn, would deal with the hardware side of the equation. The biggest issue for the note-taker was by far the hardware. Kay's assertion that something like the Dynabook could be built with 1970s technology was starting to look totally unrealistic by this point. In 1976, options for small computer components were still pretty sparse.

It wouldn't be until 78 that the team actually got their hands on some hardware at all. Planning, drafting designs, plans failing, and then redesigning ate up a few years of work. After two years of the song and dance, Fairbairn and Tesla finally settled on their microprocessor, the extremely new and untested Intel 8086. Now, this is one of the parts that I think is really interesting. Just a few years later, in 1981, IBM would launch their own 8086-based computer. For those playing along at home, that computer would be the IBM PC. IBM's system would become the personal computer. But here we have the same processor, and what would turn out to be really similar hardware in general, being put to a very different use. The PC would be a pretty capable machine for 1981,

but the note-taker was on a whole different level. Despite being worlds apart, both use the same chips, and I find that really interesting. Now, IBM tangent aside, Xerox would actually be one of the first places to get access to the now-legendary processor. The Notetaker team got their hands on some pre-release 8086 chips. So new that they still had some bugs left to be ironed out before full production. But with processors in hand, the Notetaker started to take shape pretty quickly. And for the limited technology available at the time, the finished machine was impressive. It wasn't a dynabook, and it deviated quite a bit from the design that was hashed out by the crew's desert trip. But nonetheless, Notetaker was a

futuristic piece of hardware, and it actually existing definitely helped. The finished prototype weighed somewhere in the neighborhood of 50 pounds. The fact was that it just wasn't possible back in the day to make a computer the size of a tablet. The form factor of the Notetaker is also pretty strange, at least to our modern sensibilities. The chassis was about the size and shape of a portable sewing machine, a bulky rectangular box about 22 inches by 14 inches wide and 7 inches tall. This isn't something that you could ever operate on your lap. Inside that box were all the carefully crafted circuits that made the thing work,

a motherboard built around the 8086, 256 kilobytes of RAM, a small speaker, a floppy drive, a screen control circuit, rechargeable batteries, and an ethernet card. That's right, this thing had ethernet way back in the 70s. For storage, the Dynabook used a single 5.25 inch floppy disk drive. Even one of the core features of the Dynabook, the flat panel display, is absent in this computer. It was replaced by a 7-inch CRT screen built into the case, complete with a small touch-sensitive panel mounted just in front of the screen. The final touch was a detachable lid that protected the screen and held a keyboard and a mouse. When you get down to it, the Notetaker was basically a desktop computer with a screen,

crammed into a smaller case. And thanks to the Intel processor on the motherboard, it's likely that the system would function in a pretty similar way to the later IBM PC. Or at least its internals would be similar enough to a PC enthusiast. As far as portability goes, I mean, you can move it. The form factor would later be copied and given a name, the luggable computer, and I think that sums up the situation pretty well. Sure, you can lug a full computer around with you. The system did have integrated batteries, so in theory it could operate anywhere,

but I've seen no details on how long the note-taker could run on battery power alone. My guess would be not that long, especially since a CRT tube and the motors used in a floppy drive, they draw a good deal of power. So this sounds like a pretty abysmal portable computer, at least by standards today. But let's set our preconceptions aside for a second and consider this. Up until this point, the portable computer had only existed on paper and in science fiction. The idea of taking a computer with you was a wild fantasy.

Look no further than my favorite sci-fi technology example, 2001 A Space Odyssey. Part of setting up a futuristic setting of that film is showing off some wild technology. Spaceships and AI supercomputers are just part of that setting. You can see early on in the film, more of the protagonists casually using a portable tablet-like computer to read the news. That scene was included because at the time the film was made, around the same time that Kay was first thinking of the Dynabook, the idea of a portable computer was almost an absurdity. Taking that into account, a machine like Notetaker, with all its blemishes

intact, would have been breathtaking. The other amazing part is that Notetaker had a graphical environment. Goldberg, the primary programmer of the system, had managed to squeeze an environment similar to that of Alto onto the relatively small computer. Now, we can only imagine how difficult a feat that must have been. This meant that Notetaker could be used primarily with a touchscreen and a mouse. A machine like the Alto was already a giant leap forward, and Notetaker was taking that technology to a whole nother arena. As far as what the system actually was like to use, that's where we run into some issues with sourcing.

I can't find any pictures of the Notetaker in use, except for a blurry photo of a Xerox employee sitting in front of one. I also can't confirm what the interface actually looked like. The paper, The Early History of Smalltalk, written by Alan Kay, has a few photos of what may be the note taker's screen in action, but they're labeled in such a way that it's kind of ambiguous if they're actually from a note taker or just a cropped screenshot from an Alto. The final fate of the note taker is quite a bit more concrete, though. It died out. In all, 10 prototype notetakers were produced at Xerox. The most cited reason I've seen for the failure to thrive is that the executives at

Xerox just didn't get it. Not enough people were interested in the project, so it was just scrapped after only 10 machines were built. But there is a chance the notetaker did actually spread its legs outside of Xerox Labs, at least in a limited fashion. According to Tesler, at some point in 1979, he and Fairbairn were on their way to Xerox's main office in New York. They had brought along a prototype notetaker to show off to some of the higher-ups. The duo was waiting for a late flight in SFO when

they realized that the terminal they were sitting in was totally empty. At this point, the notetaker was still an internal project, so they weren't at liberty to show it off outside the lab. So, they'd been somewhat ironically unable to use this portable computer anywhere but at a test bench. But now, they had a chance. Tesla and Fairbairn pulled out the machine, flipped it on, and in Tesla's words, quote, that was the first time a portable computer was ever used in an airport, end quote. The two would later repeat this feat on the flight east, quietly becoming the first passengers to use a computer mid-flight. But for all its promise and potential, the note-taker would never take flight again.

This leads us to our final stop on the tour of all things portable, and we're about to depart from Xerox. Well, sort of. You can never really get away from Xerox. As it turns out, Xerox PARC is really the cradle of the portable computer. And even once the idea spread, it would still be heavily informed by work done at Park. John Ellensby joined Park in the 1970s. While there, he was heavily involved with the Alto and the Alto 2, a revised version of the original. Once at Xerox, it was only a matter of time before he came face-to-face with Alan Kay. Ellensby quickly became another devotee of the Dynabook concept, and portable computing in general. Over the next few years, those ideas of what a computer could be started to clash with what he was working on at Xerox. One of his major projects at Park

was the push to commercialize the Alto. Very slowly, a few of the wonder computers did make their way outside of the lab and into some type of use, primarily in government and some academic settings. The idea was to get the machine into real-world use to figure out if people on the outside would actually be interested in the technology. One day, sources are unclear, probably 1978, Ellens B. was talking with one of those supposed early adopters of the system, an executive branch assistant at the White House. It turned out that they loved the Alto. It did everything they wanted.

But they actually never used the machine on a day-to-day basis. Ellensby recalled his contact saying something to the effect of this. Quote, I learned enough about it very quickly. It's very simple to use. I've got my secretary using it, and she's a bright person, and she's the one that uses it. Is that how you think it's going to go forward? No.

Let me tell you what I want. I want all the capabilities of your beautiful Alto, and I want it in half the size of your briefcase. The Alto was great. capabilities of your beautiful Alto, and I want it in half the size of your briefcase. The Alto was great. It was a powerful computer with amazingly user-friendly software. The problem was that it was chained down to a desk, and not everyone liked to work at a desk. This stuck with Ellensby, because what this tester was asking for was very familiar. It was the Dynabook. Or at least something similar. This was something that Xerox wasn't able to deliver. So in 1979, Ellensby left Park and started his own company called Grid Systems. Grid had the same goal as the earlier Notetaker project. Build a Dynabook. By this point, there

were already some portable computers coming into the market with similar form factors to the earlier Notetaker. But there was nothing that could come close to the Alto that was small enough for a briefcase. From the beginning, that would be John Ellensby's goal, and Grid worked aggressively to reach it. Over the course of 79, Ellensby pulled off a number of his old coworkers from Xerox to join in on his new scheme, and he'd supplement that pool of employees with other like-minded computer scientists and engineers from around Silicon Valley. The team's first big realization was that given the technology of the time, something like the Dynabook was

almost possible. They would have to use technology that was either totally new, or just about to hit the market. The upside to this is that they could make a practical, portable computer. The downside was that it would be incredibly expensive. One thing that made Grid distinct from Park was that they took a much more pragmatic approach. Kay had envisioned the Dynabook as a tool for everyone from children to adults. Ellensby was more pragmatic on the matter. While they could make a portable computer, it would be far too expensive for anyone to buy for their kids. So pretty quickly, the market was determined to be on the higher end of the continuum, mainly targeting big business and government. The final

design of the computer, which would come to be known as the Grid Compass, took a lot of cues from K's work, but everything would be slanted towards this new market. It had to be a real product after all, it couldn't just be a fantasy. One of the most striking things about the Compass is its case. It wasn't a luggable, it wasn't like the NoteTaker, and it wasn't a flat slate like the Dynabook. The machine was what we would recognize as a fully formed laptop, a computer with an integrated flat panel display that folds down over the keyboard. The Compass was the first to use this now iconic form factor.

From there, it would spread to become the default. It doesn't look exactly the same as a modern laptop. There's a protruding back section beyond the screen hinge that holds a lot of electronics. But it's really close. For something as important as the clamshell design, the details on its creation are surprisingly sparse. The case design was done by Bill Mogridge. So he's the one that came up with the idea for the folding lid.

In later interviews, he talks about how he went through a lot of design options for the case, starting with a flat slate and eventually landing on the clamshell. But he doesn't go into much details beyond just saying that, well, the clamshell just felt natural. I've combed through other interviews with X-Grid employees, and they all have a similar story. The only added details are that the folding lid helped protect the fragile display, and that it proved to be more ergonomic than a slate. The final piece that makes the chassis so unique is that it's composed entirely of cast magnesium. This was done because it offered a good sense of durability while staying relatively lightweight. After the case, the next big change you notice is the aforementioned

screen. It's a flat panel display. Here, the DynaBook and the Compass are in total agreement. Ellensby knew that early on, CRT displays were going to be a non-starter. Even a small display like the one used in the Notetaker still takes up quite a bit of space. Add to that the power needs and the hot cathode tube, and you get a recipe for a bulky, hot, and not very usable machine. But that led to the problem of a replacement. Kay's conception of a flat plasma display came from seeing demonstrations of early gas plasma screens that were developed during the PLATO project at the University of Illinois. By the end of the 1970s, these flat panel displays had developed into

larger, more reliable screens, but they weren't a good fit for grid. They could be a bit delicate, they were made out of glass. Like CRTs, they consumed a good deal of power, and they weren't self-illuminating, so you'd need a separate way to light the screen from behind. The screen looked nice, and it did a good job, but they weren't ready for on-the-go use. Eventually, Ellensby and company found a more realistic alternative. This came in the form of an electroluminescent display. These were thinner, more robust, offered a wide viewing angle, and they required no additional backlight. But, as with most components of the Grid Compass, there wasn't an off-the-shelf

display that was up to the crew's specifications. No one was making an electroluminescent display that was large enough for the new laptop. So, Grid had to do some convincing. Eventually, they talked the manufacturer into building them a batch of custom screens. The final display was a 6-inch screen with a resolution of 320x240 on orange and black pixels. Custom order parts really became the norm for Grid. Even the keyboard had to be custom made, since no company had an existing keyboard that was small enough. The only other major surprise from the exterior of the Compass is actually an omission. There's the expected slate of ports and plugs around the back and sides, but you won't find the familiar slot of

a floppy disk. The Compass didn't have an internal floppy drive. In fact, it didn't even have an internal hard drive. Those components just took up too much space, so alternatives had to be found. Of course, you could get an external floppy drive to plug into the device, but the base laptop had to be slimmed down as much as possible. So if the stock compass came with no hard drive or floppy drive, then where did all the data go? That brings us to the internals of the machine, and a lot more cutting-edge technology. All internal storage came in the form of packs of bubble memory. Now, you're probably not that familiar with this technology. I certainly wasn't coming into this. It's strange. Like, really strange.

Basically, bubble memory is a type of non-volatile RAM. Its function is similar to that of a solid state drive, but it's a little bit different. It was developed in the 1960s at, surprise surprise, Bell Labs. But by the 1970s, Intel was producing commercial bubble memory packs. It was faster than a hard drive, a whole lot smaller, and like modern flash storage, it had no moving parts. For a good decade or so, bubble memory was the bleeding edge in storage. But once hard drives started to get smaller and better, the technology kind of died out. While Grid was developing the Compass, it was the best option available, but the stuff didn't come cheap. Ultimately, the Compass would ship with only 340

kilobytes of bubble storage, which ended up being just enough for most use cases at the time. The rest of the machine's loadout gets back into recognizable territory. Strangely enough, the processor is also the Intel 8086. But don't be fooled, this wasn't a PC clone. Like the NoteTaker, it was roughly similar to the PC, but the compass would settle in on the 8086 before IBM did. The rest of the laptop is loaded out with pretty standard RAM, ROM, a custom-built miniature modem, and an integrated power supply. Conspicuous in its absence is a battery pack. That's right, the first portable computer didn't have its own source of power. If you wanted to use the thing, you still had to be near a plug. From what I can gather,

the lack of a battery seems to come down to size and weight constraints. Rechargeable batteries simply weren't small back in the day. The last big feature of the Compass was, of course, the software. This is another place where a grid takes a turn for the unexpected. Despite being close to a PC, this laptop didn't run DOS, at least not on release. It came loaded with a copy of Compass Computer Operating System, or CCOS, baked into ROM. The custom OS was developed totally in-house at Grid, and was at least partially inspired by systems at Xerox. CCOS, sometimes called GridOS, was a fully graphical operating system, but it didn't use a mouse. Once again, part of the

rationale behind this choice came down to portability. A mouse was just another thing to lug around, and then you'd be restricted to operating the laptop on or near a flat surface. The decision may have also come down to making the compass more user-friendly. Now on the surface, that may sound counterintuitive, but let me explain. The mouse still hadn't made its way very far out of Xerox, and a few research labs. The general public had no clue what a mouse was, and the market the grid was pointing at definitely would not be familiar with a pointing device. Setting up a training program wouldn't be economical or reasonable for a consumer product,

especially coming out of a startup like Grid, so a mouse was just never in the cards. Then, what did the graphical OS actually look like? It's primarily menu-driven, supplemented with control key combos. Via the keyboard, a user could traverse menus and run programs and open a file. A base install of CCOS came complete with a small suite of office programs, including a text editor, a spreadsheet, a day planner, and a terminal which could connect up to remote servers. Out of the box, the Compass was a capable tool. The other big impressive feature was multitasking.

CCOS could run multiple programs at once. This feature wasn't yet commonplace on microcomputers, and it wouldn't be for years. You could switch between programs at the push of a key, or break the screen up into multiple tiled windows to display each program simultaneously. Overall, it's a compelling environment. And it's a more practical take on systems like the Alto or Flex. People weren't used to pointing at things with a mouse, but they were used to using a keyboard. The Compass and all its accompanying software

would be completed and ready to ship in 1982. That was less than a year after the IBM PC launched, and I really like looking at the differences between these two machines. Each used the same processor, and each was marketed as a productivity tool primarily for business. The Compass was, generally speaking, a much more capable and impressive computer thanks to well-built software and really a lot of new and custom technology. The PC, on the other hand, took a much more conservative route. It was a workhorse made from easily sourced parts, but ultimately, the same high-end tech components that made the Compass possible would greatly restrict its use. On

release day, the laptop cost just over $8,000, or, adjusting for inflation, about $21,000. The PC wasn't portable, or nearly as advanced, but it cost only $1,500. With Grid, you paid a high premium for portability. A device like Kay's Dynabook was roughly possible in 1982, but it definitely couldn't be cheap. But that's not to say that the Compass was a flop. Far from it. The original niche that LNsby targeted was really taken with the computer. And there was another customer waiting in the wings. The US government, specifically the military, really, really liked Grid's hardware.

There wasn't any other option for an on-the-go computer at the time, at least none that were as portable as the Compass. That alone made the new laptop ideal for a wide range of applications within the government. The device would find its way into the halls of Congress, Humvees, and helicopters. Most impressively, in 1985, a modified grid compass was flown on the space shuttle Discovery. It was the first consumer computer in space, a feat that no other portable computer could hope to accomplish for decades to come. Alright, that does it for this episode. We looked at how the idea of a portable computer tied into the larger idea of a personal computer. The first designs for a portable system would stay tucked away on paper for decades before any physical machine would be

built that remotely resembled it. But even on paper, Alan Kay's Dynabook described a shockingly modern computer. The Dynabook concept wasn't Kay's first attempt at designing a personal computer, but building off his earlier experience, he really stuck the landing. A lot of what Kay described, both in hardware and how it was used, was spot on. Even Xerox PARC, maybe the most advanced research center for personal computing at the time, just wasn't able to keep up with Kay's futuristic vision. So, this leads to one final question. Was the Dynabook finally achieved at grid? Well, yes and no. Broadly speaking, the Compass implemented every hardware and software feature in K's earlier

plan. It was portable. It was easy to use. It was a personal computer that could be always on your person. The technology was there, even if it did diverge from the original form factor. But ideologically, Grid was still leagues away from the Dynabook. It would take a few more decades until cheap and easy-to-use portable computers would come to the market. Sure, we do have tablet computers that are eerily similar to Kay's 1972 drawings, but in a broader sense, the work done by Alan Kay and his fellow believers helped push what was possible. Belief in a personal computer, and a portable computer that could be used by anyone, anywhere, was the first step towards the systems we use today.

Thanks for listening to Advent of Computing. I'll be back in two weeks time with another piece from the story of the computer. And hey, if you like the show, there are now a few ways you can support it. If you know someone else who'd be interested, then why not take a minute to share it with them? You can also rate and review me on Apple Podcasts. And if you want to be a super fan, then you can support the show through Advent of Computing merch or signing up as a patron on Patreon.

Patrons get early access to episodes, polls for the direction of the show, and other assorted perks. You can find links to everything on my website, adventofcomputing.com. If you have any comments or suggestions for a future episode, then shoot me a tweet. I'm at adventofcomp on Twitter. And as always, have a great rest of your day.